Richard Harland (UC Berkeley) 2: The Cellular Basis of Gastrulation

My name is Richard Harland. And this is the second in the series of talks about Xenopus gastrulation and pattern formation. And in this short video, I hope to show that gastrulation, which is often thought of as a horrendously complex process, is actually fairly simple. You just have to take it apart and look at the individual cell behaviors. And there are quite a few different cell behaviors, but these all conspire to get the whole mechanism of gastrulation to happen. So, let's first start with this schematic which we saw from the first video. And so what we're going to discuss is all of the component movements that go into this complex process of gastrulation. So, we'll discuss, for example, the thinning of this roof, the crawling of cells across the blastocoel roof, and the other individual motions that contribute to the process. Okay, so here's a schematic which shows seven basic mechanisms of gastrulation. I'm only going to discuss six, but we'll take them in turn.

So, the first one is the epiboly, that is, the thinning and spreading of the blastocoel roof, so, all of this blue tissue. Because the blue tissue has to spread and cover the whole embryo. And recently, there's been a lot of progress made in understanding how this may occur. So, epiboly consists of this initial gathering of cells, and these are freeze-fracture electron micrographs looking through that blastocoel roof. You can see, at the beginning of gastrulation, we have about three cells thickness of the whole surface. But during gastrulation, there's a thinning process, and you can see these green cells, here, intercalating between one another. And that of course is going to lead to a spreading. So, there's a spreading not only of the green part — by virtue of these rearrangements of cells, the intercalating between each other — but also up here, you can see that these red cells begin to be a little stretched and are no longer quite so cuboidal.

It's subtle but it makes a difference in the overall process. And below here, it's represented schematically, where you start with the three layers and then go through this intercalation, and end up with a thinner but much wider array. This is the work of Roberto Mayor's lab. What they found is that there's a signal that's made in the red cells. It's actually a complement signal. And that acts as a chemoattractant for the green cells to compact against the red cells. And so that's thought, now, to be the major driver of this spreading process of the blastocoel roof. Let's move on to a second [process] now: vegetal rotation. So, vegetal rotation is the process whereby all of this yolky stuff, you'll remember, has to constrict move inside. And so what we're going to see is there'll be a spreading of this floor of the blastocoel while, at the same time, the outside layer will have to collapse down essentially to a point to get enveloped by the rest of the animal. And that occurs by a fountain-like movement of cells. It's also slightly asymmetric, such that this dorsal side — the future head tissue, up here — is going to end up moved against the blastocoel roof.

You can see this cleft here. So, unlike on the ventral side, the dorsal side has this cleft, the cleft of Brachet, as it's known, where there's a contact between the inner and the outer part of the embryo. So, again, let's watch what happens here. You can see the spreading, there, of the blastocoel roof. It becomes quite a large surface area before it finally collapses. The collapse of the blastocoel is thought to be mediated mostly because these cells here are very loosely attached. And the liquid moves from the blastocoel into this other cavity, the archenteron, which is the primitive gut cavity. At the same time, you can see that cells down at the bottom, they initially covered a very large surface area, but they collapse down to a small point. So again, this is thought to arise as a result of the rearrangements of cells, so that it's a sort of fountain-spreading of cells. And this has been studied recently by Rudi Winklbauer's. And here he's taken a slice through that yolky tissue and… and put it in culture. And he can see this fountain-like movement of cells.

So, these cells rearrangements are exerting force. And here you can see, also, the spreading of the blastocoel floor, whereas the initial area was much, much broader. Okay. So, that's thought to be as a motor for get… getting convergent… this… this vegetal rotation against the blastocoel roof. The underlying mechanisms are not understood so well, but at least the cellular behaviors can be described. Let's move on to bottle cell formation. We saw that as the initial pigment line, which we can see from the surface of the embryo. And actually, we see these after there's already been quite a lot of action inside the embryo. So, these internal movements have already started to occur.

And indeed, these bottle cells form slightly later. And then, as I mentioned in the first lecture, they form a kind of an inflection point, so that when other forces extend this red mesoderm, and crawling forces of the purple tissue occur, of the blastocoel roof, then rather than going outside the embryo these will flip the corner and move inside the embryo. So, here's a… a light micrograph of a stained section done by Jen-Yi Lee in the lab. And what she's done here is to stain the cells with tubulin and actin. And as is generally found with these apically constricting cells… they're called bottle cells because of the old-fashioned glass blown bottles that were used when these were first described in the late 19th century.

