mTOR Core · Lesson 01 · Foundation · 15 min

What is mTOR?

The kinase that decides whether a cell builds or recycles

The question

How does a cell decide when to grow?

What you should be able to do

After this lesson you should be able to

Explain what a protein kinase does, and why phosphorylation is enough to change what a cell is doing.
Name the four kinds of input that converge on mTOR and state which cellular programs shift when its activity shifts.
Predict what happens to mTOR signalling when one input is present and another is missing.
Distinguish a teaching summary such as ‘mTOR controls growth’ from what the underlying experiments actually measured.

Research skill Building a biological model

The core idea

A cell is under constant pressure to answer one question: is now a good time to build? Building is expensive. Making protein consumes a large share of a cell's energy budget, and a cell that starts building without the raw materials, the energy, or the external permission to do so ends up worse off than one that waited.

Cells do not answer that question with a single switch. They answer it by pooling several streams of information at once — amino acids and other nutrients, the cellular energy state, growth-factor and hormone signals, and various forms of stress. mTOR is one of the main places where those streams meet. It is a protein kinase: an enzyme that attaches phosphate groups to other proteins, changing what they do.

When the signals arriving at mTOR say conditions are good, its downstream effects push the cell toward building — protein synthesis, lipid synthesis, nucleotide synthesis, growth. When they say conditions are poor, that pressure is lifted and recycling programs such as autophagy are released. That is the whole idea in one sentence: mTOR sits between what a cell can detect about its situation and what a cell does about it.

Building is expensive. Making new protein costs a cell a large share of its energy, so a cell that starts building at the wrong moment ends up worse off than one that waited.

mTOR is one of the places where a cell decides whether now is a good moment. It is an enzyme that adds phosphate groups to other proteins, and that small chemical change alters what those proteins do.

Signals about food, energy, hormones and stress all arrive at mTOR. When they say conditions are good, the cell builds. When they say conditions are poor, the cell switches to recycling its own material instead.

The mechanism, in outline

Read the diagram left to right. On the left are the inputs: nutrients (amino acids especially), the energy state of the cell, growth factors such as insulin and IGF-1, and stress signals. In the middle is mTOR, working inside protein complexes rather than alone. On the right are the outputs: the cellular programs whose activity shifts when mTOR's activity shifts.

Two things about this picture are worth noticing straight away. First, the inputs are not interchangeable. Amino acids and growth factors reach mTOR by different molecular routes, and a cell can have plenty of one and none of the other — which is exactly the situation the pathway is built to distinguish. Second, mTOR does not simply switch outputs on and off; it shifts the balance between building and recycling. Autophagy is not merely 'mTOR off'. It is a program that mTOR activity restrains, and that becomes available when that restraint is lifted.

Read the diagram from left to right. On the left are the things a cell can detect: nutrients, how much energy it has, hormone signals from the rest of the body, and stress. In the middle is mTOR. On the right are the jobs that change when mTOR changes.

Two details matter. The inputs are not interchangeable — a cell can have plenty of food and no hormone signal, and it is built to tell those situations apart. And mTOR does not switch things on and off; it shifts the balance between building and recycling.

Interactive model

Three inputs, one kinase, three outputs

Amino acids

Growth factors

Energy

The same picture as the diagram above, with the inputs under your control. Each input reaches mTORC1 by its own molecular route: nutrients through the Rag GTPases, growth factors through Rheb, energy stress through AMPK. Node names and links come from the Atlas pathway model.

S6K1 phosphorylated · 4E-BP1 phosphorylated · ULK1 held off

Everything the cell needs is present, and everything the cell needs permission for has been granted. mTORC1 is active, the building programs run and autophagy stays restrained.

