Biologics Bioanalysis - SCIEX · of bioanalytical quantitation such as proteomics, anti-doping,...

GUIDE TO INNOVATION

Bio

log

ics

Bio

anal

ysis

Biologics Bioanalysis – Guide to Innovation

Pharmaceutical companies have leveraged advancements in basic science perhaps more than any other industry. With the advent of whole genome sequencing, sophisticated analysis of metabolic pathways, and exponential improvements in computer processing, R&D organizations have expanded their drug portfolio focus on small molecules to encompass a new class of drugs — biotherapeutic compounds and biomarkers.

Helping customers by listening to their ideas, participating in discussions, and creating cutting-edge solutions to research challenges is top priority at SCIEX. The following compendium includes key solutions for peptide and protein bioanalysis — and, more importantly, describes in detail work done by, and in collaboration with, our customers. Your success is our success, and the SCIEX team will partner with you to overcome the emerging challenges of bioanalysis, now and into the future.

Joe FoxSenior Director – Pharmaceutical Business

RUO-MKT-02-1486-A

Tokyo, Japan

Redwood Shores & Dublin, CA, U.S.A.

Concord, ONT,Canada

Framingham, MA, U.S.A.

Warrington,UK

Paris, France Darmstadt, Germany

Milan, Italy

Melbourne, Australia

Shanghai, China

Singapore

Seoul, S. Korea

For Research Use Only. Not for use in diagnostic procedures.

SCIEX global associates and sites

1,500+ associates worldwide

260+ hold PhDs or other advanced degrees

14 operating sites

Global sales and service teams

7 global demo labs

Ultimate sensitivity

Advances in the Bioanalysis of Protein and Peptide Therapeutics through Innovations in Mass Spectrometry

Technology Drives High Performance in Biomolecular Mass Spectrometry

Achieving Low-Flow Sensitivities for Peptide Quantitation Using Microflow Rates on the QTRAP® 6500 System

A Sub-picogram Quantification Method for Desmopressin in Plasma Using the SCIEX Triple Quad™ 6500 System

High-Sensitivity Quantification of the Triptorelin Decapeptide Using the QTRAP® 6500 System

High-resolution accurate mass

High-Resolution Time-of-Flight for High-Quality Quantitative Analysis

Ultrasensitive Quantitation of Exenatide Using Microflow Liquid Chromatography Systems and High-Resolution Mass Spectrometry

Investigating Biological Variation of Liver Enzymes in Human Hepatocytes

Quantification of Large Oligonucleotides using High Resolution MS/MS on the TripleTOF® 5600 System

Increasing LC/MS Assay Robustness through Increased Specificity Using High Resolution MRM-like Analysis

6-29

6-13

14-15

16-19

20-25

26-29

30-55

30-31

32-37

38-43

44-49

50-55

Con

tent

sTherapeutic Peptide Bioanalysis

The need for sensitive and selective quantification methods

RUO-MKT-02-1486-A

Orthogonal selectivity tools

Application of Differential Ion Mobility Mass Spectrometry to Peptide Quantitation

Improving Intact Peptide Quantitation with Differential Mobility Separation and Mass Spectrometry (DMS-MS)

Differential Mobility Separation Mass Spectrometry for Quantitation of Large Peptides in Biological Matrices

UGT Family of Enzymes: Quantification of Tryptic Peptides using SelexION™ Technology on the QTRAP® 6500 System

MRM3 Quantitation for Highest Selectivity in Complex Matrices

Improved Selectivity for the Low-Level Quantification of the Therapeutic Peptide Exenatide in Human Plasma

Quantification of Prostate Specific Antigen (PSA) in Nondepleted Human Serum Using MRM3 Analysis

UGT Family of Enzymes: Quantification of Tryptic Peptides

Software tools

DiscoveryQuant™ Software: Signature Peptide MRM Optimization Made Easy for Therapeutic Protein and Peptide Quantification

MultiQuant™ Software 3.0: Peptide Bioanalysis for the Regulated Bioanalytical Laboratory

56-89

56-61

62-66

68-71

72-75

76-77

78-81

82-85

86-89

90-104

90-95

96-104


6 THERAPEUTIC PEPTIDE BIOANALYSIS www.sciex.comRUO-MKT-02-1486-A

Introduction

The importance of biotherapeutics as a class of drugs has increased significantly over recent years due to their enormous potential to treat a wide array of human diseases ranging from autoimmune and inflammatory diseases to cancer, cardiovascular diseases, and rare genetic disorders. These highly promising therapeutic agents, including very small peptide chains, such as insulin, up to much larger proteins, such as antibodies and novel Fc-like fusion proteins, are extremely attractive as drug candidates because of their low toxicity and high specificity, and these compounds continue to fill the pre-clinical and clinical pipelines of many pharmaceutical companies.

The rapid growth of biotherapeutics is a good indicator of its success, with the global market valued at around US$199.7 billion in 2013 and projected to grow by 13.5% through 2020. The number of clinically approved protein and peptide therapies has jumped to over 170 products with 350 antibody-based therapies currently awaiting clinical trials, making biotherapeutics the fastest growing class of drugs in the last decade. With increasing industry interest and investment and rising demand from the medical community for these unique, targeted therapies, there is a growing requirement to develop high-throughput analytical techniques to expand biotherapeutic product lines.

To overcome regulatory hurdles and advance to clinical trials, biopharmaceutical drug development and discovery requires metabolic monitoring of a candidate drug, a process which necessitates accurate quantitation during pharmacokinetic (PK), toxicokinetic (TK), bioequivalence, and clinical drug monitoring studies—all of which are conducted in a complex biological matrices (blood, plasma, or urine). With this rapid growth in biotherapies comes increased demand for an analytical platform that is flexible, robust, and is easily integrated into pre-existing drug development workflows. Widely used for small molecule drug development, liquid chromatography-tandem mass spectrometry (LC/MS/MS) has recently made a larger impact on bioanalysis applications due to recent technological developments in analyte detection. Presented here, we demonstrate how key mass spectrometry technologies from SCIEX can coalesce into straightforward, accurate, extremely sensitive, and, most importantly, high throughput quantitative solutions. Already considered as the preferred choice for quantitation in other areas of bioanalytical quantitation such as proteomics, anti-doping, forensics, and clinical chemistry, LC/MS/MS is poised to replace and outperform other techniques for biotherapeutic analysis.

The current standard conventions for protein and peptide quantitation are based on ligand-binding assays (LBAs), such as the enzyme-linked immunosorbent assay (ELISA), or on UV

Advances in the Bioanalysis of Protein and Peptide Therapeutics through Innovations in Mass SpectrometryOverview of peptide and protein quantitation applications on the SCIEX QTRAP® System and the TripleTOF® System

Laura Baker1, Suma Ramagiri2

1Contract Technical Writer at SCIEX, Pittsburgh, PA, 2SCIEX, Concord, Canada

Eksigent MicroLC and UHPLC System QTRAP® LC/MS/MS System TripleTOF® LC/MS/MS System

MultiQuant™ Software

DiscoveryQuant™ Software

ULT

IMA

TE S

ENSI

TIV

ITY

www.sciex.com THERAPEUTIC PEPTIDE BIOANALYSIS 7

detection of individual peptides using high pressure liquid chromatography (HPLC) separations. LBAs rely on immunoaffinity detection of a unique epitope on the protein or peptide of interest, and the high specificity of the antibody-based interactions can track an analyte at high sensitivity, although the dynamic range is narrowed to just one or two orders of magnitude. Because production of unique antibodies is lengthy, assay development can often be time-consuming and expensive; in addition, LBA results are often plagued by interferences and high background from antibody cross-reactivity. UV detection and quantitation of peptides is commonly used for peptide mapping, and this analytical method can be useful after extensive sample preparation and cleanup. UV detection with HPLC also does not require the expense and time commitment of antibody production, but the applicability of this method narrows as the complexity of the sample matrix increases.

Herein, we present an extensive resource on the quantitation of peptides using SCIEX mass spectrometry instruments, revealing how sensitive and selective detection can be achieved even in the presence of high background noise. To meet bioanalytical quantitative standards and assay validation parameters, peptide bioanalysis must be sensitive enough to meet the standard benchmarks for excellency in accuracy and precision. In Section 1, we explore the highest level of sensitivity by evaluating experiments conducted on the SCIEX QTRAP® 6500 System and the SCIEX Triple Quad™ 6500 System. Due to the inherent sample complexities, bioanalysis is often negatively impacted by high background noise and interfering peaks. Section 2 illustrates how realizing superb analyte selectivity—even in biological samples with numerous, highly abundant, endogenous proteins—is driven by innovative tools such as multiple reaction monitoring cubed (MRM3) workflow and SelexION™ Differential Ion Separation Technology. Advances in high resolution mass spectrometry are detailed in Section 3, which highlights targeted workflows on the TripleTOF® 5600+ System that extend the sensitivity and selectivity

of quantitative assays due to the narrow extraction widths and high resolution TOF data. Lastly, Section 4 investigates the software tools available for robust peptide quantitation workflows that give researchers intuitive tools to automate the complex, multi-step calculations for peak area quantitation.

Each section and experiment featured in this resource includes an overview of the key challenges, benefits, and features of the bioanalytical technique presented. In this way, the technique of mass spectrometry can be put into context with other bioanalytical tools and help provide insight into its many advantages. LC/MS/MS analysis offers many attractive features for supporting biopharmaceutical drug development; however, establishing LC/MS/MS in the biopharmaceutical workflow has been slow in spite of its dominance in the small-molecule laboratory. Widely accepted and easily validated, the LBA technique remains a popular method for protein and peptide bioanalysis due to its relatively lower investment in infrastructure and ease of implementation into the high-throughput environment. Yet, even LBA methods have their drawbacks, and straightforward LC/MS/MS alternatives are sought that can support the operational challenges of accelerating the further development of biotherapies.

Key challenges of peptide bioanalysis

To understand why the pharmaceutical industry has been hesitant to fully embrace LC/MS/MS strategies for peptide quantitation, the complexities and challenges of the workflow must be fully appreciated. (For a summary of excellent reviews on LC/MS/MS protein and peptide quantitation, please see Table 1.) For both proteins and peptide quantitation, standard calibration curves are used to calculate concentration values for unknowns in biological samples; in addition the amassed data must be stringent enough to meet the rigorous benchmarks prescribed by the USFDA. For therapeutic peptides, proteolysis is omitted, and the intact peptide can be directly quantitated by MS/MS after relatively limited sample preparation (Figure 1). There is appreciably much more complexity when evaluating larger molecular weight biotherapeutics (>10 kDa), which are not always suitable in their entirety for direct MS/MS analysis. Therefore, bioanalysis of larger proteins and antibodies is based on quantitation of a small portion of the protein, typically a tryptically digested signature peptide with a m/z ratio that is unique from all other peptides in the digest mixture. When coupled with a stable isotope-labeled (SIL) internal standard, the response ratio of the signature peptide to the SIL internal standard reveals a concentration representative of the intact protein. To build this multifaceted process into the framework of regulated bioanalysis is extremely challenging in practice, which makes it easy to comprehend why LC/MS/MS quantitation of biopharmaceuticals has been slow to gain acceptance in the GLP laboratory.

Figure 1. Peptide and protein bioanalytical workflow strategies. Protein quantitation typically involves a tryptic digestion step, which is omitted during intact peptide bioanalysis, thereby simplifying the process.

For Research Use Only. Not for use in diagnostic procedures.U

LTIMA

TE SENSITIV

ITY


Evaluation of LC/MS/MS bioanalysis reveals that the major challenges for accurate and precise quantitation lie primarily in the realm of sample preparation, which includes: 1) the lengthy and extensive workflows for producing signature peptides and 2) the diminishing accuracy of quantitative measurements in highly complex biological samples due to background interferences. Because the multi-step reduction/alkylation/digestion process generates a more complex mixture than the starting sample, bioanalysis of low-level therapeutic

peptides can be extremely challenging. Achieving LLOQs in the low ng/mL range is highly dependent at this time on the optimization of sample preparation steps.9 The numerous competing background peptides are a major consideration in sample preparation, which typically requires enrichment and semi-purification of the analyte that introduce additional complexity to the workflow. Of concern from a regulatory perspective is the potential for variable peptide release during digestion of the target protein, and if digestion conditions are not well-controlled

Title Article Highlights Citation

“Analysis of biopharmaceutical proteins in biological matrices by LC/MS/MS I. Sample preparation”

• Sample-preparation aspects for quantifying biopharmaceutical proteins in body-derived fluids by LC/MS/MS

• Enrichment at the peptide level after proteolytic digestion

• Chemical derivatization of peptides for enhancing ionization efficiency

• Automation of the entire analytical procedure for routine applications in pharmacokinetic and clinical studies

Bischoff R, Bronsema KJ, van de Merbel, NC. Trends in Analytical Chemistry. 2013; 48: 41-51.

“Analysis of biopharmaceutical proteins in biological matrices by LC/MS/MS II. LC/MS/MS analysis”

• Overview of selected reaction monitoring (SRM) strategies for quantifying peptides in biological matrices

• Selection of signature peptides and internal standards

• Selectivity improvements using MS3 and differential mobility spectrometry (DMS)

• Quantitative LC/MS analysis with low-resolution and high-resolution MS

• Data-independent acquisition (DIA) for collection of all data in a single analysis

Hopfgartner G, Lesur A, Varesio E. Trends in Analytical Chemistry. 2013; 48: 52-60.

“Bioanalytical LC/MS/MS of protein-based biopharmaceuticals”

• Overview of topics relating to the bioanalysis of biopharmaceutical proteins in biological matrices

• Compares alternative quantitative methodology, such as ligand binding assays (LBAs), to mass-spectrometry-based platforms

• Review of practical aspects of the seven “critical factors” for protein sample preparation

• Special focus on the quantitation of monoclonal antibodies in serum and plasma

• Advances in selectivity, including high-resolution mass spectrometry

van den Broek I, Niessen WMA, van Dongen WD, Journal of Chromatography B. 2013; 929: 161-179.

“Liquid chromatography coupled with tandem mass spectrometry for the bioanalysis of proteins in drug development: Practical considerations in assay development and validation”

• Approaches for overcoming operational challenges due to complex sample preparation

• Development and validation of a fast, simple, and reliable LC/MS/MS peptide quantitation method that fits into current pharmaceutical workflows

• Recommendations for validating quantitative methods based on surrogate peptides

Liu G, Ji QC, Dodge R, Sun H, Shuster D, Zhao Q, Arnold M. Journal of Chromatography A. 2013; 1284:155-162.

*These review articles were reprinted with permission in the first 30 copies of this resource.

Table 1. Selected citations for further reading on protein and peptide LC/MS/MS methodologies

Figure 2. Comparison of mass spectrometric platforms for peptide quantitation.

TripleTOF® Workflows

• Versatile workflow for simultaneous qual and quant in drug discovery and development

• High resolution accurate mass workflows – SWATH™ Acquisition

• Characterization and comparability of biosimilar in research and development

• Bioanalysis and biotransformation workflows for PK/PD studies

• Retrospective data analysis and robust performance

• When sensitivity is utmost importance – low bioavailability and high clearance

• IonDrive™ Technology for low LOQs and wider linear dynamic range

• Absolute quantitation of protein/peptide therapeutics

• GLP/non GLP bioanalysis in phase I and above

• High throughput and robust performance

TripleTOF vs QTRAP for Protein/Peptide Bioanalysis

QTRAP® Workflows

ULT

IMA

TE S

ENSI

TIV

ITY


or compensated for, then irregular signature peptide release can have a lasting impact on the overall data quality.7 To overcome these drawbacks, strategies such as condensing sample prep steps and digestion optimization can lead to more straightforward method development with wider regulatory appeal. And to that end, as advances in technology deliver exceedingly more sensitive and selective mass spectrometry workflows for direct quantitation in the sub-picomolar range, sample prep protocols can be further streamlined and simplified, relying less on intricate sample enrichment and baseline reduction protocols, which will help propel this versatile and reliable MS methodology firmly into the domain of regulated biotherapeutic quantitation.

Summarized below are some of the current challenges of LC/MS/MS peptide quantitation:

• Limited quantitation range – Poor MS/MS sensitivity combined with

often poor selectivity can compromise the desired lower limits of

quantitation (LLOQ).• Impaired sensitivity in complex matrices – Very low-level peptide

detection (sub-pg/mL) can be suppressed by high background and

competing ions in biological samples. The best, previously reported

LOQ is 100 pg/mL, which is insufficient for extended-release

pharmacokinetic studies.• Low specificity – Complex biological matrices hamper data resolution

and require sophisticated sample preparation and/or advanced

instrumentation.• Co-eluting, multiply charged interferences limit accurate quantification

and also peak integration at LOQ levels.• Isobaric interferences will limit selectivity and specificity of the assay

and cause issues for accurate identification during bioanalytical method

development process.• Reduced recovery, low sensitivity – The adsorptive properties and/or

polarity of peptides can compromise recovery, and interferences from

biological matrices can negatively impact sensitivity and selectivity. • Challenging physicochemical proprieties of peptides such as non-

specific binding, poor solubility, and complex charge state envelope

result can be problematic for the design of quantitation protocols.• Limited MRM selectivity – MRM approaches and efficient UHPLC

separations may not provide adequate signal-to-noise ratios at LLOQ

due to isobaric interferences or high baseline noise.• Systematic measurement errors – Especially for ultra-low-level

quantitation, errors in measurement have a significant effect on data

accuracy and precision. • Poorly fragmenting peptides – Cyclic fragments often fragment poorly

resulting in few product ions for MRM analysis.

Key benefits of the mass spectrometry based peptide quantitation workflow

While LBAs may be primarily used in industry at this time, LC/MS/MS techniques provide many potential benefits that are grounded in the direct evaluation of the analyte’s chemical nature, rather than indirect signals stemming from an immunological interaction. Quantitative data obtained by LC/MS/MS methodology correlates well with LBA-derived concentrations.7 Unlike LBA assays that require specific antibodies for each analyte, mass spectrometry platforms have universal applicability, providing one technique for a large diversity of analytes. All types of proteins and peptides can be evaluated by LC/MS/MS without exception, and a wide diversity of other biomolecules such as lipids and carbohydrates can also be identified, providing researchers with a flexible platform for identifying non-protein impurities. LBAs are generally more limited in their applicability because of auto-antibody cross-reactivity and the lack of commercial kits for every protein of interest.11 Non-specific

Multiple Orthogonal Selectivity Tools for Protein/Peptide Bioanalysis Additional Selectivity

SelexION™ Scheduled MRM™MRM3

HPLC SeparationSample Extraction

QTRAP® 6500 System

Figure 3. Mass Spectrometry based additional orthogonal sample clean up tools such as SelexION (differential mobility spectrometry), MRM3 scan function on QTRAP LC/MS/MS System and Scheduled MRM for increase in duty cycle


LTIMA

TE SENSITIV

ITY


binding and molecular class limitations are surpassed with LC/MS/MS, which can even quantify highly homologous isoforms that are impossible to distinguish using immunoaffinity techniques. Low-level biomolecule quantitation is analogous to finding a needle in a haystack; yet LC/MS/MS is able to deliver quantitative data with excellent accuracy and precision over a wide linear dynamic range, often over 3 to 4 orders of magnitude.7 Additionally, in contrast to the repeated expense and time-consuming nature of antibody production, LC/MS/MS methods can be developed and validated within a relatively shorter amount of time for multiple targets all at once. All of these characteristics taken together, including its flexibility, good data quality, and excellent selectivity, make LC/MS/MS an attractive method for biopharmaceutical quantitation in the regulated laboratory.

Key features of SCIEX instruments for MS/MS peptide quantitation

Ongoing optimization of sample preparation steps will continue to enhance the LC/MS/MS quantitation process, but the most significant gains in protein and peptide quantitation will be realized through technological innovations in mass spectrometry instrumentation. Focused on improving sensitivity and selectivity for the detection of very low levels of proteins and peptides in very complex backgrounds, SCIEX delivers high performance instruments that can rapidly and simultaneously measure multiple analytes—powering pharmaceutical discovery and development into the future (Figure 3).

1. Sensitivity. Biopharmaceuticals are very potent, highly targeted therapies that are administered in low concentration doses and exhibit a narrow therapeutic range. Often found at circulating levels in the sub-ng/mL range, detection of biotherapies requires very highly sensitive methods, and the enhancement of ionization efficiency and ion transmission have made it possible to detect drugs and metabolites in the sub-femtogram levels. New technologies such as the IonDrive™ QJet Ion Guide underpin the sensitivity enhancements in the QTRAP 6500 System and the SCIEX Triple Quad™ 6500 System, bringing more ions to the detector through improved collisional focusing of ions. Heating and desolvation improvements in the IonDrive™ Turbo V Source and increased size and improved design of the aperture release more ions into the instrument. To fully detect the augmented signal, improvements to the dynamic range of the detector allow for accurate ion counting; the high energy conversion dynode (HED) detection system measures high ion signals without saturation to produce a linear dynamic range of over 6 orders of magnitude. These technologies are pivotal for providing continued improvements to sensitive bioanalysis.

2. Selectivity. Even if the pinnacle of sensitivity is reached, researchers will still be faced with the challenges of separating low levels of pharmaceutically active biomolecules from the highly complex biological matrix, where every endogenous compound can potentially interfere with the target signal. On the sample prep side, several strategies exist for the selective removal of competing background ions as well as enrichment of the analyte fraction. However, the required time and the potential for sample loss with additional cleanup steps makes this approach much less appealing. Currently, advances in MS selectivity are focused on methods that provide an additional degree of separation subsequent to the entrance to the MS or post MS/MS selection to help improve separation capacity in highly complex biological matrices. To maximize instrument performance when detecting low-level analytes masked by high background, SCIEX offers MRM3 scans and the SelexION™ Differential Mobility Separation Device for improved peak shapes and signal-to-noise ratios during protein and peptide quantitation. MRM3. Peak measurements obtained by multiple reaction monitoring (MRM) scans are occasionally challenged by interferences that cannot be removed without further, more elaborate sample clean-up. To provide additional specificity, the technique of MRM3 can be applied using the QTRAP Series of instruments—extremely sensitive, hybrid triple quadrupole instruments with a linear ion trap for further fragmentation of the primary product ions. Quantitation of the secondary product ions is usually not affected by competing or overlapping ions, which are filtered out in previous MRM selection steps. This reduction in baseline results in improved peak shape, higher signal-to-noise ratios, and superior LLOQs. The QTRAP® 5500 System and 6500 System are powered by eQ™ Electronics for scan speeds that are fast enough to be compatible with fast LC flow rates; and these instruments are equipped with single frequency excitation for highest selectivity of the product ion prior to secondary fragmentation. The Linear Accelerator™ Trap Electrodes provide 100-fold more sensitivity for the detection of low-level secondary fragments resulting from the use of MRM3 to resolve issues of high background noise.

ULT

IMA

TE S

ENSI

TIV

ITY


3. Differential mobility spectrometry (DMS). In some cases, if secondary product ions are not specific enough or are too low for MRM3 to be used, or method development time is too limited for prolonged MRM3 development, then additional selectivity can be gained through differential mobility spectrometry (DMS). This technique selects ions of interest based on their inherent mobility difference between a set of planar plates with high and low energy fields applied, where co-eluting interferences can be tuned out prior to analyte entrance into the mass spectrometer. SCIEX offers the SelexION™ Differential Mobility Separation Device for quickly resolving isobaric species and single and multiple charge state interferences on a timescale compatible with UHPLC and multiple MRM acquisitions, thus providing an additional, orthogonal level of separation for difficult-to-address overlapping peaks.

4. High resolution accurate mass spectrometry. Improvements to selectivity can also be gained through high resolution mass spectrometry on instruments such as the SCIEX TripleTOF® 5600+ System, which combines qualitative exploration and high resolution on a single platform. When using an MRMHR workflow, the TOF analyzer detects all the fragments from the precursor at high resolution and high mass accuracy. Using narrower extraction widths than the unit resolution of triple quadrupole-based experiments, difficult separations between background peaks and analytes can now be achieved and improved to such an extent that minimal interferences are observed. When fragment ions are extracted at these narrow extraction widths, analytes can be detected at higher specificity and at accurate mass in complex matrices.

5. Software. Evaluating the results of protein and peptide quantitation can often be time-consuming and repetitive, relying on manual peak identification and data integration—a process that does not lend itself well to the high-throughput environment. SCIEX has developed comprehensive, powerful, and easy-to-use solutions such as MultiQuant™ Software and DiscoveryQuant™ Software that simultaneously process multiple analytes. Not only do these software packages rapidly process MS scans and data, but they also support improved data integrity and security, combining unique audit trail functionality for improved regulatory compliance and an embedded digital link to the Watson LIMS system for increased confidence in data safety.

Advantages of the diversity of mass spectrometry systems

In this resource, we primarily focus on experiments conducted on two hybrid triple quadrupole instruments, the TripleTOF 5600+ System versus the QTRAP 6500 System. Each platform has distinct advantages (Figure 2): The TripleTOF is uniquely suited to qualitative discovery (as well as quantitation) due to the underlying acquisition of a full spectrum of secondary fragments at high resolution, while the QTRAP System and its augmented ion generation, transmission and detection works best for applications requiring high sensitivity and expanded linear ranges. The SCIEX QTRAP 6500 System is fully accepted for regulated bioanalysis at the Phase 1 level and above, but the TripleTOF System dominates in ease of method development and non-targeted analysis during drug discovery protocols. In the event that one application demands the benefits and strengths of an alternative MS platform, transferring methods is easy and intuitive; the two MS systems have identical source and collision cell designs based on the innovative LINAC® Collision Cell, which allows for seamless coordination of quantitative data with qualitative analysis (Figure 4).

Perspectives for the future

As technological innovations surpass the limitations imposed by biological sample complexity, LC/MS/MS biopharmaceutical quantitation will become more fully established as a routine methodology in the regulated laboratory. Time-consuming and complicated sample preparation steps will evolve to become better suited to the automated requirements of the MS-based bioanalytical workflow, and sample extraction procedures are likely to become more highly selective to achieve the sensitivities required for monitoring sub-picomolar concentrations of biotherapeutic agents. Working with highly sensitive methods based on the enhanced MS ionization efficiency and transmission has yielded promising results on the QTRAP System, producing sufficient LLOQs for low-level biomolecule quantitation needed for PK and TK studies. Additionally, distinct gains using DMS and MRM3 are adding an additional layer of selectivity, removing hard-to-separate background and leading to better signal-to-noise parameters. The potential of high resolution mass spectrometry to measure intact, high molecular weight biomolecules will gain increasing interest as technological advances push TOF sensitivities towards those of the hybrid linear ion trap instruments. By reducing the need for additional sample preparation steps with enhanced MS detection and selectivity capacities, LC/MS/MS techniques are becoming more closely aligned with the high-throughput workflows necessary for regulated bioanalysis.


LTIMA

TE SENSITIV

ITY


References1“Biopharmaceuticals – A Global Market Overview.” November 2013. The abstract

from a market research report from Reportbuyer.com. Accessed online at: http://www.prweb.com/releases/2013/11/prweb11314067.htm

2 Zhong X, Neumann P, Corbo M, Loh E. “Recent Advances in Biotherapeutics Drug Discovery and Development.” Drug Discovery and Development – Present and Future, Dr. Izet Kapetanovic (Ed.) ISBN: 978-953-307-615-7, InTech. Accessed online at: http://www.intechopen.com/books/drug-discovery-and-development-present-and-future/recent-advances-in-biotherapeutics-drug-discovery-and-development

3 Reichert J. “Which are the antibodies to watch in 2013?” mAbs. Jan/Feb 2013; 5(1): 1-4.

4 Shi T, Su D, Liu T, Tang K, Camp DG 2nd, Qian WJ, Smith RD. “Advancing the sensitivity of selected reaction monitoring-based targeted quantitative proteomics.” Proteomics. Apr. 2012; 12(8): 1074-92.

5 Hopfgartner G and Gougongne E. “Quantitative high-throughput analysis of drugs in biological matrices by mass spectrometry.” Mass Spectrom Rev. May/Jun 2003; 22(3): 195-214.

6 Ezan E and Bisch F. “Critical comparison of MS and immunoassays for the bioanalysis of therapeutic antibodies.” Bioanalysis. Nov. 2009; 1(8): 1375-1388.

7 van den Broek I, Niessen WMA, van Dongen WD. “Bioanalytical LC/MS/MS of protein-based biopharmaceuticals.” Journal of Chromatography B. 2013; 929: 161-179.

8 Food and Drug Administration. “Guidance for Industry; Bioanalytical Method Validation.” US Department of Health and Human Services. FDA. Center for Drug Evaluation and Research, Rockville, MD, 2001.

9 Bischoff R, Bronsema K, van de Merbel NC. “Analysis of biopharmaceutical proteins in biological matrices by LC/MS/MS 1. Sample preparation.” Trends in Analytical Chemistry. 2013; 48:41-51.

10 Liu G, Ji QC, Sun H, Shuster D, Zhao Q, Arnold M. “Liquid chromatography coupled with tandem mass spectrometry for the bioanalysis of proteins in drug development: Practical considerations in assay development and validation.” Journal of Chromatography A. 2013;1284: 155-162.

11 Hopfgartner G, Lesur A, Varesio E. “Analysis of biopharmaceutical proteins in biological matrices by LC/MS/MS II. LC/MS/MS analysis.” Trends in Analytical Chemistry. 2013; 48:52-61.

12 Thomson B. “Driving high sensitivity in biomolecular MS.” Genetic Engineering and Biotechnology News. Nov 2012; 32(20). Accessed at: http://www.genengnews.com/gen-articles/driving-high-sensitivity-in-biomolecular-ms/4603/?kwrd=high%20sensitivity%20in%20biomolecular%20MS.

From Biologics Characterization to Biotransformation and Bioanalysis

SWATH™ Acquisition Multiple Reaction Monitoring (MRM)

High Res XICs

TripleTOF® System QTRAP® System

MRM

Discovery Development GLP Bioanalysis

m/z

coun

ts

retention time

coun

ts

Easy Method Transfer with LINAC® Collision Cell

Figure 4. Continuity of workflows between TripleTOF to QTRAP. From product characterization during research and development process to biotransformation and bioanalysis during PK/PD analysis in preclinical and clinical studies

ULT

IMA

TE S

ENSI

TIV

ITY



LTIMA

TE SENSITIV

ITY


In applications that range from proteomics to biomarker discovery to drug development, mass spectrometry has become the tool that provides the high accuracy and specificity in trace chemical analysis. While there are many important metrics of analytical performance (accuracy, precision, limit of quantitation), they all rely heavily on two key instrumental performance characteristics – sensitivity and dynamic range. In mass spectrometry, instrument sensitivity can best be defined as the number of ions detected per molecule of analyte injected, thus accounting for all losses in ionization, transmission, and detection. Dynamic range is usually defined as the range of linear response of the instrument, limited at the low end by absolute sensitivity and, at the high end, by detector or other instrument-related saturation effects.

Over the last thirty years or more of development at SCIEX, enormous strides have been made in improving both the instrument sensitivity and the dynamic range. This improved performance has enabled new applications to be addressed by mass spectrometry, and allowed analyses to be performed more rapidly and with greater confidence and higher precision. Higher sensitivity has also enabled the use of additional capabilities and techniques that provide improved analytical specificity – such as higher mass resolution, faster scans speeds, and shorter reaction monitoring times, and techniques such as ion mobility/mass spectrometry combinations, or added levels of tandem mass spectrometry (MS/MS/MS). The growth curve of sensitivity in SCIEX triple quadrupole mass spectrometers over this time period is plotted in Figure 1, which shows a growth of nearly six orders of magnitude in absolute sensitivity since our first LC/MS/MS product, the TAGA 6000. The SCIEX QTRAP® 6500 System, our newest and highest-performance instrument, reaches new levels in both sensitivity and dynamic range. New technologies in both the ion optics and ion counting detection system have driven these performance increases.

A key step in achieving higher sensitivity is to create more ions in the source. Over the years, improvements in ionization efficiency have been achieved by increasing the efficiency of desolvation and declustering in the source. The new IonDrive™ Turbo V Source of the QTRAP 6500 System has reached a new level.

By optimizing the design of the IonDrive Turbo V Source heaters for better and more uniform distribution of heat in the region of droplet evaporation, the efficiency of creating ions in front of the sampling orifice has been improved, especially at higher liquid flow rates and with less volatile compounds.

However the sampling aperture into the vacuum still typically represents the largest area of ion losses. We have, therefore, increased the size of the orifice in order to sample more ions. Improved pumping in the interface helps maintain an acceptable core vacuum pressure without increasing the size of the turbo pumps. The gas expanding through the orifice forms a supersonic free jet with a characteristic barrel shock structure as shown in Figure 2. The high gas flow and pressure provide a strong drag force on the ions that are entrained in this jet, making it more challenging to effectively focus the ions through the next aperture. The new IonDrive™ QJet Ion Guide optics employs a two-stage RF quadrupole to capture and focus the ions to the center-line of the optics using the technique of collisional focusing, allowing the majority of gas to be pumped away. The first section is a large-diameter, RF-only quadrupole with narrow gaps between the rods in order to contain the ions. The narrow gaps minimize the radial outflow of gas, and, therefore, ion losses, while allowing the entrained ions to become collisionally

Technology Drives High Performance in Biomolecular Mass SpectrometryEnhancing the sensitivity and dynamic range of the

SCIEX QTRAP® 6500 with IonDrive™ System Technology

Bruce Thomson and Bruce Collings

SCIEX, Concord, ON, Canada

Figure 1: The growth in sensitivity of high-flow LC/MS/MS mass spectrometer systems over the last thirty years at SCIEX.