You see this apical constriction here. The cells become elongated. And actually, Holtfreter showed that these cells are quite invasive. They really have a strong impetus to move inside the embryo. While the apices constrict… and you can see here there's some irregularity in how they go about this. Some of those are quite constricted, whereas the neighbor is not so constricted. But the actomyosin contrac… contraction that's mediated there constricts these apices and leads to the beginning of an invagination movement on the outside of the embryo.

Here's a movie that shows that from the outside. So, this is an embryo that's been labeled with a membrane-targeted GFP. And so what we'll see as this plays through is that up here there's not much action, and these cells are perhaps being stretched, but down here where the bottle cells are forming you can see the constriction of the apical surfaces of these cells. And it's by no means completely uniform. There are some cells in between them that are not undergoing apical constriction. They're just sort of passively stretched by the forces from the apical constriction. But ultimately, all of these cells will start to apically constrict, and that will form the… this impetus for these cells to…

To move inside the embryo, and help the process of involution of the marginal zone, the equatorial zone of the embryo. I've already mentioned this cleft, here, the cleft of Brachet. And of course, this cleft could have two fates. One is that the tissues can remain separate, or they could merge with one another. And that's a real active process of tissue separation that these cells are maintained as different. And that can be shown by these explant experiments, again done by Rudi Winklbauer. And what he did was to take this blastocoel roof, this pale blue tissue, prospective epidermis, and invert it in culture. Then, he can put onto that little groups of cells from different regions of the embryo. And so, for example, what he can show is if he takes like tissue from the ectoderm and puts that on there, those cells will dive in and just merge with the rest of the explant.

But instead, if he takes this… this red tissue or purple tissue, either one, he can put those on the blastocoel roof and they'll maintain a separate identity and stay separate. So, this is an active difference in the cell fate of tissue separation that keeps these separate and allows this cleft to be maintained. We're now moving on to cell migration. And this is a real force-generating process as well. And this leading edge here of the endomesoderm… so the endoderm is yellow, the mesoderm is pink and red… that endomesoderm actively sticks to the overlying blastocoel roof. It's just the first cell or two that does this. And then those cells actively crawl across the blastocoel roof. So, during gastrulation, these cells have the ability to find their direction up towards the top. Now, this has also been studied in isolated cells. And again, one can use a similar assay, where one has an inverted cap from the gastrula, and take this purple tissue and put it on top.

If one does that with big pieces of tissue, then it is able to coherently migrate in a single direction. But here, what I show in this movie is that the individual cells, they clearly have the ability to sit on the underlying blastocoel roof, but they're migrating. They're migratory. And individuals, they migrate pretty randomly. But when they're in a coherent mass, there's an edge that provides a directionality to their migration. Okay, we've done the first five.

And the final one is we're going to do convergence and extension. And that is the process whereby this red tissue becomes very long. This occurs not only in the mesoderm, making the notochord, but also up in the spinal cord, the prospective spinal cord, which is just up here. This has been studied in detail, again using explants.

And this particular one is the work of John Wallingford in the lab, where he took explants that had been originally made by Ray Keller: the "Keller" explant. So, over here is a whole gastrula. And he's cut off, with a pair of eyebrow knives, this dorsal region here, peel it back and chop it off. So, you have one piece of tissue that's just a few cells thick. And this can be put nicely underneath a coverslip and squished down onto a slide. It doesn't adhere to the slide, but the cell movements and the cell behaviors in there can be visualized using this confocal microscope that's sitting underneath it.

So, to look at the behavior of the cells, we… we scatter-label those cells by injecting one of the early blastomeres, the early cells, with a membrane-targeted GFP, such that we get some but not necessarily all the cells labeled. Now, previous to this experiment, Ray Keller had already described the behavior of these cells. And in particular, at the onset of gastrulation these cells put out these lamellipodia, these slabs of… of protrusion in all sorts of directions, randomly. But during the course of gastrulation, those cells become very oriented, mediolaterally, so that they're putting out the protrusions on their sides, their left and right sides.

