Beyond one switch: reading the whole world
You have just spent this rung learning the cleanest gene switches in biology. The operon strings related genes onto one promoter; the [[molbio-lac-operon|lac operon]] turns on only when lactose is present and glucose is gone; the trp operon shuts off when tryptophan piles up. Each is, at heart, an answer to one narrow question — is this single small molecule around? That is gene regulation at its most local. But a bacterium does not only need to track one sugar. It needs to know whether the temperature just jumped, whether it is starving, whether it is alone or surrounded by a billion others. This final guide is about those wider answers.
Notice the scale jump. A repressor on the lac operator touches a few genes. But coping with a sudden heat shock means switching on dozens of protective genes at once; turning a cell into a tough dormant spore means orchestrating hundreds. A bacterium needs ways to flip whole *programs* of genes, not just one operon at a time — and it needs to do it fast, because in the microbial world conditions change in seconds. We will meet three devices that meet this need at three different layers: a swappable targeting part of the polymerase itself, a phosphate relay that runs from outside the cell to its DNA, and the surprise of the lot — a piece of mRNA that needs no protein to sense and switch at all. We close with quorum sensing, where the unit of decision stops being the cell and becomes the crowd.
Sigma factors: changing the route card
Recall from the transcription rung that bacterial [[molbio-rna-polymerase|RNA polymerase]] is, on its own, oddly incomplete. Its core can do the chemistry of copying DNA into RNA, but it cannot find the right place to start — it cannot recognize a promoter by itself. For that it needs a detachable subunit called a [[sigma-factor|sigma factor]]. Picture a delivery driver who knows how to drive but has no idea which addresses to visit; the sigma factor is the route card that tells the polymerase, "these are the promoters you start at." Core plus sigma makes the complete enzyme, the holoenzyme; and tellingly, once transcription is underway the sigma often drops off and is reused. Its whole job is choosing *where to begin*.
Here is the regulatory twist. A cell carries not one sigma factor but several, and each one steers the polymerase to a *different* set of promoters. There is an everyday housekeeping sigma that aims the polymerase at the genes a cell always needs. But when the world changes, the cell can deploy an alternative sigma that redirects the polymerase to a whole battery of genes at once. Swap the route card and the same fleet of drivers suddenly visits a different set of addresses. So by changing which sigma is active, a bacterium switches an entire program of genes on as one block — exactly the broad lever the single-operon switches could not provide.
Two examples show the power. When E. coli is suddenly overheated, proteins start to unfold and clump; the cell responds by activating a heat-shock sigma factor that switches on dozens of heat-shock proteins together — the chaperones you met in the protein-life rung that refold the damaged proteins and clear the wreckage. When the bacterium Bacillus runs out of food, a timed cascade of sporulation sigma factors turns on hundreds of genes in sequence, rebuilding the cell into a tough dormant spore. This is regulation at the level of whole gene sets, achieved by the elegant trick of swapping the part of the polymerase that decides where to read.
Two-component systems: a wire from outside to the genes
Sigma swapping reprograms genes, but something still has to *sense* the change in the world and decide which sigma — or which regulator — to deploy. Often that sensing is the job of a [[two-component-system|two-component system]], the workhorse signalling device of bacteria. Picture a doorbell wired to a notepad inside a house: someone presses the button outside, a wire carries the signal in, and a hand inside writes a note that changes what the household does. A two-component system is exactly that — a sensor at the cell surface notices something outside and passes the news, through one chemical step, to a partner inside that changes which genes are read.
- A sensor kinase sits in the cell membrane, one part facing the outside world. It is tuned to one particular cue — a change in saltiness (osmolarity), a nutrient, a metal ion, the presence of an antibiotic.
- When that cue appears, the sensor kinase grabs a phosphate group from ATP and sticks it onto itself — a self-tagging step called autophosphorylation. The signal is now a chemical flag held on the protein.
- The sensor kinase passes that phosphate to the second component, a response regulator inside the cell. Receiving the phosphate changes the response regulator's shape — the message has crossed the membrane carried by a single phosphate, not by the outside molecule itself entering.
- Most response regulators are themselves transcription factors. Once phosphorylated and reshaped, the response regulator binds specific DNA sites and switches its target genes on or off — so within minutes the cell's gene expression has answered the signal from outside.
The pathway is short and direct — sense outside, transfer one phosphate, change gene expression inside — and that brevity is why it is fast. A single bacterial species may carry dozens of these systems, each tuned to a different cue, which together let the cell respond to osmotic stress, scout for nutrients, resist antibiotics, and decide whether to settle into a biofilm. Worth being precise: the message is the *transferred phosphate*, not the outside molecule sneaking in. And we are looking narrowly at how the relay changes which genes get transcribed — the general physics of how cells signal across membranes is a much broader topic for a later rung.