Rag GTPases: RagA/B–RagC/D heterodimers; nucleotide state determines mTORC1 recruitment.The taxi that brings mTORC1 to the lysosome.Note the inversion: RagA/B is active when GTP-loaded, RagC/D when GDP-loaded. The Rags control mTORC1 *location*, not its catalytic activity.
Rheb: Small GTPase; GTP-loaded Rheb allosterically activates mTORC1.The one thing that actually switches mTORC1 on.Realigns the mTOR active site; the convergence point of the entire growth-factor arm and the reason localisation alone is not activation. Note a declared simplification in this map: Rheb is farnesylated and distributes across the endomembrane system, with a substantial ER and Golgi pool, and which pool supplies the activating Rheb is still debated. It is drawn on the lysosomal band because that is where it meets mTORC1.
AMPK: Energy-stress kinase that both activates TSC2 and directly inhibits Raptor.The low-fuel sensor.αβγ heterotrimer. Two independent arms onto mTORC1 plus a direct activating arm onto ULK1 — the reason energy stress switches growth off and recycling on in one move.
mTORC1: mTOR–Raptor–mLST8 (+PRAS40, DEPTOR); switched on only at the lysosome.The growth decision itself.Coincidence detector: nutrients supply location via the Rags, growth factors supply activation via Rheb. Neither alone is sufficient — the single most important idea in the pathway.
S6K1: Ribosomal protein S6 kinase; the standard readout of mTORC1 activity.mTORC1's best-known output kinase.T389 phosphorylation by mTORC1 is rapamycin-sensitive, which is why S6K1 became the field's default assay — and why the field long over-read rapamycin as a complete mTORC1 inhibitor.
4E-BP1: Translational repressor released from eIF4E upon multi-site phosphorylation.A cap on protein-making that mTORC1 removes.Only partially rapamycin-sensitive. This single fact explains the rapalog/Torin discrepancy and drove the whole ATP-competitive inhibitor programme.
ULK1: Autophagy-initiating kinase, inhibited by mTORC1 and activated by AMPK.The switch that starts self-digestion.mTORC1 phosphorylates S757 to block the AMPK–ULK1 interaction; ULK1 also feeds back to phosphorylate and dampen AMPK, making this a closed loop rather than a switch.

This is a simplified teaching model with a fixed set of states, not a simulation. It shows the direction each control pushes the pathway, not how much, how fast, or what any particular cell would do.

Every state of this model

Amino acids	Growth factors	Energy	Readout	What it means
HIGH	ON	normal	S6K1 phosphorylated · 4E-BP1 phosphorylated · ULK1 held off	Everything the cell needs is present, and everything the cell needs permission for has been granted. mTORC1 is active, the building programs run and autophagy stays restrained.
HIGH	ON	low	S6K1 low · 4E-BP1 unphosphorylated · ULK1 released	Raw material and permission are both there, and the cell still does not build, because AMPK reports that the energy to pay for it is not. An input that can veto on its own is not the same kind of input as one that merely contributes.
HIGH	OFF	normal	S6K1 low · 4E-BP1 unphosphorylated · ULK1 released	The situation worth remembering from this lesson. Nutrients are abundant, so mTORC1 is delivered to the right place — and the activator waiting for it there has not been switched on, because no growth factor said so. Lesson 03 is about that step.
HIGH	OFF	low	S6K1 low · 4E-BP1 unphosphorylated · ULK1 released	Two independent reasons not to build. Notice that the readout looks identical to the state above it: a single downstream measurement cannot tell you which input was missing.
LOW	ON	normal	S6K1 low · 4E-BP1 unphosphorylated · ULK1 released	Permission without material. The hormone signal arrives and the amino acids that would put mTORC1 in position do not, so the kinase never reaches the membrane where its activator sits.
LOW	ON	low	S6K1 low · 4E-BP1 unphosphorylated · ULK1 released	Nothing to build with and nothing to build it on. In a real cell this is where recycling stops being an option and becomes the main source of amino acids.
LOW	OFF	normal	S6K1 low · 4E-BP1 unphosphorylated · ULK1 released	The resting or starved state. Autophagy is not switched on here so much as let go of — the restraint mTORC1 was applying to ULK1 is simply no longer there.
LOW	OFF	low	S6K1 low · 4E-BP1 unphosphorylated · ULK1 released	Every input says no. Nothing in the model distinguishes this from the state above it, which is a fair reflection of the readouts: the pathway does not report why it is off.

Two complexes, as a first mental model

mTOR is a single protein, but it does its work inside at least two different assemblies of proteins, called mTORC1 and mTORC2. They share the mTOR kinase and differ in their partner proteins, in what activates them, and in what they act on.

For now, hold a deliberately rough version: mTORC1 is the complex most closely tied to nutrient sensing, protein synthesis and the restraint of autophagy; mTORC2 is more closely tied to signalling through AKT and to the cytoskeleton. This is a teaching model, not a full mechanistic account. The two complexes share components, they influence each other, and quite a lot of what a real cell does cannot be sorted cleanly into one box or the other. Lesson 02 takes the model apart and shows where it holds and where it leaks.

mTOR is one protein, but it works inside two different groups of proteins, called mTORC1 and mTORC2. Think of the same worker in two different teams: same person, different job.

