Chapter 17: Control of Gene Expression in Eukaryotes

Welcome to Last Minute Lecture.

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

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

For complete coverage, always consult the official text.

Humans and chimpanzees shared a common ancestor just five to seven million years ago, which, I mean, in an evolutionary time, that is practically yesterday.

Oh, absolutely.

A blink of an eye.

Right.

And when you sequence our genomes, our DNA is 96 % identical.

96%.

So how do you get a human instead of a champ?

That's a huge question.

If the structural blueprint is basically exactly the same, why are we so anatomically and cognitively different?

Welcome to the Deep Dive.

You are listening right now because you are staring down the barrel of a major genetics exam, and you need to get eukaryotic gene expression locked into your brain.

And we are here to help.

Exactly.

We are the Last Minute Lecture Team, and our mission today is to help you conquer Chapter 17 by finding the underlying logic of how your genes are actually controlled.

And that chimp -human paradox is actually the perfect place to start because scientists wrestled with that exact puzzle for decades.

Back in 1975, geneticists Mary Claire King and A .C.

Wilson hypothesized that the differences weren't really in the structural genes themselves.

So not the genes making the actual proteins.

Right.

They thought the difference was in how those genes are regulated.

They just, well, they didn't have the technology to prove it at the time.

But they do now, right?

Yeah.

Fast forward to 2009, and a researcher named Kutynowik and her colleagues finally proved them right.

They found that the big evolutionary differences lie in a relatively small number of regulatory sequences.

Wow.

Specifically, they identified a group of transcription factors called KRABZFPs that evolved very rapidly in humans.

And these proteins act like master switches.

They control massive interconnected networks that shape everything from energy metabolism to the physical development of the human brain.

Okay, so it's not the structural lumber that's different.

It's like the foreman on the construction site deciding when, where, and how fast that lumber gets nailed together.

That is exactly the core of it.

To understand what makes eukaryotic organisms, you know, humans, plants, yeast unique, we really have to look at the differences in how we regulate our genomes compared to something simpler, like bacteria.

Right, because in bacteria, genes are often clustered together into operons.

Right.

Meaning, they share one promoter and get transcribed as a single long unit.

Exactly.

Plus, bacteria don't have a nucleus.

So the moment the RNA is being made, ribosomes are just like already latching onto it to build proteins.

Which is incredibly efficient for a single -celled organism that's just trying to multiply rapidly.

But eukaryotic cells face completely different challenges.

We don't usually use operons.

Each gene has its own promoter.

Right, each gene tends to have its own promoter.

And transcription happens inside the nucleus, which creates a physical barrier from translation, which is out in the cytoplasm.

But the defining challenge, I mean, the absolute biggest hurdle of eukaryotic gene expression is chromatin.

Because there's just so much DNA.

So much.

Our DNA is so long that to fit inside the nucleus, it has to be tightly schooled around histone proteins, forming this dense structure called chromatin.

In its default state, this microscopic fortress is so packed that the transcription machinery can't even access the genes.

Which kind of feels like a design flaw.

I mean, if the DNA is wrapped up that tightly, how does the cell ever manage to read the instructions?

It literally has to physically force the chromatin open.

Yeah.

And researchers can actually see where this happens.

As genes prepare to become active, the regions around them become highly sensitive to an enzyme called DNA's eye, which cuts DNA.

Oh, so they're called DNA's eye hypersensitive sites.

Exactly.

Their existence basically proves that the chromatin relaxes into an open configuration, usually just upstream of the transcription start site.

That exposes the DNA so regulatory proteins can land.

Okay, but how is the cell doing that physically?

Is it just, you know, ripping the histones away?

There are three main mechanisms the cell uses to breach that chromatin barrier.

The first is chromatin remodeling.

If these massive protein complexes, the best studied one is called sepUISNF.

SepUISNF?

Yeah, and they don't change the chemical makeup of the histones.

Instead, they use energy from ATP hydrolysis to act like microscopic motors.

Wait, like a motor, are they sliding the DNA around?

Essentially, yes.

They bind to the DNA and the nucleosome, the DNA histone spool, and use ATP to physically twist and pull the DNA rope.

That's wild.

It is.

This can slide the nucleosome down the DNA strand or alter its conformation, which pushes the histones out of the way to expose the promoter region underneath.

