I’m getting serious about Thinking Out Loud this fall. And I’d like to get your feedback. Sharing ideas is important, and the work we’re doing – translating ideas into medicines, has a huge impact on human health.
If you want to hear me postulate in person, see the list of conferences scheduled for September here. Or, follow the links in this email back to the blog, and if you want to write me directly, please do that too.
Here are the latest ideas that I’d like to share.
I think that immunogenicity – and tolerance for that matter, is the sum of a set of signals that can be directly linked to protein sequence. Both positive signals (T effector epitopes) and negative signals (T reg epitopes) may be present. The immune system sums the signals and produces a local output. The best analogy is the rheostat – which similarly sums analog signals and produces a single output. Why so sure? Read this for example.
Negative signals are almost as important as positive ones. We’ve come to find that Tregitopes are negative signals (induce regulatory T cells) and the evidence that they are present is solid. We’ve been working away on validating the observation we first published in 2008 for quite a while. Tregitopes definitely impact protein therapeutics, and immunogenicity, and they demand your attention.
T effector epitopes act as positive signals. The good news? They can be predicted using in silico tools with a high degree of accuracy, but which tools you use determine the accuracy of your prediction. If you use off the shelf (on line) tools, you have to take the similarity between HLA pockets into account (see Greenbaum and Sydney, 2011). Proceed with caution! This similarity between HLA pockets means that predicting for a population based on HLA nomenclature based on the assumption that the names of HLA molecules reflect different structure, will lead to over-prediction for some HLA pockets.
Measuring T cell response is an art. I am sure most of you know that, but I think it’s important to emphasize that the phenotype of cells that respond to proteins is important. Simply measuring proliferation – although cheap – is not enough. Tregs proliferate. So do T effectors. Will proliferation assays discriminate and give you an accurate picture of immunogenicity? No they will not.
And finally, my favorite news – T cells talk directly to each other. Have you taught a class recently and noticed that your students are texting to each other rather than raising their hands? That’s what goes on in the immune system. Sometimes text messages (cytokines) are all that is required, but occassionally the T cells actually pass notes – as in pieces of cell membrane, with antigens included, to each other.
I’m full of ideas and want to hear your feedback. The results of these discussions and the output of “thinking out loud” will be available on my blog this fall. Link to my blogs are at the left – including a recent post on ‘naturally deimmunized’ proteins (hirudin).
These are just a few of the ideas I will be exploring . . .
Follow my blog and talk back to me on line. I look forward to our discussions.
Annie De Groot, M.D. CEO/CSO, EpiVax Inc.
Providence, Rhode Island