Specific enzymatic digestion and fragment isolation for ... · quality guidelines Q8 – Q12 for...

ZURICH UNIVERSITY OF APPLIED SCIENCES

DEPARTMENT OF LIFE SCIENCES AND FACILITY MANAGEMENT

INSTITUTE OF BIOTECHNOLOGY

Specific enzymatic digestion and fragment isolation for the critical quality attribute assessment of IgG1

Master thesis

by

Rappo, Martin Alexander

Master of Life Sciences

20.10.2017

Pharmaceutical Biotechnology

Academic supervisor:

Prof. Dr. Jack Rohrer Head of the Cell Physiology and Cellular Engineering Group ZHAW Life Sciences und Facility Management, Einsiedlerstrasse 31, 8820 Wädenswil, Switzerland

Industrial supervisor:

Dr. Christoph Rösli Senior Fellow in the Physicochemical Analytics Group Novartis Pharma AG, Werk Klybeck, Postfach, 4002 Basel, Switzerland

iii

Imprint

Keywords

Critical Quality Attributes, CQA, Monoclonal Antibody, mab, Fab, Fc, F(ab’)2, Large hinge IgG

fragment, LHF, IdeS, IgdE, Lysine Gingipain, Kgp, EndoS2

Please cite this report as followed:

M. A. Rappo, Specific enzymatic digestion and fragment isolation for the critical quality attribute

assessment of IgG1, ZHAW LSFM (2017)

Contact information

Author

Martin Alexander Rappo Speiserstrasse 12 4600 Olten Switzerland Tel. +41 79 537 64 76 E-Mail: [email protected]

Academic Supervisor

Prof. Dr. Jack Rohrer ZHAW Life Sciences und Facility Management Einsiedlerstrasse 31 8820 Wädenswil Switzerland Tel. +41 58 934 57 17 E-mail: [email protected]

Industrial Supervisor

Dr. Christoph Rösli Novartis Pharma AG Werk Klybeck Postfach 4002 Basel Switzerland Tel. +41 61 696 14 39 E-Mail: [email protected] Basel 20.10.2017

mailto:[email protected]



iv

Acknowledgement

I would like to express my very great appreciation to Christoph Rösli. I would like to thank him

for giving me the chance to work on this interesting project, the stimulating scientific

discussions and useful inputs given during the time I spend at Novartis.

I am particular grateful for the support and training given by Olivia Guerre. Also the interesting

exchange helped to approach problems with a different perspective.

I would like to offer my special thanks to Philippe Simeoni, who always offered this time

whenever a question or issue regarding CE came up.

Also Francois Griaud I would like to thank for giving me an introduction on antibody digestion.

A thanks also goes to Wolfram Kern for being so kind to let me use his work bench and all his

equipment.

The refreshing exchange with Manuel Diez inside and outside the office has been a great help

to keep motivated and stay open-minded.

I would like to thank Genovis for kindly providing me a free test sample of FabALACTICA™.

A special thanks also goes to Raphael Dreier for all the coffee and lunch breaks we had

together, the support during stressful times, extended brain-storming sessions and all the

laughs we shared.

v

Abstract

Recent ICH quality guideline updates introduced the enhanced drug development which is

based on Quality by Design approach. For this a Quality Risk Management has to be

conducted based on scientific knowledge. Information gaps are closed through analysis of

molecules with domain-specific modifications. The aim of this thesis was to evaluate methods

to prepare a model IgG1 for such studies. The antibody fragments F(ab’)2, LHF, Fab and Fc

and the Fc-glycosylation cleaved IgG with residual N-acetylglucosamine were selected. A

representative IgG1 was digested with the recently discovered enzymes IdeS, Kgp, IgdE and

EndoS2 and the digestion products were purified with common biotechnological down-stream

methods. The F(ab’)2 was successfully isolated with a yield of 55% and a purity of 96%. After

an initial digestion parameter optimization the LHF, Fab and Fc could be generated and an

isolation method was established. A following scale-up was unsuccessful, due to

reproducibility issues. A promising alternative was identified. The enzymatic glycan cleavage

and product isolation was successful with a yield of 80% and a purity of 99%.

Zusammenfassung

Die kürzlich von der ICH aktualisierten Qualitätsrichtlinien führten den erweiterten

Wirkstoffentwicklungsvorgangs ein, welcher auf dem Quality-by-Design Prinzip basiert. Dazu

wird ein Risikomanagement, basierend wissenschaftlichen Fakten durchgeführt. Um

Informationslücken zu schliessen müssen proteindomänenspezifisch veränderte Moleküle

untersucht werden. Das Ziel dieser Arbeit war die Evaluation von Methoden zur Vorbereitung

eines Modell IgG1 für solche Studien mit minimaler unspezifischer Veränderungen.

Ausgewählt wurden die IgG-Fragmente F(ab‘)2, LHF, Fab und Fc und das Fc-Polysaccharid

gespaltene IgG mit restlichem N-Acetylglucosamin. Ein repräsentatives IgG1 wurde mit den

kürzlich entdeckten Enzymen IdeS, Kgp, IgdE und EndoS2 verdaut und mittels üblichen

biotechnologisch Reinigungsmethoden aufgereinigt. Das F(ab‘)2 Fragment konnte mit einer

Ausbeute von 55% und Reinheit von 96% isoliert werden. Nach einer initialen Optimierung der

Verdauungsparameter konnte das LHF, Fab und Fc erfolgreich hergestellt und eine Methode

zur Fragmentisolation gefunden werden. Die Herstellung einer grösseren Menge war erfolglos,

jedoch konnte eine alternative Lösung identifiziert werden. Die enzymatische Spaltung des

Polysaccharids und Isolation war erfolgreich mit einer Ausbeute von 80% und einer Reinheit

von 99%.

vi

Table of content

1 INTRODUCTION ............................................................................................................ 1

2 THEORETICAL BACKGROUND.................................................................................... 5

2.1 ANTIBODIES ............................................................................................................... 5 2.1.1 Structure ............................................................................................................ 5 2.1.2 Fragment nomenclature ..................................................................................... 7 2.1.3 Glycosylation ...................................................................................................... 8 2.1.4 Mechanism of action .......................................................................................... 8

2.2 ENZYMES ................................................................................................................... 8 2.2.1 Immunoglobulin-degrading enzyme of Streptococcus pyogenes ........................ 9 2.2.2 Lysine gingipain of Porphyromonas gingivalis .................................................. 11 2.2.3 Immunoglobulin G degrading enzyme of Streptococcus agalactiae .................. 13 2.2.4 Endo-beta-N-acetylglucosaminidase of Streptococcus pyogenes .................... 13

2.3 KINETICS MODEL FOR THE KGP DIGESTION ................................................................. 15

3 METHODS AND MATERIALS ...................................................................................... 17

3.1 INSTRUMENTS, CONSUMABLES, REAGENTS AND SOFTWARE. ....................................... 17 3.2 MODEL ANTIBODY ..................................................................................................... 17 3.3 BUFFER PREPARATION .............................................................................................. 17 3.4 PHYSICOCHEMICAL METHODS ................................................................................... 18

3.4.1 Cation exchange chromatography .................................................................... 18 3.4.2 Size exclusion chromatography ........................................................................ 19 3.4.3 Capillary electrophoresis – sodium dodecyl sulfate .......................................... 20

3.5 DIGESTION AND ISOLATION ........................................................................................ 22 3.5.1 Fab, Fc and LHF .............................................................................................. 22 3.5.2 F(ab’)2 .............................................................................................................. 27 3.5.3 Chitobiose cleavage ......................................................................................... 29

3.6 ECOLOGY, SAFETY AND DISPOSAL ............................................................................. 30

4 RESULTS AND DISCUSSION ..................................................................................... 31

4.1.1 Fab, Fc and LHF .............................................................................................. 31 4.1.2 F(ab’)2 .............................................................................................................. 44 4.1.3 Chitobiose cleavage ......................................................................................... 46

5 CONCLUSION .............................................................................................................. 48

6 OUTLOOK .................................................................................................................... 49

7 REFERENCES ............................................................................................................. 50

7.1 LIST OF ILLUSTRATIONS ............................................................................................ 55 7.2 LIST OF TABLES ........................................................................................................ 56 7.3 LIST OF EQUATIONS .................................................................................................. 57

ANNEX .................................................................................................................................. I

vii

Abbreviations

Table 1: Abbreviations.

Acronym Description

Ab Antibody

ADCC Antibody-dependent cellular cytotoxicity

CDR Complementary determining region

CE Capillary electrophoresis

CH1 First constant heavy chain domain

CH2 Second constant heavy chain domain

CH3 Third constant heavy chain domain

CL Constant light chain domain

DB Digestion buffer

DP Drug product

Fab Fragment antigen binding

Fc Fragment crystallizable

FcRn Neonatal Fc receptor

FcγR IgG-Fc receptor

Fuc Fucose

Gal Galactose

GAS Group A Streptococcus

GlcNAc N-acetylglucosamine

HC Heavy chain

HPLC High performance liquid chromatography

ICH International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use

Ig Immunoglobulin

IgdE Immunoglobulin G degrading enzyme of S. agalactiae

LC Light chain

Mab Monoclonal antibody

Man Mannose

NeuGc N-glycosylneuraminic acid

NR Non-reducing

PAGE Polyacrylamide gel electrophoresis

PD Pharmacodynamics

viii

PK Pharmacokinetics

QbD Quality by Design

QRM Quality risk management

Rgp Arginine gingipain

RMT Relative migration time

Rt Room temperature

SDS Sodium dodecyl sulfate

VH Variable heavy chain domain

VL Variable light chain domain

Amino acids

Figure 1: Amino acid properties Venn diagram [1].

Amino acid A Ala Alanine C Cys Cysteine D Asp Aspartic acid E Glu Glutamic acid F Phe Phenylalanine G Gly Glycine H His Histidine I Ile Isoleucine K Lys Lysine L Leu Leucine M Met Methionine N Asn Asparagine P Pro Proline Q Gln Glutamine R Arg Arginine S Ser Serine T Thr Threonine V Val Valine W Trp Tryptophan Y Tyr Tyrosine

ZHAW LSFM, 2017 Introduction Martin Alexander Rappo

1

1 Introduction

With the creation of the first monoclonal antibody (mab) in 1975 a new era of pharmaceutical

development was initiated [2]. The mab technology increased therapeutic options and opened

new treatment opportunities for diseases with an unmet medical need, such as cancer,

immunological disorders, chronic inflammatory disease, infectious disease and many more [3].

Because mabs have a higher target specificity and reduced systemic toxicity compared to prior

predominant small molecular drugs, focus of drug research and development slowly shifted

toward this technology [4]. In 1986, the first mab was licensed [2,5]. Sixteen years later, in

2002 the first human mab was approved by the U.S. Food and Drug Administration for

pharmaceutical use in humans [3]. In 2014 the annual global mab market was approximately

$20 billion, generated by roughly 30 products [2] and is estimated to grow to $125 billion with

around 70 products on the market by the year 2020 [5]. In parallel with the transition from small

to large molecular drug research and development, regulatory authorities introduced the new

quality guidelines Q8 – Q12 for drug development, listed in Table 2 [6–8].

Table 2: List of new ICH quality guidelines.

ICH Guideline Topic

Q8(R2) Pharmaceutical development

Q9 Quality Risk Management

Q10 Pharmaceutical Quality System

Q11 Development and Manufacture of Drug Substance

Q12 Technical and Regulatory Considerations for Pharmaceutical Product Lifecycle Management

The new guidelines propose an enhanced approach for drug development compared to the

traditional. Where the traditional way defines set values and operational ranges for process

parameters, based on tests and reproducibility studies to reach their acceptance criteria the

enhanced approach integrates quality directly into the process [8]. This is achieved by

elaborating a quality risk management (QRM) based on scientific evaluations throughout the

whole product lifecycle, a so called quality by design (QbD) approach. The QbD, illustrated in

Figure 2, includes a summary of characteristics of a drug product (DP) for its ideal quality, the

so called quality target product profile (QTPP). The QTPP contains for example information

regarding the indication, the mechanism of action, the dosage form or the dosage strength [9].

Based on the QTPP a critical quality attribute (CQA) assessment is performed, listing any

physical, chemical, biological, or microbiological property of a DP having an influence on the

quality with regard to activity/potency, pharmacokinetics/-dynamics (PK/PD), immunogenicity

and/or safety. Examples of CQAs are viral impurity, host cell proteins, glycosylation or high

molecular weight species. Afterwards operation parameters affecting CQAs, the critical


2

process parameters (CPP), are defined and tested by a design of experiment (DoE). Based

on the results of these experiments, the criticality of the CQAs is adapted. Examples for CPPs

are: temperature, time and pH. Based on the findings a working area of each CPP is defined,

called the design space. At last a control strategy is introduced, ensuring the defined quality of

the final DP. The whole process is maintained and updated through the whole lifecycle and

should lead to an improved product understanding and ultimately higher patient safety [10–

14].

Figure 2: Illustration of the Quality by Design strategy [15,16].

To conduct a QRM potential risks are identified first, their likelihood is rated and the

consequences are assessed. Based on those points an impact factor for each risk is calculated

and an appropriated follow-up action is evaluated. Such can be “no action required”, “further

investigation required”, “close monitoring” or “risk prevention”. The basis for such a risk

identification for quality attributes emerges from four sources, illustrated in Figure 3: literature

& in-silico data, in-house knowledge and product comparison, project related experimental

data and specific clinical experience. The information availability is usually high for literature &

in-silico data and decreases for in-house knowledge & product comparison. Project related

experimental data and specific clinical expertise, which have both a high certainty score and

cost & time investment, are often only available in late development phases. Therefore, the

relation of effort to information gain has to be evaluated thoroughly.

Quality target product profiles (QTPP)

Critical process

parameters (CPP)

Critical quality attributes

(CQA)

Design

Space

Control

Strategy

Verification

Validation Iterative process throughout the lifecycle


3

Figure 3: Potential information sources for the quality risk management [15].

The risks considered for impact assessment of quality attributes during drug development are

divided into four categories, listed in Figure 4.

Figure 4: The four quality categories for the risk assessment quality attributes [15,17–20].

This QbD approach slowly finds its way into pharma companies, where chemistry,

manufacturing and control strategies are adapted and updated accordingly. For an adequate

risk evaluation a fundamental molecule, mechanism and process understanding is necessary.

During the CQA assessment related to the active pharmaceutical ingredient in-depth

understanding of potential structural and molecular modifications and their impact are

evaluated. As literature and in-silico data are generally rated quiet uncertain and are often not

Literature & in-silico data

In-house knowledge & product comparison

Project related experimental

data

Specific clinical

experience

Cost and duration

Certainty

Availability

Biological activity/potency describes the ability of a molecule/drug to induce a certain response in a biological system at a defined concentration [17].

Pharmacokinetics (PK) is the study of progression of a drug during absorption, distribution, metabolism and excretion. Pharmacodynamics (PD) is the study of the biological effect caused by the drug at the site of

action [18].

Immunogenicity describes the ability of a substance inducing an immunological reaction in an organism This includes the formation of anti-drug antibodies [19].

Safety according to the World Health Organization are biological, chemical or physical agents or operations that are reasonably likely to cause illness or injury [20].


4

even available for new biological entities, molecular entity-specific in-house knowledge has to

be expanded and tools to gain project-specific experimental data have to be established. So

far biopharmaceuticals have been exposed to controlled stressed conditions, like heat,

oxidation or shearing stress, for a defined period of time and degradation products were

analyzed. This approach may introduce multiple, uncontrolled modifications leading to vague

conclusions. Therefore, the development of new tools to modify a drug very specific (and

prevent undesired modifications) is necessary.

The topic of this study is to start establishing tools for the CQA assessment of

biopharmaceuticals. Potential modification are for the most part dependent on the molecular

structure of the drug substance and may vary from one project to another. As a starting point

the immunoglobulin G (IgG) 1 subclass was used in this thesis. IgG1 is currently the most used

subclass of mabs in drug development. A possibility of targeted modification is obtained

through enzymes. Especially in protein-related sciences enzymes offer a wide variety of

applications. Unfortunately, commonly used endoproteinase, such as trypsin, Lys-C, papain

or pepsin still tend to undesired over-digestion. In recent years new endoprotease were

discovered which promise to cleave IgGs sequence specific without the risk of over-digestion

[21–24]. The goal of this thesis is to screen and improve the enzymatic digestion of a model

IgG1 to obtain Fab, Fc, LHF and F(ab’)2 fragments, respectively the chitobiose cleaved IgG,

shown in Figure 5, and isolated the corresponding products. Undesired modifications have to

be prevented as far as possible. For a copy of the objective setting agreement please see

appendix page X.

