Chapter 26: Development and Evolution

Welcome to Last Minute Lecture.

This free chapter overview is designed to help students review and understand key concepts.

These summaries supplement not replaced the original textbook and may not be redistributed or resold.

For complete coverage, always consult the official text.

Welcome back to the Deep Dive.

You know, the story of evolution we all learn is, well, it's usually about a sudden change happening and then, boom, natural selection steps in and decides if that change survives.

The survival of the fittest.

Exactly.

But today, we're going to dig into something a little more fundamental.

We're asking,

where does that new selectable variation,

the raw material for selection, actually come from?

We are taking a deep dive into the revolutionary field that's searching for the arrival of the fittest.

That's a perfect way to put it.

For centuries, the whole focus was on that selective pressure.

What EVO -DEVO or Evolutionary Developmental Biology does is it shifts the spotlight back to the source of all that variation.

Which is the developmental process itself.

Exactly.

If you think about development as just the change in gene expression and cell position over an organism's life, then evolution.

Well, at its core, it's just the change of that process over geological time.

Okay, so that really frames our mission for this deep dive.

We're trying to connect the incredibly fast scale of an embryo developing hours, maybe months, to the vast, fast time scale of eons.

And this isn't a new idea, not really.

Even Charles Darwin, back in 1857 when he was formulating his theory, he was consulting with Thomas Huxley about this very thing, about the nature of variation.

And Huxley's insight was, frankly, crucial.

It's still central to EVO -DEVO today.

He observed that when you compare different organisms, the differences aren't usually, well, they're not from creating entirely new structures out of thin air.

It's more about modifying what's already there.

Modification, not invention.

That was a key.

You're just modifying parts that are already common to these different types of animals.

So the secret sauce for evolutionary change isn't always inventing a brand new gene.

It's more like finding a new use for an old one.

Right.

Which is changing where an old one gets turned on.

Precisely.

And now we have the tools, modern molecular biology, genomics, computational modeling, to actually explain how these shifts in the embryo can lead to profound anatomical differences between species.

We're talking about things like losing limbs or the expansion of a brain.

Yes.

And often it comes down to really subtle changes.

Just a shift in the when, the where, or how much a developmental gene is expressed.

Okay.

So before we get into the leads, let's unpack the central punchline of this whole deep dive right up front.

What is the single biggest mechanism that allows for all this grand evolutionary flexibility?

It's the flexibility and the modularity of our DNA's regulatory regions, specifically these little things called enhancer elements.

If you take away one thing, that's it.

Enhancers.

Why are they the key?

Well, because a change in an enhancer, a regulatory shift, can tweak the cell type or the timing or the amount of a gene's expression in one very specific part of the body.

So you can have variation emerge safely without breaking the whole system.

Exactly.

It leaves the gene's function in all the other tissues where it might be absolutely vital for life, completely untouched.

This ability to just tinker locally while the rest of the blueprint stays stable, that forms the basis of pretty much all the morphological variation you see in the animal kingdom.

All right.

Let's start at the beginning then.

The foundations of this field really do trace

to Darwin himself.

His theory of descent with modification was, I mean, it was a brilliant synthesis.

It reconciled two big observations in natural history that at the time people saw as being in opposition.

They did.

Those two forces were the concepts of unity of type and conditions of existence.

Before Darwin, naturalists were really split.

You had people like Richard Owen who championed this unity of type.

Right.

He was focused on the deep underlying similarities between animals, what we now call homology.

Exactly.

Homology.

They were asking questions like, why does the bone structure of a human hand and a mole's digging paw and a seal's flipper and even a bird's wing, why do they all follow that same basic five digit pattern?

And Darwin's answer was descent from a common ancestor.

Right.

Those homologous structures are the signature of a shared ancestry.

It's the same blueprint, just modified for different jobs.

And on the other side, you had the conditions of existence camp.

Yeah.

People like Georges Cuvier, who were all about how exquisitely adapted each organism is to its specific life.

They were focused on the differences.

Yeah.

The adaptations.

What makes a bird wing, a marvel for flight or a seal flipper so perfect for swimming?

So Darwin's genius was basically saying you're both right.

He put them together.

He said the unity of type is explained by that common ancestry and the modifications, the adaptations, are explained by natural selection, acting on those ancestral structures over millions of years.

Driven by the conditions of existence.

Exactly.

And even back in the 19th century, long before we knew about DNA, the evidence connecting development and ancestry was often right there in the earliest stages of life.

In the embryos and the larvae?

Absolutely.

Darwin was incredibly enthusiastic about what he called larval homologies.

For instance, the larval anatomy of barnacles proved definitively that they were crustaceans.

It cemented their place in the animal kingdom, even though the adults look so bizarre and sessile.

And there was another one, right?

The work on tunicates.

Yes.

Alexander Kowalewski's work on tunicates.

These are those.

