Signal, Stress, and Selection
The Mechanistic Architecture of Quiet Biology
Biological systems do not respond optimally to the accumulation of interventions. They respond to the sequencing of signals. This paper presents a mechanistic framework integrating four distinct signal domains, stress, inspection, stabilisation, and reconstruction, into a temporal model of cellular quality-control. The central principle is signal fidelity: the hypothesis that the accuracy of cellular quality-control decisions depends less on the strength of any individual input than on the clarity with which each signal can be read before the next arrives. Signal fidelity is defined here as the ratio of genuine damage signal to metabolic background noise in the environment in which quality-control decisions are made. The framework proposes that improving this ratio may allow the quality-control processes healthy cells already possess to operate with greater accuracy. The paper also addresses whether organ-level improvements observed in the context of this framework represent targeted organ intervention or systemic metabolic restoration, and concludes with a set of predictions and experimental questions that would allow the framework to be tested or falsified.
01Biology as a Sequenced System
Most therapeutic thinking assumes that biological pathways behave in a linear and additive way. More inhibition produces more effect. More activation produces more benefit. Stack the interventions and the outcomes compound. This assumption is understandable — it reflects the logic of pharmacology, where dose-response relationships are the primary analytical tool.
But cellular systems did not evolve under conditions of simultaneity. They evolved under conditions of fluctuation: feeding alternating with fasting, activity alternating with rest, stress alternating with recovery. The signalling networks that regulate growth, repair, and cell death are not designed to receive all inputs at once. They are designed to interpret sequences, to read a stress signal, respond to it, resolve the response, and return to a resting state from which the next signal can be clearly received.
This is the oscillation principle that has been central to the preceding papers of this series. When mTOR is chronically active, the cell loses its cleanup phase. When AMPK is chronically suppressed, the energy-sensing function that should gate growth degrades. When p53 cannot pulse because it is being held in a permanently suppressed state by elevated MDM2, the quality-control decisions it is supposed to make are deferred or degraded. The problem in each case is not the absence of a signal. It is the loss of the rhythm between signal and silence that gives each signal its meaning.[1]
The framework this paper describes applies that principle at the protocol level. Rather than asking which pathway to activate or suppress, it asks which signals to apply, in which order, and with what separation between them. The sequence is not a scheduling convenience. It is the proposed intervention.
The organising concept throughout is signal fidelity. This term is used here with a specific operational meaning: the ratio of genuine damage signal to metabolic background noise in the environment in which quality-control decisions are made. A cellular environment characterised by chronic insulin excess, persistent mTOR activation, and elevated inflammatory tone is a low-fidelity environment, not because damage signals are absent, but because the background noise against which they must be read is high. The framework proposes that reducing this noise, and applying stress signals in a sequenced manner that preserves their discriminatory power, may allow the cell's quality-control machinery to make more accurate decisions than it would otherwise. Signal fidelity is not yet experimentally measurable as a single parameter; it is a theoretical construct whose components, damage signal amplitude, background inflammatory tone, mTOR oscillation depth, mitochondrial quality, can each be approximated by available biomarkers.
Biology interprets sequences of signals, not stacks of inputs. When signals are layered indiscriminately, they interfere with each other, reduce the clarity of stress detection, and impair the downstream decision-making that depends on reading each signal cleanly. The goal is not more signal. It is cleaner signal.
02The Four Signal Domains
The framework organises its biological inputs into four domains, each acting on a distinct part of the cellular quality-control network, the integrated system of p53, AMPK, mTOR, autophagy, and mitochondrial surveillance machinery through which cells assess damage, allocate repair resources, and decide whether to continue, arrest, or undergo programmed death. Each domain requires temporal separation from the others to function as intended.
Signal One: Stress, revealing system weakness
Stress, at the cellular level, is not inherently pathological. It is a disclosure mechanism. Under controlled conditions, stress may reveal which cellular components are functioning well and which are not, which mitochondria are metabolically robust and which are already compromised, which cells can maintain their energetic integrity under pressure and which cannot.
The two primary stress inputs in this framework are high-intensity exercise and low-dose mitochondrial ribosome inhibition via doxycycline. Exercise induces transient reactive oxygen species production, mitochondrial metabolic demand, and AMPK activation.[2] Doxycycline at sub-antimicrobial doses inhibits mitochondrial protein synthesis, impairing oxidative phosphorylation and creating bioenergetic strain.
The theoretical interest in this mechanism lies in preclinical evidence suggesting that subsets of cancer stem cells may exhibit increased dependence on oxidative phosphorylation compared to most normal differentiated cells.[3] This is not a uniform property: cancer metabolism is characterised by substantial plasticity, and glycolytic subpopulations are well documented. The stress signal does not target any specific cell type. The hypothesis is that it may expose differential vulnerabilities that compromised cells already carry, though this remains to be established in human models.
