Eukaryotic gene regulation

Concept

Eukaryotic gene regulation

T-077Home BU-303Threads information

Statement

Enhancers, chromatin and transcription factors.

Why it matters

operon-model already showed how bacteria regulate gene expression using a single operator and repressor/activator, sufficient for a genome with relatively few genes and one basic cell type. Eukaryotic organisms face a harder problem: a single genome must specify hundreds of distinct, stably different cell types from the identical DNA sequence in every cell, and eukaryotic gene regulation is the more elaborate, combinatorial regulatory system that makes this possible — the direct mechanistic basis for epigenetic-inheritance's claim that expression states, once set, can be stably propagated.

Understanding combinatorial control is also what makes sense of a otherwise puzzling numerical fact: the human genome encodes on the order of only 1,500–2,000 distinct transcription factors, yet these are sufficient to specify many hundreds of distinct, stable cell identities, because regulation depends on which combination of factors is present, not on a one-gene-one-regulator mapping.

Hypotheses

Cis-regulatory DNA elements (promoters, enhancers, silencers) are recognised by sequence-specific DNA-binding transcription factors.This is the basic molecular interaction the entire system is built on: a transcription factor's effect on a gene depends on whether its specific binding sequence is present in that gene's regulatory DNA, and whether that DNA is currently accessible. Chromatin accessibility is a prerequisite for transcription-factor binding.DNA wrapped tightly around histones in compact chromatin is generally inaccessible to the transcription machinery until locally remodelled or opened; this is the direct link to epigenetic-inheritance, since a stably closed or open chromatin state is exactly what propagates a gene's expression status across cell division. Enhancers can act on a promoter over large genomic distances and independent of orientation, via DNA looping.Unlike a bacterial operator, which sits immediately adjacent to the gene it controls, a eukaryotic enhancer may lie tens or hundreds of kilobases away, or even within an unrelated neighbouring gene, and still regulate a distant target — a major complication for identifying which enhancer controls which gene from sequence position alone.

Proof

\text{The core promoter positions RNA polymerase II and the general transcription factors to set the transcription start site.}

A minimal set of general factors assembles at the core promoter (e.g. around a TATA box or other core promoter element) in essentially every gene, providing the basic machinery for transcription initiation, but generally only at a low, unregulated basal rate on its own. A

\text{Enhancers, often distant, bind combinations of activator transcription factors; DNA looping brings the bound enhancer complex into contact with the promoter.}

Physical looping of the intervening DNA (mediated by looping factors and the Mediator complex) allows an enhancer's bound activators to directly contact and stimulate the promoter-bound general machinery from Step 1, converting the low basal rate into a strong, regulated rate of transcription. A

\theta = \frac{[TF]^n}{K^n+[TF]^n}

Occupancy of a given binding site by its transcription factor follows a saturating, often cooperative (Hill-type) relationship with the factor's concentration; a gene's overall transcription rate is then set by the specific combination of sites occupied at a given time, not by any single site or factor alone — a small number of factors, combined in different subsets across different genes, is enough to specify a very large number of distinct expression outcomes. B

\text{Chromatin remodelling complexes and histone-modifying enzymes act together with transcription factors to open or close local chromatin accessibility.}

A gene's regulatory region must first be made accessible (Hypotheses) before the factors of Steps 1–3 can bind at all; once established, this accessible or inaccessible state can be maintained across subsequent cell divisions by the epigenetic-inheritance mechanisms this same accessibility decision feeds directly into. A

\text{Post-transcriptional regulation (splicing choice, mRNA stability, microRNA-mediated repression) adds further layers of control after transcription itself.}

Even once an mRNA is transcribed, its final protein output can still be tuned or blocked entirely by regulation acting after transcription, so the transcriptional combinatorics of Steps 1–4 are not the only regulatory layer determining a gene's ultimate expression level. A

Result

\text{Expression level} = f(\text{chromatin accessibility}, \{\text{TFs bound at promoter + distal enhancers}\}, \text{post-transcriptional control})

Reading. A eukaryotic gene's expression is not a single on/off switch but the combined output of chromatin state, which specific combination of transcription factors is bound at its promoter and enhancers, and further post-transcriptional processing — giving a relatively small toolkit of regulatory proteins the combinatorial power to specify a very large number of distinct cell identities.

Scope. Applies across eukaryotic genes broadly; the specific enhancer(s) controlling a given promoter, and the distance/orientation over which they act, must generally be determined experimentally rather than read directly off genomic position alone (Hypotheses, t3).

Corollaries & converses

operon-model's single-operator logic is the simpler, prokaryotic special case of this same general principle (regulatory DNA element bound by a sequence-specific protein controlling a promoter) without the added combinatorial and distal-enhancer complexity eukaryotic regulation requires.
epigenetic-inheritance directly extends Step 4: once chromatin accessibility is set during development, it can be propagated across cell division without needing every one of the original regulatory signals to remain continuously present.
crispr-cas9 targeted to a promoter or enhancer, rather than a coding sequence, can be used experimentally (or therapeutically) to directly test or manipulate which regulatory element controls a given gene.

