Cell Culture, Upstream – BioProcess International Conference & Exposition 2015

View PDF


What’s on the minds of groups involved in cell culture and fermentation in 2015? They’re looking to implement genomics and other “-omics” technologies in cell-line development and engineering. They’re working with new analytical tools to model, monitor, and control upstream production bioprocesses. Many are interested in the possibilities of perfusion and other continuous processes. They want the most sophisticated single-use options developed with modern bioprocessing in mind. They’re adapting cell culture knowledge to new product modalities and applying all this knowledge toward improvements in efficiency, productivity, and development speed. And they’re increasing their understanding of how process conditions affect product quality.

Lately, we’ve been hearing a lot of grumbling about a lack of innovation in bioprocess technology. But over the past few years, BPI has published the work of several authors who would beg to differ. For instance, a September 2014 supplement sponsored by Sartorius Stedim Biotech detailed the development and application of a new single-use, stirred-tank bioreactor concept based on a specially designed polymer film. That same month, a start-up company introduced its novel solid-media platform for microbial production. And over the past few months, we’ve published a four-part series (ending in the issue alongside this supplement) reporting on development of a new rotating- membrane bioreactor design for culturing adherent cells. See our archive box at the end of this chapter for a selection of other examples.

You’ll find many analytical approaches in that list as well as in this year’s BPI Conference program. We’ve always emphasized the importance of analytical methods throughout biopharmaceutical product and process development. But the quality by design (QbD) era is adding impetus to such work. Modern approaches such as high-density culture, serum-free/ animal-free media, and process monitoring and control require a higher level of cell-line and bioprocess understanding than ever before. Biosimilars must show comparability, many new products and expression systems demand specialized process conditions, and cell therapies present their own unique challenges. Luckily, the companies that supply equipment and instrumentation to the biopharmaceutical industry are rising to the challenge of helping bioprocessors develop and characterize their systems.

BPI’s marketing and digital content strategist, Leah Rosin, conducted the following interviews as the conference program came together this summer. Participants addressed insect-cell expression of viral subunit vaccines, glycosylation profiling based on metabolic data, and high-throughput cell-line engineering. Here, in Q&A format, is what they had to say.


Barry Buckland (Protein Sciences Corporation and University College London)
Barry Buckland is a senior advisor at Protein Sciences Corporation and a visiting professor at University College London in the United Kingdom. He will be part of the “Development and Production of New Modalities” session on Wednesday morning, 28 October 2015. His case study is titled, “Development of a Scalable and Productive Insect Cell Culture Based Process for Making Flublok, the First FDA Licensed Recombinant Influenza Vaccine,” and it features new data.

Abstract: The hemagglutinin (HA) protein from the influenza virus has proven to be an effective vaccine. This presentation will demonstrate that its manufacturing process is reproducible and scalable. An example is presented (for H7N9) on how the HA-based vaccine process used to make the Flublok recombinant vaccine by Protein Sciences is an ideal manufacturing platform for rapid response to pandemic influenza.

Can you describe the Flublok vaccine and the path that brought you to this point? It is the first recombinant influenza vaccine licensed by the Food and Drug Administration (FDA) in the United States. It was developed over many years by Protein Sciences Corporation as an alternative to traditional egg-based manufacturing and cell culture, virus-based platforms for making influenza vaccine. This is a novel approach expressing the HA protein as an antigen. It was licensed in 2014 and is now available in the United States.

What makes insect cell culture better for this application than other expression systems? It turns out that the choice of insect cell culture, which the company made many years ago, is actually perfect for making an influenza vaccine. Unlike nearly every other biotech product, the influenza vaccine changes more or less every year. The challenge is to make a new vaccine on a very short time table to be ready for the influenza season.

The nice thing about this system is that the insect cell culture part never changes. That process is already successfully developed and scaled up. The cells can be adapted to grow in suspension culture up to large volumes much like Chinese hamster ovary (CHO) cells. And the baculovirus expression system (BEVS) that goes along with it, can be used to make this new HA protein antigen.

Once a virus is chosen for the next flu season, what the company needs is a genetic sequence for making its HA protein. That can be inserted into the baculovirus that in turn is used to infect the insect cells, which then express the HA protein. It is actually formed on insect-cell membranes, which is probably ideal for this membrane-bound glycosylated protein.

