Taking hands-on methods out of the process is crucial, because 96-well cloning, protein expression, and assay development is laborious and expensive
Patrick McGee
Senior Editor
Most biotechnology products are proteins that must be prepared in a purified form at high volumes, and the drug discovery process also relies heavily on protein and antibody purification. Over the last several years, a great deal of effort has been directed toward automating and miniaturizing the purification process, and that work continues. “I think the quality of our work has increased success rates by screening more proteins. The number of proteins that we deliver in the end is probably not that much increased, but we have a much higher success rate, so we don’t have to drop projects relatively early, and there’s a chance that a target survives for a later stage of drug discovery. This is greatly enhanced by having tools available that allow you to simultaneously screen for a lot of different options,” says Stefan Schmidt, PhD, team leader of the Biotech Laboratory at AstraZeneca, Södertälje, Sweden.
Schmidt’s lab supplies all AstraZeneca units that require large amounts of protein, including researchers in crystallography, assay development, and high-throughput screening. “We are producing up to several hundred milligrams of protein, 50 to 200 mg recombinantly, in fermentation. Some of these proteins are the typical drug targets that you can imagine, and a very small part of that are antibodies, mainly for research purposes.”
Schmidt says his lab is now focused on automating steps in the purification process, which could include screening for the optimal culture conditions when working on a small scale. They are also investing in purifying larger amounts in parallel, not sequentially. One machine they are using is the BioOptix 10 from Teledyne Isco Inc., Lincoln, Neb., a 10-channel parallel purification system that includes a high-capacity fraction collector (20 to 60 fractions per sample). Its delivery and detection module contains 10 independently controlled pumps that can be programmed with different gradient conditions for rapid scouting. “This is something that is an elemental part of our strategy to work more in parallel and automate more to allow our customers in structural chemistry to have earlier access to proteins for the initial crystallization trials,” Schmidt says.
He adds that the research facility in Södertälje and another at Alderley Park, UK, have been testing the functionality of the 5100 Automated Lab-on-a-Chip Platform (ALP) from Agilent Technologies. The ALP is an expansion of Agilent’s 2100 Bioanalyzer, a microfluidics-based platform for the analysis of DNA, RNA, proteins, and cells, and can be used in place of labor-intensive gel electrophoresis techniques. The Protein 200 HT-2 assay for the ALP is for identification, sizing and quantitation of proteins from 14 to 200 kD; its two capillaries can analyze two samples in parallel, allowing up to 1,000 samples per day to be processed. Schmidt says the ALP is unique because it allows researchers to replace SDS-polyacrylamide gel electrophoresis procedures in the lab. “Doing this saves a lot of time and is a lot more precise. This is an essential part for all larger scale and high-throughput methods around protein purification.”
Academic automation
Some academic laboratories are automating processes as well. Tallamraju V. S. Murthy, PhD, a research associate at the Harvard Institute of Proteomics (HIP), Cambridge, Mass., has been working on automating high-throughput protein purification pipelines for biochemical and immunoassays. The process was automated because 96-well cloning, protein expression and assay development is laborious, prone to errors, expensive, and requires normalization, Murthy says. The proteins are purified using two well-known bacterial systems, an in vitro cell-free system and a cell-based system.
In bacterial cell-free protein synthesis, extraneously added DNA is transcribed and translated in vitro to produce protein. Over the last few years, researchers have designed protocols to generate highly synthetic bacterial cell extracts capable of producing hundreds of micrograms of protein, Murthy says. The cell-free approach has several advantages over cell-based systems in the expression of toxic proteins, labeling of amino acids for structural studies, and expression of mutants of a protein. Cell-free synthesis requires several ingredients such as tRNA, amino acids, and nucleotides, and commercial extracts for protein synthesis are often expensive and cannot be modified in most cases.
The HIP lab in which Murthy works adopted strategies for preparing bacterial cell extracts for protein synthesis developed by Shigeyuki Yokoyama, D.Sci., group director of the advanced protein crystallography research group at the Riken Harima Institute, Hyogo, Japan. The HIP lab chose 63 bacterial genes for protein expression and found that 50 could be successfully expressed and purified under denaturing conditions. They already have a high-throughput cell-based system in place and are now automating the cell-free system to express human and viral proteins. The automated purification and analysis platform was developed using the Biomek FX Laboratory Automation Workstation from Beckman Coulter Inc., Fullerton, Calif., and the LabChip 90 Electrophoresis System from Caliper Life Sciences Inc., Hopkinton, Mass. Cell-free systems are commercially available from companies such as Roche Applied Science and Invitrogen, as well as from academic laboratories in Japan.
