TTrends and Developments in BioProcess …€¦ · portfolio of high performance animal-free...

Spring 2008Spring 2008ISSN 1538-8786ISSN 1538-8786

Vol. Vol. 77//No. 1 No. 1 www.bioprocessinwww.bioprocessingjgjournal.comournal.com

A Publication of the A Publication of the WilBio Institute for BioProcess TechnologyWilBio Institute for BioProcess Technology

BioBioPPrrococessingessing J O U R N A LJ O U R N A LTrends and Developments in BioProcess TechnologyTrends and Developments in BioProcess Technology

©200

7 Sa

rtor

ius

Sted

im B

iote

ch

Reusable or disposable – you can rely on our Cell Culture solutions:

BIOSTAT® CultiBag RM is the new generation of Disposable Bioreactors, offering real “Culturing Convenience”.

– For simple cultivation of cells and products

– No SIP (sterilization in place) required

– No CIP (cleaning in place) required

– Validation guide and extractable data available

– Superior process control

– Comprehensive ‘ready to use’ packages

– For all your disposable applications

Sartorius Stedim BiotechUSA +1.800.368.7178Europe +49.5661.71.3400 www.sartorius-stedim.com/biostat_cultibag

turning science into solutions

It’s your Choice!

Med

ia P

rep

Cell

Cult

ure

& F

erm

enta

tion

Capt

urin

gPu

rific

atio

n &

Pol

ishi

ngFo

rm &

Fill

Engi

neer

ing

Serv

ices

research preclinical clinical commercial

1

One Automated Modular Analyzer for Fast Comprehensive

Cell Culture Analysis

Flexible Modular DesignBioProfile FLEX can be customized with 1–4 modules to consolidate up to 15 vital cell culture tests:

Module 1: Gluc, Lac, Gln, Glu, NH4+, pH, PCO2, PO2, Na+, K+, Ca++,

by ElectrochemistryModule 2: Osmolality, by Freezing Point DepressionModule 3: Cell Density/Cell Viability, by Digital ImagingModule 4: Optional Chemistries - - Phosphate by Photometry

Under Development

The Power of OneBioProfile FLEX can reduce cell culture analysis time, labor, and operating costs by consolidating multiple analyzers into a single, easy to use workstation.

One compact workstation versus 4 or 5 separate analyzersOne small, 1 mL sample conserves cell culture mass and end productOne fast, 6 minute analysis saves at least one hour over multiple instrument analysesOne integrated data source simplifies data collection, analysis, and archivingOne validation saves time over validation of multiple instruments

www.novabiomedical.com/bioprocessing0408.htm

Gluc Lac Gln Glu NH4+ pH PO2 PCO2 Na+ K+ Ca++ CD CV Osm PO4

WITHNEW OPC

CONNECTIVITYSUITE

2

New Brunswick ScientificInnovative equipment for culture growth, detection & storage

800-668-5040 • 732-287-1200 • [email protected]

For downloadable brochure see: www.nbsc.com/wbf.htm

Advanced Protein ProductionMade Easy

NEW CelliGen® 310 Bioreactor controls 1 to 4 vesselsand over 120 parameters simultaneously!

• Easily integrate multiple gas flow controllers, sensors, scales, and otherexternal devices for optimized process control.

• Auser-friendly 15" touchscreen controller integrates itall, for simplified setup,calibration and operation.

• A variety of specialized low-shear impellers optimize yields. Packed-bedbasket for high-yield production of proteins and secreted products shown.

• Compact, cGMP-compliant, and available with choice of interchangeable2.2 - 14.0 liter vessels.

Spring 2008 Spring 2008 BioProcessing JournalBioProcessing Journal • • www.bioprocessingjournal.com3

FEATURE ARTICLE

10 Virus Safety Testing for the Next Generation of Biologics

Daniel N. Galbraith

CONFERENCE EXCLUSIVE

14 Current Practices in Heart Valve Preservation (Part Two)

Kelvin G.M. Brockbank

18 Innovative Disposable Bioreactors for Membrane Protein Production Based on the Tide Principle

Lewis Ho

28 Dendritic Cell VaccineProduction Facility: From Design to Operation

Nicolas Taquet

SUPPLIER SIDE

34 How to Construct a Monoclonal Antibody Factory: A Comparison of Production Costs in Fed Batch-and Perfusion Culture with Microcarriers (Part One)

Björn Lundgren

38 Extreme Pipetting: An Issue of Mountainous ProportionsUnderstanding Barometric Pressure’s Effect on Data Integrity

George Rodrigues

42 Bioinformatics Application Has A BLAST With Exanodes Shared Internal Storage

Frank Gana

DEPARTMENTS

6 From the Publisher

44 New Products & Services

48 Advertiser Info

Trends and Developments In BioProcess Technology

Please send your article abstracts and manuscripts, editorials, cover images, press releases, and comments to: [email protected]

BioProcessing J O U R N A L

Spring 2008

On the Cover: Representative optical micrograph depicting primary smooth muscle cells post-treatment to cell electrospinning at a time point of three weeks (see BioProcessing Journal, Fall 2007, Vol. 6/No. 3, pp 42-48).

– Image courtesy of Suwan N. Jayasinghe, PhD. University College London Dept. of Mechanical Engineering

www.bioprocessingjournal.com

contentscontentsVol. 7/No. 1

Spring 2008 BioProcessing Journal • • www.bioprocessingjournal.com4

BIOPROCESSING JOURNAL EDITORIAL ADVISORY BOARD

A Publication of the WilBio Institute for BioProcess TechnologyBioProcessing Journal (ISSN 1538-8786) is published four times per year in Virginia Beach, Virginia, USA.

Periodicals postage paid at Virginia Beach, VA and additional mailing offices.

Postmaster, send change of address to: BioProcessing Journal, P.O. Box 1229, Virginia Beach, VA 23451 USA Editorial contact: [email protected] • Advertising contact: [email protected]

To ship ad materials: Production, BioProcessing Journal, 1321 Laskin Rd., Suite 201, Virginia Beach, VA 23451, USAMail: P.O. Box 1229, Virginia Beach, VA 23451, USA • Phone: 757.423.8823 • Fax: 757.423.2065

Address or information updates: [email protected]

Printed in the USA. © 2008 BioProcessing Journal. All rights reserved.

Publisher Keith L. Carson

Editor-in-Chief Manisha Trivedi

Managing Editor Marci Brown

Associate Editors Jessica Carson

M. Joan LaCount

Production & Design Joy Rogers

Advertising Coordinator Dhalia Edwards

Accounting Assistant Melissa Armstrong

Trends and Developments in BioProcess Technologywww.bioprocessingjournal.com

Darcy Birse, PhDGE Healthcare

Gregory Blank, PhDGenentech, Inc.

Bryan T. Butman, PhDGenVec, Inc.

Greg Dean, PhDMedImmune Ltd.

John J. DoughertyEli Lilly & Company

Roshni Dutton, PhDBioProcess Assist (BPA) Ltd.

Pete GagnonValidated Biosystems, Inc.

Scott HallerRegeneron Pharmaceuticals, Inc.

Alan Hardwick, PhDChromos Molecular Systems, Inc.

Arnold H. Horwitz, PhDXOMA Corporation

Beth Hutchins, PhDSchering Plough Biopharma

Darshana JaniBiogen Idec, Inc.

Donald Jarvis, PhDUniversity of Wyoming

Suwan N. Jayasinghe, PhDUniversity College London

Amine Kamen, PhDCanada’s Biotechnology Research Institute

Brian Kearns, PhDImClone Systems, Inc.

Bob Kennedy, PhDBoard of EducationTenafl y, New Jersey

Tom Kost, PhDGlaxoSmithKline

Michael LaBarre, PhDParamount Biosciences

Bruce Levine, PhDUniversity of Pennsylvania Health System

Anthony S. Lubiniecki, ScDCentocor, Inc.

Phillip Maples, PhDGradalis, Inc.

Mary Pat Moyer, PhDINCELL Corporation, LLC

Vincent NarbutBiogen Idec, Inc.

Kim Nelson, PhDCRB Consulting Engineers, Inc.

Erno Pungor, PhDBioMarin Pharmaceutical, Inc.

Robin Robinson, PhDUS Dept. of HHS, FDA

Joe SenesacIntrogen Therapeutics, Inc.

Richard Siegel, PhDCentocor, Inc.

Gale Smith, PhDNovavax, Inc.

Noelle-Ann Sunstrom, PhDNeuClone, Inc.

Martin Vanderlaan, PhDGenentech, Inc.

Michael Washabaugh, PhDMerck & Company, Inc.


Cell Engineering andCell Engineering and BankingBankingBaculovirus ExpressionBaculovirus ExpressionCell Culture Scale-UpCell Culture Scale-UpFiltrationFiltrationPurificationPurifi cationCharacterization andCharacterization and ComparabilityComparabilityViral Vectors and VaccinesViral Vectors and VaccinesAntibody DevelopmentAntibody Development and Productionand ProductionSingle-Use BioProcessingSingle-Use BioProcessing Components and SystemsComponents and SystemsFacility Design,Facility Design, Construction, andConstruction, and OperationOperationBiomarkersBiomarkersCellular Therapies,Cellular Therapies, Vaccines, and TissueVaccines, and Tissue RegenerationRegenerationRaw Material QualityRaw Material Quality ProgramsPrograms

The content ofThe content of eacheach JournalJournal issue issue concentrates on theconcentrates on the topics covered in ourtopics covered in our WilBio ConferencesWilBio Conferences..We welcome articlesWe welcome articles based on:based on:

[email protected]

5

When you need a manufacturing partner that is:

• Able to deliver both safety and productivity, with the broadest portfolio of high performance animal-free supplements

• Uniquely focused on rigorous quality and regulatory needs, with the industry’s most stringent animal-free policy

• Dedicated to quality innovation, with the industry’s mostcomprehensive automated change notification system

…choose BD and our novel portfolio of hydrolysates, cell culturemedia supplements and services.

For more information, visit us today at bdbiosciences.com/advbioor call BD Technical Service and Support at 800.638.8663.

I wish I had apartner whohas my safetyconcerns in mind fromthe beginning.”We do that.

“

BD BiosciencesAdvanced Bioprocessing800.638.8663bdbiosciences.com/advbioBD and BD Logo are trademarks of Becton, Dickinson and Company. ©2007 BD

BD Biosciences – Advanced BioprocessingHydrolysates • Cell Culture Media Supplements • Services


from thefrom the publisher publisher he biotech industry appears to be struggling instead of moving forward at the pace we have come to expect. Not enough new products are in the development and regulatory pipeline, and a number of huge revenue- producing products are coming off patent.

Five years ago, products were being licensed on a regular basis, and there were numerous small companies engaged in developing biologics, or supplying products and services to the biotech companies. Today, many of these small companies are gone, or at least their trade names and technologies are no longer visible to the casual observer.

So what happened? In my opinion, the housing market wasn’t the only victim of cheap money. Many large biotech and supply companies had access to fi nancing that was too good to pass up. With this cash, many of them decided they would please their investors by showing growth they could achieve through acquisition.

Of course, some of these acquisitions worked out fairly well, but in far too many cases they didn’t produce the desired results. What seems to have been missing is an understanding of how the acquisition would be assimilated into the existing corporate structure and culture. In addition, too much attention was paid to changing the company and trade names, and too little was invested in working out how to make the chemistries of the key players coexistent.

Then after spending lots of money on attorneys and investment bankers, plus public relations and new letterheads, the realities began to sink in: How would the new products or services be marketed, and which sales force or distribution system would sell them? Where would the products be made and the profi ts be taken?

In the case of biologics: Were these technologies that should be brought into the development process immediately to produce new products or enhance existing ones? Or, did it really matter what happened since the technology was safely tucked away and might only be used for royalty income?

But then, the fi nancing had to be paid back, and possibly at higher rates than initially expected. And the only way to do that was by cutting costs and eliminating duplication, which meant laying off or moving people, and selling buildings. But what about the value that was being purchased in the fi rst place, such as the people, expertise, technology, and good will in the marketplace? Didn’t these things matter?

Today, there are fewer companies and the desired number of new products just hasn’t materialized. But the real travesty is the loss of the small companies that were creating so many of the highly skilled jobs, and providing the innovation that was driving much of the industry. And what happened to all the displaced workers?

Instead of forming partnerships to capitalize on the benefi ts these small companies had to offer, the acquiring companies had to have it all, or nothing. And unfortunately, too many of them ended up with nothing.

Keith L. Carson, Chairman and Publisher [email protected]

T

8

2nd International

Single-UseBioProcessComponentsand SystemsEnsuring the Quality and Suitabilityof Single-Use Components

JULY 28-30, 2008SAN FRANCISCO CALIFORNIA

• Single-Use versus Reusable Components

• System Integration and Validation

• Vendor Certifi cation and Audits

• Disposal Issues• Testing Requirements• Media and Buffers• Chromatography Resins• Animal-Sourced Materials• Critical Raw Materials• Storage and Qualifi cation

As more biopharmaceutical products enter clinical trials and commercial production, it is becoming ever more challenging to guarantee the supply of disposable process components and critical raw materials. The sheer volume of some substances is testing the limits of many suppliers, while more application move toward the use of prepared and pre-sterilized materials. Concurrently, quality control expectations are expanding to keep pace with the availability of increasingly powerful analytical techniques.

As a response to these trends, reliable vendors must be identifi ed, evaluated, and then worked with closely to make sure they understand and are able to satisfy the volume, quality, documentation, and communication requirements of this rapidly growing industry.

Register Now!www.wilbio.com

757.423.8823

Meeting Held Concurrently with

Raw Materials and Contract Services

SPONSORED BY:

Topi

cs: The Williamsburg BioProcessing Foundation

The Most Trusted Source of BioProcess Technology®

�Featuring a Tour ofFeaturing a Tour ofBaxter Healthcare’sBaxter Healthcare’sSingle-Use ContractSingle-Use Contract

Manufacturing FacilityManufacturing Facility

SUB08Ad_JR.indd 1 5/4/2008 11:14:57 AM

9

The Williamsburg BioProcessing FoundationThe Most Trusted Source of BioProcess Technology®

JULY 28-30, 2008SAN FRANCISCO CALIFORNIA

Raw MaterialsRaw Materials and Contractand Contract ServicesServices

9th International

For MammalianCell Production

Topics:• Animal-Sourced Materials• Storage and Identification Systems

• In-House Testing Requirements

• Vendor Certification and Audits

• Traceability of Components

• Defining Critical Materials

• Regulatory and Legal Ramifications

• Outside Testing Services

• Material Suitability

• Managing Vendor Relationships

• Cell Lines as Raw Materials�Featuring a Tour ofFeaturing a Tour ofBaxter Healthcare’sBaxter Healthcare’sSingle-Use ContractSingle-Use Contract

Manufacturing FacilityManufacturing Facility

Meeting Held Concurrently with Single-Use BioProcessComponents and Systems

Register Now!www.wilbio.com757.423.8823

RMC08Ad_JR_MB.indd 1 5/5/2008 11:17:09 AM

FEATURE


Virus Safety Testing for the Next Generation of Biologics

Today, many if not all of the major pharmaceutical companies have specifi c budgets for large research and development pipelines

for new biologics including vaccines, monoclonal antibodies, recombinant proteins and even gene therapy projects. In addition, small innovative compa-nies are continually presenting us with novel treatment opportunities which hold great promise for the future. The reason for such huge optimism in the fi eld of biologics is that there are now a number of “blockbuster” drugs for which the safety profi le has been very successfully managed.1 This success has been achieved following a considerable amount of time and fi nancial invest-ment by both the innovators and the regulators.

There are a number of hurdles to overcome for new biologics. A very important one involves the potential for viruses to contaminate the produc-tion process. The safety of a new drug is always of paramount importance, but with the experience gained over the past 30 years, perhaps we are at a juncture where the established virus safety testing protocols for new products should be reviewed.