Figure 5: Selected digestion products to be evaluated during the master thesis including their final purpose in the CQA assessment.

F(ab‘)2

Bivalent binding (potency) Monovalent binding (potency)

Monovalent binding (potency)

LHF

Fab Fc

(Fuc-)GlcNAc-IgG

Glycosylation pairing (PK/PD)

Fc methionine oxidation distribution

Change in higher order structure

FcRn, FcγR and C1q binding (PK/PD)

ZHAW LSFM, 2017 Theoretical background Martin Alexander Rappo

5

2 Theoretical background

2.1 Antibodies

Antibodies (Ab), also called immunoglobulin (Ig), are glycoproteins and play an essential role

in the adaptive immune system. Abs are naturally either transmembrane proteins, as B-cell

receptors or soluble proteins released from plasma cells after an antigen recognition. Special

properties of such Abs are the high specificity and affinity to bind a certain antigen and initiate

an effector mechanism [25].

2.1.1 Structure

The basic structure of an Ab, illustrated in Figure 6, is Y-like shaped and consists of the stem,

the so called fragment crystallizable (Fc) responsible for the effector function (explained in

chapter 2.1.4) and the two arms, the so called fragment of antigen-binding (Fab) responsible

for the specific antigen binding [26]. Abs are axial symmetrical hetero-tetramers containing two

light chains (LC) and two heavy chains (HC). The LC consist of two domains of roughly

12.5 kDa, the variable (VL) and constant (CL) domain. The HC consists of four domains, also

one variable (VH) and three constant domains (CH1, CH2 and CH3). The two HC are interlinked

between CH1 and CH2, in the so called hinge-region via disulfide bridges [25,27]. Similarly, the

LC and the HC are liked via a disulfide bridge in the Fab region [25]. Additionally, each domain

contains an intra-chain disulfide bond [28]. The CH2 bears an N-glycosylation at position

Asn297, which plays an important role for the effector function, protein stability and the ability to

aggregate [29,30]. The highly specific binding ability of Igs origins from the hypervariable

regions within the variable region, the so called complement determining region (CDR) [25].

Each variable domain contains three CDRs and the CDR of the HC and LC lie close together

through the three-dimensional arrangement forming the so called antigen-binding pocket [31].


6

Figure 6: Illustration of the IgG1 structure with light chain in apricot and heavy chain in blue.

Figure 7: Ig Fab with CDR in red and a hapten as epitope [32].

Figure 8: 3D surface structure of an IgG1 [33].

Fragment antigen-binding (Fab): antigen binding

Hinge region: flexibility and connection

Variable region

Constant region

IgG1 structure

Fragment crystallizable

(Fc): Effector function

Gly

co

syla

tion

Heavy chain

Light chain

Fab region with hapene in binding pocket

IgG1 3D structure

LC HC

Glycosylation


7

In humans there are two different types of LCs, the κ and the λ chain and five different types

of HCs, the γ, δ, ε, α and μ chain. Depending on the HC type, antibodies are classified into five

classes or isotypes, namely IgG, IgD, IgE, IgA and IgM [25–27]. IgG have the highest half-life

of roughly three weeks and are therefore the most interesting class for therapeutic applications

[25]. The IgG class is further divided into four subclasses [25,27,34,35], depending on the

preferable antigen target, hinge region length, disulfide connection [26,28], serum

concentration, half-life, placental transfer potential, complement activation and Fc receptor

binding. Those subclasses are numbered ascending from highest to lowest relative abundance

(γ1, γ2, γ3, γ4) in human serum. IgG1 is usually found in the highest relative abundance in the

human serum compared to other subclasses and targets mainly soluble and membrane

proteins. Additionally, it has compared to other subclasses a unique disulfide connection

between LC and HC [34] and binds efficiently to the IgG-Fc receptor (FcγR) [36].

2.1.2 Fragment nomenclature

In Figure 9 common antibody fragments are listed.

Figure 9: Nomenclature of common IgG fragments [37–42].

Immunoglobulin G

IgG

Fragment antigen binding

Fab

F(ab‘)2

Fragment crystallizable

Fc

Fc/2

Large hinge fragment [40]

LHF

Fab‘ Reduced/half IgG rIgG/hIgG

Fd‘

The term Fabc is rarely used and can have two different meanings. Either as

equivalent for the intact IgG [37,42] or the IgG without CH3 domains [41].


8

2.1.3 Glycosylation

Protein glycosylation is a highly complex post-translational modification but crucial for folding,

conformation, stability & half-live, solubility, pharmacokinetics, biological activity and

immunogenicity [29,43]. Ig of the subclass γ1 undergo glycosylation in the endoplasmic

reticulum and Golgi network and contain a conserved glycosylation at approximate position

Asn297 of both CH2 domains in the Fc region. Roughly 20% of mabs contain an additional N-

glycosylation in the variable region. Usually glycosylation of Abs consists of a complex bi-

antennary glycan with core-fucosylation [44]. An illustration of transgenic animal respectively

Chinese hamster ovary cell N-glycosylation structures is shown in Figure 10. Minor size and

structural differences of the glycan from one protein to another lead to an unequal mixture,

which is called glycoform heterogeneity [44,45]. Specifically for IgG the glycosylation has an

effect on thermo-, unfolding- and protease-stability, impacts the aggregation tendency [46] and

influences the effector functions, especially the antibody-dependent cell cell-mediated

cytotoxicity (ADCC) and the complement dependent cytotoxicity (CDC) [29,46,47].

Figure 10: Fc glycosylation structure of transgenic animal/Chinese hamster ovary cells.

2.1.4 Mechanism of action

Abs can have different effects on an antigen. Important for a functional Ab is the ability to bind

a certain epitope with the binding pocket and initiate an effector function from the Fc region.

With both of these properties Abs act as adapter and bring antigen and the immune cell

together. Especially IgG1 and IgG3 are known to initiate such effector mechanisms [34].

Therapeutic mabs on the other hand usually do not depend on an effector function and in some

cases those are even undesired. A summary of the most important effector functions are listed

in the appendix on p. II.

2.2 Enzymes

Enzymes are proteins with a wide function of catalyzing bio-chemical reactions in an organism.

Some enzymes catalyze a certain reactions high specific and under physiological conditions.

Fc glycosylation of Chinese hamster ovary cells

Chitobiose core


9

Enzymes also have a high relevance in industrial and diagnostic applications. A summary of

commonly used enzymes for the mab analytic is listed in the appendix on p. III.

2.2.1 Immunoglobulin-degrading enzyme of Streptococcus pyogenes

Name Immunoglobulin-degrading enzyme of Streptococcus pyogenes

Abbreviation IdeS, streptococcal Mac-1

Catalytic type Cysteine [48]

Family Protease C66 [49]; CA clan [50]

Organism Streptococcus pyogenes

Molecular weight 36.2 kDa [50]

Molar absorbance 1.362 (calculated) [51]

Figure 11: Illustration of IdeS.

Streptococcus pyogenes is a bacteria causing strep throat and impetigo in humans [52].

Recent studies revealed its secreted cysteine endopeptidase (IdeS or streptococcal Mac-1) to

naturally protect the bacteria from the host’s immune system by cleaving surrounding Ig at the

lower hinge region.

Figure 12: IgG fragmentation of IdeS.

IdeS contains similar structural features as papain but shows only minor sequence homology.

Both can structurally be divided into a left-hand (L) and right-hand (R) domain and the catalytic

site is located at the interface in between. High specificity of substrate recognition towards the

hinge region of IgG is a unique feature of IdeS. So far only few accepted sequences of the Ig

hinge region have been identified, shown in Figure 13 [50]. This high specificity, which prevents

over-digestion is exploited for bioanalytical applications.

Catalytic domain

L R


10

Figure 13: Accepted sequence for the IdeS cleavage. Identical amino acid in black, closely chemically similar in blue, weakly chemically similar in sienna and non-similar in apricot.

The catalytic mechanism proposed for the IdeS digestion is similar to the one of the papain

cysteine protease superfamily, forming a thiolate-imidazolium ion with the catalytic dyad Cys94

and His262, shown in Figure 14 [50]. Through proton transfer from His262 to Cys94 the enzyme

is activated. The thiolate of Cys94 attacks the carbonyl of the substrate and forms an

enzyme-substrate complex. The scissile peptide bond of the substrate is broken and the C-

terminal fragment is released, where the remaining chain forms together with the enzyme a

thioester, the so called acyl-enzyme intermediate. The thioester is hydrolyzed in a

subsequent step and the N-terminal chain is also released [53].

Figure 14: Proposed catalytic cycle of the IdeS cleavage.

Asp284 and Asp286, close to the catalytic pocket, have been identified to have a positive impact

on the catalytic activity, by facilitating the correct three-dimensional orientation resp. creating

240 250

....|............|....

IgG1 ...CPAPELLG | GPSVF...

IgG2 ...---APPVA | GPSVF...

IgG3 ...CPAPELLG | GPSVF...

IgG4 ...CPAPEFLG | GPSVF...

Accepted IdeS cleavage sequence


11

a suitable electrostatic milieu. Additionally, Lys84, which is stabilized through the hydrogen

bond-salt link of Asp286, forms together with the active site Cys94 a stabilizing pocket for the

substrate, the so called oxyanion hole [50]. This is illustrated in Figure 15.

Figure 15: Catalytic pocket and stabilizing effects.

Initial kinetic studies showed a non-Michaelis-Menten behavior, but a curve of positive

cooperativity [54]. Later studies revealed an underlying two-step Michaelis-Menten kinetics,

where the first reaction is 100 times faster than the subsequent [55].

Genovis®, a life science company, offers recombinant IdeS in various forms, e.g. as lyophilisate

(FabriCATOR®), immobilized on agarose (FragIT™) or in combination with a CaptureSelect™

column (FragIT™ kit) including digestion procedure.

2.2.2 Lysine gingipain of Porphyromonas gingivalis

Name Lysine gingipain

Abbreviation Kgp

Enzyme class EC 3.4.22.47 [54]

Catalytic type Cysteine

Family Protease C25; CD clan [53]

Organism Porphyromonas gingivalis

Molecular weight 187 kDa (native) [56] 48 kDa (recombinant; catalytic and immunoglobulin superfamily-like domain) [53,57]

Molar absorbance 1.971 (recombinant, calculated) [58]

Figure 16: Illustration of recombinant Kgp.

The periodontal disease is a wide spread bacterial infection of the gums and the bacteria

supposedly to be the main cause is Porphyromonas gingivalis. P. gingivalis expresses two

Hole with catalytic pocket

Immunoglobulin superfamily- like domain


12

types of cysteine peptidase gingipain, the lysine (Kgp) and the arginine gingipain (RgpA and

RgpB). These two enzymes contribute to a large part to the extracellular proteolytic activity

responsible for its high virulence [57]. Structurally Kgp consists of five domains; the signal

peptide, a pro-domain, the Lys-specific catalytic subdomain, an immunoglobulin superfamily-

like domain, and the C-terminal domain [53,57]. Kgp cleaves multiple substrates after Lys [53]

and also plays an important role in evading the host’s immune system by degrading Ig at the

upper hinge region, illustrated in Figure 17 [54].

Figure 17: IgG fragmentation of Kgp and IgdE.

For Fc glycosylated IgG1 only one cleavage site, above the hinge region, has been identified,

shown in Figure 18. For IgG1 without glycosylation a second cleavage site in the Fc region

was observed [22]. Also IgG3 have been reported to be cleaved by Kgp but not as specific as

IgG1. For IgG2 and IgG4 no cleavage sites have been identified [54].

Figure 18: Accepted Kgp cleavage sequence of IgG1 [22].

The proposed mechanism is similar to IdeS Cys/His dyad involving Cys477 and His444, but

experiments have shown the importance of Asp388 for the catalytic cleavage [57]. Typical for

cysteine proteases is an activity increase in a mild reducing environment, like glutathione or

cysteine [53]. Kinetic studies revealed that the digestion does not follow the Michaelis-Menten

kinetic, but rather a mechanism of positive cooperativity [54], usually meaning the first ligand

binding to the enzyme simplifies a subsequent binding [59]. Similar to IdeS, Kgp kinetics could

also underlie a consecutive two-step Michaelis-Menten kinetic, but this theory was not tested

yet. It is reported that the digestion can be stopped by drastically decrease pH or addition of

iodoacetic acid resp. Nα-Tosyl-L-lysine chloromethyl ketone [54]. The mayor advantage of Kgp

over commonly used papain or Lys-C is the high specificity without the risk of over-digestion.

Genovis® offers a recombinant Kgp in a kit including a cysteine solution as reducing agent,

called GingisKHAN®.

230 240

.....|............|...

...KSCDK | THTCPPCP...

Accepted Kgp cleavage sequence of IgG1


13

2.2.3 Immunoglobulin G degrading enzyme of Streptococcus agalactiae

Name Immunoglobulin G-degrading Enzyme

Abbreviation IgdE

Catalytic type Cysteine

Family Protease C113, CA clan [60]

Organism Streptococcus agalactiae

Molecular weight 70 kDa (recombinant)


Figure 19: Illustration of IgdE.

The immunoglobulin G degrading enzymes (IgdE) of the Streptococcus species, known for

their virulence in several species, were recently discovered and have been identified to help

the bacteria evade the host’s immune system by degrading surrounding Ig. Exceptional for

those enzymes is their high specificity towards IgG of the streptococci’s main host and distinct

cleavage site. One IgdE, degrading human IgG1, origins from S. agalactiae and cleaves the

Ab similar to papain, above the hinge region, shown in Figure 20, but without further cleavage.

The involved mechanism is the same as for Kgp and is formed by the catalytic triad Cys302

His333 and Asp348. Several applicable cysteine protease inhibitors like, Z-LVG-CHN2 and

iodoacetamide have been reported [36,49,55].

Figure 20: Accepted IgdEagalactiae cleavage sequence of human IgG1 [36].

2.2.4 Endo-beta-N-acetylglucosaminidase of Streptococcus pyogenes

Name Endo-beta-N-acetylglucosaminidase

Abbreviation EndoS2

Family Glycoside hydrolases 18

Organism Streptococcus pyogenes group A (GAS) (Serotype M49) [62]

Molecular weight 95 kDa (native) [63] 92 kDa (recombinant) [24]


Figure 21: Illustration of EndoS2.

Streptococcus pyogenes, as explained in chapter 2.2.1, possesses another way to evade the

human immune system. The bacteria produces, next to IdeS, the endoglycosidase EndoS.

Catalytic pocket

230 240

.....|............|...

...KSCDKT | HTCPPCP...

Accepted IgdE cleavage sequence of IgG1

Catalytic site


14

This enzyme is hydrolyzing the glycosidic bond between the two acetylglucosamine (GlcNAc)

in the chitobiose core structure of IgG impairing an effector function and increasing bacterial

survival [62].

Figure 22: Chemical structure of the chitobiose core.

The EndoS is conserved in almost all identified S. pyogenes serotypes with the exception of

serotype M49 GAS strain. The serotype M49 GAS expresses instead the enzyme EndoS2, with

similar function but a sequence homology of only 37% [63]. Both enzymes cleave complex

type glycans. However EndoS2 cleaves in addition also high-mannose and hybrid type glycans

[65]. All human IgG subclasses and Abs of several species, such as mouse, rat, monkey,

sheep, goat, cow and horse have been identified as substrate. Importantly, the enzyme works

only on native proteins, which indicates a protein-protein interaction. Through a sequence

alignment and later active-site mutation the catalytic Glu186 and several Trp have been

identified to be of importance for the digestion [62]. Next to IgG only one other protein, the α1-

acid glycoprotein, has been identified to act as substrate and therefore EndoS and EndoS2 do

not show general chitinase activity.

Figure 23: Chitobiose cleavage of EndoS2.

Genovis® offers an EndoS (IgGZero™) and an EndoS2 (GlycINATOR®) ready-to-use kit,

containing the corresponding enzyme with a poly-his tag as lyophilisate or immobilized on

agarose.


15

2.3 Kinetics model for the Kgp digestion

To simplify result comparison and reduce enzyme consumption a model was applied for

studies on Kgp. This model is based on the reaction kinetics of a consecutive first-order

reaction as described below [66]. The kinetic order was not experimentally determined but

assumed as a first order reaction.

