A multinational team of scientists has reported developing a protein array to detect the presence, severity, and response to therapy for mouse autoimmune encepha-lomyelitis, a model for human multiple sclerosis.
The methods used in the experiment, which are described in a September Nature Biotechnology article published online Aug. 10, could have far-reaching applications in the development of diagnostics and therapeutics for a variety of immune-related disorders, according to the lead author and other scientists involved in the protein microarray field.
The paper describes how the researchers used a robotic plate contact printer modeled after the system developed by Pat Brown, a microarray specialist at Stanford, to create protein arrays consisting of both mouse and human proteins that compose the myelin proteome. “These were all proteins and peptides [for which] people had evidence that they were in the myelin sheath and could be targets — they were all the usual suspects,” lead author William Robinson, an assistant professor of medicine at Stanford, told ProteoMonitor.
The protein arrays were then soaked with mouse serum taken from mice induced with acute EAE, and were analyzed to find which antigens had a matching antibody response in the serum. The researchers found that those mice that had more severe disease with increased relapse rates also had antibodies to a greater variety of antigens — in other words, diversity corresponded with severity. This finding pointed to obvious potential clinical applications, according to Robinson. “I’m a rheumatologist, and when we see patients coming in with first onset arthritis, we want to know who’s going to develop rheumatoid arthritis, and who’s going to have just a single acute flair,” Robinson said. “Profiling could help us sort out which patients are likely to progress to RA as opposed to having a benign course — and that influences how you treat them.”
In addition to arrays that test for MS and arthritis, Robinson said his group is developing connective tissue disease arrays to study lupus and related diseases, and arrays of the HIV proteome to study responses in macaques to experimental HIV vaccines. The latter study is described in a paper set to appear soon in the Journal of Virology. Essentially any disease involving “potentially chronic infections with viruses and bacteria” could be studied with similar arrays, according to Robinson.
Mike Snyder, a protein array specialist at Yale who says he has recently submitted a similar paper to Nature Biotechnology describing an antibody-detecting array composed of the yeast proteome, said that Robinson’s array approach to predicting the severity of autoimmune disease is both powerful and marketable. “The kinds of reactivity [observed] may ultimately lead to typing subclasses of autoimmune disease. I think it’s a very powerful approach to that kind of thing,” Snyder said. “It wouldn’t take much to scale [the array methods] up and turn it into a commercial assay ... It should be easy to transfer the same technology to humans.”
In addition to predicting disease course, Robinson’s paper, entitled “Protein microarrays guide tolerizing DNA vaccine treatment of autoimmune encephalomyelitis,” also describes two other applications of his arrays: using the results to design tolerizing DNA-, peptide-, or protein-based vaccines to the disease — and “perhaps to select which patients are likely to respond to a given therapy that’s available” — and monitoring the patient’s response to therapy once it has been given. Both applications were fleshed out in the mouse experiment. As described in the paper, Robinson’s group tested a DNA-based tolerizing vaccine based on his array results, then verified that the vaccine had an effect by repeating the array analysis after treatment. Robinson discovered that those mice treated with the vaccine had less diversity in their antibody response compared to those that were not treated.
Bayhill Therapeutics of Palo Alto, a spinoff company from Stanford that Robinson co-founded, is working on a DNA therapeutic for MS based on Robinson’s work, and has announced it will enter clinical studies on an antigen-specific ther-apy for MS in the first half of 2004.
Although he chose protein arrays over DNA arrays for the initial experiment, Robinson went back to DNA when it came time to make a therapeutic, for mostly practical reasons. “It’s a lot of work to make a recombinant protein,” Robinson said. “It’s much easier to just grow up the plasmid DNA.”