The reason behind the need for virus safety testing of biologics can be traced back to the introduction of the fi rst mass-vaccine products manufac-tured during the second half of the 20th century. These products were used to prevent viral diseases which were a major cause of pathology at the time.

Such vaccines were the fi rst medicines to be produced on a large scale using cells as the basis of the production system. The best example is polio vaccine which is probably one of the greatest success stories of our time. Mass- vaccinations have succeeded in controlling and virtually eliminating this disease.

The cells used in polio vaccine pro-duction were originally prepared from animal organs. Back then, many of the donors were sourced from uncontrolled populations with very little screening for health status. There is published evidence that adventitious viruses pres-ent in the animal cells were introduced into the polio vaccine.2

There have been a number of vaccines and products manufactured from human blood or blood derivatives which have resulted in the transmis-sion of adventitious agents (particularly

viruses) to the recipient. All of these incidents led to the creation of test-ing guidelines for vaccines which were implemented in Europe and the United States to ensure patient safety. With improvements in aseptic techniques and air handling and a general lifting in cGMP standards for the production of drugs came huge improvements in the safety of these products.

With the advent of recombinant DNA technology in the 1980s, proteins could be manufactured to order. This offered the opportunity to treat disease using a completely new array of tech-niques. Cell culture methods proved to be extremely useful as cells could be transformed with the appropriate gene of interest and then frozen in banks to be recovered for production when required. The use of cells is similar to that of the vaccine industry, however there were considerable differences with

By Daniel N. Galbraith

Daniel N. Galbraith, MSc, PhD ([email protected]), is the head of operations with BioOutsource Ltd, United Kingdom.


the safety profi le of these cells when compared with primary cells sourced directly from animals.

It should also be noted that the regulators who were providing guidance to the early biologics manufacturers were presented with a product that had not been manufactured in the aseptic stainless steel facilities now used to manufacture recombinant proteins or monoclonals. Some of the fi rst mono-clonals were produced using mouse ascites fl uid, a technique reliant on animals and very prone to contamina-tion. Virus safety testing standards for these new products were based on guidance designed for the only other cell-derived products being manufac-tured— vaccines.3,4 Much of the regula-tory guidance for biologics production involves virus detection techniques that have been around for some time, as far back as the 1960s or earlier.

The safety testing for biologics uses a three-staged approach to assure safety: 1) testing the starting materials such as the cells and media used in the produc-tion; 2) testing of “in-process” samples once the cell culture period has ended; and 3) testing the virus reduction capacities of the downstream process-ing. The fi rst two parts are similar to the safety testing strategies laid out in the vaccine guidelines of the European Pharmacopeia and the United States Code of Federal Regulations. The third step is clearly not useful in the manu-facture of a virus vaccine.

The virus testing strategies described in almost all regulations for biologics splits viruses neatly into three areas: 1) the testing for adventitious viruses; 2) the testing for retroviruses; 3) and the testing for species-specifi c viruses. Testing for adventitious viruses covers those which are likely to be introduced in either the starting materials or as a breakdown of the aseptic protocols used during the manufacturing pro-cess. Retrovirus testing is normally concerned with the endogenous retroviruses which infect the production cell line. Testing for species-specifi c viruses is again concerned with viruses inadvertently introduced from the pro-duction cell line.

Despite these three neat areas, viruses themselves do not fall easily into one specifi c category—some viruses can be considered both adventitious and species-specifi c or retrovirus. Guidelines specify the types of testing necessary to fulfi l the safety require-ments. They normally include two tests (in vitro and in vivo) which have the ability to detect a broad range of adventitious viruses. The in vitro test is a cell-based assay which detects viruses by inoculating a sample onto particular “detector cells” that are susceptible to adventitious viruses. The types of cells (such as Vero or MRC-5) considered suitable for use are specifi ed in the guidelines. Samples are observed over a specifi ed period of time for the presence of cytopathic events which can signal the presence of viruses. The in vivo assay is similar to the in vitro method but rather, samples are inoculated into laboratory animals (Guinea pigs, adult or suckling mice, and embryo-nated chickens eggs) and subsequently observed for pathological markers indicating virus infection.

Virologists have been using these types of assays since the turn of the 1900s and they are still popular today. The assays are generally simple to carry out but require a very high level of expertise to determine the presence or absence of virus in a test sample. One diffi culty in interpretation is that many biologic drugs cause sample-related pathology and abnormal conditions in the test animals or cell cultures.

A major issue with assessing the potential risk of contamination posed by most viruses is that there has been so little work carried out on the vast majority of viruses and their inherent ability to infect cells used in produc-tion such as Chinese hamster ovary (CHO). Most risk strategies are based on what is known about a relatively few model viruses. An example of where this model theory fails is that of Bovine polyomavirus (BPyV), once thought to be a high risk for contamination of production cell lines due its ubiquitous nature and the use of fetal calf serum (FCS) to supplement growth media for cell culture.5 BPyV viruses are a well-

known risk for the contamination of laboratory cell cultures. Hence, it was thought that should BPyV be present (contaminating cell cultures in the typi-cal polyoma pattern of infection), the in vitro assay would identify it. However, in vitro culture of the virus described an unusual pattern of growth which included a long period of latency for the virus without identifi able replica-tion. This suggested that infection of cells would not be identifi ed using the tried-and-tested techniques and additionally—maybe there were other viruses that would not be identifi ed using the familiar techniques.

Retrovirus testing can become a complicated regimen under the current guidelines and requires a number of techniques to be applied to satisfy the regulations. Retroviruses are perhaps one of the most interesting virus fami-lies yet described. They are ubiquitous agents present in all animal species investigated thus far. Their ability to integrate into the host cell genomes and remain silent does present a number of diffi culties when testing for their presence. There are three strategies used to detect retroviruses: 1) molecu-lar techniques to detect the present of the reverse transcriptase enzyme, an enzyme common to all retroviruses but absent in almost all other viruses; 2) electron microscopy to observe the virus particles; and 3) cell culture techniques used to detect the virus by infection of cells and subsequent cell pathology.

Retroviruses, although ever-present in all of the common production cell lines (CHO and NS0), do have a consid-erable disadvantage with respect to their ability to be transmitted. They are very easily inactivated when outside of the host cell and in adverse conditions and typically have very specifi c requirements related to the cell receptors to which these viruses will bind.

The fi nal viruses to consider are termed “species-specifi c viruses” which may be introduced dependant on the type of the production cells or any animal-derived reagent used. Testing for these requires an array of techniques which are targeted at specifi c viruses or


virus families. These techniques can be prescriptive such as the mouse antibody production (MAP) assay to test viruses which may be present in mouse cells or be generic such as the polymerase chain reaction (PCR) assay used to detect the presence of viruses in a human produc-tion cell line.

Using the CHO cell as an example, a testing package for regulatory submis-sion in Europe or USA is described in Table 1. This example includes an assay to determine any specifi c bovine viruses (assuming the cells have been cultured in media containing FCS some time in their history), and an assay to determine any specifi c porcine viruses (assuming the cells have been passaged using por-cine trypsin). A package of work such as this would normally take 6-8 weeks to complete for the master cell bank (MCB) and around 4 weeks for an “in-process” test.

When considering a risk-based strategy for a product manufactured on CHO cells, the MCB presents the high-est risk of contaminating viruses as typi-cally these are cells which have been in a less controlled environment than that of cGMP manufacturing. Cells may have been cultured near other cell lines, some of which may have been express-ing endogenous or other viruses. The reagents used to culture the cells may not have been quarantined, tested and released as in a cGMP environment and therefore may have contained viruses.

Evidence has shown that the high-est risks of virus contamination are related to: 1) the use of animal prod-ucts in the process; and 2) as a result of a breakdown in cGMP, allowing contamination by operators or other external sources. Viruses have been introduced following the use of bovine products during production which can be considered an extremely high-risk practice. Inadequate personnel training or lapses in cGMP compliance have also resulted in compromised product integrity. Publications have described that, in addition to the contaminating viruses from FCS, other kinds of virus infection in production processes have been noted6— one in particular isMinute mouse virus (MMV). This virus

has been shown to be transmitted to CHO production from wild mice. This may be considered to be a breakdown of cGMP as appropriate precautions were not undertaken to ensure the integrity of materials used in production.

Therefore, the production process should have a viral risk assessment to ascertain when and where these hazards might enter the process. To be a risk to the product, viruses must be able to survive and replicate in the environ-ment provided, otherwise they will be inactivated or diluted out during pas-sage. A prerequisite for virus replica-tion is that it must be able to attach to and enter the host cell—in the preced-ing example, CHO cells.

To be “successful” in sustaining an infection, a contaminating virus will use one of three strategies: 1) continuously replicate in the production cells; 2) inte-grate into the DNA of the host cell such as accomplished by the retroviruses and some Parvoviruses; and 3) have the abil-ity to remain latent (such as members of the Herpes virus family) where it can remain in cell culture without causing signifi cant cell pathology but retain suffi cient viral functions to maintain infection.

Detecting a replicating virus can be relatively easy. As in the case of most viruses, an infection of cells results in

a cytopathic effect that be identifi ed by the in vitro assay. When attempting to detect viruses which are slow- growing or do not exhibit cytopathic effect on infection, this can prove diffi cult. Retroviruses fall into this category but have their own suite of tests to detect their presence or absence. The dif-fi culty comes when trying to detect the presence of an integrated virus.

By their nature, retroviruses can remain integrated without expression of viral proteins. Detecting such viruses requires molecular techniques designed to target the genome. Latent viruses require specifi c detection assays and some must be induced out of the latent stage for detection to be successful. The in vivo tests and other assays are not capable of detecting the presence of these inapparent viruses. In many cases, the use of cell- or animal-based assays do not provide the solution.

There are technologies such as molecular-based techniques which can be applied to detect the presence of viruses without the observation approaches of in vitro and in vivo assays. A simple process which has become very commonplace is the polymerase chain reaction (PCR) technique. All viruses for which there is sequence known can potentially be identifi edusing PCR. One limitation to PCR is

TABLE 1. An example of a CHO cell production testing package for regulatory submission.


that only a single virus type (or rarely, a virus family) can be detected using this single assay. This would makeattempting to detect all possible viruseswhich can infect CHO cells for example, an immensely time-consuming and costly task.

While this does pose a challenge for molecular-based techniques, now there are microarray techniques which can be targeted to thousands of genetic sequences to detect a wider spectrum of viruses. The overall sensitivity of these detection methods remains to be defi ned. Techniques are already being published which describe the viral detection methodology based on a combination of viral genomics and long oligonucleotide DNA microarray tech-nology. Highly-conserved sequences within a viral family have been used on the microarray, and as such, uniden-tifi ed or newly-evolved virus family members could be detected. Coupling this detection to the amplifi cation of the target region by PCR would increase the sensitivity of the assay to close to 100 copies of virus or less in a test sample.

Another especially interesting technique involves adapting the cell used in the in vitro assay to detect the contaminating viruses. Cells can be manipulated to signal infection of virus rather than relying on a more obvious cytopathic effect. Following virus infec-tion, cells will initiate a number of intra and extracellular signals—one of the most widely studied being the inter-

feron cascade. If a marker gene such as luciferase were coupled to the switching on of such genes then these cells would signal the presence of virus infection very early on, and even latent viruses could be identifi ed in this way. The choice of cells would also be important because depending on the cell recep-tors available on the production cells, this will dictate the viruses which are able to enter and replicate in these cells. The production cells could be modifi ed using the reporter gene and then used as a detector cell line to indicate virus infection or alternatively, the traditional cell types used to detect viruses such as Vero or MRC-5 cells could be used.

In Conclusion

Viruses are a known risk factor for biologics produced on cell lines. The number of agents being identifi ed is ever-increasing—along with the associ-ated risks which require assessment. The techniques we have used since the inception of vaccines in the 1950s have been successful in ensuring the safety of the majority of products. In this, there is no doubt. However, as we move to the future and treat many more patients with an increasing array of novel drugs, the chances of a mistake or breakdown in-processes become ever more likely. With respect to the viral risk of these products, the detection techniques we have been using provide a fair degree of reassurance. The biologics industry is now aware that there are additional

techniques which can be applied to provide us with increased reliability. Should we begin to remove the require-ment for some of the traditional tech-niques which may have been important at one stage but have now outlived their usefulness? Through risk-based analysis and thorough validation of the new techniques, older protocols could be replaced. This will improve the indus-try’s profi le with the wider society and maintain the highest levels of safety and integrity in our products.

REFERENCES

1. Blockbuster Drugs 2006. La Merie Business Intelligence, July 2006.2. Immunization safety review: SV40 contamina-tion of polio vaccine and cancer, October 2002. Institute of Medicines report. Stratton K, Almario DA, and McCormick MC, Editors, Immunization Safety Review Committee. 3. Points to consider in the production and testing of new drugs and biologicals produced by recombinant DNA technology. US FDA’s Center for Drugs and Biologics, 1985.4. Carthew P. Is rodent virus contamination of monoclonal antibody preparations for use in human therapy a hazard? Journal of General Virology (1986) 67, 963-974.5. Nairn C, Lovatt A, Galbraith DN. Detection of infectious bovine polyomavirus. Biologicals, Vol.31/No.4, Dec. 2003, pp. 303-306.6. Vilcek S. Identifi cation of pestiviruses contaminating cell lines and fetal calf sera. Acta Virol 2001 Apr;45(2):81-6.

A Call for PapersA Call for PapersOur peer-reviewed Journal provides valuable information on industry trends, key people, and the technologies and services available from the industry’s leading supplier companies.

Find out more:

www.bioprocessingjournal.com or [email protected]




� PART TWO IN A TWO-PART SERIES �

Current Practices in Heart Valve Preservation

By KELVIN G.M. BROCKBANK

This is the second and fi nal part of this article. Part One was published in the Fall 2007 issue of the

BioProcessing Journal. In this part, the journey of an allograft heart valve is continued, discussing current practices from donation to just prior to implantation in the surgery.

Kelvin G.M. Brockbank, PhD ([email protected]), Cell & Tissue Systems, Inc., 2231 Technical Parkway, North Charleston, SC 29406; and the Georgia Tech/Emory Center for the Engineering of Living Tissues, Georgia Institute of Technology, 315 Ferst Drive, Atlanta, GA 30332. This article is based on a presentation given at The Williamsburg BioProcessing Foundation’s 12th international Cell & Tissue BioProcessing meeting held in Austin, Texas, October 29-31, 2007.

Section 4: Tissue Storage and Transportation Conditions

Storage and shipping variables may signifi cantly impact cryopreserved product quality. Some of the variables that are important aspects of any tissue preservation process are summarized in Figure 4.

In this particular article, the focus is on temperature control issues. The AATB (American Association of Tissue Banks) Standards for Tissue Banking1 state that cryopreserved cardiac allografts shall be maintained at tem-peratures of –100°C or colder. Heart valves are usually stored below –135°C in vapor phase nitrogen. There have been few published studies of higher (warmer) storage temperatures. Most heart valves are transported using dry shippers that maintain vapor phase nitrogen temperatures. These contain-ment devices are expensive, and the

costs for two-way shipping are sig-nifi cant due to their size and weight. Considerable savings could be had if dry ice shippers with temperature excur-sions warmer than the AATB standard (of up to –70°C) could be employed.

