Researchers at biotech company Cell Signaling Technology have developed an antibody-production method that uses a combination of mass spec-based proteomics and next-generation sequencing to identify antigen-specific antibody sequences in the serum of immunized animals.
The approach, which the company has patented and which was detailed in a paper published this week in Nature Biotechnology, could prove a more efficient and effective technique for monoclonal antibody production, aiding development of agents for research and therapeutic purposes, Roberto Polakiewicz, CST's chief scientific officer and an author on the study, told ProteoMonitor.
As the researchers noted in their paper, the mammalian immune response generates a variety of antibodies via changes in the genome of B-cells, each of which produces a specific monoclonal antibody to the broad polyclonal response attacking the initiating infection.
Researchers and clinicians have used such polyclonal antibodies for over 100 years, Polakiewicz said. However, he noted, this has been done "without really knowing for sure the complexity and composition of the polyclonal antibody you are using."
The advent of monoclonal antibody technologies provided researchers with methods – such as B-cell immortalization, single-cell sorting, and phage display – for generating antigen-specific monoclonal antibodies. However, Polakiewicz said, these approaches also have their downsides.
Specifically, he said, because B-cells do not survive well in culture, "there is an inherent challenge of cell attrition … Most of these [B-cells] die when you try to immortalize them, so you essentially get a very limited view of the original [B cell] repertoire."
Additionally, he said, with single-cell methods "the fact that you culture those cells doesn't necessarily mean that the antibodies in circulation derived from those cells. Essentially, you are making a leap. The [cultured] cells can produce an antibody, but that doesn't mean it was really the antibody that was doing the job in circulation."
To address these challenges, the CST team turned to a proteomic method, using LC-MS/MS analysis on a Thermo Scientific LTQ Orbitrap Velos to identify the monoclonal components of the polyclonal antibodies circulating in the serum of infected animals. These monoclonal sequences could then be synthesized and cloned and the antibodies expressed and screened for specificity and activity.
The CST researchers' approach was not without its own challenges, however. In particular, they ran up against the fact that, because an individual animal's B-cell antibody repertoires are constantly changing in response to foreign antigens, no proteomic reference databases existed for these proteins.
This meant the researchers had to build their own reference databases, which they did using next-generation sequencing on a Roche 454 Life Sciences platform to sequence RNA from splenic B-cells taken from the animals used in the study. They then searched their LC-MS/MS data against this database using Sequest.
The high level of conservation between the sequences of antibody V-regions – which are responsible for antigen binding – also presented a challenge, Polakiewicz said, in that it made the project more analogous to a proteomic study of protein isoforms than of distinct gene products.
"On the one hand you have the very highly conserved peptides that belong to the framework of the antibody, and then you have the highly divergent sequences from the [complementarity determining regions]," he said. This meant that a conventional shotgun mass spec approach where proteins could be identified on the basis of a relatively small number of peptide matches was insufficient.
To solve this problem the researchers "used four different proteases [trypsin, pepsin, elastase, and chymotrypsin] to generate multiple fragments belonging to a given antibody," Polakiewicz said. "By using those fragments [generated by the four different proteases] you get a high level of coverage of the whole [V-region], and that's how we were confident that not just one peptide but a group of peptides covering the whole [V-region] were mapped to a given [antibody]."
To make their identifications with sufficient confidence, the researchers required identification of at least 12 unique peptides — greater than 65 percent overall coverage, and greater than 95 percent coverage of the hypervariable complementarity determining region 3.
Advances in mass spec technology – particularly the improved mass accuracy of newer machines – were also key to making the method work, he said. "There's been significant progress in instrumentation. You need [the method] to have the mass accuracy that's allowed by the newer instruments that wasn't available [in machines] a few years back."
In the Nature Biotechnology study the researchers used the technique to develop antibodies against the proteins PR A/B, pMET, Lin28A, Sox1, and phospho-p44/42. They provided performance details on the anti-PR A/B antibodies, noting that two anti-PR A/B monoclonals generated via their proteomic method "exhibited superior signal and specificity in western blot analysis and immunohistochemistry and also reacted specifically in flow cytometry and immunofluorescence assays where [a] polyclonal mixture did not."
Since submitting the paper for publication, the company has developed more antibodies using the method, Polakiewicz said, adding that, while the workflow is still being developed, it has the potential to be significantly faster than conventional monoclonal methods. While traditional hybridoma approaches typically take between seven and 10 weeks, the proteomic technique takes roughly three weeks, he said.
CST, which specializes in providing reagents and services for the study of cell signaling pathways and the analysis of protein post-translational modifications, doesn't yet offer antibodies produced using the technique, Polakiewicz said. He noted that at the moment it is exploring the technique as a platform for therapeutic applications.
The company currently has several agreements for pharma work, including deals with Ventana Medical Systems for antibodies for detecting epidermal growth factor receptor mutations and for detecting ALK rearrangements via immunohistochemistry as part of a companion diagnostic for determining which non-small cell lung cancer patients will respond to crizotinib, marketed by Pfizer as Xalkori.
Polakiewicz noted, however, that while CST has not "deployed this method as a routine [antibody] development method yet … it could be deployed as a routine method. There are certain things that need to be addressed, but it definitely could be used routinely."
Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.