OCRP 2012 Investigator Vignette

Title: Targeting Cell Surface Proteins in Molecular Photoacoustic Imaging to Detect Ovarian Cancer Early

Investigator: Charles Drescher, MD; Fred Hutchinson Cancer Research Center

I have a grant currently through the DoD OCRP looking at targeting cell surface proteins for photoacoustics to detect ovarian cancer early. And this is motivated by the understanding that the conventional imaging tests that we have are really not sufficient to detect ovarian cancer in an early curable phase. Perhaps one of the most promising imaging strategies for ovarian cancer early detection is photoacoustic imaging. And this approach combines a molecular contrast agent with a hybrid imaging probe that uses nanosecond pulses of laser light to excite tissue based on different physical characteristics. And those excited tissues then generate sound waves of particular ranges which a traditional ultrasound probe can then use to kind of reconstruct into a 3D image of the tissue.

And the current research funding that I have focuses on only one step in that process and that's basically to try to identify the proteins that are expressed on a surface of an ovarian cancer cell so that we can then target them with a molecular imaging probe. We have adapted a sort of a new type of chemistry to be able to very selectively and specifically bind what are known as N-linked glycoproteins. And glycoproteins are proteins that have sugar attached to them and they're very enriched on the surface of cells and particularly on the surface of tumor cells. So we adapted a particular chemical strategy that allows us to essentially bind these glycoproteins on intact viable ovarian cancer cells. And once we bind them, then we can use different affinity capture methods to basically strip them off of the cells. And then we subject them to mass spectrometry so that we can then go ahead and identify them.

This is just a summary of where we are with these analyses. And so what you can see is an OVCAR3, which is a cell line, we essentially identified a total of 275 proteins that contain the motif. And then we went on using bioinformatic methods to try to determine whether in fact these proteins are cell surface proteins. And we used either gene ontology annotation and a number of prediction algorithms and then, in addition, literature searches based on experimental data. And we were able to confirm that of these 275 proteins that were identified on OVCAR3 cell lines that contain the motif, 235 of them are in fact cell surface proteins.

If you contrast that to the peptides that don't contain this motif, you can see where there's 196 and of those really only 14 were ultimately confirmed to be cell surface proteins. So you can see that our method is very highly selective for identifying the proteins of interest. And we subsequently prioritized these proteins for those that we think might be excellent targets for molecular imaging probes and then gone on to validate them with more kind of standard immunohistochemical means just looking to see are these proteins expressed in ovarian cancer cells from patients' tissue samples that are collected in surgery. And in fact, so far eight out of the nine that we've selected have actually validated.

So this is milk fat globulin 8 and what you can see here is very specific staining in ovarian cancer cells. And if you look carefully you can see that there are some cytoplasts that are staining but this sort of deep line in these pictures demonstrate that there's abundant membrane expression. This is another one of our candidates, TAC-STD2. This is a protein that's involved in cell signaling, may play a role in invasion and proliferation of ovary cancer, and again we see abundant staining on the membrane and cytoplasmically in serious ovary cancer.

So we feel like we're off to a good start and we're optimistic that we, you know, will ultimately find some markers that can be translated really hopefully after some animal studies pretty rapidly into the clinic.