Driving Sensitivity by Orders of Magnitude

Year

Rela

tive

Sens

itivi

ty

1E+6

1E+5

1E+4

1E+3

1E+2

1E+1

1E+0

1980 1990 2000 2010 2020

ULT

IMA

TE S

ENSI

TIV

ITY

Bruce Collings Bruce Thomson


focused. The second section is a smaller diameter quadrupole that provides the final stage of ion-beam focusing while the gas escapes. The transmission efficiency of the ions into the next chamber is approximately 50%, an impressive figure of merit considering the larger orifice diameter, and higher pressure and higher gas velocity.

The increased rate of ion generation in the source, and improved transmission efficiency in the ion optics results in a higher ion flux reaching the detector for a given amount of sample injected. At the detector, ions are detected and registered with very high efficiency using a pulse counting system with a very low noise level. The challenge with pulse counting has always been to be able to measure high ion signals without saturation. The new high energy conversion dynode (HED) detection system powered by IonDrive™ Technology provides a very significant improvement in this area, extending the upper level of ion counting while maintaining the ability to register single ion events for the best signal-to-noise ratios at the detection limit. The improved dynamic range can be seen in Figure 3, which compares the new HED detection system to the standard CEM detection system.

In Figure 3, the measured count rate of the first isotope of reserpine is plotted against the true count rate as determined from the known ratio and intensity of its fourth isotope. The new system uses high-energy ion-to-electron conversion and a low impedance multi-channel continuous dynode detector with a closely coupled transimpedance amplifier system that allows high count rates to be sustained without loss of signal. Arrival rates of up to 200 million ions per second can be achieved resulting in a detector linear dynamic range of more than six orders of magnitude. With the sensitivity and dynamic range improvement described above, the QTRAP® 6500 System provides a new level of analytical performance, as evidenced by the ability to detect and quantify sub-femtogram amounts of biomolecules injected

on-column as shown in Figure 4. Demands for ever decreasing detection limits will continue to drive the need for newer and better methods of ionization, transmission, and detection in the future. However, the growth curve of sensitivity will become more and more difficult to maintain as we approach the limit of measuring and detecting nearly every ion injected.

Figure 2: Cross-sectional view of the IonDrive™ QJet showing the supersonic free jet (supersonic flow region in red) and the gas flow along the axis.

ULTIM

ATE SEN

SITIVITY

Mea

sure

d C

ount

Rat

e (c

ps)

True Count Rate (cps)

Figure 3: Dynamic range of the high energy conversion dynode (HED) detection system compared to the standard CEM detection system.

Figure 4: Signal from 500 attograms of verapamil injected on-column monitored in MRM mode using the transition 455/165.



Key challenges of nanoflow quantitation

Reduced throughput – Nanoflow platforms lack the robustness and high-throughput required for multi-sample drug metabolism studies.

Inflexible and complicated interface assembly – Installing and troubleshooting nanoflow fittings is time-consuming, making variable-rate method development cumbersome.

Limited options for sensitive peptide quantitation – Easier and faster peptide quantitation methods that meet nanoflow standards for sensitivity are needed for pharmaceutical applications.

Key benefits of microflow peptide analysis on the QTRAP®

6500 System

Ultra-sensitive peptide quantitation – LLOQs obtained on the robust microflow platform meet or exceed nanoflow standards by 2-fold – even in complex matrices.

Accelerated throughput – Easy-to-use microflow workflow provides >2-fold faster run times, suitable for high-throughput analysis.

Easy hardware assembly – Microflow components take only a few minutes to interchange – realizing a single, adaptable LC platform for peptide quantitation.

Method portability – No loss of sensitivity is observed when transferring nanoflow regimens to a microflow platform on the QTRAP® 6500 System.

Key features of the microflow workflow on QTRAP® 6500 System

IonDrive™ Technology – Increased detector dynamic range and signal-to-noise improvements are due to ionization efficiency, ion sampling, and ion transmission enhancements.

Mass range of m/z 5–2,000 – Comprehensive mass range provides the versatility needed for peptide quantitation.

Flexible and reproducible chromatography – The nanoLC 425 System supports a wide range of rates – nano to microflow – providing unparalleled method flexibility.

Introduction

To obtain the best sensitivities and ionization efficiencies for peptide analysis, nanoflow chromatography is often used in combination with hybrid triple quadrupole/linear ion trap mass spectrometry to provide an established, highly-sensitive quantitation method. However, sample processing times are typically slowed by the extended chromatography run times obtained at sluggish, nL/min flow rates. Shifting to quicker microflow rates (3–50 µL/min) has improved sample run time, but the higher flow rates cause dampening of the ionization efficiency compared to nanoflow, diminishing sensitivity and driving up LLOQs.

To find a good balance between sensitivity, robustness, and throughput, we compared the LLOQs of various tryptic peptides using nano- or microflow rates – including bradykinin, a nine amino-acid peptide involved in vasodilation. Recent evaluation of peptides on the QTRAP® 5500 System indicated a 3-fold loss in sensitivity when transitioning from nanoflow to microflow chromatography (4 µL/min).1 However, other peptide quantitation experiments have demonstrated a roughly 3–5-fold improvement in sensitivity when moving to the QTRAP® 6500 System from the

Achieving Low-Flow Sensitivities for Peptide Quantitation using Microflow Rates on the QTRAP® 6500 System High-throughput, sensitive, microflow analysis of bradykinin and other peptide standards with a

hybrid triple quadrupole linear ion trap coupled with the Eksigent ekspert™ nanoLC 425 System

Kelli Jonakin and Christie Hunter

SCIEX, Redwood City, CA, USA

ULT

IMA

TE S

ENSI

TIV

ITY


5500. From these two studies, we hypothesized that nanoflow peptide assays currently performed on the QTRAP® 5500 System could be upgraded to microflow rates on the QTRAP® 6500 System, resulting in similar sensitivities, but with improved robustness and increased throughput.

To realize the most efficient strategy for large-scale peptide analysis, this study explores the LLOQs and the speed of analysis for a range of peptide standards, including bradykinin, using two, hybrid triple quadrupole/linear ion trap systems operating at different flow rates. Variable chromatography was executed using an ekspert™ nanoLC 425 system, which has the flexibility to support micro and nanoflow rates in a single system. Coupled with recent advances in IonDrive™ Technology for higher-sensitivity detection, microflow chromatography provides a step forward in productivity and ease-of-use, meeting or exceeding sensitivity levels established for nanoflow peptide analysis.

Methods

Sample preparation

The beta-galactosidase digest mixture and the 6-peptide mixture containing the bradykinin 2–9 fragment (monoisotopic mass 904.9681) were obtained from SCIEX. The six protein digest containing carbonic anhydrase and five other proteins was obtained from Michrom BioResources. Protein-precipitated (crashed) plasma matrix was prepared by mixing equal volumes of plasma and acetonitrile, followed by centrifugation.

Chromatography

LC system: Eksigent ekspert™ nanoLC 425 System with 0.1–1 μL/min or 1–10 μL/min flow module in combination with the Eksigent cHiPLC® System in trap and elute mode

Nanoflow settingsColumn: Eksigent ChromXP™ C18 cHiPLC, (75 μm × 15cm)

Injection: 2–5 μL

Flow rate: 300 nL/min

Microflow settingsColumn: Eksigent ChromXP™ C18 (300 μm × 15 cm)

Injection: 2–5 μL

Flow rate: 4 μL/min

Mass spectrometry

Nanoflow settingsSystem: QTRAP® 5500 System

Interface: NanoSpray® Ion Source

Microflow settingsSystem: QTRAP® 6500 System

Interface: IonDrive™ Turbo V Source with 25 μm ID electrodes (Figure 2)

Data processing

MRM transitions were optimized for each peptide and used on both instruments. Standard concentration curves were performed to evaluate impact of flow rates and separation times on the two

Figure 2: IonDrive™ Turbo V Source. The optimized geometry and larger diameter heaters provide more efficient heat transfer to a larger cross section of the spray cone. Equipped with the lower inner diameter hybrid electrodes from Eksigent, this source provides a robust, easy-to-optimize solution for microflow chromatography.

ULTIM

ATE SEN

SITIVITY

Figure 1: Comparing signal intensities of standard protein digest. A beta-galactosidase digest was analyzed by nanoflow LC on a QTRAP® 5500 System (top) and compared to microflow LC on the QTRAP® 6500 System (bottom). Similar signal intensities were observed with similar separation quality but with >2-fold faster total run times with microflow LC.



different MS systems. All samples were analyzed in triplicate. Lower limits of quantification (LLOQ) were determined using MultiQuant™ Software.

Results and discussion

Microflow LC for peptide quantitation

Previous peptide studies that used mass spectrometry for quantitation demonstrated lowered ionization efficiencies during microflow analysis.2 To improve sensitivities for chromatographic runs conducted at 4–50 μL/min on the QTRAP® 6500 System, the IonDrive™ Turbo V Source was used to provide high-efficiency ionization and increased ruggedness (Figure 2). For best performance at microflow rates, the sources were plumbed with the hybrid electrodes specifically designed for microflow.2 These electrodes significantly reduced post-column dead volumes for minimized dispersion and sharper peak widths. In this work, we used the 25 µm ID electrode, ideal for 300 µm ID columns and 3–25 µL/min flow rates.

Comparing the sensitivity differences across LC/MS platforms

To better understand the sensitivity of peptide detection at variable flow rates, a series of experiments were performed to compare LLOQs of peptide standards obtained using a nanoflow platform on the QTRAP® 5500 System versus a microflow platform on the QTRAP® 6500 System. First, the magnitude of signal intensities of beta-galactosidase peptides (10 fmol on column) were compared, and analysis of the resulting spectra indicated that similar intensities were obtained for peptides analyzed by microflow versus nanoflow (Figure 1). Total run time was reduced 2-fold in the microflow experiment, while preserving good peak resolution. Preliminary data indicate that higher LC flow rates do not impede the detection and resolution of peptide peaks analyzed using the QTRAP® 6500 System, laying the groundwork for further microflow studies.

Tryptic peptides from the six protein digest were evaluated under microflow and nanoflow conditions in a simple matrix to assess

sensitivity. Peak areas were calculated for increasing protein concentrations and were assembled into concentration curves. LLOQs determined for peptides analyzed under microflow and nanoflow conditions were compared for individual peptides within the mixture (Table 1). When using the microflow configuration on the QTRAP® 6500 System, a 2-fold lower LLOQ was observed on average for each individual peptide with some variation across the group.

To establish if signal intensity improves under microflow conditions for a more complex mixture of peptides, the MRM signals for eight peptides from the six protein mixture were assessed (using an on-column concentration just above the LLOQ for the group of peptides). Signal intensities for peptides from the mixture under microflow conditions were elevated over those obtained under nanoflow conditions (Figure 3). Additionally, chromatographic run times, shortened by 25 min. under microflow conditions, allowed for more rapid peak elution while preserving good peak resolution (Figure 3). Focusing on peak intensity for one particular peptide (2y

5 from carbonic anhydrase) within the peptide mixture from the six peptide mixture reveals improving LLOQs (Figure 4) on the microflow platform. The LLOQ achieved under microflow conditions (1.9 amol on column) was ~2-fold more sensitive than that obtained by nanoflow (3.8 amol on column).

To evaluate peptide response in a more complex matrix system – protein-precipitated or crashed plasma – peak intensities and elution times for the bradykinin peptide in the SCIEX six-peptide mixture were assessed for both LC/MS/MS platforms. The introduction of competing ions and background noise from the plasma did not impact the intensity, elution times or resolution of microflow peaks over the nanoflow peaks. LLOQs from the microflow experiments (6.3 amol on column, Figure 5) were lower than those achieved under nanoflow conditions (12.5 amol on column) and were indicative of equivalent or slightly better sensitivity when using microflow on the QTRAP® 6500 System.

ULT

IMA

TE S

ENSI

TIV

ITY

Table 1: Lower limits of quantification (LLOQ) obtained for eight tryptic peptides on the two LC/MS systems Standard concentration curves in simple matrix were generated and the LLOQs were determined using both nanoflow on QTRAP® 5500 System and microflow on QTRAP® 6500 System. The results for the peptides show some variation observed across peptides but on average a 2-fold lower LLOQ was seen on the microflow QTRAP 6500 System.

PeptideFragment Ions

Summed for QuantQTRAP® 6500 System – 4 µL/min

LLOQ (amol on column)QTRAP® 5500 System – 300 nL/min

LLOQ (amol on column)

Sensitivity Improvement on Microflow QTRAP®

6500 System

IDALNENK 2y4, 2y6, 2y7 .48 1.9 4.0

TPEVDDEALEK 2y102+, 2y7, 2y8 3.8 3.8 1.0

VLVLDTDYK 2y5, 2y6, 2y7 1.9 3.8 2.0

AEFVEVTK 2y5, 2y6 3.8 3.8 1.0

ATEEQLK 2y5 7.6 3.8 0.5

DGPLTGTYR 2y5 1.9 3.8 2.0

VGDANPALQK 2y5, 2y6, 2y7 .95 3.8 4.0

VLDALDSIK 2y6, 2y7, 2y8 1.9 3.8 2.0

Average Difference: 2.1

Standard concentration curves in simple matrix were generated and the LLOQs were determined using both nanoflow on QTRAP® 5500 System and microflow on QTRAP® 6500 System. The results for

the peptides show some variation observed across peptides but on average a 2-fold lower LLOQ was seen on the microflow QTRAP® 6500 System.


ULTIM

ATE SEN

SITIVITY

Conclusions

An easy-to-use, high-throughput, and highly-sensitive workflow was established on the QTRAP® 6500 System using microflow chromatography for performing targeted peptide quantitation. The flexibility and reproducibility of the Eksigent ekspert nanoLC 425 makes it an ideal LC system for labs performing a broad range of proteomics workflows, including both nano and microflow rate applications. Assays currently performed using nanoflow LC on the QTRAP® 5500 System can be easily translated to the microflow QTRAP® 6500 System for accelerated sample analysis with similar sensitivities.

High-throughput capacities were realized with decreased peak retention times under microflow conditions – while preserving peak resolution and intensity.

Similar or lower LLOQs attained using microflow chromatography on the QTRAP® 6500 System met or exceeded nanoflow sensitivity standards for robust peptide quantitation.

Highly reproducible microflow chromatography on the Eksigent ekspert™ nanoLC 425 ensured accurate peptide quantitation.

References1 Exploring the Sensitivity Differences for Peptide Quantification in the Low Flow Rate

Regime – Eksigent ekspert™ nanoLC 400 System. SCIEX technical note 6560212_02.

Poster # TP08 – 151

2 Higher Sensitivity and Improved Resolution Microflow UHPLC with Small Diameter Turbo V™

Source Electrodes. SCIEX technical note 4590211-01

Figure 4: LLOQ comparisons for carbonic anhydrase peptide DGPLTGTYR from the six protein digest. The MRM signal at LLOQ for the carbonic anhydrase peptide with microflow (left) and nanoflow (right) chromatography. An approximate 2-fold improvement in sensitivity is obtained when using the microflow chromatography on the QTRAP® 6500 System.

Figure 5: Quantification of bradykinin using microflow LC on a QTRAP® 6500 System. MRM signals for bradykinin at LLOQ (top figure) were obtained using microflow conditions. The standard concentration curve (bottom figure) of bradykinin in protein-precipitated plasma provided an LLOQ of 6.3 amol on column. Linearity was very good across the limited dynamic range interrogated (>4 orders of magnitude in this example). The equivalent experiment using nanoflow on the QTRAP® 5500 resulted in an LLOQ of 12.5 amol on column for the same y7 fragment ion. Assay performance metrics are listed in the table for the bradykinin standard curve.

Analysis of peak areas for various concentrations of bradykinin 2y7 peptide on the QTRAP® 6500 System

Conc. (ng/mL) Mean (n=3) Std. Dev. %CV Accuracy (%)

6.3 5.411 0.8429 15.58 85.88

12.5 12.63 1.120 8.87 101.01

25 25.55 2.747 10.75 102.18

50 46.26 5.584 12.07 92.52

100 106.8 6.212 5.81 106.83

200 194.2 4.001 2.06 97.09

400 427.9 8.055 1.88 106.98

800 841.3 9.923 1.18 105.16

4,000 4,000.7 84.15 2.10 100.19

20,000 20,560 236.3 1.15 102.78

100,000 99,370 1641 1.65 99.37

Figure 3: Signal intensity comparison of eight tryptic peptides from the six protein mixture at concentrations near LLOQ. Peptide signals from the six-protein digest (7.6 amol on column) are shown for (top) the QTRAP® 5500 System under nanoflow conditions and (bottom) the QTRAP® 6500 System under microflow conditions. Microflow signal intensities showed a small improvement over the nanoflow data.



Key challenges of desmopressin quantitation

Impaired sensitivity in complex matrices – Very low-level peptide detection (sub-pg/mL) can be suppressed by high background and competing ions in biological samples.

Poor data quality – Precision and accuracy can be compromised at low peptide levels, giving results below accepted bioanalytical standards.

Key benefits of peptide quantitation on the Triple Quad™

6500 LC/MS/MS System

High sensitivity – Very low level peptide detection in human plasma (at sub pg/mL concentrations) is enabled by IonDrive™ Technology.

Excellent precision and accuracy at the LOQ level – Data quality (for LOQ, LQC, MQC and HQC levels) met or exceeded USFDA bioanalytical method validation criteria.

High throughput – High sensitivity was achieved under high-flow conditions (0.750 mL/min), optimal for multi-sample analysis.

Unique features of the Triple Quad™ 6500 System for low-level peptide detection

IonDrive™ Turbo V Source – Increased ionization efficiency and heat transfer contribute to sensitivity enhancements, including improved signal-to-noise.

IonDrive™ QJet Ion Guide – Increased ion sampling improves method efficiency and ruggedness.

IonDrive™ High Energy Detector – Innovative detector technology boosts dynamic range and sensitivity.

Introduction

Low-level peptide detection has a number of applications in clinical studies and in the pharmaceutical discovery and development processes, highlighting the increasing relevance of sensitive and selective mass spectrometric platforms in the

bioanalytical laboratory. Regulatory requirements demand intensive and rigorous quantitation of therapeutic peptides during pharmacokinetic, bioequivalence, and metabolic studies. In addition, drug discovery and development strategies seek

A Sub-picogram Quantification Method for Desmopressin in Plasma using the SCIEX Triple Quad™ 6500 SystemA high-throughput method for detecting ultra-low levels (0.5 pg/mL) of a therapeutic

peptide in human plasma using an SCIEX Triple Quad™ 6500 LC/MS/MS System

and UHPLC Chromatography

Rahul Baghla1, Swati Guttikar2, Dharmesh Patel2, Abhishek Gandhi2, Anoop Kumar1, and Manoj Pillai1

1SCIEX, 121, Udyog Vihar, Phase IV, Gurgaon, Haryana, India2 Veeda Clinical Research India, Ahmadabad, India

ULT

IMA

TE S

ENSI

TIV

ITY

Figure 2: Unique features of Triple Quad™ 6500 System.

Abhishek Ghandi


ULTIM

ATE SEN

SITIVITY

to monitor and quantitate peptide biomarkers in complex biological samples, necessitating highly-selective separations of low concentration analytes from high background noise and prominent levels of competing ions. The SCIEX Triple Quad™ 6500 LC/MS/MS System, equipped with IonDrive™ Technology for enhanced detector performance, has demonstrated particular strength in the detection of low-level amounts of small molecules, and in this study, we extend the augmented signal-to-noise, broad dynamic range, and the efficient method development capacities of the Triple Quad 6500 System to the detection of sub-picogram levels of a therapeutic peptide under high-throughput conditions.

We have developed a reliable, fast, and sensitive method for the detection of a nine-amino-acid peptide, desmopressin (1 desamino-8-D-arginine, vasopressin), which is structurally similar to the hormone arginine vasopressin, but contains a deaminated first amino acid and dextro-arginine (rather than levo-) in the eighth position. Therapeutically, desmopressin reduces urine production, restricting water elimination from the kidneys by binding to the V2 receptors in renal-collecting ducts, thereby facilitating increased reabsorption. The longer half-life of desmopressin over vasopressin offers some therapeutic advantages, and typical doses of desmopressin to treat diabetes insipidus and bedwetting range between 0.200 to 1.20 mg per day, resulting in very low plasma concentrations. In this bioanalytical study, we have established a sensitive and selective LC/MS/MS method for the quantitation of desmopressin in human plasma, detecting peptide levels as low as 0.500 pg/mL with excellent accuracy and precision. This technique should facilitate additional mass spectrometric method development for accurate quantitation of a range of therapeutic peptides in biological matrices on the Triple Quad 6500 System.

Materials and methods

Sample preparation

Plasma samples (1000 µL) containing 2% desmopressin standard and 50µL internal standard were vortexed and spiked with 50µL of orthophosphoric acid (OPA). Samples were extracted on weak

cation exchange cartridges conditioned with methanol followed by 100mM ammonium acetate solution. After loading, samples were washed in three steps: 1) 2% OPA:methanol (80:20 v/v); 2) 2% NaOH:Methanol (60:40 v/v); and 3) water:methanol (60:40 v/v). Analytes were eluted with 5% acetic acid in methanol, dried under nitrogen at 40 ºC, and reconstituted with 0.1% acetic acid (150 µL) prior to analysis by mass spectrometry.

Chromatography

LC system: GL Sciences LC 800 System

Column: Agilent 300 Extend C18 (150 x 2.1 mm, 3.5 μm)

Column temp.: 40 °C

Injection: 50 μL

Flow rate: 0.750 ml/min

Mobile phase: A) water, 0.1% acetic acid B) acetonitrile, 0.1% acetic acid

Gradient: Time/min A% B%

0 85 15

1.5 85 15

3.5 50 50

3.51 85 15

5 85 15

Mass spectrometry

Analysis of desmopressin and desmopressin-d5 required different mass spectrometry settings (Table 1). The MRM transition monitored for desmopressin was m/z 525.4/328.0 and 537.9/328.0 at a dwell time of 50 msec. Five replicate injections were performed at all concentrations.

Data system: Triple Quad™ 6500 System

Interface: IonDrive™ Turbo V Source in positive ion mode

Figure 3: Structure of desmopressin. Figure 4: Structure of internal standard, desmopressin-d5..



ULT

IMA

TE S

ENSI

TIV

ITY

Figure 6: High signal-to-noise ratio for desmopressin. The signal-to-noise ratio was calculated for desmopressin extracted from plasma at LLOQ level (0.500 pg/mL in plasma, S/N = 60.7).

Table 1: Compound-dependent parameters for desmopressin and desmopressin-d5 on the Triple Quad™ 6500 System.

InstrumentParameter Desmopressin Desmopressin-d5

CUR 40 40

TEM 600 °C 600°C

ISV 5500 5500

GS1 50 50

GS2 60 60

CAD 10 10

DP 50 71

EP 10 10

CE 23 23

CXP 12 12

Figure 7: Desmopressin technical replicates. Chromatograms of six LLOQ quality control samples (0.502 pg/mL) for precision and accuracy calculations are shown (Table 2).

Figure 5: Desmopressin MRM signal (shown in left side) panes for multiple concentrations and desmopressin D5 MRM signal (shown in right side panes).


Data processing

All Triple Quad 6500 System data was processed using MultiQuant™ Software. The concentration curves were analyzed using a linear fit with a 1/x2 weighting. Data acquired on the Triple Quad 6500 System was processed using the quantitation tools within Analyst® 1.6 Software.


Method analysis and data quality

The desmopressin quantitative assay was validated by generating an internal standard curve using standards alone and standards spiked into human plasma. Left side pane of Figure 5 shows representative peaks for A) blank extract, B) plasma spiked with

0.5 pg/mL desmopressin and the right side pane of Figure 5 shows the MRM response from the internal standard. Standard concentrations varied from 0.5 to 100 pg/mL generating an LLOQ in plasma of 0.5 pg/mL resulting in a signal to noise ratio of 60.7 (Figure 6). Reproducibility of the assay was assessed by multiple technical replicates of the same sample (n = 6, Figure 7) on an LLOQ quality control sample of 0.5 pg/mL. The calibration curve for desmopressin in plasma shows excellent linearity over 2.5 orders of magnitude concentration range with an r value of >0.99 (Figure 8).

The data collected for a single calibration curve are presented in Table 2. Analyte retention time and internal standard peak retention times were consistent, with both eluting at approximately 2.6 min. The calculated concentration correlates

ULTIM

ATE SEN

SITIVITY

Sample IDSample

Type

Analyte Retention Time (min)

Analyte Peak Area

IS Retention Time (min)

IS Peak Area

Area Ratio

Analyte Conc. (pg/mL)

Calculated Conc. (pg/mL) % Accuracy

AQM 18122013 Unknown 2.60 119046 2.57 119,422 0.997 N/A 60.309 N/A

BLANK 01 Blank 0 0 0 0 #DIV/0! N/A #DIV/0! N/A

BLANK+IS 01 Unknown 0 0 2.57 120,418 0 N/A No Peak N/A

STD A 01 Standard 2.61 2,148 2.58 141,964 0.015 0.500 0.507 101.40

STD B 01 Standard 2.60 3,567 2.57 157,232 0.023 1.000 0.967 96.73

STD C 01 Standard 2.60 8,190 2.57 168,629 0.049 2.498 2.544 101.85

STD D 01 Standard 2.60 20,050 2.57 180,711 0.111 6.448 6.344 98.39

STD E 01 Standard 2.60 51,086 2.57 187,669 0.272 16.122 16.167 100.28

STD F 01 Standard 2.60 94,706 2.57 145,728 0.65 40.304 39.173 97.19

STD G 01 Standard 2.60 218,389 2.57 163,086 1.339 80.608 81.158 100.68

STD H 01 Standard 2.60 321,985 2.57 187,366 1.718 100.760 104.268 103.48

LLOQ QC 01 Qual. Control 2.62 2,395 2.58 153,948 0.016 0.502 0.533 106.19

LQC 01 Qual. Control 2.60 5,296 2.57 160,710 0.033 1.492 1.593 106.75

MQC 01 Qual. Control 2.60 114,782 2.57 180,711 0.635 40.692 38.277 94.07

HQC 01 Qual. Control 2.60 244,061 2.57 183,047 1.333 81.384 80.806 99.29


LQC 02 Qual. Control 2.6 5,433 2.57 154,564 0.035 1.492 1.727 115.74

MQC 02 Qual. Control 2.60 119,912 2.57 183,163 0.655 40.692 39.465 96.99

HQC 02 Qual. Control 2.60 227,932 2.57 179,913 1.267 81.384 76.760 94.32


LQC 03 Qual. Control 2.61 5,266 2.58 153,330 0.034 1.492 1.678 112.44

MQC 03 Qual. Control 2.60 117,712 2.57 178,847 0.658 40.692 39.678 97.51

HQC 03 Qual. Control 2.60 228,311 2.58 175,270 1.303 81.384 78.936 96.99


LQC 04 Qual. Control 2.61 4,940 2.58 155,534 0.032 1.492 1.520 101.90

MQC 04 Qual. Control 2.60 102,226 2.58 156,822 0.652 40.692 39.294 96.56

HQC 04 Qual. Control 2.61 222,646 2.58 167,585 1.329 81.384 80.516 98.93


LQC 05 Qual. Control 2.61 5,299 2.58 161,308 0.033 1.492 1.587 106.33

MQC 05 Qual. Control 2.60 101,960 2.58 157,530 0.647 40.692 39.013 95.87

HQC 05 Qual. Control 2.60 225,914 2.58 169,163 1.335 81.384 80.937 99.45


LQC 06 Qual. Control 2.61 5,414 2.58 161,549 0.034 1.492 1.627 109.05

MQC 06 Qual. Control 2.61 103,636 2.58 159,865 0.648 40.692 39.076 96.03

HQC 06 Qual. Control 2.60 223,129 2.58 168,042 1.328 81.384 80.471 98.88

Table 2: Full analysis of precision and accuracy measurements for desmopressin (batch 1 samples) in human plasma.



ULT

IMA

TE S

ENSI

TIV

ITY

Nominal Concentration (pg/mL)

Desmopressin LLOQ QC LQC MQC HQC

PA BATCH

01

0.502 1.492 40.692 81.384

1 0.533 1.593 38.277 80.806

2 0.538 1.727 39.465 76.760

3 0.587 1.678 39.678 78.936

4 0.526 1.520 39.294 80.516

5 0.562 1.587 39.013 80.937

6 0.560 1.627 39.076 80.471

Mean 0.5510 1.6220 39.1338 79.7377

S.D (+/-) 0.02287 0.07302 0.48653 1.62681

C.V. (%) 4.15 4.50 1.24 2.04

% Nominal 109.76 108.71 96.17 97.98

N 6 6 6 6

PA BATCH

02

7 0.519 1.590 39.191 79.548

8 0.491 1.283 40.359 82.140

9 0.490 1.509 39.486 78.094

10 0.571 1.436 39.828 78.854

11 0.526 1.387 40.624 78.472

12 0.680 1.319 40.355 79.635

Mean 0.5462 1.4207 39.9738 79.4572

S.D (+/-) 0.0719 0.1159 0.5636 1.4443

C.V. (%) 13.17 8.16 1.41 1.82

% Nominal 108.80 95.22 98.24 97.63

N 6 6 6 6

PA BATCH

03

13 0.418 1.364 41.098 79.992

14 0.602 1.446 39.814 80.103

15 0.520 1.399 40.988 79.854

16 0.463 1.350 39.391 80.937

17 0.563 1.274 39.960 80.577

18 0.528 1.332 39.188 82.162

Mean 0.5157 1.3608 40.0732 80.6042

S.D (+/-) 0.05867 0.05867 0.80200 0.86337

C.V. (%) 11.38 4.31 2.00 1.07

% Nominal 102.73 91.21 98.48 99.04

N 6 0 6 6

Table 3: Precision and accuracy calculations for individual batches of desmopressin samples.

Nominal Concentration (pg/mL)

Desmopressin Sample LLOQ QC LQC MQC HQC

0.502 1.492 40.692 81.384

1 0.533 1.593 38.277 80.806

2 0.538 1.727 39.465 76.760

3 0.587 1.678 39.678 78.936

4 0.526 1.520 39.294 80.516

5 0.562 1.587 39.013 80.937

6 0.560 1.627 39.076 80.471

7 0.519 1.590 39.191 79.548

8 0.491 1.283 40.359 82.140

9 0.490 1.509 39.486 78.094

10 0.571 1.436 39.828 78.854

11 0.526 1.387 40.624 78.472

12 0.680 1.319 40.355 79.635

13 0.418 1.364 41.098 79.992

14 0.602 1.446 39.814 80.103

15 0.520 1.399 40.988 79.854

16 0.463 1.350 39.391 80.937

17 0.563 1.274 39.960 80.577

18 0.528 1.332 39.188 82.162

Mean 0.5376 1.4678 39.7269 79.9330

S.D (+/-) 0.05691 0.14051 0.73499 1.36511

C.V. (%) 10.56 9.59 1.85 1.71

% Nominal 107.57 98.53 97.64 98.21

N 18 18 18 18

Table 4: Mean precision and accuracy calculations for desmopressin for three batches of measurements from different days.

well with the actual spiked-analyte concentration in plasma matrix with a percent accuracy of the standard curve very close to 100% for all concentrations of standard, and the quality control samples had a percent accuracy of 110%. Table 3 shows the individual statistics for three separate batch runs of desmopressin. Data from Table 2 are taken from Batch 3. Table 4 shows the mean values for the percent accuracy and % CV for three separate batch runs. For the LLOQ quality control, the mean accuracy was calculated to be 108% with a %CV of 10.5%.

Figure 8: Calibration curve of desmopressin in plasma from 0.500 pg/mL to 100.760 pg/mL. The method has shown excellent linearity over the concentration range with r = 0.9996.


ULTIM

ATE SEN

SITIVITY

The percent recovery and plasma matrix effect were evaluated by comparing the peak areas for standard curve samples with and without plasma (Table 5). The mean percent recovery was calculated to be 93%. The recovery of the internal standard was calculated to be 78% (Table 6).

Conclusions

A highly sensitive and high-throughput bioanalytical method was developed and validated for the detection of ultra-low-levels of the therapeutic peptide, desmopressin, in human plasma on the SCIEX Triple Quad™ 6500 LC/MS/MS System.

Method sensitivity for desmopressin detection was exceptional (0.5 pg/ml or 2.5 fg on column), and demonstrated high-reproducibility and cost effectiveness with good precision and accuracy.

Analyte recovery is 92.7%, even under high-throughput conditions.

Total run time for each sample was only 5 min, using a flow rate rapid enough for high-throughput analysis in the bioanalytical laboratory.

Acknowledgements

The authors are indebted to Dr. Venu Madhav, Chief Operating Officer (COO), Veeda Clinical Research, India, for his encouragement and support for the successful completion of the work.

References1 Friedman, F and Weiss JP. “Desmopressin in the treatment of nocturia: clinical evidence and

experience.”Therapeutic Advances in Urology. 2013; 5(6): 310-317.

2 Neudert, L, Zaugg, M, Wood, S, Struwe, P. “A high sensitivity dual solid phase extraction LC/

MS/MS assay for the determination of the therapeutic peptide desmopressin in human plasma.”

Celerion white paper.

LQC RESPONSE MQC RESPONSE HQC RESPONSE

Sample ID Extracted Unextracted Extracted Unextracted Extracted Unextracted

01, 013 5,923 6,282 145,169 148,017 292,234 290,315

02, 014 6,645 6,157 135,153 145,270 267,547 291,138

03, 015 6,572 6,319 145,616 153,630 302,953 288,730

04, 016 5,879 6,446 141,823 156,298 270,214 318,201

05, 017 5,823 6,072 122,242 154,238 240,994 304,383

06, 018 5,263 5,825 114,853 146,425 260,462 287,647

Mean 6,017.5 6,183.5 134142.7 150,646.3 272,400.7 296,735.7

S.D 517.00 218.46 12860.35 4,634.26 22,289.26 12,153.15

C.V 8.59 3.53 9.59 3.08 8.18 4.10

N 6 6 6 6 6 6

% Recovery 97.32 89.04 91.80

Mean 92.72

SD (+/-) 4.216

CV (%) 4.547

N 3

Table 5: Recovery of desmopressin from plasma at three different concentrations, LQC, MQC and HQC, was 92.72%.