This is the example I will present. The process offers fast turn-around to make a new influenza vaccine for each flu season. And if a pandemic occurs, it could be used to make a vaccine for that as well.

Can you share a little more about your case study? I’ll also present this case study for the H7N9 avian influenza virus, which does infect and is lethal to humans. This is for preparedness if that ever becomes a pandemic. Protein Sciences has made a vaccine against the virus and scaled- up the process. A pandemic flu vaccine is about to be tested in the clinic. It is built on the scaffold of the same universal process used for the seasonal flu vaccine.

Is there a specific facility ready to use this technology for rapid response? Or are you looking for partner companies to assist in such mass production? The BEVS is key to such a rapid response to a pandemic. Within a few days of getting the genetic sequence, we can insert that into the baculovirus and start to get expression. It’s always the same cell line, which already has been approved for use in making seasonal flu vaccine. So the cell line itself does not have to be requalified.

Scale-up is similar to that for a CHO cell culture process. Such facilities are in use throughout the world for making antibodies and other therapeutic proteins. At least conceptually, any of those facilities could be readily converted to making a vaccine for a pandemic flu, with modest changes. That makes this a very attractive proposition. And other facilities in different parts of the world could be readily mobilized.

Aside from speaking, why are you attending the BPI Conference? This conference is a unique forum to get up to date on what’s happening in the bioprocessing arena. You can learn best practices of what’s out there in this rapidly changing environment. This is an exciting time to be working in this field, whether it’s biotechnology or vaccines. Tremendous progress is being made in biology and bioprocess engineering. I’m excited for the opportunity to attend and learn from other experts in the area.

>> Listen to the full interview

Michael Butler (University of Manitoba)
Michael Butler is a distinguished professor in the department of microbiology at the University of Manitoba in Canada. He will be preceding the “Integration of Process Analytics” session on Tuesday afternoon, 27 October 2015, with a featured presentation titled, “Modeling of Glycosylation: Predicting Profiles of Glycosylation from Metabolic Data.” It will feature new data.

Abstract: Single-glycoform monoclonal antibodies (MAbs) can be targeted for specific therapeutic functions. The glycosylation profile can be controlled through cellular glycoengineering, media manipulation, or enzymatic remodeling. These methods could be used strategically in production of an antibody. The expected outcomes of such strategies will be evaluated with specific examples of commercial MAbs.

How is the glycosylation pattern of a MAb controlled? There are different methods for controlling glycosylation patterns. In a normal bioprocess, antibodies are produced in a heterogeneous mixture of glycans. It is of increasing interest to restrict those glycans so that we can have either restricted glycoform patterns or a single glycoform type. The value is that we can relate functions to particular structures.

I will illustrate with examples three methods that can be used to control glycosylation. The first one is based on molecular biology. In the MabNet program, we’ve recently published our ability to transfect cells with enzymes to allow for a particular type of glycoform.

The second strategy is control of the bioprocess itself. It is important to have a range of substrates to form the precursors for glycosylation, particularly carbohydrates (e.g., glucose). They are very important in controlling glycosylation. In normal fed-batch culture processes, a carbohydrate source is typically provided at a low concentration. That may be good for productivity or cell growth, but it might affect glycosylation. Controlling the bioprocess (the media, dissolved oxygen, and pH) is key to determining a product’s glycoform profile.

The third method I will present is enzymatic remodeling. Within the purification process, we can apply enzymes such as galactosidase or transfer enzymes to modify antibody glycoforms in downstream processing.

Can you provide some background on studies you’ve conducted? Over the past five years or so, I’ve led a network in Canada that involves eight universities with different laboratories and principal investigators, molecular biologists, cell biologists, and chemical engineers. We’ve been looking at some of these methods and platform technologies. Funded by the country’s Natural Sciences and Engineering Research Council (NSERC, www. nserc-crsng.gc.ca), this “Network for Production of Single-Type Glycoform Antibodies” will carry on until about 2016. I invite anyone who would like more detail about its activities to see our list of publications online at www.mabnet.ca. We have produced some 36 graduate students from this extensive program across Canada.

