On Wednesdays, I try to break away from the lecture format and prompt the students to talk about the science of development. We’re working our way through Sean Carroll’s Endless Forms Most Beautiful, and yesterday we talked about chapters 3 and 4.

Chapter 3 has an overview of basic molecular biology — transcription and translation, that sort of thing — and since these are junior and senior students who’ve already heard that a few times, we skipped right over it and they explained to me what master genes are, with specific examples of homeobox-containing genes like the Hox genes and Pax6. They caught on fast that what we call master genes are actually just transcription factors located high up in a regulatory hierarchy.

I think we also got across a less-than-naive idea of the evolution of Hox genes. There is a recognizable, conserved motif in each of these genes, but the proteins are far more than just their homeodomains, and can exhibit considerable variation — necessary functional variation, because the expression of different Hox genes are going to have distinct morphological consequences.

Chapter 4 has a general theme of maps and geography — what does it mean for a cell to be in a particular position and to have a particular fate? We also get into details. This is a very fly-centric chapter, and we get a picture of early development in the fly and the specific patterning and positional organization in the early embryo of that organism, with an introduction to many genes we’ll be hearing much more about during the course of the term. We also got enough information on vertebrate development that I could ask them to play the compare and contrast game: what’s different and what’s the same in fly and mouse development? I’m trying hard to be the Reese’s Peanut Butter Cup of development in this class: it’s so easy to say, “they’re the same!” and focus on common molecular mechanisms, or to say “they’re different!” and talk about the numerous quite radical innovations between them (especially in the fly, which is a weird, highly fine-tuned machine for rapid robust development). I’m trying to get across that both statements are absolutely true, and they really taste great together.

Friday is their first exam. Next Monday, class will be an overview of nematode development, to prime them for the lab exercises for the next two weeks which will be all about photomicrography of worm development and behavior, and also more details about early fly embryology to get them prepared for a couple of weeks of nothin’ but flies. I also warned them that next Wednesday we’ll be discussing chapter 5 in Carroll, just chapter 5, because I’ve found in the past that that’s usually the brain-clogger chapter, with all its talk of boolean logic and gates and circuits.

We began today with chocolate. Always a good thing at 8am, I think — so I brought a candy bar to class. Then I told the students that I loved and respected them all equally and that they all had equal potential, but that I was going to mark just one person as special by giving them that candy bar^*. So I asked them how I could decide who should get it, telling them right off that dividing it wasn’t an allowed solution, and that yes, this could be an openly unfair process.

There were lots of suggestions: we could do it by random chance. I could throw it into the middle of the room and let them fight over it. We could analyze everyone’s DNA and give it to the most average person…or the most genetically unusual. I could just give it to the first person to raise their hand, or the person closest to me, or the person farthest from me. We could have a competition of some sort, and the winner gets it. I could give it to the person who wants it most, or who needs it most.

The point I was making is that this is a common developmental problem, that you have a potentially uniform set of cells and that somehow one or a few have to be distinguished as different, and carry out a different genetic program than another set of cells. One cop-out is to invoke mosaicism: that is, they aren’t uniform, but inherit different sets of cytoplasmic determinants that make them different from the very beginning, but that even in that case, these determinants aren’t detailed enough to specify every single cell fate in most organisms. Even with an initial prepattern, you’re eventually going to end up with a field of cells, like the dorsal side of the fly wing, and within that uniform field, some cells will have to be programmed to be epithelial, others to be bristles, others to be neurons. And that means that in every organism, even the most classically mosaic, you’ll reach a point where cells have to process information from their environment and regulate to build differential structures.

And with that I went on to talk about some animals that were judged as being mostly mosaic in character: molluscs, tunicates, echinoderms, and nematodes. Even here, these animals all required complex molecular interactions to build their embryos.

For example, I’d earlier used echinoderms as classic examples of regulative development. You can dissociate them at the 4-cell stage and each blastomere can go on to build a complete embryo. But at the 8-cell stage, when the cleavage plane separates an animal half from a vegetal half, that’s no longer true: the top four cells when isolated are animalized, forming only a ciliated ball, while the bottom four cells are vegetalized, only making a static blob with a bit of a skeleton inside. Clever experiments can quantitatively juggle these cells around, removing just the bottom 4 cells (the micromeres) at the 16-cell stage, or assembling composite embryos with different ratios of the different tiers of cells, and get different degrees of development. Even when you’re discussing an organism in which you’d call the pattern of development mosaic, it absolutely depends on ongoing cell:cell signaling at every step, and the final form is a consequence of interactions within the embryo. It’s a mosaic-regulative continuum.