Fails without

Drop the chromatin-accessibility requirement (Hypotheses): if transcription factors could bind their target sequence regardless of chromatin packing state, genes would be constitutively "readable" at all times, and the stable, heritable silencing that epigenetic-inheritance describes (a closed chromatin state blocking factor binding) would have no mechanism to actually prevent transcription.
Drop distance/orientation-independent enhancer action (t3 Hypothesis): if regulatory elements could only act when immediately adjacent to their target promoter (as in the bacterial operator model), the many disease-associated regulatory variants found by genome-wide association studies at large distances from any gene would have no plausible mechanism for influencing expression at all.

Common errors

Assuming eukaryotic genes are controlled by a single, simple on/off operator as in the bacterial operon-model, rather than by combinatorial input from multiple factors and elements (Step 3).
Assuming an enhancer must be located immediately next to the gene it regulates, rather than potentially tens or hundreds of kilobases away and reachable only via looping (Hypotheses, t3).
Confusing a promoter (where general transcription machinery assembles, Step 1) with an enhancer (a distal element that boosts transcription rate via looping, Step 2) — the two are functionally and often physically distinct regulatory DNA elements.
Assuming a single transcription factor determines a gene's expression outcome on its own, rather than recognising that the specific combination of factors bound together is generally what sets the outcome (Step 3).

Discussion

François Jacob and Jacques Monod's 1961 operon model first established the basic logic of gene regulation by a sequence-specific DNA-binding protein, in bacteria. As molecular tools extended to eukaryotic cells over the following decades, it became clear that eukaryotic regulation is substantially more elaborate: multiple enhancers, often acting at a distance, combine with chromatin-level control in a way the simpler bacterial operator model does not require.

Because a relatively modest number of transcription factors can be combined in an enormous number of possible subsets, combinatorial control is also what allows evolution to generate new cell types and expression patterns largely by rewiring which existing regulatory elements a gene responds to, rather than by evolving entirely new regulatory proteins from scratch for every new expression pattern.

Common misconception: that a gene's nearest enhancer is necessarily the one that regulates it. Because enhancers can skip over intervening genes and act on a more distant target instead, the enhancer nearest a gene in linear genomic sequence is not reliably the one that actually controls its expression; determining true enhancer–promoter pairs typically requires additional (e.g. chromatin-looping) experimental evidence.

Worked examples

\text{Three transcription factors, A, B, C, each independently present or absent, combine at a gene's regulatory region.}

If the gene is transcribed only when a specific subset (say, A and B together, but not C) is bound, then \(2^3=8\) possible combinations of the three factors' presence/absence exist, of which only one specific combination activates this particular gene — the same three factors can specify entirely different outcomes at other genes requiring a different subset. A

\text{Generalising: } n \text{ independent regulatory factors combinatorially specify up to } 2^n \text{ distinct regulatory states.}

A relatively modest transcription-factor repertoire can therefore specify a very large number of distinct target-gene responses, since the number of possible combinations grows exponentially with the number of factors, not merely additively (Step 3 of the Proof). A

n\text{ factors} \Rightarrow \text{up to } 2^n \text{ combinatorial regulatory states}

Reading. Combinatorial logic, not a larger regulatory-protein repertoire, is the main reason a limited number of transcription factors is sufficient to specify the very large number of distinct cell types found in a complex multicellular organism.

Scope. An idealisation (independent, binary present/absent factors); real regulatory logic also involves graded (not purely binary) occupancy (Step 3's Hill function) and factor-factor cooperativity, but the combinatorial-scaling argument holds qualitatively regardless.

Problems

A gene's promoter is accessible and its transcription-factor binding sites are all correctly recognised, yet the gene is still not transcribed. Suggest an explanation consistent with the Proof.
Solution
Accessibility and correct binding-site sequence are necessary but not automatically sufficient; the specific combination of activating factors required (Step 3) may not currently be present at sufficient concentration, or the relevant enhancer may not be correctly looped into contact with the promoter (Step 2) — both additional requirements beyond mere site accessibility and sequence recognition.
A disease-associated genetic variant lies 80 kilobases from the nearest gene, with several other genes in between. Explain, using the Hypotheses, how this variant could still plausibly affect that distant gene's expression.
Solution
Enhancers are not required to be adjacent to their target gene; DNA looping (t3 Hypothesis, Step 2) can bring a distal regulatory element into direct contact with a specific promoter even when other genes lie physically between them in linear sequence. The variant may disrupt (or create) a transcription-factor binding site within such a distal enhancer that loops specifically to the distant gene rather than to any of the intervening genes.
Explain why two different cell types with an identical genome and identical transcription-factor repertoire could nonetheless maintain stably different expression patterns for the same set of genes.
Solution
Even with the same available factors, chromatin accessibility (Step 4) can differ stably between the two cell types — one may have a given regulatory region open and the other closed, propagated in each lineage via epigenetic-inheritance's maintenance mechanisms. Since Step 4 establishes accessibility as a prerequisite for any factor binding at all, a closed region remains transcriptionally silent regardless of which factors are otherwise present in the cell.