What specific tools and techniques are used in your work? Key tools for structural analysis of glycosylation have involved high-performance liquid chromatography (HPLC), typically the hydrophilic-interaction liquid chromatography (HILIC) method of glycol analysis. We can combine that with a sequencing technique using a series of enzymes that break down the glycan. We normally complement that with mass spectrometry. To connect the structural basis of our work with the functionality of particular glycan types, we use binding assays (e.g., surface plasmon resonance, SPR) and cell-based assays.

Other work that we’ve done — notable for the tools we’ve used — involves a unique prototype dielectric spectroscopy probe to analyze samples from our bioreactors. This monitors electrical properties of cells to tell us something (for example) about the stages they go through. So we use that novel technique to monitor our bioprocesses as well.

Can you talk a little bit more about your other research and that of your laboratory? Our research is entirely aimed at using mammalian cells for biopharmaceutical production. This has been an emphasis for me throughout my career. I’ve been interested in the metabolism of producer cells, initially their energy metabolism and now that associated with glycosylation. Some methods we use could be predictive, so we could translate them for use in large-scale production of biopharmaceuticals.

Clearly, this is an expanding industry. As academics, we can identify many problems and look at them on a small scale over the long term. We don’t have 1,000-L bioreactors or fermentors; we do things at a small, laboratory- bench scale. But with that, we can look at some problems associated with scale- up and industrial production. My research involves a lot of collaboration with industry. The lab is staffed by a mixture of graduate students, post- doctoral researchers, and technicians. We are sponsored by scientific grants from government agencies (e.g., NSERC), and we conduct sponsored research for individual companies.

Aside from speaking, why are you attending the BPI Conference? It has become a premier event on the calendar for people in my field looking at biopharmaceutical production. I like the fact that there are many delegates: well over 1,000, maybe approaching 2,000 now. Boston is a very good location because it is the heart of a biotechnology hub. I’m very pleased that a number of academics go to the BPI Conference as well. By attending, I can talk to people, interact, and form an ongoing network, which is important for evaluating future work and research in this area.

>> Listen to the full interview

Trent Munro (Amgen)
Trent Munro is scientific director of cell science and technology, leading cell-line development for product and process development at Amgen, Thousand Oaks, CA. He will be joining us for the “Optimizing Interface with Discovery and Applying ‘Omics and Systems Biology” session on Tuesday morning, 27 October 2015. His featured presentation is titled, “High- Throughput Multiparametric Clone Screening Approach for the Generation of Tailored Production Cell Lines.”

Abstract: Production cell-line development requires precise control of product quality to match predetermined ranges. Amgen now has key capabilities and expertise to address the challenge of producing tailored cell lines. We present a case study targeting and matching a specific clone profile. Data generated is used to further develop models for predicting future campaigns.

Can you briefly describe your case study? I lead the cell-line development team here at Amgen, and my topic will be methods we’ve developed and techniques we’ve looked at for increasing our throughput during cell- line development. What that means for us is incorporating scale-down cell culture models for automated analysis of many cultures — far more than we could do in the past. We need to have that supported by analytical techniques to generate data sets that help us make data-driven decisions as we move through product/process development.

Each new biologic molecule has its own set of parameters that need to be monitored in development. We are primarily interested in productivity, which helps us narrow down to those cell lines that we feel will be useful in future manufacturing settings. Each molecule has its own set of key product quality attributes (PQAs). For proteins such as antibodies, the normal parameters people look at include glycosylation and percentage of high– molecular-weight (HMW) moieties.

We try to understand the point at which we are analyzing enough cells to get the full spread of diversity from a given population. That’s been really interesting, trying to work out at exactly what point we get a diminishing return because there’s no further diversity in the population.

What’s the value of retrospective statistical analysis. How can it be used in future cell-line development projects? Like many other companies, we are interested in defining (particularly in cell-line development) the design space that we’ve got to play with. For us, that means trying to understand the range for any particular attribute that we can match from a clone-screening point of view, and then which of those attributes are influenced by the bioprocess. The interplay between those things is key.

As we develop platforms and process-based knowledge, we can map out that design space. For a glycan, you can go from a Range A to Range B, and that could be matched with a particular host cell line. So we know that we have to screen a certain number of clones to fill that range.