I also described very superficially the work of Davidson and Cameron on specification events in echinoderms. These interactions can be drawn as a kind of genetic circuit diagram, where what you’re seeing is the pattern of genes being switched on and off. We can describe a cell type as the output of mappable gene circuitry, and we can even identify modules of networks of genes associated with a particular kind of cell, and that we can also see a limited number of genes that mediate interactions.

Next week I promised to start going into more detail, when we start talking about early fly development and axis decisions. The next class we’re actually going to switch gears a bit and discuss Sean Carroll’s Endless Forms Most Beautiful.

Slides used in this talk (pdf).

^*Yeah, I lied again. I brought enough candy bars for everyone, and after we’d generated a list of ways to share just one, I gave them to everyone. They’ll never trust me again.

You really can’t teach a class by lecturing at them…especially not an 8am class. But sometimes there is just such a dense amount of information that I have to get across before the students know what to ask that I have to just tell them some answers. My compromise to deal with this eternal problem is to mix it up; some days are lecture days, others are discussion days. And today was a discussion day.

I’ve been talking at them for the past two weeks, basically working to bring them up to a 1950s understanding of the field of developmental biology, with a glimmering of the molecular answers to come and some of the general concepts, so that they’re equipped to start thinking about the contemporary literature. So I had them read this paper before class:

Kerszberg M, Wolpert L (2007) Specifying Positional Information in the Embryo: Looking Beyond Morphogens. Cell 130(2):205–209.

And then today they got into small groups and tried to explain it to each other. I primed them by suggesting that they try to define the terms positional information, gradient, morphogen, and polarity, and mentioned that I was playing a dirty trick on them, giving them a paper to introduce a basic concept that at the same time was pointing out some of the difficulties and problems of the idea, so I expected them to also do some critical thinking and question the concepts.

So they went at it. It went well; they had some lively conversations going on, which I always worry won’t happen with early morning classes. I find it helpful to ask students to try to poke holes in an idea, rather than just recite by rote what the paper says — it sends them hunting rather than gathering.

Several students noted that having a simple continuum of a molecule begs the question of how that gets translated into many discrete cell types; why does having one concentration of a morphogen make a cell differentiate into a thorax, while a very slightly lower concentration means it differentiates into an abdomen? It’s all well and good to suggest that a couple of overlapping gradients can specify position, like laying out a piece of graph paper with coordinates on it, but it doesn’t explain how that gets translated into position-specific tissues. I was most pleased that several of them, while groping for an answer, related it to the lac operon in E. coli, and brought up the idea of thresholds of gene activation. Yay! That so sets up future discussions about early fly embryogenesis, where that is exactly the answer.

I think they also got the idea that an explanation for general specification of body parts, for example, may not apply for explaining polarity within a body part; we may have to think about some kind of hierarchy of regulation, where we progressively partition the embryo into smaller and smaller units, with different mechanisms at different scales. They might be catching on to the depths of the problems to come.

Another way the paper primed the students was that it very briefly introduces a whole bunch of specific molecules: dpp, bicoid, sonic hedgehog, activin. They got a very general idea of the broad roles these molecules play, all part of my devious grand plan. When we start talking about the details of how animals set up dorsal-ventral polarity, for instance, and dpp/BMP start coming up in more specific contexts, I want them to be familiar old friends — molecules they already knew casually and informally, and now see doing very specific things, and interacting with another set of molecules, which now also joins their circle of pals. Before I’m done with them, they’re all going to regard these developmental signals and regulators as part of their family!

My development class also has a lab. The last few weeks have all been teaching them how to get good optics on a research scope, and how to take photomicrographs with my PixeLink camera system. Today I’m going to show them how to appropriately process an image for publication, so they’ll learn a few digital enhancement tricks and a few ethical rules. I lay down a few laws about using image processing on scientific data:

What you must do to your image:

• You must archive the original data and work with a copy. If I ask to see the original after you’ve enhanced the image up the wazoo, you better be able to show it.

• You must document every step and every modification you make. You’re going to describe everything either on the image itself or in a figure legend; if this were to be published, you’d probably include it in the Methods section.

• You must explain the scale and orientation of the image. The scale is usually shown by including a scale bar; orientation may be shown either by including annotations (text describing landmarks in the image) or an explanation in the figure legend, such as that it is a sagittal or horizontal section.

• You must save the image in a lossless format, such as .png or .psd or .tiff. Do not save it in a lossy format like .jpeg, which can add compression artifacts.

What you may do to your image:

• You may crop and rotate the image.

• You may adjust the contrast and brightness for the whole image.

• You may carry out simple enhancements, like applying a sharpening filter or unsharp masking, to the whole image…but remember, document everything!