Roughly: mTORC1 is the one tied to food and to making protein, mTORC2 is tied to other signalling. This is a rough picture on purpose. Lesson 02 shows where it stops being true.

What this lesson is not claiming

'mTOR controls growth' is a useful summary and a poor literal statement. mTOR is one node in a network with many inputs and many parallel outputs, and almost everything in the outline above was worked out in cells and in laboratory animals rather than in people. When you meet a claim about mTOR and human health later in the Atlas, the evidence tier attached to it is doing real work — keep looking at it.

What does the evidence say?

These are Atlas studies, with the Atlas's own evidence tier. Each card links to the full record — nothing here restates it.

DRAFT1: a mammalian protein that binds to FKBP12 in a rapamycin-dependent fashion and is homologous to yeast TORs1994

Discovery of the protein RAFT1 (today's mTOR) as the direct target of the FKBP12-rapamycin complex; founding paper of the entire mTOR field.

Study page →

DRictor, a novel binding partner of mTOR, defines a rapamycin-insensitive and raptor-independent pathway that regulates the cytoskeleton2004

Discovery of Rictor and the SECOND mTOR complex, mTORC2.

Study page →

DAblation in mice of the mTORC components raptor, rictor, or mLST8 reveals that mTORC2 is required for signaling to Akt-FOXO and PKCalpha, but not S6K12006

The foundational genetic 'dissection' of the two complexes in living mice.

Study page →

Almost every arrow in this lesson's diagram was drawn from work in cultured cells and in mice. The mapping from those arrows to human physiology is where the evidence is thinnest, and it is where the Atlas's evidence tiers matter most.

Work through it

Each of these asks you to commit to something — a prediction, a reading of two studies, a design — before it answers. Everything here is in the page, so nothing is lost if you would rather just read it.

Predict → observe → explain

Nutrients without permission

Predict — commit before you look

A cell is sitting in medium rich in amino acids, but insulin and other growth-factor signalling have been withdrawn. Predict what happens to mTORC1 output before you look.

Observe — what was actually measured

DPRAS40 is an insulin-regulated inhibitor of the mTORC1 protein kinase2007

Biochemistry in human cells: the mTORC1-associated protein PRAS40 was identified and its behaviour followed with and without insulin signalling.

PRAS40 acts as an inhibitor sitting inside mTORC1. Without insulin it clamps the complex shut; insulin-driven AKT phosphorylation of PRAS40 releases that restraint so the complex can be switched on.

Study page →

Explain

This is one of the mechanisms behind the behaviour you predicted. Growth-factor signalling does not add a push to a complex that nutrients have already started — it removes a restraint that was holding the complex closed. So a cell with plenty of amino acids and no hormone signal has the material and not the permission, and its output stays low. Lesson 07 follows the same branch further and finds a second brake outside the complex.

What this evidence supports

That an inhibitory subunit inside mTORC1 links insulin signalling directly to the complex's activity in these cells
That the growth-factor input works by releasing restraint rather than by adding activation

What it does not establish

How much of the response of an intact tissue to a meal runs through this particular subunit
That the two inputs behave as a strict logical AND in every cell type — the branches are not perfect gates, and how low is low matters

Show the expected answer

B. Output stays low — something inside the complex keeps it restrained until a growth-factor signal releases it

Observe — what was actually measured

DPRAS40 is an insulin-regulated inhibitor of the mTORC1 protein kinase2007

Biochemistry in human cells: the mTORC1-associated protein PRAS40 was identified and its behaviour followed with and without insulin signalling.

PRAS40 acts as an inhibitor sitting inside mTORC1. Without insulin it clamps the complex shut; insulin-driven AKT phosphorylation of PRAS40 releases that restraint so the complex can be switched on.

Study page →

Explain

What this evidence supports

That an inhibitory subunit inside mTORC1 links insulin signalling directly to the complex's activity in these cells
That the growth-factor input works by releasing restraint rather than by adding activation

What it does not establish

How much of the response of an intact tissue to a meal runs through this particular subunit
That the two inputs behave as a strict logical AND in every cell type — the branches are not perfect gates, and how low is low matters

Compare the evidence

Two ways of finding out what a protein does

	D RAFT1: a mammalian protein that binds to FKBP12 in a rapamycin-dependent fashion and is homologous to yeast TORs	D Ablation in mice of the mTORC components raptor, rictor, or mLST8 reveals that mTORC2 is required for signaling to Akt-FOXO and PKCalpha, but not S6K1
Model system	Mammalian cell culture	Knockout mice (raptor/rictor/mLST8)
Perturbation	A drug used as a probe: the FKBP12–rapamycin complex was used as bait to pull out whatever it binds in mammalian cells.	Genetics in a living animal: the genes for raptor, rictor and mLST8 were deleted in mice, one at a time.
Readout	Identification of the bound protein — RAFT1, the protein now called mTOR.	Which animals survive, and which downstream signalling is lost — deleting raptor was lethal early, while deleting rictor or mLST8 removed signalling to AKT and left S6K1 signalling in place.