The framework proposes that the stress signal is most informative when applied before the inspection machinery is fully active. If stabilising signals are present simultaneously, they may dampen the dysfunction signals that the inspection phase depends on reading. Whether stress-first sequencing genuinely improves the discrimination of dysfunctional cellular components, rather than simply combining two independent stressors, is among the key testable predictions of the framework.
A mechanistic boundary requires acknowledgement here. Because doxycycline inhibits mitochondrial translation broadly, prolonged or excessive administration may reduce expression of proteins, including components of the PINK1/Parkin mitophagy pathway, that are required to coordinate the autophagic clearance the inspection phase depends on. This creates a calibration constraint: the temporal gap between Signal One and Signal Two must be wide enough to allow the stress signal to accumulate and disclose differential vulnerability, but not so wide that the mitophagy-competent machinery required for Signal Two is itself impaired. The optimal duration of this window is among the protocol parameters most in need of preclinical calibration.
Signal Two: Inspection, the autophagic decision window
Stress without a subsequent inspection phase is incomplete. The damage disclosed by the stress signal requires an evaluation mechanism that can act on what has been revealed. mTOR inhibition via weekly low-dose rapamycin provides this window.[1]
When mTOR is suppressed, growth signalling pauses and autophagy, the cellular process for identifying and clearing damaged or redundant components, is activated. Autophagy in this context functions as a quality-control process, biasing the system toward the identification and removal of damaged or dysfunctional components. Preclinical evidence suggests that the combination of mitochondrial stress and mTOR-mediated autophagy induction may produce synergistic effects, with stressed cells undergoing autophagy-dependent clearance that neither intervention produces alone.[4] The translation of this finding to human systems has not yet been established.
The oscillation research in this series has established that the inspection window is more productive when periodic rather than continuous. Chronic autophagy activation is not a healthier version of rhythmic autophagy. A cell locked in permanent inspection mode cannot rebuild properly. The benefit, where it exists, appears to emerge from the alternation between inspection and recovery, which is why the weekly rapamycin dosing produces a transient inhibition window followed by full mTOR recovery before the next dose.[5]
Signal Three: Stabilisation, refining the signal environment
Following stress and inspection, the cellular system requires a stable and coherent baseline from which to evaluate what has been found and initiate the appropriate response. This is the proposed role of microbiome-derived butyrate, produced through the fermentation of partially hydrolysed guar gum by gut bacteria.[6]
Butyrate acts as an inhibitor of histone deacetylases, enzymes that suppress gene expression by removing chemical tags from the proteins around which DNA is wound. Through HDAC inhibition, butyrate may influence the expression of genes involved in autophagy, mitophagy, and apoptotic signalling. Evidence from preclinical studies suggests it can activate AMPK through a pathway involving PPAR-δ and PGC-1α, and through AMPK activation may contribute to mTOR suppression and enhanced mitochondrial quality control. It also appears to reduce inflammatory signalling through NF-κB suppression, which in this framework's terms may lower the background noise component of the signal fidelity ratio.[6] These effects are concentration-dependent and context-dependent, and their magnitude in human systems varies.
Butyrate's proposed function in this framework is signal refinement rather than signal intensity. The hypothesis is that it may improve aspects of the signalling environment in which stress and inspection signals are being read, reducing the denominator in the signal fidelity ratio rather than increasing the numerator. The distinction matters because an agent that broadly suppresses inflammatory and stress responses, applied during the stress phase rather than after it, could reduce rather than improve the accuracy of the quality-control process.
Signal Four: Reconstruction, rebuilding the baseline
After the clearance of dysfunctional components, the system enters a rebuilding phase. The question this phase must answer is not only what to build, but whether the post-clearance state is meaningfully better than the pre-stress baseline.
The exercise literature offers the most direct evidence here. Repeated cycles of mitochondrial stress followed by adequate recovery are associated with improvements in mitochondrial number, efficiency, and metabolic flexibility, outcomes that do not occur without the alternation between stress and rest.[2] The proposed mechanism involves mitochondrial biogenesis driven primarily through the PGC-1α pathway, activated by both AMPK and SIRT1. What the clearance phase may contribute, hypothetically, is a reduction in the proportion of dysfunctional mitochondria into which new synthetic capacity is directed. Reconstruction following clearance may therefore tend toward a higher-quality mitochondrial population than reconstruction alone, though this specific sequence has not been formally tested in human systems.