Most of the published evidence for tissue storage and shipping tem-peratures support using vapor phase nitrogen storage temperatures below –135°C; not –100°C. We are unaware of any studies done in the vicinity of –100°C. There is one paper reporting no changes in viability after one week at –80°C,13 but there aren’t any known publications covering the effects of dry ice at temperature variations of –79.6°C (when fresh) to –70°C (after

contamination with water). Extensive investigation is needed because some processors employ shipping conditions that can result in tissue exposures of up to –70°C. Furthermore, it is not known whether or not there are negative consequences associated with placing tissues that have been exposed to such warm temperatures back in vapor phase nitrogen. Theoretically, damage to the tissue matrix may occur due to changes in ice crystal structure during the warming/cooling cycle, and if the rate of re-cooling is too rapid, there may be an increase in the risk of tissue fractures:

• What is an acceptable cooling rate?

• How many times can tissue be warmed

FIGURE 4. Critical issues for storage and distribution of cryopreserved tissues. (Figures 1–3 are shown in Part One of this article.)

Selection of Materials

• Packaging, labeling — compatible with process

Inventory Management

• Large inventories may require sophisticated inventory management (e.g., bar coding)

Temperature Control

• Thermal cycling can degrade samples

• Avoid exposure to temperatures above the glass transition

Container Integrity

• Avoid submersing samples in liquid nitrogen

• Container leakage — source of contamination


and then cooled back to vapor phase nitrogen temperatures for long term storage without negative consequences?

• How warm can a heart valve be — and for how long — without suffering detri-mental consequences upon recooling?

Temperature exposures warmer than –80°C have not been investigated.

If the storage temperature is suffi -ciently low — below the glass transition point of the cryopreservation solution (approximately –123°C for DMSO cryoprotectant formulations) — little, if any, change occurs in biological materi-als.9,10 It has been demonstrated that human heart valve leafl ets retain their ability to synthesize protein for at least two years when stored below –135°C.13 Degradative processes may occur at and above the solution’s glass transi-tion temperature. For example, it has been shown that cells in cryopreserved human heart valve leafl ets may be negatively affected by prolonged storage at –80°C for periods greater than one week.13

Fractures can be attributed to abrupt changes in temperature, espe-cially in tissues which were not cooled in a controlled manner to an appropri-ately low temperature (–80 to –100°C) prior to immersion in vapor or liquid nitrogen. Furthermore, going rapidly

from nitrogen storage temperatures directly into warming solutions may also result in fractures. The immersion of cryopreserved human valves directly into liquid nitrogen for as little as fi ve minutes can cause tissue fractures.76

It is important to avoid rapid submersion in liquid nitrogen during both storage and also transport from one storage facility to another, and this problem came to light after a hospital’s valve storage system overfi lled during an automatic refi ll cycle. Following normal heart valve thawing procedures in the operating room, it was discovered that there were numerous (full-thick-ness) fractures of the valve conduit.77 This phenomena has been reproduced experimentally.76

Abrupt temperature-dependent changes in aqueous glycerol solutions were described 40 years ago.78 The for-mation and disappearance of fractures depends on the interaction of several factors, in particular: a) the mechanical properties of the material; b) the con-centration of solute; c) the temperature gradients; d) the overall temperature; and e) the rate of temperature change.79

Mechanical forces are present in cryopreserved tissues.80,81 Without a racking system for storage of the tissues, liquid nitrogen freezers may be subject to much greater temperature gradients

and fl uctuations than storage freez-ers operated with a complete racking system in place — even though most of the racks may be empty. The formation of fractures may also be related to mis-handling. Manufacturers anecdotally report that the incidences of fractures increase with time when contained in hospital-controlled freezers. The particular processes employed for cryo-preservation are critical.

The causes of stress fractures, tears or cracks in cryopreserved tissues are not clear. The tissues may be pulled apart by stresses within the tissue, or cut apart by ice growth. Fractures can be shallow and superfi cial, or full thickness (Figure 5). Whether stresses are caused by the differing thermal properties of frozen and unfrozen fractions, dissolved gases or some other cause is not known. It is possible to experimentally induce stress fractures by plunging tissues such as aortic arteries into liquid nitrogen (Figure 5). However, fractures do not occur every time, even under such extreme circumstances, and clinically, tissue fractures are rarely reported by surgeons when the rewarming guide-lines discussed in the next section are followed.

Slow immersion in liquid nitrogen is probably less problematic that rapid immersion. Slow immersion can occur

FIGURE 5. Cryopreserved aorta with experimentally-induced stress fractures. The fractures were induced by plunging aortic tissues from vapor phase nitrogen into liquid nitrogen. (Photographs provided by Professor Wolfi nbarger from studies conducted at Old Dominion University).


when the solenoid valve on a storage unit fails and the storage freezer is over-fi lled with liquid nitrogen. These issues need extensive study. For an in-depth discussion of solid mechanics issues in cryobiology please see the recent review of Rabin and Steif.90

Section 5: Processing Steps and Valve Handling Performed in the Surgical Suite

Thawing of Cryopreserved Tissues:It is generally accepted that rapid

thawing of cells enhances survival. This observation is especially important for rapidly-cooled cells and has been suggested to favor cell survival by sup-pressing the phenomenon known as recrystallization. Recrystallization is a phenomenon common to solutions that have been frozen under nonequilib-rium conditions. Slow cooling typically results in the formation of large crystals, whereas rapid cooling produces smaller crystals. Small ice crystals are unstable because of their high surface energy, and they tend to re-form into large crys-tals to improve their thermodynamic stability. Such transitions readily occur during warming and may result in cell damage that was not present during the actual freezing event. Cells frozen by a slow cooling process have fewer intra-cellular ice crystals and larger crystals, and thus are presumably less sensitive to the rate of thawing.

Rapid thawing is the preferred route for rewarming cryopreserved heart valves as it restricts recrystalliza-tion. The process is normally accom-plished by fi rst warming the tissues to approximately –100°C followed by rapid immersion of the frozen valve in a large volume of water warmed to 42°C. Thawing is normally completed in less than six minutes. As long as care has been taken during packaging and when handling the valve during the thaw-ing process, little mechanical damage occurs.

The warming rate may also play a major role in the formation of tissue fractures or cracks when the tissue is rewarmed from vapor phase nitrogen temperatures. Traditionally, most valve

processors have incorporated safeguards in their warming procedures to reduce the initial warming rate of their tissues. For instance: the valve is removed from the storage freezer or shipper at some distance from the operating room and is wrapped in a towel and carried with-out additional cooling if the distance is short, or placed on dry ice if the distance is more than a couple of rooms away.

The objective of both of these strate-gies is to warm the tissue slowly from below –135°C through the glass transi-tion (–123°C) to –100°C. Alternatively, a set time before initiation of the rapid thaw can be determined such that the tissue reaches approximately –100°C before placing the tissue in a water bath. Wassenaar et al.82 demonstrated a correlation between the formation of cracks in cryopreserved aortic grafts and rapid initial thawing. In our heart valve research we routinely place the valve packages on dry ice for ~50 minutes or a –80°C freezer for a longer period prior to placing them in 37°C water.

Cryoprotectant Removal:After thawing cryopreserved tis-

sues, the cryoprotective agents must be removed. Although the mechanism for DMSO toxicity has not been deter-mined, its ability to affect cells has been well documented.83-86 Until recently, the cryoprotectants were serially diluted following thawing using cold solu-tions and the heart valves were fi nally placed in cold 300–320 mOsm medium

without DMSO. Cryoprotectants have been removed gradually by chang-ing the concentration of the extracel-lular solution in a stepwise fashion because DMSO enters and leaves cells at a slower rate than water. Therefore, there is a tendency for cells to swell as its environment returns to isotonic-ity. Even though the cell volumes will return to normal as the DMSO equili-brates, excessive volumetric excursions and the associated osmotic water fl uxes can result in cell damage. However, the benefi ts of slow, stepwise cryoprotectant removal must be weighed against the increased exposure times to DMSO. Because the cytotoxic effects of DMSO are temperature dependant, most DMSO elution protocols are performed using refrigerated (4– 6°C) solutions.

Carpenter and Brockbank87 intro-duced a one-step method of cryopro-tectant removal from cryopreserved cardiovascular tissues in 1992. Cell viability results following cryoprotec-tant elution by one-step methods in heart valves demonstrated that single-step methods yielded essentially the same viability results as the historically employed multi-step approach.87 An explanation for cell survival in heart valves following cryoprotectant elution by the one-step method is this: Cells embedded within a tissue matrix may survive large changes in their environ-ment more readily than cell suspensions because the movement of solutes may be restricted through the ion-exchange action of the macro-molecular matrix.

HEA

RT V

ALV

E IM

AG

ES K

IND

LY P

ROV

IDED

BY

LIF

ENET

OF

VIR

GIN

IA


The addition of an impermeant solute may also help to reduce the risk of osmotic shock to heart valve cells dur-ing one-step cryoprotectant removal protocols.87

Mannitol is commonly employed as an impermeant solute, with respect to cells, in the washout solutions employed for cryoprotectant elution from tissues. In contrast to heart valves, studies in other tissue models such as that of Taylor using a corneal model have indicated that serial dilution of DMSO was preferred to one-step removal.88 The major advantage of the one-step method appears to be that it is easier for the operating room staff to perform than the multi-step procedure in which the valve must be moved to the next step in the dilution series every few minutes. The major theoretical disad-vantage of the one-step method is that more residual DMSO may remain in the valve resulting in an increased risk of patient reactions to the DMSO and exhaled DMSO breakdown products.

More recently, a gradient-based method of cryoprotectant removal was developed that combines positive features of both the multistep and the single step methods.89 From the per-spective of tissue handling, this method is simple since it only involves one step. However, the method is essen-tially composed of many dilution steps performed in a continuous manner with the resulting reduction in risk of osmotic shock associated with multi-step methods. The continuously-stirred diluting solution is perfused through and around the tissue, maximizing the tissue DMSO gradient. The tissue DMSO gradient is also maximized by the continuous clearance of the DMSO as it is eluted from the tissue.

Like the one-step process, the continuous process may incorporate a nonpermeating solute in the wash solution. An additional benefi t of the continuous process is that washout may be performed at room temperature to permit more rapid restoration of meta-bolic imbalances. However, providing there are no concerns regarding cell viability, and with appropriate matrix validation, there is probably no reason

that room temperature or higher could not be employed for either single or multistep procedures. The development of more sensitive quantitative methods for evaluation of the consequences of process changes that may impact heart valve matrix quality and clinical perfor-mance is needed.

Conclusion

The conditions of tissue storage and transportation promoted by the AATB standards are likely highly conservative and have signifi cant opportunities for improvement. Poor funding because of the small market size for allograft heart valves and the lack of accepted extracel-lular matrix test methods have limited the amount of research done in this area. Cell viability — the major heart valve quality control method used for years — is no longer considered to be important for valve storage.

ACKNOWLEDGEMENTS

I would like to thank Elizabeth Greene and Alma Vasquez for their assistance in manuscript preparation; Dr. Katja Schenke-Layland, David Geffen School of Medicine at UCLA, Los Angeles, for her multiphoton and second harmonic micro-graphs in Part One of this article; and Professor Lloyd Wolfi nbarger, Old Dominion University, Norfolk, Virginia, for his discussion and photos on tissue fractures in Part Two.

REFERENCES

(Numbers 1, 9, 10, and 13 were also discussed in Part One.)

1. Standards for Tissue Banking. 11th Edition, American Association of Tissue Banks, 2006.9. Karow AM. Biophysical and chemical consid-erations in cryopreservation. In: Organ Preserva-tion for Transplantation. Karow AM, Pegg DE, (Eds), Dekker, New York, p 113, 1981.10. Mazur P. Freezing of living cells: mecha-nisms and implications. Am J Physiol 247:125, 1984.13. Brockbank KGM, Carpenter JF, Dawson PE. Effects of storage temperature on viable biopros-thetic heart valves. Cryobiol 29:537, 1992.76. Adam M, Hu JF, Lange P, Wolfi nbarger L. The effect of liquid nitrogen submersion on cryopre-served human heart valves. Cryobiol 27:605-614,

1990.77. Wolfi nbarger L, Adam M, Lange P, Hu JF. Microfractures in cryopreserved heart valves: valve submersion in liquid nitrogen revisited. Applications of Cryogenic Technology, Plenum Press, 10:227-233, 1991.78. Kroener C, Luyet B. Discontinuous change in expansion coeffi cient at the glass transition temperature in aqueous solutions of glycerol. Biodynamica 10:41-45, 1966.79. Kroener C, Luyet B. Formation of cracks during the vitrifi cation of glycerol solutions and disappearance of the cracks during rewarming. Biodynamica 10:47-51, 1966. 80. Rubinsky B, Lee C, Bastacky J, Onik G. The process of freezing in the liver and the mecha-nisms of damage. In: Proceedings, CRYO 87, 24th Annual Meeting, 1987.81. Rajotte R, Shnitka T, Liburd E, Dossetor J, Voss W. Histological studies on cultured canine heart valves recovered from –196°C. Cryobiol 14:15-22, 1977.82. Wassenaar C, Wijsmuller EG, van Herwerden LA, Aghai Z, van Tricht C, Bos E. Cracks in cryo-preserved aortic allografts and rapid thawing. Ann Thorac Surg 60:S165-167, 1995.83. Barnett RE. The effects of dimethylsuphoxide and glycerol on Na+/K+–ATPase and membrane structure. Cryobiol 15:227, 1978.84. Miranda AF, Nette G, Khan S, Brockbank, KGM, Schonberg M. Alteration of myoblast phenotype by dimethylsulfoxide. Proc Natl Acad Sci USA 75:3826-3830, 1978. 85. Katsuda S, Okada Y, Nakanishi I. Dimethyl sulfoxide induces microtubule formation in cultured arterial smooth muscle cells. Cell Biol Int Rep 11:103-110, 1987.86. Katsuda S, Okada Y, Nakanishi I, Tanaka J. The infl uence of dimethyl sulfoxide on cell growth and ultrastructural features of cultured smooth muscle cells. J Electron Micros 33:239-241, 1984.87. Carpenter JF, Brockbank KGM. Process for preparing tissue for transplantation. US Patent #5,160,313: 1992.88. Taylor MJ. Clinical cryobiology of tissues: pres-ervation of corneas. Cryobiol 23:323-353, 1986.89. Braendle L, Linthurst A, Burkart M, Brock-bank KGM, Wolfi nbarger L. A comparative study of cryoprotectant removal methods, Presented at the 35th Annual Meeting of the Society for Cryobiology, 1998.90. Rabin Y, Steif PF. Solid mechanics aspects of cryobiology. In: Advances in Biopreservation, Edited by Baust JG, CRC Press, chapter 8; 359-382, 2007.



FIGURE 1. The principle of tide technology.

Innovative Disposable BioreactorsFor Membrane Protein Production

Based on the Tide Principle

Lewis Ho, PhD ([email protected]), is a partner with IDL Bioservices, LLC, Lawrenceville, GA, USA. This article is based on a presentation given at The Williamsburg BioProcessing Foundation’s fi rst international Single-Use BioProcessing Components & Systems meeting held in Concord, California, July 16-18, 2007.

Membrane proteins such as hERG (human Ether-a-go-go Related Gene) and GPCRs (G-protein-

coupled receptors) have been widely used as favorite targets for discovery of therapeutic drugs to treat cardiac arrhythmia, diabetes, epilepsy, cancer, glaucoma and many other indications. They are also widely used in cell-based assays to test new pharmaceuticals for safety in the early stages of drug discovery.

Therefore, the demand for mem-brane proteins has signifi cantly increased in recent years in the pharma-ceutical and biopharmaceutical indus-tries. As a result, cell culture technology is playing more of an essential role in the production of these recombi-nant proteins. Commercially, most of

these proteins are expressed in CHO (Chinese hamster ovary) and HEK293 cell lines. These cell lines are commonly grown as adherent cells for membrane proteins and are scaled up in roller bottles or microcarrier bioreactors for production. The roller bottle system is labor-intensive to use, and microcarrier bioreactor setups also demonstrate great limitations in effi ciency and scale-up capabilities. In recent years, because of lower capital requirements, faster turn-around times, and easier-to-meet regu-latory requirements for validation, etc., the use of disposable bioreactors has gained signifi cant interest and attention in the industry. Among those, the Wave Bioreactor™ is the most well-known in the industry. However, the Wave Bioreactor is not suited for adherent cell culturing. It cannot provide suffi cient steady surface area for cell attachment.