PA Batch No. 03

Sample ID Extracted (MQC) Unextracted

01, 013 251,778 301,602

02, 014 241,864 297,780

03, 015 253,224 305,778

04, 016 256,487 313,985

05, 017 217,971 316,703

06, 018 208,773 300,178

Mean 238,349.5 306,004.3

S.D 20,164.07 773,5.25

% C.V. 8.46 2.53

N 6 6

% Recovery 77.89

Table 6: Recovery for desmopressin-d5 from plasma at the MQC level was 77.89%.



Key challenges of high-throughput peptide quantification in plasma

Low accuracy – Detection of triptorelin in plasma at higher accuracy is needed for drug development regulatory requirements and therapeutic monitoring.

Diminished sensitivity – Detection at low pg/mL levels is challenging in complex matrices.

Substandard data quality – Precision and accuracy are compromised at low peptide levels, giving results below accepted bioanalytical standards.

Key benefits of IonDrive™ Technology for high-throughput peptide quantification

Excellent linearity – Dynamic range was linear over a wide peptide concentration in a complex matrix.

Ultrasensitive method – Triptorelin LLOQs of 5 pg/mL on the QTRAP® 6500 System were improved 8-fold over those obtained on the 5500.

Accurate and precise measurements – Data quality met or exceeded validation criteria over the standard curve range.

Key features of IonDrive™ Technology for high-throughput peptide quantification

IonDrive™ QJet Ion Guide – Increased ion sampling improves method efficiency and ruggedness.

IonDrive™ High Energy Detector – New detector technology boosts dynamic range and sensitivity.

Mass range of m/z 5 – 2,000 – Comprehensive mass range provides the versatility needed for peptide quant.

Ion Drive™ Turbo V Source – Increased ionization efficiency and heat transfer contribute to sensitivity enhancements, including improved signal-to-noise.

High-Sensitivity Quantification of the Triptorelin Decapeptide using the QTRAP® 6500 SystemIonDrive™ Technology delivers improved sensitivity for peptide

detection under high-flow conditions on a hybrid triple

quadrupole/linear ion trap mass spectrometer

Kelli Jonakin1, Simon Wood2, Laurence Meunier2

1SCIEX, USA, 2Celerion, Switzerland

Figure 1: IonDrive™ Turbo V Source and IonDrive™ QJet Ion Guide. High efficiency heaters provide efficient desolvation and ion production for high sensitivity at high flow rates. Dual RF stages maximize ion sampling from a large orifice while increasing ion transfer efficiency into the Q0 region – without increasing vacuum load in the analyzer region.

ULT

IMA

TE S

ENSI

TIV

ITY

Laurence Meunier Simon Wood


Introduction

Very high levels of sensitivity at high-flow rates have been achieved for peptide analysis on the SCIEX 6500 Series of mass spectrometers, and new methods on these instruments have demonstrated long-term robustness during multi-sample runs. Integral to the enhanced levels of sensitivity obtained on the 6500 Series, the new IonDrive™ Technology (Figure 1) boosts ionization efficiency, sampling, and transmission of ions, driving up the detector dynamic range, improving signal-to-noise measurements, and accelerating scan speeds. The new heater design of the IonDrive™ Turbo V Source improves desolvation and ion production, allowing for high-levels of sensitivity even at high flow rates. The innovative, dual QJet® Ion Guide increases the efficiency of ion transmission while maintaining simplicity and robustness. Together, these advances in detector technology and ion transmission augment the capacity for sensitivity in any peptide quantitation method.

Triptorelin, a ten-amino-acid synthetic peptide, is a gonadotropin-releasing hormone agonist (GnRH agonist) used in the treatment of hormone-responsive cancers such as prostate cancer and breast cancer (Figure 2). Used in men, triptorelin reduces the amount of testosterone in the blood, which limits the growth of prostate cancer. When given to women, triptorelin reduces the production of estrogen. Many clinical and pharmaceutical applications require low-level detection of triptorelin, and the demands of processing multiple biological samples requires a robust and high-throughput strategy. Here, we demonstrate a LC/MS/MS detection strategy for monitoring low levels of triptorelin in rat and human plasma at pg/mL levels using a QTRAP® 6500 System and fast chromatographic flow rates for sample run times of <5 minutes.


Sample preparation

A standard curve for the triptorelin peptide in rat plasma was prepared over a concentration range of 0.005–4.0 ng/mL. Samples underwent solid phase extraction (SPE), were dried down to completion, and then were reconstituted in H2O/10% MeOH/0.02% acetic acid (100 μL).

Chromatography

Direct injection workflowLC system: Shimadzu LC20AD LC System

Column: Ascentis Express Peptide ES C18, 2.7 μm (2.1 x 50 mm, Sigma Aldrich)

Guard column:


Injection: 20 μL

Flow rate: 300 μL/min

Mobile phase: A) water B) methanol, 0.02% acetic acid

Gradient: Time %B

0 15

1.2 60

1.8 60

1.81 90

2.3 90

2.31 15

4 15

Mass spectrometry

A QTRAP 6500 System equipped with an IonDrive™ Turbo V Source was operated in positive ESI mode. The MRM transition monitored for triptorelin was m/z 656.5/249.1 at a dwell time of 50 ms. Source and compound dependent parameters are shown in Table 2. Five replicate injections were performed at all concentrations.

Sensitivity was compared to previously obtained data from the QTRAP® 5500 System (courtesy of Celerion) using identical chromatography and MS methods. Source conditions on the IonDrive™ Turbo V Source were slightly altered compared to those optimized for the Turbo V™ Source on the 5500 System (Table 2).

Table 1. Chromatography gradient

ULTIM

ATE SEN

SITIVITY

Figure 2: Chemical structure of triptorelin (MW = 1311.5 g/mol).



Data processing

All QTRAP 6500 System data was processed using MultiQuant™ Software and the SignalFinder™ Algorithm. The concentration curves were analyzed using a linear fit with a 1/x2 weighting. Data obtained on the QTRAP 5500 System was processed using the quantitation tools within Analyst® Software.

Measuring triptorelin sensitivity in rat plasma

This triptorelin quantification method was initially optimized and validated on the QTRAP® 5500 System. A concentration curve was constructed from peak area measurements of peptide in rat plasma matrix. An LLOQ of 40 pg/mL was obtained for triptorelin (on column) on the QTRAP 5500 System; assessments of the data quality demonstrated a %CV of 9.7% and an accuracy of 102.8%. Figure 3 shows a representative chromatogram at the LLOQ of 40 pg/mL (0.305 fmol on column) from the QTRAP 5500 System.

A comparable chromatographic strategy was translated to the QTRAP 6500 System to evaluate sensitivities obtained on both instruments for peptide quantitation. The signal at the lower limit of quantification for triptorelin in plasma (n = 5) is shown in Figure 4, with very good reproducibility and S/N. The LLOQ obtained with this method was 5 pg/mL (0.038 fmol of triptorelin on column). The matrix blank and signal at the determined LLOQ (5 pg/mL) can be found in Figure 5.

The data (Figure 4) showed excellent linearity across the concentration curve range analyzed, from 5 pg/mL to 4 ng/mL on column. The statistics for this analysis are shown in Table 3. The coefficients of variation (%CV) and the accuracies of the curve fall well within commonly accepted bioanalytical validation criteria throughout the range of concentrations measured.

The observed sensitivity increase for this assay in rat plasma on the QTRAP 6500 System (5 pg/mL) was found to be ~8x improved over the QTRAP 5500 System (40 pg/mL) (Table 3).

ULT

IMA

TE S

ENSI

TIV

ITY

Figure 4: Quantification of triptorelin on the QTRAP® 6500 System. (Top) The signal measured at the LLOQ (5 pg/mL) in buffer is shown (one replicate, 0.038 fmol on column) with a CV of 12.2%. (Bottom) Using linear (1/x2) regression, good accuracy was achieved across the range of concentrations analyzed. Replicate triptorelin measurements (n= 5) were well within accepted bioanalytical method validation criteria (statistics are shown in Table 3).

InstrumentParameter QTRAP® 6500 System QTRAP® 5500 System

CUR 25 30

TEM 500 700

ISV 5500 5000

GS1 50 50

GS2 60 60

CAD 11 High

DP 90 180

EP 10 10

CE 39 39

CXP 12 12

Table 2: Source and compound-dependent parameters for the QTRAP® 6500 System and the QTRAP® 5500 System.

QTRAP® 6500 System QTRAP® 5500 System

Conc.(ng/mL)

On-Column(fmol)

Accuracy(%)

% Coefficientof Variation

Accuracy(%)

% Coefficientof Variation

0.005 0.038 97.1 12.2 – –

0.01 0.076 106.9 6.7 – –

0.02 0.152 98.9 7.0 – –

0.04 0.305 93.4 9.1 102.8 9.7

0.08 0.61 106.9 0.9 93.8 2.1

0.2 1.52 106.5 0.5 96.7 4.0

0.8 3.05 98.6 1.2 99.8 1.3

1.6 12.2 97.3 1.2 99 0.15

4 30.5 94.4 3.3 103.4 4.0

Table 3: Statistics for the quantification of triptorelin in buffer. Replicates (n = 5) were run at every concentration, and the statistics for accuracy and precision were computed. The LLOQ for triptorelin on the QTRAP® 6500 System (bold) was 0.038 fmol of the molecule on column (5 pg/ml), 8-fold improved over the LLOQ obtained on the QTRAP® 5500 System at 0.305 fmol on column (40 pg/mL).


Conclusions

High-throughput peptide quantitation at high sensitivity was performed on the SCIEX QTRAP® 6500 System equipped with IonDrive™ Technology.

Ultra-low levels of triptorelin were detected in plasma on the QTRAP® 6500 System under high-flow conditions – giving an LLOQ of 5 pg/mL (0.038 fmol of peptide on column).

The high-flow data obtained on the QTRAP 6500 System provided detection limits for triptorelin of 5 pg/mL – an ~8x improvement in LLOQs previously developed on the QTRAP 5500 System.

References1 The SCIEX Triple Quad™ 6500 and QTRAP® 6500 Systems for Bioanalysis – A New Level of

Sensitivity, SCIEX Technical Note, Publication 5780212-0

ULTIM

ATE SEN

SITIVITY

Figure 5: Detection of triptorelin on the QTRAP® 6500 System. Peak areas for triptorelin present in the matrix at 0 pg/mL (left) and at the LLOQ of 5 pg/mL (right).



With the ability to collect data at rates as high as 50 Hz in either MS or MSMS mode, the TripleTOF® 5600 system offers unique way to support both qualitative and quantitative analysis. First and foremost, compatibility with UPLC separation where the system can easily be set to collect more than 12 data points across most LC peaks. Secondly, the high resolution (>30K) and the high mass stability (<2ppm rms) provides the ability to extract narrow ion chromatogram (<10mDa) to achieve selectivity in MS mode that is comparable to MRM analysis on QqQ systems. The advantage of this approach is that generic data acquisition can be used

HIG

H R

ESO

LUTI

ON

AC

CU

RA

TE M

ASS

For many years, quantitative analysis were carried on triple quadrupole (QqQ) MS instruments, while qualitative analysis would be carried on instruments with high resolution and mass accuracy, such as Q-ToF MS systems. As laboratories are looking into improving efficiency in streamlining their decision making process, research ranging from drug metabolism to proteomics are looking at instrument that would provide both qualitative and quantitative data simultaneously. As we looked at the attributes of MS system that deliver typical quantitative analysis, their main features are: high duty cycle, high sensitivity, high selectivity and wide dynamic range. In addition to all of the front end development associated with QqQ systems such as the Turbo V™ source and the QJET® ion optics, key technologies were integrated into the TripleTOF® 5600 system that enabled high performance quantitative analysis. First, operating the accelerator/pulser at 30kHz provides high duty cycle extraction of the ion beam exiting the collision cell. To match this capability, a 40-GHz multichannel TDC detection system that ensures high rate data collection was integrated into the system. Secondly, operating at 15 kV TOF acceleration voltage with high transmission grids (~92% transparency) assisted in maintaining a great deal of the sensitivity gains from the front end changes. Both of these technologies ensure high efficiency extraction of the ion beam to provide high sensitivity. On the qualitative front, it was also important to improve the performance of the system in terms of resolution as well as mass accuracy. To improve resolution above 30,000 resolution, the ion optic was optimized to transfer ions with coherent ion trajectories in the pulser region over a distance of 2.5m. And finally, the last key attribute of the system was to ensure to maintain mass accuracy <2ppm RMS over long periods of analysis. In the case of the TripleTOF® 5600 system, this is done in a 2 step fashion: first the mass accuracy is established via scheduled introduction of calibrant as part of batch, and secondly the mass precision is maintained by dynamic monitoring of background ions that are determined adaptively to analysis conditions.

High Resolution Time-of-Flight for High Quality Quantitative AnalysisYves Leblanc1SCIEX, Concord, ON, Canada

Figure 1: Analysis of verapamil in diluted urine samples (2x). Data was collected at 30K resolution in MS mode. The width of the XIC was set to mimic single quadrupole analysis (XIC = 0.7 Da) and high resolution mode (XIC = 10mDa)

XIC Peak Width0.7 Da (Unit resolution)

10 mDa (High resolution)

Time, min

Time, min 1.0 2.0 3.0 4.0

1.0 2.0 3.0 4.0

Yves Leblanc


for analysis, thus the need for tuning. Equally important is the ability to obtained reliable peak area from narrow-XIC which is indicative of the instrument stability in terms of mass accuracy. Figure 1 shows the benefit of reducing the XIC width in order to gain selectivity in the detection of verapamil in diluted urine. As the XIC width is reduced from unit resolution (0.7Da to mimic quadrupole isolation) to 10mDa, all interferences are eliminated and a single LC peak is detected. This approach can be further extended to peptide analysis by summing the signal associated with both the charge state distribution as well as the isotope distribution for each ion. Figure 2 shows the observed isotope distribution associated with the +3 an +4 ions of neuromedin U (NMU, seq.: YKVNEYQGPVAPSGGFFLFRPRN). As can be seen here, the isotope contributes to a large portion of the ion signal and each one of them can be combined to give the appropriate S/N for selectivity and proper detection of LC peaks. For neuromedin U spiked in protein precipitated, considering the top 4 isotopes, more than 80% of the signal can be captured for detection. This approach was shown to improve linearity and precision for the detection of NMU.

An additional benefit of the TripleTOF® 5600 system is the ability to record MSMS spectra that can also be processed post-LC. This mode of operation is referred to as MRM-HR from which selective fragment(s) can be used as representative of the peptide LC peak. Figure 4 shows an example associated with the ability to extract multiple fragment ions associated a given peptide, in this particular case a phosphorylated peptide. This also provides the ability to compare the experimental mass spectrum to library entries to ensure the proper peptide was detected.

HIG

H R

ESOLU

TION

AC

CU

RA

TE MA

SS

Figure 2: Isotope distribution of neuromedin U for the dominant charge state (Z=+3 and Z=+4). When the highlighted isotopes are extracted and combined, then 91% of the signal association with the peptide, thus improving the overall sensitivity of the system.

Figure 3: Neuromedin U spiked into protein precipitated plasma. For each charge states, the top 4 isotopes were extracted and summed for each charge states. The 3 charge states can be also be combined to further improve the sensitivity as little noise is captured in the process.

Figure 4: Narrow XIC associated with dominant fragment masses of the illustrated peptides. The full scan mass spectrum can also be compared to library spectrum to gain further confidence in the detection of the peptide.

Z = +4

Z = +3

SUMMED top 4 Isotope

+3

+2

1388

1000

500

0

565

400

200

0

51

40

20

0

Time, min

Time, min

Time, min

0 1.0 2.0 3.0 4.0 5.0

0 1.0 2.0 3.0 4.0 5.0

0 1.0 2.0 3.0 4.0 5.0

[PGQ]-QSPASPPPLGGGAPVR

[email protected]

+4



Key challenges of exenatide quantitation in biological samples

Insufficient sensitivity – The best, previously reported LOQ is 100 pg/mL; extended-release pharmacokinetic studies demand lower levels of detection.

Limited quantitation range – Analytical range of ELISA-based method is <2 orders of magnitude; at least 3 orders of magnitude is desired in bioanalysis.

Low specificity – Complex biological matrices hamper data resolution and require sophisticated sample preparation and/or advanced instrumentation.

Systematic measurement errors – Especially for ultra-low level quantitation, measurement errors have a significant effect on data accuracy and precision.

Key benefits of MicroLC and MRMHR workflow

Robust sensitivity – Up to 10-fold better than the best reported result (LLOQ of 10 pg/mL using the nanoLC trap-and-elute method) over a linear range 3 orders in magnitude.

Improved specificity – Background noise is significantly reduced with MRMHR offering the high selectivity.

Excellent accuracy and precision – %CV <14% across the whole analytical range; deviation of accuracy <18% including LOQ.

Reduced operating costs – MicroLC usage results in >90% savings on solvents and waste disposal.

Key features of MicroLC and MRMHR workflow

Dedicated microflow UHPLC system

Seamlessly integrated microLC ESI interface

MRM-like quantitation – with high resolution fragment ion spectrum.

Fast cycle times – compatible with UHPLC speed

Wider linear dynamic range of 4 orders – with TripleTOF® 5600 system

HIG

H R

ESO

LUTI

ON

AC

CU

RA

TE M

ASS Ultrasensitive Quantitation of Exenatide Using

Microflow Liquid Chromatography Systems and High-Resolution Mass SpectrometryA high resolution multiple reaction monitoring (MRMHR) method for low-level peptide quantitation developed

on an SCIEX TripleTOF® 5600 LC/MS/MS System coupled with Eksigent ekspert™ microflow ultrahigh pressure

liquid chromatography (μUHPLC) systems

Leo Jinyuan Wang1, Daniel Warren2, and Anthony Romanelli2

1SCIEX, Redwood City, CA; 2SCIEX, Framingham, MA;

Eksigent ekspert™ microLC µUHPLC Optimized Micro Flow ESI SourceTripleTOF® 5600 System


Introduction

Exenatide, marketed as Byetta and Bydureon, is an effective medication for treating type 2 diabetes. This 39-amino-acid peptide is a synthetic version of exendin-4, a hormone excreted in Gila monster saliva. Exenatide displays biological properties similar to human glucagon-like peptide-1 (GLP-1), a regulator of glucose metabolism and insulin secretion. While exenatide shares 50% amino acid homology with GLP-1, exenatide is more resistant to metabolic degradation and, thus, has a therapeutic advantage due to its longer pharmacological half-life.

Reported methods for quantifying exenatide in plasma include immunoassay (ELISA)1,2 and liquid chromatography mass spectrometry (LC/MS).3,4 Quantitation by immunoassay suffers from limited analytical range, endogenous interferences, lack of reproducibility and specificity, and a requirement for expensive antibodies. Alternatively, LC/MS detection methods have excellent selectivity (when using MS/MS or high resolution MS/MS (MS/MSHR)), high sensitivity, a wide analytical range (usually greater than 3 orders of magnitude), and good reproducibility. Currently published MS/MS methods report adequate sensitivity for exenatide quantitation (LLOQs of 100 pg/mL).3 These levels may be sufficient to quantify quick-release exenatide formulations in plasma (i.e., Byetta); however, quantifying ultra-low levels of the slow-release formulation in plasma remains challenging.

Microflow ultra-high-performance liquid chromatography (μUHPLC) has gained substantial popularity for peptide detection, where low sensitivity, slower throughput, and rising operational costs have been constant challenges. The advantages

of μUHPLC/MS methods over highflow LC include approximately 14-fold sensitivity gain5, fast gradient separation, reduced source contamination, and shrinking solvent consumption and disposal costs.6

We developed an ultra-sensitive μUHPLC-MRMHR method suitable for quantitating low-level exenatide in plasma, resulting in an LLOQ of 10 pg/mL that gives excellent linearity, accuracy, and precision. In addition, two injection workflows were compared to determine the impact of sample handling on sensitivity: 1) a direct injection method on the ekspert microLC 200 System and 2) a large-volume injection method combined with a trap-and-elute protocol on the ekspert nanoLC 425 System.


Sample preparation

Proteins were precipitated from of plasma by mixing plasma (1 mL), acetonitrile (3 mL), and formic acid (0.4 mL). After vortexing for 15 sec, the mixture was centrifuged (4000 rpm for 15 min.). The supernatant was transferred and stored at -20 °C.

Exenatide stocks and isotope-labeled internal standard were kindly supplied by GlaxoSmithKline. Stock solutions were prepared in 20% acetonitrile and diluted in rat plasma in series from 1–100 ng/mL. For the trap and elute workflow, multiple concentrations of calibration standards were prepared as follows: 10, 20, 50, 100, 200, 500, 1000, and 2000 pg/mL. Dilutions were spiked with internal standard (1 ng/ml) and 20% rat plasma. For the direct injection workflow, calibration standards were prepared in rat plasma from 50–50,000 pg/mL (10 concentrations).

HIG

H R

ESOLU

TION

AC

CU

RA

TE MA

SS For Research Use Only. Not for use in diagnostic procedures.


μUHPLC-MRMHR analysis

Direct injection analysis was performed using an Eksigent ekspert™ microLC 200 System; and the large volume injection analysis with the trap-and-elute workflow was performed on an ekspert nanoLC 425 System. A high-resolution quadrupole time-of-flight mass spectrometer, the SCIEX TripleTOF® 5600 System, was employed to capitalize on the instrument’s high specificity with analytes in biological samples. The parameters and instrument configurations are listed as follows:

Chromatography

Direct injection workflow

LC system: Eksigent ekspert microLC 200 μUHPLC with 20–200 μL/min flow module

Column: Eksigent ChromXP™ C18CL, 1 × 50 mm, 3 μm

Guard column: C18, 1 × 5 mm, 5 μm


Injection: 10 μL (10 μL loop with 20 μL loading volume)

Flow rate: 150 μL

Mobile phase: A) water, 0.1% formic acid B) acetonitrile, 0.1% formic acid


0 95 5

0.5 95 5

2 10 90

4 10 90

4.1 95 5

5.0 95 5

Trap-and-elute workflow

To analyze samples using the trap-and-elute workflow, two gradient pumps (set for 5-50 μL/min flow rates) and two valves (6-port, 2-position injection valve and 10-port, 2-position switching valve) were configured for synchronized sample loading, transferring, injection, and elution. The detailed flow diagram is illustrated in Figure 1.

LC system: Eksigent ekspert nanoLC 425 μUHPLC with 5–50 μL/min flow modules (2x)

Column: Eksigent ChromXP C18CL, 0.5 × 50 mm, 3 μm

Trap column: C18, 0.5 × 5 mm, 5 μm

Column temp.: ambient

Injection: 50 μL (50 μL loop with 60 μL loading volume)

Gradient 1: Loading pump

Flow rate: 50 μL

Mobile phase: A) water, 0.1% formic acid B) acetonitrile/isopropanol (1:1), 1% trifluoroethanol, 0.1% formic acid

Gradient 1: Time/min A% B%

0 100 0

1.5 100 0

1.6 0 100

5.5 0 100

5.6 100 0

6.0 100 0

In-line filter: Micro filter with 1 μm titanium frit for 1/32” PEEKsil tubing

Gradient 2: Eluting pump

Gradient 2: Time/min A% B%

0 95 5

0.5 95 5

2 10 90

4 10 90

4.1 95 5

5.0 95 5

Flow rate: 40 μL


Chromatography and MS synchronization was achieved with the nanoLC 400 System autosampler method detailed below:

Initialize Autosampler device

Valve Switch injection valve to load

HIG

H R

ESO

LUTI

ON

AC

CU

RA

TE M

ASS

Figure 1: Trap-and-elute workflow flow diagram


Valve Switch ISS-A valve to load (10-1)

Wait for 1 sec

Needle wash Pre-wash 1x using wash solvent 1

Wait for gradient 2 to be ready

Wait for gradient 1 to be ready

Get sample Fill loop 60 µL: 2 mm from bottom at 2 µL/s

Wait for 5 sec

Start Start gradient 1

Valve Injector inject

Wait for 1 min 30 sec

Start Gradient pump 2

Valve Switch ISS-A valve to inject (1-2)

Wait for 1 min 40 sec

Valve Switch ISS-A valve to load (10-1)

Needle wash Wash 2x cycles, inner wash solvent 2, outer wash solvent 2

Needle wash Wash 2x cycles, inner wash solvent 1, outer wash solvent 1

Mass spectrometry

System: TripleTOF 5600+ System

Interface: DuoSpray™ Ion Source with 65 μm electrode

Ion source gas 1 (GS1): 55


Curtain gas (CUR): 20

Temperature (TEM): 650 °C

Ion spray voltage 5500

Floating (ISVF):

Scan type: TOF Product Ion Scan of 833 and 844

Scan experiment 1: 220–420 with high sensitivity mode and enhanced mass 396.2

Scan experiment 2: 220–420 with high sensitivity mode and enhanced mass 402.2

Accumulation time: 0.15 sec for each scan experiment

Declustering potential: 141

Collision energy: 30

Collision energy spread: 5

Calibration: Automated calibration with CDS

HIG

H R

ESOLU

TION

AC

CU

RA

TE MA

SS Figure 2: Typical chromatograms and TOF MS/MS spectra at 100 pg/mL exenatide in plasma.

Figure 3A: Chromatograms of low-level exenatide in plasma using the nanoflow-based, trap-and-elute workflow.

Figure 3B: Chromatograms of low-level exenatide in plasma using the microflow-based direct injection workflow.



proportionally with increasing concentrations of exenatide for both chromatography methods.

Peptide residue removal between sample runs

Ensuring an uncontaminated system after repeated sample exposure is critical for accurate quantitation of ultra-low levels of peptides. An exceptionally hydrophobic peptide, exenatide sticks persistently to the sampling path and the column’s stationary phase, and a strong, organic wash (see method) is required to effectively remove exenatide from needle surfaces and sampling paths. Following the injection of a high concentration of sample (3333 pg/mL, Figure 4), the exenatide residue was quickly reduced to a very low level after three injections of wash solution (<2 pg/mL based on estimated peak height) and eliminated completely after five injections.

HIG

H R

ESO

LUTI

ON

AC

CU

RA

TE M

ASS

Figure 4: Effectiveness of the system cleaning protocol between sample runs.

Figure 5A: Calibration curve with exenatide concentration data obtained from the direct injection workflow.

Figure 5B: Calibration curve developed from exenatide concentration data obtained using the trap-and-elute workflow.

Injection #1: Blank Matrix

Injection #2: Exenatide 3333 pg/mL in Matrix

Injection #5: 1st Blank Matrix

Injection #7: 3rd Blank Matrix

Injection #3: 1st Wash Solution(autosampler strong wash solvent)

Injection #4: 2nd Wash Solution

Injection #6: 2nd Blank Matrix

Effective System Cleaning

Data processing

The MRMHR transitions for exenatide quantitation were 838/396 and 838/299. These transitions were determined to be optimal based on MS/MS analysis (not shown). Data were quantitated using MultiQuant™ Software.


μUHPLC-MRMHR method development

An ultrasensitive method for quantitating ultra-low levels of exenatide in plasma was developed using high-resolution mass spectrometry, relying on the quantitation of peaks obtained from MRM scans specific to the intact peptide (or fragment of the peptide?). To evaluate the effects of different sample injection methods and flow rates on the sensitivity of the exenatide detection method, two chromatography workflows were created: 1) a simple, direct-injection, high-throughput workflow was developed on a more affordable microLC 200 system that produced excellent sensitivity (LLOQ of 50 pg/mL); and 2) a sophisticated, flexible, large-volume-injection, trap-and-elute workflow was established on a nanoLC 425 system that produced the best sensitivity between the two systems (LLOQ of 10 pg/mL).

A representative total ion chromatogram (TIC) of exenatide (100 pg/mL) in plasma (analyzed using the nanoflow trap-and-elute workflow) is shown in Figure 2. Peaks eluting at 1.3 min were extracted for MS/MS analysis, and MRMHR transitions of 838/396 and 838/299 were followed, providing excellent specificity. To ensure high quality data, it was only necessary to acquire only two MRMHR experiments due to the high resolution achieved using the narrow MS/MS scan windows.

Ultra-low level amounts of exenatide can be reliably quantitated by the nanoflow-based, trap-and-elute workflow (Figure 3A) and by the microflow-based, direct-injection workflow (Figure 3B) using the selected MRMHR transitions. Exenatide detection in plasma using both the nanoflow- and microflow-based protocols reveals the maintenance of peak shape over a wide range of concentrations. Additionally, the detector response increases


Assessment of quantitative results

Exenatide quantitation methodology was evaluated for sensitivity, linearity, accuracy, and precision for both the direct injection and trap-and-elute chromatography workflows.

Sensitivity was defined by the LLOQ achieved – the lowest concentration in calibration standards that satisfies both signal-to-noise (S/N >10) and statistical requirements (precision and accuracy with <20% deviation). The LLOQ from the direct-injection method was determined to be 50 pg/mL with a dynamic range of 50–50,000 pg/mL (Figure 5A). The trap-and-elute workflow significantly reduced the LLOQ to 10 pg/mL (Figure 5B), enabling exceptionally low-level analysis of exenatide in plasma within a dynamic range of 10–2,000 pg/mL. An excellent linear response across a wide concentration range was demonstrated for both chromatography workflows (Figure 5A and 5B).

Both microflow methods demonstrated good data reproducibility and accuracy within an acceptable experimental range, and peak areas were calculated from multiple injections at each concentration (n>3). Method precision and accuracy are summarized in Table 1. For the direct-injection workflow, the %CV was observed within 11%, and data accuracy ranged from 86–118%. For trap-and-elute workflow, %CV was observed within 14%, and accuracy was from 89.5–112%.

Conclusions

A high-resolution method for peptide quantitation was developed for the TripleTOF® 5600 system using a trap-and-elute, microflow workflow (on an Eksigent ekspert nanoLC 425 System) that provided an extremely reproducible and highly-sensitive method for the ultra-low-level quantitation of exenatide (LLOQ = 10 pg/mL). An additional microflow direct injection workflow (on the ekspert microLC 200) was created for a higher-throughput, but less sensitive method of exenatide quantitation (LLOQ = 50 pg/mL).

References1 Fineman, M.; Flanagan, S.; Taylor, K.; Aisporna, M.; Shen, L.; Mace, K.; Walsh, B.; Diamant, M.;

Cirincione, B.; Kothare, P.; Li, W.-I.; MacConell, L. Pharmacokinetics and pharmacodynamics of

exenatide extended-release after single and multiple dosing. Clin Pharmacokinet. 2011, 50 (1),

65-74.

2 Lin, Y.-Q.; Khetarpal, R.; Zhang, Y.; Song, H.; Li, S. S. Combination of ELISA and dried blood

spot technique for the quantification of large molecules using exenatide as a model. Journal of

Pharmacological and Toxicological Methods. 2011, 64 (2), 124-128.

3 Zhang, J.-F.; Sha, C.-J.; Sun, Y.; Gai, Y.-Y.; Sun, J.-Y.; Han, J.-B.; Shao, X.; Sha, C.-N.; Li, Y.-X.;

Liu, W.-H., Ultra-high-performance liquid chromatography for the determination of exenatide in

monkey plasma by tandem quadrupole mass spectrometry. Journal of Pharmaceutical Analysis.

2013, 3 (4), 235-240.

4 Kehler, J. R.; Bowen, C. L.; Boram, S. L.; Evans, C. A. Application of DBS for quantitative

assessment of the peptide Exendin-4; comparison of plasma and DBS method by UHPLC–MS/MS.

Bioanalysis. 2010, 2 (8), 1461-1468.

5 Christianson, C. C.; Johnson, C. J. L.; Needham, S. R. The advantages of microflow LC–MS/MS

compared with conventional HPLC–MS/MS for the analysis of methotrexate from human plasma.

Bioanalysis. 2013, 5 (11), 1387-1396.

6 Yang, M.; Gong, X.; Schafer, W.; Arnold, D.; Welch, C. J. Evaluation of micro ultrahigh pressure

liquid chromatography for pharmaceutical analysis. Analytical Methods. 2013, 5 (9), 2178.

HIG

H R

ESOLU

TION

AC

CU

RA

TE MA

SS

Direct Injection Workflow Trap-and-Elute Workflow

Concentration

(ng/mL) Mean %CV %Accuracy

Concentration

(pg/mL) Mean %CV %Accuracy

0.05 0.045 5.8 90.3 10 10.6 12.2 106

0.1 0.086 10.1 86 20 18.2 13.2 91.2

0.25 0.23 6.62 90.7 50 45 6.84 90

0.5 4.33 1.18 86.6 100 103 6.6 103

1 1.18 1.55 118 200 196 6.47 98

2.5 2.59 5.45 103 500 447 8.84 89.5

5 5.17 3.96 103 1000 1124 6.23 112

10 11.6 2.72 116 2000 2190 4.71 109

25 27.1 1.3 108

50 47.9 2.33 95.8

Table 1: Summary of Quantitation Performance.