Reconstruction also involves the rebalancing of p53–MDM2 dynamics that govern whether the cell enters a maintenance or growth-oriented state going forward, and the restoration of metabolic efficiency in tissues that have undergone the stress and inspection phases.[7]
Exercise contributes to reconstruction as well as to the initial stress signal. This dual role is sometimes read as a contradiction: how can the same stimulus serve both phases? The answer lies in kinetic separation. Acute, high-intensity exercise constitutes the stress spike: a rapid, transient surge in ROS production, mitochondrial demand, and AMPK activation that raises the signal fidelity numerator. The post-exercise recovery window, characterised by restored insulin sensitivity, nutrient routing toward repair, and PGC-1α-driven mitochondrial biogenesis, constitutes the reconstruction wave. These are not two different effects of the same intervention. They are two temporally distinct biological states produced by the same initiating event. Exercise is not a single signal superimposed on the framework. It is a self-contained, compressed instantiation of the entire framework cycle: stress spike followed by recovery wave, damage disclosed followed by renewal initiated. Its presence in both Signal One and Signal Four is not a design inconsistency. It is evidence that the oscillatory principle the framework is built on is already operating in biology's most well-studied adaptive system.[2]
The four signals form a proposed cycle: stress may reveal differential vulnerability, inspection acts on what has been disclosed, stabilisation refines the environment in which that evaluation occurs, and reconstruction replaces what was cleared. The cycle is the intervention. Disrupt the sequence and the cycle produces noise rather than selection.
03A Note on Organ-Level Observations
The liver research that informed part of this framework raises a question worth addressing directly: when the liver shows improvement in the context of this protocol, reduced fat accumulation, improved insulin sensitivity, reduced inflammatory markers, is that a consequence of liver-targeted intervention, or of something more systemic?
The answer, from the mechanistic architecture described here, is the latter. The protocol does not target the liver. It targets the systemic metabolic environment: chronic insulin excess, mTOR hyperactivation, suppressed AMPK, elevated inflammatory tone, impaired mitochondrial quality. The liver, as the primary site of insulin signalling regulation and the organ most directly affected by the metabolic field conditions the framework addresses, is among the first and most visible responders to systemic improvement.
This is not a trivial distinction. A liver-targeted intervention and a systemic metabolic restoration that benefits the liver are different therapeutic strategies with different mechanistic justifications and different implications for what else may improve alongside them.
The retatrutide Phase 2 data illustrate the principle: liver fat reduction of up to 82% was achieved not through any liver-specific mechanism but through the combination of reduced hepatic lipogenesis, improved insulin sensitivity, and glucagon-driven fat oxidation operating systemically.[8] The liver improved because the metabolic field improved. The organ followed the system.
The same principle applies throughout this framework. Improvements in inflammatory markers may reflect systemic inflammatory tone reduction. Improvements in insulin sensitivity may reflect systemic metabolic restoration. Improvements in mitochondrial quality may reflect the effects of a cellular environment in which quality-control processes can run more effectively throughout the body, not only in the tissue that happens to be measured. The framework is a field protocol. Organ-level improvements are evidence that the field is responding, not evidence that organs are being targeted.
Organ-level improvements are not the target of this framework. They are its downstream signal. When the metabolic field, insulin signalling, inflammatory tone, mTOR oscillation, mitochondrial quality, improves systemically, every organ operating within that field may benefit. The liver is an early and visible indicator, not the endpoint.
04p53 and the Quality of Cellular Decision-Making
p53 functions as one of the principal integrators of cellular stress signals. It receives inputs from DNA damage, metabolic stress, and mitochondrial status, and responds with repair instructions, cell cycle arrest, or death signals depending on what it detects, operating within a network that also includes p16, RB, FOXO, NF-κB, the BCL-2 family, and ATM/ATR, and whose outputs depend heavily on the metabolic context in which it operates. As the Lahav laboratory's research has shown, p53 does not simply activate and remain activated. It pulses, and the frequency and number of those pulses appear to carry information about the nature and severity of the stress the cell is experiencing.[9]
The quality of p53's responses depends in part on the quality of the signals it receives. In a cellular environment characterised by high metabolic noise, chronic insulin signalling, elevated MDM2, persistent mTOR activation, mitochondrial dysfunction generating excess reactive oxygen species, p53's ability to discriminate between genuine damage and background noise may be reduced. Cells that should be cleared may persist. Cells that should repair may do so less efficiently. The quality-control network is operating in a low signal fidelity environment.