Figure 24: Proposed consecutive reaction scheme of an IgG and Kgp.

The IgG reacts in a first reaction step with the kinetic constant k1 to the LHF and a Fab and in

a subsequent step with k2 to two Fabs and one Fc. k1 is calculated using Equation 1 and kinetic

constants for different digestion parameter are compared. The higher the kinetic constant the

faster the digestion.

Equation 1: Determination of the kinetic constant k1 [66].

𝑘1 =1

𝑡ln[𝐴]0[𝐴]𝑡

It is assumed that k1 and k2 are almost the same, as neither the enzyme nor the substrate

change by a lot. Only the accessibility of the substrate of the second reaction is presumably

higher compared to the first and k1 is therefore slightly smaller than k2.

𝑘1 ≈ 𝑘2 𝑘1 < 𝑘2

The molar concentration of the intermediate product is calculated according to Equation 2.

Equation 2: Calculation of the intermediate product concentration [I].

[𝐼] =𝑘1

𝑘2 − 𝑘1(𝑒−𝑘1𝑡 − 𝑒−𝑘2𝑡)[𝐴]0

The molar concentration of the product is calculated according to Equation 3.

Equation 3: Calculation of the product concentration [P].

[𝑃] = (1 +𝑘1𝑒

−𝑘2𝑡 − 𝑘2𝑒−𝑘1𝑡

𝑘2 − 𝑘1) [𝐴]0

The time point where the intermediate product is highest is calculated according to Equation

4.

k1 k2

A I P


16

Equation 4: Calculation of the time point of the maximal intermediate concentration is reached.

𝑡[𝐼]𝑚𝑎𝑥 =1

𝑘1 − 𝑘2ln (

𝑘1𝑘2)

Transformed back to purity the digestion progress is expected to behave as shown in Figure

25.

Figure 25: Simulated Kgp digestion progress based on a consecutive first-order kinetic.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Purity

in A

%

Time in min

Simulated Kgp digestion progress for a consecutive first order reaction kinetic

IgG1

LHF

Fc

Fab

ZHAW LSFM, 2017 Methods and materials Martin Alexander Rappo

17

3 Methods and materials

3.1 Instruments, consumables, reagents and software.

High performance liquid chromatography (HPLC) was performed on instruments from Agilent,

the capillary electrophorese (CE) on BECKMAN-COULTER instruments and the content was

measured on a Nanodrop™ by Thermo Scientific. All enzyme were purchased from Genovis®.

Analytical data evaluation was performed on Chromeleon™ 6.8. A detailed list of all

instruments, consumables, reagents and software can be found in the appendix on p. IV - VI.

3.2 Model antibody

During this studies a model IgG1 was used. The used mab is developed by Novartis and

blinded in this report for the protection of the company’s intellectual property. It is simply

referred as mIgG1 standing for model IgG1. The mIgG1 belongs to the IgG1 subclass with a κ

LC and the standard mab solution is formulated at approximately 200 mg/mL in 5 mM L-

histidine/L-histidine hydrochloride at pH 5.8.

3.3 Buffer preparation

Table 3: List of used buffer.

Name Composition

CaptureSelect™ binding buffer 10 mM sodium phosphate, 150 mM sodium chloride pH 7.4

CaptureSelect™ elution buffer 100 mM glycine pH 3.0

CE-SDS NR gel buffer Beckman Coulter gel buffer, Cat. no. 391163

CE-SDS NR sample buffer 100 mM tris base, 10 g/L SDS, pH 7.00

CE-SDS red gel buffer Beckman Coulter gel buffer, Cat. no. 391163 + water (1:1.2)

CE-SDS red sample buffer 50 mM tris base, 5 g/L SDS, pH 8.5

cOmplete™ solution 1 Tablet in 50 mL Kgp DB

EndoS2 digestion buffer 10 mM sodium phosphate, 150 mM NaCl, pH 7.4

Formulation buffer 5 mM L-histidine/L-histidine-HCl pH 5.8

IdeS digestion buffer 10 mM sodium phosphate, 137 mM NaCl, 2.7 mM KCl, pH 6.0

IgdE digestion buffer 150 mM sodium phosphate pH 7.0

Kgp digestion buffer 100 mM tris pH 8.0/7.3/6.0

Protein L binding buffer 20 mM sodium phosphate, 150 mM NaCl, pH 7.2

Protein L elution buffer 100 mM sodium citrate, pH 2.0 – 3.5

Reducing solution 20 mM cysteine


18

3.4 Physicochemical methods

Several physicochemical techniques were used for sample analysis and fragment isolation.

Detection was always by UV and the digestion progress was evaluated according to purity by

capillary electrophoresis (CE).

3.4.1 Cation exchange chromatography

Cation exchange chromatography (CEX) is a liquid chromatography technique and belongs to

the ion-exchange chromatography. It is used to separate species with a different net surface

charge on an ionic stationary phase. It was used to detect the rate of deamidation, shown in

Figure 26, by monitoring purity of the acidic species.

Figure 26: Deamidation of asparagine.

During this study a salt based separation gradient was applied with parameter listed in Table

4.

Table 4: Applied CEX parameter.

Parameter Setting

Sample injection concentration 1 mg/mL

Sample injection volume 50 μL

Sample temperature 5°C

Mobile phase A 25 mM sodium phosphate, pH 6.0

Mobile phase B 25 mM sodium phosphate, 250 mM sodium chloride, pH 6.0

Flow rate 1.0 mL/min.

Column ProPac™ WCX-10


19

Column temperature 25°C

Detection 220 nm

Gradient Time in min %B

0 10

30.0 37

30.1 100

32.0 100

32.1 10

40.0 10

An example chromatogram is displayed in Figure 27.

Figure 27: Example cation exchange chromatogram.

3.4.2 Size exclusion chromatography

Size exclusion chromatography (SEC) separates species according to their hydrodynamic

radius. This technique was used to separate and collect the native IgG1, LHF and Fab fractions

using the parameters listed in Table 5.

Table 5: Applied SEC parameter for fraction collection.

Parameter Setting

Sample injection amount 40 – 400 μg

Sample injection volume 10 – 100 μL

Sample temperature 5°C

Mobile phase 150 mM potassium phosphate, pH 6.5

Flow rate 0.4 mL/min.

12 13 14 15 16 17 18 19 20 21 22

Absorb

ance in m

AU

Time in min

Example CEX chromatogram of mIgG1

Acidic variants Main Basic variants


20

Column TSKgel G3000SWXL; 2 – 3 columns in series

Column temperature 30°C

Detection 210 nm

Run time 55 – 85 min

3.4.3 Capillary electrophoresis – sodium dodecyl sulfate

CE is, like the polyacrylamide gel electrophoresis (PAGE), a physicochemical separation

technique but instead of a planar gel the separation occurs in a thin capillary [67]. Sodium

dodecyl sulfate (SDS) denatures the protein and ensures a similar charge to mass ratio of all

proteins, thus migration time increases with higher molecular weight. The technique is superior

to SEC in regard to peak resolution of fragments but has the disadvantage that sample

fractionation is not possible.

Non-reducing conditions

For non-reducing (NR) CE-SDS, the protein sample was pre-diluted to 6 mg/mL in water, 20 μL

of this solution were further diluted in 75 μL sample buffer and 5 μL 250 mM iodoacetamide

solution were added for alkylation. Afterwards the protein was denatured at 70°C for 10 min,

the solution was cooled down to room temperature (rt), centrifuged and analyzed by CE-SDS

with parameters listed in Table 6.

Table 6: List of CE-SDS NR parameter.

Parameter Setting

Auto sampler temperature 20°C

Injection type Outlet

Injection voltage 5 kV

Capillary length 30 cm

Capillary diameter 50 µm

Capillary temperature 25°C

Polarity Positive

Voltage 15 kV

Detection 214 nm

An example CE-SDS NR electropherogram of the Kgp digestion, with assigned peaks is shown

in Figure 28.


21

Figure 28: Example CE-SDS NR electropherogram of the Kgp digestion.

Reducing conditions

For reducing (red) CE-SDS the protein sample was pre-diluted to 6 mg/mL in water, 20 μL of

this solution were further diluted in 75 μL sample buffer and 5 μL 2-mercaptoethanol were

added for reduction. Afterwards the protein was denatured at 70°C for 10 min, the solution was

cooled down to rt, centrifuged and analyzed by CE-SDS with the parameters listed in Table 7.

Table 7: List of CE-SDS red parameter.

Parameter Setting

Auto sampler temperature 20°C

Injection type Inlet

Injection voltage 5 kV

Capillary length 30 cm

Capillary diameter 50 µm

Capillary temperature 25°C

Polarity Reverse

Voltage 14.2 kV

Detection 214 nm

An example CE-SDS red electropherogram of the EndoS2 digestion, with assigned peaks is

shown in Figure 29.

15 30 45 60 75 90 105 120 135 150 165 180 195

Absorb

ance in m

AU

MW in kDa

Example CE-SDS NR electropherogram of the Kgp digestion


22

Figure 29: Example CE-SDS red electropherogram of the EndoS2 digestion.

3.5 Digestion and isolation

3.5.1 Fab, Fc and LHF

Figure 30: Flow chart of the Fab, Fc and LHF generation and isolation.

Stability evaluation in digestion buffer with and without reducing solution

Increased pH and temperature are usually used to enforce asparagine deamidation

in proteins [68] and reducing agents lead to breaking of disulfide bridges and loss of

tertiary structure. As all three conditions are present in the Kgp digestion, specified

in the Genovis protocol the rate of deamidation and reduction was evaluated.

15 17 19 21 23 25 27 29 31 33 35

Absorb

ance in m

AU

Time in min

Example CE-SDS red. electropherogram of the EndoS2

digestion

Undigested Digested

Dig.

AC

BE

SEC

BE

Control

Digestion

IPC

Protein L affinity chromatography

Buffer exchange

Buffer exchange

Size exclusion chromatography

Purity analysis

IPC

IPC

In-process control

In-process control

In-process control


23

Deamidation stability: 20.0 µL mIgG1 solution were diluted in 788 µL digestion

buffer (DB) to 5.00 mg/mL (final pH 7.87), split in two equal portions, kept at rt resp.

at 37°C/300 rpm for 1, 2 and 24 h and were analyzed by CEX and CE-SDS NR.

Reduction stability: 20.0 µL mIgG1 solution were pre-diluted in 60.0 µL DB to

50.5 mg/mL and 24.8 µL of this pre-dilution solution were further diluted in 50 µL

Genovis reducing solution (RS) and 175.2 µL DB, kept at 37°C/300 rpm for 2 h and

analyzed by CE-SDS NR.

Digestion progress of Kgp and influence factors

Enzyme ratio: To assess the initial digestion rate and test the applicability of the

model, the initial digestion was performed according to the provided protocol from

Genovis [22]. 20.0 µL mIgG1 solution were diluted in 60.0 µL DB and 24.8 µL

(1’250 µg) of this solution were further diluted in 50.0 µL DB, 125.8 µL Kgp 10 U/µL

(1’258 U) and 50.0 µL fresh RS to 5.00 mg/mL mIgG1. The digestion solution was

incubated at 37°C/300 rpm for 5, 15, 30, 60 and 120 min and tested by CE-SDS NR.

The 5 min sample was injected twice to prove that digestion stops in presence of 1%

SDS.

In the next experiment, the enzyme amount was reduced by a factor of ten. 20.0 µL

mIgG1 were diluted in 60.0 µL DB and 24.8 µL (1’250 µg) of this stock solution were

further diluted in 162.7 µL DB, 12.5 µL Kgp 10 U/µL (125 U) and 50.0 µL fresh

reducing agent to 5.00 mg/mL. The digestion solution was incubated at

37°C/300 rpm for 5, 15, 30, 60 and 120 min and tested by CE-SDS NR. After first

data evaluation the experiment was repeated with pull points from 6 to 15 min every

minute.

Enzyme inhibition: As the intermediate product LHF is of interest slowing down or

stopping the digestion progress would simplify the procedure. In this experiment the

influence of cooling the digestion mixture to rt resp. 5°C, acidifying to pH 5.5 and/or

addition of cOmplete™ protease inhibitor was evaluated. As extreme pH changes

(e.g. by spiking TFA) might lead to undesired modifications, the maximum pH drop

was set to pH 5.5, where most IgG1 are stable.

20.0 µL mIgG1 were diluted in 60.0 µL DB and 5.00 µL of this pre-dilution, 2.50 µL

Kgp 10 U/µL and 10.0 µL fresh reducing agent were diluted in 32.5 µL corresponding

DB, digested for 15 min at 37°C/300 rpm, 22°C or 5°C and analyzed by CE-SDS NR.

Cysteine influence: To evaluate the influence of cysteine on the enzyme activity

several digestions at different concentration levels, origins and age were tested and


24

reaction constants compared. As a rate of reduction also the LC peak, at a relative

migration time (RMT) of 0.54, was observed through the digestion.

20.0 µL mIgG1 were diluted in 60.0 µL DB to 50.5 mg/mL and the digestion solution

were prepared as described in Table 8 with the corresponding RS, incubated for

15 min at 37°C/300 rmp and analyzed by CE-SDS NR.

Table 8: Dilution table of the cysteine influencing digestion.

µL

Genovis Self-made/fresh Self-made/7d-aged

2.0 mM 0.4 mM 2.0 mM 1.0 mM 0.2 mM 1.0 mM 0.2 mM

DB pH 8.0 37.5 41.5 37.5 40.0 42.0 40.0 42.0

mIgG1 50.5 mg/mL

5.0 5.0 5.0 5.0 5.0 5.0 5.0

Kgp 10 U/mL

2.5 2.5 2.5 2.5 2.5 2.5 2.5

RS 20 mM 5.0 1.0 5.0 2.5 0.5 2.5 0.5

Genovis 2.0 and 0.4 mM and fresh/self-made 2.0 mM were then repeated in a three

times larger volume and analyzed after 5, 15, 30, 60 and 120 min.

To evaluate the reaction kinetics of the Kgp digestion in absence of cysteine 9.9 µL

mIgG1 (50.5 mg/mL) were diluted in 85.1 µL DB and 5.0 µL Kgp (10 U/µL) were

added. Afterwards, the digestion mixture was incubated at 37°C/300 rpm for 15, 30

and 60 min and analyzed by CE-SDS NR.

pH influence: To be able to evaluate the impact of the pH and better control the

digestion rate mIgG1 was digested at pH 7.3 (slightly basic) and 6.0 (slightly acidic).

20.0 µL mIgG1 were diluted in 60.0 µL DB to 50.5 mg/mL and 9.9 μL of this pre-

dilution solution were further diluted in 75.1 μL DB pH 7.3 resp. 6.0, 5.0 μL Kgp

(10 U/μL) and 10.0 μL RS were added. Afterwards, the digestion mixture was

incubated at 37°C/300 rpm for 15, 30 and 60 min and analyzed by CE-SDS NR.

Digestion parameter optimization: To optimally use the reagents included in the

kit and digest to 20 – 30% LHF within a reasonable time frame the enzyme content

was set ten times lower compared to the standard protocol and the cysteine

concentration was reduced to 1.25 mM. pH 6.0 and 8.0 were tested on those

conditions. 20.0 µL mIgG1 solution were diluted in 60.0 µL DB to 50 mg/mL and

19.8 μL of this pre-dilution solution were diluted in 157.7 μL DB at pH 6.0 resp. 8.0,

10 μL Kgp solution (10 U/μL) and 12.5 μL RS were added, incubated at

37°C/300 rpm for 5, 15, 30, 60 and 120 min and analyzed by CE-SDS NR.


25

Fc isolation: To confirm the identity of the Fc peak (RMT = 0.73) a Protein L affinity

chromatography (AC) was performed. 20.0 µL mIgG1 were diluted in 60.0 µL DB to

50.5 mg/mL, 24.8 μL of this pre-dilution solution were further diluted in 187.7 μL DB,

12.5 μL Kgp (10 U/μL) and 25 μL RS were added and incubated for 2 h at

37°C/300 rpm. After full digestion a sample of 24 µL was taken for CE-SDS NR, the

remaining volume was loaded on a Protein L AC pre-equilibrated in binding buffer

(BB) and eluted with elution buffer (EB) according to Table 9. Acidic elution fractions

were neutralized with 60 µL 1 M tris pH 8.0 per milliliter fraction.

Table 9: Elution of Fc isolation.