Well, they look like little sacks attached to rocks.

Marine filter feeders.

Not very chordate -like.

Not at all.

But Kowalewski showed that the tunicate larvae had anotochord and pharyngeal pouches.

These are structures that are characteristic only of chordates.

Like us.

So that developmental observation linked invertebrates directly to vertebrates.

It was a massive piece of evidence.

It showed that developmental change was the engine of evolution, even if they couldn't see the molecular gears turning back then.

Which brings us to the core paradox that Ivo Devo really had to solve.

The paradox of change.

I mean, development is so complex, so integrated, timing and signaling, everything has to be just right.

It's a high -wire act.

So how can this incredibly fine -tuned system change dramatically, like lose a whole structure or shift a body part, without causing the entire organism to just fail?

Right.

If you mess up the blueprint, you should kill the organism.

The answer, as modern Ivo Devo has shown us, lies in two really critical preconditions that apply to basically all multicellular life.

Modularity and molecular parsimony.

Okay, let's tackle modularity first.

You hear this framed as divergence through dissociation.

What exactly is a developmental module?

A module is just a discrete,

interacting unit of development.

It can be a physical thing, like the morphogenetic field for the heart or for a limb, or it can be a functional thing, like a specific signaling pathway.

Like the WANT or BMP pathways that you hear about all the time.

Right.

Think of them as independently managed subroutines in the organism's big developmental operating system.

And the key evolutionary idea is dissociation.

Why is that so important that they could evolve independently?

Because it allows for local innovation without causing a global collapse.

If the module that controls, say, the formation of your heart is totally separate from the module controlling your arm, then a mutation that dramatically changes your arm.

Would automatically give you a heart attack.

Exactly.

It won't cripple the heart.

You mentioned an experiment by Victor Twitty that demonstrated this just beautifully.

Oh, it's a classic.

From the mid -20th century, Twitty took a limb bud, that's the precursor tissue for a limb, from the larva of a large species of salamander, and he grafted it onto the flank of a larva from a small species of salamander.

The host larva grew up, and it was its normal, genetically determined small size, but the grafted limb...

Let me guess.

It grew to be huge.

It grew to its own genetically predetermined large size.

It was completely comically disproportionate to the rest of the host's body.

So that limb module was completely ignoring the global growth signals of its new host.

Completely.

The genetic program defining the size of that limb field was dissociated, or independent, from the program defining the overall body size.

And that independence is what gives evolution its playground.

You can adjust the limb growth dial without touching the heart function dial.

Right.

That makes the paradox much, much less severe.

Modularity is the safety net.

So what about the second precondition, then?

Molecular parsimony.

This must have been a real shock when molecular biology first got going.

It was astounding.

Molecular parsimony, or what we call the small toolkit, is this observation that despite the incredible morphological diversity you see across all animal phyla...

From jellyfish to insects to us...

The development of all these creatures relies on the same basic types of molecules?

The same core machinery.

Transcription factors, the signaling molecules, the adhesion molecules.

Yes.

Evolution isn't constantly reinventing its fundamental building blocks.

It's using the same high -performance, multipurpose toolkit.

The BMP pathway, the Wnt pathway, the hox genes in every single animal lineage.

It's like being given the same set of Lego bricks and being told to build a jellyfish, a beetle, and a human with them.

So if that toolkit is the same everywhere,

the key to all this change has to be in the instruction manual.

Let's really zero in on how that first precondition modularity is organized at the genetic level.

This brings us back to those enhancers you mentioned.

Right.

We established that anatomical parts are modular, but the DNA regions controlling them are even more modular.

Think of the gene itself, the part that codes for the protein.

That's the constant component.

It's the hammer, or the nail.

The enhancer elements are the specific localized switches.

They're the instructions that dictate where and when that hammer gets used.

Can you break down that structure a bit more?

Sure.

A single developmental gene can have multiple discrete enhancer elements associated with it.

Each one of those enhancers is a binding site for specific transcription factors, and it only directs the gene's expression within a specific cell type at a specific time in a specific tissue.

They are totally independent instructions.

Why is that structure so profound for evolution?

Why don't we just see more mutations in the protein coding part of the gene itself driving these big changes?

Because of something called pleiotropy.

The idea that one gene does many, many jobs.

If a gene like Pac -6 mutates in its coding sequence, it would probably make a defective protein everywhere it's expressed.

In the eyes, in the brain, in the pancreas.

And that would lead to a severe, probably lethal problem.

Right.

Evolution just stops right there.

But if the mutation only hits one specific enhancer.

Then you get a viral selectable change.

Exactly.

The mutation only messes up the instruction manual for the one module controlled by that one enhancer.

Let's say the hindlimb.

The gene's function in the brain, the eyes, the heart.

It all remains completely normal.

The organism survives, and that new variation can be tested by selection.

This is why you said that mutations in enhancers are now seen as the most important cause of morphological change.