In this study, we have found that the innovative new bioreactors developed by Cesco Bioengineering work very well

for this purpose. Oxygen transfer limitations and

shear stress sensitivity are two major factors which make bioreactor scale-up diffi cult. In addition, the durability of disposable bioreactor materials further limit production scales. Single-use bio-reactors based on the tide principle have been found to eliminate these obstacles, making scale-up signifi cantly easier than with conventional bioreactors.

Cells are adsorbed or entrapped in the matrix and are constantly mixed and nourished with necessary oxygen and nutrients through the dynamic tide fl ow motion that is created. No further aeration and agitation is required and thus, less shear stress occurs. Above all, the bioreactor’s tide fl ow pattern permits scale-up directly to any size because its mass transfer is only mod-estly dependent on scaling factors. The immobilized cells in the matrix allow for: 1) cell-free exchange or harvest of culture medium; and 2) the continued

By LEWIS HO


replacement of commercially-available, large-volume disposable bags. Thus, the total bioreactor volume can be consid-ered unlimited.

Because of the matrix’s large surface area and ease of operation, this biore-actor can produce extremely high cell densities in the production of viruses, proteins, etc. Greater levels of produc-tivity are achieved when compared to conventional bioreactors with microcar-riers (e.g., T-fl asks and roller bottles). In this study, two disposable BelloCell® and TideCell® bioreactor systems were used to investigate the production of the most common membrane proteins such as hERG and GPCR.

Tide Principle Unlike most bioreactors, this novel

bioreactor utilizes the thin liquid fi lm to transfer oxygen to the cells directly from open gas phase instead of bulk liquid. Similar to an ocean’s ebb and fl ow, slow up and down motions are created to transport medium through a stationary matrix where cells reside. As illustrated in Figure 1, the matrix is submerged by the medium and nutrient transfer is facilitated when the tide comes in. As the tide goes out, the matrix emerges and is exposed to the air where oxygen transfer takes place. This is referred to in the industry as the “tide principle.”

The signifi cance of this unique methodology is as follows: 1) extremely high oxygen transfers result from the direct exposure to the air; 2) it is virtu-ally independent of scale factoring with oxygen transfer because bulk mixing of liquid medium to determine its mass-transfer effi ciency is unnecessary; and 3) shear stress is very low due to the high oxygen transfer capacity. All of these attributes are ideal for cell culturing and simplicity of scale-up.

Carriers

As mentioned above, the basic prin-ciple behind this bioreactor methodol-ogy lies in the immobilization of cells in a stationery carrier while allowing the cell-free medium to submerge and

expose the carrier-dependent cells in cycles. Therefore, the carrier plays an essential role. It is preferable to have a large surface area dedicated to the effi cient attachment of adherent or semi-suspension cells.

The FibraCel® disk (New Brunswick Scientifi c) made of nonwoven poly-ester fi ber and polypropylene backing has a surface area of 1,200 cm2/g. It has been widely used in packed bed bioreactors such as their CelliGen® Plus. In Figure 2, two carriers are are shown, both developed by Cesco Bioengineering for separate applications.

Bioreactors Based on the Tide Principle

Cesco Bioengineering has developed two types of disposable bioreactors for different scales of operation based on the tide principle and the previ-ously-mentioned carriers: BelloCell for laboratory scale, and TideCell for pilot and production scale. Figure 3 shows a BelloCell-500 bioreactor system consisting of two components: 1) a sterile single-use culture bottle; and 2) a BelloStage® bellow compressor. The BelloCell-500 is a 500 ml clear poly-ethylene (PE) bottle with a stationary

FIGURE 2. Bionoc II and Bionoc D carriers. Bionoc II (left): Made of 100% polyethylene terephthalate nonwoven fabric with treated surface; has high equivalent surface area (2,000 cm2/g carrier; is rigid and inert; and requires enzymatic treatment for cell removal. Bionoc D (right): Made of alginate base and soluble without requirement of enzymatic treatment for cell removal; has high equivalent surface area (1,500 cm2/g carrier); and requires serum in the medium.

FIGURE 3. The BelloCell-500 system.


100 ml matrix loaded with carriers in the upper chamber and a fl exible 500 ml bellow container in the lower chamber. The 500 ml of medium is raised and lowered alternatively by the BelloStage to submerge and expose the carriers. This action creates a dynamic interface between air and medium on cell surfaces to maximize nutrient uptake, waste removal, and oxygen transfer.

The BelloCell-500P bioreactor system (Figure 4) is similar to the BelloCell-500 system. However, it has one-inlet tubing on the top of central tube and one-outlet tubing extend-ing through the central tube to the bottom of matrix. The inlet tubing is connected to a 2 or 4 L medium bottle through a peristaltic circulation pump, and timer to control the fl ow rate (the BelloFeeder® system).

With the BelloCell-500P, medium from this separate large-volume res-ervoir bottle continuously exchanges the medium inside the BelloCell bottle. Normally, just one reservoir of medium is suffi cient for the entire run, but if an extended culture of secreted protein is desired, this system can be set up to simply replace the reservoir with a fresh one as needed.

In Figure 5, the schematic diagram and virtual system of a pilot plant production system is shown. The system consists of a TideCell reactor enclosed in a chamber with constant temperature and CO2 control, and a shaker in a chamber with constant temperature, CO2 and pH control. Reactor sizes can range from 1.0–2.5 L with a carrier volume of 80%. The shaker bag’s volume can range from 20–50 L.

Figure 6 shows a TideCell 20/50 virtual pilot plant unit. The reactor volume can be ranged up to 100 L and the medium bag can be up to 2,000 L (as shown in Figure 7). While the volume of the matrix vessel can be greater than 100 L, the largest dis-posable bag commercially available is currently 1,000 L. However, fresh 1,000 L bags of nutrient medium can be easily interchanged to replace the nutrient-depleted ones.

Production of Membrane Proteins

Genetically-engineered HEK293 cell cultures were used for expressing hERG and GPCR, respectively.

Early studies were conducted pri-marily with lab-scale BelloCell-500 and

500P systems. Subsequent studies were performed with the recently-developed pilot production TideCell system. The roller bottle R-850 system was used as control to compare the performance of each system studied. For the materials and methods used, refer to Ho et al.1,2

FIGURE 4. The BelloCell-500P system.

FIGURE 5. A schematic diagram of the TideCell pilot plant system.


hERGCells reached >80% confl uency on

the T-fl ask surfaces in less than 72 hrs, with an average of 1.35 x 106 cells/ml-1 (Ho et al.2). Some of the cells were then inoculated to 850 cm2 roller bottles (200 ml of medium) at cell density of 2.5 x 105 cells/ml-1. After 80 hrs, the

cells on the roller bottle reached >80% confl uency and were harvested with an average cell density of 1.315 x 106 cells/ml-1 or 2.63 x 108 total cells per bottle (Table 1). The cell viability was 95%. The average specifi c hERG expression (Bmax) was 2.49 pmole of [3H]astemizole that was bound per

total membrane protein, or a total hERG protein expression of 55 pmole of [3H]astemizole that was bound per roller bottle.

The BelloCell-500 was used for cell culturing (Ho et al.1,2). The tem-perature was controlled at 37˚C and CO2 controlled initially at 5%. It was adjusted later as needed to maintain a 6.8–7.4 pH. The bottle was fi lled with 500 ml medium and inoculated with 4 x 105 cells/ml-1. The top holding time (THT) was used to hold the bellow up on the top and allow the entire matrix to submerge in the medium. By using the bottom holding time (BHT), the bellow was then drawn down on the bottom, allowing the entire matrix exposure to the air.

After inoculation, the BelloStage controlled the up/down speed at 2.0 mm/sec-1 with THT of 20 sec for the fi rst 2–5 hrs, to assure cell attachment to the matrices. Then up/down speed was reduced to 1.5 mm/sec-1, THT changed to 0 sec, and BHT adjusted to 50 min, as found to be optimal for expression. During the entire run, the substrate and metabolite concentrations (includ-ing glucose, glutamine, ammonia and lactate) were monitored once a day. The medium was replaced to maintain a glu-cose level above 1g/l-1 and/or lactic acid or ammonia below 3 g/l-1. Disk samples were taken from the bottle periodically for cell density measurements by a CVD (crystal violet dye) nucleus staining method.

As the cell density reached a desired minimum level (>3 x 106 cells/ml-1 or total cell count of >1.5 x 109) or the glucose uptake rate (GUR) reached plateau, the run was terminated and the bottle removed for processing to release the cells from the matrices.

To simplify further the medium exchange process, a new BelloCell-500P bioreactor was also used to study the production of the same hERG protein. Approximately two days after inocula-tion, when the glucose concentration in BelloCell-500P dropped to 1-1.5 g/l-1, the recirculation started at a rate of 60 ml/hr. As GUR reached a plateau, indi-cating that the cell density had reached the desired level (>3 x 106 cells/ml-1 or

FIGURE 7. A schematic diagram of the TideCell production system.

FIGURE 6. The TideCell 20/50 system.


total cell count of >1.5 x 109), the run was terminated and the bottle removed for processing to release the cells from the matrices.

To scale up for the system for large-scale of production, a new TideCell 20/50 system was used for the study. The matrix vessel was fi lled with 1,000 ml medium and inoculated with 4 x 105 cell/ml-1 or a total cell count of 2 x 109, obtained from one BellCell-500 or 500P. The same operating procedure (as with the 500P) was followed, except that one extra 20 L bag of fresh medium was added to supplement the depleted glucose during the run. A 50 L bag can

also be used as a direct scale-up with the BelloCell-500P run.

Figures 8–10 depict typical GUR and metabolite concentration profi les including pH, glucose, ammonium and lactate for the eight-day semi-batch cultures using a BHT of 50 min. Similar profi les for the nine-day culture using a batch/recirculation process with a BHT of 50 min are shown in Figures 11–13. Table 1 summarizes the results of all experiments. Results show that:

1) High cell density can be achieved in both BelloCell and TideCell systems. A cell density difference of 5.65–6.57 versus 1.315 x 106 cells/ml was shown

between the BelloCell/TideCell systems and the roller bottle system.

2) Cell counts of 2.68–3.06 x 109 in BelloCell-500 and 500P, and 3.06 x 1010 in TideCell 20/50 was obtained in one unit as compared to 2.6 x 108 in R-850 roller bottles. In other words: one BelloCell can be equivalent to 10–12 roller bottles, and one TideCell 20/50 has the equivalency of 120 R-850 roller bottles, in terms of cell mass.

3) The medium utilization effi ciency for the hERG protein production was 1.78–2.25-fold better in the said systems when compared to the roller bottle system.

TABLE 1. Summary of experiments on hERG membrane protein production.


4) In terms of pmole [3H]astemizole bound/mg protein, the specifi c expres-sion was 64–85% better in the said systems than the roller bottle system.

5) In terms of total hERG pro-tein production, one BelloCell can be equivalent to 17–21 R-850 roller bottles, or one TideCell 20/50 is equivalent to 212 of them.

6) In general, both the BelloCell-500P and TideCell 20/50 are equivalent on cell growth and protein production. The scale-up of BelloCell to TideCell should be straightforward, just as the results have indicated.

GPCRThe cells reached ~80% confl uency

on the surface of T-fl asks in 72 hrs (Ho et al.1). The cells were then digested with 0.05% trypsin and an average cell density of 1.05 x 106 cells/ml-1 was obtained using GPCR production medium. The cells were then inoculated to three 850 cm2 roller bottles (200 ml medium) at a density of 2.5 x 105 cells/ml-1. After 48 hrs, a total medium replacement was conducted because the medium had run out of glucose.

In 80 hrs, the cells on roller bottles reached ~80% confl uency and cells were harvested at an average cell density of 6.1 x 105 cells/ml-1. The cell viability was 94%. Table 2 shows the specifi c receptor X binding activity Bmax of 2.3 pmol/mg-1 of total membrane pro-tein, or a total binding activity of 22.5 pmoles for GPCR-X production per bottle. The specifi c receptor Y binding activity Bmax of 1.97 pmol/mg-1 of total membrane protein, or a total binding activity of 20.76 pmoles for GPCR-Y production per bottle, is illustrated in Table 3.

As described by Ho et al.1 as well as with hERG studies shown above, BelloCell-500, 500P and TideCell bio-reactor systems were used for produc-tion of GPCR-X and Y. The operating procedures were identical except that the medium contained no additional glucose and used a regular BHT of 1 min. All metabolite profi les were similar to those shown above for hERG production.

Tables 2 and 3 summarize the exper-

FIGURE 8. Typical glucose and GUR profi les of semi-batch culture (BHT = 50 min).

FIGURE 9. Typical lactate and pH profi les of semi-batch culture (BHT = 50 min).

FIGURE 10. Typical glutamine and ammonium profi les of semi-batch culture (BHT = 50 min).


imental results of GPCR production using BelloCell and TideCell systems in comparison to the roller bottle system. Results show that:

1) Relatively high cell densities can be achieved in both BelloCell and TideCell systems. The cell density dif-ferences of 2.8 for BelloCell and 13.98 for TideCell versus 0.6 x 106 cells/ml for roller bottle are shown. Cell counts of 1.36–1.4 x 109 in the BelloCell-500 and 500P, and 1.39 x 1010 in the TideCell 20/50 was obtained in one unit, as com-pared to ~1.2 x 108 in the R-850 roller bottles. In other words: one BelloCell can be equivalent to ~12 roller bottles; or one TideCell 20/50 is equivalent to ~120 R-850 roller bottles, when it comes to cell mass. In general, the cell density was lower in GPCR than that in the hERG system due to a low glucose content in the medium.

2) The nutrient requirements for a unit of cellular growth should be about the same. Therefore, the cellular effi ciency, as defi ned by the number of cells per amount of medium utilized, was about the same for all systems. However, there may be signifi cant dif-ferences in gene expression and produc-tion effi ciency because of the specifi c operating conditions for the bioreactor system utilized and its applied control scheme.

3) The medium utilization effi ciency for GPCR protein production was 1.52–2.13-fold better in the TideCell and Bellocell systems than the roller bottle system.

4) The specifi c expression was 39–61% better in the TideCell and BelloCell systems when compared to the roller bottle system, in terms of pmole radioligand X and Y bound/mg total membrane protein.

5) Regarding total GPCR protein production, one BelloCell can be equivalent to 16–20 roller bottles and one TideCell 20/50 run is equivalent to 156 R-850 roller bottles.

6) In general, the BelloCell-500P and TideCell 20/50 are equivalent for cell growth and protein production. The scale-up of BelloCell to TideCell should be straightforward, just as the results indicate.

FIGURE 11. Typical glucose and GUR profi les of the batch/recirculation culture (BHT = 50 min).

FIGURE 12. Typical lactate and pH profi les of the batch/recirculation culture (BHT = 50 min).

FIGURE 13. Typical glutamine and ammonium profi les of the batch/recirculation culture (BHT = 50 min).


Summary

Based on the tide principle, a disposable BelloCell bioreactor for lab-scale, and a TideCell bioreactor for large-scale of cell cultures, werepresented — both developed by Cesco Bioengineering Co., Ltd. The two most common membrane

proteins (hERG and GPCR) were produced in these new bioreactor systems and the roller bottle system. Then the results were compared.

Results have demonstrated that these exceptional systems can be much more effi cient than the

commonly-used roller bottle system in the production of mem-brane proteins. The scalability of these bioreactors for production of hERG and GPCR proteins has also been successfully demonstrated.