Okay, so that's basically the brute force method.

But there's a second mechanism that seems a lot more elegant, right?

Histone modification, or what's often called a histone code.

Yes, the histone code.

I'm trying to visualize the chemistry here.

The histone proteins have these positively charged amino acid tails sticking out, and the DNA backbone is negatively charged, so they stick together like magnets.

That magnetic attraction is exactly what keeps the chromatin locked down.

Okay, so if the cell wants to open a gene,

could we think of it like greasing the skull,

adding some sort of chemical to those tails so the DNA thread just unwinds?

That is a brilliant way to picture it.

And the grease, in this case, is usually an acetyl group.

When enzymes called acetyltransferases add acetyl groups to those positively charged histone tails, it neutralizes their positive charge.

So the magnetic attraction to the DNA weakens.

Right.

It disrupts the entire 30 nanometer chromatin fiber, destabilizing the structure and turning genes on.

And conversely, enzymes called deacetylases can strip those groups away.

To restore the positive charge and turn genes off.

Exactly.

They repack the chromatin.

What about methylation?

Because I see that mentioned all the time alongside acetylation.

Methylation is adding methyl groups to those same histone tails, but its effect depends on exactly which amino acid gets methylated.

Think of it less like grease and more like signaling flags.

Why do they?

Yeah, for instance, a specific modification called H3K4Me3, which just means adding three methyl groups to lysine 4 on histone H3, is a flag frequently found near the promoters of active genes.

It acts as a beacon, recruiting other proteins that help open the chromatin.

I want to make sure this makes sense in a real biological system, rather than just abstract chemistry.

The textbook brings up the Arabidopsis plant and how it controls when it flowers.

That's a great example.

So there's a gene called FLC that essentially acts as an emergency break on flowering.

As long as FLC is active, the plant won't flower, because it's evolutionarily programmed to wait out the cold of winter.

Yes, FLC creates a repressor protein that actively blocks flowering.

But then you have this other gene, FLD.

And FLD encodes one of those decetylase enzymes we just talked about.

Exactly.

So when the plant has experienced enough winter cold and it's finally time to bloom, FLD produces its enzyme.

That enzyme travels to the FLC gene,

strips the acetyl groups off its histones, restores that magnetic charge, and tightly packs the chromatin around FLC.

Repressing the repressor.

Exactly.

It shuts off the emergency break and the plant finally flowers.

It's just amazing how a tiny chemical change on a histone dictates an entire season of growth.

Now the text also mentions a third mechanism,

DNA methylation.

How is that different from what we just talked about?

It's a really crucial distinction.

Histene methylation modifies the protein spools.

DNA methylation is adding methyl groups directly to the cytosine bases of the DNA sequence itself.

Oh, directly onto the DNA.

Right.

This happens heavily at areas called CPG islands, which are regions rich in cytosine and guanine nucleotides usually found near promoters.

And while histone modification can turn things on or off, heavy DNA methylation almost exclusively locks genes down, repressing transcription long -term.

Okay.

I have to pause here because this always bugs me when I study molecular biology.

How do we actually know this?

Like how does a scientist point at a genome and say, ah, yes, there is an acetylated histone sitting exactly 200 base pairs upstream of this specific gene?

Well, they use a technique called chromatin immunoprecipitation, or Chi -Chi for short.

It's essentially molecular fishing.

Molecular fishing.

Let's look at crosslink Chi -Chi or X -Chi -P.

First, scientists treat living cells with

This is a fixative.

It freezes everything in place by creating strong covalent bonds between the DNA and whatever protein happens to be touching it at that exact millisecond.

So they basically freeze the action, then what?

Then they elise the cell, breaking it open, and use sound waves to smash the long chromatin strands into tiny fragments.

So now you have a soup of fragmented DNA, some of which has proteins glued to it.

Okay.

I'm with you.

To find the specific protein they want, say, a transcription factor they drop in antibodies designed to grab only that one specific protein.

Oh, I see.

Because the protein is permanently glued to the DNA by the formaldehyde.

When the antibody fishes out the protein, the DNA comes along for the ride.

You've got it.

You pull out the antibody,

reverse the crosslinking to separate the protein and the DNA, and then you sequence the DNA that's left over.

That's so smart.

Right.

When you map those short sequences back to the entire human genome, you see peaks of sequence reads at specific locations.