What do both studies support?

That mTOR is a real, identifiable protein whose activity is required for normal growth signalling, and that its functions are separable rather than one indivisible job.

Where do they differ?

In what they can be asked. The drug-based experiment identifies a molecule and defines it by what a compound binds — so everything it tells you is filtered through that compound. The mouse genetics removes a component entirely, in a whole animal, and reads out survival and signalling; it answers what is required, and cannot say what is sufficient or what happens acutely. Notice also the asymmetry in what each one cannot see: a drug reveals nothing about functions it does not block, and a knockout reveals nothing about a function that compensation has already covered.

What experiment would help next?

An acute, reversible removal of the same components in an adult animal — degradation or conditional deletion rather than a germline knockout — so that the difference between 'required for development' and 'required now' stops being confounded.

Scientific caution

‘mTOR controls growth’

What this evidence supports

That mTOR activity shifts the balance between building programs and recycling programs in the systems where this has been measured
That several unrelated kinds of information reach the same kinase, which is what makes it worth calling an integrator

What it does not establish

That mTOR is the only route to growth, or that removing it stops growth in every cell
That the arrows in this lesson have the same weight in a human being as in a cultured cell
That any single measurement of mTOR activity captures what the whole pathway is doing

Why?

Almost everything in this lesson was worked out in cultured cells and in laboratory animals, and the summary sentence compresses a network with many parallel inputs and outputs into one verb. The compression is useful for learning and it is not a finding; when you meet a claim about mTOR and human health later in the Atlas, the evidence tier attached to it is doing real work.

Think

If nutrients are abundant but growth-factor signalling is low, what would you expect mTOR signalling to do? Explain your reasoning before revealing.

Ask yourself what each input is evidence of, from the cell's point of view.

Think first, then reveal

The usual expectation is that the two inputs are not additive but closer to a requirement pair: nutrient availability tells the cell it could build, growth-factor signalling tells it that the organism is asking it to. In the standard model, mTORC1 output stays low when either one is missing, because the growth-factor and amino-acid branches converge on different steps of the same activation mechanism — which is the subject of Lesson 03. If you predicted 'partly on', that is also a reasonable reading: the branches are not perfect AND-gates, and in real cells the result depends on cell type and on how low 'low' is.

Check yourself

Three questions, easiest first. Nothing is recorded and there is no pass mark — the explanation after each answer is the part worth reading.

Question 1 · Warm-upWhat kind of molecule is mTOR?

Show the answer

B — A protein kinase — an enzyme that attaches phosphate groups to other proteins. mTOR is an enzyme. It changes what other proteins do by phosphorylating them, and everything downstream in this course follows from that one activity.

Question 2 · Step upA cell has abundant amino acids but almost no growth-factor signalling. In the standard model, what happens to mTORC1 output?

Show the answer

B — It stays low — the two branches reach mTORC1 by different routes and neither alone is sufficient. Nutrients tell the cell it could build; growth factors tell it the organism is asking it to. The two branches converge on different steps, which is why the pathway behaves more like a requirement pair than a sum.

Question 3 · HarderA review states flatly that ‘mTOR controls growth’. What is the strongest reason to read that as a scoped summary rather than a literal fact?

Show the answer

B — mTOR is one node in a network with many parallel inputs and outputs, and most of the underlying work is in cells and laboratory animals. The sentence is a useful shorthand and a poor literal statement. Both halves matter: mTOR is not the only route to growth, and the arrows behind the summary were drawn mostly in cells and mice rather than people.

Go deeper

mTOR Akt/PKB S6K1 4E-BP1 mTORC1 mTORC2 Protein synthesis Autophagy

Follow a guided route through the mechanism:

What does a cell actually do when mTORC1 fires?Follow what actually changes inside a cell once mTORC1 fires.
How does a cell know it has enough raw material to grow?See how the nutrient input reaches mTORC1 in the first place.

Concepts introduced in this lesson

Kinase and phosphorylation · Signal integration · Anabolic vs catabolic programs · Protein complex

← CurriculumNext: mTORC1 vs mTORC2 →