They then proceed to crawl between each other, so that initially, for instance, this array of four cells will become an array of four cells, here, that's in a line. So, it goes from a squat series of cells to a long, skinny array of cells. And this is a potent force-generating mechanism in gastrulation that drives the elongation of the prospective notochord and the prospective spinal cord. So, this is that happening in a dish. For technical reasons, what's been done is to take two of those explants and sandwich them together. And then they do this behavior very well. Even though they're sitting on agarose, they have nothing to crawl on. And so you can see here that the head end of this explant does not undergo this behavior so much, but this spinal cord area does. And if we use molecular markers, we can tell that this is all spinal cord, here, whereas down here we have the mesoderm, the not…

The perspective notochord, we have mesoderm. Now in this case, we keep it flat, underneath pressure from the coverslip, whereas in the normal embryo all of this stuff down here would have crawled underneath and up the inside. But this illustrates the autonomous behavior of these cells in this macroscopic view. Let's go to the… the confocal view now. And here, we're going to discuss an experiment where we've manipulated planar cell polarity signaling. So, using a special internal deletion of the molecule Dishevelled, which is involved in planar cell polarity, we're going to interfere and see what that does to these cells. It had previously been shown by Sergei Sokol that if you use this reagent in the whole embryo it prevents the convergence and extension behavior.

But the question we had was, is this a very specific effect on the cell behaviors, or is it a sort of a nonspecific toxic effect on the cell behaviors? Now, as you'll see, we can resolve that clearly by looking at the behavior of the cells in detail. So, let's start this movie going. And what you'll see is the one on the right is actually doing something, whereas the one on the left is doing very little. This is the control. And what this is showing… these cells are aligned sort of from left to right. And they have these extensions on their surface. You can see these lamellipodial extensions. So, this is the phase where they're getting oriented. They put out at least one extension. That extension is very stable. And so, over the course of time, it's thought that that can exert traction on the neighboring cells so that the cells crawl between each other.

This is the case where we put in that dominant-negative reagent, and you can see a very different behavior. But importantly, if this was a very nonspecific effect, a toxic effect, we would expect the cells to do nothing and not crawl between each other. Instead, what happens is these cells are actually hyperactive. They're putting out these protrusions that I think you can see on this cell, for instance. Some of these protrusions are put out and then taken back quite quickly. So, the protrusions are unstable, and therefore unable to sustain a force generation. And the other thing is that if one looks at the shape of these cells, they're much more randomly shaped.

The cells in the control actually become elongated, whereas the cells in the manipulated case lose that orientation. So, we think it's that lack of polarity in these cells, the… and the lack of ability to sustain traction, that causes them to essentially be non-productive. They're like a bunch of kids with ADD. They're running around with lots of energy but not achieving a whole lot. So, instead of that sustained and disciplined behavior, we have very active behavior, but without the polarity and the sustained lamellipodial contact that allows this intercalation of cells. And the… the consequence to the embryo, is I said, is the lack of convergence and extension. And we can see that in the whole embryo.

The top embryo is a control. That's undergoing its proper convergence and extension movements. And here we see that particularly in the neural plate. As the neural plate extends and comes together to make the neural tube. In contrast, the bottom embryo has been injected with this dominant-negative reagent. And you can see it… although it can close the blastopore, it fails to undergo much convergence and extension of the neural plate. The consequence is that the neural folds… although the neural folds try to form, they never quite get close enough to touch. And so this mimics a human condition that is sadly quite prevalent in children, that of spina bifida, a very extreme form of spinal bifida, in this case, that results from the loss of that planar cell polarity signaling. And indeed, it's now known that one of the predisposing factors in human birth defects of spina bifida can be defects in the planar cell polarity components.

So, those are the movements that I've summarized. And as you've seen, they can, between them all, get together to give this rather complex overall result, but each individual cell behavior is relatively simple. So in this movie, the embryo ends up pointing at the sky, but if we turn it over so that the anterior is now at the anterior end, you can start to see that this is really a tadpole. So, we have a head over here.

We're gonna have the spinal cord that's forming up here. Here's the primitive gut and the future anus. And if that just goes along a little bit further, with the extension of the axis, we can see that it really starts to form something that looks like an animal, with the eye going to be forming over here; a spinal cord; this long stiff rod, the notochord; and then the gut beneath.

So, as far as I'm concerned, as an embryologist, this is pretty much it. The rest is just elaboration of this initial pattern, and the growth of the animal into an adult. So, that's this short presentation on the cell movements. In the next presentation, we're going to discuss how the signaling events happen during early development to make the cells different from one another, and give rise to the formation of the nervous system, in particular..

You May Also Like