That proves your protein was physically sitting at that exact spot when the formaldehyde hit.

And is that the only way to do it?

No.

There's also native ChiIP, which skips the crosslinking step.

That's usually used to find those modified histones we just discussed, since histones are naturally tightly wrapped by the DNA anyway.

That is incredibly clever.

All right.

So the chromatin barrier is breached.

The DNA is exposed.

We are entering the transcription control room.

But what actually presses the start button for RNA polymerase?

You need two main components.

First is the basal transcription apparatus.

This is a massive complex of RNA polymerase and general transcription factors that assemble at the core promoter right next to the gene.

But they don't do much on their own, right?

No.

This apparatus is pretty sluggish on its own.

It only drives a minimal baseline level of transcription.

To blast transcription into high gear, you need specific transcription factors binding to the regulatory promoter located further upstream or to distant sequences called enhancers.

The yeast GAL4 system is a really great way to visualize this switch.

GAL4 is a specific transcription factor that yeast cells use to turn on the genes for digesting galactose, which is a sugar.

Right.

GAL4 has these physical structures called zinc fingers that perfectly fit into a specific DNA sequence called UASG.

So GAL4's entire job is to sit on that UASG sequence and stimulate the basal transcription apparatus.

Right.

It wants to start the party.

But the cell doesn't want to waste energy making galactose digesting enzymes if there's no galactose around.

So it produces a second protein, GAL80, which acts like a bouncer.

A bouncer.

I like that.

Yeah.

GAL80 binds directly to GAL4, physically blocking it from interacting with a transcription apparatus.

So the switch is basically in the off position.

But then you eat some sugar and galactose, the VIP, enters the yeast cell.

The galactose binds to a third protein, GAL3.

This galactose GAL3 complex walks right up to the GAL80 bouncer and interacts with it.

Causing a conformational change.

Exactly.

This causes GAL80 to change its shape and let go of GAL4.

Suddenly GAL4 is free to interact with the basal apparatus and transcription skyrockets.

It's a beautifully responsive biological circuit.

Now, you brought up enhancers earlier, and they present one of the weirdest physical realities of the genome.

They really do.

Enhancers are regulatory sequences that can act from huge distances, thousands, sometimes hundreds of thousands of base pairs away from the core promoter they control.

And this is where I always get tripped up.

If an enhancer works from half a million base pairs away,

the DNA has to physically loop backward so the proteins on the enhancer can touch the basal apparatus.

Correct.

But if the DNA is just a big plate of spaghetti looping around, how does the enhancer not accidentally bump into the wrong promoter and turn on a completely unrelated gene nearby?

That is the exact problem the cell has to solve.

And the solution is insulators.

Insulators are DNA sequences that block the effect of an enhancer, but only if they are positioned between the enhancer and the promoter.

Okay, so how do they actually block it?

They achieve this by organizing the genome into topologically associated domains, or TADs.

TADs.

How do those work?

Specialized proteins like CTCF in mammals bind to the insulator sequences and physically pinch the chromatin into massive loops.

Think of TADs as isolated neighborhoods.

An enhancer can freely float around and interact with any promoter inside its own looped neighborhood.

But the CTCF proteins at the base of the loop act like a fence.

They prevent the enhancer from looking outside its TAD to activate a promoter in the next neighborhood over.

So the genome isn't just a linear string.

It's a 3D city with distinct walled off zip codes.

Okay, so let's say the enhancer connects, the apparatus fires up, and RNA polymerase takes off.

Is it just a straight sprint to the end of the gene?

Not necessarily.

Sometimes RNA polymerase starts transcribing, gets 24 to 50 nucleotides down the DNA track, and then it just, well, it spalls.

It pauses and waits.

Why would it do that?

I mean, why start if you aren't going to finish?

Speed of response.

A classic example is the heat shock genes in Drosophila fruit flies.

These genes produce proteins that protect the cell from extreme stress, like sudden heat.

So they need to be fast.

Very fast.

By preloading RNA polymerase and letting it stall just past the start line, the cell is primed.

It doesn't have to wait for chromatin remodeling or transcription factor assembly when a crisis hits.

The moment the heat stress arrives, the stalled polymerases are released simultaneously to flood the cell with protective proteins instantly.

Which perfectly sets up the concept of coordinated control.