They really are.

And the stickleback fish case study.

I mean, it makes this concept unforgettable.

It's a perfect, rapid, natural experiment that shows enhancer evolution in action.

Tell us about the three -spined stickleback.

Okay.

So the three -spined stickleback is a fish.

The marine populations have these large pelvic spines, which are basically modified pelvic fins, and they're really important for defense against predatory fish in the ocean.

But their changes when they move from the ocean to freshwater.

Dramatically.

Around 12 ,000 years ago, as the last glaciers retreated, these marine sticklebacks colonized thousands of new freshwater lakes.

And in these lakes, the main predators were often these big aquatic insects.

And the insects could use the spines against the fish.

They actually used the spine as a kind of handle to grab onto the fish.

So suddenly being spineless was a huge advantage.

Selection rapidly favored the loss of the spines.

So you end up with two distinct populations.

Spine marine fish and spineless freshwater fish.

The perfect setup for a genetic experiment.

It is.

Researchers mated them to map out where the genetic difference was, and they found it.

They mapped the major pelvic difference to a single spot on chromosome 7, which happens to contain the PITS -1 transcription factor gene.

And PITS -1 is known to be involved in making hind limbs in vertebrates.

Correct.

So the obvious first thought is, OK, the freshwater fish must have a broken PITS -1 protein.

A mutation in the gene itself.

But that's the classic assumption.

And this is where the evo -devo plot twist comes in.

It is.

When they sequenced the protein coding region of PITS -1, it was absolutely identical between the spined and the spineless fish.

The protein was fine.

It was physically capable of doing its job.

So the problem wasn't the tool.

It was the instruction for where to use the tool.

Exactly.

They looked at the expression pattern.

In both fish, PITS -1 was expressed normally in the thymus, the olfactory neurons, the mouth cartilage, all these places where it's essential for survival.

But in the freshwater fish, the expression in the pelvic region was just gone or severely reduced.

It was silent right where the spine needed to form.

That's dissociation in action.

The mutation hit only the pelvic module.

They eventually located the culprit, a specific enhancer module about 2 .5 kilobases long that normally sits upstream of the PITS -1 gene.

Its only job is to direct expression to the developing hind limb or pelvis.

And in the freshwater fish, that enhancer was broken.

It had a mutation, a deletion in both cases that just rendered it non -functional.

What was the definitive proof?

The definitive proof was taking that functional marine enhancer and experimentally inserting it into the embryos of spine -deficient freshwater fish.

And what happened?

They grew spines.

They grew pelvic spines.

The ability to make the spines was always there.

The freshwater fish just lacked the genetic instruction, the enhancer, to turn that ability on in the right place at the right time.

That is just.

It's such powerful evidence that evolution is working on the control panels, not the machine itself.

And this modularity, it also allows evolution to do something called co -option or recruitment.

Co -option is one of the most common ways that major new things happen in evolution.

If you have a functional module for making something, evolution can just tweak a regulatory switch to use that same module somewhere else for a new purpose.

Give us an example of this recruitment.

Okay.

Take the sea urchin larval skeleton.

Most echinoderms, like starfish or sea cucumbers, they only turn on the genes for making their skeleton plates when they're adults.

That's the ancestral role of that module.

Right.

But in sea urchins, that entire skeleton -forming module was recruited much, much earlier into the larval stage of development.

And how did that happen?

A few changes in the enhancer of a single control gene caused that adult module to be co -opted into the larva.

It allowed the larval cells to build a rigid skeleton way earlier than their ancestors.

It's an example of just changing the time the module gets activated.

What about the beetles?

You mentioned they're the most diverse group of animals on earth.

Their success is tied directly to their hardened forewings, the

They act like a protective shield for the delicate flight wings underneath.

Now, in all insects, there's a gene called apterus that defines the dorsal or top part of the wing.

Okay.

But in beetles, apterus was co -opted to do a new job.

In the foring, apterus now activates the genetic subroutine for making hard, chitinous exoskeleton plates.

That's what creates the tough elytra.

So it's borrowing the make exoskeleton program and applying it to the wing.

And at the same time, it also represses the genes needed for flight in that foring.

Meanwhile, in the hindwing, apterus just keeps doing its normal ancestral job so that wind stays flexible and ready for flight.

This co -option of one subroutine into one specific wing module is a huge reason why beetles were able to colonize so many different niches.

Let's pivot now to that second foundational idea,

molecular parsimony and this concept of deep homology.

We've established that the developmental toolkit is small and it's conserved across, well, all animal phyla.

What does it tell us about our common ancestor?

It tells us that the common ancestor of all animals already had a really sophisticated set of instructions for building a complex body.

It's not just that we share genes, it's that those genes, which we call orthologs, often perform the same fundamental jobs.

Like patterning the main body axis.

Exactly.