Direct Injection Trap-and-Elute

LLOQ (pg/mL) 50 10

Injection Volume (μL) 10 50

Linear Range (pg/mL) 50–50,000 10–2,000

%CV <11% <14%

Accuracy 86%–118% 89.5%–112%

Run Time 6 min 10 min

Table 2: Comparison of microLC workflow features for exenatide quantitation



MS/MSALL with SWATH™ Acquisition is a data independent workflow, performed on the TripleTOF® 5600+ system, which is of great interest today as it provides a higher level of reproducibility and comprehensiveness in proteomics data1. In a SWATH acquisition experiment, a wide Q1 isolation window is stepped through the precursor mass range, transmitting multiple analytes into the collision cell. The transmitted ions from each step are fragmented and a composite MS/MS spectrum is measured in the TOF MS Analyzer at high speed and high resolution. Post-acquisition, the peptides of interest are quantified by generating fragment ion extracted ion chromatograms (XICs) and measuring their peak areas.

For drug development, understanding the protein expression levels of drug metabolizing enzymes responsible for phase I and II bio-transformations (Figure 1) is a fundamental aspect of assessing drug-drug interactions, and evaluating drug safety and efficacy. Targeted quantitation using multiple-reaction-monitoring (MRM) has been used to quantitatively profile these enzymes in liver hepatocytes or microsomes. However, an MRM experiment normally focuses on a limited set of selected proteins for quantitation and requires significant upfront assay development work. In this work, the MS/MSALL with SWATH™ Acquisition method was used to analyze large numbers of proteins and multiple enzyme families involved in drug metabolism.

Key Advantages of Targeted Quantitation using MS/MSALL with SWATH™ Acquisition

• High quality protein quantitation strategy for biological samples• ‘MRM-like’ quality quantitation obtained on large numbers of proteins

and peptides• No method development required to target and quantify large numbers

of proteins and peptides in a single run• Easy transition to future MRM assays• Data independent acquisition using the TripleTOF® 5600+ Systems and

MS/MSALL with SWATH™ Acquisition• High sensitivity and speed of MS/MS acquisition• Easy data processing using the SWATH™ Acquisition MicroApp in

PeakView® Software• Post-acquisition extraction of large numbers of high resolution

sequence specific fragment ions of the targeted peptides and proteins

to generate peak areas with high specificity

HIG

H R

ESO

LUTI

ON

AC

CU

RA

TE M

ASS Investigating Biological Variation of

Liver Enzymes in Human HepatocytesMS/MSALL with SWATH™ Acquisition on the TripleTOF® Systems

Xu Wang1, Hui Zhang2, Christie Hunter1

1SCIEX, USA, 2Pfizer, USA


HIG

H R

ESOLU

TION

AC

CU

RA

TE MA


• Targeted multiple peptides per protein to confirm differential

quantitation on proteins from key protein families that share high

sequence homology• High quality, reproducibility LCMS using Eksigent ekspert nanoLC 425

System with cHiPLC® System.

Methods

Sample Preparation: Tissue cells (0.625 million) were resuspended in 1mL extraction buffer from ProteoExtract Native Membrane Protein Extraction Kit (EMD, Billerica, MA) with protease inhibitor. The cells were then lysed for 10 min at 4°C. The lysate was mixed with 250µL 100mM NH4HCO3 /3.6% DOC buffer and shaken for 20 minutes. Reduction agent (50µL of 100 mM DTT) was added and incubated for 10 min at 95 °C to disrupt disulfide bonds, followed by alkylation of free sulfhydryl groups with 50µL of 5mM iodoacetamide at room temperature in the dark for 30 min with continuous shaking. Extracted proteins were digested with trypsin (1:50) at 37 °C for 18 hrs. The digestion was stopped with addition of 0.2% formic acid/H2O solution, then

vortexed and centrifuged at 10,000 g for 5 min. The supernatant was transferred to a new eppendorf tube and dried down in speed vacuum for 3hrs at 50°C.

Chromatography: Tryptic digests were separated using an Eksigent ekspert nanoLC 425 with a cHiPLC® column (75 µm x 150 mm, 300 Å pore size ChromXP™ column) running a flow rate of 300 nL/min. The gradient was 90 mins as follows: 5 % B for 2 min, from 5 % B to 50 % B in 60 min, from 50% B to 90 % B in 8 min, 98% B for 5 min, from 98 % B to 5 % B in 5 min, and 5 % B for 10 min. (10 μL sample) Mobile phase A consisted of H2O and 0.1% formic acid, and mobile phase B consisted of acetonitrile and 0.1% formic acid. The column oven was operated at 35 ˚C. Sample injection volume was 10 µL.

Lipophilic

R

R RR

O

O R — SG

R — SO3H

R — OH R—SH R—NH2 R — Ac

R — GI

Phase I Phase II

Oxidation

HydrolysisReduction

Glutathioneconjugation

Sulfation

Acetylation

Glucuronidation

Electrophiles

Hydrophilic

Nucleophiles

Para

llel D

enat

ure

& D

iges

t

SWA

TH™

Acq

uis

itio

nID

A

. . .

1 2 3

984 99

13 Liver Hepatocyte Samples

Apply Ion Library

Hepatocyte Samples

Rel

ativ

e Q

uan

tifi

cati

on

(%

)

140

120

100

80

60

40

20

0

EST1FLSDLOGDPR

EST1AGQLLSELFTNR

EST1EVAFWTNL

EST1YLGGTDDTVK

Hepatocyte Samples

Rel

ativ

e Q

uan

tifi

cati

on

(%

)

140

120

100

80

60

40

20

0

EST2IQELEEPEER

EST2FTEEEEQLSR

091 098 DAD I2G IDE KMI MRS NQT RML ROE SED VCM YAA

091 098 DAD I2G IDE KMI MRS NQT RML ROE SED VCM YAA

Figure 1. Phases of Detoxification. Drug metabolism is the biochemical modification of xenobiotics or drugs using a specialized enzymatic system. These metabolic pathways modify the chemical structure of the foreign compounds to detoxify and prepare for excretion. Understanding the biological variation across individuals as well as the protein expression changes in response to drug is critical during drug discovery. © Wikipedia.

Figure 2. Experimental Design. Protein extracts from thirteen liver samples were digested in parallel. The pooled sample was created and analyzed using an IDA experiment in order to generate a spectral ion library. The individual samples were analyzed in triplicate using SWATH™ Acquisition. Resulting data was then analyzed using targeted data extraction as guided by the ion library.

Figure 3. Reproducible Quantitation of Liver Proteins. Quantitation for individual proteins of interest can be easily extracted from the SWATH™ acquisition dataset. Shown here are two different esterases, EST1 - liver carboxylesterase 1 and EST2 - cocaine esterase. There is very good agreement between the multiple peptides per protein highlighting the reproducibility of quantitation.



HIG

H R

ESO

LUTI

ON

AC

CU

RA

TE M

ASS

Mass Spectrometry: Eluant from the column was sprayed using the NanoSpray® Source into a TripleTOF® 5600+ system (SCIEX). Data were acquired using an MS/MSALL with SWATH™ acquisition method with a Q1 window size of 25 Da and a mass range of 400-1000 m/z (cycle time 2.5 sec). Information dependent acquisition (IDA) experiment was performed on the pooled sample to obtain peptide identifications to generate ion library. Thirteen individual hepatocytes samples (labeled 098, 091, DAD, I2G, IDE, KMI, MRS, NQT, RML, ROE, SED, VCM, YAA) were analyzed in triplicate by SWATH acquisition (Figure 2). Protein/peptide data were loaded into Skyline for MRM assay development and samples were also run in triplicate by MRM on the QTRAP® 6500 System.

Data Processing: The pooled sample was analyzed with ProteinPilot™ Software 4.5 beta to create a spectral library of proteins and peptides in the sample. The SWATH Acquisition data was processed using the SWATH™ Acquisition MicroApp 1.0 in PeakView® Software. Only proteins that were identified at a 1% global FDR were used in SWATH acquisition processing. Fragment ion XICs were summed to obtain peptide peak areas, and the areas for multiple peptides per protein were summed to obtain protein areas. Statistical analysis including principal component analysis (PCA) and t-tests were conducted with MarkerView™ Software 1.2. Data analysis of the MRM data was performed using MultiQuant™ Software 2.1.

PC2

Sco

re%

Res

po

nse

PC2

Load

ing

-10000 0 10000 -0.05 0.00 0.05 0.10 0.15

091 098 098 DAD I2G I2G IDE KMI KMI MRS NQT NQT RML ROE ROE SED VCM VCM YAA

PC1 Score

Sample (by index)

PC1 Loading

12000

8000

4000

0

-4000

-8000

-12000

-16000

100

80

60

40

20

0

0.16

0.12

0.08

0.04

0.00

-0.04

-0.08

-0.12

Scores for PC1 (27.7%) versus PC2 (19.5%), Pareto Loadings for PC1 (27.7%) versus PC2 (19.5%), Pareto

CYP2C8

CYP2A6

091

091091

098

098

098

098

I2G

I2G

I2G

I2G

DAD

DAD

DAD

DAD

IDE

IDE

IDE

IDE

KMI

KMIKMI

KMI

MRS

MRSAGXT

EPHX1

CPS1

CYP3A4

ECHS1

GLUD1

ALDH1B1

ARG1

ALDH1A1

ADH1B

ALDOB

GSTA2

PCK2CES1

HLO17B13

ACAA2

MRS

MRS

NQT

NQT

NQTNQT

RML

RMLRML RML

ROE

ROE

ROE

ROE

SED

SED

SED

SED

VCM

VCM

VCMVCM

YAA

YAAYAA

091

Figure 4. Analyzing Global Protein Expression Differences Across the Samples. After the data extraction was performed on the SWATH acquisition data, the protein areas were loaded into MarkerView™ Software for statistical analysis. PCA was performed to obtain the Scores and Loadings plots (top) for easy visualization of the proteins that are showing the highest degree of change across the multiple samples. As an example, the protein expression differences across the different liver hepatocyte samples for the two cytochrome P450 proteins, CYP2C8 and CYP 2A6, are shown.


High Quality Quantification on Large Number of Proteins

The pooled sample was used to generate a spectral ion library containing more than 2000 proteins. From the SWATH™ acquisition, an average of 1987 proteins was quantifiable across the 13 samples. Highly reproducible results were obtained for enzyme families of interest, such as 19 CYP proteins, 12 UGT proteins, and 7 GST proteins. The reproducibility was assessed across three technical replicates. The quantitative profiling of the key metabolizing enzymes potentially helps to discover the expression variations of these enzymes, as well as their correlation/anti-correlation properties across the sample set and ultimately within populations when larger sample sizes are available. This systematic understanding of protein expression is essential during the drug development process.

Figure 3 (previous page) shows the relative comparison of the two phase II metabolism enzymes (liver carboxylase EST1 and EST2) across 13 samples. For comparison, the peptide signal was normalized against sample 091. A good correlation of quantitative differences was observed across multiple peptides of each of the two proteins, demonstrating that good reliability was observed in the SWATH™ Acquisition data.

Because of the comprehensive nature of SWATH acquisition, it overcomes some common limitations existing in the MRM based targeted methods, such as limited multiplexing capabilities and the fact that only targeted analytes will be detected and quantified. Unlike a targeted method, SWATH acquisition creates a permanent record of MS and MS/MS spectra of all detectable species in the sample, and all these information can be extracted from the data. In addition, the data can be further interrogated

when more information arises and a researcher wants to ask more questions of the study.

Also because of the high information content of the technique, statistical analysis and visualization strategies are important. PCA analysis can provide one way of quickly detecting proteins that are differing between the samples. The scores plot (Figure 4, top left) shows how the samples are different from each other and the loadings plot (Figure 4, top right) shows which specific features are responsible for this difference. Proteins of interest can then be selected and viewed for their changes across the whole sample set (Figure 4, bottom). Here, the expression differences between two members of the Cytochrome P450 enzyme family are displayed. The protein expression of these two proteins track each other fairly closely across the samples except for a few individuals were differences are observed.

HIG

H R

ESOLU

TION

AC

CU

RA

TE MA

SS

% R

esp

on

se


Sample (by index)

100

80

60

40

20

0

CYP3A4

UGT1A6

CYP3A4

HSD17B13

GSTK1

UGT2B7

UGT2B15

098I2GDAD IDE KMI

MRS

NQTRML

ROESED

VCM

YAA

091

% R

esp

on

se

091 098 098 DAD I2G I2GDE IDE KMI MRS MRS NQT RML RML ROE SED SED VCM YAA

Sample (by index)

100

80

60

40

20

0

098

I2GDAD

IDE

KMI MRSNQT

RML

ROESED VCM

YAA091

% R

esp

on

se


Sample (by index)

100

80

60

40

20

0

098I2GDAD IDE KMI

MRS

NQTRML

ROESED

VCM

YAA

091

% R

esp

on

se


Sample (by index)

100

80

60

40

20

0

098

I2G

DAD

IDE

KMIMRS

NQT

RML

ROE

SED

VCM

YAA

091

Figure 6. Visualize Protein Profiles for Different Expression Behaviors. Using the PCA analysis in MarkerView™ Software, proteins that show similar correlated expression patterns as well as anti-correlated patterns can be readily found (Left). Using the pattern of CYP3A4 enzyme for comparison, the UGT1A6 enzyme was found to show a similar protein expression pattern while the HSD17B13 protein pattern was found to have a more opposite behavior to CYP3A4. On the right of the figure, two proteins UGT2B7 and UGT2B15 show very similar patterns to each other and also high variation across the individual samples. In contrast, the GSTK1 protein shows very minimal variation across individual samples.

% R

esp

on

se


Sample (by index)

100

80

60

40

20

0

CYP3A4

CYP3A5

098I2GDAD IDE

KMI

MRS

NQTRML

ROE SEDVCM

YAA

091

Figure 5. Specific Quantitation of Closely Related Protein Isoforms. Three peptides that uniquely distinguish the closely related 3A4 and 3A5 CYP450 enzymes were summed to represent protein intensity. The absolute signal intensities were used to compare protein expression across the samples.



HIG

H R

ESO

LUTI

ON

AC

CU

RA

TE M

ASS Mining Data for Interesting Protein Expression Patterns

The quantitative profiles of the phase I and II enzymes are needed to facilitate drug development, understanding which proteins show higher or lower biological variation is important to understand and therefore a straightforward assay for measuring this is an advantage. Correlation as well as variation in protein expression can be easily assessed after PCA analysis (Figure 4). In the results, multiple proteins were found to be correlated, such as CYP3A4 and CYP3A5, which are two major phase I drug metabolizing enzymes in cytochrome P450 superfamily (Figure 5). As their sequence homology is very high, >80%, the quantitation from the SWATH acquisition data was performed using unique peptides to each protein isoform. Previously published data also demonstrated the positive correlation between these two proteins2. In addition, a higher correlation between UDP glucuronosyltransferases UGT2B7 and UGT2B15, and between CYP3A4 and UGT1A6 was observed (Figure 6, top left and right).

As PCA is a multivariate statistical analysis, the anti-correlation of proteins can also be discovered. Figure 6 (bottom left) displays an anti-correlated behavior between CYP3A4 and HSD17B13. At this moment, it is not clearly understood what biological events are behind this observation. But either anti-correlated or correlated behavior of inter- and intra- protein family isoforms may indicate their networking relations or metabolic activities.

In addition to the proteins discussed above, proteins were observed with minimum population variation. As shown in Figure 6 (bottom right), Glutathione S-transferase kappa1 (GSTK1) was consistently expressed across the 13 samples (with ~10% variation). These examples illustrate how much information is present within the SWATH acquisition data for mining and also demonstrate how valuable the quantitative protein expression information is for the evaluation of drug biotransformation during the early drug development.

Good Correlation with MRM Assay Results

After quantitative profiling with SWATH™ Acquisition during the early stages of research, MRM assays to proteins of interest can be quickly developed for better sensitivity and throughput3. As the MRM strategy is the well accepted approach for this type of assay, it was of interest to compare the quantitative results. Skyline software was used to develop an MRM assay for a selected set of proteins and the assay was run on the same set of samples. Comparable results were observed between the MRM results generated with QTRAP® 6500 system and the SWATH™ Acquisition generated with the TripleTOF® 5600+ system as illustrated by a selected peptide for the CYP1A2 protein (Figure 7).

Conclusions

MS/MSALL with SWATH™ Acquisition provides a powerful acquisition strategy for the quantitative profiling of a large number of proteins key in the investigation of drug metabolism.

• Multiple drug metabolizing enzymes were quantitatively profiled in a

single assay across multiple samples.• As the SWATH™ Acquisition requires very little method development, it

is easy to establish and use.• Principal Component Analysis (PCA) provides one method of statistical

analysis that makes data interpretation easier.• With the SWATH acquisition approach, comparable quantitation results

were obtained as with the MRM strategy.

References1. MS/MSALL with SWATH™ Acquisition - Comprehensive Quantification with

Qualitative Confirmation using the TripleTOF® 5600+ System. SCIEX Technical Note 3330111-03.

2. Lin Y. S. et al., (2002) Co-Regulation of CYP3A4 and CYP3A5 and Contribution to Hepatic and Intestinal Midazolam Metabolism. Mol. Pharmacol. 62: 162-172.

3. Discovery to Validation: Transition from SWATH™ Acquisition to Targeted MRM Analysis for Quantitative Proteomics Pipeline - Using the TripleTOF® 5600+ System and QTRAP® 6500 System. SCIEX Technical Note 7940213-01.


HIG

H R

ESOLU

TION

AC

CU

RA

TE MA



Quantification of Large Oligonucleotides Using High Resolution MS/MS on the TripleTOF® 5600 SystemThomas Knapman1, Vicki Gallant1 and Martyn Hemsley2

1SCIEX, Warrington, UK, 2Covance, Harrogate, U

Key Challenges of Oligonucleotide Bioanalytical Assay

1. The bioanalysis of oligonucleotides as therapeutics requires sensitive, specific and robust analysis.

2. Many ELISA-based oligonucleotide measurements do not accurately distinguish large metabolites from the full-length oligonucleotide of interest.

3. ELISA- and UV-based measurements have limited dynamic range which complicates quantitative analysis of oligonucleotides in complex matrices.

Key Benefits of MRMHR Workflow for Oligonucleotide Bioanalytical Assay

1. High sensitivity MS/MS enables the quantitative MRMHR

workflow, providing high selectivity in biological matrices.

2. High resolution, accurate mass MS/MS spectra enable qualitative verification of oligonucleotide sequences.

3. The MRMHR workflow offers a dynamic range of two to three orders of magnitude.

Unique Features of MRMHR Workflow on TripleTOF® 5600 System

1. Summing of multiple ion transitions to increase both sensitivity and selectivity of quantitation.

2. Accelerated method development times, since ion transitions can be selected post-acquisition to eliminate background interferences.

3. High multiplexing due to high acquisition speeds (up to 100 spectra per second) for simultaneous quantitation of multiple species, including multiple oligonucleotide sequences and/or their metabolites.

Introduction

Quantitative analysis of synthetic oligonucleotides in biological matrices is an important aspect of pharmacokinetic (PK), toxicokinetic (TK) and metabolic pathway studies in drug development1. With an increasing number of oligonucleotide based drugs in research pipelines, the acceleration of the drug development process by reducing the time spent on method

Figure 1: The TripleTOF® 5600 System has the speed and sensitivity to deliver high-throughput targeted quantitation of many species in a single run. This example focuses on using MRMHR workflow to quantify a synthetic oligonucleotide from human plasma, demonstrating post-acquisition fragment ion selection and summing product ions to achieve the highest possible sensitivity and selectivity.

HIG

H R

ESO

LUTI

ON

AC

CU

RA

TE M

ASS

!

!

Inte

nsi

ty 339.1952

306.0494

743.3495762.3451

529.2072

761.3530

763.3511744.3533

770.0837

908.15041043.1871

200 400 600 800 1000

0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

Inte

nsi

ty

Inte

nsi

ty

Time, min

Mass/Charge, Da

Time, min

20

15

10

5

0

25

20

15

10

5

0

180

160

140

120

100

80

60

40

20

0

3.683

0.7311.228 2.913 3.375 3.856

Figure 2: Principle of MRMHR Workflow Quantitation. Looped full scan MS/MS spectra are acquired for each precursor. Selected fragments are then extracted post-acquisition using narrow extraction widths (in this case 50 mDa) to produce high resolution XICs. These XICs can be used individually or summed, depending on which provides the best selectivity and sensitivity for quantitation.


HIG

H R

ESOLU

TION

AC

CU

RA

TE MA

SS

development, and by performing simultaneous qualitative structural analysis with quantitative analysis are crucial advantages in any potential quantitation approach1.

Current LC/MS approaches to oligonucleotide quantitation predominantly use multiple reaction monitoring (MRM), however the complex fragmentation pathways of oligonucleotide species coupled with the variability of matrix effects mean that it can be difficult to predict the sensitivity and selectivity of a given MRM transition without significant optimization1. These effects limit the utility of low resolution quantitation methods both in terms of the achievable limits of quantitation and in sample throughput, particularly when quantifying large numbers of potential drug candidates of different sequences, and their metabolites.

Materials and Methods

Sample Preparation

The synthetic DNA Oligonucleotide 1 was spiked into human plasma over a concentration range of 0.025 to 10 nM. Oligonucleotide 2 was used as an internal standard.

LC Conditions

LC System Shimadzu Prominence XR UFLCAnalytical column Waters Acquity BEH, 50 x 2.1 mm,

1.7 µm, temp.= 60 ºCAnalytical flow 0.40 ml/min (initial 24 hour flush)Mobile Phase A Water (15 mM TEA, 400 mM HFIP)Mobile Phase B 50:50 Methanol:Water (15 mM TEA,

400 mM HFIP)

Gradient conditions

Time (min) Mobile phase A% Mobile phase B%

0.5 75 25

5 50 50

5.1 0 100

5.5 0 100

5.6 75 25

6 75 25

MS Conditions

MS System TripleTOF® 5600+ system with a DuoSpray™ Source

Ionization Mode ESI with Negative ModeTOF MS range m/z 100-2500 at 250 msec

accumulation timeMRMHR 2 product ions each 250 msecCollision energy spread -40 ± 4 eVSource temperature 550°C

Software

Data acquisition Analyst TF® 1.5.1 SoftwareData review PeakView® 1.2 SoftwareDeconvolution BioAnalyst® SoftwareQuantitation MultiQuant™ 2.1 Software

Results and Discussion

MS and MS/MS Analysis of Oligonucleotides. TOFMS analysis of Oligonucleotide 1 showed a charge state envelope consisting of [M-5H]5-, [M-6H]6- and [M-7H]7- ions (Figure 3) with a resolution of approximately 36,000. Inspection of the TOFMS spectrum showed that the system passivation process had reduced adduct formation to less than 5% relative to the fully protonated form, thus facilitating quantitation from the [M-6H]6- peak (data not shown).

The principle of the MRMHR workflow for quantitation is to acquire full scan TOF MS/MS spectra for each species of interest, and to use high resolution extracted ion chromatograms (XICs) for quantitation, summing multiple transitions where appropriate to achieve optimum sensitivity and selectivity (Figure 2). To develop an MRMHR workflow assay for oligonucleotides 1 and 2, full scan MS/MS spectra were acquired for m/z 761.9 for Oligonucleotide 1, and m/z 745.6 for Oligonucleotide 2. The full scan MS/MS spectra were also used to verify the sequence of Oligonucleotide 1 and 2 (Figure 4). The MS/MS spectra were deconvoluted using BioAnalyst® Software to enable the singly and multiply-charged fragment ions to be plotted on a mass scale. The sequences were subsequently verified by matching theoretical sequence ions to the fragment ions observed in the deconvoluted spectra.

600 700 800 900

Inte

nsi

ty

652.8268

761.9676

761.4653

761.6334

761.8009

762.1338

762.3002

761.4653 914.1618

914.5622

914.3628

913.9599

914.7624

652.6840

Mass/Charge, Da

761.0 761.5 762.0 762.5 763.0 763.5

Inte

nsi

ty

20000

15000

10000

5000

0

20000

15000

10000

5000

0

761.6334

761.8009

761.9676

762.1338

762.3002

762.4681

Mass/Charge, Da

Figure 3: TOFMS Analysis of Intact Oligonucleotide 1 acquired on the TripleTOF® 5600 system (top). Isotope resolution of higher charge states is achieved with a resolution of 30,000-40,000, as illustrated by zooming in on the 6- charge state (bottom). This resolution is achieved at TOF MS scan speeds of 10 MS spectra per second.



MRMHR Workflow Assay Development. Having acquired full scan TOF MS/MS spectra, fragment ions can be selected and extracted post-acquisition for use in quantitation. Figure 2 shows the stepwise process of selecting fragment ions from a full scan MS/MS spectrum, generating multiple high resolution XICs from the selected fragment ions, and summing the XICs to optimize the signal-to-noise. For transitions arising from Oligonucleotides 1 and 2, an extraction width of 50 mDa was used to generate XICs; however it is possible to collect the MS/MS spectra in high resolution mode (>30,000 resolution) and extract with narrower windows to improve selectivity in any given assay if required.

The post-acquisition selection of fragments is a significant advantage of the MRMHR workflow, since the selectivity of specific fragment ions in matrix cannot necessarily be predicted prior to data acquisition. In the case of Oligonucleotide 1, the three most intense fragment ions in the MS/MS spectrum gave poor selectivity when extracted (Figure 5), and therefore could not be used for quantitation. In contrast, other less intense transitions showed excellent selectivity, and were subsequently included in the assay. The final XIC trace was achieved by summing 25 different fragment ion XICs (Figure 5). The final data processing using the summed XICs was performed using MultiQuant™ Software. TOF MS vs. MS/MS Quantification Strategies. The MRMHR

workflow for quantitation from complex matrices offers significant advantages over the full scan TOF MS approach to quantitation, due to both the high selectivity of the MS/MS based XICs and the ability to remove fragment ions with background interferences, which can impact significantly on the achievable limits of detection and quantification.

600 1000 1500 2000 2500

Inte

nsi

ty, c

ps

m/z, Da

4000

3000

2000

1000

0

4369

600 1000 1500 2000 2500 3000 3500 4000 4500

Inte

nsi

ty, c

ps

Mass, Da

1.5e4

1.0e4

5000.0

0.0

1.8e4

0.5 1.5 2.0 3.5 4.5

Inte

nsi

tyIn

ten

sity

Inte

nsi

tyIn

ten

sity

Time, min

300

250

200

150

100

50

0

0.5 1.5 2.0 3.5 4.5Time, min

25

20

15

10

5

0

0.5 1.5 2.0 3.5 4.5Time, min

400

300

200

100

0

0.5 1.5 2.0 3.5 4.5Time, min

150

100

50

0

TOF MS

0.5 nM

0.25 nM

0.1 nM 0.1 nM

0.2 nM

0.5 nM

MRMHR Workflow

0.5 1.5 2.0 3.5 4.5

Inte

nsi

ty

Time, min

500

400

300

200

100

0

0.5 1.5 2.0 3.5 4.5

Inte

nsi

ty

Time, min

100

80

60

40

20

00.5 1.5 2.0 3.5 4.5

Inte

nsi

ty

Time, min

600

400

200

0

0.5 1.5 2.0 3.5 4.5

Inte

nsi

ty

Time, min

40

30

20

10

00.5 1.5 2.0 3.5 4.5

Inte

nsi

ty

Time, min

300

200

100

0

0.5 1.5 2.0 3.5 4.5

Inte

nsi

ty

Time, min

150

100

50

0

Figure 4: High Quality MS/MS of Oligonucleotide 1 for Characterization and Quantitation Method Development. (top) The full scan MS/MS approach of MRMHR workflow enables full qualitative analysis of any targeted analyte. Full sequencing of Oligonucleotide 1 was achieved on the mass reconstructed MS/MS spectrum (bottom) for fragment ion selection.

Figure 5: Post-Acquisition Extraction of Structurally Specific Ions. In the case of Oligonucleotide 1, the three most intense fragment ions (top left) are non-selective in plasma at low concentrations and therefore summing of these XICs does not provide a good assay (top right). Because the full scan MS/MS spectrum is acquired in the MRMHR workflow, this allows different fragment ions to be selected and extracted for quantitation post-acquisition (bottom left, summed bottom right), and therefore requires significantly less method development than traditional MRM approaches.

Figure 6: Selectivity of MRMHR Workflow in Complex Matrices Allows Better LLOQs to be Obtained. In the case of Oligonucleotide 1, background interferences in full scan TOF MS result in higher limits of detection and quantitation (0.5 nM), while the selectivity of the MRMHR workflow allows quantitation of concentrations less than 0.1 nM.

HIG

H R

ESO

LUTI

ON

AC

CU

RA

TE M

ASS


Figure 6 shows a comparison of the TOF MS and MRMHR workflows applied to the analysis of Oligonucleotide 1. In the case of the TOF MS quantitation, each peak in the isotopic envelope of the [M-6H]6- charge state was extracted using a 10 mDa extraction window (Figure 6, left). The extracted ion chromatograms from the TOF MS approach show significant matrix interferences that seriously impact the limit of quantification, which is approximately 0.5 nM using this approach.

In contrast, the specificity of the MRMHR workflow produced significantly lower limits of detection and quantification (Figure 6, right). This improvement is due to the significant reduction in noise from the complex matrix.

Figure 7 shows the calibration plot of Oligonucleotide 1 using the MRMHR workflow with Oligonucleotide 2 used as an internal standard. The %CV values for each concentration analyzed are shown in the embedded table. The correlation coefficient of the response is in excess of 0.99 over ~2 orders of magnitude; the linearity of response at concentrations in excess of 10 nM was not investigated. The lower limit of quantification (LLOQ) using the MRMHR workflow was 0.05 nM (Figure 7), while the LLOQ was 10 fold higher when the same oligonucleotide was analyzed by full scan TOF MS workflow (Figure 6).

Conclusions

1. The TripleTOF® 5600 system offers sensitive, high-resolution analysis of large oligonucleotides, with the opportunity to perform both qualitative and quantitative analysis in a single run.

2. Using a targeted MRMHR workflow, looped full scan TOF MS/MS spectra of a ~4.5 kDa synthetic oligonucleotide were acquired with a resolution of ~16,000 and XICs were generated from specific fragment ions to achieve a highly sensitive and selective quantitative assay.

3. The ability to quantify oligonucleotides from complex matrices with minimal assay optimization offers the opportunity for high-throughput analysis of potential oligonucleotide-based therapeutics. Upfront method development is highly simplified and consists of specifying a theoretical oligonucleotide m/z and a collision energy, the remainder of the analysis being done post-acquisition.

References1. Z. J. Lin, W. Li and G. Dai, J. Pharm. Biomed. Anal. 44 (2007) 330-341.

Note: Thanks to Randy J. Arnold, SCIEX help write and organize this material.


© 2014 SCIEX. The trademarks mentioned herein are the property of SCIEX Pte. Ltd. or their respective owners. SCIEX™ is being used under license.

Publication number: : 9480114-01

HIG

H R

ESOLU

TION

AC

CU

RA

TE MA

SS

Are

a R

atio

1 2 3 4 5 6 7 8 9

Concentration Ratio

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

Calibration for Oligo 1: y= 0.09896 x + 0.00175 (r=0.99943) (weighting: 1/x)

Conc (nM)

0.025

0.05

0.1

0.25

0.5

2.5

5

9

10

% CV

26.71

3.12

7.67

2.52

1.48

0.73

0.28

0.91

4.85

Figure 7: Standard Concentration Curve for Oligonucleotide 1 in Matrix using MRMHR Workflow. Concentration curve for Oligonucleotide 1 in plasma, using Oligo-nucleotide 2 as an internal standard. Using MRMHR, excellent linearity was observed, with a lower limit of quantitation of 0.05 nM.



Handling Reference Standard Material

The Oligonucleotide API (active pharmaceutical ingredient in GMP Studies) or test article (in GLP Studies) is highly hygroscopic and the water content of an oligonucleotide is related to its environment. Additionally, concentrated solution such as those observed in a clinical setting can be very viscous; both of these traits can lead to inaccuracies in the Assay. To prevent inaccuracies due to the hygroscopic nature of the API, one of two options are presented. The drug substance can be handled only in a room that enables precise temperature and humidity control. The glaring limitation of this approach is that not every site has this available. The alternate option is to equilibrate the API to ambient temperature and humidity prior to handling and using the UV absorbance (typically at 260 nm) and purity to determine the actual concentration. The latter approach assumes knowledge of the oligonucleotide purity. The viscosity issue can be overcome by relying on a gravimetric technique as opposed to a volumetric one.

Dealing with Non-Specific Binding

Oligonucleotides are prone to non-specific binding with the container/closure system, components in biological matrices and components of the LCMS system used for analysis. Non-specific binding to sample containers are more pronounced at lower concentrations and can be helped by either storing the solutions in Type I glass containers or using EDTA or other preservative with plastic container. Regardless of oligonucleotide concentration EDTA is a useful preservative as it will chelate divalent cations which are required for nuclease activity. To prevent non-specific binding to components of the LCMS system, one option is to replace as much stainless steel as possible and replace with PEEK or similar tubing; of course this is only applicable to relatively low pressure systems as the PEEK tubing will not tolerate as high a pressure. Another little trick is to add a minute amount (µM quantities) of EDTA to the mobile phase again to chelate any divalent cations present. Dealing with the non-specific binding of oligonucleotides to components in the biological matrices is more involved and intricate given that oligonucleotides carry lots of negative charge on their phosphodiester-based backbone.

The negative charge imparts a strong affinity for ubiquitous cations such as Na+ and K+ as well as other matrix components.

Sample Extraction Tips

There are many ways to separate these matrix components from the therapeutic oligonucleotide including protein precipitation, liquid-liquid extraction, solid phase extraction and various combinations of these procedures. Simple protein precipitations with organic solvents such as acetonitrile are met with limited success as they are prone to low recoveries and the ubiquitous cations are not necessarily removed.

