Era of Hope 2011 Investigator Vignette

Title: Going Fishing with RNA Binding Proteins

Investigator: Ulus Atasoy, MD, MA; University of Missouri

Post-transcriptional gene regulation plays a very significant but underappreciated role in breast cancer. The RNA immunoprecipitation techniques which identify the small fish, genes whose RNA levels don’t go up, can be powerful tools to identify novel breast cancer targets. We work with post-transcriptional gene regulation. What that means is we’re interested in what RNA binding proteins and micro-RNAs are doing to the mRNA that’s made from the DNA blueprint in the cell. That’s transcription. And post-transcription by definition is everything that happens to the mRNA blueprint as it’s made and exported from the nucleus into the cytoplasm.

We work with an RNA binding protein called HuR. it stabilizes many of the mRNAs and about 10 years ago Hanahan and Weinberg had a seminal publication where they asked the question what properties do malignantly transformed cells have to acquire to become cancerous, and they came up with about six of them. And literature indicated that HuR was controlling genes in five out of the six different areas of the acquired capabilities. So the hypothesis that we wanted to test is that HuR is a tumor maintenance gene that regardless of how the breast cancer develops or other cancers as well, the cell still has to acquire those traits, those tools in its toolbox so to speak to enable it to become a cancer cell, to become a self-sufficient tumor, and then metastasize at distant sites. So what we did in our DoD Award was we did these HuR immunoprecipitations from an estrogen receptor negative cell line, the MDA-MB-231 and an estrogen receptor ER positive cell line MCF7 and we asked if we can identify different classes of genes that HuR may be regulating.

So this is an immunoprecipitation done three times with matching controls, hybridized against 12 different arrays and each one of these arrays has about 47,000 mRNAs on it. So this is a statistical compilation of lots of data and so this is a log-2 scale so it goes up by log-2. With our control, many of the genes do not come down with HuR, which is what you would expect. It means that it’s not nonspecific, so in the upper left quadrant here these are mRNAs that are associated with HuR in the triple negative breast cancer, the 231s.

Down here in the lower right quadrant are mRNAs that are associated with HuR in the estrogen receptor positive. And the ones here are mRNAs that are associated with HuR in both cell lines. So are these new genes, old genes; were these genes previously known? So you can classify them into arbitrarily three classes. The Class 1, basically old friends that we knew HuR regulated; we knew that yes, HuR regulated—[Inaudible], Prothymosin, HIF1-alpha, other cancer genes. Then Class 2 would be known cancer targets that we didn't know HuR regulated—new friends.

But the third class I think is the most interesting class. These are genes that do not have any known cancer function but they definitely come down in the HuR pellet, so we’re wondering whether HuR is actually playing a role in regulating these genes and whether they may be novel HuR targets. We picked two of these genes called CD9—this is a tetraspanin molecule; it’s involved in cell-to-cell adhesion so it determines invasiveness and potentially metastasis of the breast cancer. And we picked another one, Calmodulin2 which is involved in calcium metabolism in the cell. Both CD9 and Calmodulin were in this quadrant; they were both highly expressed in both breast cancer cell types and they came down greatly enriched in the HuR pellet. And so and the rest of the figures here, we basically systematically verified using several different approaches that the HuR protein really does interact with the CD9 and the Calmodulin 2 mRNA.

Now the power of this approach and the innovation we believe lies in the fact that we have identified genes that have known cancer function but we didn't know were regulated by HuR and we’ve kind of plugged some of these genes into a modified Hanahan—Weinberg acquired capabilities model. So these ones in bold are new putative HuR targets that we know have cancer relevance that we identified from this approach. But as I intimated earlier, I think the more exciting approach is those genes that are associated with HuR and HuR may be regulating them but we have no idea whether they have any role in cancer. So these may be novel target genes.

We’re talking about one RNA binding protein, HuR, which is a stabilizer. There are other RNA binding proteins, most of them actually—that are de-stabilizers, and people speculate there may be upwards of 3,000 RNA binding proteins in human cells. Only about two to three dozen have been cloned and are actively being studied. So this is the RNA binding portion of the equation; then you have the micro-RNA binding portion of the equation and it’s clear from the emerging evidence that RNA binding proteins act in concert with micro-RNAs to affect the stability and translation of their mRNA targets.

So what I’m saying is we should look at both RNA binding proteins and micro-RNAs to get a better understanding of post-transcriptional gene regulation in breast cancer.