TABLE 2. Summary of experiments on GPCR-X membrane protein production.

TABLE 3. Summary of experiments on GPCR-Y membrane protein production.


ACKNOWLEDGEMENT

The author wishes to thank King-Ming Chang of Cesco Bioengineering for valuable comments and technical support.

REFERENCES

1. Ho L, Greene C, Schmidt A, Huang L. Cultivation of HEK293 cell line and production of a member of the superfamily of G-protein coupled receptors for drug discovery applications using a highly effi cient novel bioreactor. Cytotech, 2004, 45: 117-123. 2. Ho L, Ou JJ, Shaw SY. A simple control scheme applied in a new bioreactor system for better expression of hERG membrane protein. J of Biotech, (2008, currently under review).

NOTES

TideCell®, BelloCell®, BelloStage®, BelloFeeder®, and BioNOC® are registered trademarks of Cesco Bioengineering Co., Ltd, Taiwan, R.O.C.

Wave BioreactorTM is a trademark of GE Healthcare Wave Products Group.

FibraCel® and CelliGen® are registered trademarks of New Brunswick Scientifi c Co., Inc.

The Most Trusted Source of BioProcess Technology ®

Phone: 757.423.8823 • Fax: 757.423.2065 • www.wilbio.com • E-mail: [email protected]

• Proprietary Technology and Patents • Collaborations• Contract Services • Biosimilar Products • Harmonization of Regulatory

Guidelines• Scale-Up• Media and Assay Development

• Feed and Harvest Strategies • Purifi cation • Applications• Antibodies• Recombinant Proteins • Viral Vaccines • Viral Gene Vectors • Cellular Therapies

Topics:

5th International Meeting • October 6-8, 2008 • Singapore

BBiiooPProcessrocessTTechnechnoologylogy--AsAsiiaa

Sponsored By:

CALLCALL FORFOR PAPERSPAPERSPPlease send your conference presentation abstractslease send your conference presentation abstracts fforor

our upcoming meetings toour upcoming meetings to [email protected]@wilbio.comFind out more at www.wilbio.com or call 757-423-8823Find out more at www.wilbio.com or call 757-423-8823

WilBio meetings provide a thorough coverage of process development and production issues while focusing on narrowly defi ned applications

The Most Trusted Source of BioProcess

Technology®

27

We want to help you attend our upcoming

conferences by Reducing Your Travel CostsSM.

Register for one of our meetings by the Travel

Savings Dates shown below and we’ll pay $50

per night toward the cost of your room at the

conference hotel (maximum of four nights).

For more information on each meeting, go to our

website homepage (www.wilbio.com). You can then

register online, or by printing and faxing the form

provided with the conference information (PDF fi le).

Be sure to include promo code TSD801 on the form

or in your online registration.

We look forward to seeing you at our conferences

and making your travel less expensive.


When & Where Conferences Features Travel Savings Date

June 30 - July 2, 2008Amsterdam,

The NetherlandsBioProcess Technology — Europe

Featuring a tour of theCrucell Manufacturing Facility

June 15, 2008

July 28-30, 2008San Francisco, CA

Single-UseBioProcess

Components and Systems

RawMaterials

and Contract Services

Concurrent MeetingsFeaturing a tour of the

Baxter Healthcare Single-Use Contract

Manufacturing Facility

June 15, 2008

September 8-10, 2008Philadelphia, PA

Cell Culture Scale-Up

Purifi cation of Biological

Products

Concurrent MeetingsFeaturing a tour of the

Merck Biologics Pilot PlantJune 15, 2008

October 6-8, 2008Singapore BioProcess Technology — Asia

Featuring a tour of the Lonza Contract

Manufacturing FacilityJune 15, 2008

October 20-22, 2008Houston, TX

Cell and Tissue BioProcessing

Viral Vectors and Vaccines

Concurrent MeetingsFeaturing tours of

Baylor College of Medicine and Introgen Therapeutics

June 30, 2008

November 3-5, 2008Santa Barbara, CA Filtration of Biological Products Expanded Content July 14, 2008

December 2-4, 2008Santa Barbara, CA

CellBanking

CellEngineering

Concurrent Meetings August 12, 2008

December 8-10, 2008Baltimore, MD

Characterization andComparability of Complex Biologics

Focus on Cellular and Viral Products

September 14, 2008

Travel ToTravel To OurOurConferenceConference

Any WaAny Way Youy YouChooseChoose

aand Knd KnownowYou’llYou’ll SaveSave

MoneyMoney When When You SnoozeYou Snooze!!

Register Now!Register Now!

www.wilbio.comwww.wilbio.com757577.423.8823.423.8823

ReducingYourTravelCosts_MB.indd 1 5/30/2008 12:27:53 PM



Dendritic Cell Vaccine Production Facility: From Design to Operation

By NICOLAS TAQUET* et al.

Nicolas Taquet, MS ([email protected]) is the technical director at the Baylor Institute for Immunology Research in Dallas, Texas. In 2004, he became the project manager to oversee the design, construction, and qualifi cation of an entire large-scale vaccine manufacturing facility for Phase II/III clinical trials. This article has been co-written by: Lee K. Roberts, PhD;‡ Susan Burkeholder;† Patricia Phipps;† Jennifer Finholt;† Lynnette Walters;† A. Carson Harrod, PhD;† Jacques Banchereau, PhD;† and A. Karolina Palucka, MD, PhD.† *Mr. Taquet is the corresponding author. This article is based on a presentation given at The Williamsburg BioProcessing Foundation’s 12th international Cell & Tissue BioProcessing meeting held in Austin, Texas, October 29-31, 2007.

†Baylor Institute for Immunology Research, Dallas, Texas. ‡ODC Therapy, Inc., Dallas, Texas.

Melanoma is a com-mon and deadly form of skin cancer. The American Cancer Society has estimated

that there will be 62,480 new cases and 8,420 deaths from melanoma in the United States in 2008.1

Since 1999, the Baylor Institute for Immunology Research (BIIR) in Dallas, Texas has carried out six melanoma vaccine clinical trials. These trials are conducted under Investigational New Drug Applications (INDs) fi led with the FDA and are paving the way toward the development of a potential cell-based vaccine therapy for melanoma patients.

Development of a Frozen Vaccine Against Melanoma Skin Cancer

Patients with advanced melanoma cannot eliminate their cancer cells, oftentimes due to their immune systems becoming tolerant to the cancer cells. Our strategy is to edu-cate a patient’s immune system to direct cytotoxic T-cells to eliminate the melanoma cells. To accomplish this, a patient’s own dendritic cell progenitors (or precursors) are

placed in cell culture. Differentiated dendritic cells are a type of white blood cell that induces and regulates immune responses (Figure 1). They are loaded with melanoma antigens (proteins unique to the cancer cells) for injection as a vaccine back into the patient. A patient receives several injections of the antigen-loaded dendritic cell vaccine.

Our clinical results have shown that this vaccine therapy is safe and well-tolerated by the patients, and can lead

to both elimination of the cancer and long-term survival in some patients.The ability to provide a patient- specifi c cancer vaccine has taken years to develop. Initially, each vaccine was manufactured fresh, which took up to nine days of preparation followed by lengthy release testing. This approach was very expensive and would have required numerous cell processing centers located in close proximityto the patient’s clinic in order to provide a fresh, effi cacious product. We have decreased our manufactur-ing time by developing new processes for the manufacture of dendritic cell vaccines.

Today, we manufacture these vaccines in three days. This allows us to produce more vaccines using the same facility and number of personnel. We have also developed a frozen dendritic cell vaccine, which enables us to ship it anywhere in the United States and possibly elsewhere in the world to the patient’s local physician.

The frozen vaccine process also allows us to manufacture and release only one batch of vaccine needed for multiple injections into the patient. This has simplifi ed and streamlined the process as well as considerably reduced the production costs.

FIGURE 1. Monocyte-derived dendritic cell (white blood cell that recognizes and then binds foreign substances) used to provide a patient-specifi c cancer treatment.


Dendritic Cell Vaccine Production Facility Design

In order to move product develop-ment forward, the majority of biotech companies and academic institutions involved in cell-based therapies need new facilities in order to scale up pro-duction capabilities and comply with evolving regulatory requirements.

Some institutions choose to use a contract manufacturing organization (CMO) to benefi t from established expertise while others support their clinical development programs with their own dedicated production facility. The main challenges in establishing a dedicated pilot-scale production facility are described hereafter.

BIIR has used a dedicated Class 10,000 (ISO 7) cleanroom to manufac-ture the vaccines for clinical trials. It takes a minimum of two years for the design, fundraising, construction, and commissioning of a new production facility. Because of this, it was quite diffi cult to anticipate all of the future needs and the probable changes to the manufacturing process, regulatory requirements, resources, and future activity. This was like trying to solve a multi-variable equation in math with numerous unknowns.

In 2004, BIIR designed and built, with the help of Holtz BioPharma Consulting and K-Tec Cleanroom Systems, a three-cleanroom production facility able to accommodate multiple Phase II clinical trials or one large-scale Phase II/III trial (see Figures 2 and 3). The design was driven by select-ing equipment (Scientifi c Southwest Resources, Inc.) that would accommo-date most of the existing and antici-pated dendritic cell vaccine processes and was fl exible enough to house future changes in the required equipment.

After considering the likely vac-cine production processes that rely on short-term cell culture, the decision was made to incorporate the cell culture incubators within the cleanrooms of the production suite. In contrast to this strategy, some CMOs recommend placing the cell culture incubators in a centralized location. This helps

them to minimize overloading of their production cleanrooms with various cell incubations while segregating the potential contamination risk away from the cell manipulation areas. Instead, we have integrated enough incubators in each of the cleanrooms to smoothly accommodate a seven-day manufactur-ing schedule with cleaning and mainte-nance included.

Because we use a closed cell culture system for our dendritic cell vaccine processes, cell culture media evapora-tion during the short-term cell incuba-tion is not an issue. The culture bags are permeable to CO2 but are waterproof. Our Sanyo incubators are equipped with CO2 infrared sensors so that the incubator chambers do not require any ambient humidity from a water pan,

FIGURE 2. View of cleanroom #1 in the dendritic cell vaccine production facility at Baylor Institute for Immunology Research.

FIGURE 3. View of cleanroom #2.


thus minimizing the contamination risk. However, the incubator chambers still require extensive cleaning between patient vaccine batches even though we use a closed system. Paradoxically, the cleaning process, which involves a thor-ough wiping with sanitizing wet wipes, will catch most of the free particles but can displace particles from one site to another. This can cause both nonviable and viable particle counts to peak. As a result, locating the incubators and cen-trifuges next to the cleanroom exhaust grilles has helped control the environ-ment by extracting particles, aerosols, humidity or any hazardous gases from the cleanrooms (see Figure 4).

The single 6´ biosafety cabinet in each of the production suite’s clean-rooms is located as far as possible from the exhaust grilles to work with the room’s unidirectional airfl ow. Our manufacturing process has a one-way fl ow from the introduction of the patient’s apheresis blood into the cleanroom to the fi lling and freezing of the vaccine vials. The manufactur-ing personnel also have a one-way fl ow through the cleanrooms with separate entrance and exit doors. While the raw materials enter the cleanrooms via a pass-through window connected to other areas within the production suite, the biohazard wastes (mainly packaging materials and cell media within sealed bags) are moved out of the cleanrooms through the exit door (see Figure 5).

The two cleanrooms currently used for dendritic cell vaccine manufac-ture are mirrors of each other, with a common air exhaust within the middle wall. They provide redundancy to each other by having the exact same produc-tion equipment. The exhaust grilles were staggered to prevent any potential cross-contamination between the two cleanrooms. The fi rst merging of air within the cleanroom exhausts occurs further up the wall in the plenum. A single-pass air fl ow allows simultaneous processes in each cleanroom without any potential cross-contamination. The room pressure, temperature, and humidity are monitored by two inde-pendent alarm systems (one monitors and controls the environment and the

second serves only as a monitor). Easy access for inspecting the air supply lines and pipes was included in the design.

In order to maximize the quality of the vaccines, every component involved in the manufacturing process is dispos-able and is outsourced to specialized vendors. The reagents and ancillary materials come triple-packaged, steril-ized, and certifi ed. Our quality assur-ance unit inspects each received lot of production materials, reviews and verifi es the accompanying certifi cates, and releases them for use in vaccine manufacturing prior to their entering the production suite. Using disposable versus reusable materials helps the qual-ifi cation of raw materials and increases the productivity of the manufactur-ing personnel by minimizing cleaning required between vaccine batches.

Cleanrooms and Support Area

All of the equipment and materials were carefully selected to provide ease of cleaning, durability, and adequate performance. The critical production facility is a sealed box made of modular

walls. The walls were assembled within BIIR in a period of three weeks, and an epoxy fl oor with 4˝ coves was installed. Contracting with K-Tec Cleanroom Systems enabled us to prepare the shell at the site while the modular walls were delivered and stored at their warehouse. This saved time on the overall con-struction schedule and it resulted in a self-contained facility isolated from the surrounding elements of the building.

Two redundant 70-ton direct-expansion (DX) units supply 7,500 cfm (60 air changes per hour of 100% outside air in a single pass). One unit is at rest while the other one is in use. They are alternated to make sure each unit has the same the amount of hours on it, to change the intake pre- fi lters and to clean the coils. The DX unit dehumidifi es, and either cools or heats the outside air prior to its introduc-tion into the production suite. Purifi ed steam from two redundant humidifi ers is added to maintain a humidity level of around 45% within the production suite (see Figure 6). When an abnormal temperature, humidity, or pressure differential is detected, the second DX

FIGURE 4. View of cleanroom #2. Cleanroom #1 is visible through the window. Locating the incubators or a centrifuge next to the cleanroom exhaust grilles helps control the environment by extracting particles, aerosols, humidity, or any hazardous gases.


FIGURE 5 (above). Schematic diagram of the facility. The manufacturing process has a one-way fl ow from the apheresis to the frozen doses of vaccine. The personnel also have a one-way fl ow with exits distinct from entrances. While the raw materials enter the production suites via a pass-through window, the biohazard wastes move out through the exit door.

FIGURE 6 (right). Top view of mechanical support. A common supply duct enters the facility via an underground crawl space. The exhaust duct (against the building) leads to two exhaust fans. Redundant purifi ed steam generators (which humidify the air in the facility) are visible at the top of the picture.


unit starts and, in about six minutes, reaches the proper level of cooling/heat-ing. Automated dampers, together with the fan blower variable frequency drive (VFD), allow ramping up of the second unit as the fi rst unit ceases operation, thus keeping a consistent airfl ow above 7,500 cfm that minimizes the impact of the changeover on the production suite.

The building has an electrical feed from two different transformers and an automated switch. If both transformers from the power plant happen to fail, power is provided by an emergency generator that covers the entire produc-tion suite and its support equipment, such as the DX units, exhaust fans, humidifi ers, water purifi cation system, and medical air production system. A centralized uninterrupted power supply (UPS) covers the entire production suite by buffering the electrical changes so that the computer-based equipment does not reboot. By itself, this UPS can support the entire production suite in full production for 2.5 hours if the emergency generator fails, giving us enough time to shut down the opera-tion if necessary.

The dendritic cell vaccine process requires liquid nitrogen for freezing and storage of each patient’s vaccine batch in nitrogen vapor. Two outside bulk liquid nitrogen tanks (a 3,000 gallon and a 400 gallon backup) are fi lled every three weeks at night by a local liquid nitrogen distributor. A 750 pound outside tank and eight backup cylin-ders provide CO2 for the cell culture incubators.