Several liquid-liquid extraction techniques have met with some success, particularly using a phenol/chloroform solution for the LLE. Often, the LLE will include a step that adds a detergent or other modifier to break up any complexes between the oligonucleotide and matrix components. The proteins will partition into the organic layer while the oligonucleotide is left in the aqueous portion along with other polar matrix components. These remaining matrix components can and do interfere with LCMS. For this reason, the liquid-liquid extract can be further treated by solid phase extraction. Adsorption of the oligonucleotide to a reverse phase solid packing material can be enhanced by the addition of a modifier to both the loading and elution solvent; typically an ion pairing agent and/or ammonium hydroxide are used. Once the oligonucleotide is adsorbed onto the solid phase, washing and elution. This combination of LLE followed by SPE is laborious, time consuming and the ability to automate for a larger number of samples is limited. A few years ago Phenomenex came out with their Clarity OTX system which is a mixed-mode (weak anion exchange and reverse phase) SPE cartridge along with buffers designed to work with the cartridge to clean up oligonucleotide samples. The Clarity OTX offers a quicker method for sample preparation from biological matrices with adequate recoveries.

HIG

H R

ESO

LUTI

ON

AC

CU

RA

TE M

ASS Considerations for Handling Therapeutic

Oligonucleotide Reference Standard and Sample Extraction TipsCindy Sanderson

SCIEX, Framingham, USA


HIG

H R

ESOLU

TION

AC

CU

RA

TE MA



Increasing LCMS Assay Robustness through Increased Specificity using High Resolution MRM-like AnalysisMRMHR Analysis using the TripleTOF® 5600 System

Christie L Hunter1, Hasmik Keshishian2, Steven A. Carr2

1SCIEX, CA, USA; 2Broad Institute, MA, USA

Interference in targeted quantitative assays can come from different sources and can degrade the quality of the quantitative data and ultimately the quantification limits. Many assays are developed using a single sample or a small pool, however when the assay is used on a larger sample set, interferences are discovered that confound the results. When running MRM assays on human biological fluids or tissue, it is imperative that MRM assays are developed to be robust to the presence of interferences that are possible and likely. Robustness to interference can come in two forms: the assay can be robust to the detection of interferences or the assay can be robust by avoiding interferences. Robustness to the detection of interferences can be achieved by designing assays to contain multiple peptides per protein and multiple transitions per peptide so that any interfered peaks can be removed from the final dataset. Robustness to interference through avoidance can be achieved by increasing the assay specificity.

Recent innovations on the TripleTOF® 5600 System have enabled a new acquisition strategy, where looped MS/MS spectra are collected at high resolution and then fragment ions are extracted post-acquisition to generate MRM-like data. The technique is sensitive and fast enough to enable quantitative performance similar to high end triple quadrupole instruments. But the resolution capabilities are far greater, thereby enabling a degree of selectivity that cannot be reached using standard MRM on a triple quadrupole platform. This workflow on the TripleTOF® 5600 System is termed the MRMHR workflow.

To understand the ultimate utility of this type of acquisition strategy on complex biological samples, a set of peptides dosed into depleted human plasma. Injections were performed in triplicate on the TripleTOF® 5600 System and the lower limits of quantification (LLOQ) using various fragment ions to the peptides were measured. Post-acquisition extractions of varying width were analyzed and the impact on specificity and sensitivity were examined.

HIG

H R

ESO

LUTI

ON

AC

CU

RA

TE M

ASS

Inte

nsi

ty, c

ps

Mass/Charge, Da

Time, min

Tim

e, m

ins

300 500 700 900

23.0 24.0 25.010000

0

Figure 1. Looped High Resolution MS/MS – MRMHR. In this analysis, the instrument is set up to acquire full scan MS/MS data on a fixed precursor, over and over again across the LC run. The Q1 is fixed, the peptide is fragmented in the collision cell and the full scan TOF MS/MS is acquired. After data acquisition, extracted ion chromatograms on sequence specific ions are generated. Multiple fragment ions can be monitored or even summed together because the full scan MS/MS is always acquired. After this extraction, the processing for quantification is now very similar to how one would handle MRM data acquired on a triple quad or QTRAP® system.


Methods

Sample Preparation: Human plasma was depleted of the top 14 proteins using a MARS14 depletion cartridge. The sample was reduced, akylated, and digested, providing a solution of 0.9 µg/µL plasma to act as the matrix for the experiment. Samples for the response curve were generated as previously described1,2, using the same set of peptides as in those studies. C12 versions of the peptides were spiked into plasma at varying amounts covering the concentration range of 0.005 -250 fmol/uL, while C13 stable isotope labeled versions of the corresponding peptides were dosed at fixed 5 fmol/uL level. For each sample, a 1µL injection was performed.

Chromatography: The sample was analyzed using the Eksigent nanoLC-Ultra® 2D System combined with the cHiPLC®-nanoflex system in Trap-Elute mode. The cell lysate was loaded on the cHiPLC trap (200 µm x 500 µm ChromXP C18-CL, 3 µm, 300 Å) and washed for 10 mins at 4 µL/min. Then, an elution gradient of 5-35% acetonitrile (0.1% formic acid) in 45 mins was used on a nano cHiPLC column (75 µm x 15 cm ChromXP C18-CL, 3 µm, 300 Å). Trap and column were maintained at 30 ºC for retention time stability.

Mass Spectrometry: Eluant from the column was sprayed using the NanoSpray® Source and heated interface into a TripleTOF® 5600 system (SCIEX). The acquisition method was set up with two acquisition time periods with 18 looped MS/MS (high sensitivity

mode) across them. Each had 200 msec accumulation time. The Q1 and CE were defined for each targeted peptide and Unit resolution isolation was used on the Q1 quadrupole.

Data Processing: All data were processed using MultiQuant™ Software 2.0. Fragment ion extractions were performed at 0.7 Da and 0.02 Da widths and the LLOQs were compared.

High Resolution MS/MS Spectra

Full scan MS/MS spectra can be acquired on the TripleTOF® 5600 System using either a high sensitivity (resolution > 15 000) or high resolution mode (resolution >30 000). Depending on the specificity required, the data can also be extracted post-acquisition using variable width windows. In this experiment, looped MS/MS spectra was acquired in high sensitivity mode, typical fragment ion peak resolutions of around 20 000 were observed. After acquisition, two different XIC widths were used on the same dataset, 0.7 Da to simulate what quadrupole resolution would look like, and 0.02 Da to investigate the effects of higher resolution.

Easy Assay Development for MRMHR

When performing MRMHR quantification, full scan MS/MS is acquired all the time; so only the Q1 and collision energy must be determined upfront of collecting the dataset. It is during the post-acquisition data processing where the optimal fragment ions will be selected for use. Many fragment ions can be assessed after data collection for quantitative performance, such as lack of interference, best sensitivity, best combination for summing, etc.

As an example of interference assessment, multiple fragment ions for PSA peptide LSEPAELTDAVK were investigated for presence of interferences by using the blank matrix injection. The internal standard heavy peptide is present in the blank and XICs to heavy peptide fragment ions can be used to determine the elution time of the peptide (Figure 4, top). From here, the same elution

HIG

H R

ESOLU

TION

AC

CU

RA

TE MA

SS

Inte

nsi

ty

Mass/Charge, Da

m/z, Da

Resolution – 22 400

m/z, Da

500 1000

448.0 449.0 450.0 451.0 963.0 964.0 965.0 966.0

200

100

100

0

Resolution – 22 800

449.1902

289.0948133.0423

612.2523

747.3465

715.34011092.5103

1252.5404

863.4083

964.4534449.1903

985.4558

964.4531

y4

y5

y6

y7y8

y9

y10

14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 330

Inte

nsi

ty

20x

Time, min

MBP

.YLA

SAST

MD

HA

R

Myo

.LFT

GH

PETL

EK

PSA

.IVG

GW

ECEK

CRP

.ESD

TSY

VSL

K

HRP

.DTI

VN

ELR

HRP

.SSD

LVA

LSG

GH

TFG

K

PSA

LSEP

AEL

TDA

VK

APR

.AG

LCQ

TFV

YG

GC

R

CRP

.GY

SIFS

YATK

Figure 2. LCMS of Peptides in Plasma. Nine synthetic peptides and their C13 stable isotope labeled counterparts were monitored in human plasma. Good separation was achieved between the peptides such that two acquisition time periods were used. Very good retention time stability was achieved and peak shapes on all peptides were of very good quality. Looped high resolution MS/MS scans (MRMHR scans) were developed to all 18 peptides.

Figure 3. High Resolution Extracted Ion Chromatograms of Fragment Ions. In this experiment, MS/MS spectra were acquired in high sensitivity mode (resolution > 15000). To understand the effects of resolution on specificity for the fragment ions, post-acquisition XIC extraction was done at either 0.02 Da or 0.7 Da window width.



HIG

H R

ESO

LUTI

ON

AC

CU

RA

TE M

ASS

time for the light peptide fragment ion XICs can be analyzed (Figure 4, bottom). In this case, y8+, y9+ and y102+ fragments are determine to be free of interferences in the matrix blank, while the other three fragment ions show a small amount of interference at 20.1 mins that could impact the lower limits of quantification (LLOQ) for this peptide in matrix.

Again, because the underlying data is full scan MS/MS data, any number of individual fragment ions can be evaluated using MultiQuant™ Software for quantitative performance on the dataset. The standard concentration curves can be assessed to determine which fragment ions provide the best LLOQs (Table 1). Also, the fragment ions can be evaluated across the biological samples to determine which remain free of interferences across the sample set. This flexible post-acquisition processing is only possible because of the full scan data.

When the noise is lower across a number of fragment ions, summing can sometimes improve observed LLOQs. Better ion statistics of summed signal can improve reproducibility and therefore LLOQs. Post-acquisition data processing decisions can be made because the full scan MS/MS data is always present. This can be different for every peptide and every fragment ion, therefore needs to be determined during data processing through exploring the different combinations. Table 1 shows an example for the AGLCQTFVYGGCR peptide for Aprotinin protein. Multiple fragment ions were evaluated for their individual sensitivity and then a subset were summed together. For this peptide, the summed case provides a slight improvement in the LLOQs obtained.

Impact of Fragment Ion Extraction Resolution on Specificity

In MRM analysis, both the Q1 and Q3 quadrupoles are set to transmit windows of ions of about 1Da wide. With MRMHR, the fragment ions are analyzed as much higher resolution and therefore the post-acquisition XICs can be generated with very tight extraction windows of the user’s choice. To assess the impact of tighter resolution on specificity and sensitivity, the same dataset was analyzed using both wide 1 Da window extraction (to simulate MRM specificity) and narrow 0.02Da window extraction.

For one of the PSA peptides analyzed, the y7 ion is the most dominant fragment ion and therefore should provide the most

sensitive quantitative data. However, the wide extraction analysis of this y7 ion (Figure 5 top) yielded a substantially worse 40 fold lower LLOQ than the narrow extraction case (Figure 5 bottom). Because the full scan data is stored, the MS/MS spectra can be investigated for the source of the interference. The full scan MS/MS at the point of peptide elution is shown in Figure 6 (bottom). Zooming in on the y7 ion m/z, it is clear that there is another fragment ion from another co-eluting peptide that has a very similar mass that cannot be resolved with quadrupole resolution (Figure 6 middle). However, if a very narrow extraction window is used, the y7 ion for PSA peptide can be selectively extracted to provide clean quantification data (Figure 6 top).

When there is little background or competing interferences, the high resolution extraction will not have a significant benefit. This is illustrated by the HRP peptide DTIVNELR (Figure 7). Using the definition of LLOQ as better than 20% CV and between 80-120% accuracy, the LLOQs obtained for both the narrow and wide extraction windows are quite similar, 25 amol on column for the 07 Da extraction window and 50 amol on column for the 0.02 Da extraction window. Both peaks look very nice and the background noise is very low. The small improvement in LLOQ in the wider extraction case could be due to the fact that the entire fragment ion peak is extracted at this width providing improved ion statistics through more ion counts and therefore better reproducibility at the lowest level.

Table 1. Advantage of Summing Fragment Ions. For peptide AGLCQTFVYGGCR, four different fragment ions were analyzed both individually and summed.

Heavy

19.93

19.03

20.09

y102+

y92+

y82+

y72+

y6

y8

Light

Inte

nsi

ty, c

ps

Inte

nsi

ty, c

ps

Time, min

19.5 20.0 20.5

19.5 20.0 20.5

1500

1000

500

0

150

100

50

0

Figure 4. Easy Assay Development from Full Scan MS/MS Data. For the blank injec-tion, XICs on fragment ions for both the heavy (top) and light peptides (bottom) for PSA peptide LSEPAELTDAVK were generated. Heavy peptides were added as an internal standard and good signal at 19.9 min was observed. As there was no light peptide added in the blank, minimal signal should be observed. However, signal was observed in the y92+, y82+, y72+ ions for the light peptide indicating the presence of interferences that will impact the LLLOQs for this peptide. These fragment ions can now be avoided in the final data processing.

Peptide FragmentLLOQ on column % CV

% Accuracy R2

APR.AGLCQTFVYGGCRy4, y5, y6, y7,

y8, y9, y1025 14.6 101 0.9998

APR.AGLCQTFVYGGCR y5 50 5.5 97 0.9998

APR.AGLCQTFVYGGCR y7 100 8 97 0.9996



MRMHR Quant – 0.02 Da XIC


HIG

H R

ESOLU

TION

AC

CU

RA

TE MA

SS

0 5 10 15 20 25 30 35 40 45

Concentration Ratio

0

10

20

30

40

50

60

Calibration for IVGGWECEK.y7.Light: y = 1.31107 x + 0.68662 (r = 0.99963) (weighting: 1 / x)

0 5 10 15 20 25 30 35 40 45

Concentration Ratio

0

10

20

30

40

50

60

70

Calibration for IVGGWECEK.y7.Light: y = 1.40048 x + 0.01438 (r = 0.99820) (weighting: 1 / x)

R2 = 0.9980.02 Da XIC

R2 = 0.9990.7 Da XIC

0.7 Da XIC

0.02 Da XIC

Are

a R

atio

Are

a R

atio

Row Component Name Actual Conc Num. V Mean Standard Devia Percent CV Accuracy Value #1 Value #2 Value #3

1 IVGGWECKy7.Light 5.00 0 of 1 N/A N/A N/A N/A 9.591e2


3 IVGGWECKy7.Light 25.00 0 of 3 N/A N/A N/A N/A 8.915e2 9.878e2 5.313e2







10 IVGGWECKy7.Light 10000.00 2 of 2 1.149e4 1.655e2 1.44 114.89 1.137e4 1.161e4



Row Component Name Actual Conc Num. V Mean Standard Devia Percent CV Accuracy Value #1 Value #2 Value #3


2 IVGGWECKy7.Light 10.00 0 of 1 N/A N/A N/A N/A N/A

3 IVGGWECKy7.Light 25.00 0 of 3 N/A N/A N/A N/A 3.681e0 2.330e1 N/A



6 IVGGWECKy7.Light 250.00 0 of 3 2.367e2 1.737e2 7.34 94.69 1.740e2 2.244e2 2.490e2







Figure 5. Higher Resolution Fragment Extraction Can Improve LLOQ. For the PSA peptide IVGGWECEK, the most sensitive fragment ion is the y7 ion. The LLOQ for this pep-tide using a 0.02 Da extraction width was 250 amol on column. However, when a wider extraction window of 0.7 Da was used, the LLOQ was significantly impacted, only 10 fmol on column. Figure 6 illustrates at a spectral level this case.



HIG

H R

ESO

LUTI

ON

AC

CU

RA

TE M

ASS

Conclusions

• Biological matrices are very complex and specificity is an important

factor for assay robustness • TripleTOF® 5600 System has the MS/MS sensitivity and speed to perform

MRM-like analysis• Post-acquisition extraction of fragment ions from the high resolution

TOF MS/MS data allows for high specificity MRM-like data (MRMHR) to

be obtained• MRMHR can provide better quantitative detection limits in the presence

of interferences and increase assay robustness• Selecting best fragment ions or summing multiple fragment ions post-

acquisition can improve detection limits• Assay development is simplified as fragment ion selection is post-

acquisition

References1. Keshishian H et al. (2007) Molecular and Cellular Proteomics, 6, 2212-2229. 2. Addona TA, et al. (2009) Nature Biotechnology, 27, 633 – 641.


© 2016 SCIEX. The trademarks mentioned herein are the property of SCIEX Pte. Ltd. or their respective owners. SCIEX™ is being used under license.Publication number: 2780411-01

Figure 7. Measuring the Impact of Extracted Ion Width on LLOQ for Peptide DTIVNELR. For the wider extraction window of 0.7 Da, the standard curve was linear down to 25 amol on column (left). The background noise was minimal in this example, so when the higher resolution extraction was performed the LLOQ results were very similar (50 amol, right).

25 amol0.7 Da XIC13.3% CV

50 amol0.02 Da XIC10.7% CV

5fmC13, 0.9ug mtrx, 0.025fmC12 3-DTIVNELR_v6.Light (Standar...Area: 1.730e3, Height: 1.479e2, RT: 23.99 min

5fmC13, 0.9ug mtrx, 0.050fmC12 2-DTIVNELR_v6.Light (Standar...Area: 1.670e3, Height: 1.277e2, RT: 24.01 min

Inte

nsi

ty

Inte

nsi

ty

Time, min Time, min23.0 23.5 24.0 24.5 25.0 23.0 23.5 24.0 24.5 25.0

140

120

100

80

60

40

20

0

130

120

110

100

90

80

70

60

50

40

30

20

10

0

23.99

24.01In

ten

sity

, cp

s

Inte

nsi

ty, c

ps Inte

nsi

ty, c

ps

Mass/Charge, Da

Mass/Charge, Da

Mass/Charge, Da

200 400 600 800 1000 1200

864.0 864.5 865.0 865.5 866.0 866.5

865.2 865.3 865.4 865.5 865.6

2400

2000

1600

1200

800

400

0

300

200

100

0

300

200

100

0

233.1291(1)

310.1768(1)532.9292(3)

553.7768

677.8365(2)759.3689(2)

969.4881(1)

0.7 Da

0.02 Da

Peptide

Interference

Figure 6. Investigating MS/MS Interference for PSA Peptide IVGGWECEK. The full scan MS/MS is shown (Bottom). Zooming in on the y7 fragment ion (middle), the spectra from two different injections are overlaid. The blue trace is the injection of 0.5 fmol of the C12 peptide on column, the pink trace is a more concentrated injection of 10 fmol on column. The top pane shows a further enlargement of the fragment ion. The peptide fragment of interest is the signal that increased from 0.5 fmol to 10 fmol. An extraction width of 0.7 obviously is not sufficient to resolve the target fragment from a nearly isobaric interference. An extraction width of 0.02 Da however, is sufficient and provides a clean XIC and a much improved LLOQ.


HIG

H R

ESOLU

TION

AC

CU

RA

TE MA



Application of Differential Ion Mobility Mass Spectrometry to Peptide QuantitationUsing SelexION™ Differential Mobility Separation Technology for better selectivity for peptides in

complex mixtures on the SCIEX Triple Quad™ 5500 LC/MS/MS System

A Zerr, L Meunier, SW Wood, P Struwe

Celerion Switzerland AG, 8320 Fehraltorf, Switzerland

OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS

Key scientific challenges of peptide quant assays

Reduced recovery, low sensitivity – The adsorptive properties and/or polarity of peptides can compromise recovery, and interferences from biological matrices can negatively impact sensitivity and selectivity.

Limited quantitation range – Poor MS/MS sensitivity combined with often poor selectivity can compromise the desired lower limits of quantitation (LLOQ).

Limited MRM selectivity – MRM approaches and efficient UHPLC separations may not provide adequate signal-to-noise ratios at LLOQ due to isobaric interferences or high baseline noise.

Key benefits of differential mobility separation (DMS) for peptide quant assays

Background noise reduction enhances LLOQs – For cases where background noise limits LOQ, DMS provides an additional level of selectivity, orthogonal to the mass spectrometer and LC system.

Better sensitivity even with less refined sample prep – Selectivity gains from DMS permit less selective sample preparations, allowing for overall improvements in sensitivity due to more efficient extractions and better recovery.

Selectivity improvements overcome sensitivity losses – DMS is often accompanied by a loss in absolute sensitivity, but the gains in selectivity improve the potential for real gains in LLOQ.

Key features of SelexION™ differential mobility separation technology for peptide quant assays

Separation of diverse species reduces baseline noise – SelexION Technology separates isobaric species, and single and multiple charge state interferences to reduce background levels and achieve better selectivity and LOQs –while retaining compatibility with UHPLC/MRM speeds.

Simple installation and maintenance – DMS is truly orthogonal to LC and MSMS; installs in minutes with no tools required and no need to break vacuum. Device maintenance is minimal and very straightforward.

Shortened assay cycle – SelexION™ Technology can potentially reduce chromatographic runtimes.

Efficient separation process – Planar geometry results in short residence times, high speeds, and minimal diffusion losses for maximum sensitivity and UHPLC compatibility.

Chemical modifiers for further selectivity – Introducing chemical modifiers to the homogenous fields of the SelexION™ Device cell allows for amplification of the separation capacity and adds a new dimension of selectivity.

SelexION™ Curtain PlateUpdated version of the traditional curtain plate to accommodate the differential ion mobility cell. Maintains the same level of robustness and stability associated with the original design.

Differential Mobility CellCompact and simple design allows the cell to be installed without the use of any tools and in less than two minutes.


Compatible with high-throughput, regulated environments – SelexION™ Technology provides ruggedness and stability to enable high performance quantitative bioanalysis under GLP settings.

Introduction

The quantitative determination of therapeutic peptides to support pharmacokinetic and toxicokinetic studies can sometimes be challenging. Poor MS/MS sensitivity combined with poor selectivity fragments can compromise the desired lower limits of quantitation (LLOQ). In addition, the adsorptive properties and/or polarity of peptides can compromise recovery, and interferences from biological matrices can negatively impact sensitivity and selectivity. In such cases, MRM approaches – even when combined with efficient UPLC separations – may not be sufficient to provide adequate signal-to-noise ratios at LLOQ in the presence of isobaric interferences or high baseline noise.

Differential ion mobility spectrometry (DMS) may provide a useful tool in such instances by providing an additional, orthogonal degree of selectivity. Although DMS analysis is often accompanied by a loss in absolute sensitivity, the gains in selectivity may be sufficient enough to realize real gains in the LLOQ. Alternatively, selectivity gains from DMS may permit a less selective sample preparation to be used, while still delivering overall improvements in sensitivity due to improved extraction recovery.

This poster presents two case studies were the SelexION™ Differential Mobility Separation Device was evaluated for the quantitation of therapeutic peptides. Selectivity, sensitivity and precision of peptide measurements obtained in experiments with and without DMS were compared. The potential benefits and limitations of the technique are discussed.


Sample preparation

Celerion proprietary peptides, A and B, were used for all experiments.

Chromatography

Peptide A-LC system:

Pumps: Series 200 Micro Pump from Perkin Elmer

Auto sampler: Pal CTC from CTC Analytics

Column: Onyx monolithic C18 100x3 mm

Column temperature: Room temperature

Injection: 20 uL

Flow rate: 0.8 mL/min

Mobile phase A: Methanol/Water/Formic acid 5:95:3 v/v/v

Mobile phase B: Methanol/Water/Formic acid 95:5:0.2 v/v/v

Gradient: Time (min) %A %B

0 100 0

0.5 100 0

1.0 35 65

2.0 25 75

4.0 10 90

7.0 0 100

7.5 100 0

Peptide B-LC system:

Pumps: Acquity Binary Solvent manager from Waters

Auto sampler: Acquity Sample Manager/Organizer from Waters Column Ascentis Peptide ES C18 50x2.1 mm 2.7 um

Column temperature: Room temperature

Injection: 20 uL


Mobile phase A: 0.02% Acetic acid aqueous

Mobile phase B: Methanol

Gradient: Time (min) %A %B

0 90 10

1.0 90 10

3.0 5 95

3.1 90 10

4.0 90 10

OR

THO

GO

NA

L SELECTIV

ITY TO

OLS

Figure 1: Optimization of separation voltage. The optimization of SV is performed by constant infusion at low analyte flow in solution whilst ramping the compensation voltage.



OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS

SelexION™ device settings

The SelexION Device needs only a few minutes for installation onto the SCIEX Triple Quad™ 5500 LC/MS/MS System and can be accomplished without breaking the MS vacuum. For the best performance, equilibrating the electrode for 20–30 min at the desired temperature (low/med/high) is required.

Within the device, ions are separated by differential mobility due to an individual molecule’s size and shape. An optimized combination of separation voltage (SV) and compensation voltage (COV) separates the analyte from background ions.

Optimization of these parameters is very simple and can be performed as part of instrument tuning. The optimization of SV is performed by constant infusion at low analyte flow in solution whilst ramping COV (Figure 1). The optimal combination of separation and compensation voltages gives the most separation whilst maintaining maximum peak intensity. Although optimal SV is usually obtained at around 4500 V, a lower value can be chosen to ensure system robustness and stability. As COV is influenced by mobile phase and source conditions, on column COV optimization is performed by injection of analyte at a flow rate and mobile phase composition comparable to the intended LC conditions (Figure 2).

Data processing

Samples were acquired with the Analyst® 1.5.2 Software. Quantification was completed with MultiQuant™ Software.

Figure 2: Optimization of compensation voltage. As compensation voltage (COV) is influenced by mobile phase and source conditions, on column COV optimization is performed by injection of analyte at a flow rate and mobile phase composition comparable to the intended LC conditions. In some instances, additional selectivity may be achieved by use of a modifier (e.g., methanol, acetonitrile, isopropanol, acetone) introduced into the SelexION™ Device at low flow. No modifier was used in the case studies presented.

Table 1: MS/MS, MRM, and SelexION™ Device parameters for peptides A and B on the SCIEX QTRAP® 5500 System.

MS/MS Settings

Peptide A Peptide B

Ion source/polarity ESI/Positive ESI/Positive

CAD High High

CUR 30 30

TEM 700 °C 700 °C

Gas 1 70 50

Gas 2 50 60

Ion Spray Voltage 5500 V 5000 V

MS/MS Settings

Peptide A Labeled IS for A Peptide B Labeled IS for B

Transitions 1029.3/136.0 1106.7/123.0 656.4/249.0 661.4/249.0

Dwell Time (msec) 150 100 100 100

Resolution Q1/Q2 Unit Unit Unit Unit

SelexION™ Device Settings

Peptide A Peptide B

DT (temperature) Low High

DR (throttle gas) Off Off

COV 11.5 15.0

DMO -3 -3

SV 3500 3500

Spiked level of peptide A (ng/mL)

8 20 50 100

Precision (CV%)

without SelexION™

DeviceN/AP 25.0 13.7 6.0

Precision (CV%)

with SelexION™

Device6.3 7.8 8.1 6.3

n 6 6 6 6

Table 2: Lower limit of quantitation (LLOQ) data for peptide A.


OR

THO

GO

NA

L SELECTIV

ITY TO

OLSFigure 3: Peptide A at 20 ng/mL from a protein precipitation extract from human

plasma without DMS.

Figure 6: Peptide B at 0.04 ng/mL extracted with solid phase extraction with DMS.

Figure 7: Peptide B at 0.08 ng/mL extracted with protein precipitation without DMS. Figure 8: Peptide B at 0.08 ng/mL extracted with protein precipitation with DMS.

Figure 4: Peptide A at 20 ng/mL from a protein precipitation extract from human plasma with DMS.

Figure 5: Peptide B at 0.04 ng/mL extracted with solid phase extraction without DMS.



OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS


Case study 1

In this case study, a proprietary peptide (peptide A, MW 4,113.7 g/mol) was evaluated. This peptide is known to exhibit adsorptive characteristics, and solid phase extraction (SPE) cleanup (reverse phase or mixed mode) from human plasma results in very low recoveries. As a consequence, protein precipitation with methanol was the only feasible extraction approach from human plasma. Additionally, due to poor fragmentation, a wide selection of fragments was not available for quantitation, and fragment m/z 136 was the only suitable fragment displaying adequate sensitivity. Under these conditions, the resulting LLOQ is severely compromised by the lack of selectivity, despite separation using a 6 min chromatographic gradient (Figure 3). Without DMS, using MRM with +ESI (Table 1), an LLOQ of only 50 ng/mL is achievable from human plasma. An elevated baseline and a number of closely-eluting peaks were observed for spectra obtained for peptide A, which required careful set-up of peak integration parameters.

Using the same protein precipitation procedure and LC gradient conditions, DMS was added to the workflow (optimized parameters, Table 1), and peptide A spectra were evaluated for any improvement in selectivity. A significant improvement was observed. Despite a loss of absolute signal (approximately a factor of 5), a reduction of background interference of approximately a factor of 20 was observed. This resulted in an overall gain in S/N of an approximate factor of 4-5 (Figure 4). This facilitated a lowering of the feasible LLOQ from 50 to 8 ng/mL (Table 2) without changing extraction or gradient LC conditions.

Case study 2

In this case study the quantitation of a therapeutic peptide (peptide B, MW 1,311 g/mol) in rat plasma was evaluated. This peptide was extracted from rat plasma using polymer-based, reversed-phase SPE. A recovery of 65% was achieved. Samples were chromatographed on a fused-core peptide column using a methanol/water gradient with formic acid as the acid modifier.

Using +ESI-MRM (Table 1), a range of 0.04–10 ng/mL could be routinely achieved. At LLOQ a S/N of 10 (analyte intensity 1,000 counts, background 100 counts) resulted in a precision of 7.4% (Figure 5). Applying DMS to this method resulted in approximately a 10-fold decrease in background with absolute analyte sensitivity exhibiting approximately a 6-fold decrease. Whilst S/N improved to 16 at LLOQ (0.04 ng/mL), there was no marked improvement in precision at this level (Table 3). A lowering of LLOQ could also not be facilitated under these conditions. Background, however, was almost completely eliminated allowing for easier and more consistent peak integration (Figure 6).

A protein-precipitation approach was also evaluated for peptide B. As absolute recovery was compromised using SPE, it was anticipated that a lowering of the LLOQ by a factor of 2 could be achieved by combining recovery gains of protein precipitation with selectivity gains from DMS. Without the benefits of SPE cleanup in this instance, an LLOQ of only 0.08 ng/mL could be achieved without DMS due to high background and closely eluting isobaric interferences (Figure 7). With DMS, all background interferences were removed (Figure 8). However, due to the inherent loss of absolute signal associated with DMS, the resulting LLOQ achieved was limited to the LLOQ demonstrated for SPE extraction.

Conclusions

Differential ion mobility spectrometry provides a useful additional or orthogonal selectivity during the quantitation of peptides (and conventional small molecules). For applications involving peptide quantitation in particular, selectivity gains may be significant as separation of multiply-charged analyte precursors from singly-charged background interference. The true gain in sensitivity as a function of absolute sensitivity and selectivity will be analyte dependant and will also be influenced by choice of MRM transition, chromatographic separation and extract cleanliness

Often for peptides sensitivity is already compromised by a number of factors including low bioavailability, adsorption to surfaces, formation of multiple charge states and poor or non selective fragmentation. In these instances additional tools to aid lowering of LLOQs are to be welcomed. In some cases, true gains in sensitivity may not be realised or required, but improvements in selectivity may bring other benefits – namely, simpler extraction methods, shorter chromatographic runs, or improved peak integration.

References1 A. Zerr, L.Meunier, S.Wood, P. Struwe. Application of Differential Ion Mobility Mass Spectrometry

to Peptide Quantitation. Nov 14–16 2012. Celerion Poster Presentation. EBF 2012 Open

Symposium, Barcelona.

Accuracy (%) Precision (%)

Concentration (ng/mL) + DMS - DMS + DMS - DMS

0.04 100 100 10.2 7.4

0.08 101 101 6.0 4.8

0.2 97.7 97.6 4.7 2.2

0.8 97.0 97.2 2.4 0.9

1.6 102 102 2.2 0.6

10 103 103 1.3 1.2

Linear regression r value

0.9976 0.9987

Table 3: Linearity, accuracy, and precision for peptide B peak area measurements with and without DMS. Intra-run replicates, n=6, for SPE extracts with and without DMS. Concentration curves were analyzed with linear fit and a 1/x2 weighting.


OR

THO

GO

NA

L SELECTIV

ITY TO

OLS



OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS Improving Intact Peptide Quantitation

with Differential Mobility Separation and Mass Spectrometry (DMS-MS)Using SelexION™ Technology coupled with the QTRAP® 5500 System for additional selectivity and

separation of peptides in biological matrices

J. Larry Campbell and J.C. Yves Le Blanc

SCIEX, 71 Four Valley Drive, Concord, ON, L4K 4V8 Canada

Key challenges of large peptide quantitation

Poorly fragmenting peptides – Cyclic fragments often fragment poorly resulting in few product ions for analysis.

Challenging Physico-chemical proprieties of peptides such as non-specific binding, poor solubility, complex charge state envelope makes peptide quantification makes peptide quantification challenging.

Sub pg level sensitivity – Need of very low levels of quantitation (pg/mL range) and lack of target functional groups for sample clean-up.

Key benefits of SelexION™ Technology for peptide quantitation

SelexION™ Technology (differential mobility spectrometry) separates isobaric species, single and multiple charge state interferences and reduces background levels to achieve better selectivity and thereby LOQs – all while being compatible with UPLC/MRM speeds.

SelexION™ Technology is the only ion mobility technology with the ruggedness and stability to enable high performance quantitative bioanalysis under GLP settings.

SelexION™ Technology is truly orthogonal to LC and MS/MS with no tools required and no need to break vacuum to install.

The QTRAP 6500 System enables MRM3 capabilities that provide an order of magnitude of selectivity improvement over standard MRM techniques and can enable faster separations without the need for long LC run times to eliminate/reduce background interference.

Key features of SelexION

Chemical Modifiers – The introduction of chemical modifier adds a new dimension to selectivity and dramatically increases separation capacity.

Planar geometry results in high speed and minimal diffusion losses for maximum sensitivity and UHPLC compatibility.

