[PRESENTATIONS] [OVERVIEW] [FACULTY] [REFERENCES]

Colleen M. Brophy, MD: Stress and Cardiovascular Disease at the Cellular and Molecular Level

I think I'd like to start off my talk with one of the comments that Chuck made when he was addressing surgical residents attending the fundamentals of surgical research course last year at the AAS as president of the AAS. And Chuck said that, "The foundation in your research should be 'What disease are you going to cure?'" Pretty profound. And that's a big way to think of your research. To a certain degree, though, that's how I started about seven years ago. And the disease that we were planning on curing was vasospasm. We ended up here looking at stress and I'm going to kind of tell the story of how our research has evolved over the past seven years.

[ Slide 01 ] What is the Off Signal?

We were interested in how smooth muscle cells function. ...and more specifically we asked the question, "How does it relax? What is the off signal?" These are typical physiologic responses of strips of vascular smooth muscle in a muscle bath. When you treat them with contractile agonists and they contract. There are two ways the muscles can relax. One is, remove the contractile agonist and they'll passively relax. The other is in the continued presence of the contractile agonist, and in this case thrombin, adding agents which increase cyclic nucleotide concentrations inside cells cause the muscles to relax. In this case, the NO donor sodium nitroprusside or in this case, the adenylate cyclase activator, forskolin. So we asked the question "what is the off signal: how do muscles relax?"

[ Slide 02 ] Substrate Proteins

Now it's appropriate that Tim Billiar came in front of me, not just because he's in alphabetical order, he also started a bit of this discussion in the concept that nitric oxide causes smooth muscle relaxation. And really this is one of the fundamental concepts in vascular biology. That endothelial cells produce a diffusible substance that actually goes right across the cell membrane, activates guanylate cyclase, and then most slides at that point have this big arrow that goes right to relaxation. And that's the point we're interested in. That big arrow. But we know there's another step in that process. And that is cyclic GMP and cyclic AMP respectively which activate specific protein kinases, PKA and PKG. Protein kinases are enzymes that take a phosphate, add it to specific substrate protein and change the function of those proteins. So we asked the question "What are the specific substrate proteins that are activated, that are phosphorylated, when muscles relax?"

[ Slide 03 ] Model of Vasospasm

In order to answer that question we needed a model. We needed a model in which muscles did not relax or a model of vasospasm. And this was the fundamental observation in the laboratory that's been extremely reproducible, and that is while most muscles when treated with contractile agonists contract, followed by relaxing agents that relax, there's one muscle that's very unique and that's an umbilical artery smooth muscle obtained from postpartum umbilical cords. These muscles can be treated with milimolar concentrations of sodium nitroprusside and do not relax. Similarly, treated with high concentrations of forskolin they don't relax. This gave us a model in which to look at situations in which muscles relax normally, and when muscles don't relax. We used another tool, and that is we radiolabeled all of the phosphate pools inside cells with p32, treated these muscles with sodium nitroprusside or forskolin, homogenized the muscles and then separated the proteins. We then separated the proteins in two dimensions, one based their charge, the other based on their relative mobility through an acrylamide gel and then applied these gels to x-ray film we found that in this muscle that relaxed normally, there was an increase in the phosphorylation of two 20 kD phosphoproteins. And we did not see these proteins getting phosphorylated in the umbilical arteries.

[ Slide 04 ] Identification of PP20

So the next year or two's worth of work, summarized on the slide, and we obtained some sequence from those proteins, and when we put the sequence through the computer programs, it was highly homologous to alpha-B-crystalline.

[ Slide 05 ] Heat Shock Protein

I didn't know what a heat shock protein was or what they did when we got that answer. But we quickly ran to the library to find out what they were, and we found out that heat shock proteins were initially discovered in the late '60s in Drosophila larvae. If you heated up the larvae, there was an decreased overall gene expression and increase in the expression of a group of proteins that were termed heat shock proteins. There's been a large body of work that has subsequently had been done in cultured cells that has demonstrated that a variety of stressors: osmotic stress, chemical stress, heat stress will cause an induction of these stress proteins. Some of them are also expressed constitutively, which suggests that they may have a role in normal cellular functions. One of the underlying themes associated with heat shock proteins is they function as molecular chaperones or assist in protein-protein interactions, and protein folding.