Challenge 1: Implementation of Evolving Regulatory Requirements

The regulatory requirements for an early Phase I/II clinical trial are differ-ent from those for a Phase II/III trial. The early phase requirements include progressive implementation of current Good Manufacturing Practices (cGMP) and the determination of the optimal product. Total cGMP compliance and a fully-characterized product are expected for licensure in later phases of clinical and product development.

The speed of the cGMP implemen-

tation is critical. At the beginning of product development, it is important not to exhaust the limited resources by overloading them with too many operational changes. However, imple-menting those requirements too slowly may impact the predefi ned regulatory timeline.

The known requirements for most therapeutic products are listed in the current FDA Guidelines. While it may take a minimum of two years to get a new vaccine production facility con-structed and operational, new regula-tory requirements may arise in the meantime that can delay or signifi cantly affect progression to commercialization of the product. Therefore, the facility design and construction project have to stay fl exible to accommodate and record any changes. The justifi cation and the review of any changes need to be tracked and well documented. Additionally, each FDA reviewer may have his or her own specifi c comments or questions about a manufacturing process or the production facility, which could prompt a modifi cation. Over time, new reviewers may ask novel questions that could affect the process or operations. It is a learning exchange where the biotech or academic institu-tion will mature its process and manu-facturing competency, and the FDA will observe and regulate new emerging science and technology.

Challenge 2: Anticipation of Future Activity and Needed Resources

Starting one or several clinical trials takes time. The development of the process, preparations for the IND, the initiation of the clinical sites, and the accrual of patients require consider-able effort by several individuals. As it can take six to eight months to fully train and qualify our key manufactur-ing personnel, this coordination can be very challenging. The recruitment of appropriate personnel can also be diffi cult in a small market. Vacations or illness of strategic personnel can also have a dramatic impact on the manufacturing schedule. A projection of growth can be diffi cult in the early

phases of product development. While the estimated yearly number of patients per disease is known, the potential number of patients that can be accrued at a given clinical location with com-peting clinical trials may be hard to determine. This aspect of early phase clinical trials is usually unpredictable. A wave of patients may be enrolled into a trial, thus requiring several shifts of work. Conversely, holidays, vacations, and illness can disrupt the manufactur-ing schedule by limiting the number of production personnel available at any given time. Our annual maintenance is scheduled during slow periods, but patient enrollment does not always coincide with these schedules.

Unexpected, less-than-satisfactory environmental monitoring results may quarantine some equipment and impact the schedule. Or if a vendor announces a recall, discontinues a critical material, has a backorder situation, or some reagents require additional testing, the schedule for manufacturing will be altered. Dealing with simultaneous clinical trials, each with different cell incubation times, will give scheduling coordinators headaches. Planning for a worst-case scheduling scenario is a nec-essary exercise to balance production capacity against the need for supplying vaccines to the clinical trials. The cost of running a full-capacity facility can only be justifi ed if the activities have been optimized.

Challenge 3: Manufacturing versus Operational Costs

Once a new facility is activated, there will always be certain fi xed operational necessary costs. At our facility, these include: utilities, equip-ment maintenance and certifi cation, sterile garment leasing, subcontracted sanitization, environmental monitoring, liquid nitrogen for vaccine storage, and CO2 for our 24 incubators. During the transition phase from lab-scale to a fully GMP-qualifi ed facility, these operating costs can be quite high. Operating expenses necessary for keeping the current facility operational are added to those that are essential for the new


facility’s qualifi cation period. It should also be remembered

that these fi xed operational costs are incurred regardless of the utilization of the facility. Thus, personnel expenses must be factored in, especially for the core group of highly skilled technicians. Reducing the overall cost of operations is obviously linked to maintaining full production capacity. This can also be achieved by activating individual cleanrooms within the production suite only when needed and by optimizing the use of space towards one centralized location.

In our case, the manufacturing cost-per-patient vaccine batch varies with the process and the activity. Slow periods of patient accrual require smaller lots of high-priced raw materials, which may make it diffi cult to justify, considering the high overall costs of operation. On the other hand, high-production periods will require a larger inventory of stock materials. It helps in lowering expenses to order in bulk, although these materials must be purchased up front and then used as needed. While the initial expenditure will eventually average out over time, it is sometimes diffi cult to justify a large cash outlay. Anticipating stock quantities will remain an enduring challenge given the ambiguity of predicting produc-tion needs in early phases of clinical development.

The goal is to implement an average activity for our production facility with a consistent patient schedule to best control critical variables and expenses (e.g., garment inventory, sanitization, and environmental monitoring fre-quencies). By achieving this goal, it is possible to maintain a consistent opera-tional cost and lower cost-per-vaccine batches.

To date, we have found our facility design to be appropriate to our needs. However, in 2008 as we move from supporting two clinical trials to initiat-ing fi ve separate clinical trials including a randomized trial in melanoma in May and a Phase II trial in HIV in October, we will have the true test of the full-scale performance, qualifi cation and capacity of our facility.

ACKNOWLEDGEMENTS

The author wishes to thank C. Samuelsen; B. Holtz, PhD (Holtz BioPharma Consulting); S. Ivie (Scientifi c Resources Southwest, Inc.); and M. Wells (K-Tec Cleanroom Systems) for their help. Baylor Health Care System (www.biir.org) funded and also operates this new facility.

REFERENCES

1. <http://www.cancer.org/docroot/CRI/content/CRI_2_2_1X_How_many_people_get_melanoma_skin_cancer_50.asp?rnav=cri>

SUPPLIER SIDE


� PART ONE IN A TWO-PART SERIES �

How to Construct a Monoclonal Antibody Factory:A Comparison of Production Costs in Fed Batch

and Perfusion Culture with Microcarriers

Björn Lundgren ([email protected]) is marketing manager1; Peter Bösch is a chemistry faculty member2; and Christian Kaisermayer, PhD, is the project manager of cell culture applications and support1. *Mr. Lundgren is the corresponding author. This article is based on a presentation given at The Williamsburg BioProcessing Foundation’s second international Cell Culture Scale-Up meeting held in Waltham, Massachusetts, September 24-26, 2007.

1. GE Healthcare, Uppsala, Sweden. 2. The Vienna University of Technology, Vienna, Austria.

By PETER BÖSCH, BJÖRN LUNDGREN*, and CHRISTIAN KAISERMAYER

In the fi rst part of this series, we will summarize this technology and provide background information. Part Two will discuss how

to construct a plant, review the fi nancial aspects of fed batch ver-sus perfusion culture, and provide conclusions—to be published in the Summer 2008 issue of the BioProcessing Journal.

Introduction

A basic engineering study has been performed to evaluate three different strategies for the production of mono-clonal antibodies (MAbs) from Chinese hamster ovary (CHO) cells. Cells are expanded in suspension culture and are then inoculated into either fed batch or

perfusion culture for MAb production. The fi rst strategy, which is also the cur-rent industry standard, uses fed batch culture with the cells in suspension in a stirred tank fermenter. The second strategy uses perfusion culture with the cells immobilized on Cytopore™ micro-carriers in a stirred tank fermenter. The third and fi nal strategy is perfusion culture with Cytoline™ microcarriers in a fl uidized bed fermenter. Perfusion cultures, while leading to a somewhat lower product titer, were characterized by a much smaller equipment footprint. This in turn led to a >30% reduction in investment costs and a 12% reduction in MAb production costs calculated over fi ve years of depreciation and ten years of production time.

We have endeavored to answer the following question: “Is microcarrier perfusion technology a viable economic alternative to fed batch culture?” The information presented here supports this assumption, even without taking into account improvements in MAb activity. Genzyme Corporation, for example, has claimed a four-fold increase in overall productivity through the use of perfusion technology.1

The Goal of This Study

We will show the positive fi nancial consequences of introducing microcar-rier-based perfusion into large-scale biopharmaceutical production. The study has been performed as a collabo-ration between GE Healthcare, respon-sible for defi ning the process parameters and process design, and Vogelbusch GmbH, responsible for all pre-engineer-ing work.

Monoclonal Antibody Market Potential

In 2004, the total market for bio-pharmaceuticals was worth nearly $60 billion. By 2005 that share had grown by more than 15% to about $70 bil-lion. And this double-digit growth is forecasted to continue for the next few years. 40% of the biopharmaceuticals currently in development are MAbs.2 Six of the 19 MAbs approved by the FDA had global sales of more than $500 million in 2004. Both Infl iximab (Remicade, Centocor) and Rituximab (Rituxan, Genentech) had global sales of over $2 billion in 2004.3


In November 2007, Remicade became the fi rst anti-TNF biologic to treat one million patients worldwide.4

MAb Production Improvement Goals

As MAbs are therapeutically effec-tive only at doses 100- to 1,000-fold higher than other recombinant proteins such as hormones or growth factors, relatively large amounts are needed. For example, a yearly cumulative dose of 3–5 g of Herceptin (Trastuzumab, Roche) is required for the treatment of breast cancer.

The yearly global consumption of a successful antibody can thus reach several hundred kg, making the application of MAbs a rather expensive treatment. Additionally, fermentation capacities are limited. In 2006, total production capacities were about 2.3 million liters. However, the demand is predicted to increase to between 3.5 and 4 million liters by 2011— consequently, a shortfall in production capacity for recombinant proteins has been pre-dicted.4 Intensive efforts are required to step up this capacity. In addition to the construction of new production plants, increasing overall productiv-ity and improved yields from future manufacturing processes will be among the drivers leading to higher production capacity.

High volumetric productivity (i.e., maximum product output per liter of reactor volume) is a key parameter in the effort to increase productivity.4 This can be achieved by increasing the viable cell concentra-tion in the reactor through the use of perfusion culture.

The two overriding goals in any process are fi rstly to improve the bio-logical activity/effi cacy, and secondly to minimize costs by improving reac-tor output and product quality. The latter parameter is usually defi ned by increased product homogeneity, improved glycosylation and reduced aggregate formation.5 The combina-tion of these factors leads to a longer half life, a reduced immunogenicity and facilitates downstream processing.

The Positive Consequences of Perfusion Culture

Changing to perfusion culture makes more effi cient use of the bioreac-tor due to a higher cell concentration and the fact that the generated cell mass is exploited for production over a longer time period. The volumetric pro-ductivity of the vessel is thus increased. The use of perfusion culture means that steady-state conditions can be achieved over the entire production phase, which is more similar to the homeostasis that is maintained in vivo.

In order to perform perfusion culture, the cells must be retained in the bioreactor to avoid fl ushing out. Voisard et al.6 have investigated the potential for different cell retention

techniques. The principal possibilities were: fi ltration (spin fi lters, cross-fl ow fi ltration) and sedimentation (centrifu-gation, gravity settlers and hydrocy-clones). Cells can also be immobilized on microcarriers.8,9 The advantage of using microcarriers is that the cells are attached to them so that separation is already achieved. Macroporous carriers also protect the cells from bubble-induced shear stress when sparging is used for oxygen supply, necessary in high density cultures. It has also been shown that perfusion improves product quality by increasing the relative glycan content and the extent of sialylation.10,11

The consequences of perfusion on reactor size can be seen in Figure 1. Reactor output is compared in Figure 2. A comparison of the effects of batch

FIGURE 1. Perfusion can increase fermenter productivity: the same volume with 10x more productivity, or the same productivity in a 10x smaller reactor.

FIGURE 2. Perfusion reduces downtime of the production fermenter for better utilization of equipment. Higher cell concentration gives higher volumetric productivity.


and perfusion culture on metabolite levels is provided in Figure 3. In a fed batch culture, nutrient levels can be kept constant while waste accumulates and eventually terminates the process. In perfusion culture, on the other hand, nutrient and waste levels are kept at steady-state concentrations and the levels can be adjusted during the pro-cess by changing the medium renewal rate. Improvements in product quality through the use of perfusion culture are listed in Table 1.

Achievable Advantages

Such advantages include: 1) maxi-mized volumetric production rates; 2) robust, reproducible and economical process; 3) scale-up to production bio-reactors; and 4) impact of culture stabi-lization on product quality (decreased cell turnover and lysis rate, lower pro-tease/sialidase release, and higher sialic acid content). A four-fold increase in productivity was achieved by switching from fed batch to a perfusion process,1 and in published literature, even higher increases in volumetric productivity have been shown. Ozturk12 and Yang et al.13 both reported a ten-fold increase when perfusion was compared to a fed batch process.

Summarizing the Benefi ts of Perfusion Technology

• Lower capacity fermenters produce higher cell concentrations.

• Smaller plants save in investment and running costs.

• Minimal cellular stress occurs.

• There is an increased specifi c productivity and improved quality.

• Reduced downtime of the fermenter can be realized.

• Not as much cell debris in supernatant simplifi es downstream processing.

• Smaller downstream equipment can be utilized.

• Chromatography gels can be used to full cycling capacity.

REFERENCES

1. Genzyme Corp. In a case study presented at the IBC Cell Culture and Upstream Processing conference in December 2003.2. Tauzin B. 418 biotechnology medicines in testing promise to bolster the arsenal against disease. Biotechnology – Medicines in Development 2006, p. 1-52.3. Reichert J et al. Monoclonal antibody successes in the clinic. Nature Biotechnology 2005; 23, 9: 1073-1078.4. <http://sev.prnewswire.com/medical-pharmaceuticals/20071106/NETU07806112007-1.html>.5. Werner R. Economics aspects of commercial manufacture of biopharmaceuticals. Journal of Biotechnology 2004; 113: 171-182.6. Spearman M et al. Production and glycosylation of recombinant beta interferon in suspension and Cytopore microcarrier cultures of

CHO cells. Biotechnology Progress 2005, 21, 1: 31-39.7. Voisard D et al. Potential of cell retention techniques for large-scale high-density perfusion culture of suspended mammalian cells. Biotechnology and Bioengineering 2003; 82, 7: 751-765.8. Chengzu X et al. High density and scale-up cultivation of recombinant CHO cell line and hybridomas with porous microcarrier Cytopore. Cytotechnology 1999; 30: 143-147.9. Chen Z et al. High-density culture of recombinant Chinese hamster ovary cells producing prothrombin in protein-free medium. Biotechnology Letters 2001; 23: 767-770.10. Lipscomb M et al. Effect of production method and gene amplifi cation on the glycosylation pattern of a secreted reporter protein in CHO cells. Biotechnology Progress 2005; 21: 40-49.11. Goldman M et al. Monitoring recombinant interferon gamma N-glycosylation during perfused fl uidized bed and stirred tank batch cultivation. Biotechnology and Bioengineering 1998; 60: 596-607.12. Ozturk S. Engineering challenges in high density cell culture systems. Cytotoechnology 1996; 22: 3-16.13. Yang J et al. Achievement of high cell density and high antibody productivity by a controlled fed perfusion bioreactor process. Biotechnology and Bioengineering 2000; 69: 74-82.

NOTES

CytoporeTM and CytolineTM are trademarks of GE Healthcare Companies. ©2008 General Electric Company – All rights reserved.

• Growth on Cytopore carriers increases productivity and reduces aggregate formation of gamma-interferon compared to suspension culture.6

• Perfusion increases glycan content and sialylation of secreted alkaline phosphatase.10

• Increased glycosylation of gamma-interferon produced in perfusion cultures compared to fed batch.11

TABLE 1. Product quality with perfusion cited in published literature.

FIGURE 3. Nutrients and fermentation byproducts stay at the same concentration to maintain steady-state conditions.