Extension ringfor coupling of TurboV Source

Curtainplate

DMScell

Orificeplate

(A) (B)

Figure 1: (A) Components associated with SelexION™ Technology mounted on 5500 QTRAP® System. (B) Cross section of the DMS cell showing the introduction of the chemical modifier, typically set at 1.5% (or higher) of the total curtain gas flow supplied. Cell dimensions are 1 x 10 x 30mm (gap height x width x length).


OR

THO

GO

NA

L SELECTIV

ITY TO

OLS

Highly robust, reproducible, and stable for use in regulated bioanalysis.

Easy to maintain, and can be Installed or removed in minutes with no need to break vacuum or use any tools.

Introduction

Differential mobility spectrometry (DMS) can be used to separate isobaric species and co-eluting interferences by filtering selected compounds based upon the difference in the ions’ mobilities during the high- and low-field periods of the applied asymmetric waveform. This waveform, termed the separation voltage (SV) is applied across the gap between the two planar electrodes. Applying a compensation voltage (CoV) ensures the selective transmission of a particular species through the DMS device at a specific SV, while the other, unwanted chemical species in the mixture are filtered out. Peptides tend to have CoVs that are several volts higher than singly-charged ions, and this characteristic provides a way to easily discriminate amongst multiply-charged peptides when using DMS, allowing for detection, with some degree of selectivity, using the single ion monitoring (SIM) mode. Furthermore, the addition of a chemical modifier into the transporter gas significantly alters CoVs, increasing the capacity of DMS to separate these molecules [1-2].

During the drug discovery and development process, it is necessary to separate and quantify peptides from a biological matrix with a high degree of selectivity and sensitivity. For peptides that do not fragment well (such as cyclic peptides) or ones that over-fragment (resulting in too many product ions), analysis by SIM is essential, but often confounded by high background noise and low sensitivity. To overcome these challenges, we applied SCIEX SelexION™ Technology, a planar differential mobility separation device, to a mixture of intact peptides and explored how DMS can selectively filter peptides from chemical noise. In this current work, we show that DMS mass spectrometry provides gas phase separation of peptides, while reducing background noise and co-eluting interferences, improving the selectivity of peptide detection in a manner comparable to MRM detection.


Sample preparation

Two matrices were used to generate chemical noise: 1) protein-precipitated plasma produced using perchloric acid (HClO4) and 2) trypsin-digested plasma. For HClO4-precipitated plasma, plasma (1 mL) was vortexed with 5% HClO4 (v:v) for 5 min. After centrifugation (16,500 g, 10 min.), the resulting supernatant was mixed with 100 μL of 10% ammonium hydroxide (v:v). The precipitated plasma was diluted 2-fold prior to injection on column. Digested plasma was generated according to conventional protocols (overnight trypsin incubation). To eliminate undigested large proteins post-incubation, plasma was precipitated with acetonitrile (1:1 v:v) and diluted 2-fold prior to injection on column. Angiotensin I, angiotensin II, angiotensin IV, neurotensin, dynorphin A, melittin and exenatide were spiked

at different concentration levels in both matrices. Exenatide was obtained from ChemPep (Wellington, FL), and all other peptides were obtained from Sigma Aldrich (St. Louis, MO) and used without further purification.

Chromatography

LC System: Eksigent ExpressHT™ LC system

Column: Poros (1 x 10mm, 7 μM)

Injection: 1 μL (fixed loop)

Flow rate: 175 µL/min



0 98 2

0.5 98 2

5 75 25

6 20 80

7.5 20 80

Mass spectrometry

System: SCIEX QTRAP® 5500 System

System interface: Turbo V™ Source and the reduced volume electrospray ionization (ESI) probe (65μm ID) for peak dispersion minimization.



Curtain gas (CUR): 20

Temperature (TEM): 500 °C

IonSpray voltage

Floating (ISVF): +5500 V

Scan type: Single-ion monitoring (SIM)

Accumulation time: sec for each scan experiment

Declustering potential:

Collision energy: 10 V

Collision energy spread: 0 V

SIM conditions were used for all peptides. Retention time (RT) and CoV values are listed in Table 1. SIM was used to evaluate the additional selectivity that would originate from the DMS cell.



OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS DMS cell conditions

SelexION™ Technology, a planar DMS device, attaches between the curtain plate and orifice plate of the QTRAP 5500 System (Figure 1A). Drawn towards the orifice by the curtain gas as it flows to the MS orifice, ions fluctuate between the flat plates when an asymmetric waveform is applied. An ion’s net drift is based on the difference in the ion’s mobility in the high field (K(E)) and low field mobility (K(0)) conditions of the applied waveform (Table 1). A compensation voltage (CoV), a small DC offset between the plates, is applied as a filtering voltage – and must be optimized for each ion for specific analytical conditions. Isopropanol (IPA), acetonitrile (ACN), and methanol (MeOH) can be used as chemical modifiers and introduced into the transport gas via the curtain gas (Figure 1B), prior to an ion’s entrance into the DMS device, thereby altering the separation characteristics of analytes [1-2]. However, this feature was not evaluated in the present study. The SV and the associated CoV were tuned for each analyte. The DMS can also be used in transparent mode, whereby the SV and CoV voltages are turned off, allowing the MS system to operate in conventional mode (as if the DMS device was not installed).


Elimination of co-eluting, multiply-charged interferences using SelexION™ Technology

Table 1 lists the compensation voltages (CoV) obtained for all peptides under LC conditions. Several peptides have optimum CoVs in the +10.0 to +11.5 V range (typically the highest charge state) or unique CoV values can be found for some of charge states. In many cases, the charge state associated with these unique CoVs is not necessarily the highest possible charge state, but is for the charge state with the highest response. Figure 2 shows the distribution of the optimum CoV observed for each peptide versus the peptide’s m/z – and includes values obtained

for several singly-charged compounds at similar separation voltages, aptly demonstrating that singly-charged species (including peptide-based) can be discriminated from the multiply-charged signals.

To investigate the extent of background noise reduction when using DMS, individual peptides were spiked into protein-precipitated or digested plasma and were monitored in SIM mode.

Figure 3 shows the chromatogram obtained for angiotensin I, II and IV when monitored in SIM mode. In the presence of both matrices, DMS data showed significant baseline reduction over data collected in transparent mode (to mimic a non-DMS-filtered triple quadrupole environment). Peptides spiked into perchloric

Angio I

HCIO4 Precipitated Plasma

DMS ON DMS OFF DMS OFFDMS ON

Digest Plasma

Angio II

Angio IV

Figure 2: Distribution of observed CoV values as a function of mass-to-charge. Charge states are color coded. A separation voltage (SV) of 3500 V was used. Included are values for non-peptide singly-charged species under similar conditions.

Figure 3: Detection of angiotensin I, II and IV in SIM mode under optimized DMS conditions and DMS in transparent mode (mimic regular MS system) in different matrices. For angiotensin 1, z=+4, +3 and +2 are represented by blue, pink and orange trace, respectively. For angiotensin II and IV, z=+3, +2 and +2 are represented by the blue, pink and orange traces, respectively.


OR

THO

GO

NA

L SELECTIV

ITY TO

OLS

DMSTransparentMode

DMSOptimized(SV 3400)

Figure 4: Detection of angiotensin I, II and IV in SIM mode under optimized DMS conditions and DMS in transparent mode (to mimic a non-DMS environment) in different matrices. For angiotensin 1, z = +4, +3 and +2 are represented by blue, pink and orange trace, respectively. For angiotensin II and IV, z = +3, +2 and +2 are represented by the blue, pink and orange traces, respectively.

Figure 5: Selectivity of melittin spiked in digested plasma at different levels when detected by LC-DMS-SIM. Comparing DMS at optimum values (DMS ON) versus DMS operated in transparent mode (DMS OFF). Concentration are in fmol/μL and chromatograms are for the melittin z = +5 ion.



OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS

acid-precipitated plasma were selectively detected using SIM. For all peptides filtered by DMS, both the noise level and extra LC peaks were eliminated from the resulting chromatograms, showing very low baselines for both matrices (Figure 3).

For samples analyzed in digested plasma, most noise in the baseline was reduced for all peptides, except for the z = +2 ion of angiotensin II, which displayed additional LC peaks that were unresolved from the analyte peak (Figure 3). When compared to data collected in transmission mode for angiotensin II, the DMS-filtered data showed a significant reduction in chemical interferences. Because of the high fragmentation efficiency of the angiotensin II z = +2 ion, the combination of DMS with MS/MS detection could increase selectivity for this peptide; however, it is likely that selectivity gains by MS/MS detection would be offset by the loss of signal incurred by high levels of secondary peptide fragmentation during MS/MS.

The fragmentation characteristics of peptides can significantly impact the sensitivity of their detection. Many intact peptides, such as melittin, do not fragment easily and generate low ion counts; DMS filtering is expected to improve the sensitivity of low-fragmenting peptides due to a lessening of chemical noise and the increased selectivity inherent in the DMS method. Chromatograms obtained for the +6, +5 and +4 charge state of melittin spiked in digested plasma show a significant drop in baseline noise and elimination of additional peaks for the +6 and +5 charge states when filtered with DMS (in SIM mode) and resulted in no additional loss of signal (Figure 4). The enhancement of selectivity by DMS improved melittin detection in complex matrices, and required minimal tuning of DMS parameters (SV/CoV). Overall, DMS improves the selectivity of melittin in a concentration-dependent manner, as demonstrated by heightened analyte peaks over baseline when compared to chromatograms obtained without DMS (Figure 5.)

To increase the separation capacity of the DMS, a chemical modifier (e.g., 2-propanol) can be added in the curtain gas (Figure 1B) [1-2]. Preliminary results using chemical modifiers such as 2-propanol (IPA) and acetone for the analysis of intact peptides have not significantly improved the separation capability of the DMS cell (data not shown). In general, the peptide CoVs shift towards a lower-value charge state after exposure to the chemical modifier, but the observed shift (∆CoV) remains the same for all charge states for a given peptide. Occasionally , the increased proton affinity of the chemical modifier can lead to proton abstraction from select peptide charge states, resulting in signal loss. This phenomena can, in some cases, assist in further reducing chemical interference but can also lead to additional tuning requirements (monitoring of additional charge states). This concept will be explored in future work, as well as the effects of adding affinity labels (e.g., SCIEX iTRAQ® Reagents) to further enhance peptide separation by DMS.

Conclusions

Differential mobility spectrometry (DMS) using SelexION™ Technology separated peptides from co-eluting interferences, improving selectivity in a complex biological background by reducing background noise.

Significant baseline reduction was observed for angiotensin I, angiotensin II, and angiotensin IV spiked into plasma when filtering peptides using DMS.

Selectivity for melittin in plasma was improved by DMS, with chromatograms showing fewer interfering peaks and diminishing background noise, even over a range of peptide concentrations (0–100 fmol/μL).

References1 Schneider, B.B.; Covey, T.R.; Coy, S.L.; Krylov, V.E.; Nazarov, E.G., Anal.Chem. 2010 (82)

1867-1880

2 Schneider, B.B.; Covey, T.R.; Coy, S.L.; Krylov, V.E.; Nazarov, E.G, ThOC am 08:50,

Proceedings of 58nd ASMS Conference on Mass Spectrometry and Allied Topics, Salt Lake

City, May 23-27 (2010)

Figure 6: Calibration curve for melittin spiked in digested plasma and monitored in SIM mode for the Z=+5 ion. Duplicate injections were made and a Wagner fit was used.

© 2016 SCIEX. For Research Use Only. Not for use in diagnostic procedures. The trademarks mentioned herein are the property of SCIEX Pte. Ltd. or their respective owners. SCIEX™ is being used under license.

SCIEX SelexION™ Technology

A NEW DIMENSION IN SELECTIVITY

PUSHING THE LIMITS IN SELECTIVITY

Ion mobility spectrometry for quantitative and

qualitative applications

SelexION™ technology on the SCIEX Triple Quad™ 6500 and QTRAP® 6500

systems delivers a new dimension of selectivity and performance

for any application requiring the separation of isobaric species, isolation

of challenging co-eluting contaminants and reduction of high

background noise.

Explore a new dimension at www.sciex.com/selexion



OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS Differential Mobility Separation Mass

Spectrometry for Quantitation of Large Peptides in Biological MatricesUsing SelexION™ Technology Coupled with QTRAP® 5500 System

Derek T. Wachtel, Sanjeev T. Forsyth & Robert W. Busby

Ironwood Pharmaceuticals, Cambridge, MA

Key challenges of large peptide quantitation

Poorly fragmenting peptides – Cyclic fragments often fragment poorly resulting in few product ions for analysis.

Challenging Physico-chemical proprieties of peptides such as non-specific binding, poor solubility, complex charge state envelope makes peptide quantification very challenging.

Sub pg level sensitivity – Need of very low levels of quantitation (pg/mL range) and lack of target functional groups for sample clean-up.

Key benefits of SelexION™ Technology and the QTRAP® 5500 System for large peptide quantitation

Separation based on analyte structure – SelexION™ Technology is a planar differential mobility separation device (DMS) that separates peptides based on difference in their chemical and structural properties.1

More options for selectivity – SelexION™ Technology adds an orthogonal level of separation and selectivity prior to the instrument orifice (Figure 1).1

Compatible with high performance bioanalysis – SelexION™ Technology is compatible with fast cycle times required for monitoring multiple MRM transitions combined with narrow HPLC peaks.

Key features of SelexION™

Chemical Modifiers – The introduction of chemical modifier adds a new dimension to selectivity and dramatically increases separation capacity.

Planar geometry results in high speed and minimal diffusion losses for maximum sensitivity and UHPLC compatibility.

Highly robust, reproducible, and stable for use in regulated bioanalysis.

Easy to maintain, and can be Installed or removed in minutes

with no need to break vacuum or use any tools.

Introduction

The ability to quantify large peptides in a biological matrix with adequate selectivity and sensitivity depends on several factors: 1) the presence of multiple charge states, 2) varied fragmentation efficiency due to size and structure, and 3) the complexity of the matrix to cause varied background and interferences. Tryptic digestion can reduce the size of the peptide and improve fragmentation, but this process introduces an additional step, which is undesirable in a high-throughput environment.

Detection limits for MRM acquisition can be heavily affected by the fragmentation characteristics of the peptide. Sensitivity is reduced if the peptide resists fragmentation or if it fragments too extensively (spreading the ion current across many product ions). Cyclic peptides are especially difficult to detect because they tend to have poor MS/MS efficiency. For these cases, high sensitivity can sometimes be achieved by monitoring the intact

Derek Watchel


Figure 1: High selectivity quantification using SelexION™ Technology on the QTRAP® 5500 System. SelexION™ Technology1 is an easy-to-install differential mobility device available on a QTRAP® 5500 or 6500 System that is used to provide additional selectivity to any quantitative experiment. An asymmetric waveform alternates between high field and low field, and ions will have a net drift towards one of the plates based on their high and low field mobility difference. A compensation voltage (CoV) is applied as the filtering voltage, which is tuned for the compound of interest. Other co-eluting isobaric or non-isobaric species tune with different CoVs and will be filtered away.

OR

THO

GO

NA

L SELECTIV

ITY TO

OLS

peptide in single ion monitoring (SIM) mode. However, SIM methods are not routinely utilized due to the resulting reduced selectivity and high background levels from monitoring only the parent ion without fragmentation.

Differential mobility separation (DMS) mass spectrometry adds an

additional level of selectivity to LC/MS/MS providing gas phase separation of isobaric species and co-eluting interferences to reduce background noise. Here, the utility of a SIM workflow combined with DMS for specificity was investigated for quantifying large therapeutic peptides in a protein-precipitated plasma matrix.


Sample preparation

To extract the peptides from plasma, an aliquot of each standard (100 µL) was mixed with 400 µL of 5% formic acid in acetonitrile (with 150 ng/mL peptide). After vortexing for 10 min, the sample was centrifuged at room temperature for 10 min (21,000 x g). The supernatant (425 µL) was transferred to a clean 96-well plate and evaporated to dryness at 50 ºC. The sample was reconstituted with acetonitrile (30 µL) and shaken for 15 min. After mixing with 2.5% formic acid in water (20 µL), the sample was shaken for an additional 15 min.

Chromatography

LC system: Acquity UPLC System (Waters)

Column: Thermo-Hypersil Gold (2.1 x 50 mm, 1.9 μm)


Injection: 5 μL




0 95 5

0.5 95 5

3 5 95

Mass spectrometry

System: QTRAP® 5500 System

Interface: SelexION™ Technology device and Turbo V™ Source

DMS parameters

To maximize signal intensity for each peptide, the following optimal parameters were used:

DMS temp: low

Modifier: none

Separation voltage: 3800

Compensation voltage: 15.5

DMO offset: -5.5

DMS resolution → off

Figure 2: Multiple reaction monitoring (MRM) scans of peptide PN1944 using the QTRAP® 5500 System. Multiple reaction monitoring (MRM) scans provide a high degree of selectivity for detecting peptides in complex matrices. However, detection of some large peptides is limited by their fragmentation pattern, which ultimately limits the sensitivity of the MRM mode – either the peptide does not fragment well, or the peptide fragments into many product ions. The LLOQ (125 ng/mL) achieved in rat plasma using this approach was not sufficient to achieve detection limits required for the experiment.

Sourceextension

ring

Curtainplate

DMScell



OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS Analysis is done using a single ion monitoring (SIM) acquisition

strategy. Single ion monitoring (SIM) is a variation of an MRM experiment where the parent ion is monitored in Q1 and Q3 without fragmentation in Q2. For large peptides, SIM methods can provide larger signal without the need for extensive compound optimization.

Data processing

Results were analyzed using Analyst® Software 1.5 quantification tools.


High-throughput peptide quantitation

Optimizing the extraction method and chromatographic conditions for every test compound being evaluated is not desirable for high-throughput analyses. To overcome the challenges of evaluating peptide levels in complex biological matrices, a simple extraction method and high-throughput chromatographic conditions were developed and optimized for selectivity and sensitivity for a diverse set of peptides (~30 amino acids, 4 kDa) in multiple lots of rat plasma prior to analysis by LC/MS/MS.

Time-saving, generic MS acquisition methods are attractive in a high-throughput environment, but these methods are often limited by background noise, delivering reduced sensitivity and limiting the ability to generate a suitable pharmacokinetic (PK) profile for a peptide. MRM methods are the gold standard for high sensitivity quantification, but occasionally, the fragmentation properties of large peptides can limit the ultimate sensitivity achieved (Figure 2).

Monitoring the intact form of the peptide using SIM can provide better sensitivity (Figure 3, top), but the resulting spectra have much higher background noise that further impacts detection limits. Combining the higher intensity of SIM with the added selectivity of differential mobility separation (DMS) can provide the necessary selectivity for improving detection limits. Compared to those achieved with SIM alone, the signal-to-noise ratio shows a significant improvement when using DMS to detect peptides due to the substantial reduction of baseline noise (Figure 3, bottom).

Assay Performance using DMS with SIM

The lower limit of quantification (LLOQ) for the DMS + SIM acquisition strategy was 4 ng/mL (Figure 4, top), approximately a 30-fold improvement over the LLOQ from conventional MRM scans (125 ng/mL, Figure 2) . The statistics for the peaks observed at 4 ng/mL are shown in the table (Figure 4); good reproducibility and accuracy were achieved.

Figure 3: Signal-to-noise ratio improves using DMS. (Top) Poor selectivity is observed during the analysis of peptide PN1944 in SIM mode (monitoring precursor m/z in Q1 and Q3 with no collision energy for higher sensitivity). The peptide signal at 16 ng/mL is largely obscured by background noise. (Bottom) Analysis of the same sample with DMS shows greatly improved signal-to-noise for the target peptide.


Figure 4: Lower limit of quantitation for a 4 kDa peptide in plasma using the DMS+SIM high-throughput quantification workflow. (Top) The chromatograms for the blank plasma and the LLOQ using the DMS + SIM approach are shown. (Bottom) The statistics for the measurements at the LLOQ are shown. Using the DMS+SIM strategy provides an ~30-fold improvement in detection limit over the conventional MRM approach.

Theoretical Concentration

(ng/mL) Replicate Sample

Measured Concentration

(ng/mL)

Relative Error (%)

Average Relative

Error (%)

CV (%)

4

1

1 3.5 -12.5

2.2 13.5

2 4.2 5

3 4.84 21

4 5.08 27

2

1 4.17 4.3

2 4.25 6.3

3 4.35 8.7

4 4.3 7.5

3

1 3.86 -3.5

2 3.15 -21.3

3 3.61 -9.8

4 3.75 -6.3

OR

THO

GO

NA

L SELECTIV

ITY TO

OLS

Conclusions

Differential Mobility Separation (DMS) using SelexION™ Technology provides an orthogonal level of selectivity by separating components based on their chemical properties and ion mobility.

Matrix interferences and background noise can be significantly reduced to improve selectivity and thus sensitivity.

An LLOQ of 4 ng/mL for a 30 amino acid peptide (~4 KDa) was achieved in protein-precipitated rat plasma, providing an over 30-fold improvement in the MRM analysis strategy.

The DMS + SIM acquisition strategy combined with a high-throughput sample preparation and LC workflow achieves the desired LLOQ (4 ng/mL) for PK studies of large therapeutic peptides.

References1 SCIEX SelexION™ Technology: A New Solution to Selectivity Challenges in Quantitative Bioanalysis

– Differential Mobility Separations Enhanced with Chemical Modifiers: A New Dimension in

Selectivity. SCIEX Technical Note, Publication 2960211-01.

2 Multiple Mass Spectrometric Strategies for High Selectivity Quantification of Proteins and

Peptides: Solve the Most Challenging Quantitative Problems with QTRAP® 5500 System. SCIEX

Technical Note, Publication 5310212-01.



Key challenges of peptide selectivity

Co-eluting multiple charge interference limits accurate quantification and also peak integration at LOQ levels.

Isobaric interferences will limit selectivity and specificity of the assay and cause issues for accurate identification during bioanalytical method development process.

Key benefits of SelexION™ Technology and the 6500 Series for tryptic peptide quantitation

Background or interfering ions can be tuned out prior to any MS filtering. This leads to a boost in performance through several avenues; LOQ improvement, reduction in background or interfering ions, isobaric separation or an accelerated runtime.

Whether facing the challenge of resolving a chromatographic interference, eliminating a high baseline, or separating isomers, SCIEX SelexION Technology offers a powerful new tool to help the bioanalytical scientist solve tough selectivity challenges.

Key features of SelexION

SelexION™ Technology (differential mobility) separates isobaric species, single and multiple charge state interferences and reduces background levels to achieve better selectivity and thereby LOQ’s – all while compatible with UPLC/MRM speeds.

Excellent inter-day and intra-day precision and accuracy can be achieved with SelexION Technology, to satisfy bioanalytical validation requirements.

DMS is truly orthogonal to LC and MSMS with no tools required and no need to break vacuum to install and provide ruggedness and stability to enable high performance quantitative bioanalysis under GLP settings.

SelexION™ Technology provides the potential to reduce chromatographic runtimes.

OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS UGT Family of Enzymes: Quantification of

Tryptic Peptides using SelexION™ Technology on the QTRAP® 6500 SystemPart 3 of 3: Using SelexION™ Technology on the QTRAP 6500 LC/MS/MS System

for additional selectivity

by separating multiple charge-state ions in tryptic digests

Suma Ramagiri, Loren Olson, Gary Impey, Carmen Fernandez Metzler

SCIEX, 71 Four Valley Drive, Concord, ON, L4K 4V8 Canada Pharma Cadence,

Carmen Fernandez


OR

THO

GO

NA

L SELECTIV

ITY TO

OLS

Introduction

In this tech-note series on improving peptide quantitation, we again use the UDP-glucuronosyltransferase (UGT) enzyme family to highlight an additional selectivity option for the complex analysis of peptides in biological matrices. We previously discussed the importance of UGT enzymes in glucuronidation which have a major role in phase II metabolism – arguably the most important human pathway for elimination of the top 200 drugs. Our studies focus on the quantitation of very low-levels of UGT peptides, providing absolute protein concentrations used to calculate the necessary pharmacokinetic parameters and dosing schedules for drug trials.

LC/MSMS technologies enable highly precise and accurate quantitation of the UGT enzymes, but require advanced analytical performance to reach desirable LOQs. In Part 1, we introduced the 6500 series of instruments that provide up to 5-fold improvement in the LOQ over other MS systems – using the standard MRM approach and microflow LC strategies to assist with UGT peptide detection in the low nanogram/mL level. Then, in Part 2, we investigated how MRM3 scans can offer improved selectivity through a fragmentation pathway specific to the ion of interest – to help eliminate closely, and, even, co-eluting interferences.

As researchers encounter more complex samples, the separation performance and LOQs must be enhanced through a variety of methods. Improvements to standard MRM approaches are limited to adjustments in ion pairs, selection of different CE profiles, modifications in sample preparation and/or chromatography to enhance the signal on analytes of interest. Often these adjustments are very time-consuming and don’t provide significant improvement over the original methodology. QTRAP Technology offers MRM3 capabilities, providing an additional selectivity option for those analytes with a generous spectrum of fragment ions. If the fragment ions demonstrate an inadequate signal for further quantitation or are not specific enough to be advantageous, there is another truly orthogonal technique available on both the QTRAP and triple quadrupole systems – SelexION™ Technology.

In this third and final installment, we highlight a selectivity approach that uses SelexION™ Technology to take advantage of an analyte’s differential mobility in high and low energy fields for an additional level of separation in LC/MS/MS workflows.


Sample preparation

Standard tryptic digestion procedures were applied to 10 lots of human liver microsomes. Standard curves were made using digested rat liver microsomes and BD Supersome Human UGT products.

Chromatography

LC system: Eksigent ekspert microLC 200

Column: Eksigent ChromXP™ 3C18-300-CL (1.0 x 50 mm, 3 μm, 300 Å)


Injection: 5 μL

Flow rate: 50 µL/min μL



0 100 0

0.5 100 0

5 75 25

6 20 80

7.5 20 80

Mass spectrometry

System: QTRAP® 6500 System

Interface: SelexION™ Technology device in positive MRM mode

Data processing

The MRM transitions were. These transitions were determined to be optimal based on MS/MS analysis (not shown). Data were quantitated using MultiQuant™ Software.

Figure 1: SelexION™ Technology (Differential Mobility Separation (DMS): The DMS cell is easily mounted without any tools, and consists of two planar plates that direct ionized sample in the gas flow laterally towards the MS orifice. An asymmetric waveform is applied which alternates between high field and low energy fields, resulting in a separation voltage (SV). As the ions move back and forth between the plates, they are dragged towards the exit with the gas flow and will have a net drift towards one of the electrodes based on their high and low field mobility. A small DC offset, the compensation voltage (CoV), is applied to one electrode and must be optimized to ensure transmission of the ion of interest through the cell. The COV can be considered a filtering voltage, and tuning individual peptides at specific COV values separates the peptide from unwanted background and interfering peaks.

Sourceextension

ring

Curtainplate

DMS Dimensions1x10x30 mm

DMScell



OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS


Elimination of co-eluting multiple charge interference using SelexION™ Technology

When optimizing MRM transitions for quantitation of low-level tryptic peptides from tissues samples, co-eluting interferences can occasionally contaminate peptide MS/MS spectra making sequence confirmation difficult, as observed with the analysis of the IEIYPTSLTK peptide of UGT isoform 2B7 (UGT2B7). Using the QTRAP 6500 System coupled with the MIDAS™ Workflow, the MS/MS spectra initially indicated that only y ions were present, along with many additional, unexplainable peaks (Figure 2). The QTRAP MIDAS™ Workflow is a single-injection, targeted workflow only available on the QTRAP Series of instruments, where MRM transitions are used as survey scans to quantify the key peptides in complex samples. By employing Q3 as a fully functional ion trap, the QTRAP System simultaneously identifies and confirms the identity of peptides by capturing full scan MS/MS spectra. An alternative approach using looped enhanced product ion (EPI) scans with Q0 trapping produced a much stronger signal, revealing a chimeric MS/MS spectra resulting from a closely-eluting contaminant. To eliminate the contaminating signals and provide a pure MS/MS spectra of IEIYPTSLTK, SelexION™ Technology was used to filter the peptide of interest form the background. The EPI scan is the linear ion trap version of a regular product ion scan, but the response is more intense (by orders of magnitude) because the ions of interest are

collected and stored in the trap. As a result, the EPI scan produces a high quality MS/MS spectrum for a specific peptide fragment. The fragmentation occurring in the collision cell provides the information-rich MS/MS spectrum typical of collisionally-activated dissociation fragmentation. The additional information gained from the b ions discovered in the cleaner spectrum, confirmed the identity of the IEIYPTSLTK peptide.

Reduction in LC analysis time while sustaining selectivity with SelexION™ Technology

Coupling SelexION™ Technology to the 6500 series platforms enables exploration of a new dimension in selectivity prior to MRM analysis. Along with the reduction of background noise and removal of interfering ions, SelexION™ Technology provides the additional benefit of shortening the chromatographic runtime. Extensive sample preparation prior to chromatography can be reduced as well by SelexION™ Technology because of the elimination of multiple charge-state ions – which may interfere with the detection of selected peptides in complex matrices (e.g., tryptic digests). Selected ion monitoring (SIM) assays for poorly-fragmenting peptides (e.g., cyclic peptides) are also feasible with SelexION™ Technology.

A deeper look at how SelexION™ Technology separates multiple charge states?

The DMS offers a degree of selectivity that is dependent on the chemical nature of the analyte. For intact peptides, the multiply

Enhanced product ion (EPI)

EPI scan with Q0 TrappingDMS OFF

EPI scan with Q0 TrappingDMS ON

10X increase insignal intensity

Elimination of multiplecharge statebackground noise

Multiple charge state interferences

No interferences

IDA

EPI with Q0 Trapping

DMS to purify spectra (COV=9C)

IEIYPTSLTKInterference Peak

Figure 2: The QTRAP® 6500 System offers multiple options to enhance sensitivity, while still acquiring qualitative/sequence information. In this case Q0 trapping was leveraged to boost sensitivity, while a looped EPI scan provided full scan MS/MS sequence data at UHPLC speeds. The resulting chimeric spectrum – resulting from a co-eluting interference – would have taken much longer to identify using conventional triple quadrupole technology. SelexION™ Technology was then leveraged to remove the interference, yielding a clean MS/MS spectrum for quantitation.


OR

THO

GO

NA

L SELECTIV

ITY TO

OLS

charged ions (Z > +2) tend to optimize at elevated compensation voltages (CoVs). As a consequence of the higher CoVs, the ion of interest is moved into a CoV space that is often free of other interfering ions. Figure 3 illustrates the separation of multiple charge states from a mixture of BSA/Myoglobin/ß-Gal digests.

Conclusions

SelexION Technology takes advantage of the mobility differences that ions experience as they travel through a set of plates with high and low fields applied. Background or interfering ions can be tuned out prior to any MS filtering. This leads to a boost in performance through several avenues; LOQ improvement, reduction in background or interfering ions, isobaric separation or an accelerated runtime.

The spectrum for EIYPTSLTK peptide of UGT isoform 2B7 was deconvoluted when interfering ions were removed by SelexION Technology, allowing for further sequence confirmation and quantitation for this peptide.

Multiple charge states can be separated from each other due to the increase in CoV relative to the number of charges per ion, allowing for a mixture of isobaric, multiply charged species to be separated.

References1 Fernández-Metzler C. (August 2013) “Peptide Quantification on the QTRAP® Mass Spectrometers

with MicroflowLC: Bridging the Best of Small Molecule and Proteomic Analysis.” SCIEX Mass

Spec Webinar Series. Retrieved at: http://www.sciex.com/events/webinars/peptide-quantification-

on-the-qtrap-mass-spectrometers-with-microflowlc--bridging-the-best-of-small-molecule-and-

proteomics-analyses.

1“UGT Family of Enzymes: Quantification of Tryptic Peptides. Part 1 of 3: The QTRAP® 6500

Platform and MicroLC Provide the Combination of Sensitivity, Specificity and Robustness for the

Quantitation of UGT Enzymes.” (White Paper) SCIEX. Accessed November 2013. Retrieved at:

www.sciex.com

1 UGT Family of Enzymes: Quantification of Tryptic Peptides. Part 2 of 3: Accelerating MRM3

Workflows on QTRAP® 6500 System for Enhanced Selectivity in Complex Matrices like Tryptic

Digests.” (White Paper) SCIEX. Accessed November 2013. Retrieved at: www.sciex.com

1 UGT Family of Enzymes: Quantification of Tryptic Peptides. Part 3 of 3: Using SelexION™

Technology for Additional Selectivity by Separating Multiple Charge State Ions in Tryptic Digests.”

(White Paper) SCIEX. Accessed November 2013. Retrieved at: www.sciex.com



OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS MRM3 Quantitation for Highest Selectivity

in Complex MatricesSCIEX QTRAP® 5500 System Innovations

Mass spectrometry has transformed quantitative analysis to become the method of choice for many assays. More recently, LC/MS/MS has revolutionized quantitative bioanalysis. While single MS filtering offers advantages over non-mass selective techniques, the use of tandem mass spectrometry (MS/MS, or MS2) eliminates interferences and results in a dramatic increase in selectivity which yields a very low baseline, excellent limits of quantification, and very good linearity. As a result, the Multiple Reaction Monitoring (MRM) experiment performed on triple quadrupole mass spectrometers has become the technique of choice for highly sensitive and selective quantification in biological matrices.

In some cases, interferences cannot be eliminated using MRM. More elaborate sample cleanup and chromatography is required to eliminate these interferences. If a high baseline or matrix interference cannot be eliminated, the result is a compromised Lower Limit of Quantification (LOQ) as the detection of compounds in complex matrices is limited by signal-to-noise rather than by raw instrument response. In such cases, the addition of a third MS stage has been shown to greatly increase selectivity and eliminate the high baseline or chromatographic interference. The result is a lower LOQ and better chromatographic peak shape.