We noted earlier that eukaryotes don't have operons to turn on multiple genes at once, like bacteria do.

So how does a eukaryotic cell coordinate a massive response to a single stressor?

Instead of clustering genes together, eukaryotes use response elements.

These are short consensus sequences found in the promoters or enhancers of different genes scattered all over the genome.

Let's look at the metallothionane gene, which makes a protein that protects cells from heavy metal toxicity.

I was reading this part and it felt like the gene is an employee answering to multiple different Let's follow that logic.

Upstream of this single metallothionane gene, you have multiple copies of a sequence called a metal response element, or MRE.

When heavy metals enter the cell,

specific transcription factors bind those MREs and spike transcription.

But right next to the MREs, there is also a TE response element that responds to a completely different protein called AP1, and a GRE element that responds to steroid hormones.

So because this one gene has multiple different response elements in its promoter, it can be activated by heavy metals or by AP1 or by hormones.

It responds to whatever boss is shouting at the moment.

Yes.

And the flip side of that coordination is even more powerful because the exact same MRE sequence is placed in front of dozens of different genes across the entire genome.

A single signal heavy metals entering the cell causes transcription factors to bind to MREs everywhere simultaneously.

So the cell mounts a massive coordinated defense without needing a bacterial operon.

Exactly.

We've covered a lot of ground, but the RNA has finally been transcribed.

You might think the story is over, but we are just entering the post -transcriptional editing room.

Just because you have a pre -mRNA strand doesn't mean you know what protein you're actually going to get.

Far from it.

Alternative splicing allows a single gene to produce multiple functionally different proteins.

And this is not some rare exception.

An estimated 95 % of multi -exon genes in humans undergo alternative splicing.

95%.

Yeah.

The cell uses consensus sequences at the splice sites and specialized regulatory proteins like SR proteins to dictate exactly which exons are kept as the final instructions and which are thrown in the trash.

The textbook's case study on this is the Drosophila sex determination cascade.

And it is just a masterpiece of biological logic.

It revolves around three genes acting like falling dominoes.

It starts with the sex -lethal gene, or XXL.

How does the fly decide whether to turn XL on or off?

It's based on the ratio of X chromosomes.

In female embryos, which have two X chromosomes, the XXL gene is activated and transcribes a functional XXL protein.

This XSL protein then travels over to the pre -mRNA of the second gene in the cascade, called Trey.

The SSL protein physically binds to the Trey RNA and blocks the default splicing site, forcing the spliceosome to use a downstream site instead.

This specific alternative cut creates a functional Trey protein.

And that Trey protein continues the cascade.

Right.

The Trey protein, along with another factor, goes to the pre -mRNA of the third gene, DSX, and directs it to splice into a female -specific version.

This ultimately produces a female fruit fly.

But the real genius is what happens in the male.

Walk us through the male cascade.

In an XY male embryo, the SSL gene is never turned on, so there is no XSL protein.

Without it, the spliceosome looks at the Trey pre -mRNA and just uses the first default splice site it sees.

But using that upstream site is a fatal error for the protein, isn't it?

Yes.

Because that default upstream segment happens to contain a premature stop codon.

When the ribosome tries to translate that mRNA later, it hits the stop codon early, aborts the process, and produces a broken, non -functional Trey protein.

Because there's no functional Trey protein to guide the final step,

the DSX pre -mRNA defaults to a male -specific splice pattern, causing male traits to develop.

It is a brilliant system, one pre -mRNA, entirely different body plans,

purely based on where you cut the tape.

It's an elegant cascade.

But, you know, even after splicing is finished, the mRNA has to survive the shredding room, RNA degradation.

The amount of protein a cell can synthesize doesn't just depend on transcription rate.

It depends on how long the mRNA survives in the cytoplasm.

The mRNA's lifespan is basically a ticking clock.

And the countdown timer is the poly -A tail at the end of the RNA.

Usually, yes.

The mRNA has a five prime cap at the front and a long tail of adenine nucleotides at the back.

Poly -A binding proteins cling to that tail and physically protect the five prime cap by looping the RNA around.

Okay, so they guard the front by holding the back.

Right.

But over time, cellular enzymes slowly chew away the poly -A tail.

Shortening the timer.

Exactly.

Once the tail gets critically short, the poly -A binding proteins fall off.