Whether you're looking at a sea anemone, a worm, or a human, the balance of BMP signaling levels generally specifies the dorsal -ventral axis, your back versus your front, and the wind and hawks pathways specify the anterior -posterior axis, your head to your tail.

This conservation is so deep that the basic architectural principles of how you build an animal were laid down more than 600 million years ago.

They were.

And probably the most famous illustration of this conserved identity is the PAC6 gene.

The master eye -building gene.

The master regulator for light -sensing organs.

Yes, it governs the development of eyes in pretty much every animal that has them.

From a squid, to an insect, to a primate.

The gene sequence itself is so deeply conserved that the protein is largely interchangeable between species.

And the experiment that proved this is genuinely one of the most mind -bending in all of biology.

It really is.

Researchers took the PAC6 gene, the full protein -coding sequence, from a mouse, and they expressed it in the developing leg tissue of a fruit fly.

And the result?

The result was that the fly grew a functional compound insect eye, complete with its little ommatidia growing right out of its leg.

So a mammalian gene told an insect cell to build an insect eye.

That's the ultimate proof that the underlying make -an -eye program is ancient and universal.

It is.

Now deep homology takes this a step further.

It's not just about single genes, it's about the conservation of entire signal transduction pathways.

And the mechanism for forming the central nervous system is a prime example of this.

It involves the interaction between cordon and BMP4.

And this gets into the difference between protostomes like insects and deuterostomes like us.

Right.

In both groups, the nervous system forms along the midline of the body.

And that formation is controlled by a balance of two signals.

An activator that wants to make skin, or epidermis, and an inhibitor that promotes nervous tissue instead.

In invertebrates, the activator is called DPP and the inhibitor is SOG.

In vertebrates, the activator is BMP4 and the inhibitor is cordon.

And these molecules are orthologues of each other.

They're the same ancestral genes.

So the basic pathway is identical.

The inhibitor blocks the activator, which carves out a specific zone where nervous tissue is allowed to form.

Exactly.

But what makes this so astonishing, what makes it deep homology, is that the overall body plan is inverted between the two groups.

In vertebrates, our nervous system runs along our dorsal side, our back.

In insects, it runs along their ventral side, their belly.

But the chemical language used to specify that midline is identical.

Identical.

It's incredibly strong evidence that the central nervous system originated only once in the ancestor of all bilaterian animals and that ancestor used this exact BMP -chordon balance.

That really helps explain why the toolkit is so constrained.

If these pathways are so fundamental, you can't really mess with them.

But if the core toolkit is conserved, how do we get new things?

How do new capabilities arise?

Through a process called duplication and divergence.

This is what generates gene families, or what we call paralogs.

And the mechanism is simple.

It's basically just an error in DNA replication that leads to the accidental copying of an entire gene.

I remember the biologist Susumu Ono used this analogy of a watchful police force.

It's a perfect metaphor.

Natural selection is that police force.

It makes sure the original functional gene stays constrained and keeps doing its vital, known job.

But the duplicate copy is now, well, it's unencumbered.

It's free from that immediate pressure.

It can accumulate mutations without killing the organism.

Exactly.

And this duplicate can then drift, or it can acquire totally new functions.

That process is called sub -functionalization.

And the hox genes, the ones that dictate body segment identity, are the absolute poster children for duplication.

They are.

The ancestor of all bilaterians probably had just a handful of hox genes.

Through successive duplication events, sometimes entire genomes duplicating, sometimes just regional duplications, we now have four distinct clusters of hox genes in mammals.

39 genes in total, on different chromosomes.

And their organization is famously conserved.

Their linear order on the chromosome often matches the order in which they're expressed along the head -to -tail axis of the body.

That's still interchangeable.

We know that the human HOXB4 gene, which specifies the identity of our upper thoracic segments, is so similar to its fruit fly ortholog, a gene called deformed, that you can put the human gene into a fly embryo, and it will do the fly gene's job.

So new capabilities come from duplicating an ancient, vital tool, and then letting the copy evolve to do a new, specialized job.

Speaking of which, let's talk about the human brain.

You mentioned a critical duplication event that's directly linked to our massive cognitive expansion.

Yes, the history of the sRGAP2 gene.

So all other non -human mammals have one functional copy of sRGAP2, and this gene makes a protein that's involved in regulating how neurons migrate and branch out in the cerebral cortex.

But in the human lineage, something different happened.

We duplicated it, not just once, but twice.

And the second duplication event, which happened roughly 2 .4 million years ago, was really significant because it was only a partial duplication.

It created a new gene we call sRGAP2C.

And what's the consequence of having just a partial copy?

Well, sRGAP2C is expressed in the cerebral cortex, and it makes a shorter, truncated protein.

And crucially, this short protein acts as a dominant negative.

It actually inhibits the function of the normal ancestral sRGAP2 protein.

So it's actively sabotaging the ancestral process.