What became apparent to us was that a point came at which, even if we analyzed many more clones, we wouldn’t see attributes outside of a certain range. That helped us think about this statistically and come up with a model to understand how many cells we need to analyze to find the design space of a given molecule.

Aside from speaking, why are you attending the BPI Conference? It is an opportunity to see what is going on in the industry at large. Many conferences I go to are fairly specific within an upstream area or even cell-line development alone. The BPI Conference attracts people from a spectrum of backgrounds, different companies (large and small), and also people from research-based organizations. I think that is a nice mix.

The other key part of this meeting is its attendance by vendors and suppliers. It is a great time to catch up with those companies and see what is happening at the cutting edge, see what new tools are being introduced, and bring that information back to Amgen.

>> Listen to the full interview

These interviews have been edited from transcripts for space and style.

Cell Culture and Upstream Processing Sessions

Tuesday, 27 October 2015

8:00–9:45 am Optimizing Interface with Discovery and Applying ‘Omics and Systems Biology

10:15–11:45 am Applying Novel Approaches and Tools to Cell-Line Engineering and Development

1:25–3:00 pm Integration of Process Analytics

Wednesday, 28 October 2015

10:30 am–12:00 pm Development and Production of New Modalities

1:40–3:15 pm Perfusion and Continuous Processing in Cell Culture

4:00–5:30 pm What is Driving Improvements in Efficiency, Productivity, and Timelines in Cell Culture

Thursday, 29 October 2015

8:00 am–12:00 pm High-Throughput Approaches to Process Development Impact of Process Conditions on Product Quality

2:00–5:00 pm Process Characterization, QbD, and Technology Transfers

Cell Culture and Upstream Articles from the Archives Online at www.bioprocessintl.com

Hemmerich J, Kensy F. Automation of Microbioreactors. September 2013.

Allen D. Upstream Chemistry Analysis in Cell-Based Process Development. September 2013.

Moncaubeig F. Simpler and More Efficient Viral Vaccine Manufacturing. October 2013.

Skaletsky E, Tucker E, Maggio E. Enhancing Antibody Production. November 2013.

Shimoni T, Moehrle V, Srinivasan V. Process Improvements Increase Production Capacity of a Legacy Product. November 2013.

Glauche F, et al. Design of Experiments for Fed-Batch Process Development in Shaken Cultures. January 2014.

Gazzano-Santoro H, et al. Ready-to-Use Cryopreserved Primary Cells. February 2014.

Szczypka M, et al. Single-Use Bioreactors and Microcarriers. March 2014.

Shimoni Y, et al. Qualification of Scale-Down Bioreactors: Validation of Process Changes in Commercial Production of Animal-Cell- Derived Products, Parts 1 and 2. May and June 2014.

Liu M, et al. Fed-Batch Cell Culture Process Development: Implementing a Novel Nutrient Additive for a Robust, High-Titer, Scalable Process. September 2014.

Pandiripally V, Lee J, Rader RA. A Novel Solid-Media E. coli Platform: Comparison with Standard Fermentation Processes. September 2014.

Sunil N, et al. Expansion of Human Mesenchymal Stem Cells: Using Microcarriers and Human Platelet Lysate. September 2014.

Lange I, Chhatre S, Zoro B. Reducing Timelines in Early Process Development: Using a Multiparametric Clone-Selection and Feed-Optimization Strategy. November 2014.

Janas M, et al. Perfusion’s Role in Maintenance of High-Density T-Cell Cultures. January 2015.

Simón M. Bioreactor Design for Adherent Cell Culture — The Bolt-On Bioreactor Project, Parts 1–4. January, April, May, and September 2015.

Wright B, et al. A Novel Seed-Train Process: Using High-Density Cell Banking, a Disposable Bioreactor, and Perfusion Technologies. March 2015.

Vanderlaan M, et al. Hamster Phospholipase B-Like 2 (PLBL2): A Host-Cell Protein Impurity in Therapeutic Monoclonal Antibodies Derived from Chinese Hamster Ovary Cells. April 2015.

Dhanasekharan K, et al. Rapid Development and Scale-Up of Biosimilar Trastuzumab: A Case Study of Integrated Cell Line and Process Development. April 2015.

Leave a Reply