Without their protection, an enzyme sweeps in and chops off the five prime cap.

Without the cap, nucleases rapidly chew the mRNA to pieces from the five prime and down.

That's brutal.

Much of this destruction takes place in specialized cellular structures called P -bodies.

Interestingly, P -bodies can also temporarily sequester and store mRNA, pausing translation until the cell needs it later.

Okay, we are in the home stretch.

What if the cell wants to silence an mRNA that's already out there before the poly -A tail naturally degrades?

It uses RNA interference, or RNAi.

This mechanism is massive.

It's estimated to regulate up to 30 % of human genes.

The cell produces tiny RNA molecules, specifically small interfering RNAs CERNase and microRNAs mRNAs.

These small RNAs form complexes with proteins and physically pair up with complementary sequences on the target mRNA in the cytoplasm.

Like a heat -seeking missile finding its exact target.

Precisely.

If the match is perfect, the complex uses an enzyme called slicer to cleave the mRNA right down the middle, destroying it.

And if it's not a perfect match?

If the match is imperfect, which is common with mRNAs, the complex just parks itself on the mRNA, physically blocking the ribosome so the instructions can never be translated.

This is vital for biological development.

Oh, like the heart example.

Right.

There's a specific microRNA called MIR11 that controls the programmed growth of the vertebrate heart.

If that microRNA fails to silence its target transcripts at the exact right time, the heart tissue develops abnormally.

Wow.

And finally, even if the mRNA survives the P bodies and the microRNAs and arrives at the ribosome, the cell still has one last checkpoint.

Translational control.

The final gate.

Right.

Translation can be throttled by the sheer availability of ribosomes charged tRNAs or initiation factors.

The textbook's example of d -cells perfectly illustrates why this is useful.

You have millions of t -cells, which are vital immune cells circulating in your blood right now.

Most are sitting dormant in the G0 phase of the cell cycle.

Okay.

They have transcribed the mRNAs they need to mount an immune response, but they aren't translating them.

Why bother making the mRNA if you aren't going to use it?

Because transcribing DNA and processing chromatin takes time.

By having the mRNA already waiting in the cytoplasm, the t -cell is a loaded spring.

Ah, I see.

When a viral antigen finally binds to the t -cell receptor, it triggers a sudden massive increase in the availability of initiation factors, specifically one called EIF3.

So EIF3 is the missing key.

Once it's available, it allows the ribosomes to finally bind to all those waiting mRNAs.

And protein synthesis spikes seven to tenfold almost instantly.

Which allows the t -cell to rapidly divide and fight the infection much faster than if it had to start all the way from scratch at the chromatin level.

And just to close the loop on the entire process, even after the protein is completely built, you have post -translational modifications.

It really never stops.

The control literally never stops.

The cell can attach a small protein called ubiquitin to mark a protein for immediate destruction or add methyl and acetyl groups to the protein to change its folding and function.

So bring this all back to the chimpanzee.

We are 96 % identical in our DNA sequence.

The evolutionary difference isn't just about having slightly different protein building blocks.

The difference is the entire multilayered control board we just walked through.

It's the chromatin unwinding just a little bit differently in a human embryo.

It's the enhancers looping to different neighborhoods via TADs.

Exactly.

It's the alternative splicing, cutting the brain development instructions differently, and the microRNAs silencing certain messages while leaving others alone.

That 4 % difference in regulatory sequences runs interconnected networks that dictate building a human instead of a chimp.

And this leaves us with an incredible realization about our own lives.

We spent a lot of time talking about the histone candidate and CPG methylation.

Right, the epigenetic changes.

Yes.

These epigenetic changes dictate gene expression by opening or closing chromatin without ever altering your underlying DNA sequence.

We know these marks can be heavily influenced by the environment.

Really?

Like what?

Well, consider how the environmental factors in your own life, periods of high stress, your diet, exposure to toxins, might be actively modifying your epigenome right now.

Oh, wow.

You are constantly adding or removing those acetyl and methyl groups, potentially altering how your very own genes are being expressed at this exact moment.

That is wild.

Your genome isn't a static textbook you just read from start to finish.

It's a living, breathing control room reacting to everything you do.

And with that, you have a conceptual roadmap to conquer eukaryotic gene expression.

Thank you, and good luck on your genetics exam from the Last Minute Lecture Team.