Number of steps

Volume in mL Buffer Fraction volume in mL

Fraction name

1 1 BB 1.2 Flow through

4 1 BB 4 x 1 Wash 1 – 4

5 1 EB pH 3.5 5 x 1 Elution 1 – 5

The content of each fraction was determined by Nanodrop. The rinsing fraction was

concentrated using a 0.5 mL 10K-centrifugal filter, the concentration was determined

and the purity analyzed by CE-SDS NR.

SEC screening

To generate material for the preliminary trials 4.00 mg mIgG1 were digested

according to the optimized conditions and quenched after 30 min with 84 μL

trifluoroacetic acid 0.5%.

Number of columns: To evaluate the optimal number of serially connected columns

required to reach the needed resolution, 40 μg mIgG1 were injected on two and three

columns in series.

Column load: To evaluate the maximal injection amount without losing resolution

40, 120, 200, 300 and 400 µg mIgG1 were loaded on three columns in series.

Fractionation range: To evaluate an appropriate range to collect desired fractions

20 runs of 40 µg mIgG1 each were injected on three columns in series and collected

according to Table 10. Collected fractions were pooled, concentrated with 15 and

0.5 mL 10K-centrifugal filters and analyzed by Nanodrop and CE-SDS NR.

Table 10: SEC fractionation range.

Fragment name Fraction range in min.

IgG 59.0 – 61.0

LHF 63.5 – 66.0

Fab (+ Fc) 71.5 – 75.0


26

Protein L AC and SEC loading buffer evaluation: To generate fragments free of

Fc and to stop further fragmentation the digestion was repeated with 8.00 mg mIgG1.

After 30 min digestion, the mixture was loaded on a Protein L AC column and eluted

as described in Table 11. Acidic elution fractions were neutralized with 60 µL 1 M tris

pH 8.0 per milliliter fraction.

Table 11: Elution of Protein L AC.

Number of steps


Fraction name


4 1 BB 4 Wash

8 1 EB pH 2.0 8 Elution

The concentration of each fraction was measured. Elution fraction 2 – 8 were pooled,

concentrated with a 15 mL 10K-centrifugal filter to approximately 1 mL and analyzed

by Nanodrop. 60 µL (413 µg) were analyzed by SEC.

The sample was then buffer exchanged to SEC mobile phase with a 0.5 mL 10K-

centrifugal filter, rinsed and diluted with 2 x 500 µL mobile phase. Afterwards, the

content was determined. 8 fraction collection runs of 85 µL (8 x 406 µg) were

performed. The collected fractions were pooled, concentrated and buffer exchanged

into formulation buffer in a 4 mL 10K-centrifugal filter to approximately 150 µL

sample. The fractions were analyzed by Nanodrop and CE-SDS NR.

Sample stability: To evaluate the sample stability in mobile phase the injection

sample was analyzed before and after fraction collection (T1d) by CE-SDS NR.

Digestion reproducibility and fragment isolation: To evaluate the reproducibility

the digestion was repeated with 4.00 mg mIgG1. After 30 min a sample was

analyzed by CE-SDS NR. The remaining solution was loaded on a Protein L AC

column and eluted as described in Table 12. Acidic elution fractions were neutralized

with 60 µL 1 M tris pH 8.0 per milliliter fraction.

Table 12: Elution of Protein L AC for reproducibility.

Number of steps


Fraction name


4 1 BB 4 Wash

10 1 EB pH 2.0 10 Elution

The eluate was concentrated and buffer exchanged to approximately 150 µL and

rinsed with 250 µL formulation buffer. The concentration was measured and the

samples diluted with additional 800 µL formulation buffer to 3.8 mg/mL. 7 fraction

collection runs of 100 µL (7 x 380 µg) were performed. The fractions were pooled,


27

concentrated and buffer exchanged into formulation buffer in a 4 and 0.5 mL 10K-

centrifugal filter to approximately 42 µL and rinsed with 2 x 10 µL formulation buffer.

The fractions were analyzed by Nanodrop and CE-SDS NR.

Digestion progress of IgdE

During the time of this master’s thesis Genovis launched their IgdE product, called

FabALACTICA™. Its cleavage pattern is the same as Kgps, apart from the cleavage

site shift of one amino acid. Compared the GingisKHAN®, FabALACTICA™ is

superior in regard to digestion at neutral pH and absence of reducing agent. However

it has a lower digestion rate. In this experiment it was evaluated if IgdE is suitable for

LHF generation.

10.0 µL mIgG1 solution was pre-diluted in 40.5 µL IgdE DB to 40.0 mg/mL, 5.0 µL of

this pre-dilution solution (200 µg) were further diluted in 10.0 µL DB to 10 mg/mL and

5.0 µL IgdE 40 U/µL (200 U) were added, incubated at 37°C/300 rpm for 3 h and

analyzed by CE-SDS NR.

This procedure was repeated with 800 µg mIgG1 and analyzed after 30, 50, 60 and

90 min and a third time for time points at 5, 10, 15, 20 and 25 min.

3.5.2 F(ab’)2

Figure 31: Flow chart of the F(ab)2 generation and isolation.

IdeS digestion and Protein L AC pH 3.5 elution

The IdeS digestion was performed according to the Genovis protocol. 4.95 µL mIgG1

were diluted with 80.0 µL DB and 15.0 µL IdeS (66.7 U/mL) were added. The solution

was incubated at 37°C/300 rpm for 30 min. After the digestion a sample of 18.0 µL

was taken and analyzed by CE-SDS NR.

Afterwards the Protein L AC was performed according to GE Healthcare Instructions

29-0078-77 AC [69]. The digestion mixture was loaded on the column and eluted as

Dig.

AC

BE

Control

Digestion

IPC

Affinity chromatography

Buffer exchange

Purity analysis

In-process control


28

listed in Table 13. Acidic elution fractions were neutralized with 60 µL 1 M tris pH 8

per milliliter buffer.

Table 13: Elution of Protein L AC after IdeS digestion.

Number of steps


Fraction name

1 0.6 BB 0.68 Flow through

5 1.0 BB 5 x 1 Wash

5 1.0 EB 5 x 1 Elution

The content of each fraction was determined by Nanodrop. Flow through and wash

fraction 1 were pooled, concentrated with 4 and 0.5 mL 10K-centrifugal filter to

approximately 20 µL, rinsed with 20 µL formulation buffer and analyzed by CE-SDS

NR. All elution fraction were pooled, concentrated with 4 and 0.5 mL 30K-centrifugal

filter and rinsed with 2 x 10 µL formulation buffer. Concentration of both samples

were determined.

Over-digestion screening and Protein L AC pH 2.0 – 3.0 elution

To check whether the digestion is complete after 30 min and exclude over-digestion,

the digestion time was extended to 60 min. To evaluate the correct pH value to elute

the F(ab’)2 fragment from the column a stepwise elution gradient of pH 3.0, 2.5 and

2.0 was applied.

The digestion conditions are identical to the digestion before. After 60 min digestion

a sample of 18.0 µL was taken for CE-SDS NR and the remaining volume was

loaded on the Protein L column, washed and eluted according to Table 14.

Table 14: Protein L AC purification table.

Number of steps


Fraction name

2 (vial rinse) 0.45 BB 1 Flow through

4 1.0 BB 4 Wash

8 0.25 EB pH 3.0 5 pH 3.0 elution

3 1.0

8 0.25 EB pH 2.5 5 pH 2.5 elution

3 1.0

8 0.25 EB pH 2.0 5 pH 2.0 elution

3 1.0

The protein content of each fraction was determined. The flow through, wash and all

fractions of the identical EB (pH 2.5 and 2.0) were pooled individually, concentrated

with a 4 and 0.5 mL 10K-centrifugal filter and each sample was analyzed by

Nanodrop and CE-SDS NR.


29

FragIT™ kit digestion and fragment isolation

For simple enzyme removal and F(ab’)2 isolation Genovis offers the FragIT™ kit,

containing agarose-immobilized IdeS and a CaptureSelect™ column with an

immobilized 13 kDa lama antibody fragment [70]. The digestion and fragment

isolation were performed according to the FragIT™ kit instructions [70] including Fc/2

elution and all steps for maximal recovery. To evaluate if the columns could be

reused the whole procedure was repeated at the same day and after 17 days.

49.5 µL mIgG1 solution (10 mg IgG) were diluted with 450 µL DB to 20 mg/mL,

loaded on the digestion column and rotated end-over-end for 30 min at rt. Afterwards

the solution was collected through centrifugation at 100 g for 1 min and the column

was washed with 2 x 1 mL BB. The content was determined and a sample for CE-

SDS NR was taken. The solution was loaded on the CaptureSelect™ column, rotated

end-over-end for 30 min at rt, collected by centrifugation at 200 g for 1 min and

washed with 2 x 1 mL BB. The Fc/2 was collected through re-suspending with

2 x 1 mL CaptureSelect™ EB and collecting at 200 g resp. 1000 g at the final step.

Afterwards, the fractions (4.5 mL F(ab’)2 and 2.5 mL Fc/2-fraction) were

concentrated and buffer exchanged to formulation buffer with 4 and 0.5 mL 10K-

centrifugal filter to approximately 42 µL and rinsed with 3 x 20 µL. Samples were

analyzed by Nanodrop and CE-SDS NR. This procedure was repeated after 150 min

and 17 days.

3.5.3 Chitobiose cleavage

Figure 32: Flow chart of the EndoS2 digestion and fragment isolation.

EndoS2 digestion

The EndoS2 digestion was performed according to the Genovis protocol [24] at

20 mg/mL. For stability reasons the pH of the DB was set to 6.5.

Dig.

AC

BE

Control

Digestion

IPC

Affinity chromatography

Buffer exchange

Purity analysis

In-process control


30

5.90 µL mIgG1 (1.19 mg) were diluted in 24.1 µL DB and 30.0 µL EndoS2 40 U/µL

(1’200 U) added. The solution was incubated at 37°C/300 rpm for 30 min. After the

digestion a sample of 9.0 µL analyzed by CE-SDS red. The remaining digestion

mixture was buffer exchanged with a 100 K-spin filter, the membrane was rinsed with

2 x 20 µL fresh formulation buffer and the sample was analyzed by Nanodrop and

CE-SDS red.

Immobilized GlycINATOR®

To ensure complete enzyme removal immobilized EndoS2 (immobilized

GlycINATOR® MidiSpin) is used for digestion. The digestion was performed

according the Genovis standard protocol for A0-GL6-100 Version 17.1.1 [71]

including steps for maximum recovery.

49.5 µL mIgG1 (10 mg) were diluted in 450.5 µL DB, loaded on the equilibrated

column and rotated end-over-end for 30 min at rt. Afterwards, the solution was

collected through centrifugation at 100 g for 1 min and the column was washed with

2 x 1 mL BB. All fractions were pooled, concentrated and buffer exchanged into

formulation buffer with 4 and 0.5 mL 10K-centrifugal filters to approximately 42 µL.

The filter was rinsed with 2 x 20 µL formulation buffer and the sample was analyzed

by Nanodrop and CE-SDS red.

3.6 Ecology, safety and disposal

Standard laboratory safety equipment consisting of lab coat and protective goggles were used

during all tasks. During acid or base handling regular nitrile and during handling of organic

solvents butyl gloves were used. Contaminated solutions and consumables were discarded in

the chemical waste container. Powdered SDS and 2-mercaptoethanol were only handled

under the hood, as SDS creates irritating dust and 2-mercaptoethanol creates toxic fumes and

may be fatal in contact with skin [72].

ZHAW LSFM, 2017 Results and discussion Martin Alexander Rappo

31

4 Results and discussion

4.1.1 Fab, Fc and LHF


Deamidation stability: A negligible deamidation rate of < 0.5% was detected within

the two hours in DB for both temperatures. However after 24 h an increase of 0.7

and 2.6% was detected for rt and 37°C, respectively. The chromatogram is shown in

the appendix on p. VII. Neither fragmentation nor aggregation was detected by CE-

SDS NR during the measured time frame.

Stability in digestion buffer

Figure 33: Purity trend of the stability in digestion buffer by CEX (deamidation).

Figure 34: Purity trend of the stability in digestion buffer by and CE-SDS NR (fragmentation/aggregation).

Reduction stability: The relative area increase of the sum of reduced species (L,

H, HL, HH, and HHL) after 2 h at 37°C is below 0.1% and therefore consider non-

critical. The electropherogram overlay is displayed in Figure 35.

7.5

8

8.5

9

9.5

10

10.5

11

11.5

12

12.5

0 4 8 12 16 20 24

Purity

(Acid

ic)

in A

%

Time in h

Purity of acidic variants by CEX

95

95.5

96

96.5

97

97.5

98

98.5

99

99.5

100

0 4 8 12 16 20 24

Purity

(Main

) in

A%

Time in h

Purity of main by CE-SDS NR


32

Figure 35: CE-SDS NR electropherogram of the mIgG1 stability in digestion buffer with reducing agent.

Digestion progress of Kgp and influence factors

Enzyme ratio: The consistency of the CE-SDS NR double injection of the digestion

sample after 5 min proves, that the digestion stops in presence of 1% SDS. The

digestion according to the Genovis protocol is almost complete after 5 min (appendix

p. VII) and is therefore not suitable for the generation of LHF. With the adjusted

enzyme-protein ratio of 1:10 a maximum LHF concentration of 24 area% is reached

after 14 min and the digestion is complete after 2 h. Linear regression of the kinetic

ln([A]/[A]0) VS. time-plot, displayed in the appendix on p. VIII shows a coefficient of

determination (R2) of 0.9683. This high value is an indication of first order-like

behavior [66]. A comparison of the measured and the simulated values, presented

in Figure 36, shows a good overlap of the IgG1 and LHF curve. Fc and Fab

concentrations are slightly overestimated by the simulation. As the generation of LHF

has priority over Fc and Fab this model is considered fit for purpose.

7 9 11 13 15 17 19

Absorb

ance in m

Ab

Time in min.

CE-SDS NR electropherogram - Digestion buffer stability with reducing agent

T0 T120 min

LH HL HH

HHL

Ma

in


33

Figure 36: CE-SDS NR purity progress of the 1:10 Kgp-to-IgG digestion and comparison of the simulated values.

Enzyme inhibition: An evaluation of the reaction constant k1 (see Figure 37) and

tmax(LHF) (see Figure 38) revealed following inhibition potential:

Cooling > Acidify to pH 5.5 > cOmplete™

Figure 37: Reaction constant k1 in dependency of temperature for tested inhibitory factors (standard digestion, cOmplete™ and pH 5.5).

Figure 38: Calculated time point of maximal LHF concentration in dependency of temperature for tested inhibitory factors (standard digestion, cOmplete™ and pH 5.5).

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0 15 30 45 60 75 90 105 120

Purity

in A

%

Time in min

1:10 enzyme:protein digestion progress - measured VS. calculated

IgG1(measured)

IgG1(calculated)

LHF (measured)

LHF (calculated)

Fc (measured)

Fc (calculated)

Fab (measured)

Fab (calculated)

R² = 0.9946

R² = 0.9999

R² = 0.9654

0.00

0.02

0.04

0.06

0.08

0.10

0.12

5 15 25 35

Reaction c

onsta

nt

k1

Temperature / °C

Kinetic constant

0

20

40

60

80

100

120

5 15 25 35

t ma

x(L

HF

) in

min

Temperature / °C

Time of maximal LHF conentration

Inhibitory factor influence


34

The cOmplete™ protease inhibitor cocktail will be omitted for further studies, as the

inhibition potential is low. Additionally, as the composition is not clearly defined

tracking during fragment isolation and full removal cannot be ensured. For further

studies cooling to 5°C potentially in combination with acidifying and enzyme removal

for minimal molecule impact will be applied.

Cysteine influence: The comparison between digestions a different cysteine

concentrations, shown in Figure 39 and Figure 40, revealed a high influence of

cysteine on the Kgp digestion rate. The Kgp digestion rate is drastically reduced resp.

the digestion time is increased in absence of cysteine.

Figure 39: Kinetic constants of the Kgp digestion for different cysteine concentration of Genovis, self-made/fresh and self-made/aged RS.

Figure 40: Calculated time point of maximal LHF concentration of the Kgp digestion for different cysteine concentration of Genovis, self-made/fresh and self-made/aged RS.

The self-made cysteine solution had the highest digestion rate resp. lowest digestion

time for all concentrations compared to other conditions. The digestion using aged

self-made cysteine solution was slower than freshly made RS for all concentrations.