[ Slide 06 ] PP20 = HSP20

When we were doing the literature search we found that our sequence had complete identity with another small heat shock protein that had recently been identified and published, and that is P20. We talked to the investigator in Japan, Dr. Kato, and asked him if he would be willing to establish a collaboration with us because we thought we had a functional relevance to his protein. He sent us antibodies and our protein was immunoreactive with his antibodies. He also sent us purified protein, which could be phosphorylated in vitro with PKA and PKG. And we've published those results in collaboration with him.

[ Slide 07 ] HSP20 in Vasorelaxation - 1

So we think that this particular small heat shock protein may be an important mediator downstream in regulating smooth muscle relaxation. Thus, least part of the answer to that big arrow may lead to HSP20.

[ Slide 08 ] HSP20 in Vasorelaxation - 2

We've done subsequent studies that show that not only is HSP20 phosphorylated during pharmacologic activation of cyclic nucleotide signaling pathways in intact strips of muscle, but it's also phosphorylated during endothelial dependent vasorelaxation induced by flow or acetylcholine.It is not phosphorylated in this particular muscle that's refractory to relaxation. There may be changes in the macromolecular associations of this protein when it's phosphorylated. Finally HSP20 is actin associated. So it is associated with proteins at the level of the contractile machinery.

[ Slide 09 ] The Effect of Synthetic Peptides on Contractile Responses

This is correlative data that doesn't really prove a direct association between vasorelaxation and this protein, so what we did is we identified where on the molecule it was phosphorylated. We then produced phospho-peptide analogs and inhibitors, introduced them into permeabilized cells, and showed that we could inhibit agonist-induced contraction with the introduction of a phosphorylated peptide analogue. We think this has tremendous potential therapeutic applications and are currently working with a chemist to produce a permeable form of this peptide and hope to use it to treat vasospasm in diseases such as subarachnoid hemorrhage.

[ Slide 10 ] Stress

Well, the title of my talk was "Stress and Cardiovascular Disease". Now this is something that is relatively well known by the people in this audience and we know that stress can lead to a variety of vascular diseases, such as hypertension and atherosclerosis. And I've talked about one stress protein that is constitutively present in smooth muscle. Well there's some evidence that there's another stress protein that's induced by stress, HSP27, another small heat shock protein. It's associated with HSP20, that's how HSP20 was first found. It was a byproduct of the purification of HSP27. Muscles treated to contract have an increase in the phosphorylation of this protein and others have implicated HSP27 in contraction.

[ Slide 11 ] Arsenite Treatment of BCA

So we said, well let's see what happens when we stress intact strips of vascular smooth muscle. And we chose, in this case, arsenite treatment but we found similar results with heat shock. And we found that if you stress the muscle with arsenite there was really no change in the basal tone. It didn't contract. There was no change in the magnitude of the contractile responses to high potassium or to serotonin. However there was significantly impaired relaxation to sodium nitroprusside and this was abrogated by a kinase inhibitor that inhibited the pathway that leads to the phosphorylation of HSP27.

[ Slide 12 ] Arsenite Increases HSP27 Phosphorylation

There were increases in the phosphorylation of HSP27 in response to stress as we demonstrated using phosphorylation state-specific antibodies that we have developed in our laboratory.

[ Slide 13 ] AS TX Inhibits HSP20 Phosphorylation

And there were also decreases in the cyclic nucleotide dependent phosphorylation of our particular stress protein, HSP20.

[ Slide 14 ] Current Hypothesis

So our current hypothesis is that HSP20 and HSP27 coordinately regulate vascular tone, probably by direct interactions with specific elements of the contractile machinery. And the way we envisioned this model, driving back from a heat shock protein symposium at Wood's Hole this past summer, was that it's kind of like driving out of the rental car place. That the muscles can contract, but there's some latch that holds that contraction in place and maybe HSP27 is that latch. And the phosphorylation of HSP20 releases that latch.

[ Slide 15 ] Future Questions

In the future we plan to try to understand whether stress of intact organisms will lead to impaired relaxation and our preliminary data suggests very much "yes". Are these small heat shock proteins involved not only in acute changes in vessel caliber, but also chronic changes that are associated with vascular remodeling and our preliminary data also suggests "yes". And as I indicated, what are the mechanisms by which these small heat shock proteins may be modulating the specific elements of the contractile machinery and modulating vasomotor tone? None of this work takes place in a vacuum and I'd like to thank my entire lab group for all their wonderful efforts and also my husband for his continued support. And this is for Chuck, and I'm not sure it's going to be safe for my little girls to spend much time with his little boys when they get older. They're essentially the same age, but they are darling wonderful little kids and kind of the joy of my life. Thank you very much.

Page last modified on April 3, 2002 .