• You may splice multiple images together to produce a photomontage; you can also insert panels with enlarged or otherwise enhanced regions of the image, as long as it is absolutely clear what you’ve done.

What you may not do to your image:

• You must not carry out selective modifications of portions of the image; you cannot sharpen the cell you care about and then reduce the contrast for other regions, for instance. You should not burn or dodge regions of the image.

• No pixel operations or retouching: you are not allowed to go into the image and paint your data into existence!

We do a lot of this preliminary basic stuff because I run the course out of my research lab, rather than a student lab. I want to make sure they’re not going to break anything, and also that they know how to do good imaging, a skill they’ll find useful in other courses and in research (years ago when I taught this stuff, we’d also do black&white darkroom work — nobody does that any more, so now it’s all photoshop). The goal is to get them all able to churn out lovely photographic data, so later I can just hand them some nematodes or fruit fly embryos and tell them to do their own experiments and observations, just show me the pretty pictures when they’re done.

It’s a good life, being able to sit back and let students bring me gifts of biological beauty. I think they’ll also be posting some of these to their blogs.

I was bad. I didn’t post my summary last week, so this is actually what I taught a week ago and what I taught today.

[Read more…]

This is an abbreviated summary of my class lecture in developmental biology today. This was the first day of class, so part of the hour was spent on introducing ourselves and going over the syllabus, but then I gave a lightning fast overview of the history of developmental biology.

Classical embryology began with Aristotle, whose work was surprisingly good: he approached the problem of development with relatively few preconceptions and fairly accurately summarized what was going on in the development of the chick. Most of this old school embryology is descriptive and was really a narrow subset of anatomy, but there were a few major conceptual issues that concerned the old investigators, in particular the question of preformation (the plan of the embryo is laid out in the egg) vs. epigenesis (the plan of the embryo emerges progressively). Aristotle, by the way, was on the right side of this debate, favoring epigenesis.

In the 19th century, development was seen as a progressive process that paralleled the hierarchical organization of nature — that is, developmental biology, what there was of it, was coupled to the great ladder of being. This is not an evolutionary idea, but reflects the view that there was a coherent pattern of greater and lesser development that was part of a coherent divine plan for life on earth. The German ‘Natural Philosophers’ pursued this line of reasoning, often to degrees that now look ridiculous in hindsight. In contrast, there were developmental biologists like Karl Ernst von Baer who wanted nothing to do with a cosmic teleology but instead preferred to emphasize observation and data, and simple minimal hypotheses.

In the late 19th century, developmental biology split into two directions. One was a dead end; Ernst Haeckel basically lifted the explanatory framework of the natural philosophers, replaced divinity with evolution, and tried to present development as a parallel process to evolution. Von Baer had already demolished this approach, and despite a few decades of popularity Haeckelian recapitulation died as a credible framework for studying evolution in the early years of the 20th century. The other direction developmental biology took was Wilhelm Roux’s Entwicklungsmechanik, or experimental embryology. This was an approach that largely eschewed larger theoretical frameworks, and focused almost exclusively on observation and experimental manipulation of embryos. It was a successful discipline, but also divorced mainstream developmental biology from the evolutionary biology that was increasingly influential.

As examples of Entwicklungsmechanik, I discussed Roux’s own experiments in which he killed one cell in a two-cell embryo and saw partial embryos result, an observation that fit with a preformationist model, but more specifically a mosaic pattern of development, in which patterns of development were encoded into the cytoplasm or cortex of the egg. Those experiments were seriously flawed, however, because the dead cell was left attached to the embryo, and could have deleteriously affected development. The experiments of Hans Driesch were cleaner; he dissociated embryos at the four cell stage, cultured each blastomere independently, and discovered that each isolated cell developed fully into a complete, miniature larva.

Driesch, unfortunately, interpreted these results to imply that there was an entelechy, or guiding intelligence outside the embryo, and that the only conceivable explanation was the existence of purpose behind embryology. This was also a dead end; the modern explanation for the phenomena is that they regulated, that is, that cells determine their fate by interacting with one another, rather than some kind of cosmic plan. And that’s really going to be a major focus of this course: how do cells communicate with one another, how are genes regulated to set up coherent and consistent patterns of gene expression that produce the organized cell types we find in an adult multicellular plant or animal?

That set up the next lecture. Entwicklungsmechanik, while representing a solid and productive research program, quickly reached its limits, because what we really needed to examine were those patterns of gene expression rather than trying to infer them from observations of morphology. The big breakthrough was the melding of developmental biology and molecular biology — most of the modern developmental biology literature focuses on examining interactions between genes. So on Wednesday we’ll get another fast overview of the molecular genetics research program, and a bit of evo-devo.

Slide thumbnails (PDF)