The highest digestion rate and lowest digestion time was determined at a

concentration of 1 mM, even though similar results were obtained for a concentration

of 2 mM cysteine. A possible explanation for the lower kinetic constant of Genovis’

RS compared to the self-made might be small differences regarding final

concentration. The cysteine in aged RS is most probably partly oxidized and not able

to provide the optimal reducing environment anymore.

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.20

Kin

etic c

onsta

nt

k1

Cysteine concentration in mM

Kinetic constant

0

50

100

150

200

250

Tim

e o

f [L

HF

] ma

xin

min

Cysteine concentration in mM

Time point of the maximal LHF concentration

Cysteine concentration influence


35

The purity progress of the peak at RMT 0.54 through the digestion time, displayed in

Figure 41, shows a comparable pattern for all tested cysteine concentrations. The

purity initially increases and later decreases again. As expected, the rate of reduction

increases with increasing cysteine content, but is throughout low. All experiments

using the RS of Genovis show a purity maximum at 30 min, whereas the self-made

RS shows its maximum after 60 min, which is significant higher compared to the

Genovis RS at equivalent concentration. This is an indication that this peak does not

only contain LC but also another unidentified species, which is formed and degraded

during the digestion.

Figure 41: Progress of the 25 kDa-peak formation during the IgG1 Kgp digestion at different cysteine concentrations.

Therefore the cysteine concentration is considered crucial for the digestion progress

and Genovis and self-made RS do not deliver equal results.

pH influence: The experiment showed that the pH has a large impact on the

digestion kinetic. At pH 8.0 the highest and at pH 6.0 lowest kinetic constant was

measured. This is visualized in Figure 42 and Figure 43.

0%

1%

2%

3%

4%

5%

6%

7%

8%

9%

0 15 30 45 60 75 90 105 120

Purity

in A

%

Time in min

Purity of the peak at a RMT = 0.54 through the digestion course at different cysteine conentrations

Genovis 4 mM

Genovis 2 mM

Genovis 1 mM

Self-made 2 mM


36

pH influence

Figure 42: Kinetic constants of the Kgp digestion at different pH.

Figure 43: Calculated time point of maximal LHF concentration of the Kgp digestion at different pH.

Digestion parameter adjustment: It was observed that upon reconstitution of the

cysteine according to the Genovis protocol, a suspension was obtained. The cysteine

only fully dissolved after further dilution for digestion. The purity of IgG, LHF, Fc and

Fab at pH 6.0 and 8.0 are visualized in Figure 44 and Figure 45, respectively. The

digestion at pH 6.0 is relatively slow and the maximum concentration of LHF is not

reached until the upper time limitation of 120 min. The digestion at pH 8.0 on the

other hand shows a maximum LHF concentration after 30 min followed by a shallow

decrease, allowing the scientist to perform the digestion in a reasonable time frame.

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

6 6.5 7 7.5 8

Kin

etic c

onsta

nt

k1

pH

Kinetic constant

0

5

10

15

20

25

30

35

6 6.5 7 7.5 8

Tim

e o

f [L

HF

] ma

xin

min

pH

Time point of the maximal LHF concentration


37

Digestion parameter adjustment

Figure 44: Digestion course of the parameter adjusted Kgp digestion at pH 6.0.

Figure 45: Digestion course of the parameter adjusted Kgp digestion at pH 8.0.

Fc isolation: The Protein L AC successfully bound the Fab and the Fc could be

collected with a recovery of 47%. Later Fab elution, displayed in Figure 46, was

minimal with a recovery of 20%. A possible explanation is, that the pH of the EB was

not low enough to reverse the binding.

Figure 46: Protein content of the Protein L affinity chromatography by Nanodrop.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0 15 30 45 60 75 90 105 120

Purity

in A

%

Time in min

pH 6.0

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0 15 30 45 60 75 90 105 120P

urity

in A

%

Time in min

pH 8.0

0.00

0.05

0.10

0.15

0.20

0.25

0.30

Concentr

ation in m

g/m

L

Step

Protein content of the Protein L affinity column fractions


38

The Fc could successfully be isolated with a final concentration of 3.43 mg/mL by

Protein L AC and the peak at RMT 0.73 in CE-SDS NR could be assigned. The

corresponding electropherogram is shown in Figure 47.

Figure 47: CE-SDS NR electropherogram of the Kgp digestion and concentrated Protein L flow through.

SEC screening

Number of columns: Two SEC columns in series are capable of separating IgG,

LHF and Fab. Three columns in series prolong the run time from 55 to 85 min,

increase the resolution and an additional peak between LHF and Fab could be

separated. Baseline separated is not reached in both cases. Corresponding

chromatograms are shown in Figure 48.

6 7 8 9 10 11 12 13 14 15 16

Absorb

ance in m

AU

Time in min

CE-SDS NR electropherogram before and after Protein L AC

Kgp digestion

Protein L flow through


39

Figure 48: SEC chromatogram overlay of 2 and 3 columns in series.

Column volume: Testing column loads from 40 to 400 μg mIgG1 showed no

significant loss of peak resolution. The SEC chromatogram overlay is displayed in

Figure 49.

Figure 49: SEC chromatogram of the injection amount evaluation.

Fractionation range: The chosen fractionation range was sufficient to isolate the

desired fragments. Importantly, the chromatographic overlays (Figure 50) revealed,

that the TFA spiking did not fully quench the digestion and the sample did

fragmentation continued at 5°C (shown by decreasing peaks at retention time 59.6

0 10 20 30 40 50 60 70 80

Absorb

ance in m

AU

Time in min

SEC chromatogram of the Kgp digest - 2 VS. 3 columns

2 columns 3 columns

IgG

1

LH

F

Fab

IgG

1

LH

F

Fab

55 60 65 70 75 80 85

Absorb

ance in m

AU

Time in min

SEC chromatogram - Increasing column load

40 µg 120 µg 200 µg 300 µg 400 µg


40

(IgG) and 64.5 min (LHF) as well as an increase in the Fab peak at retention time

72 min).

Figure 50: SEC chromatogram overlay of the fraction collection trial of 20 times 40 μg injection.

A summary of the isolated fragment is listed in Table 15.

Table 15: Summary of isolated fragments during SEC trials.

Fragment Volume in µL

Conc. In mg/mL

Mass in µg Recovery in %

Purity in A%

IgG1 120 0.10 11 4 86

LHF 120 0.18 22 14 75

Fab+Fc 120 0.81 64+32 45+30 98

Figure 51: CE-SDS NR overlay of the isolated IgG, LHF and Fab+Fc fractions.

9 10 11 12 13 14 15 16

Absorb

ance in m

AU

Time in min.

CE-SDS NR electropherogram - Purity of isolated fragments

IgG LHF Fab+Fc

71%

27%

75% 86%


41

Protein L AC and SEC loading buffer evaluation: The Protein L AC could

successfully remove the Fc part and the enzyme (see appendix p. VIII) and stop

further digestion. As a trade-off, a small amount of IgG, LHF and Fab was lost during

the wash steps. The intact IgG purity by CE-SDS NR was higher than expected.

Reason for this could possibly be a slower digestion rate and imply issues regarding

reproducibility. Telquel injection on the SEC showed a different chromatographic

pattern, displayed in Figure 52, than previous injections without prior Protein L AC.

A possible explanation for this behavior is the different buffer composition influencing

the hydrodynamic radius.

Figure 52: SEC chromatogram of the Kgp digestion fragment isolation with and without Protein L AC.

Prior buffer exchange into SEC mobile phase improved pattern resemblance on the

SEC column, shown in Figure 53. Peaks before the IgG peak indicate sample

instability of mixed sample but also isolated fragments. The lower UV trace of one of

the sample injections can be explained by a smaller injection volume. A summary of

isolated fragments is listed in Table 16. Reason for the low LHF purity is the relatively

high IgG content in the initial sample tailing into the LHF fraction.

Table 16: Summary of isolated fragments during SEC injection buffer evaluation.


Conc. In mg/mL


Purity in A%

IgG1 130 4.55 591 89 98

LHF 110 1.29 142 21 45

Fab 110 1.24 136 15 95

43 45 47 49 51 53 55 57 59 61 63 65 67 69 71 73 75 77 79 81 83

Absorb

ance in m

AU

Time in min

Comparison: Fraction collection after digest and after Protein L AC

Without Protein L AC With Protein L AC IgG fraction

LHF fractio Fab fraction


42

Figure 53: SEC chromatogram overlay of the fragment isolation after Kgp digestion, Protein L AC and buffer exchange in mobile phase.

Sample stability: The CE-SDS NR electropherogram of the SEC injection sample

before and after fraction collection, displayed in Figure 54, shows no significant

changes. A possible explanation are reversible aggregates, which dissociate in CE-

SDS NR sample buffer.

Figure 54: CE-SDS NR electropherogram of initial and one-day aged injection sample.

Digestion reproducibility and fragment isolation: CE-SDS NR comparison of two

digestions with identical parameters but different GingisKHAN® batches (see Figure

35 40 45 50 55 60 65 70 75 80 85

Absorb

ance in m

AU

Time in min

Kgp digestion after Protein L AC and desalting: Fraction collection

Injection 1 Injection 2 Injection 3 Injection 4 Injection 5

Injection 6 Injection 7 IgG fraction LHF fraction Fab fraction

9 10 11 12 13 14 15 16 17 18 19

Absorb

ance in m

AU

Time in min

CE-SDS NR electropherogram - Injection sample stability in SEC mobile phase at 5°C

Before fraction collection After fraction collection


43

55) showed inconsistent digestion progress. A possible explanation for this are

differences of the RS, as a different solubility was overserved during reconstitution.

Figure 55: CE-SDS NR electropherogram overlay of two Kgp digestions with identical parameter apart from different GingisKHAN® batches.

Buffer exchange into formulation buffer improved the sample stability and less peaks

before IgG were observed.

Figure 56: SEC chromatogram overlay of the fragment isolation after Kgp digestion, Protein L AC and buffer exchange in formulation buffer.

9 10 11 12 13 14 15 16 17

Absorb

ance in m

AU

Time in min

CE-SDS NR electropherogram - Comparison of two 30 min Kgp digestions

Digest 1 Digest 2

35 40 45 50 55 60 65 70 75 80 85

Absorb

ance in m

AU

Time in min

SEC chromatogram - Kgp digestion after Protein L AC and buffer exchange in formulation buffer: Fraction collection

Injection 1 Injection 2 Injection 3 Injection 4 Injection 5

Injection 6 IgG fraction LHF fraction Fab fraction


44

A summary of isolated fragments is listed below.

Table 17: Summary of isolated fragments during SEC fraction collection.


Conc. In mg/mL


Purity in A%

IgG1 20 6.21 124 4 97

LHF 20 2.24 45 9 41

Fab 20 2.32 46 15 82

Digestion progress of IgdE

The IgdE digestion progress, displayed in Figure 57, showed a much lower digestion

time required than the 16 –18 h suggested in the Genovis protocol. The maximum

LHF concentration of 29% was detected after 10 min and would even allow reduction

of the enzyme-to-protein ratio to slow down the digestion. The higher intermediate

concentration, the poly-his tag for simple enzyme removal and mild digestion

conditions make IgdE superior to Kgp for the LHF generation.

Figure 57: Digestion progress of IgdE by CE-SDS NR.

4.1.2 F(ab’)2

IdeS digestion and Protein L AC pH 3.5 elution

The IdeS digestion was complete after 30 min. Fc/2, IdeS and some other, minor

impurities could successfully be removed by the Protein L AC. The F(ab’)2 could not

be eluted with EB at pH 3.5.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0 15 30 45 60 75 90 105 120 135 150 165 180

Purity

in A

%

Time in min

Digestion progress of the IgdE digestion

IgG

LHF

Fc

Fab


45

Over digestion screening and Protein L AC pH 2.0 – 3.0 elution

After 1 h of digestion no sign of over-digestion could be detected by CE-SDS NR.

Elution at pH 3.0 was very slow, at pH 2.5 moderate and at pH 2.0 acceptable. The

electropherogram is displayed in the appendix on p. IX.

Figure 58: Elution profile of Protein L AC at pH 3.0, 2.5 and 2.0.

FragIT™ kit digestion and fragment isolation

The on-column IdeS digestion was complete after 30 min and the Fc/2 could

successfully be removed with the CaptureSelect™ column, as shown in Figure 59.

Figure 59: CE-SDS NR electropherogram of the IdeS digestion and isolated products.

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.1

Concentr

ation in m

g/m

L

Elution buffer volume in mL

Protein L AC elution at pH 3.0 - 2.0

pH 3.0

pH 2.5

pH 2.0

7 8 9 10 11 12 13 14 15 16

Absorb

ance in m

AU

Time in min

CE-SDS NR electropherogram - FabIT kit digestion and purification

Undigested

Digested

F(ab)2

Fc/2


46

Initial and second digestion at the same day show the same digestion rate. The

digestion after 17 days showed a loss in activity and is after 30 min not complete. No

differences of digestion products was observed. A summary of the isolated fragments

is listed in Table 18.

Table 18: Summary of isolated fragments for FragIT™ kit digestion and fragment isolation.

Run Volume in

µL Conc. in mg/mL


Purity in A%

1 100 31.65 3’165 48 91

2 100 36.12 3’612 55 96

3 100 12.71 1271 19 97

Figure 60: CE-SDS NR electropherogram overlay of the three FragIT™ digestion.

4.1.3 Chitobiose cleavage

EndoS2 digestion

The digestion according to the Genovis® protocol and the subsequent buffer

exchange were successful. The enzyme could not be removed but reduced to below

0.5%. 40 µL chitobiose-cleaved mIgG1 27.06 mg/mL (1’082 µg = 94% yield) could

be isolated.

7 8 9 10 11 12 13 14 15

Absorb

ance in m

AU

Time in min

CE-SDS NR electropherogram - Overlay of FragIT™ digestion at different time points

1st run (T0) 2nd run (T150min.) 3rd run (T17d)


47

Figure 61: CE-SDS red. electropherogram of the EndoS2 digestion and buffer exchange.

Immobilized GlycINATOR®

The Genovis® digestion procedure for the immobilized GlycINATOR® was

successfully applied to mIgG1. The digestion product was be concentrated and

buffer exchanged into formulation buffer. 62 µL chitobiose-cleaved mIgG1 at

125 mg/mL (7’750 µg = 80% yield) were isolated.

15 17 19 21 23 25 27 29 31 33 35

Absorb

ance in m

AU

Time in min.

CE-SDS red. electropherogram of the soluble EndoS2

digestion and buffer exchange

Reference Digestion T30min Buffer exchange

ZHAW LSFM, 2017 Conclusion Martin Alexander Rappo

48

5 Conclusion

The influence of the undesired modifications, deamidation and reduction during digestion

where assessed and are considered minimal (> 0.5%) within 2 The Kgp digestion of the mIgG1

using GingisKHAN® was successful and the LHF could be detected as intermediate by CE-

SDS NR after reducing the enzyme-to-protein ratio to 1:10. Cysteine concentration and pH of

the DB was classified as a crucial influence parameter for the digestion kinetic. Through

digestion parameter optimization for LHF isolation the time for the maximal LHF was adjusted

to 30 min at 37°C, which is considered an appropriate time frame to handle. Of the evaluated

digestion inhibition parameter (temperature, pH and cOmplete™ inhibitor cocktail) nothing fully

inhibited but slowed down further fragmentation. Inhibition potential is in decreasing order:

cooling to 5°C, acidify to pH 5.5 and adding cOmplete™. Therefore enzyme removal by AC is

proposed to stop further digestion. Through Protein L AC the Fc could be isolated and following

elution at pH 2.0 the IgG, LHF and Fab mixture could be eluted. Small quantities could be

separated by SEC with a purity of > 75%. Scale-up and reproducibility experiments revealed

issues regarding inconsistent digestion progresses for different batches. Slow digestion led

finally to large IgG contamination of the LHF fraction. This digestion kinetic variations probably

origin from RS differences. Experiments on the recently launched FabALACTICA™ (IgdE)

showed promising results with an acceptable digestion time, mild digestion conditions, less

influencing factors and a poly-his tag on the enzyme for simple removal and digestion stop.

Also the outlook on the immobilized IgdE, which is according to Genovis® supposed to be

launched in fall 2017 makes IgdE and better alternative for LHF generation than Kgp.