Yes, but in a good way, an adaptive way.

Inhibiting the ancestral sRGAP2 leads to two major developmental changes in the cortex.

First, it slows down the rate of neural cell division, which extends the period of brain growth.

And second, and maybe more importantly, the dendrites, the branching connection points on the neurons, they become larger and much, much more complex.

So this partial duplication effectively created a break that slowed down development and allowed for much more complex wiring.

And the timing is just.

It's perfect.

This duplication happened around the same time we see the rise of Australopithecus and then the genus Homo.

It coincides with the first known use of stone tools and this rapid expansion of primate brain size.

It's a direct to molecular explanation for one of the most profound changes in our own anatomy.

Okay, so we have the raw materials,

modularity, and this conserved, duplicable toolkit.

Now let's get into the four primary ways that evolution, as Francois Jacob said, tinkers with these components.

Jacob's insight was that evolution doesn't design from scratch.

It's just combines existing parts in new ways, mainly by messing with the regulatory genes that build the embryo.

And these four major mechanisms are all defined by how the gene expression pattern is changed.

We'll look at changes in location, which is heterotopy, changes in time, heterochrony, changes in amount, heterometry,

and changes in kind heterotopy.

Let's start with A, heterotopy, a change in location.

This is a spatial change where a gene is expressed in a different physical place than it was in its ancestor.

The evolution of the batwing is a really elegant example.

In most mammals, as our digits develop, there's a process of programmed cell death apoptosis that removes the webbing between our fingers to sculpt them.

Right, but bats keep that webbing.

They keep it, and they do that by expressing a signaling molecule called FGF8 in that interdigital webbing, a place where it's not expressed in other mammals.

That's the heterotopy.

And what does FGF8 do there?

It does two things.

First, it inhibits the BMP signals that would normally trigger that cell death, so it saves the webbing.

And second, it's a strong mitotic signal, so it promotes cell division, which helps to elongate the digits.

So a single spatial shift in one growth factor results in a completely new limb structure, a functional wing.

It does.

And a far more complex series of heterotopies explains the construction of the turtle shell, which is one of the great morphological novelties in all of vertebrate evolution.

What's so unique about the turtle shell, developmentally speaking?

It's unique because it incorporates the ribs.

Instead of forming an internal rib cage like ours, a turtle's ribs migrate outward, laterally, to fuse with a set of bony dermal plates.

That's what forms the rigid carapace, the top part of the shell.

And that requires a whole sequence of spatial shifts.

It does.

It starts with the expression of FGF10 in the dermal layer of the trunk, right where the shell is going to form.

This FGF10 acts as a kind of recruitment signal.

It attracts the precursor cells of the ribs to migrate out into the dermis.

That's a completely new location for ribs.

A totally new location.

So the dermis is basically telling the ribs where to go.

Then, once the ribs are settled in this new dermal position, they start making BMP signaling molecules.

And the dermal cells around these relocated ribs respond to those BMPs by turning into bone themselves.

And those become the dermal plates that make up the rest of the shell.

Exactly.

The whole thing is this tightly orchestrated cascade of sequential heterotopies that just fundamentally reorganizes the entire vertebrate trunk.

Okay.

Next up is B, heterochrony, a change in time.

This is a shift in the relative timing or the rate of two different developmental processes.

Heterocrony is absolutely critical for defining the differences between species.

I mean, think about the extended growth phase of the human brain compared to a chimpanzee brain.

Our brain development is essentially paused at an earlier, more malleable stage for a much longer time, and that allows for greater complexity.

And snakes are a great example of developmental acceleration, right?

A fantastic example.

The huge number of vertebrae in snakes, some have up to 500, is a direct product of developmental acceleration.

The snake embryo's overall growth rate is normal, but the molecular clock that controls the segmentation of the body axis, it cycles nearly four times faster than it does in a lizard.

So you're packing more body segments into the same amount of developmental time.

Exactly.

More time units are packed into the same window, and you get a massively elongated hyper -segmented body.

On the flip side, you can have deceleration leading to reduction.

Take the loss of digits in the lizard genus hemiurgus.

Right.

Some of those species have lost their limbs entirely.

And that loss of digits correlates perfectly with the shorter duration of shh, or sonic hedgehog, expression in the limb bud.

Shush is the key signal that patterns the digits.

So less time for the shh signal to do its work means fewer digits are formed.

It shows that the duration of a signal is just as critical as its presence or absence.

That brings us to c, heterometry, a change in amount.

This is a quantitative shift in how much of a gene product you get or the size of a structure.

Let's go back to the blind Mexican cavefish.

These fish live in total darkness, and they've lost their eyes.

And this condition is caused by the overproduction of sonic hedgehog protein in the area where the eye should form.

The high level of shh suppresses the Pac -6 master regulator, and so the eyes just fail to develop.

Losing your sight seems like a negative, but how is this an adaptive heterometry?