Riboswitches: an mRNA that senses and decides on its own
So far every device used proteins to do the sensing and switching. Now the surprise. A [[riboswitch|riboswitch]] is a stretch of mRNA that senses a small molecule and turns its own gene on or off — with no regulatory protein involved at all. Imagine a length of ribbon that ties itself into one of two knots depending on whether a particular bead is threaded onto it: with the bead it knots one way, without it knots another, and which knot it makes decides whether a gate downstream is open or shut. A riboswitch is RNA behaving exactly like that ribbon — its fold, not a protein, is the switch.
Remarkably, the riboswitch is built right into the front of the mRNA it controls, usually in the untranslated region ahead of the coding part. It has two cooperating pieces. The first is an aptamer — a precisely shaped RNA pocket that grips one specific small molecule, often a vitamin, an amino acid, or a metabolite. The second is an expression platform — a neighbouring stretch that folds one way or the other depending on whether the aptamer has caught its target. When the metabolite is abundant and binds, the expression platform refolds into a shape that shuts the gene down: typically it forms a transcription terminator hairpin that makes the polymerase stop early, or it hides the ribosome binding site so translation cannot start. When the metabolite is scarce, the RNA folds the other way and the gene stays on. The same molecule is sensor and switch in one.
A riboswitch in the 5' untranslated region of its own mRNA:
metabolite SCARCE metabolite ABUNDANT
---------------- -------------------
aptamer empty metabolite --> [aptamer pocket]
| |
platform folds ON platform refolds OFF
| |
polymerase reads through terminator hairpin -> polymerase STOPS
(or ribosome site exposed) (or ribosome site hidden)
| |
GENE ON --> enzyme made GENE OFF --> enzyme not made
Logic: a gene that MAKES the metabolite shuts itself off
once enough metabolite is around — feedback, no protein needed.Why does this matter beyond its sheer cleverness? Riboswitches show that RNA, not only protein, can directly read the chemical state of a cell and regulate genes — a striking echo of the idea that early life may have run on RNA before proteins took over the regulatory jobs. They typically govern the very genes that make or import the metabolite they sense, giving a tidy negative-feedback loop: enough product around, and the gene that makes it quietly shuts itself off. Two honest caveats. Most riboswitches turn genes *off* when their metabolite is abundant, but some work the opposite way, so each must be checked case by case. And because certain riboswitches are unique to microbes and absent in us, they are being explored as targets for new antibiotics — a promising but still developing idea, not a finished drug.
Quorum sensing: when the crowd decides
The last device changes the unit of decision. Everything so far was one cell reading its own surroundings. [[quorum-sensing|Quorum sensing]] is a bacterium asking a question no single cell can answer alone: how many of us are there? Imagine everyone in a crowd quietly humming the same note. With only a few people the hum is too faint to hear; but once enough are humming together, the sound swells loud enough that everyone notices and acts. Bacteria do this with chemistry instead of sound — each cell releases a little signal molecule, and only when there are enough cells does the signal build up high enough to trigger a coordinated change in gene expression. It is counting by chemistry.
The signal molecule each cell makes and secretes is called an autoinducer. As a population grows in a confined space, the autoinducer accumulates, so its concentration stands in for how crowded the neighbourhood is. Once it crosses a threshold, it binds a receptor or regulator inside the cells and flips a whole set of genes — often ramping up its own production too, a positive feedback that makes the switch snap sharply once the quorum is reached. The textbook example is a marine bacterium that glows only when packed densely inside a squid's light organ: one glowing cell would be a pointless waste, but billions together make a useful light. Quorum sensing also coordinates biofilm formation and the timing of when pathogens release their toxins — which is why disrupting it is being studied as a way to disarm infections.
One honest qualifier closes the picture. Quorum sensing tracks population density only indirectly — through how high a secreted signal builds up — so it is not bacteria literally counting one another. In a tight, poorly mixed pocket the signal can pile up because it cannot diffuse away, not because the cells are truly numerous, so what the system really senses is a blend of density and confinement. Keep that nuance, and you have the honest version: bacteria do not count heads, they read a chemical that *usually* tracks how many of them there are. With that, you have walked the whole of bacterial gene control — from one repressor on one operator to a billion cells deciding together — and you are ready for the far more layered story of regulation in eukaryotes in the next rung.