The four-signal framework addresses this at each stage. The stress phase may increase the amplitude of genuine damage signals relative to background, raising the numerator. The inspection phase opens the window in which p53 and the autophagy system can act on what they detect. The stabilisation phase aims to reduce inflammatory background noise, lowering the denominator. The reconstruction phase aims to restore the metabolic clarity that supports appropriate p53 pulsing in response to future stresses.
The goal is not to force p53 into action, or to maximise autophagy, or to permanently suppress mTOR. It is to create an environment in which the cellular quality-control network, of which p53 is one principal node, may make more accurate decisions about what to keep and what to discard. That is a more modest-sounding objective than targeted cancer therapy, and a more difficult one to measure. It is also, this framework argues, the more durable one.
p53's function depends in part on the clarity of the signal environment in which it operates. Background metabolic noise may suppress discrimination. Reducing that noise, improving signal fidelity, may allow quality-control decisions to be made more accurately. The framework does not direct p53 to make a specific decision. It aims to create conditions under which better decisions become more likely.
05Why Timing Is Not Optional
The principle of temporal separation is not a refinement of this framework. It is its foundation. Remove the separation and the four-signal architecture collapses into an undifferentiated set of interventions whose interactions cannot be controlled and whose proposed mechanism, sequential phase completion, no longer applies.
When doxycycline-induced mitochondrial stress and rapamycin-induced autophagy run simultaneously rather than sequentially, the inspection system is activated before the stress has had time to disclose differential vulnerability across cell populations. The result may be less specific clearance. When butyrate's stabilising and anti-inflammatory effects are present during the stress phase, they may dampen the dysfunction signals that give the inspection phase its targets. When the reconstruction phase, protein synthesis, mitochondrial biogenesis, mTOR re-activation, begins before the clearance phase has completed, components that might have been cleared may instead be incorporated into the rebuilt system.
Each phase should complete before the next begins. This reflects the temporal structure of the cellular processes involved. Autophagy requires time to identify, sequester, and degrade its targets. Mitochondrial biogenesis requires time to produce functional new mitochondria. The AMPK–mTOR regulatory triangle requires time to transition between its catabolic and anabolic states.[10] Running conflicting signals simultaneously does not produce the same outcomes faster. It produces different and potentially less useful outcomes.
The framework as a state machine
Each phase occupies a distinct biological state, and each state must be fully entered and exited before the next begins. The phases are not additive; they are sequential and mutually exclusive in their dominant biology. If phases are collapsed into simultaneity, phase contrast is lost. The signals do not cancel each other pharmacologically; the individual mechanisms continue to operate. What is lost is the sequencing logic that gives the cycle its proposed selectivity. A protocol that delivers all four signals simultaneously is not a faster version of this framework. It is a different and less specified intervention.
Temporal separation preserves phase contrast. Each phase produces a distinct biological state, stressed, inspecting, stabilising, rebuilding. The contrast between those states is what may make the cycle selective. Collapse the contrast and the cycle loses its proposed logic, not because the signals cancel, but because the sequencing that gives them meaning no longer exists.
06The Systemic Nature of the Framework
It is worth being explicit about what kind of intervention this framework represents, because it is easy to misread it as a collection of targeted pathways or a protocol designed for a specific organ or disease.
It is neither. The Quiet Biology framework is a systemic metabolic intervention. Its targets, mTOR oscillation, AMPK activation, mitochondrial quality, inflammatory tone, signal fidelity, are not specific to any organ, cell type, or disease. They are properties of the metabolic field in which every cell in the body operates. When that field responds well, when chronic excess is reduced, insulin sensitivity improves, mTOR cycles rather than running continuously, and mitochondrial quality is maintained, every organ and every regulatory system within it may benefit.
The upstream metabolic dysfunctions that this framework addresses, chronic insulin excess, mTOR hyperactivation, inflammatory tone, impaired mitochondrial quality, are shared across a range of chronic conditions, including metabolic syndrome, type 2 diabetes, and early-stage prostate cancer management. The evidence supporting the framework's relevance varies substantially between these conditions. The metabolic and hepatic literature is considerably more developed; the prostate cancer application rests on mechanistic plausibility and shared upstream biology rather than disease-specific clinical trial data. The framework does not claim equivalent evidence across conditions. It claims a shared mechanistic target.
The four-signal architecture is the proposed mechanism through which the protocol may improve the field. Stress reveals differential vulnerability. Inspection acts on it. Stabilisation refines the signal environment. Reconstruction replaces what was cleared. Each cycle may leave the metabolic field slightly more capable of accurate self-regulation than before. Over time, the accumulation of those cycles is the biology of health maintenance, not as a fixed state to be achieved, but as an ongoing process to be supported.