Because that elevated shlevel has a positive effect.

It also increases the size of the jaw structure and the number of taste buds.

Since the fish can't see, selection has driven this adaptive tradeoff.

You lose the expensive, now useless eyes, but you get an enhancement of your feeding and tasting senses.

The novelty is purely in the quantitative change of one molecule's expression.

But the most powerful example of heterometry has to be the molecular explanation for Darwin's Oh, this is a true cornerstone of Ivo Devo.

The 15 species of finches on the Galapagos have this incredible spectrum of beak shapes, each one perfectly adapted to a different food source, and we've now traced that entire spectrum of adaptive radiation back to the specific amount and timing of just two signaling molecules in the developing phase.

Which two molecules are we talking about?

The first is BMP4, bone morphogenetic protein 4.

Higher levels of BMP4, especially when it's expressed earlier in development, lead directly to the development of broad deep beaks, the kind used for crushing hard seeds in the ground finches.

So more BMP4 means a deeper, wider beak.

Exactly, and researchers have shown this experimentally.

If you artificially enhance BMP4 expression in a chick embryo, you can induce a broader, deeper beak shape.

If you inhibit it, the beak becomes narrow and shallow.

What about the long, pointy beaks you need for probing into cactus flowers?

That's controlled by a different molecule, calmodulin, or CAM.

Gene expression studies showed that the sharp -beaked cactus finches had calmodulin levels up to 15 times greater in their beak primordia than the blunt -beaked finches.

Calmodulin is a calcium -binding protein that promotes elongation and pointing in the cartilage.

So this means that natural selection is effectively targeting the enhancers that control the quantitative output of just BMP4 and calmodulin in that one facial module.

That's right.

Changing the dial on those two signals accounts for the huge morphological diversity across the entire lineage.

It's a remarkable demonstration of how evolution can achieve these macroevolutionary changes through simple quantitative tweaks to existing control genes.

And finally, that brings us to D, heterotypy, a change in kind.

This is where the mutation actually affects the protein coding region of a gene, and it alters the function of the protein itself.

As you said earlier, this is the riskiest type of change.

It is, because it so often leads to a pleiotropic disaster.

But when a coding region change happens in a gene that's only active in a few tissues, or when the change grants some profound new function, it can be stabilized by selection.

The evolution of internal gestation in placental mammals is a great example of this.

Right.

Internal gestation, pregnancy,

it requires the mother's body to manage this incredibly complex relationship with the fetus.

Immune suppression, nutritional support, all mediated by hormones like prolactin.

Exactly.

And key to this whole process is the HOX11 gene, a transcription factor.

In the lineage that led to placental mammals, the HOX11 protein sequence actually changed.

But the change didn't give it a completely new job, it changed its interactions.

It changed who it partners with.

Precisely.

The new mammalian HOX11 protein gained the ability to stably associate with another transcription factor, FOXA1a.

And this new HOX11 -FoxA1a complex is required to turn on the expression of prolactin, specifically in the uterine cells, which allows them to become the decidual cells that are essential for maintaining a pregnancy.

And the HOX11 from non -placental relatives, like an opossum or a chicken, can't form that complex.

It can't.

So a mutation that altered the protein's ability to bind a partner molecule, a heterotype, was absolutely essential for the evolution of the mammalian uterus.

Wow.

And there's one more defining heterotype that explains the insect body plan, right?

Specifically, why they have exactly six legs.

Yes, unlike most other arthropods.

Spiders have eight, centipedes have many more.

What stops insects from forming legs on their abdomen?

What is it?

It's a heterotypic change in a highly conserved HOX gene called ultra -abithrax, or UBX.

In insects, UBX is expressed in the abdominal segments.

And the key mutation in the insect lineage added a small polyalanine region to the end of the UBX protein.

And what does that little addition do?

It acts as a gain -of -function mutation.

It gives UBX a new inhibitory role.

It allows UBX to actively repress the expression of another gene, distalus, in the abdomen.

And distalus is the gene required to start making a leg.

So by repressing distalus in the abdomen, the insect is constrained to only having six thoracic legs.

Exactly.

And the proof is that when researchers engineered the UBX gene of a brine shrimp, which normally has abdominal legs, to include that specific insect polyalanine region, the resulting shrimp embryo repressed distalus in its abdomen and lost its abdominal legs.

That one single heterotypic change is a defining feature of the entire insect class.

Okay, so we've established this enormous flexibility and the variety of ways evolution can use the developmental toolkit.

But now we have to look at the other side of the coin.

What can't happen?

Let's talk about developmental constraints.

I mean, why are there only a few dozen major animal body plans?

Why are some anatomical structures just never seen?

Constraints are what limit the range of possible viable phenotypes.

They're the guardrails on the evolutionary road and they come in three main flavors.

Physical, morphogenetic, and pleotropic.

Physical constraints seem like the easiest to grasp.