Sensitive LC MS/MS/MS analysis requires instrumentation with three key performance features: the ability to perform MS3, ultra high sensitivity to overcome the loss in ion current due to

multiple fragmentation steps, and fast scan speeds to keep up with fast LC. These requirements have been difficult to achieve with results comparable to MRM mode until the development of the SCIEX QTRAP® 5500 LC/MS/MS System.

Key Features of the QTRAP® 5500 System for MRM3 Quantification

MRM3 quantification – On the QTRAP® 5500 System, an MS3 scan is performed with a fast cycle time and using a narrow scan range centered around the second generation product ion to be used for quantification. This type of scan is referred to as MRM3 (Figure 1).

Faster linear ion trap scan speeds – Scan speeds up to 20,000 Da/sec enable MS3 scans with an HPLC compatible cycle time such that extracted ion chromatograms (XICs) of second generation product ions can be extracted and integrated with a sufficient

number of data points across the chromatographic peak.

Better in-trap fragmentation – The new Linear Accelerator™ Trap with pulsed gas valve implemented in the QTRAP® 5500 System provides faster, more efficient in-trap fragmentation (Figure 2).

Highest available ion trap sensitivity – The QTRAP® 5500 System features the highest sensitivity commercial ion trap mass analyzer.

High selectivity – Unit isolation of precursor ion in Q1 followed by excitation and fragmentation at unit resolution in the ion trap provides the highest available selectivity in MRM3 analysis (Figure 3).

Speed

The speed and efficiency of ion-trap fragmentation has also been greatly improved on the QTRAP® 5500 System. Collision gas is now introduced through a high speed pulsed gas valve that enables a rapid increase in pressure in the LIT (Figure 2). Together with an increase in the RF drive frequency, this results in increased fragmentation efficiency and reduced excitation time of 25 ms or less.

In addition, the scan speed of the linear ion trap has been increased to a maximum of 20,000 through the use of the faster eQ™ electronics. This enabling fast MRM3 scan cycles at highest sensitivity.

Selectivity

The combination of features on the QTRAP® 5500 System provides the highest level of selectivity. The analyte ion of interest is isolated in the Q1 quadrupole with a user–selected resolution, usually unit resolution (0.7 Th FWHH). It is then fragmented in the Q2 collision cell, providing a broad range of product ions to be selected in the ion trap.

The in-trap fragmentation is achieved through the application of a single wavelength / narrow band excitation. As shown in Figure 3, this allows very selective fragmentation. The C12 isotope of a product ion can be specifically excited and fragmented to completion with minimal fragmentation of the C13 isotope. This provides further selectivity advantages in the removal of interfering background. The combination of these features


OR

THO

GO

NA

L SELECTIV

ITY TO

OLS

provides unprecedented selectivity and flexibility in the design of the optimal MRM3 quantification experiments.

Sensitivity

Linear Accelerator™ Trap Technology has resulted in ground breaking improvement in the handling of ions inside the linear ion trap of the QTRAP® 5500 System, resulting in up to 100x more sensitivity. Trapped ions are manipulated within the linear ion trap through the use of auxiliary DC fields provided by the addition of small electrodes (Figure 4, top). Ions are gently moved toward the extraction region of the linear ion trap during the cooling period by a voltage applied to the trap collar. A potential barrier is created by increasing the potential on the auxiliary electrodes just before the mass scan to complete the ion concentration process. The application of this axial field has a significant effect on sensitivity (Figure 4, bottom left).

In addition, a radio frequency is applied to the exit lens of the Linear Accelerator Trap resulting in further sensitivity gains (Figure 4, bottom right). These two innovations enable better than unit resolution to be obtained in the trap scan modes at these very high scan speeds.

Removal of Tough Interferences

Innovations in scanning speed, selectivity and sensitivity on the QTRAP® 5500 System enable successful implementation of the MRM3 workflow for a wide range of analytes 3,4. Sometimes, background noise or interferences can limit the detection of an analyte. Shown in Figure 5 is an example of an interference that has the same MRM transition as Clenbuterol and elutes at the

same retention time. Use of MRM3 can completely remove this interference and enable a much lower detection of this analyte.

Conclusions

MRM3 is an effective strategy for quantitation of analytes when high background or interferences make standard MRM quantitation difficult (Figure 5).

MRM3 can be used to achieve similar LOQ‘s with less sample preparation and simplified or faster chromatography.

MRM3 has been successfully applied to the detection and quantitation of small molecules, peptides, and protein biomarkers.

MRM3 is significantly improved on the QTRAP® 5500 System technology making it a useful tool for quantitation in tough matrices.

The QTRAP® 5500 LC/MS/MS System is the highest performance triple quadrupole and linear ion trap system available on the market today providing users with many powerful quantitative and qualitative tools.

References1Collings BA, (2007) Fragmentation of ions in a low pressure linear ion trap. J. Am. Soc. Mass

Spectrom. 18, 1459-1466.

2Collings BA, and Romaschin AR, (2009) MS/MS of ions in a low pressure linear ion trap using a

pulsed gas. J. Am. Soc. Mass Spectrom. 20, 1714-1717.

3Fortin T. et al, (2009) Multiple Reaction Monitoring Cubed for Protein Quantification at the Low

Nanogram/Milliliter Level in Nondepleted Human Serum. Anal. Chem. ePub. Oct 19, 2009.

4Niessen J. et al, (2009) Human platelets express OATP2B1, an uptake transporter for atorvastatin.

Drug. Metab. Dispos. Fast Forward. Feb 23, 2009.



Improved Selectivity for the Low-Level Quantification of the Therapeutic Peptide Exenatide in Human Plasma MRM3 quantitation for highest selectivity in complex mixtures on the

SCIEX QTRAP® 5500 System

Yan Xu, John Paul Gutierrez, Tian-Sheng Lu, Haiqing Ding, Katie Piening, Erin Goodin,

Xiuying Chen, Kristin Miller, Yong-Xi Li

Medpace Bioanalytical Laboratories, Cincinnati, OH

Key challenges of peptide quantitation





Key benefits of MRM3 peptide quantitation

High selectivity – Because of the multiple fragmentation steps in MRM3, the resulting spectra have lower backgrounds and fewer interfering, co-eluting contaminants.

Improved sensitivity – Detection limits in very complex matrices can often be improved using MRM3 analysis by removing interferences at the low end of the concentration curve.

Key features of QTRAP® 5500 System for MRM3 peptide quantitation

Fast scanning speed – Improvements to the QTRAP® 5500 Systems has enabled faster and more sensitivity MRM3 analysis.

Unique hybrid linear ion trap MS – Q1 is used for precursor ion selection (unit resolution), and Q2 for the first fragmentation step in transmission mode. No low mass cut-off associated with the first fragmentation step and higher collision energies delivers higher speed, greater selectivity, and more flexibility in the choice of the first product ion.

Introduction

Pharmaceutical research is increasingly focused on the development of biotherapeutics, which in turn is driving increased interest in the use of LC/MS/MS techniques for the quantitative analysis of proteins and peptides. To measure these peptide therapeutics in plasma, researchers have traditionally relied upon ligand-binding-based assays such as ELISA for pharmacokinetic and drug development studies. However, these immunoassays are fraught with selectivity problems, often displaying high background and a dynamic range too limited for accurate and precise concentration measurements at low levels. On the other hand, LC/MS/MS methods offer an increased dynamic range and fast method development, as well as significant improvements in the accuracy of bioanalysis.

Exenatide is a large therapeutic peptide that has been approved for the treatment of diabetes mellitus type 1 and 2. This peptide enhances glucose-dependent insulin secretion by the pancreatic beta-cell, acting as a regulator of glucose metabolism and insulin secretion. In recent years, immunoenzymatic assays have been the primary method for quantitating exenatide, but the time-consuming development of antibodies, and the lack of precision and accuracy in quantitative measurements have compelled the development of new bioanalytical strategies for detecting peptides in biological fluids. Because of its highly

Figure 1: Structure of Exenatide. Exenatide is a large, 39- amino acid peptide (MW = 4,186.6 Da) that acts as a regulator of glucose metabolism and insulin secretion.

OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS

Dr. Yong-Xi Li


selective capacity, an MRM3 LC/MS strategy1 was evaluated using the SCIEX QTRAP® 5500 System for reproducible and robust peptide detection in human plasma. We found that MRM3

scans (Figure 2) on a hybrid triple quadrupole linear ion trap mass spectrometer using high flow chromatographic conditions can provide high-throughput, as well as highly sensitive and selective, measurement of exenatide in biological fluids, demonstrating the precision and accuracy necessary for the regulated laboratory environment.


Sample preparation

Human plasma samples containing exenatide were dried using a gas vortexer (TurboVap) under nitrogen and reconstituted. In all steps, pH values and organic phase were carefully controlled.

Chromatography

LC system: Shimadzu UFLC LC-20ACXR

Column: Reverse phase C-18 (2.0 x 30 mm, 5 µm)

Injection: 5 μL


Mobile phase: A) water, 0.1% formic acid B) methanol, 0.1% formic acid


0 98 2

0.5 98 2

5 5 95

Mass spectrometry

Mass spectrometry system: SCIEX QTRAP® 5500 System

LC/MS analysis using the MRM3 acquisition strategy.1 Using the MS3 scan, the trap was filled using dynamic fill time (DFT), and the instrument was scanned at 10,000 Da/sec. The trap excitation time was 25 msec, giving a total cycle time of 0.17 sec. MRM3 analysis used the transitions: 838→396→202 for exenatide quantitation.

OR

THO

GO

NA

L SELECTIV

ITY TO

OLS

Figure 3: MRM3 assay design. An enhanced mass spectrometry (EMS) scan (top) is used to select the dominant parent ion; the most intense fragment ions are identified using enhanced product ion (EPI) mode (middle); and MS3 fragmentation is used to select the best secondary fragments to extract (bottom).

Figure 4: MRM3 significantly improved selectivity for exenatide in plasma over MRM analysis. Extracted ion chromatograms show the results of an MRM (top) and an MRM3 (bottom) scan for the detection for exenatide. Elimination of chromatographic interferences and background noise improves the LOQ for exenatide in plasma.

Figure 2: Schematic of MRM3 for quantitative analysis by LC/MS. The parent ion is first selected in Q1 and then fragmented in Q2. Product ions are trapped and isolated in the linear ion trap, followed by resonance excitation for secondary fragmentation. Second-generation product ions are then scanned out to the detector.



Data processing

Samples were acquired with the Analyst® 1.5 Software. Quantification was completed with MultiQuant™ Software, version 2.

Figure 5: Calibration curves for exenatide quantitation in human plasma using MRM and MRM3 analysis. (Top) MRM measurements of exenatide (250-1,000 ng/mL) in plasma and (bottom) MRM3 measurements from 5–1000 ng/mL are compared. The graph displaying the MRM3 data shows significantly improved linearity (R2 = 0.996) over the data obtained by MRM analysis.

OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS LLOQ LQC MQC MHQC HQC

Conc (ng/mL)

5.00 15.0 50.0 800 1800

5.14 15.9 46.4 779 1650

4.32 16.9 47.2 767 1549

5.65 12.0 41.7 821 1521

4.54 13.5 43.7 729 1641

3.69 17.1 50.0 658 1745

4.22 17.4 45.3 751 1672

Mean 4.59 15.5 45.7 751 1630

SD 0.701 2.22 2.85 54.8 82.4

%CV 15.3% 14.3% 6.2% 7.3% 5.1%

RE -8.2% 3.2% -8.6% -6.1% -9.5%

Table 1: Accuracy and precision of QC samples for exenatide quantitation.



MRM3 exenatide assay development

To determine the MRM3 transitions for exenatide quantitation, enhanced MS (EMS) mode was used to scan the plasma samples. The multiply-charged parent ion [M+5H]5+ at m/z 838.3 was selected as the first precursor (Figure 3, top). Upon fragmentation, the resulting predominant product ion m/z 396.4 was selected as the second precursor ion (enhanced product ion (EPI) scan, Figure 3, middle) and then fragmented in linear ion trap to generate the MS3 spectrum (Figure 3, bottom). The major fragment ion m/z 202.2 was selected as the second-generation product ion for MS3 quantification.

Assay performance for exenatide

Use of MRM3 analysis resulted in significantly improved selectivity for exenatide in human plasma extracts. Figure 4 shows a comparison of MRM3 vs. traditional MRM quantitation. Baseline was lower and chromatographic interference from the plasma matrix was completely eliminated in the MRM3 scan. The fast scanning speed of the QTRAP® 5500 System (10,000 Da/sec) provided a sufficient number of data points across the analyte’s chromatographic peak for good reproducibility. The improved detection performance resulted in excellent assay performance at the LLOQ and four QC levels as shown in Table 1. Accuracy and %CV for six replicates demonstrate that the MRM3 approach is capable of quantitative performance suitable for development-grade bioanalytical assays.

Conclusions

A sensitive and selective bioanalytical assay for exenatide in human plasma was successfully developed using MRM3 analysis on the QTRAP® 5500 System.

The increased selectivity of MRM3 eliminated baseline noise and chromatographic interferences, delivering analytical performance that was superior to traditional MRM scans.

MRM3 assays demonstrated the potential for excellent linearity achieving a dynamic range of 5–2,000 ng/mL, compared with a range of less than 250–1,000 ng/mL for traditional MRM.

Accuracy and reproducibility of the MRM3 assay met the requirements for a development stage bioanalytical assay.

Acknowledgements

The authors would like to acknowledge Gary Impey, Xavier Misonne, Johnny Cardenas, Suma Ramagiri, and Mauro Aiello from SCIEX for their support and valuable advice during the completion of this work.

References1 MRM3 Quantitation for Highest Selectivity in Complex Matrices. SCIEX Technical Note, Publication

0920210-01

OR

THO

GO

NA

L SELECTIV

ITY TO

OLS



OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS Quantification of Prostate Specific Antigen

(PSA) in Nondepleted Human Serum using MRM3 AnalysisHigh-throughput analysis of protein biomarkers using the SCIEX QTRAP® 5500 System

Yan Fortin T1, Salvador A2, Charrier JP1, Lenz C3, Bettsworth F1, Lacoux X1, Choquet-Kastylevsky G1, Lemoine J2

1BioMerieux, France, 2University of Lyon, France, 3SCIEX, Germany

Introduction

Over the last ten years, protein biomarkers of disease were discovered – from a comparatively small number of proteomics samples – using mass-spectrometry-based applications. Verification and validation of these biomarkers has been minimal, limiting the translation of these proteins into viable, routine clinical assays. Traditionally, clinical researchers have used immunoassays to quantify low-level proteins in biological samples; however, development of these tests is very time-consuming, and the results often lack reproducibility and specificity. Analytical techniques that sustain the rapid development of proteomics assays with high specificity and sensitivity are needed, driving a rapidly growing interest in LC/MS-based strategies for protein quantitation.

The use of multiple reaction monitoring (MRM) scans combined with stable-isotope-labeled proteins and peptides for the quantification of proteins has been actively explored over the last few years and shows great promise for clinical applications. Some key requirements that must be met for MRM scans to be applicable to the clinical environment are: high specificity, robustness, very high throughput, and high sensitivity – detecting low-abundance proteins in the ng/ml to pg/ml concentration range in human plasma. In keeping with the high-throughput laboratory environment, sample preparation must also be simple and robust.

Another limitation hindering the widespread use of MRM for biomarker verification has been the use of nanoflow chromatography. While it greatly increases the overall sensitivity of LC/MS/MS experiments and requires significantly reduced amounts of sample, nanoflow chromatography does not offer the sample throughput, reproducibility, and robustness required for implementation in clinical research laboratories.

To overcome these challenges, we developed a novel MRM approach that combines the use of fast-flow analytical chromatography with the new, highly-selective mass spectrometry technique, MRM3 – a scan mode that significantly reduces MRM background noise and spectral complexity by producing and analyzing second-generation product ions. This approach enables robust detection of low-level protein biomarkers from human serum at low ng/mL concentrations and provides an efficient method of protein quantitation applicable for high-throughput, multi-sample analysis.

Figure 1: MRM3 scans for quantitative analysis by LC/MS. Analyte precursor ions are selected in Q1 and fragmented in Q2; the resulting product ions are collected in the linear ion trap (LIT). A suitable fragment ion is isolated and fragmented in a second step using resonance excitation. Second generation product ions are collected and scanned out of the LIT to the detector.


OR

THO

GO

NA

L SELECTIV

ITY TO

OLS


Sample preparation

After denaturation with 6 M urea and reduction with 30 mM dithiothreitol, human serum samples were alkylated with 50 mM iodoacetamide. Proteins were digested with trypsin overnight at 37 °C (1:30 w/w enzyme to substrate ratio). After desalting using reversed-phase cartridges (Oasis HLB, 3 cm3, Waters), samples were fractionated using an MCX cartridge with elution at pH 5.5 using a methanol/acetate-buffer mixture (Waters).2

Chromatography

Column: Waters Symmetry C18 reversed phase (2.1 x 4100 mm, 3.5 µm)




0 95 5

0.5 95 5

30 60 40

Mass spectrometry

System: SCIEX QTRAP® 5500 System

Interface: Turbo V™ Source

MRM and MRM3 scan settings

MRM analysis was performed using unit resolution for both Q1 and Q3 quadrupoles. MRM3 analysis was performed using the MS/MS/MS scan function. The precursor ion was isolated in Q1 using unit resolution. First generation product ions were generated in the Q2 collision cell using an optimized collision energy and trapped in the Q3 linear ion trap for 200 msec. A suitable first-generation product ion was subjected to resonance excitation for 25 msec to produce second-generation fragments. Ions were scanned out of the ion trap at 10,000 Da/sec (total cycle time 350 msec/peptide).2 Q0 trapping further increased sensitivity.

Data processing

Data was processed using MultiQuant™ Software. MRM peak areas were integrated, either individually or summed together. MRM3 peak areas were determined by summing the integration of up to four granddaughter ions.

Figure 2: MRM3 assay design. The dominant parent ion is selected from an EMS scan (top), and the most intense fragment ions are identified using the EPI mode (middle), and MS3 fragmentation is used to select the best second-generation fragments to extract (bottom). Multiple second-generation fragments can be used to generate MRM3 XICs (lower panel).

Figure 3: MRM quantification of PSA (5 ng/ml) in human serum. High-flow HPLC coupled with MRM analysis was used to detect PSA after depletion, digestion, and fractionation of the serum sample.



OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS


MRM detection of PSA

Recently, PSA in human serum was detected using MRM analysis on a 4000 QTRAP® System with high-flow chromatography.1 HSA depletion followed by mixed-cation-exchange fractionation and HPLC separation enabled low-level detection at 5 ng/mL (Figure 2). Depletion of highly-abundant proteins is difficult to automate, and additional strategies were explored that removed this step and accelerated the workflow. The highly-selective MRM3 scan mode (Figure 2) was evaluated for its ability to maintain a high-level of sensitivity while enhancing analyte separation from co-eluting interferences. Detection of PSA and other low-abundance proteins in human serum at very low levels was facilitated by sufficient production of second-generation product ions in the linear ion trap of the QTRAP 5500 System; multiple product ions can be summed to generate MRM3 quantitative results, boosting the sensitivity of the assay (Figure 2, lower panel).

Using MRM3 scans for PSA quantification

To assess the intrinsic gain in specificity of the MRM3 method compared to the conventional MRM operating mode, a tryptic digest of human serum from a female donor was spiked with tryptic digests of bacterial proteins TP171, TP574, TP435, and core NS4, and the human prostate-specific antigen (PSA) protein over a range of concentrations (0–1,000 ng/ml). One average, the limits of detection for the five model proteins improved by 3- to 5-fold when detected using the MRM3 scan mode over the traditional MRM mode.2 MRM3 analysis of the model proteins resulted in spectra with decreased background noise and fewer co-eluting interferences compared to spectra obtained by the regular MRM mode; representative spectra for PSA detected by both scan methods are shown in Figure 4, and clearly illustrate the heightened specificity for the analyte.

The selectivity gains from MRM3 scans positively impacted the quantitative data obtained from standard concentration curves constructed from MRM3 peak area data, which showed excellent linearity over three orders of magnitude for the core NS4 peptide ALESFWAK (Figure 5). The accuracy and precision of the MRM3 data for the core NS4 peptide met bioanalytical standards, giving a %CV between 0.8–16% for MRM3 data,an improvement over the %CV range of 0.4-19.3% for MRM data alone (Table 1). An improved LLOQ was shown for the core NS4 peptide when using the more selective MRM3 experiment over the standard MRM – lowering the LLOQ to 10 ng/mL and enhancing the capacity of LC/MS/MS for very-low level peptide detection.

Figure 4: Reduction in background interference with MRM3 analysis. PSA spiked into human serum (female) at 50 ng/ml was analyzed by MRM scan (left) or by MRM3 scan (right). A significant reduction in background was observed when an MRM3 scan was used for PSA detection.

Figure 5: Improved detection of bacterial core NS4 peptide, ALESFWAK, in human plasma. MRM analysis yielded a quantitation limit (LLOQ) of 50 ng/ml (right panels). MRM3 detection resulted in an improved LLOQ of 10 ng/ml.

Concentration(ng/mL) 10 50 100 500 1000

MRM3 Mean 11.3 47.0 92.4 516.7 992.7

Precision 16.0 6.7 6.4 3.2 0.8

Accuracy 13.2 -5.9 -7.6 3.3 -0.7

MRM Mean 54.7 101.1 511.3 994

Precision 8.17 19.3 1.3 0.4

Accuracy 9.4 1.1 2.3 -0.6

Table 1: Accuracy and precision analysis of quantification data for the core NSA peptide, ALESFWAK, in human plasma.


OR

THO

GO

NA

L SELECTIV

ITY TO

OLS

Conclusions

Protein quantitation in human serum at low ng/ml concentrations was demonstrated using high-flow chromatography and MRM3 analysis.

Critical detection levels of 4-10 ng/ml of circulating PSA were accurately determined in human serum using a robust two-step sample preparation protocol combined with MRM3 analysis.

Improvements in speed and sensitivity to the QTRAP® 5500 System make MRM3 analysis a robust quantitative strategy for peptides in complex matrices when significant background interferences are present.

References1 Fortin T, et al. (2009) Clinical quantitation of prostate-specific antigen biomarker in the low

nanogram/milliliter range by conventional bore liquid chromatography-tandem mass spectrometry

(multiple reaction monitoring) coupling and correlation with ELISA tests. Mol. Cell; 11(8): 1006.

2 Fortin T, et al. (2009) Multiple reaction monitoring cubed for protein quantitation at the low

nanogram/milliliter level in nondepleted human serum. Anal. Chem; 15: 9343.

3 MRM3 Quantitation for Highest Specificity in Complex Matrices. SCIEX Technical Note, Publication

0920210-0.



OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS UGT Family of Enzymes:

Quantification of Tryptic PeptidesAccelerating MRM3 workflows on the QTRAP® 6500 System for enhanced

selectivity in complex matrices like tryptic digests

Suma Ramagiri, Loren Olson, Gary Impey, Carmen Fernandez-Metzler

SCIEX, 71 Four Valley Drive, Concord, ON, L4K 4V8 Canada

PharmaCadence Analytical Services, LLC Hatfield, PA 19440

Key challenges of peptide quantitation





Key benefits of MRM3 peptide quantitation

High selectivity – Because of the multiple fragmentation steps in MRM3, the resulting spectra have lower backgrounds and fewer interfering, co-eluting contaminants.

Improved sensitivity – Detection limits in very complex matrices can often be improved using MRM3 analysis by removing interferences at the low end of the concentration curve.

Key features of QTRAP® 5500 System for MRM3 peptide quantitation

Fast scanning speed – Improvements to the QTRAP® 5500 Systems has enabled faster and more sensitivity MRM3 analysis.

Unique hybrid linear ion trap MS – Q1 is used for precursor ion selection (unit resolution), and Q2 for the first fragmentation step in transmission mode. No low mass cut-off associated with the first fragmentation step.

Introduction

In Part 1 of this series, we introduced the UGT (UDP-glucuronsyl-transferase) family of enzymes and presented a framework for understanding why quantitating UGTs is important for the drug discovery and development process.1 In short, UGTs are responsible for glucuronic acid conjugation of xenobiotics, an important route of elimination for over 200 drugs – making this a pathway of common focus during the development of pharmaceuticals. Quantitation of UGT enzyme expression and its absolute cellular level provides pharmaceutical scientists with essential information for further characterization of tissues and cell lines used for drug metabolism studies and for studies on inductive pathways that effect UGT enzyme expression.

Improvements to standard MRM approaches for peptide quantitation are limited to adjustments in ion pairs, manipulation of different CE profiles, and development of alternate sample preparation and chromatographic strategies to enhance the signal on analytes of interest. Often these adjustments are very time-consuming and don’t provide significant improvement over the original methodology. QTRAP® Technology offers MRM3 capabilities (Figure 1), providing an additional selectivity option for those analytes with a generous spectrum of fragment ions.

The common problems experienced when quantitating low-level peptides in complex matrices are a lack of sensitivity and the presence of variable background interferences. This tech note describes fast and novel strategies available on the QTRAP® 6500 Series of instruments that provide up to 5-fold better LOQs than the standard MRM approach – with no sacrifice in throughput. The MRM3 technique conducted on the QTRAP® 6500 System quickly and easily provides better selectivity and sensitivity for peptide detection – ultimately improving the efficiency of assay development and execution for the bioanalysis of UGTs.

Carmen Fernandez


OR

THO

GO

NA

L SELECTIV

ITY TO

OLS


Sample preparation

Standard tryptic digestion procedures were applied to 10 lots of human liver microsomes. Standard curves were made using digested rat liver microsomes and BD Supersome Human UGT products.

Chromatography

LC system: Eksigent ekspert microLC 200

Column: Eksigent® ChromXP™ 3C18-300-CL (1.0 x 50 mm, 3 μm, 300 Å)


Injection: 5 μL

Flow rate: 50 µL/min μL

Mobile phase: A) water, 0.1% formic acid

B) acetonitrile, 0.1% formic acid


0 100 0

0.5 100 0

5 75 25

6 20 80

7.5 20 80

0 100 0

8 100 0

10 100 0

Mass spectrometry

System: QTRAP® 6500 System Used in positive MRM mode

Data processing

MRM3 scans are fragmentation pathway specific routes to the ion of interest (Figure 1). The following MRM3 scans were used to quantitate UGT tryptic peptides: 1) for DIVEVLSDR y7 ion, 523.3/817.4/589.3; 2) for YIPCDLDFK y6 ion, 585.3/637.3/303.2, 3) for TILDELIQR y6 ion, 558.8/773.4/515.3 for RT 4.96 min. and 558.8/773.4/515.3 for RT 5.33. MRM3 scans were processed using Analyst® Software (Figure 2). All quantitative results were processed using MultiQuant™ Software.


Elimination of tryptic peptide background interferences using the MRM3 workflow

Analysis of tissue samples from human subjects can be problematic due to high-levels of variable background noise and co-eluting, interfering contaminants – and is especially true for the detection and analysis of low-levels of tryptic peptides in tissue matrices. To address these selectivity issues, a quantitative MS3-style acquisition mode, MRM3, can help remove interfering signals quickly. These MRM3 assays can be developed in minutes, an improvement over the lengthy amount of time (often days)

Figure 1. Description of MRM3 Scans on the QTRAP® 6500 System: The precursor ion is selected in Q1, followed by fragmentation in Q2. The residual precursor ion and resulting fragments are captured in the Q3 linear ion trap (LIT) for a designated fill time. A specific fragment ion is selected for further analysis, and upon isolation in Q3. this ion is fragmented to form second-generation fragment ions. These ions are rapidly scanned out of the linear ion trap (LIT) and serve as the analytical signal for the MRM3 experiment. By reducing background and eliminating interferences, this technique results in LOQs almost an order of magnitude lower than analogous MRM. Compatible with fast flow rates, MRM3 acquisition can be combined with higher throughput chromatography, resulting in greater efficiencies.

Figure 2: Analyst® Software MRM3 acquisition method. An example of an Analyst®

Software acquisition method showing five MRM3 experiments including 10 MRM scans with a duty cycle of only 600 msec all in the same analytical run.



OR

THO

GO

NA

L SE

LEC

TIV

ITY

TO

OLS

Figure 3: A) An MRM-based chromatogram of the DIVEVLSDR.y7 tryptic peptide shows a background interference peak at RT 2.97 min and also one at RT 4.31, near the parent peak RT 4.38. B) The MRM3 technique used on the same peptide in 2A completely eliminated the interfering peaks (at RT 4.31 and 4.38), improving parent peak integration for a better %CV. C) An MRM-based chromatogram of the YIPCDLDFK y6 tryptic peptide shows an interfering signal at RT 4.29 min (~ 40% less intense than the parent peak at 4.50 min). D) The MRM3 technique completely eliminated the interfering peak (RT 4.29 min), improving the parent peak integration for better %CV.

Figure 4: MRM3 workflows for multiple second-stage UGT fragment ions provide selectivity gains. A) TheTILDELIQR.y6 tryptic peptide chromatogram shows two peaks at RT 4.96 min and 5.34 min for the MRM transition 558.8/773.4. B) An MRM3 scan (m/z 200-900) shows multiple ions (m/z 515.3 and 755.4) unique to the chromatographic peak, RT 4.96 min. C) The MRM3 chromatogram shows only one peak at RT 4.96 min for the transition 558.8/773.4/515.3+755.4. D) An MRM3 scan (m/z 200-900) shows multiple ions (m/z 529.4 and 658.4) unique to the chromatographic peak, RT 5.33 min. E) An MRM3 chromatogram shows no isobaric interference with only one peak at RT 5.33 min for the transition 558.8/773.4/515.3+755.4.


For Research Use Only. Not for use in diagnostic procedures.O

RTH

OG

ON

AL SELEC

TIVITY

TOO

LS

typically required for sample preparation or chromatography improvements. To improve detection and quantitation over MRM-style experiments, MRM3 scans include an additional level of ion selection and fragmentation that removes or significantly reduces co-eluting interferences and background peaks, allowing for easier peak integration and shortened run-times for improved efficiency. MRM3 chromatograms of two UGT tryptic peptides, the y7 DIVEVLSDR and y6 YIPCDLDFK, reveal the potential of this technique to simplify background, improve S/N, and increase throughput with faster chromatography (Figure 3).

Traditional MRM experiments may also produce spectra with multiple second-stage fragment ions, which impedes the straightforward quantitation of the peptide of interest. An extra level of selectivity gained from MRM3 experiments eliminates the isobaric interferences and yields cleaner spectra for a higher degree of separation for the peak of interest. The close elution of two peaks (from second stage fragment ions) was observed in the MRM chromatogram of the UGT y6 tryptic peptide TILDELIQR (Figure 4). MRM3 experiments on each of the peak removed the corresponding isobaric interference and facilitated further isolation of the single peak of interest.

Faster MRM3 scans for simultaneous, multi-analyte analysis at UHPLC speed

The QTRAP® 6500 System enables MRM3 scans that are twice as fast as those conducted on previous generations of QTRAP® Technology, enabling faster chromatography and more MRM3 experiments for multiple analytes in a single injection. Automated MRM3 method-building makes defining parameters effortless, while also making the MRM3 workflow fast, reproducible, and easy-to-use – increasing throughput and selectivity at the same time. Faster MRM3 scans are achievable on QTRAP® 6500 System due to improved ion processing and manipulation along the ion path, resulting in faster cycle times. Improvements to the detector capacity with IonDrive™ Technology provide an increased linear dynamic range, producing calibration curves with an extended concentration range. Faster linear ion trap scan speeds of up to 20,000 Da/sec enable MRM3 scans compatible with UHPLC-compatible cycle times. This allows for extracted ion chromatograms (XICs) of second generation product ions to be integrated with a sufficient number of data points across the chromatographic peak, increasing the precision and accuracy of the quantitative peptide data.

The sensitivity and selectivity of the MRM3 peptide quantitation assay can be further enhanced by summing multiple second stage fragment ions for an increase in signal intensity. A UGT 2B7 tryptic peptide, analyzed by MRM3, showed a two-fold increase in peak area when two second-generation ions were summed versus only one ion (Figure 5A). Quantitation data obtained from summing multiple MRM3 fragment ions was used to produce the calibration curve for the UGT 2B7 peptide (Figure 5B), which showed excellent linearity over three orders of magnitude.

Conclusions

MRM3 quantitation strategies offer the added benefit of a second stage of fragmentation over traditional MRM assays, raising the selectivity and sensitivity of detection for low-level peptides in complex biological matrices, such as tryptic digests of cellular extracts.

MRM3 scans improve signal-to-noise signals for analytes of interest in the most complex matrices by eliminating or reducing background interferences.

The faster MRM3 scan on the QTRAP® 6500 System accelerates acquisition time, making it compatible with UHPLC runs, while still maintaining a sufficient number of points across chromatographic peaks for highly precise and accurate quantitation of low-level analytes.

Figure 5: Summing multiple UGT ions for increased selectivity and sensitivity. A) Multiple, second-generation daughter ions from an MRM3 chromatogram of UGT 2B7 peptide were summed (blue trace), enabling a ~2-fold increase in signal intensity (Pink trace: chromatogram of only one second-generation ion). B) Linear calibration curve showing MRM3 transition for peptide UGT 2B17 (524.8/862.5/523.4).


Key challenges in MRM optimization for protein and peptide quantification

Choosing a unique peptide – A peptide that is unique to the protein of interest within a given background and is also sufficiently sensitive and selective.

Choosing the best MRM for the peptide – Multiple charge states are possible for a given peptide which in combination with the many product ion possibilities leaves many MRMs to be screened.