Digestion with FabRICATOR® (soluble IdeS) was complete after 30 min and Fc/2 was removed

by Protein L AC. The F(ab‘)2 was successfully generated by applying standard FragIT™

(immobilized IdeS) procedure and the desired fragment was isolated by CaptureSelect™

without exposure to acidic solution. The maximal recovery was 55% and purity 96% and no

undesired modifications were detected. The procedure was repeated with identical outcome

at the same day using the same mab and digestion kit.

The chitobiose cleavage by GlyCINATOR® (soluble EndoS2) was complete after 30 min at

37°C but a simple 100K-centrifugal filtration was not sufficient to fully remove the enzyme.

Immobilized GlyCINATOR® provided a full digestion after 30 min at rt and full enzyme removal

with a yield of 80%.

ZHAW LSFM, 2017 Outlook Martin Alexander Rappo

49

6 Outlook

Following experiments should include identification and in-depth analysis of undesired

modification by mass spectrometry.

Also other types of IgG1, mutants, other applicable subclasses or proteins should be tested on

the implemented methods.

As soon as the immobilized FabALACTICA™ (IgdE) is launched, digestion progress and

applicability for the LHF generation should be evaluated. Important is also a reproducibility and

batch-to-batch comparison.

Isolated digestion products have to be tested for regarding CQA. This could include: cell

bioassay for fragments and higher order structure analysis by differential scanning calorimetry

and circular dichroism in the far UV for the chitobiose chelated product.

ZHAW LSFM, 2017 References Martin Alexander Rappo

50

7 References

[1] R.O. Esquivel, M. Molina-Espiritu, F. Salas, C. Soriano, C. Barrientos, J.S. Dehesa, J.A. Dobado, Decoding the building bloks of life from the perspective of quantum information, Licens. InTech. (2013) 641–669. doi:http://dx.doi.org/10.5772/55160 1.

[2] J.K.H. Liu, The history of monoclonal antibody development - Progress, remaining challenges and future innovations, Ann. Med. Surg. 3 (2014) 113–116. doi:10.1016/j.amsu.2014.09.001.

[3] A.L. Nelson, E. Dhimolea, Development trends for human monoclonal antibody therapeutics, Nat. Rev. Drug Discov. 9 (2010) 767–774. doi:10.1038/nrd3229.

[4] H.M. Shepard, G.L. Phillips, C.D. Thanos, M. Feldmann, Developments in therapy with monoclonal antibodies and related proteins, Clin. Med. J. R. Coll. Physicians London. 17 (2017) 220–232. doi:10.7861/clinmedicine.17-3-220.

[5] D.M. Ecker, S.D. Jones, H.L. Levine, The therapeutic monoclonal antibody market., MAbs. 7 (2015) 9–14. doi:10.4161/19420862.2015.989042.

[6] ICH, ICH Quality Guidelines, (2017). http://www.ich.org/products/guidelines/quality/article/quality-guidelines.html (accessed August 2, 2017).

[7] U.S. Department of Health and Human Services Food and Drug Administration, Q8, Q9, & Q10 Questions and Answers -- Appendix: Q&As from Training Sessions (Q8, Q9, & Q10 Points to Consider), (2012). https://www.fda.gov/drugs/guidancecomplianceregulatoryinformation/guidances/ucm313087.htm (accessed August 2, 2017).

[8] EMA, EMA guideline Q11 - Development and Manufacture of Drug Substances (chemical entities and biotechnological / biological entities), 2011. doi:10.5639/gabij.2012.0103.025.

[9] N. Alt, T.Y. Zhang, P. Motchnik, R. Taticek, V. Quarmby, T. Schlothauer, H. Beck, T. Emrich, R.J. Harris, Determination of critical quality attributes for monoclonal antibodies using quality by design principles, Biologicals. 44 (2016) 291–305. doi:10.1016/j.biologicals.2016.06.005.

[10] ICH, ICH Q8(R2) Pharmaceutical Development, 2009.

[11] ICH, ICH Q9- Quality Risk Management, 2005. doi:10.1007/s11095-007-9511-1.

[12] ICH, ICH Q10 - Pharmaceutical quality system, 2008. doi:EMEA/CHMP/ICH/214732/2007.

[13] ICH, ICH Q11 - Development and Manufacture of Drug Substances (Chemical Entities and Biotechnological/Biological Entities, 2012. doi:10.5639/gabij.2012.0103.025.

[14] ICH, ICH Final Concept Paper Q12 - Technical and Regulatory Considerations for Pharmaceutical Product Lifecycle Management, 2014.

[15] Novartis Brandlab, Assets, (2017).

[16] S. Kim, Quality by Design (QbD) Framework, (2014). https://www.youtube.com/watch?v=XlPlDV8ogyo (accessed September 25, 2017).

[17] Antibody Solutions, Potency assays, (2017). https://antibody.com/solutions/by-applications/potency-assays/.

[18] J. Carlsson, E. Forssell Aronsson, S.-O. Hietala, T. Stigbrand, J. Tennvall, E.T. Boder, W. Jiang, A. Labeling, I. Absorption, D. Biotransformation, E. General, C. Multiple, C.L. Maynard, C.O. Elson, R.D. Hatton, C.T. Weaver, American Society of Health-System Pharmacists, Introduction to Pharmacokinetics and Pharmacodynamics,


51

Pharmocokinetics and Pharmcodynamics. 1 (2010) 1–18. doi:10.1007/SpringerReference_184421.

[19] S. Ait-Oudhia, M.A. Ovacik, D.E. Mager, Systems pharmacology and enhanced pharmacodynamic models for understanding antibody-based drug action and toxicity, MAbs. 20 (2016) 00–00. doi:10.1080/19420862.2016.1238995.

[20] WHO, Annex 7 Application of Hazard Analysis and Critical Control Point ( HACCP ) methodology to pharmaceuticals, WHO Tech. Rep. Ser. No. 908, 2003 Annex. (2003) 99–112.

[21] Genovis, FabRICATOR instructions, (n.d.). https://www.genovis.com/wp-content/uploads/instructions-fabricator-2000-5000.pdf.

[22] Genovis, GingisKHAN instructions, (n.d.). https://www.genovis.com/wp-content/uploads/INSTRUCTIONS_GingisKHAN_17.1.1.pdf.

[23] Genovis, FabALACTICA instructions, (n.d.). https://www.genovis.com/wp-content/uploads/INSTRUCTIONS-for-FabALACTICA-2000u_version-17.1.1.pdf.

[24] Genovis, GlycINATOR instructions, (n.d.). https://www.genovis.com/wp-content/uploads/INSTRUCTIONS-for-GlycINATOR_2000u_version-17.1.1.pdf.

[25] C. Schütt, B. Bröker, Grundwissen Immunologie, 2., Springer, Heidelberg, 2009.

[26] S.N. WEI WANG, SATISH SINGH, DAVID L. ZENG, KEVIN KING, Antibody Structure, Instability, and Formulation, J. Pharm. Sci. 96 (2007) 1–26. doi:DOI 10.1002/jps.20727.

[27] R. Jefferis, M.-P. Lefranc, Human immunoglobulin allotypes, MAbs. 1 (2009) 1–7. doi:10.4161/mabs.1.4.9122.

[28] H. Liu, K. May, Disulfide bond structures of IgG molecules, MAbs. 4 (2012) 17–23. doi:10.4161/mabs.4.1.18347.

[29] A.R. Costa, M.E. Rodrigues, M. Henriques, R. Oliveira, J. Azeredo, Glycosylation: impact, control and improvement during therapeutic protein production., Crit. Rev. Biotechnol. 8551 (2013) 1–19. doi:10.3109/07388551.2013.793649.

[30] S. Ha, Y. Ou, J. Vlasak, Y. Li, S. Wang, K. Vo, Y. Du, A. MacH, Y. Fang, N. Zhang, Isolation and characterization of IgG1 with asymmetrical Fc glycosylation, Glycobiology. 21 (2011) 1087–1096. doi:10.1093/glycob/cwr047.

[31] Boundless, Antibody Proteins and Antigen Binding, (2014). https://www.boundless.com/microbiology/textbooks/boundless-microbiology-textbook/immunology-11/antibodies-141/antibody-proteins-and-antigen-binding-714-5819/ (accessed August 9, 2017).

[32] R.C. Hillig, S. Urlinger, J. Fanghänel, B. Brocks, C. Haenel, Y. Stark, D. Sülzle, D.I. Svergun, S. Baesler, G. Malawski, D. Moosmayer, A. Menrad, M. Schirner, K. Licha, Fab MOR03268 Triggers Absorption Shift of a Diagnostic Dye via Packaging in a Solvent-shielded Fab Dimer Interface, J. Mol. Biol. 377 (2008) 206–219. doi:10.1016/j.jmb.2007.12.071.

[33] E.O. Saphire, Crystal Structure of a Neutralizing Human IgG Against HIV-1: A Template for Vaccine Design, Science (80-. ). 293 (2001) 1155–1159. doi:10.1126/science.1061692.

[34] G. Vidarsson, G. Dekkers, T. Rispens, IgG subclasses and allotypes: From structure to effector functions, Front. Immunol. 5 (2014) 1–17. doi:10.3389/fimmu.2014.00520.

[35] abcam, Antibody structure and isotypes, (2017) 1–4. http://docs.abcam.com/pdf/antibody-guide/antibody-structure-and-isotypes.pdf.

[36] C. Spoerry, P. Hessle, M.J. Lewis, L. Paton, J.M. Woof, U. Von Pawel-rammingen,


52

Novel IgG-Degrading Enzymes of the IgdE Protease Family Link Substrate Specificity to Host Tropism of Streptococcus Species, PLoS One. 11 (2016) 1–20. doi:10.1371/journal.pone.0164809.

[37] D.L. Chabner, Bruce A.; Longo, Cancer Chemotherapy and Biotherapy: Principles and Practice, 5th ed., Wolters Kluwer: Lipincott Williams & Wilkins, 2011. https://books.google.ch/books?id=0U4aj4GZWCIC&pg=PA468&lpg=PA468&dq=antibodyfragment+%22Fabc%22&source=bl&ots=F4243KCPtq&sig=T-PqR_9hSSqNjr8R4XlhxZe41QA&hl=en&sa=X&ved=0ahUKEwj63fDelsrVAhXMuhQKHV1KD_UQ6AEIOzAC#v=onepage&q&f=false.

[38] Thermo Fisher Scientific, Antibody Fragmentation, (n.d.). https://www.thermofisher.com/ch/en/home/life-science/antibodies/antibodies-learning-center/antibodies-resource-library/antibody-methods/antibody-fragmentation.html (accessed August 9, 2017).

[39] Absolute Antibody, Antibody Fragments, (n.d.). http://absoluteantibody.com/antibody-resources/antibody-engineering/antibody-fragments/ (accessed August 9, 2017).

[40] E. O’Connor, M. Aspelund, F. Bartnik, M. Berge, K. Coughlin, M. Kambarami, D. Spencer, H. Yan, W. Wang, Monoclonal antibody fragment removal mediated by mixed mode resins, J. Chromatogr. A. 1499 (2017) 65–77. doi:10.1016/j.chroma.2017.03.063.

[41] IMGT, Organization in domains of an IgG1 immunoglobulin and of its fragments, (2001). http://www.imgt.org/IMGTeducation/Tutorials/IGandBcells/_UK/3Dstructure/Figure1.html (accessed May 2, 2017).

[42] R.S. Shochat, Dan; Hansen, Hans J.; Wu, Method for radiolabeling antibody fragments, US5514363 (A) ― 1996-05-07, 1993. https://worldwide.espacenet.com/publicationDetails/biblio?II=0&ND=3&adjacent=true&locale=en_EP&FT=D&date=19960507&CC=US&NR=5514363A&KC=A#.

[43] J. Rohrer, Master Course: Comparative Physiology Glycosylation in Eukaryotes, (2017).

[44] F. Higel, A. Seidl, F. Sörgel, W. Friess, N-glycosylation heterogeneity and the influence on structure, function and pharmacokinetics of monoclonal antibodies and Fc fusion proteins, Eur. J. Pharm. Biopharm. 100 (2016) 94–100. doi:10.1016/j.ejpb.2016.01.005.

[45] P.M. Rudd, R. a Dwek, Glycosylation: heterogeneity and the 3D structure of proteins., Crit. Rev. Biochem. Mol. Biol. 32 (1997) 1–100. doi:10.3109/10409239709085144.

[46] K. Zheng, C. Bantog, R. Bayer, The impact of glycosylation on monoclonal antibody conformation and stability, MAbs. 3 (2011) 568–576. doi:10.4161/mabs.3.6.17922.

[47] K. Zheng, M. Yarmarkovich, C. Bantog, R. Bayer, T.W. Patapoff, Influence of glycosylation pattern on the molecular properties of monoclonal antibodies, MAbs. 6 (2014) 649–658. doi:10.4161/mabs.28588.

[48] MEROPS, Summary for peptidase C66.001: IdeS peptidase, (2017). https://www.ebi.ac.uk/merops/cgi-bin/pepsum?mid=C66.001 (accessed August 11, 2017).

[49] C. Spoerry, J. Seele, P. Valentin-Weigand, C.G. Baums, U. von Pawel-Rammingen, Identification and Characterization of IgdE, a Novel IgG-degrading Protease of Streptococcus suis with Unique Specificity for Porcine IgG, Am. Soc. Biochem. Mol. Biol. (2016). doi:10.1074/jbc.M115.711440.

[50] K. Wenig, L. Chatwell, U. von Pawel-Rammingen, L. Björck, R. Huber, P. Sondermann, Structure of the streptococcal endopeptidase IdeS, a cysteine


53

proteinase with strict specificity for IgG., Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 17371–6. doi:10.1073/pnas.0407965101.

[51] N. Fittipaldi, S.B. Beres, R.J. Olsen, V. Kapur, P.R. Shea, M.E. Watkins, C.C. Cantu, D.R. Laucirica, L. Jenkins, A.R. Flores, M. Lovgren, C. Ardanuy, J. Liares, D.E. Low, G.J. Tyrrell, J.M. Musser, Full-genome dissection of an epidemic of severe invasive disease caused by a hypervirulent, recently emerged clone of group A Streptococcus, Am. J. Pathol. 180 (2012) 1522–1534. doi:10.1016/j.ajpath.2011.12.037.

[52] P.P.P. Cleary, Streptococcus pyogenes disease and molecular pathogenesis, HSTalks, USA, 2009. https://hstalks.com/t/1480/streptococcus-pyogenes-disease-and-molecular-patho/.

[53] M.W. Robinson, J.P. Dalton, Cysteine Proteases of Pathogenic Organisms, Springer Science+Business Media, New York USA, 2011. doi:10.1007/978-1-4419-8414-2.

[54] B. Vincents, A. Guentsch, D. Kostolowska, U. von Pawel-Rammingen, S. Eick, J. Potempa, M. Abrahamson, Cleavage of IgG1 and IgG3 by gingipain K from Porphyromonas gingivalis may compromise host defense in progressive periodontitis, FASEB J. 25 (2011) 3741–3750. doi:10.1096/fj.11-187799.

[55] C. Spoerry, Streptococcal Immunoglobulin degrading Enzymes of the IdeS and IgdE Family, 2017.

[56] N. Pavloff, P. a Pemberton, J. Potempa, W.A. Chen, R.N. Pike, V. Prochazka, M.C. Kiefer, J. Travis, P.J. Barr, P.J.J.B. Chem, Molecular Cloning and Characterization of Porphyromonas gingivalis Lysine-specific Gingipain, J. Bilogical Chem. 272 (1997) 1595–1600.

[57] I. De Diego, F. Veillard, M.N. Sztukowska, T. Guevara, B. Potempa, A. Pomowski, J.A. Huntington, J. Potempa, F.X. Gomis-Rüth, Structure and mechanism of cysteine peptidase gingipain K (Kgp), a major virulence factor of porphyromonas gingivalis, J. Biol. Chem. 289 (2014) 32291–32302. doi:10.1074/jbc.M114.602052.

[58] I. de Diego, F. Veillard, M.N. Sztukowska, T. Guevara, B. Potempa, A. Pomowski, J.A. Huntington, J. Potempa, F.X. Gomis-Rüth, Structure and Mechanism of Cysteine Peptidase Gingipain K (Kgp), a Major Virulence Factor of Porphyromonas gingivalis in Periodontitis, J. Biol. Chem. 289 (2014) 32291–32302. doi:10.1074/jbc.M114.602052.

