克里斯托弗休斯：在体外血管生成的研究模型（面试）

Hi, my name's Dr.Christopher Hughes. I'm a professor at the University of California Irvine, and my lab works on angiogenesis. And we're particularly interested in the genes that regulate the process of angiogenesis.

We have optimized an in vitro assays based on the work of Nelson and, and in this assay we can look at all the different processes that are going on during angiogenesis, we coat tedex beads, small beads with endothelial cells. Then we embed these in fibrin gels. We provide growth factors.

And over the course of several days, the endothelial cells sprout from the beads. They migrate, they align, they proliferate, they make tubes, they branch and they form anastomosis. So that by about 10 days we have a complete capillary like Network in culture.

Well, some of the problems with envivo and angiogenesis Assays, for example, those that are done in mice is that it's much, it's not very easy to manipulate the system. It's hard to visualize when you've got this going on inside an animal, you, you can't really see angiogenesis happening. And if you want to manipulate the system in our in vitro cultures, it's very easy to genetically manipulate the endothelial cells, turn genes on and turn them off, and look at the effects of that on sprouting angiogenesis.

And it's much harder to do that in the whole animal. You have to make knockouts or knockins. So those are the two major advantages.

It's, it's easy to manipulate And it's easy to visualize. So over the years, there have been A number of different in vitro agenesis assays developed and they, they all have their advantages and disadvantages. And in particular a well known one is the matrix gel assay where endothelial cells are placed on top of this extracellular matrix called matrigel.

And the problem there is that the, the endothelial cells make cord like structures, but they never really make proper aluminized vessels. You don't see sprouting, you don't see anastomosis. There are collagen gel models, which can be good.

There are many of those You don't see proliferation of the endothelial cells. You don't tend to see anastomosis. The fibrin gel assay that we have developed is, is based on a model by Nelson Kar published in the mid nineties.

And we've now dramatically improved that such that we can see all the stages of angiogenesis. We can watch the initial sprouting, the migration. We see proliferation of cells, we see lumen formation, we get branching, and we finally get anastomosis to form a complete vascular network in vitro.

So this assay really is, we believe that the, the best in vitro system around if you want to look at all those different stages of angiogenesis. Another advantage that we have with this in vitro assay is that we can do laser capture, micro dissection, we can spin down the culture so it becomes very flat, and then use the laser capture micro dissection microscope. We can actually take out individual cells and in particular we are interested in the tip cells.

And so we are cutting those out and we can look at gene expression, particularly in the tip cells versus the trunk cells, the cells that are actually making the lumen. And our hypothesis is that there will be differential gene expression between those two populations of cells. We can look at that, it would be pretty much impossible to do that in an in vivo system.

So we think that's another major advantage of using this in vitro Assay. The early work that was done in this, as I mentioned by Nelson Recon, they used microvascular endothelial cells. And there's a lot of thought that microvascular endothelial cells, the cells you would want to use for an assay like this because in vivo, it's the microvascular EC that's really forming new vessels.

We decided to try and do this with ve. These are human umbilical vein endothelial cells for a couple of reasons. One of which is that these cells are very easy to obtain.

You can go to the hospital, you can get umbilical cords from the delivery suites, and you can isolate the cells very easily. And also these cells have been used in culture for many years. They're very well characterized.

And so we set up this assay and optimized it so that we could use these cells for those reasons. And interestingly, it turns out that vex, these large vessel endothelial cells appear to take on the form of microvascular endothelial cells and culture. They lose a lot of genes that are normally expressed by large vessel ec and they begin to express genes that are normally associated with microvascular endothelial cells.

And they are perfectly capable of forming capillary like sprouts of making lumens and making what all the world look like real capillaries in culture. So they're easy to use, ve are easy to use, and they do what we want them to do. They sprout and, and they make proper aluminized vessels.

When we want to quantitate this assay, there are really a couple of ways we can go. We can look at really quantitative criteria such as the number of sprouts, the length of sprouts, the diameter of the sprouts, the number of branch points, the number of anastomosis, but we also look at qualitative criteria. And so we can look at are they forming lumens?

And that's really probably what we're most interested in. In a lot of the assays we do is their proper lumen formation. And that's really a more of a, a qualitative assessment, although as the vessels get larger, one can actually Measure the diameter of those lumen as well.

So we use fibroblasts In these cultures. We embed the beads into the fibrin gel and we plate fibroblasts on top of the gel. The fibroblasts are absolutely essential for proper sprouting and lumen formation.

In the absence of fibroblasts, we get some sprouting. However, the endothelial cells tend not to stay attached to each other. What happens is they, they migrate away from each other.

And so any structure that you're developing on this sprout tends to fall apart. And after a few days, the the endothelial cells undergo apoptosis. So it looks like fibroblasts are making a factors that are essential for angiogenesis, at least in this in vitro system.

And we've been very Interested in knowing what those factors are. What we've found is that The fibroblasts secrete something into the medium, so we can take condition medium from the fibroblasts and that will support proper sprouting and lumen formation. So we're definitely dealing with a soluble factor.

So obviously we're very interested to know what it is that the fibroblasts are making that's so important for angiogenesis in this system. We've tried all the usual suspects, VEGF, basic fg F are, are in the culture, adding more does does not substitute for fibroblasts. We've also tried angiopoietin one, TGF beta, H, GF, et cetera, et cetera.

So far we've not found the vital factor. So what we're doing now is purifying factors from the condition medium from fibroblasts. So we're putting the condition medium over various columns, cation, exchange, size exclusion, and we're getting fractions of that that have activity.

And we're now following that activity and we're taking those fractions, finding which ones are active, and from there we can run those on gels, make sure that we've got just a, a relatively few number of bands on the gel and then we can cut those out and get them sequenced by mass spec. And, and that's giving us a lot of interesting days. We have quite a few candidate genes now that we are testing individually to see if they are Actually the factor that we're, that we're looking for.

So we have tested a lot of Known angiogenic factors and we know that veg GF and basic FGF are actually essential for, for the proper sprouting in this assay. But they're not the whole story. There are derived factors that are important as, as I mentioned, and one way we've looked at this is we've taken tumor cells, which are highly angiogenic in vivo, they generate large tumors in mice and they're highly vascularized.

And we've looked at these tumor cells and they express very high levels of of vegf and we've put those into our cultures. So we've substituted for the fibroblasts with these tumor cells and they do not support robust angiogenesis in this assay. And that, that's very interesting.

What that suggests is, and this has been suggested by other people as well, is that there's a crosstalk between the the tumor, the the S stroma, which is the fibroblasts and other support cells in the tissue and the endothelial cells. And it may well be that the tumor cells produce some factors that work directly on the endothelial cells, but they also produce others that are stimulating the fibroblasts and other stromal cells to make factors that then also support robust Angiogenesis.