Optimizing MS parameters – Manual tuning can be tedious, and optimizing via LC injections is time consuming, particularly when monitoring multiple MRMs per peptide and multiple peptides per protein.

Key benefits of using DiscoveryQuant™ Software for signature peptide optimization

Increase productivity with reduced method development time – By using an automated infusion or flow-injection based tune with DiscoveryQuant™ Software instead of LC with step parameters.

Save time with automated workflow – Predicted transitions from the Skyline software are validated on a real digest sample quickly. DiscoveryQuant™ Software is used to automatically optimize DP, CE, CXP and EP.

Unique features of DiscoveryQuant™ Software for signature peptide quantification

Two experiment workflow – The QuickTune and FineTune experiments allow you determine the balance between throughput and ultimate sensitivity.

Always have access to your results – All results from a DiscoveryQuant™ optimization experiment are stored in the DiscoveryQuant™ database for easy retrieval and review.

Signature Peptide MRM Optimization Made Easy for Therapeutic Protein and Peptide QuantificationIan Moore and Suma Ramagiri

SCIEX, Concord, ON

SOFT

WA

RE

TOO

LS

Protein ormAb

In silicoDigest

LC-MRM Injection to Screen for Peptides and Product Ions

Infusion or FlowInjection Tuning,

2 Minutes per Peptide

LC-MRM Injectionswith Step Parameters for Optimization of

DP, CE, CXP

LC-MRM InjectionOptimization

DiscoveryQuant™Software

160

to 2

40 m

inut

es

30 m

inut

es

FineTuneproduct ions

known

FineTune performs:MRM based fine tuning to maximize sensitivity of:DP, EP, CE, and CXPConfirms precursor ion

QuickTuneproduct ions

unknown

User reviewand validation

QuickTune identifies:Precursor ionUp to 7 product ionsA corresponding CE for each product ion

Database(all optimized compounds)

Figure 1: Time saving advantages of peptide optimization using DiscoveryQuant™ Software.

Figure 2: Schematic overview of the DiscoveryQuant™ Software tuning and optimization workflow.


Global database sharing – The data from multiple systems can be shared enterprise wide through the global DiscoveryQuant™ database connectivity.

Automatic MRM method generation – Using data from the database and DiscoveryQuant™ Analyze optimized MRM methods can be generated automatically.

Introduction

Protein based therapeutics are a rapidly expanding component of many pharmaceutical companies’ drug portfolio. Monoclonal antibodies (mAb) used in the treatment of cancer are one class of protein therapeutics that has achieved success. In addition to proteins, smaller therapeutic peptides have achieved approval for a wide variety of indications in metabolic, cardiovascular and infectious diseases. In order to support this rapidly expanding new class of drug molecules, the rapid development of sensitive and selective bioanalytical methods are required.

Historically, protein and peptide quantification has been done using ligand binding assays (LBA) but LBAs suffer from inherent variability, lack of specificity, narrow dynamic range, and time consuming method development. As an alternative to LBAs LC/MS/MS methods are both sensitive and selective, have a wide dynamic range, and have been a staple in the quantitation of small molecule drugs. Bioanalytical methods for proteins and mAbs generally require digestion of the sample with a proteolytic enzyme like trypsin followed by direct analysis of one or more of the proteolytic peptides. Unlike bioanalytical method development for small molecules the product ions of a peptide analyte can be predicted using known ion types (a, b, c, x, y, z). An excellent starting point for the development of an LC/MS/MS method for peptides is Skyline software (MacCoss Lab Software), which will provide a list of the possible product ions of a given peptide plus an estimate of the DP of the precursor ion and a CE for the product ions. The next step in the method development is to determine which proteolytic peptides are actually produced by the digestion reaction, and which product ions are actually formed in the collision cell and for a given peptide. Lastly, for the product ions formed the CE and CXP needs to be optimized to achieve maximum sensitivity. DiscoveryQuant™ Software is the ideal tool to perform this optimization. DiscoveryQuant™ Optimize Software allows for optimization of compound dependent parameters via flow injection or infusion and then populates a database with these parameters.

DiscoveryQuant™ Optimize Software offers two options for tuning and optimization: QuickTune and FineTune. The QuickTune experiment is used to identify product ions and is comprised of a precursor ion scan, a DP optimization for the precursor and product ion scans at user defined CEs. The product ion masses and an associated CE are then stored in the DiscoveryQuant™ database. The FineTune experiment can then optimize DP, CE, CXP and EP (Figure 2) using MRM transitions loaded from the DiscoveryQuant™ database for a seamless and automated

optimization that provides maximum quantitative sensitivity. The DiscoveryQuant™ database can also be manually populated with MRM information loaded from an external source like Skyline. In this way FineTune can be used to optimize DP, CE, CXP and EP without running a QuickTune experiment first.

This technical note describes the results of experiments where DiscoveryQuant™ Software was used to optimize compound dependent parameters and improve upon the sensitivity of methods obtained from the output of Skyline software for the quantitation of peptides.


Sample preparation

Trypsin digested E. coli BGAL from the SCIEX mass spectrometer standards kit, Part No. 4368624 was diluted to 0.5 pmol/µL in 50% acetonitrile in water with 0.1%formic acid. Infusion was performed at 2 µL/min using an Eksigent microLC electrode (25 µm) in the Turbo V™ ion source. LC/MS/MS injections were performed on a 0.10 pmol/µL (5 µL injection) sample at 0.25 mL/min with the standard SCIEX Turbo V™ electrode.

Data workflow

The BGAL peptide sequence (UniProt #P00722 ) was pasted into Skyline and in-silico digested with trypsin. Tryptic peptides between 9 and 25 amino acids were selected while excluding cysteine containing sequences. Skyline was setup to export up to six ‘y’ ions with masses above the doubly charged parent m/z. The molecular weight range of the peptides ranged from 1098.55 to 2445.97 Da. Only doubly charge peptides were selected and a list of doubly charged peptides and their ‘y’ ions was exported as an Analyst method in the .csv format. Using Excel, this .csv file was formatted into a table that could be imported to the DQ database and saved as a .txt file.

LC conditions

LC System: Shimadzu LC-30 Nexera System

Analytical column: Phenomenex Aeris Peptide XB-C18, 3.6 µ, 2.1 mm x 150 mm

Analytical flow : 0.25 ml/min

Mobile Phase A: Water (0.1 % formic acid)

Mobile Phase B: Acetonitrile (0.1 % formic acid)

Gradient: Time (min) Mobile phase A% Mobile phase B%

1.0 97 3

10.0 50 50

11.0 5 95

13.0 5 95

14.0 97 3

SOFTW

AR

E TOO

LSFor Research Use Only. Not for use in diagnostic procedures.


MS conditions

MS system: QTRAP® 4500 System with a Turbo V™ Ion Spray Source

Ionization mode: ESI with Positive Mode

Software

Data acquisition: DiscoveryQuant™ 2.1.2 Analyst® 1.6.1 Software

Quantitation: MultiQuant™ Software


E. Coli BGAL (1024 amino acids) was digested in-silico with trypsin using Skyline and the tryptic peptides selected for optimization are shown in Table 1.

Skyline software assigned DPs in the range of 71 to 120 V for the precursor ions of Table 1 and CEs in the range of 27 to 66V for the 6 product ‘y’ ions of each precursor. These values were used to construct an LC-MRM method that was used to analyze a sample of the BGAL tryptic digest (0.10 pmol/µL). The LC peak areas for each peptide MRM were calculated and used for comparison.

The Skyline information (product ion masses, DP and CE) was imported into the DiscoveryQuant™ Software database. A FineTune experiment was then used to optimize: the DP between 5 and 150 V, the CE between ±20V of the Skyline assigned CE and the CXP between 2 to 30V. EP was not optimized and was kept at 10 V. Total infusion time for each peptide was 1.0 minute. At a flow rate of 2.0 µL/min ~40 µL of sample was consumed or approximately 2.3 µg of protein. An example of the Optimize FineTune data (DP, CE) for peptide APLDNDIGVSEATR is shown in Figure 3. The Skyline CE for the product ions of this peptide was

SOFT

WA

RE

TOO

LS PeptideMonoisotopic

MassCharge State

m/z

IDPNAWVER 1098.546 +2 550.3

TDRPSQQLR 1099.573 +2 550.8

HQQQFFQFR 1264.610 +2 626.8

ELNYGPHQWR 1298.616 +2 633.3

VDEDQPFPAVPK 1340.661 +2 650.3

LWSAEIPNLYR 1360.714 +2 671.3

LPSEFDLSAFLR 1393.724 +2 681.4

APLDNDIGVSEATR 1456.716 +2 697.9

QSGFLSQMWIGDK 1495.713 +2 729.4

YSQQQLMETSHR 1506.689 +2 748.9

LSGQTIEVTSEYLFR 1741.889 +2 754.4

VNWLGLGPQENYPDR 1756.853 +2 872.0

IENGLLLLNGKPLLIR 1775.103 +2 879.4

WSDGSYLEDQDMWR 1786.726 +2 888.6

LQGGFVWDWVDQSLIK 1889.968 +2 894.4

DVSLLHKPTTQISDFHVATR 2264.191 +2 946.0

YGLYVVDEANIETHGMVPMNR 2407.130 +2 1133.1

YDENGNPWSAYGGDFGDTPNDR 2445.973 +2 1204.6

Table 1: List of the peptide sequences chosen for optimization.

Figure 3: DP and CE optimization data from the peptide APLDNDIGVSEATR using DiscoveryQuant™ FineTune. The CXP ramping data is shown below in Figure 4.

Figure 4: DP and CXP optimization data from the peptide APLDNDIGVSEATR using DiscoveryQuant™ FineTune.

Figure 5: Partial calibration curves of peptide APLDNDIGVSEATR. Maroon squares represent data from the DiscoveryQuant™ Software optimized LC-MRM method and blue diamonds represent data from the Skyline MRM method.


SOFTW

AR

E TOO

LS

35 V and the actual optimized CE was between 41 and 45 V for the 5 product ions.

DiscoveryQuant™ Software ranks product ions in the database based on intensity of the CE ramping experiments. Therefore, the most intense product ion is known prior to starting LC/MS/MS analysis. In the absence of library matching spectra this information is not known using Skyline alone. With the intensity of the product ions determined by the DiscoveryQuant™ Software an LC/MS/MS method can be made for the digest sample including only the most intense MRMs for each peptide. This decreases the overall cycle time of the method and allows an increase in dwell time to improve S/N for each transition.

DiscoveryQuant™ Analyze Software was used to build an LC-MRM method. Only the MRM of the most intense precursor/product ion

pair was selected from the database. The 0.10 pmol/µL sample was analyzed with this method and peak areas were compared to peak areas generated from the Skyline MRM method. The data in Table 2 shows the changes in peak area and signal to noise from the LC-MRM method generated using DiscoveryQuant™ FineTune compared to the Skyline MRM method.

Of the 18 peptides tested 11 showed an increase in signal to noise ratio and peak area >10% and 7 were unchanged (±10%). The average gain for the 10 peptides showing improvement was 94%. Calibration standards of peptides IDPNAWVER and APLDNDIGVSEATR were prepared (Figure 5) and a partial calibration curve constructed over 3 orders of magnitude. Both the signal to noise and peak area gains was consistent across all standards.

The QuickTune feature in DiscoveryQuant™ Software includes a product ion scan. Since not all peptide product ions can be described by a, b, c or x, y, z ion types the feature was used to analyze the BGAL digest for peptide product ions that are not supported in Skyline. In this workflow, only the peptide sequences need to be entered into the DiscoveryQuant™ Software batch setup table. The QuickTune settings were set to scan for product ions from 700 amu up to the mass of the singly charged precursor ion using collision energies of 15, 25, 35, 45 and 55 V. In addition

Peptide Q1/Q3

Retention Time (min)

0.10 pmol/µL BGAL

Avg. Area Gain (N=3)

Avg. S/N Gain (N=3)

DVSLLHKPTTQISDFHVATR1133.1 /

1472.7 4.78 308% 296%

YGLYVVDEANIETHGMVPMNR1204.6 /

1713.8 6.88 164% 157%

TDRPSQQLR550.8 /

728.4 6.68 153% 147%

IENGLLLLNGKPLLIR888.6 /

1023.7 6.89 91% 86%

APLDNDIGVSEATR729.4 /

832.5 7.02 53% 55%

IDPNAWVER550.3 /

660.3 6.04 49% 46%

YDENGNPWSAYGGDFGDTPNDR1224.0 /

1754.7 3.81 48% 49%

YSQQQLMETSHR754.4 /

760.3 7.37 31% 33%

LQGGFVWDWVDQSLIK946.0 /

1289.7 7.15 31% 34%

ELNYGPHQWR650.3 /

780.4 7.28 16% 13%

LPSEFDLSAFLR697.9 /

1184.6 8.61 10% 11%

VDEDQPFPAVPK671.3 /

755.4 8.20 9% 11%

WSDGSYLEDQDMWR894.4 /

979.4 7.37 4% 4%

QSGFLSQMWIGDK748.9 /

964.5 6.08 1% 0%

VNWLGLGPQENYPDR879.4 /

1075.5 5.99 0% 2%

HQQQFFQFR633.3 /

1000.5 5.30 -1% 1%

LWSAEIPNLYR681.4 /

1062.6 5.98 -3% -4%

LSGQTIEVTSEYLFR872.0 /

1143.6 5.62 -10% -10%

Table 2: Changes in peak area and signal to noise ratio of peptides that were opti-mized with DiscoveryQuant™ Software compared to un-optimized mass dependent parameters from Skyline.

Figure 6: QuickTune results and product ion spectrum for peptide QSGFLSQMWIGDK.

Peptide Q1/Q3Retention Time (min)

0.10 pmol/µL BGAL

Avg. Peak Area (N=3)

Avg. S/N (N=3)


964.57.37 4.30E+04 3.65E+02


740.17.37 8.59E+04 7.48E+02

Gain 2.00 2.05

Table 3: Gains in peak area and signal to noise ratio for peptide QSGFLSQMWIGDK based on a product ion pair identified using QuickTune.



to the product ion scans a DP ramp and an enhanced resolution precursor ion scan were performed. The data from peptide QSGFLSQMWIGDK are displayed in Figure 6.

The most intense product ion in the product ion mass window is 740.1 amu. The product ion spectrum was visualized with PeakView® 2.0 software using Biotools and ion 740.1 was not assigned to either a, b, c or x, y, z ion types. The product ion was included in an LC-MRM method and used to analyze the same 0.10 pmol/µL BGAL digest. The peak area and signal to noise ratio of the 748.9/740.1 pair was 2.02 fold greater that the 748.9/964.5 pair. In addition to maximizing the sensitivity of product ion types supported by Skyline, DiscoveryQuant™ Software can be used to increase the sensitivity of target peptides by identifying product ions not of the a, b, c or x, y, z type.

Conclusions

Optimizing the mass dependent parameters with DiscoveryQuant™ Software for transitions generated by Skyline increases the peak area and signal to noise ratio of the majority of peptides from a protein digest.

Optimizing peptides with DiscoveryQuant™ Software using infusion is fast, at 1 min per peptide, while requiring little sample, ~2 µL per peptide.

The QuickTune feature of DiscoveryQuant™ allows for the identification of product ions not of the traditional a, b, c or x, y, z ion types which can boost sensitivity for certain peptides.

SOFT

WA

RE

TOO

LS

Exceedingly sensitive.Sharply focused.

THE 6500 SERIES WITH IONDRIVE™ TECHNOLOGY

PUSHING THE LIMITS IN SENSITIVITY

© 2016 SCIEX. For Research Use Only. Not for use in diagnostic procedures. The trademarks mentioned herein are the property of SCIEX Pte. Ltd. or their respective owners.

See what couldn’t be seen. Until now.

The new 6500 LC/MS/MS series with multi-component IonDrive™ technology is

SCIEX’s most sensitive triple quadrupole, improving signal-to-noise and LOQ of up to

5X over the widest range of compounds and flow rates, and a 20X increase in detector

dynamic range – with no compromise in mass range.

When merged with the Eksigent ekspert™ microLC 200 system, the functionally stack-

able design reduces lab space by 100%, while minimizing maintenance costs and

reducing mobile phase costs by up to 95%.

The new SCIEX 6500 Series. It’s the farsighted successor to a long line of leading SCIEX

mass spec systems.

Explore visionary sensitivity at www.sciex.com/6500



Key challenges of software applications for peptide quantitation

Complicated software interfaces – Difficult-to-navigate software screens make new user training challenging.

Non-integrated, multi-step workflows – Performing multiple peak quantitation functions using separate software programs is time-consuming and inefficient.

Tedious manual integration.

Data integrity and authenticity.

Key benefits of MultiQuant™ Software 3.0 for peptide quantitation

Increased efficiency – MultiQuant Software can handle large data sets with thousands of MRM transitions for fast data processing and data review.

Fast, easy software training – Easy-to-use interface reduces the time required to train new operators for improved consistency.

Single software solution – Processes data from both small molecule drugs and biotherapeutics (proteins and peptides). Processes multiple data types (MRM, MRM3, TOF MS, SWATH™ Acquisition and MRMHR data) on multiple instruments (SCIEX Triple Quad™ Systems, QTRAP® Systems and TripleTOF® Systems).

Reduced manual data integration – Powerful and robust integration algorithms (MQ4) increase performance by automating peak integration with minimal manual intervention. Advanced query functionality quickly identifies samples that deviate from bioanalytical standards.

Improved data integrity and security – Ensures data integrity throughout with 21 CFR compliant features such as locking of results, secure reporting, robust Watson digital link, audit trail log and e-signatures. New and innovative audit trail functionality will allow easy search and present QA reviewers with track changes mode.

Key features of MultiQuant™ Software 3.0 for regulated bioanalysis

Easy to navigate screens – Fast, user-friendly interface with linked panes, custom queries, and one-click metric plots.

Compliance with regulatory audit tools – New audit trail with robust management and precise control.

Easy-to-access historical data – All versions of the calculated results are stored, and hyperlinked to the audit trail, allowing the reviewer to quickly see the data before and after a change.

Secure data reporting and exporting – Results table locking and secure export features ensure data integrity.

Searchable audit trail – Allows for a search of specific events or samples across all results tables within a project for easier QA review.

Robust Watson digital link – Secure data transfer with audit trail log.

Introduction

Recent innovations in bioanalysis – such as automated sample preparation and ultra-high pressure liquid chromatography (UHPLC) – have positively impacted sample throughput for peptide quantitation experiments, accelerating sample processing as well as data output. Peptide quantitation data analysis requires multiple, time-consuming, and labor-intensive steps including peak integration, review, and proper audit trail tracking. As a result, an upsurge in quantitation data can cause significant workflow bottlenecks, highlighting the need for rapid data processing and advanced automation of multi-step quantitative measurements and calculations to maintain a smoothly-running operation.

MultiQuant™ Software 3.0 quantitation package was designed for the bioanalytical laboratory with the goal of improving workflow efficiency and facilitating regulatory compliance through an easy-to-use interface, powerful data analysis functions, and innovative audit/review features – cohesively integrated within Analyst® Software. MultiQuant Software handles large bioanalytical data

MultiQuant™ Software 3.0: Peptide Bioanalysis for the Regulated Bioanalytical LaboratorySingle software solution delivers an easy-to-use platform for peptide quantitation on diverse

MS instruments, while providing innovative tools for rapid data processing, improved data integrity

and security, as well as unique audit trail functionality – for regulatory compliance

SOFT

WA

RE

TOO

LS


For Research Use Only. Not for use in diagnostic procedures.SO

FTWA

RE TO

OLS

sets using intuitive navigation panes, while retaining the identical software administration settings that were previously configured in Analyst Software, such as access controls and user roles. The overlapping features between MultiQuant and Analyst Software greatly simplify many administrative functions, including software installation management, authorizations for access, data review privileges, and review of data uniformity, introducing a new level of time-savings into the regulated laboratory environment.

MultiQuant Software delivers a versatile and flexible software solution designed for diverse analytes – from small molecules to larger biomolecules – with the capacity to process a wide variety of peptide quantitation data captured from MRM, MRMHR, and MRM3 scan functions, in addition to data from TOF MS and full scan experiments. The intuitive software interface retains an elegant simplicity, making software operation easy for any user regardless of experience level. Designed for multiple mass spectrometric platforms (including the QTRAP® Systems, Triple Quad™ Systems, and TripleTOF® Systems), MultiQuant Software is incredibly adaptable, supporting even UV-DAD (ultraviolet diode array detector) and ADC (analog-to-digital converter) data sets, an application often overlooked in other conventional quantitation programs.

Here, we highlight the new innovations in this single software solution – and demonstrate how MultiQuant Software 3.0 significantly improves the speed and ease-of-use for quantitative peptide processing through the following functional upgrades:

• Additional automation of quantitation methods and calculations

• Intuitive software interface design for straightforward navigation

• Fully-integrated and novel audit trail functionality for achieving regulatory compliance

• Enhanced validation procedures for fail-safe data integrity and secure reporting

Product features

Automation of quantitative functions

Automatic method creation – Peptides are routinely quantitated with high specificity using the multiple reaction monitoring (MRM) mode, and for complex samples, numerous MRM transitions are monitored simultaneously in the same run. Setting up individual quantitation methods for multiple transitions is typically quite tedious and time-consuming, but MultiQuant Software streamlines method creation with a laborsaving process. Using an Analyst Software ID column (Protein.Peptide.Fragment.IS Indicator) as a source for pre-defined MRM transitions (Figure 2), MultiQuant Software automatically creates individual MRM quantitation methods – eliminating the need to re-enter peptide names or retention times. Peptide groups are automatically configured, and peak elution times are defined by the retention times characterizing other MRM transitions for a particular peptide.

Automated absolute quantitation of peptides

For pharmaceutical research in the regulated environment, absolute quantitation of biotherapeutics is essential for regulatory compliance as well as for furthering pharmacokinetic research. Calculation of the absolute concentration requires information from a large number of samples, and as a result, can be quite monotonous and taxing, putting a strain on computing resources. To reduce the time-consuming aspects of absolute concentration determination, MultiQuant Software relies on pre-set functions to generate calibration curves, to determine unknown sample concentrations, and to measure internal standards. Additionally, MultiQuant Software links calibration data sets obtained for endogenous peptides and their corresponding labelled standards for easier navigation amongst multiple standard curves for a particular peptide. Some of the specific automated features for peptide standard curve generation include:

• Additional concentration and accuracy parameters for all the unknown samples calculated upon loading the calibration curve

• Straightforward generation of standard curves that are easily manipulated

• Creation of a statistics table for quick assessment of standard curve quality using %CV and accuracy (Figure 3, top).

• Instant creation of metric plots from results table data to visually assess data quality (Figure 3, bottom).

Intuitive software interface

Speed and ease-of-use – MultiQuant™ Software was designed with an intuitive, easy-to-use interface for streamlined navigation of quantitation data – consolidating data functions for rapid processing and manipulation. The well-organized, user-friendly software format reduces the time required for new operator training, improving user consistency and confidence.


SOFT

WA

RE

TOO

LS

Figure 1: Easy-to-use interface. Intuitive user workspaces make data navigation easy, with dynamic linking of results panes, analytes, integration and one-click metric plots.

Figure 2: Automatic creation of quantitation methods. Using a naming convention for proteins and peptides pre-defined in Analyst® Software, MultiQuant Software automatically builds quantification methods (bottom) based on the MRM method ID column (top). Multiple MRM transitions per peptide are considered part of a single analyte group for automatic retention time determination and interference analysis.



The navigation panes in MultiQuant Software make it easy to transition between information screens for different analytes. In addition, users can control display features with a high level of specificity, presenting internal standard and analyte results alone or combined, along with selective review of specific MRM transitions. Users can quickly build new data processing methods and evaluate the data in dynamically linked panes, automatically highlighting the chromatographic peaks and the corresponding integration when a sample is selected (Figure 4). In a single click, the user can view all analytes or a specific analyte. The results tables and peak review are automatically updated.

Fully-integrated audit trail management

Innovative audit trail features – Accurate and secure record keeping is essential for the regulated laboratory, but the additional documentation steps required for an audit trail often put burdensome demands on resources, introducing additional, time-consuming procedures into the workflow. MultiQuant Software 3.0 presents a new, robust audit trail, with innovative features such as Grouped Audit Events, providing powerful audit trail management and efficient review functionality integrated directly into the quantitation methodology. By consolidating required audit trail steps, significant time savings are realized, speeding up and automating the peptide quantitation process.

The new Audit Map Editor allows the laboratory administrator precise control over audit trail functionality using an intuitive, well-designed interface (Figure 5) to ensure consistency in the record-keeping process. For example, when a user is prompted to enter a reason for a specific auditable event, the administrator can pre-define the selections in a drop-down menu, limiting the user’s choices. The user can be allowed enter any reason without restrictions, or users can be constrained to a specific list of allowed reasons. Furthermore, an administrator can fine tune the audit process by configuring reason selection options separately for each auditable event.

MultiQuant Software 3.0 introduces an additional level of time savings and efficiency by combining audit trail tasks and curbing the number of user prompts just to related events. For example, the number of electronic signatures collected during data processing is reduced through the Grouped Audit Events function (Figure 6), which merges the individual responses to selected auditable events into one all-encompassing e-signature. When the user enters the expected concentration values for a multi-point standard curve, the software will prompt only once for the electronic signature, instead of prompting at each point; however, the audit trail will still document each recognized event separately. Other Grouped Audit Events include multiple, discrete changes to the integration parameters, whereby each change is recorded individually in the audit trail, but the operator must only provide one signature, rather than multiple signatures, for every parameter update. This consolidation of tasks improves the speed of data processing without sacrificing compliance.

Figure 3: Peak area statistics table and metric plots. With a single click, useful tables of statistics (top) or informative metric plots (bottom) can be quickly generated for visual assessment of data quality. The change in area of the internal standard for one peptide is plotted versus injection number (bottom). Multiple peptides or multiple MRMs can be plotted by selecting multiple rows from the left hand navigation bar (top).

SOFTW

AR

E TOO

LS


SOFT

WA

RE

TOO

LS

Figure 4: One stop easy data integration and regression set up. Key data processing parameters such as RT width, noise percentage, curve fitting etc will be chosen up to process 1000’s of sample batch unattended.

Figure 5: New audit map editor. The editor provides intuitive control over auditable events. The administrator can specify which events are audited. For each event, a list of pre-defined reasons can be specified and e-signatures can be required.



Powerful review functionality – All versions of a data set, before and after manipulations and calculations, must be collected and saved for the audit trail. MultiQuant™ Software 3.0 introduces a dedicated data review function that links all versions of the data to the audit trail. Now reviewers can view data before and after an auditable change (Figure 7). Data review and audit trail view panes are linked for easy and efficient data review. The software also prevents accidental changes in the data by warning if the data has been changed during review. This warning feature can be activated by the administrator.

Quality assurance review features – Audit trails need to be searchable, so that reviewers can locate specific events rapidly and effectively during the quality assurance process, capturing and

linking related events to the audit trail. MultiQuant Software 3.0 design facilitates efficient quality assurance practices, allowing supervisors to concurrently review multiple batches within a project – providing essential time-savings – as well as search all project tables to find a specific event or action captured in the audit trail (Figure 8). In addition, the reviewer can track a single sample across multiple batches, easily locating the original injection and re-injections in different results tables. The raw data for each injection and the resulting peak area integration are linked to each other and can be easily checked and tracked concurrently, aiding the validation of experimental data and ensuring that bioanalytical data standards are fulfilled.

Figure 6: Grouped audit events. The system logs multiple change to integration methods in the audit trail and prompts for electronic signature once.

Figure 7: Powerful audit trail viewer. The viewer displays events and the associated data with hyper-linking. The data is shown before and after each audited event. The reviewer can quickly spot the impact of a change. Results tables and peak review are automatically linked to the audit trail.

SOFTW

AR

E TOO

LS


SOFT

WA

RE

TOO

LS

Figure 8: Audit trail search capability. All manual integrations within a project can be quickly located and relevant versions of the data automatically displayed and linked to the audit trail. MultiQuant 3.0 software showing secure data transfer and audit trail log before exporting to Watson LIMS™.


For Research Use Only. Not for use in diagnostic procedures.SO

FTWA

RE TO

OLS

Validation of data integrity and security

Results table locking for improved data integrity – Untracked and unwanted alterations to experimental data puts a laboratory’s reputation at risk. To sustain outstanding data integrity throughout the results table generation process, administrators of MultiQuant Software can secure the data through a data locking feature to prevent further changes. The locking/unlocking authority defined by the administrator can also be applied to data prior to export to laboratory information management software (LIMS) or prior to printing a report. The software also requires the operator to save the results prior to printing or exporting. A dedicated data review function has been added so that supervisors can review the data without accidentally changing the data in the process.

Watson digital link – secure data transfer with audit trail log – Data export to LIMS has traditionally been identified as a weak link in ensuring data integrity since text files have the potential to be tampered with after export and prior to LIMS import. To address this issue, MultiQuant™ 3.0 Software now has a robust link with Watson LIMSd that allows for results transfer in a secure and compliant environment by encrypting the data prior to export. After receipt of the data in the LIMS system, the LIMS software can decrypt the data, all-the-while maintaining the data in a secured environment. This system of data transfer achieves higher productivity in the GLP laboratory – eliminating multiple data management tasks and

liberating researchers from concerns about data hacking. Key steps for secure data transfer between MultiQuant 3.0 Software and Watson LIMS are displayed in Figure 9.

Enhanced compliance through checksum validation and secure reporting – Once a data file has been transferred, the data integrity must be re-verified and a tamper-resistant transfer process must be demonstrated. MultiQuant 3.0 Software provides a Data File Checksum feature, which will reject data with an invalid checksum whenever a wiff file (data file) is created. Analyst Software generates a checksum value using an algorithm based on the MD5 public encryption algorithm and saves the value into the file. When the checksum is verified, the software calculates the checksum, comparing it to the checksum stored in the file. This powerful validation process ensures data safety, and boosts regulatory confidence in the analytical output.

Even with these tamper resistant features, the generation of final reports is also prone to errors, and the graphic display of data could be manipulated or corrupted during the data evaluation process. To maintain the highest level of data integrity, MultiQuant 3.0 Software also includes a powerful secure reporting functionality, ensuring that the GLP license holder is the only user who can edit the software’s encrypted report templates, limiting access to data output and preventing unwanted tampering.

Figure 9: Robust Watson Digital Link – Secure Data Transfer with Audit Trail Log. MultiQuant™ 3.0 software showing results transfer to Watson LIMS in file menu options.


SOFT

WA

RE

TOO

LS Conclusions

MultiQuant Software 3.0 significantly reduces the amount of time required for data processing and review by automating and consolidating repetitive data processing tasks.

Increased efficiency – MultiQuant Software can handle large data sets with thousands of MRM transitions using intuitive user interface for fast data processing and data review

Fast and easy software training – Easy-to-use user interface reduces the amount of time required to train new operators, and improve consistency

Single software solution to process both small molecule drugs and biotherapeutics such as proteins and peptides. Process multiple data types such as MRM, MRM3, TOF MS, SWATH™ acquisition and MRMHR on multiple instrument’s such as SCIEX Triple Quad™ System, QTRAP® LC/MS/MS System and TripleTOF® System

Save time by reducing tedious manual integration

Powerful and robust integration algorithms such as MQ4 will increase performance by automating peak integration with minimal manual intervention

Advanced query functionality will quickly identify samples that deviate from bioanalytical regulations

Data integrity and security

Ensures data integrity throughout with 21 CFR compliant features such as locking of results, secure reporting, robust Watson digital link, audit trail log and e-signatures

New and innovative audit trail functionality will allow easy search and present QA reviewers with track changes

References1 MRM3 Quantitation for Highest Selectivity in Complex Matrices. SCIEX Technical Note, Publication

0920210-01

2 To download a trial version of MultiQuant Software, please visit: https://licensing.sciex.com/

download/index

3 MultiQuant™ Software 2.0 with the Signal Finder™ Algorithm: The Next Generation in

Quantitative Data Processing.” SCIEX Technical Note, 2010.



SCIEX internal expertise

Suma Ramagiri

Gary Impey

Joseph Fox

Debadeep Bhattacharya

Yves LeBlanc

Larry Campbell

Bruce Thomson

Bruce Collings

Igor V Chernushevich

Bradley Schneider

Tom Covey

Elliot Jones

Kelli Jonakin

Tony Romanelli

Shaw Xia

Leo Jinyuan Wang

Anoop Kumar

PHARMA BUSINESS UNIT

RESEARCH APPLICATIONS

RESEARCH HARDWARE AND SOFTWARE

APPLICATION SUPPORT

Your success is our successWe take it personally

As an SCIEX customer you have access to an excellent customer support organization.

Wherever you are, we’re there with you as a trusted partner to answer questions, provide solutions,

and maximize lab productivity.

Our customer support organization has access to the latest product updates, software revisions,

methods and repair procedures to make sure that you stay on top of your game.

When you have questions, we have answers.

Learn more at sciex.com/customersupport, or locate your local account representative at

sciex.com/contactus

AB Sciex is doing business as SCIEX.

© 2016 AB Sciex. For research use only. Not for use in diagnostic procedures. The trademarks mentioned herein are the property of the AB Sciex Pte. Ltd. or their respective owners. AB SCIEX™ is being used under license.

9680314-01 04/2016 RUO-MKT-02-1486-A

Headquarters500 Old Connecticut Path | Framingham, MA 01701 USAPhone 508-383-7700sciex.com

International SalesFor our office locations please call the divisionheadquarters or refer to our website atsciex.com/offices

Biologics Bioanalysis - SCIEX · of bioanalytical quantitation such as proteomics, anti-doping,...

Documents

Transcript of Biologics Bioanalysis - SCIEX · of bioanalytical quantitation such as proteomics, anti-doping,...