They're just dictated by the laws of physics.

Absolutely.

You're never going to find a six -foot tall insect because the physics of diffusion limits how big an animal with an external respiratory system can get.

You're never going to find a vertebrate with functioning rotating limbs like a wheel.

Because you can't maintain blood flow and nerve function to something that's continuously rotating.

Exactly.

It just violates basic physiological and physical limits.

Morphogenetic constraints are a bit more subtle.

These are imposed by the construction rules of the developmental program itself.

Think about our own limbs, vertebrate limbs.

They're built according to very strict developmental rules.

For example, in 300 million years of vertebrate evolution, a middle digit has never evolved to be shorter than the digits on either side of it.

The forearm bones are never found more proximal than the upper arm bone.

The underlying chemical instructions for patterning the limb only allow for variation within a certain set of boundaries.

And a major mathematical source for these constraints comes from Alan Turing's work back in 1952, the reaction diffusion model.

Yes.

Turing mathematically proved how you can get patterns, stripes, spots, even limb digits to arise spontaneously from two uniformly distributed chemical substances.

It requires two components.

An activator, we'll call it P, and an inhibitor, S.

Okay.

P stimulates the production of more P and also more S.

But S has to inhibit P, and crucially, S has to diffuse through the tissue much faster than P does.

So the fast -moving inhibitor stops the activator from creating peaks that are too close together.

Precisely.

P starts to form a concentration peak, but the resulting surge of the fast -moving inhibitor, S, rushes out and suppresses any other P peaks from forming nearby.

This process generates these stable periodic standing waves of concentration.

That's what defines the spacing of your fingers or the arrangement of scales on a fish.

But the math dictates that based on the size and shape of the tissue and the diffusion rates, only certain patterns or wavelengths are actually possible.

It's like the mathematical limits on the resonant frequency of a musical instrument.

You can only produce certain notes based on the size and shape of the instrument.

And if evolution tries to produce a pattern that falls outside those permitted wavelengths, the organism just fails to develop correctly.

The constraints are literally written into the chemistry and geometry of the tissue.

So finally, let's talk about the very practical pleiotropic constraints.

This is where a single gene, if you change it, causes a whole cascade of lethal failures in multiple critical systems.

And this brings us back to that really striking example of the mammalian cervical vertebrae.

Almost all mammals, from a tiny shrew to a giant long -necked giraffe, have exactly seven cervical or neck vertebrae.

Why is that number so incredibly fixed?

Why can't evolution just give us eight or reduce a giraffe's neck to six?

Because the hox genes that specify where the neck ends and the thorax begins, the number of cervical ribs are also highly pleiotropic.

They're active in other critical modules, including those that control stem cell proliferation.

And what's the consequence if you do vary that number?

Well, statistical analysis of human populations shows that embryos that vary from that norm of seven have an extremely high rate of prenatal death.

And for the few who do survive, there's a significantly elevated risk of developing childhood cancers.

Wow.

So the gene expression change needed to alter the skeleton also messes up cell growth and survival.

It does.

So selection is incredibly strong against any variation in this trait.

And that's what cements seven as the fixed, unchangeable number for mammals.

For this final part of our deep dives, we really need to broaden our definition of evolution's raw material.

It's not just about simple DNA sequence changes.

We've seen how IboDivo connects these microscale changes to macroscale effects.

But what if the source of selectable variation comes from?

Well, from outside the genome entirely.

Let's look at selectable epigenetic variation.

Epigenetic inheritance is fascinating.

It allows changes that are induced by the environment or by the internal state of a cell to be passed across generations without changing the underlying DNA sequence at all.

It works on the chromatin structure, how the DNA is packaged in red.

And this offers a really potent pathway for evolution to act on existing developmental plasticity.

The first example of this is something called epiallelies.

These are inherited variants of chromatin structure usually passed on through DNA methylation.

Right.

Think about the toadflax plant.

It can have this rare asymmetrical flower structure called poloria.

It's been known since the 18th century and it's stably inherited.

But it's not caused by a mutation in the cycloidea gene, which normally controls flower symmetry.

So what's the cause?

It's caused by the hypermethylation of that gene.

The gene is so heavily methylated that it's basically locked down, non -functional.

It behaves just like a mutated allele, but the DNA sequence itself is totally normal.

And this methylation pattern, this epiallel, is stably transmitted through the seeds.

That seems like a perfect way for the environment to leave a lasting mark.

It is.

We see this with environmental triggers that affect the germline.

Diet, for instance.

If you feed a pregnant mouse a diet that's rich in methyl donors, the specific methylation pattern at the agouti locus, which controls coat color and obesity risk, is transmitted through the egg not just to her immediate pups, but to the grand offspring.

That is true intergenerational inheritance through something other than the primary DNA code.

Absolutely.

And likewise, exposure to endocrine disruptors like vinclozolin can alter the DNA methylation patterns in the germline of male mice.