[59] D.E. Koshland, K. Hamadani, Proteomics and models for enzyme cooperativity, J. Biol. Chem. 277 (2002) 46841–46844. doi:10.1074/jbc.R200014200.

[60] MEROPS, Summary for C113.001, (2017). https://www.ebi.ac.uk/merops/cgi-bin/pepsum?id=C113.001 (accessed August 16, 2017).

[61] MEROPS, MER1020437 - IgdE peptidase ({Streptococcus suis}), (2017). https://www.ebi.ac.uk/merops/cgi-bin/aaseq?mernum=MER1020437 (accessed September 25, 2017).

[62] J. Sjögren, W.B. Struwe, E.F.J. Cosgrave, P.M. Rudd, M. Stervander, M. Allhorn, A. Hollands, V. Nizet, M. Collin, EndoS 2 is a unique and conserved enzyme of serotype M49 group A Streptococcus that hydrolyses N-linked glycans on IgG and α 1 -acid glycoprotein, Biochem. J. 455 (2013) 107–118. doi:10.1042/BJ20130126.

[63] J. Sjögren, W.B. Struwe, E.F.J. Cosgrave, P.M. Rudd, M. Stervander, M. Allhorn, A. Hollands, V. Nizet, M. Collin, endo-beta-N-acetylglucosaminidase [Streptococcus pyogenes], EndoS 2 Is a Unique Conserv. Enzym. Serotype M49 Gr. A Streptococcus That Hydrolyses N-Linked Glycans IgG α 1 -Acid Glycoprotein. (2013). https://www.ncbi.nlm.nih.gov/protein/AGU16859.1 (accessed September 5, 2017).

[64] N. Fittipaldi, S.B. Beres, R.J. Olsen, V. Kapur, P.R. Shea, M.E. Watkins, C.C. Cantu, D.R. Laucirica, L. Jenkins, A.R. Flores, M. Lovgren, C. Ardanuy, J. Liñares, D.E. Low, G.J. Tyrrell, J.M. Musser, Full-Genome Dissection of an Epidemic of Severe Invasive


54

Disease Caused by a Hypervirulent, Recently Emerged Clone of Group A Streptococcus, Am. J. Pathol. 180 (2012) 1522–1534. doi:10.1016/j.ajpath.2011.12.037.

[65] J. Sjögren, E.F.J. Cosgrave, M. Allhorn, M. Nordgren, S. Björk, F. Olsson, S. Fredriksson, M. Collin, EndoS and EndoS2 hydrolyze Fc-glycans on therapeutic antibodies with different glycoform selectivity and can be used for rapid quantification of high-mannose glycans, Glycobiology. 25 (2015) 1053–1063. doi:10.1093/glycob/cwv047.

[66] J. de P. Peter W Atkins, Kurzlehrbuch Physikalische Chemie, 4., WILEY-VCH, Weinheim, 2008.

[67] G. Schwedt, Analytische Chemie, 2nd ed., WILEY-VCH, Weinheim, 2008.

[68] Y. An, Y. Zhang, H.M. Mueller, M. Shameem, X. Chen, A new tool for monoclonal antibody analysis Application of IdeS proteolysis in IgG domain-specific characterization, MAbs. 6 (2014) 879–893. doi:10.4161/mabs.28762.

[69] GE Healthcare Life Sciences, HiScreen TM Capto TM L HiTrap TM Protein L, 1, (2009) 1–20.

[70] Genovis AM, FragITkitTM instructions, (n.d.). https://www.genovis.com/wp-content/uploads/instructions-fragit-kit-midi.pdf.

[71] Genovis, Immobilized GlycINATOR, (n.d.). https://www.genovis.com/wp-content/uploads/INSTRUCTIONS-for-Immobilized-GlycINATOR®-Midispin.pdf.

[72] Merck, Safety data sheet, (n.d.). http://www.sigmaaldrich.com/ (accessed September 19, 2017).

[73] U. Kishore, K.B.. Reid, C1q: Structure, function, and receptors, Immunopharmacology. 49 (2000) 159–170. doi:10.1016/S0162-3109(00)80301-X.

[74] M.J. Smyth, J.A. Trapani, Granzymes: exogenous porteinases that induce target cell apoptosis, Immunol. Today. 16 (1995) 202–206. doi:10.1016/0167-5699(95)80122-7.

[75] R. Rojas, G. Apodaca, Immunoglobulin transport across polarized epithelial cells., Nat. Rev. Mol. Cell Biol. 3 (2002) 944–956. doi:10.1038/nrm972.

[76] Merk, Papain, (n.d.). http://www.sigmaaldrich.com/life-science/biochemicals/biochemical-products.html?TablePage=16410606 (accessed September 5, 2017).

[77] Sigma-Aldrich, Endoproteinase Lys-C from Lysobacter enzymogenes, (2014) 1–2. https://www.sigmaaldrich.com/content/dam/sigma-aldrich/docs/Sigma/Bulletin/p3428bul.pdf.

[78] Genovis, FabULOUS Instructions Version 15.1.2, (n.d.) 1–2. https://www.genovis.com/wp-content/uploads/INSTRUCTIONS-for-FabULOUS-2000u_version-15.1.2.pdf.

[79] Merk, Pepsin from porcine gastric mucosa, (n.d.). http://www.sigmaaldrich.com/catalog/product/sigma/p6887?lang=de&region=CH&gclid=Cj0KCQjw0ejNBRCYARIsACEBhDNaOVcOkZHowiEafTuS0rmBj8sS9Z8rQeWiRxnhcHt7fFUWnMAfKGUaAmbsEALw_wcB (accessed September 14, 2017).

[80] I.M. Rosenberg, Protein analysis and purification: Benchtop techniques, 2., Birkhäuser, Boston MA, 2005. doi:10.1007/b138330.

[81] Genovis, GingisREX instructions, (n.d.) 1–2. https://www.genovis.com/wp-content/uploads/INSTRUCTION-gingisrex_17.1.1.pdf.

[82] Genovis, Smart Enzymes, (2017). https://www.genovis.com/products/ (accessed September 25, 2017).


55

7.1 List of illustrations

FIGURE 1: AMINO ACID PROPERTIES VENN DIAGRAM [1]. .......................................................... VIII FIGURE 2: ILLUSTRATION OF THE QUALITY BY DESIGN STRATEGY [15,16]. .................................. 2 FIGURE 3: POTENTIAL INFORMATION SOURCES FOR THE QUALITY RISK MANAGEMENT [15]. .......... 3 FIGURE 4: THE FOUR QUALITY CATEGORIES FOR THE RISK ASSESSMENT QUALITY ATTRIBUTES

[15,17–20]. ...................................................................................................................... 3 FIGURE 5: SELECTED DIGESTION PRODUCTS TO BE EVALUATED DURING THE MASTER THESIS

INCLUDING THEIR FINAL PURPOSE IN THE CQA ASSESSMENT. .............................................. 4 FIGURE 6: ILLUSTRATION OF THE IGG1 STRUCTURE WITH LIGHT CHAIN IN APRICOT AND HEAVY

CHAIN IN BLUE. ................................................................................................................. 6 FIGURE 7: IG FAB WITH CDR IN RED AND A HAPTEN AS EPITOPE [32]. ......................................... 6 FIGURE 8: 3D SURFACE STRUCTURE OF AN IGG1 [33]. .............................................................. 6 FIGURE 9: NOMENCLATURE OF COMMON IGG FRAGMENTS [37–42]. ........................................... 7 FIGURE 10: FC GLYCOSYLATION STRUCTURE OF TRANSGENIC ANIMAL/CHINESE HAMSTER OVARY

CELLS. ............................................................................................................................. 8 FIGURE 11: ILLUSTRATION OF IDES. ......................................................................................... 9 FIGURE 12: IGG FRAGMENTATION OF IDES. .............................................................................. 9 FIGURE 13: ACCEPTED SEQUENCE FOR THE IDES CLEAVAGE. IDENTICAL AMINO ACID IN BLACK,

CLOSELY CHEMICALLY SIMILAR IN BLUE, WEAKLY CHEMICALLY SIMILAR IN SIENNA AND NON-SIMILAR IN APRICOT. ........................................................................................................ 10

FIGURE 14: PROPOSED CATALYTIC CYCLE OF THE IDES CLEAVAGE. ......................................... 10 FIGURE 15: CATALYTIC POCKET AND STABILIZING EFFECTS. ..................................................... 11 FIGURE 16: ILLUSTRATION OF RECOMBINANT KGP. .................................................................. 11 FIGURE 17: IGG FRAGMENTATION OF KGP AND IGDE. .............................................................. 12 FIGURE 18: ACCEPTED KGP CLEAVAGE SEQUENCE OF IGG1 [22]. ............................................ 12 FIGURE 19: ILLUSTRATION OF IGDE. ....................................................................................... 13 FIGURE 20: ACCEPTED IGDEAGALACTIAE CLEAVAGE SEQUENCE OF HUMAN IGG1 [36]. ..................... 13 FIGURE 21: ILLUSTRATION OF ENDOS2................................................................................... 13 FIGURE 22: CHEMICAL STRUCTURE OF THE CHITOBIOSE CORE. ................................................ 14 FIGURE 23: CHITOBIOSE CLEAVAGE OF ENDOS2. .................................................................... 14 FIGURE 24: PROPOSED CONSECUTIVE REACTION SCHEME OF AN IGG AND KGP. ....................... 15 FIGURE 25: SIMULATED KGP DIGESTION PROGRESS BASED ON A CONSECUTIVE FIRST-ORDER

KINETIC. ......................................................................................................................... 16 FIGURE 26: DEAMIDATION OF ASPARAGINE. ............................................................................ 18 FIGURE 27: EXAMPLE CATION EXCHANGE CHROMATOGRAM. .................................................... 19 FIGURE 28: EXAMPLE CE-SDS NR ELECTROPHEROGRAM OF THE KGP DIGESTION. .................. 21 FIGURE 29: EXAMPLE CE-SDS RED ELECTROPHEROGRAM OF THE ENDOS2 DIGESTION. ........... 22 FIGURE 30: FLOW CHART OF THE FAB, FC AND LHF GENERATION AND ISOLATION. .................... 22 FIGURE 31: FLOW CHART OF THE F(AB)2 GENERATION AND ISOLATION. ..................................... 27 FIGURE 32: FLOW CHART OF THE ENDOS2 DIGESTION AND FRAGMENT ISOLATION. .................... 29 FIGURE 33: PURITY TREND OF THE STABILITY IN DIGESTION BUFFER BY CEX (DEAMIDATION). .... 31 FIGURE 34: PURITY TREND OF THE STABILITY IN DIGESTION BUFFER BY AND CE-SDS NR

(FRAGMENTATION/AGGREGATION). ................................................................................... 31 FIGURE 35: CE-SDS NR ELECTROPHEROGRAM OF THE MIGG1 STABILITY IN DIGESTION BUFFER

WITH REDUCING AGENT. .................................................................................................. 32 FIGURE 36: CE-SDS NR PURITY PROGRESS OF THE 1:10 KGP-TO-IGG DIGESTION AND

COMPARISON OF THE SIMULATED VALUES. ........................................................................ 33 FIGURE 37: REACTION CONSTANT K1 IN DEPENDENCY OF TEMPERATURE FOR TESTED INHIBITORY

FACTORS (STANDARD DIGESTION, COMPLETE™ AND PH 5.5). ........................................... 33 FIGURE 38: CALCULATED TIME POINT OF MAXIMAL LHF CONCENTRATION IN DEPENDENCY OF

TEMPERATURE FOR TESTED INHIBITORY FACTORS (STANDARD DIGESTION, COMPLETE™ AND

PH 5.5). ......................................................................................................................... 33


56

FIGURE 39: KINETIC CONSTANTS OF THE KGP DIGESTION FOR DIFFERENT CYSTEINE

CONCENTRATION OF GENOVIS, SELF-MADE/FRESH AND SELF-MADE/AGED RS. ................... 34 FIGURE 40: CALCULATED TIME POINT OF MAXIMAL LHF CONCENTRATION OF THE KGP DIGESTION

FOR DIFFERENT CYSTEINE CONCENTRATION OF GENOVIS, SELF-MADE/FRESH AND SELF-MADE/AGED RS. ............................................................................................................. 34

FIGURE 41: PROGRESS OF THE 25 KDA-PEAK FORMATION DURING THE IGG1 KGP DIGESTION AT

DIFFERENT CYSTEINE CONCENTRATIONS. ......................................................................... 35 FIGURE 42: KINETIC CONSTANTS OF THE KGP DIGESTION AT DIFFERENT PH. ............................. 36 FIGURE 43: CALCULATED TIME POINT OF MAXIMAL LHF CONCENTRATION OF THE KGP DIGESTION

AT DIFFERENT PH. .......................................................................................................... 36 FIGURE 44: DIGESTION COURSE OF THE PARAMETER ADJUSTED KGP DIGESTION AT PH 6.0. ...... 37 FIGURE 45: DIGESTION COURSE OF THE PARAMETER ADJUSTED KGP DIGESTION AT PH 8.0. ...... 37 FIGURE 46: PROTEIN CONTENT OF THE PROTEIN L AFFINITY CHROMATOGRAPHY BY NANODROP. 37 FIGURE 47: CE-SDS NR ELECTROPHEROGRAM OF THE KGP DIGESTION AND CONCENTRATED

PROTEIN L FLOW THROUGH. ............................................................................................ 38 FIGURE 48: SEC CHROMATOGRAM OVERLAY OF 2 AND 3 COLUMNS IN SERIES. .......................... 39 FIGURE 49: SEC CHROMATOGRAM OF THE INJECTION AMOUNT EVALUATION. ............................ 39 FIGURE 50: SEC CHROMATOGRAM OVERLAY OF THE FRACTION COLLECTION TRIAL OF 20 TIMES

40 ΜG INJECTION. ........................................................................................................... 40 FIGURE 51: CE-SDS NR OVERLAY OF THE ISOLATED IGG, LHF AND FAB+FC FRACTIONS. ........ 40 FIGURE 52: SEC CHROMATOGRAM OF THE KGP DIGESTION FRAGMENT ISOLATION WITH AND

WITHOUT PROTEIN L AC. ................................................................................................ 41 FIGURE 53: SEC CHROMATOGRAM OVERLAY OF THE FRAGMENT ISOLATION AFTER KGP

DIGESTION, PROTEIN L AC AND BUFFER EXCHANGE IN MOBILE PHASE. ............................... 42 FIGURE 54: CE-SDS NR ELECTROPHEROGRAM OF INITIAL AND ONE-DAY AGED INJECTION

SAMPLE. ......................................................................................................................... 42 FIGURE 55: CE-SDS NR ELECTROPHEROGRAM OVERLAY OF TWO KGP DIGESTIONS WITH

IDENTICAL PARAMETER APART FROM DIFFERENT GINGISKHAN® BATCHES. ......................... 43 FIGURE 56: SEC CHROMATOGRAM OVERLAY OF THE FRAGMENT ISOLATION AFTER KGP

DIGESTION, PROTEIN L AC AND BUFFER EXCHANGE IN FORMULATION BUFFER. ................... 43 FIGURE 57: DIGESTION PROGRESS OF IGDE BY CE-SDS NR. .................................................. 44 FIGURE 58: ELUTION PROFILE OF PROTEIN L AC AT PH 3.0, 2.5 AND 2.0. ................................. 45 FIGURE 59: CE-SDS NR ELECTROPHEROGRAM OF THE IDES DIGESTION AND ISOLATED

PRODUCTS. .................................................................................................................... 45 FIGURE 60: CE-SDS NR ELECTROPHEROGRAM OVERLAY OF THE THREE FRAGIT™ DIGESTION. 46 FIGURE 61: CE-SDS RED. ELECTROPHEROGRAM OF THE ENDOS2 DIGESTION AND BUFFER

EXCHANGE. .................................................................................................................... 47 FIGURE 62: CEX CHROMATOGRAM OF THE DIGESTION BUFFER STABILITY AT 37°C. ................... VII FIGURE 63: DIGESTION PROGRESS OF THE DIGESTION ACCORDING TO GENOVIS® STANDARD

PROTOCOL. .................................................................................................................... VII FIGURE 64: KINETIC MODEL PLOT LN([A]/[A]0) IN DEPENDENCY OF TIME TO JUSTIFY A FIRST ORDER

KINETIC MODEL. ............................................................................................................. VIII FIGURE 65: CE-SDS NR ELECTROPHEROGRAM OF THE PROTEIN L AC FLOW THROUGH AND