“When you have a lot of proteins in [a] high-throughput [environment], the cell-free system has much more of an advantage because you already have a bacterial cell extract and you just add your DNA to the bacterial extract and the protein pops out, so it’s very simple to operate,” Murthy says. Although the cell-free system can produce proteins in two to three hours and has a higher throughput, the protein yield and purity are lower than that produced in a cell-based system. The cell-based approach, however, is time consuming and lower throughput. “With our experience in protein chemistry and the available instrumentation, we have developed robust pipelines for protein purification and purified several hundred proteins,” Murthy says, adding that a 90% correlation of expression and purification was observed between the systems. The cell-free system can be used for rapid expression in high-throughput assays requiring microgram levels of proteins with about 70% purity, and for screening of soluble proteins. The 96-well cell-based system can be used for high-throughput assays requiring tens of micrograms of proteins with about 90% purity. “We are currently using these proteins for target discovery and to globally understand the biological networks in collaboration with other laboratories. Moreover, the information is also being used to understand the parameters that govern bacterial expression of proteins.”
Vendors responding
Vendors have been responding to the need for better purification tools and techniques. Late last year, Pall Corp., East Hills, N.Y., introduced three new protein purification kits, the Enchant Life Sciences Kit for albumin depletion and two Enchant Life Sciences Kits for immunoglobulin G (IgG) purification. The kits provide the first step in isolating new drug targets by removing these abundant proteins from human and animal-derived serum. The all-inclusive kits include the protocol, purification columns, and necessary buffers needed to uncover small, low-abundant biomarkers.
Albumin and IgG comprise about 80% of the total protein content in human serum, so there is a clear need for diagnostic tools to rapidly deplete these unwanted proteins. The albumin kit allows researchers to remove albumin from samples in 10 minutes using five steps. The IgG purification kits are reusable gravity-based columns with high binding capacities for the effective purification or depletion of IgG; they can be used to purify a variety of immunoglobulin molecules and isotypes from a broad range of species.
Pall also recently introduced a version of its membrane chromatography for endotoxin removal. The Mustang Q ion exchange membrane differs from earlier membranes due to a slightly different membrane base, tighter pore size, and chemistry that relies more on a dual mode chromatographic separation as opposed to a single mode separation as in straight ion exchange. “This operates on both a hydrophobic interaction as well as an ion exchange interaction, and that is a benefit for endotoxin, because endotoxin has both of those chemistries hanging off of it,” says John Jenco, PhD, senior staff scientist and chromatography applications manager at Pall.
Last March, Pall announced that a collaborator, BioMarin Pharmaceutical Inc., Novato, Calif., had become the first company to receive licensure from the US Food and Drug Administration and the European Medicines Evaluation Agency for a drug product using a Pall dual membrane-based process. The new purification process eliminated the need for DNA testing and enhanced removal of potential virus contaminants from protein-based drugs, the company said at the time. The step that incorporated the Mustang Q was able to remove DNA to more than 100 million-fold below the level of detection. This allowed BioMarin to ensure the purity of Aldurazyme (laronidase), an enzyme replacement therapy for the treatment of mucopolysaccharidosis I, a genetic disorder. Following testing and validation of this process step, the company is exempt from batch-release DNA testing.
Jenco says these new technologies allow researchers to do things in a more timely fashion because they help improve processing speed and robustness. “There’s no point in running a process if it’s not going to give you a decent amount of purified material at the end of the day. . . . We offer our membrane chromatography devices in 96-well plate format so you can basically set up a number of conditions of pH and conductivity for both binding and elution, and set it up on things like robotics to rapidly screen process development.”
Scientists at Promega Corp., Madison, Wis., automated high-throughput protein purification by using the company’s MagneHis Protein Purification System in conjunction with the Biomek FX and Biomek 2000, which are capable of processing 96-sample formats, moving plates, and manipulating liquid reagents throughout a procedure. MagneHis is designed to provide a reliable method for the purification of polyhistidine-tagged, expressed proteins. Paramagnetic precharged nickel particles are used to isolate histidine (His)-tagged protein directly from a crude cell lysate. With a tube format, His-tagged protein can be purified on a small scale using less than 1 mL of culture or on a large scale using more than three liters of culture. MagneHis was automated for the rapid purification of polyhistidine-tagged proteins on several workstations; the MagneHis cell lysis reagent does away with the need for manual intervention because there are no sonication or centrifugation steps during the purification process. This allows for a simple, easily automated process, chemistry that can be scaled up for larger sample volumes, and, most importantly, high-quality purified proteins.
Fractionating proteins
Other vendors have taken a different approach to sorting out proteins. Proteome Lab products from Beckman Coulter Inc. focus on fractionation as opposed to purification of proteins. The company’s PF 2D Protein Fractionation System is a chromatographic approach that includes improved detection of low-abundance species, membrane or hydrophobic proteins, and low molecular weight proteins, as well as a contamination-free liquid flow path and simple automation. The PF 2D is designed to effectively resolve the thousands of proteins present in cell lysates, and detailed protein maps can be constructed for easy comparison using the Proteome Lab Software Suite. It also uses chromatofocusing followed by nonporous reverse-phase chromatography to provide ultra-high resolution of proteins.
“Our system is fully automated and we supply it complete with chemistry, the columns, the buffers, and the instructions,” says John Hobbs, strategic planning manager in the company’s proteomics group. “We’re trying to produce a standard way to fractionate proteins so that people can actually compare their results with others because they’re using the same methodology and the same chemistry. That’s important in proteomics because typically people do want to know how their results compare with others.”
Filed Under: Drug Discovery