And these altered patterns persist and cause developmental problems like prostate and kidney defects in their grand offspring, and even their great grand offspring, who are never exposed themselves.

That's a staggering implication.

Environmental input today could determine selectable traits three generations from now.

The second epigenetic system you mentioned involves symbiont variation.

This is when an organism's phenotype is actually conferred by inherited symbiotic microorganisms.

They're usually passed down maternally through the egg cytoplasm.

And the key advantage here is that microorganisms evolve incredibly quickly, fast generation times, high rates of horizontal gene transfer.

So they basically act as a rapid response adaptive system for their host.

Exactly.

Take the pea aphid.

They harbor all sorts of different bacterial symbionts.

One of them, rickettsiella, has strains that can change the aphids color from red to green, which provides selectable camouflage depending on what plant they're on.

Another symbion, hamiltonella defensa, gives them robust protection against parasitic wasps.

The aphid inherits this adaptive advantage from its microbe, not from its own slow evolving genome.

And the last concept here ties all of these ideas together, genetic assimilation.

This is where an environmentally induced plastic phenotype becomes fixed or canalized by selection into the genotype itself.

Right.

This is the idea that developmental plasticity, the ability of an organism to change its form in response to the environment, actually paves the way for genetic evolution.

The environmentally triggered trait is already widespread in the population.

It's already on subtle genetic modifiers that make that trait appear even without the environmental trigger.

And the classic proof of this came from Conrad Waddington's famous experiments with food flies.

Yes.

Waddington exposed developing fly embryos to ether vapor.

This caused what's called a phenocopy, a non -genetic mimic of the bithorax mutation.

It resulted in flies with four wings instead of two.

He then selectively bred the flies that showed this four wing trait generation after generation, all while still exposing them to ether.

And then eventually he took the ether away.

And after about 20 generations, he had flies that produced the four winged phenotype, even when no ether was applied.

The plastic environmentally induced trait had become genetically assimilated.

Later research showed this happened by selecting for underlying alleles of the ultra -bithorax gene that just made the developmental system easier to trigger, even by its own internal fluctuations.

And we see this happening in the wild with the Australian tiger snake.

We do.

On the mainland, tiger snakes have mixed diets.

They have developmental plasticity.

They can grow large heads if they consistently eat large prey, but they're born with smaller heads.

But on islands where only large prey is available, selection for large heads is intense.

And within just a few thousand years, the island snakes have become genetically fixed or canalized.

Meaning they're just born with large heads, no matter what their diet is.

That's right.

They are born large headed, even if you feed them small prey in a lab from birth.

This whole process eliminates that long random waiting period for a large head structural mutation to just pop up.

It allows evolution to accelerate by capitalizing on a developmental capacity that's already there in the population.

This deep dive into IvoDivo really shows how integrating development in genetics provides that essential missing link in Darwin's theory.

I think we've really delivered on the promise made by those 19th century pioneers like Vilhelm Rue.

We now have the molecular and the mechanical understanding to explain how a mutation actually translates into a complex, viable, and selectable phenotype.

So we've established that ontogeny, an individual's development, doesn't just recapitulate phylogeny or evolutionary history.

We now understand, as the biologist Walter Garstang said, that ontogeny creates phylogeny.

Yes.

By combining population genetics, the study of selection, with developmental genetics, the study of construction, we can finally start to explain the construction and the vast scope of biodiversity.

We've seen so many examples of evolution, just kinkering, relocating ribs to make a turtle shell, adjusting the timing of growth factors to form a dolphin's flipper, or changing the amount of BMP to sculpt a finch's beak.

But let's end with a final provocative thought, and this is about the origin of true novelty.

Tinkering explains modifications, but what about completely new cell types, like the vertebrate neural crest cell, which forms all the bones of your face, or the mammalian uterine decidual cell, which is absolutely essential for pregnancy?

Those are fundamental new developmental subroutines, and the latest research suggests that incredibly complex enhancer modules that you need to regulate the timing and location of these novel genes might not have arisen just through internal mutation and rearrangement of existing DNA.

So what's the alternative hypothesis?

Where do they come from?

Well, it's speculated that these crucial new enhancers might have actually been acquired from external sources, introduced into our ancient ancestors' genomes by viral insertions, or through genetic contributions from symbiotic microbes.

So you're saying the building blocks for macroevolution, for the sudden leak to a new body component, could literally arrive in our genome via a virus or a microbe.

It forces us to acknowledge that the raw material for the next great evolutionary innovation might be acquired horizontally, not just vertically through inheritance.

The boundary between what we consider self and non -self is far, far blurrier than we ever imagined.

A truly stunning synthesis to end on.

That wraps up our deep dive into development and evolution.

Thank you for joining us for a concise recap of the most important ideas and a warm thank you from the Last Minute Lecture Team.