Does the pattern matter more than the average?
Most experiments on mTOR measure how much of it is active: more after a meal, less after rapamycin. But inside a single cell, mTORC1 activity is not a fixed level. It rises and falls with the cell cycle and with the time of day. The open question is whether that pattern over time — not just the average — is what decides outcomes like autophagy or growth.
This page sorts the evidence in this atlas into three piles: what has been measured, what has only been modelled, and what nobody has done yet.
Same average. Different pattern.
Not tested. The experiment that would settle the question holds the average fixed, changes only the shape, and measures autophagy or growth. No study in this atlas has done it.
Schematic, not data
How far up the evidence goes
Each rung needs the one below it. Click a rung to see the studies.
Measured: mTORC1 activity is patterned in time
Two clocks mTORC1 keeps time with
Click a phase or a study.
Shading shows low versus high only (JOS2024). Phase lengths are not to scale, and the daily peak differs between tissues.
JOS2024 — mTORC1 activity oscillates throughout the cell cycle, promoting mitotic entry and differentially influencing… (2024)
In human and mouse cell lines, mTORC1 activity was lowest in mitosis and G1 and highest in S and G2, set through the TSC complex and independently of Akt and Mek/Erk.
WANG2026C — Fluorescent protein ticker tape (FPTT): Multiplexed recording of transcriptional dynamics in living cells and… (2026)
WANG2026C, a live recording platform reporting mTOR-driven transcription rather than mTORC1 kinase activity, saw the same cell-cycle pattern in synchronised HEK293T cells, with a different method from JOS2024 but explicitly following it up.
RAM2018 — mTOR signaling regulates central and peripheral circadian clock function (2018)
Starts from mTOR activity oscillating over 24 hours in many tissues, then shows that mTOR in turn sets the period and amplitude of the clock in cells, ex vivo tissue and mice.
OKA2013 — Circadian regulation of mTOR by the ubiquitin pathway in renal cell carcinoma (2013)
In mouse kidney tumours, phosphorylated mTOR followed a 24-hour rhythm, driven by the clock through the ubiquitin ligase Fbxw7.
LIP2015 — The Circadian Protein BMAL1 Regulates Translation in Response to S6K1-Mediated Phosphorylation (2015)
S6K1, downstream of mTORC1, phosphorylated the clock protein BMAL1 rhythmically, and protein synthesis rates oscillated over the day in a BMAL1-dependent way.
What made this measurable is a small set of reporters:
- ZHO2015 — TORCAR, the first genetically encoded reporter of mTORC1 kinase activity (FRET).
- BOU2020 — AIMTOR, a bioluminescence reporter with versions for separate compartments.
- GIN2026 — A way to read how a signal changes over time from still images of many cells.
Measured: timing changes the outcome
These are the results closest to the question. None of them isolates pattern from average, and several are one step removed from mTORC1 itself — each card says how.
KUB2012 — Temporal coding of insulin action through multiplexing of the AKT pathway (2012)
A pulse of insulin and a sustained dose became transient and sustained AKT signals, and S6K answered only the transient one. Same molecules, different pattern, different output. One step upstream of mTORC1, in a rat liver cell line.
KUB2018 — In Vivo Decoding Mechanisms of the Temporal Patterns of Blood Insulin by the Insulin-AKT Pathway in the Liver (2018)
The same selective decoding held in rat liver in vivo, with insulin delivered in different time patterns.
JOS2024 — mTORC1 activity oscillates throughout the cell cycle, promoting mitotic entry and differentially influencing… (2024)
Cells in G1, when mTORC1 is lowest, were more sensitive to autophagy induction from the same partial inhibition or nutrient drop. When in the cycle it happened changed the outcome.
GIN2026 — Inferring feedback regulation from static snapshots of a single signal (2026)
Feedback on AKT acted only in a narrow window around G1/S rather than all the time.
OKA2013 — Circadian regulation of mTOR by the ubiquitin pathway in renal cell carcinoma (2013)
Everolimus given when tumour mTOR was at its daily peak improved survival of tumour-bearing mice.
ARR2015 — Alternative rapamycin treatment regimens mitigate the impact of rapamycin on glucose homeostasis and the… (2015)
Weekly or every-fifth-day rapamycin spared glucose tolerance and immune function that daily dosing impaired, while still inhibiting mTORC1. A schedule imposed from outside, and side effects rather than benefit.
LIP2017 — Aberrant Proteostasis of BMAL1 Underlies Circadian Abnormalities in a Paradigmatic mTOR-opathy (2017)
In tuberous sclerosis mice, where mTOR is constantly on, the circadian clock kept time poorly; lowering BMAL1 rescued the behavioural rhythm.
Modelled, not measured
Studies whose readout is a mathematical model. They propose mechanisms and make predictions; they do not show that a cell behaves that way.
- GOR2026 — simulates rapamycin among other drugs in a whole-body ageing model; it is not about mTORC1 dynamics
- GUE2020 — proposes how mTORC1 could set the period and amplitude of the circadian clock
- LU2026 — treats mTORC1 as a rhythm driven by the timing and quality of amino acid intake
4 more pair a model with their own time-course data, so they count as measured: DAL2012, DAL2016, KUB2012, KUB2018.
Where the evidence sits
Studies on this page by what they show and in what system. An empty column is a finding, not a gap in the drawing.
Missing
The decisive experiment
No study in this atlas holds the average mTORC1 activity constant while changing only its pattern — pulses against a steady level with the same time-average — and then measures autophagy or growth. Until that is done, every result above is compatible with the average still being what counts.
The loops, watched live
None of the feedback arms on the map has been followed in living cells. The only arm with time-resolved evidence in unperturbed cells is S6K1 to IRS1, and it comes from fixed-cell reconstruction (GIN2026) and from population time courses fitted as a model (DAL2012), not from live imaging.
Humans
Every measurement of mTORC1's own rhythm in this atlas comes from cell lines or mice. No human study here follows mTORC1 activity over the cell cycle or the day.
From cell to organism
Whether cell-level oscillation connects to ageing or lifespan is untested in this atlas. The nearest link is indirect: KHA2014 found higher mTORC1 activity in mice lacking the clock protein BMAL1, and rapamycin extended their lifespan.
The feedback loops that could make mTORC1 pulse are laid out, arm by arm, on the timing page.
On Oliver’s reading list
This is the question behind the curator’s own reading list. 8 of its 16 studies appear on this page: JOS2024, BOU2020, GIN2026, LU2026, WANG2026C, ARR2015, ZHO2015, KUB2012.
Method
Every study cited here is in the atlas corpus and links to its record, where the abstract and evidence grade can be checked. Which pile a study sits in comes from its recorded readout type (Signal_Readout), which is set only where the abstract or methods show it. The one-line summaries are the curator’s reading of each abstract. The build fails if any cited study leaves the corpus.
Data are CC BY 4.0 — see Data & Citation.