ELUTION. ....................................................................................................................... VIII FIGURE 66: CE-SDS NR ELECTROPHEROGRAM OF THE IDES DIGESTION AND PROTEIN L AC. .... IX FIGURE 67: CE-SDS RED ELECTROPHEROGRAM OF THE DIGESTION BY IMMOBILIZED ENDOS2 AND

BUFFER EXCAHNGED SAMPLE. .......................................................................................... IX

7.2 List of tables

TABLE 1: ABBREVIATIONS. ...................................................................................................... VII TABLE 2: LIST OF NEW ICH QUALITY GUIDELINES. ...................................................................... 1 TABLE 3: LIST OF USED BUFFER. ............................................................................................ 17 TABLE 4: APPLIED CEX PARAMETER. ..................................................................................... 18 TABLE 5: APPLIED SEC PARAMETER FOR FRACTION COLLECTION. ............................................ 19 TABLE 6: LIST OF CE-SDS NR PARAMETER. ........................................................................... 20


57

TABLE 7: LIST OF CE-SDS RED PARAMETER. .......................................................................... 21 TABLE 8: DILUTION TABLE OF THE CYSTEINE INFLUENCING DIGESTION. ..................................... 24 TABLE 9: ELUTION OF FC ISOLATION. ...................................................................................... 25 TABLE 10: SEC FRACTIONATION RANGE. ................................................................................ 25 TABLE 11: ELUTION OF PROTEIN L AC. ................................................................................... 26 TABLE 12: ELUTION OF PROTEIN L AC FOR REPRODUCIBILITY. ................................................. 26 TABLE 13: ELUTION OF PROTEIN L AC AFTER IDES DIGESTION................................................. 28 TABLE 14: PROTEIN L AC PURIFICATION TABLE. ...................................................................... 28 TABLE 15: SUMMARY OF ISOLATED FRAGMENTS DURING SEC TRIALS. ...................................... 40 TABLE 16: SUMMARY OF ISOLATED FRAGMENTS DURING SEC INJECTION BUFFER EVALUATION. . 41 TABLE 17: SUMMARY OF ISOLATED FRAGMENTS DURING SEC FRACTION COLLECTION. .............. 44 TABLE 18: SUMMARY OF ISOLATED FRAGMENTS FOR FRAGIT™ KIT DIGESTION AND FRAGMENT

ISOLATION. ..................................................................................................................... 46 TABLE 19: MOST IMPORTANT EFFECTOR FUNCTIONS OF IGG. ..................................................... II TABLE 20: LIST OF IG FRAGMENTING ENZYMES. ........................................................................ III TABLE 21: LIST OF IG DEGLYCOSYLATING ENZYMES. ................................................................. III TABLE 22: LIST OF USED INSTRUMENTS. .................................................................................. IV TABLE 23: LIST OF USED CONSUMABLES. ................................................................................. IV TABLE 24: LIST OF USED REAGENTS. ........................................................................................ V TABLE 25: LIST OF USED SOFTWARE. ....................................................................................... VI

7.3 List of equations

EQUATION 1: DETERMINATION OF THE KINETIC CONSTANT K1 [66]. ............................................ 15 EQUATION 2: CALCULATION OF THE INTERMEDIATE PRODUCT CONCENTRATION [I]. .................... 15 EQUATION 3: CALCULATION OF THE PRODUCT CONCENTRATION [P]. ......................................... 15 EQUATION 4: CALCULATION OF THE TIME POINT OF THE MAXIMAL INTERMEDIATE CONCENTRATION

IS REACHED. ................................................................................................................... 16

Specific enzymatic digestion and fragment isolation for the critical quality attribute assessment of IgG1

Annex

Novartis 20.10.2017

I

Table of content

I. MECHANISM OF ACTION ............................................................................................. II

II. ENZYMES IN MAB ANALYTICS ................................................................................... III

I.1. FRAGMENTATION ....................................................................................................... III I.2. DEGLYCOSYLATION ................................................................................................... III

III. INSTRUMENTS .......................................................................................................... IV

IV. CONSUMABLES ........................................................................................................ IV

V. REAGENTS ................................................................................................................. V

VI. ENZYMES .................................................................................................................. VI

VII. SOFTWARE ............................................................................................................... VI

VIII. CHROMATOGRAM AND ELECTROPHEROGRAM ................................................. VII

I.3. FAB, FC AND LHF ..................................................................................................... VII I.4. F(AB’)2 ...................................................................................................................... IX I.5. CHITOBIOSE CLEAVAGE ............................................................................................. IX

IX. COPY OF THE OBJECTIVE SETTING AGREEMENT ................................................ X

X. STATEMENT OF AUTHORSHIP .............................................................................. XII

XI. REPORT IN DIGITAL FORMAT ............................................................................... XIII

XII. POSTER ................................................................................................................... XIV

ZHAW LSFM, 2017 Annex Martin Alexander Rappo

II

I. Mechanism of action

Table 19: Most important effector functions of IgG.

Function Mechanism

Complement activation After epitope binding on a cell an IgG undergoes a conformational change. Two bound Igs next to each other allow for binding of C1q, a protein complex, at the CH2 (Glu318, Lys320, Lys322). Subsequently, the complement cascade is initiated, which leads to pore formation and finally the destruction of the cell [25,73].

Ab-dependent cellular cytotoxicity (ADCC)

Cell bound IgGs allow natural killer-cells to bind the Fc region with a Fc receptor (FcR), the FcRs are cross-linked and initiate the release of granzymes and leading to apoptosis of the cell [25,74].

Opsonization Igs bound to bacteria present their Fc part to phagocytes and simplify the internalization [25].

Blocking Ig binding to a receptor may prevent the access for ligands and their physiological function [25]. A special form of blocking is the binding of an Ig to the CDR of another Ig, so called anti idiotypic Ab. This leads to the elimination of this respective Ab [25].

Masking Binding of non-cytotoxic Ig may prevent cytotoxic T-cells resp. Igs to recognize and bind an antigen [25].

Neutralization Small molecular toxins might act as an antigen and through binding by an Ig the binding to a receptor and the toxic effect is prevented [25].

Receptor binding Ab binding to a receptor can have three possible effects:

No reaction The receptor is not effected by the Ab [25].

Agonistic Ab The Ig binding itself imitates a ligand binding and triggers a response. This effect is only relevant for auto-immune disease [25].

Antagonistic Ab The Ig blocks the access of the receptor for ligands and prevents a physiological reaction [25].

Inhibition Some specific soluble immune complexes have the ability to suppress the activation of B-cells. This is important to reduce the antigen specific B-cell response [25].

Penetration Igs are able to cross polarized epithelial cells through cell-mediated transcytosis. For IgG the neonatal Fc receptor (FcRn) is of importance and even allows passing of the placenta [25,75].

Precipitation Binding of polyclonal Ab on multiple epitopes first forms soluble complexes and leads later to precipitation [25].

Agglutination Linking multiple particular antigens, e.g. on the surface of cells, leads to clump [25].


III

II. Enzymes in mab analytics

I.1. Fragmentation

Table 20: List of Ig fragmenting enzymes.

Application Name Trade name Information

Fab + Fc Papain Papain Cleaves hydrophilic-R/K ↓ not V [76]. Over-digestion possible.

Lys-C Lys-C Cleaves C-terminal of K. Over-digestion possible. [77]

SpeB FabULOUS™ Cleaves human IgG1: KTHT ↓ CPPCPAPEL under reducing conditions. [78]

Kgp GingisKHAN® Cleaves human IgG1: KSCDK ↓ THTCPPCP under mild reducing conditions.

IgdE FabALACTICA™ Cleaves IgG1: KSCDKT ↓ HTCPPCP [23]

F(ab’)2 Pepsin Pepsin Cleaves the peptide bond of hydrophobic and aromatic amino acids [79]. Risk of oxidation.

IdeS FabRICATOR® Cleaves human IgG1, 3 and 4: CPAPELLG ↓ GPSVF and human IgG2: CPAPPVA ↓ GPSVF. [21]

Peptide mapping Trypsin Trypsin Cleaves C-terminal of K/R-not P [80].

Rgp GingisREX™ Cleaves C-terminal of R [81].

Removal of C-terminal K

Carboxypeptidase B

Carboxypeptidase B

Cleaves K and R at the C-terminus [80].

I.2. Deglycosylation

Table 21: List of Ig deglycosylating enzymes.

Cleavage site Name Trade name Information

-GlcNAc | Asn- PNGase F PNGase F Complex, high-mannose and hybrid type oligosaccharides of N-linked glycoproteins [80].

-GlcNAc | GlcNAc- EndoS IgGZERO® Only complex oligosaccharides of IgG and the α1-acid glycoprotein [82].

-GlcNAc | GlcNAc- EndoS2 GlycINATOR® Complex, high-mannose and hybrid type oligosaccharides of IgG and the α1-acid glycoprotein [82].


IV

Removal of α2,3-, α2,6- and α2,8-linked sialic acids

Two sialidase of Akkermansia muciniphila

SialEXO™ Cleaves N- and O-glycans[82].

O-glycosylated proteins of core 1 and core 3

endo-α-N-acetylgalactos-aminidase

OglyZOR™ Required prior sialic aids removal [82].

O-glycosylated proteins N-terminally of the S/T glycosylation site

Endoprotease of Akkermansia muciniphila

OpeRATOR™ Improved enzyme activity with removed sialic aids [82].

III. Instruments

Table 22: List of used instruments.

Type Model Manufacturer

Centrifuge 0.5 mL tubes 1-14 SIGMA

Centrifuge 4 & 15 mL tubes 3-18KS SIGMA

CE-SDS NR/Red. PA800plus Pharmaceutical Analysis System

BECKMAN COULTER

HPLC (CEX and SEC) 1100 Series Agilent Technology

HPLC fraction collector 1260 Infinity Agilent Technology

Thermo shaker Thermomixer comfort EPPENDORF

Tube rotator plate Tube rotator BOEKEL

Water bath Eco Silver Lauda

Nanodrop™ NanoDrop1000 Spectrophotometer

Thermo Scientific™

IV. Consumables

Table 23: List of used consumables.

Description Manufacturer Item number

Amicon® Ultra – 0.5mL Centrifugal Filters Ultracel® - 10K

Merk UFC501096

Amicon ® Ultra – 4 Centrifugal Filters Ultracel® - 10K

Merk UFC801024

Amicon ® Ultra – 15 Centrifugal Filters Ultracel® - 10K

Merk UFC901096

Stericup® 150ml Millipore® Express PLUS

Millipore SCGPU01RE


V

0.22µm PES

Protein L affinity chromatography HiScreen™ Capto™ L HiTrap™ Protein L, 1 mL

GE Healthcare 29048665

Crimp vials 6 mL + 20 mm crimp caps with septum

Agilent Technology 9301-1419 + 9301-1425

V. Reagents

Table 24: List of used reagents.

Name Supplier Part number Purity / Conc. MW in g/mol

2-mercaptoethanol

SIGMA-ALDRICH 63689 ≥ 99.0% 78.13

cOmplete™ protease inhibitor cocktail

Roche 11 697 498 001 n.a. n.a.

Cysteine SIGMA-ALDRICH 30089 98.5% 121.16

Hydrochloric acid Merk 1.09060 1 M 36.46

Iodoacetamide SIGMA-ALDRICH I1149 ≥ 99% 184.96

L-Histidine SIGMA-ALDRICH H8000 ≥ 99% 155.15

L-Histidine hydrochloride monohydride

SIGMA-ALDRICH H8125 ≥ 98% 209.63

ortho-phosphoric acid

Merk 1.00552 85% 97.99

Potassium chloride

SIGMA-ALDRICH P9541 ≥ 99.0% 74.55

SDS-MW Gel buffer

BECKMAN COULTER

A30341 n.a. n.a.

Sodium chloride Honeywell | Fluka 71380 ≥ 99.5% 58.44

Sodium hydroxide solution

Merk 1.09137 1 M 40.00

Sodium phosphate dibasic anhydrous

SIGMA-ALDRICH S7907 ≥ 99.0% 141.96

Sodium phosphate monobasic monohydrate

SIGMA-ALDRICH 71507 ≥ 99.5 137.99

Trifluoroacetic acid

Thermo Scientific 28903 Sequencing grade 114.02

Tris-HCl Promega H5123 ≥ 99.0% 157.56


VI

Trisodium citrate dihydrate

SIGMA-ALDRICH S1804 100% 294.10

Trizma® base SIGMA-ALDRICH T1503 99.9% 121.14

UltraPure™ Tris-HCl pH 8.0

Gibco 15568025 1 M 157.60

VI. Enzymes

Name Supplier Part number Amount in units MW in kDa

FabALACTICA™ Genovis A0-AG1-020 2000 70

FabRICATOR® Genovis A0-FR1-020 2000 38

FragIT™ kit midispin

Genovis A0-FR6-100 n.a. 38

GingisKHAN® Genovis B0-GKH-020 2000 50

GlyCINATOR® Genovis A0-GL1-020 2000 92

Immobilized GlycINATOR® MidiSpin

Genovis A0-GL6-100 n.a. 92

VII. Software

Table 25: List of used software.

Name Version Developer

ChemDraw Ultra 14.0 PerkinElmer

Chromeleon 6.8 Thermo Scientific™

GPMAW 10.0 Lighthouse data

Open Lab CDS Chemstation Edition for LC & LC/MS Systems

C.01.05 [41] Agilent Technologies

PyMOL Molecular Graphics System

1.5.0.5 SCHRÖDINGER


VII

VIII. Chromatogram and electropherogram

I.3. Fab, Fc and LHF


Figure 62: CEX chromatogram of the digestion buffer stability at 37°C.

Figure 63: Digestion progress of the digestion according to Genovis® standard protocol.

12 13 14 15 16 17 18 19 20 21 22

Absorb

ance in m

AU

Time in min.

CEX chromatogram - Digestion buffer stability at 37°C

T0@37°C T1h@37°C T2h@37°C T24h@37°C

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0 15 30 45 60 75 90 105 120

Purity

in A

%

Time in min

Genovis protocol digestion procedure

IgG

LHF

Fc

Fab


VIII

Figure 64: Kinetic model plot ln([A]/[A]0) in dependency of time to justify a first order kinetic model.

Figure 65: CE-SDS NR electropherogram of the Protein L AC flow through and elution.

y = -0.0838x - 0.2684R² = 0.9683

-5.5

-4.5

-3.5

-2.5

-1.5

-0.5

0.5

0 10 20 30 40 50 60

ln([

A]/[A

] 0

Time in min.

Kinetic plot: ln([A]/[A]0) VS. time

ln([A]/[A]0)

Linear (ln([A]/[A]0))

9 10 11 12 13 14 15 16 17

Absorb

ance in m

AU

Time in min.

CE-SDS NR: Pooled and conentrated Protein L AC fractions

Protein L Wash Protein L elution


IX

I.4. F(ab’)2

Figure 66: CE-SDS NR electropherogram of the IdeS digestion and Protein L AC.

I.5. Chitobiose cleavage

Figure 67: CE-SDS red electropherogram of the digestion by immobilized EndoS2 and buffer excahnged sample.

7 8 9 10 11 12 13 14 15 16 17

Absorb

ance in m

AU

Time in min.

CE-SDS NR electropherogram of the Protein L affinity chromatography and F(ab')2 isolation

Reference 1h digestion Flow through

Wash Elute pH 2.5 Elute pH 2.0

15 20 25 30 35 40 45 50

Absorb

ance in m

AU

Time in min.

CE-SDS red. electropherogram of the immob. EndoS2 chitobiose cleavage and buffer exchange

Reference EndoS2 digestion Buffer exchange


X

IX. Copy of the objective setting agreement


XI


XII

X. Statement of authorship

I certify hereby, that I have written this report autonomous, without outside assistance and I

have used only allowed resources. I affirm particularly, that I have mentioned all literal and

paraphrase transfers as such and assigned the original author and document.

Any violation will lead to actions according to §39 and §40 of the General Academic

Regulations for Bachelor’s and Master’s degree programs at the Zurich University of Applied

Sciences of 29.01.2008.

Location Date Signature


XIII

XI. Report in digital format


XIV

XII. Poster

Specific enzymatic digestion and fragment isolation for ... · quality guidelines Q8 – Q12 for...

Documents

Transcript of Specific enzymatic digestion and fragment isolation for ... · quality guidelines Q8 – Q12 for...