37

Purifi cation of Biological

Products

4th International

OptimizingRecovery and Purity While

Minimizing Cost

Topics:• Platform Development

• Process Optimization

• Chromatography Media Selection

• Sensor and Process Automation Technology

• Clarifi cation Techniques

• Ultra and Diafi ltration Methods

• Viral Clearance

• Aggregate, DNA, and Protein Removal

• Endotoxin and Leachate Removal

• Protein-A and Alternative Capture Methods

• New In-Process Testing Methods

�Featuring a TourFeaturing a Tour ofof Merck’s BiologicsMerck’s Biologics

Pilot PlantPilot Plant

Meeting Held Concurrently With

Cell CultureScale-Up

Applications:• Antibodies

• Recombinant Proteins

• Viral Vaccines

• Viral Gene Vectors

• Plasmids


September 8-10, 2008 • Philadelphia, Pennsylvania

Register Now!www.wilbio.com

757.423.8823

PUR08_MB.indd 1 5/4/2008 4:23:03 PM

SUPPLIER SIDE


� THE FIRST IN A SERIES OF MISSIONS �

Extreme Pipetting:An Issue of Mountainous ProportionsUnderstanding Barometric Pressure’s

Effects on Data Integrity

George Rodrigues, PhD (email: [email protected], phone: 207-854-0860) is Senior Scientifi c Manager at ARTEL, Westbrook, Maine.

Climb Every Mountain, George Rodrigues, PhD, Pharmaceutical Formulation & Quality July/August Issue, © 2007 Wiley Periodicals, Inc.

By GEORGE RODRIGUES, PhD

This article is the fi rst in a series of about Extreme Pipetting, which will focus on the impact of environmental condi-

tions on data integrity.

This particular article will highlight how barometric pressure affects pipetted volumes.

Introduction

One cold day in early winter, a group of scientists prepared to trek to the top of Mount Washington, the highest peak in the White Mountains of New Hampshire. At an elevation of 6,288 feet, this mountain is known as the “home of the world’s worst weather.” Clocking the globe’s high-est recorded surface wind speed of 231 miles per hour, an average year-round temperature that is below freezing, and 21 feet of snow per year, Mount Washington is no paradise. Despite these extremes, ARTEL scientists loaded their equipment and prepared for the three-hour journey to the summit.

No, this was not a “Lewis and Clark” expedition, nor a new winter Olympics event. This was the start of ARTEL’s Extreme Pipetting Expedition, a year-round scientifi c study seeking to quantify the impact of environ-mental conditions on liquid volume and data integrity. By testing pipette performance under extreme conditions, ARTEL will help pipette users more fully understand how their instruments will perform in various laboratory envi-ronments so that they can eliminate a potential source of error and strengthen quality assurance.

Like any good scientifi c expedition, the ARTEL journey’s mission is critical. Why? Because environmental condi-tions are a major source of volume variation and laboratory error. While pipetting does not normally conjure images of extreme weather and moun-tainous heights, laboratories do work in or with extreme conditions, whether pipetting hot and cold liquids or work-ing in walk-in incubators and glove boxes. Given the small volumes used in today’s laboratories, volume inaccu-racy of just 1.0 µl can signifi cantly alter results.


Mission #1 – Barometric Pressure

The fi rst stop on ARTEL’s Extreme Pipetting Expedition was Mount Washington, where expedition mem-bers sought to isolate the effect of baro-metric pressure on pipetted volumes. While there are relatively few laborato-ries around the world operating at this elevation and low barometric pressure, how many actually operate at zero elevation? And how many laboratories at varying elevations outsource pipette calibration to fi rms operating at abso-lute sea-level? To provide an illustra-tion of the potential for error, ARTEL has measured pipette performance at sea level and will compare it to pipette performance at over 6,000 feet.

For the Mission #1 experiment, conducted in the laboratory at Mount Washington’s Weather Observatory, ARTEL tested three different pipettes from leading manufacturers for dis-pensing accuracy and precision using the ARTEL PCS® (Pipette Calibration System). A robust, portable system, the PCS is unaffected by environmen-tal conditions. The PCS also provides standardized, National Institute of Standards and Technology (NIST)-traceable results for a scientifi cally sound comparison of data regardless of the environment in which data are taken or pipettes are used.

At Mount Washington, the opera-tors pipetted 22 data points for each target volume with each of the pipettes. To generate a standardized comparison, ARTEL tested the same pipettes prior to the mission using the PCS in ARTEL’s accredited laboratory (ISO 17025) under normal, controlled conditions at sea-level. ARTEL then compared average volume dispensed in Mount Washington with the average volume dispensed in the control laboratory.

Data collected show that at this altitude, air displacement pipettes (most commonly used in labs) under- delivered by 1–10% compared to the data taken in the controlled laboratory. The errors measured were greatest at the lowest volumes.

Table 1 shows data taken with fi xed-volume pipettes. Here, the 1 µl pipette

TABLE 1. Fixed-volume pipette data comparing a control lab to extreme conditions on Mount Washington.

FIGURE 1. Fixed-volume pipettes at low barometric pressure (6,288 feet above sea level).


under-delivered by 9.8% on average. This is compared with the 200 µl pipette, which under-delivered by 1.3% on average (Figure 1).

This volume variation is explained by air’s lower density at higher altitudes. Air displacement pipettes contain air gaps between the piston and the liquid being pipetted that act as a cushion. When an operator aspirates, presses his/her thumb down on the plunger at the top of the pipette and inserts the pipette into the liquid, air is trapped in the pipette tip. This is called the dead air volume. When the plunger is released, the trapped air acts as a spring that connects the piston to the liquid and pulls the liquid up and into the tip. If air is less dense, less liquid is pulled into the pipette tip and subsequently dispensed, resulting in under-delivery and data inaccuracy.

In the next phase of the experiment, an operator pipetted with variable- volume pipettes at their respective upper-and lower-range target volumes (see Table 2). The fi rst set of data showed that operator technique was not a source of error, and therefore only one operator was necessary to obtain the second set of data. The data show a dramatic error rate when the variable-volume pipettes were set to dispense less than full-scale volume, and this error grew as smaller volumes were tested. For example, volumes varied by over 30% when using the 2 µl pipette at its low-volume range (Figure 2).

To isolate barometric pressure as the source of volume variability, ARTEL controlled for other sources of error, including operator technique, temperature, humidity and the measuring method. To control for this,ARTEL used the PCS to measure volumesdispensed by syringes at Mount Washington. Syringes work through positive displacement instead of air displacement, and would therefore be unaffected by the environment.

The results generated with syringes can be found in Table 3. The error rates for all syringes were below 1%, showing that the PCS was performing with less than 1% error. This confi rms that the PCS was unaffected by change in elevation.

TABLE 2. Variable-volume pipette data comparing a control lab to extreme conditions on Mount Washington.

FIGURE 2. Variable-volume pipettes at low barometric pressure (6,288 feet above sea level).

TABLE 3. Syringe data comparing a control lab to extreme conditions on Mount Washington.


To rule out pipetting skills as the source of variation, the pipetting technique of the expedition members was standardized and qualifi ed using the ARTEL Method. Humidity was controlled by pre-wetting the pipette tips, and temperature was not a factor since testing was conducted inside the observatory laboratory at an ambient temperature equal to that in the control laboratory.

What Does This Mean?

Consider a laboratory that sends pipettes to a service located at sea-level for calibration. Even when newly- calibrated, the pipettes would likely under-deliver once back in the labora-tory and produce inaccurate data. The data shows the importance of regularly verifying pipettes in the environment in which they are used.

The good news is that the magnitudeof volume variation for each pipette was relatively constant, which facili-tates correction within the laboratory. For example, the 10 µl pipette under- delivered by 2.6% and 2.8% in replicate tests. To compensate for this repeatable volume variation, laboratories have two options: 1) adjust the internal mecha-nism of the pipette for the specifi c envi-ronmental conditions; or 2) adjust the delivery setting. In this second option, because the pipette under-delivered by an average of 2.7%, setting the pipette to deliver 10.27 µl would deliver an actual volume of 10 µl.

Regular onsite verifi cation of liquid handling instrumentation is critical to accounting for volume variations caused by environmental conditions which fl uctuate between laboratories,and can even fl uctuate within a laboratory. Referred to as Interim Verifi cation, Quick Calibration or Functional Checks by scientists and regulators, this will help eliminate errors and ensure data integrity.

What’s Next?

Stay tuned for the Summer 2008 issue of BioProcessing Journal, which will discuss the fi ndings of the Extreme

Pipetting Expedition’s second mission: exploring how delivered volume is affected by the temperature of pipet-ted liquids. Yellowstone National Park, home to Old Faithful and well-known for mineral springs of varying tempera-tures, was selected as the site for Mission #2. Yellowstone is also distinguished as the birthplace of a critical component of PCR testing, Taq DNA polymerase, derived from bacteria fi rst isolated in the park’s Fountain Geyser region.

Thermal disequilibrium is common

in laboratories working with extremely hot or cold liquids. For example, restric-tion enzymes are frequently handled at freezing temperatures in nucleic acid work, while inoculum is often pipetted at temperatures as high as 37°C in cell culture processes.

Laboratory managers and technicians are encouraged to view the complete data by logging on to:

www.artel-usa.com/extreme

ANIMALS CAN BEUNPREDICTABLE.

LET’S KEEP THEM OUT OF YOUR PRODUCTION.

Reduce safety concerns, improve performance

and facilitate the regulatory process with our

new CellPrime animal-free supplements.

Today’s threat of animal pathogens presents a new set of challenges, not to

mention a new list of compliance requirements. That’s why Millipore has teamed

up with Novozymes, the world leader in bioinnovation to offer two, groundbreak-

ing animal-free supplements. CellPrime rAlbumin AF-S and CellPrime rTransferrin

AF eliminate the risk of TSE, without sacrificing performance. In fact, studies show

CellPrime rAlbumin AF-S promotes equivalent or greater production and yield

compared to plasma fractionated albumin, while CellPrime rTransferrin AF

frequently out performs animal derived transferrin and chemical chelators across

multiple cell lines. See for yourself by ordering your samples today, and keep

producing without the fear of contamination or regulatory headaches.

For samples and more information, visit www.millipore.com.

Millipore is a registered trademark of Millipore Corporation. The M mark is a trademark of Millipore Corporation. CellPrime is a trademark of Millipore and Novozymes. Novozymes is a registered trademark of Novozymes A/S.

TM

SUPPLIER SIDE


Bioinformatics Application Has a BLASTWith Exanodes Shared Internal Storage

Frank Gana ([email protected]) is the Business Development Director for Seanodes SA which is headquartered in Paris, France.

As scientifi c research has become more sophisticated, the fi eld of bioinfor-matics — where computer technology and biology meet — has become increas-ingly critical to our understanding of the natural world. Entire databases of biolog-ical data are created, indexed, organized, and analyzed, requiring sophisticated and robust tools. Bioinformatics often make use of mathematical computations, algorithms, artifi cial intelligence, model-ing, and other complex applications.

The most widely used of these tools today is known as BLAST (Basic Local Alignment Search Tool). As the acronym suggests, BLAST is intended to be a high-speed application. BLAST fi nds statisti-cally-probable similarities and patterns in data such as DNA sequences, proteins, or nucleotides. There’s no question that BLAST has revolutionized the industry it serves, and there’s also no question that all this analytical power requires high-performing processors.

The Challenge

BLAST requires greatly accelerated performance in order to keep pace with demanding search analyses. Just three years ago, a network fi le sys-tem (NFS) array was more than adequate to manage BLAST queries, but recent improvements in processor speeds, as well as a need for additional servers, means an NFS array is now an unavoidable, productivity-slashing bottleneck— one that becomes far more expensive and time-inten-

sive to manage as the bottlenecks persist and workarounds are sought.

Overburdened arrays can no longer serve data at the pace the central process unit (CPU) needs it. As a result, CPU rates are getting lower and lower and are now far from their optimal speed, and all resources, including humans, spend far more time waiting for data than analyzing and interpreting it.

Organizations with a limited tech-nology acquisition budget but a strong internal commitment to solving prob-lems are challenged to look beyond the typical answers to this severe decline in performance. After an in-depth evalua-tion of business needs, some companies have discovered a shared internal storage (SIS) approach, that makes use of unused disk space inside computing nodes, will overcome this challenge.

The Solution

The pioneers in SIS, Seanodes, is able to achieve a twofold increase in BLAST speeds thanks to its Exanodes™ software. Exanodes transforms the stor-age components inside computing servers into a high-end virtualized storage pool, enabling users to quickly identify and

take advantage of under-utilized internal storage capacity. With each node func-tioning as both a computing and storage server, the need and expense of buying and maintaining dedicated storage arrays is eliminated, thereby enabling full infra-structure consolidation.

In addition to increasing BLAST query performance, users realize increased reliability as the NFS server is no longer a single point of failure. The simplicity of the Exanodes model reduces management time and complexity, and physical installation takes less than two hours.

Just as the fi eld of bioinformatics is changing our understanding of grand biological questions like the human genome, shared internal storage is chang-ing the way we look at network storage technology because the SIS platform radically alters the economics and pos-sibilities in data storage and application processing.

When we compare return-on-invest-ment fi gures for Seanodes’ Exanodes versus additional servers and the gains realized by the SIS model, Exanodes is an investment that is between seven and ten times more profi table. Exanodes can allow end-users to cancel their storage

and server investment for one year, as they are able to do twice the work without adding any components in the infrastructure.

In these times of budget restrictions, this technology allows IT directors to keep pace with growing business needs while keeping their CFOs happy.

Exanodes™ is a trademark of Seanodes.

By FRANK GANA

43

September 8-10, 2008 • Philadelphia, Pennsylvania

CellCultureScale-Up

3rd International

Mammalian Cell Line Development, Scale-Up, and Productivity

Topics:• Gene Selection and Insertion

• Cell Line Development

• Expression Enhancement

• Media Development

• Sensor and Process Automation Technology

• Feed and Harvest Strategies

• Oxygen Delivery

• Cell Characterization

• Master and Working Cell Bank Development

• Viral Infection Optimization

�Featuring a TourFeaturing a Tour ofof Merck’s BiologicsMerck’s Biologics

Pilot PlantPilot Plant

Meeting Held Concurrently with

Purifi cation of Biological Products

Register Now!

www.wilbio.com757.423.8823

Applications:• Antibodies

• Recombinant Proteins

• Viral Vaccines

• Viral Gene Vectors


CSU08_MB.indd 1 5/4/2008 2:45:56 PM


NEW PRODUCTS & SERVICESStereomicroscope Illumination

With LED Technology Carl Zeiss (Thornwood, NY)

has enhanced the performance of its Stemi DV4 stereomicroscope with a new illumination system. The Stemi DV4 on the C LED stand features integrated LED illumination units for refl ected light, transmitted light and mixed light. The LEDs provide excel-lent light intensity with low power consumption and a long service life of the illumination system (~25,000 operating hours).

Thanks to white light illumina-tion in daylight quality, the objects are visualized in high contrast and natural colors. To protect the specimen from exposure to heat, the LED light doesn’t contain infrared wavelengths.

These benefi ts are particularly important when 3D observation of small objects is required in high quality and at higher magnifi cations, e.g. at routine workstations in biomedical laboratories, in industrial production and quality inspection, and for training in schools and scientifi c institutions.

Compared to the previous halogen illumination system, the C LED stand offers brighter and more homogeneous illumination of the object fi eld – with considerably lower costs as power con-sumption has been halved and the long service life of the LED systems makes bulb replacement unnecessary.

The Stemi DV4 stereomicroscope on the C LED stand permits the use of digital photo or video cameras via an eyepiece adapter. AxioVision software from Carl Zeiss is also available for image capture and evaluation.

www.zeiss.com/micro

New Release of NuGenesis SDMS for Windows Vista

Waters Corporation (Milford, MA)just announced a new release of its NuGenesis® scientifi c data manage-ment system (SDMS) for the Business and Enterprise editions of the Microsoft®Windows® Vista® operating system platform (32 bit, Service Pack 1). Waters® NuGenesis® SDMS 7.1 Service Release 3 (SR3) also supports Microsoft Offi ce 2007 with SP1 and includes updates for SDMS Vision Publisher™ that allow it to be validated on Vista and Offi ce 2007. The release of software is available immediately worldwide.

Other operating systems supported by SR3 include Microsoft Windows 2000 and XP. Organizations using NuGenesis SMDS version 7.1 under a maintenance contract are entitled to a free upgrade to the newly released software.

Waters NuGenesis SDMS, a key offering in the Waters Laboratory Informatics Suite, is an application-independent software and database platform that creates a common electronic repository for scientifi c information throughout laboratory-dependent organizations. Waters NuGenesis SDMS captures, stores, and retrieves scientifi c information and lets scientist focus on science by eliminat-ing the headache of paper and data fi le backup.

The NuGenesis SDMS platform includes the SDMS Vision Publisher, an authoring and reporting tool that optimizes the utilization of informa-tion collected and cataloged by SDMS. With an open and fl exible design, Vision Publisher easily adapts to fi t and support the existing workfl ows in

your lab.Waters NuGenesis SDMS improves

the value of information by enabling scientists to unify, share, and fully re-utilize scientifi c data generated from the vast array of laboratory applications.

www.waters.com/waters/nav.htm?cid=10067099

New Interfaces for AutomatedBioReactor Monitoring

Groton Biosytems (Boxboro, MA) now has ARS-M interfaces available to automatically collect and deliver sam-ples to both the Beckman ViCELL™ and the innovatis Cedex™ automated cell culture analyzer.

To help automate the entire cell culture process, the ARS-M provides a sterile and automatic connection between the bioreactor sample and the cell counter, and ensures rapid delivery of samples for accurate online cell counts.

Advantages to using the ARS-M interfaces are:

* increased productivity;* increased sample frequency; * an optimized process;

* higher yields; and

* reduced costs.The Groton Automated Sampler

(ARS) system provides kits to enable you to monitor your process 24 hours a day, 7 days a week through a pro-prietary sterile interface to a variety of analytical instruments.

The modularity of the ARS system provides you with the ability to imple-ment automated sampling of up to eight bioreactors or fermentors. In addition, the ARS permits near real-time online, hands-free monitoring.

www.grotonbiosystems.com


NEW PRODUCTS & SERVICESProduction-Scale Bioreactor

System That Fits on the BenchNew Brunswick Scientifi c (Edison,

NJ) has developed a cGMP-compliant system for pilot and production-scale cell culture applications. The CelliGen 510 is a modular system designed to provide the fl exibility needed to meet changing process requirements, at any time, pre-or post-delivery.

Ideal for growth of any mam-malian, insect or plant cell line, this versatile sterilizable-in-place benchtop or mobile bioreactor system comes in 19.5 L and 40 L sizes. The CelliGen 510 is capable of batch, fed-batch and perfusion modes.

Multiple gas fl ow options, special-ized impellers, single and redundant probes, options for spray balls, valida-tion packages and much more, enable customization to your needs. A built-in load cell precisely measures vessel contents, enabling integrated control of pumps for automatic addition of fresh media, pH, DO, or foam control agents, or harvesting.

The system includes control of up to 32 loops from an easy-to-use touchscreen interface with multiple analog inputs and outputs provided for connection of up to 14 external devices. A built-in load cell measures vessel volume for precise additions and harvesting; and 24 vessel penetrations provide fl exibility to position sensors, sprayballs and more.

www.nbsc.com/bi6.htm

Time-Saving/Single-UseBioreactor Sampler

Aseptika Ltd (Huntingdon, UK) has launched their Bioreactor SamplerTM product line. With a snap, sample, seal–sampling bioreactors and fermentors only takes about 15 seconds.

Monitoring, assaying and record-ing parameters off-line are an essential part of running and maintaining a bioreactor or fermentor. Because the lab environment is not always totally clean and sterile, taking samples from a bioreactor can contribute to bacterial, viral or chemical contamination of the vessel. Mammalian cell cultures are particularly at risk because of the rich media used and the length of time for which the culture is often run.

The Bioreactor Sampler (patent applied for) has a unique system that withdraws samples through its non-return valves. Simply fl ush the sample line into a waste collector, snap the sin-gle-use valve, sample into the sample syringe, and then seal the USP Class VI tubing and remove the syringe.

The Bioreactor Sampler is:

* usable with any type of bioreac-tor or fermentor;

* single-use: no cleaning, autoclav-ing and non-contaminating;

* fl exible and modular: 6, 12 or 18 samples as standard, with additional capacity using individually packed Sample Modules;

* a unique unidirectional fl ow system constructed with medical-grade materials (USP Class VI); and

* made under Class 10,000 clean room conditions and gamma irradi-ated to 25-40 kGy to ensure sterility.

www.aseptika.com

Collagen for 3D Cell CultureInvitrogen (Carlsbad, CA) is now

offering GIBCO® Collagen I, Rat Tail to customers for their 3D cell cul-ture needs. Scientists are increasingly using high-fi delity cell culture models that are more predictive of disease states and drug response, often using

extracellular matrix (ECM) proteins to closely mimic how tissues grow in vivo.

Invitrogen’s collagen is a widely-used ECM in cell culture applications, and it facilitates cell attachment, growth, differentiation, migration and tissue morphogenesis during development.

GIBCO Collagen I is high quality, forms a clearer, fi rmer gel and includes protocols for each application.

www.invitrogen.com/3d-cellculture

PCR Detection in Real-TimeBio-Rad Laboratories (Hercules, CA)

announces the new CFX96™ real-time PCR detection system. The CFX96 is very simple to install and delivers sensitive, reliable detection for applica-tions including absolute quantifi cation, genetic variation analysis, and gene expression.

The fl exible, easy-to-use CFX Manager™ software accommodates the needs of all users, whether performing a real-time PCR experiment or analyzing


NEW PRODUCTS & SERVICESa comprehensive gene expression study.

Key benefi ts of the Bio-Rad CFX96 system include:

* long lasting solid-state compo-nents with LEDs and photodiodes for reliable results;

* quick installation and setup with factory-calibrated optics;

* fast thermal cycling to produce results in less than 30 minutes;

* a thermal cycler gradient feature optimizing reactions in a single experiment;

* minimal sample and reagent usage that offer reliable results obtained with low reaction volumes;

* experiments tailored to need by use one of several data acquisition modes including a single channel fast scan option for SYBR Green; and

* software that can send email notifi cation with an attached data fi le upon run completion and up to four instruments that may be controlled by a single computer.

www.bio-rad.com/gemomics

Portable Raw Material AnalysisThe SmartProbe Analyzer by FOSS

NIRSystems, Inc. (Laurel, MD) is based on XDS near-infrared technol-ogy and provides the next generation of dedicated NIR analysis for rapid non-destructive measurement of liquid and solid chemical and pharmaceutical formulations.

The XDS SmartProbe Analyzer is ruggedly manufactured for either the warehouse or plant environments. Sensitive identity and quality tests of liq-uid or solid actives are performed directly in the original shipping container.

The ergonomic, hand-held design

is straightforward—simply place the probe in the sample and depress the trigger. Pass/fail results are displayed on the handle after each test.

The FOSS XDS platform ensures decreased method development time and seamless method transfer. It is ideal for use in a GMP environment where 100% identifi cation of materials, as per PIC/s or other regulations, is required.

Identifi cation, qualitative and quantitative methods are easily derived with the advanced, user-friendly, networkable Vision soft-ware. Precise and accurate analysis is accomplished with the press of a key or click of a mouse.

FOSS NIRSystems provides a complete range of rapid scanning, near-infrared spectrophotometers for laboratory and process applications.

www.foss-nirsystems.com

Smaller DiameterMonolithic HPLC Column

Phenomenex Inc. (Torrance, CA) has added a new 50 x 2.0 mm ID column to its Onyx monolith line. The 30% smaller diameter column requires less sample volume and reduces solvent consumption. The Onyx columns, which contain solid rods of fused silica, produce results at high fl ow rates with low backpressure, reducing run times over conventional

columns. The permeable media separates viscous biological samples that clog traditional columns, making Onyx ideal for drug metabolism stud-ies in pharmaceutical companies and contract labs.

For fast gradient conditions, fl ow rates on the Onyx 2.0 mm ID can be increased during re-equilibration without backpressure constraints, sig-nifi cantly reducing overall sample cycle

times. The increased speed makes the columns ideal for high-through-put applications. Onyx columns are also available in 3, 4.6 and 10 mm diameters.

www.phenomenex.com

High-Purity/Solvent-ResistentPVDF Ventilation System

Asahi/America Inc. (Malden, MA) has introduced the latest addition to their Purad piping line: the Purad ProVent high purity PVDF ventila-tion system. The Purad ProVent system features superior chemical resistance properties that make it ide-ally suited for critical exhaust applica-tions such as solvent, acid and caustic fumes.

The new Purad ProVent system isoffered in sizes ranging from 2˝(63 mm)to 16˝ (400 mm) with standard fi ttings, damper valves, sanitary adapters and fl ex connects.

Purad ProVent is FM 4910 listed for clean room materials and FM 4922 for fume exhaust ducts. It can be easily fabricated onsite to fi t actual piping needs; is constructed with lightweight material, making it easier to handle and support; and provides reliable resistance and long life.

Resistant to solvents, the versatile Purad ProVent exhaust system replaces the costly installation of steel-lined vent systems. This new system is easy to work with, using ir fusion, butt fusion or hot air welding; no glues are required. The system can be custom fabricated in the fi eld for lower instal-lation costs.

www.asahi-america.com


NEW PRODUCTS & SERVICESOPC Connectivity Suite for

Cell Culture Analyzer Nova Biomedical (Waltham, MA)

has a new Connectivity Suite which is a complete plug and play solution for use with the BioProfi le® FLEX Cell Culture Analyzer that facilitates Nova’s groundbreaking bi-directional auto-mation and product integration.

The BioProfi le FLEX Connectivity Suite features complete connectivity with all OPC-compliant systems including bioreactor controllers, data historians, overall IT infrastructure, and plant manager systems.

BioProfi le FLEX is an “all-in-one” analyzer that provides immediate measurement of key chemistries, cell density/viability, and osmolality in cell culture media to provide a total picture of cell growth in a single instrument. Consolidating these key tests into a one instrument saves time and labor, elimi-nates errors, and improves compli-ance in every phase of the bioprocess. BioProfi le FLEX is compatible with the FDA’s PAT initiative to communicate, capture, and control bioprocess data for cell culture processes.

BioProfi le FLEX is offered in a series of modules to meet specifi c user needs. Multiple confi gurations are available building on the base chemistry module and adding cell density/viability and/or osmolal-ity modules as required. BioProfi le FLEX offers individual sampling via syringe, multiple sampling from a choice of tray confi gurations, or automated sampling via an optional On-Line Autosampler from up to ten bioreactors.

www.novabio.com

Carbon NanotubesMethodologies

Microfl uidics (Newton, MA), a wholly owned subsidiary of MFIC Corporation, recently announced their methods for processing bulk carbon nanotubes. Microfl uidics has devised a methodology to process, deagglomerate, purify and stabilize these processed nanotubes so they can be used in an optimal manner, thus achieving enhanced functionality and performance.

The company has identifi ed and demonstrated numerous operational protocols utilizing its Microfl uidizer® processor systems to prepare these bulk nanotubes for their ultimate spe-cifi c uses. These applications capitalize on the unique mechanical, electrical and thermal properties of carbon nanotubes, and require dispersion of the nanotubes in liquid media and independent length reduction of the carbon nanotubes.

These nanotubes are noted for their tubular shape, extremely small diameter with respect to length, and exceptional physical, mechanical and

electrical properties. When prepared in diameters in the 10-50 nm range, the unique characteristics of strength, conductivity and quantum behavior become highly desirable and are thus incorporated into a multitude of com-mercial products.

Microfl uidizer processors are ideal for conditioning carbon nano-tubes (and nanomaterials in general) for many applications including medical therapeutics and diagnostic biolabels.

www.microfl uidicscorp.com

Product Enhancements inFlow Meters/Controllers

Brooks Instrument (Hatfi eld, PA)has announced several mass fl ow meter/controller product updates. The Model 4800 Series now has digital communications via RS-232 ( standard) and a downport connection option, and a local operator interface (LOI) option. Also, it is the fi rst of Brooks’ mass fl ow devices to be fully RoHS compliant.

The one-to-one RS-232 digital communication protocol allows the customer to provide a setpoint and receive the process value, select one of nine gases for the calibration curve, access the valve override function and the re-zero function. Optional down-port connections allow top mounting for ease of maintenance and installa-tion, and saves space because there is no point-to-point tubing necessary.

The RoHS Directive bans new electrical and electronic equipment containing more than agreed-to levels of lead, cadmium, mercury, etc. from being distributed in the EU. The 4800 Series is Brooks’ fi rst, fully RoHS-com-pliant MFC.

A local operator interface (LOI) option allows the user to view, control, and confi gure the 4800 Series devices. A 28 mm x 11 mm backlit display mounts directly on top of the instrument pow-ered via a DC adapter. It measures the fl ow signal level from the device and displays it on an LCD. A jog dial button allows the user to adjust the fl ow, per-form a valve override, zero the device, and set device confi guration parameters. It also allows the user to connect a remote controller to the device.

www.emersonprocess.com/brooks/


ADVERTISER INDEXContact the companies whose products and services are advertised in the BioProcessing Journal

Advertiser Page Advertiser Page

Asahi Kasei Inside Back Cover877.PLANOVA (752.6682)[email protected] www.PlanovaFilters.com

ATCC 33800.638.6597

www.atcc.org/passagenumber.cfm

BD Biosciences 5 800.638.8663Fax: 410.316.4770 www.bdbiosciences.com/advbio

Millipore 41800.645.5476Fax: 800.645.5439 www.millipore.com

Nova Biomedical 1US: 800.458.5813 / Canada: 800.800.263.5999

www.novabiomedical.com/bioprocessing0408.htm

New Brunswick Scientific 2800.668.5040Fax: 732.287.4222 www.nbsc.com/wbf.htm

Pall Life Sciences Back Cover800.645.6532Fax: 516.484.5228 www.pall.com/biopharm

Sartorius Stedim Biotech Inside Front Cover800.368.7178Fax: 631.254.4253 www.sartorius-stedim.com/biostat_cultibag

WilBio 7, 8, 9, 13, 26, 27, 37, 43, 48 757.423.8823 Fax: 757.423.2065 www.wilbio.com

BioProcessing J O U R N A LTrends and Developments

In BioProcessIn BioProcessTechnologyTechnology

For Details:www.bioprocessingjournal.com

[email protected]@bioprocessingjournal.com

Phone 757-423-8823

ASAHI KASEI MEDICAL

20 years ago, a new pla�orm was launched byAsahi Kasei Medical, with the introduc�on of Planova filters: the first validated virus-removal filters in the industry.

Today, Asahi Kasei Medical con�nues to evolveits virus-removal pla�orm products, capable ofsuperior LRVs and excellent product recovery.

Asahi Kasei Medical’s patented hollow-fibertechnology results in an extremely robustvirus-removal system suited for a wide range of biotherapeu�cs.

Planova filters are available in mul�ple surfaceareas for seamless scalability all the way fromthe laboratory to your commercial scale process.

YOUR VIRUS REMOVAL PLATFORM

ASAHI KASEI MEDICAL

www.PlanovaFilters.com

An enabling technology standard thatresults in process design innova�ons

PLATFORM

プラットホーム

plataforma

plateformepiattaforma

платформа

plattform

ASAHI KASEI MEDICAL is committed to manufacture all of its products in a manner that is safe for our environment

Contact Informa�on:Toll Free: 1-877-PLANOVA (752-6682)E-mail: [email protected]

TTrends and Developments in BioProcess …€¦ · portfolio of high performance animal-free...

Documents

Transcript of TTrends and Developments in BioProcess …€¦ · portfolio of high performance animal-free...