07Predictions and Experimental Questions
A framework paper is most useful when it generates hypotheses that can be tested. The following predictions follow directly from the mechanistic architecture described above. Each is falsifiable by available or feasible experimental methods. Taken together, they constitute the empirical agenda that would allow the signal fidelity model to be supported, refined, or rejected.
Hypothesis 1: Stress-first sequencing improves mitochondrial discrimination
If signal fidelity depends on applying stress before inspection, then sequential administration of doxycycline followed by rapamycin should produce greater selective clearance of dysfunctional mitochondria than simultaneous administration at equivalent doses. This is testable in preclinical models using mitochondrial membrane potential assays, mtDNA copy number, and respiratory capacity measurements in cell populations with known proportions of compromised mitochondria. A finding that simultaneous administration produces equivalent or superior clearance would challenge the stress-first sequencing hypothesis.
This hypothesis has a second, independent dimension arising from the doxycycline calibration constraint identified in Section 2. If doxycycline administration beyond an optimal window impairs PINK1/Parkin pathway expression and thereby blunts the autophagic capacity required for Signal Two, then there should be an inverted-U relationship between doxycycline exposure duration and subsequent mitophagy efficiency. The interval that maximises stress signal accumulation while preserving mitophagy-competent machinery is itself a testable parameter, measurable by comparing PINK1 and Parkin protein levels and mitophagy flux across varying doxycycline exposure durations prior to rapamycin administration.
Hypothesis 2: Post-clearance reconstruction yields a superior mitochondrial population
If the reconstruction phase is not simply a return to baseline but a rebuilding into a higher-quality mitochondrial population, then measures of mitochondrial function, respiratory capacity, coupling efficiency, mtDNA integrity, should be higher following stress-inspection-reconstruction cycles than following equivalent periods of recovery without the preceding clearance phase. The exercise literature supports this directionally; it has not been tested for this specific protocol sequence.
Hypothesis 3: Background inflammatory tone modulates quality-control accuracy
If signal fidelity depends in part on the ratio of damage signal to background noise, then reducing inflammatory tone prior to stress application should improve the selectivity of subsequent clearance. This predicts that butyrate pre-treatment followed by doxycycline and rapamycin should produce more selective outcomes than doxycycline and rapamycin alone. Markers of NF-κB activity, IL-6, and TNF-α could serve as proxies for background noise; selectivity of clearance could be assessed by the ratio of dysfunctional to functional mitochondria cleared.
Hypothesis 4: Signal fidelity correlates with measurable metabolic outcomes
If the signal fidelity construct has predictive validity, then individuals or cell populations with lower background metabolic noise, lower fasting insulin, lower CRP, lower inflammatory cytokines, better mTOR oscillation depth, should show greater quality-control accuracy in response to the same stress stimulus than those with higher background noise. In human systems, this predicts that metabolically healthier individuals may show greater mitochondrial quality improvement per cycle of the protocol. This is the most clinically tractable prediction, as its components are measurable with standard laboratory methods.
These hypotheses are not exhaustive. They are the predictions that most directly test the novel claims of the framework: the sequencing hypothesis, the reconstruction hypothesis, the noise-reduction hypothesis, and the signal fidelity construct itself. Falsification of any of them would require revision of the relevant section of the framework. Confirmation of all of them would not prove the framework correct, but would substantially increase its mechanistic credibility and justify progression to human observational studies.
A framework is not a conclusion. It is a structured set of claims about how things might be connected, accompanied by the predictions that would follow if the connections are real. The predictions above are what distinguish this framework from a sophisticated essay.
08Honest Limitations
The mechanistic reasoning in this paper rests on well-replicated science. The oscillation principle for mTOR and autophagy, the mitochondrial dependency of cancer stem cell subsets in preclinical models, the p53 pulsing dynamics, the butyrate–AMPK–mTOR connection, and the exercise–AMPK–SIRT1 axis are each supported by multiple independent research groups across multiple experimental systems.
The application of these mechanisms as a sequenced, temporally separated protocol in a clinical human context is a different matter. Human trial evidence for this specific four-signal combination, at these doses and timings, does not yet exist. The synergistic preclinical data for rapamycin and doxycycline in combination derives from cell-line models.[4] The butyrate data is similarly predominantly preclinical. The individual components have human evidence. The combination in this specific sequence does not.
There are also meaningful uncertainties within the mechanistic argument itself. Whether stress-first sequencing genuinely improves the discrimination of dysfunctional cellular components has not been directly tested. Whether the post-clearance reconstruction phase yields a meaningfully superior mitochondrial population is plausible from the exercise literature but has not been formally demonstrated for this protocol. Whether signal fidelity, defined here as a ratio of damage signal to background noise, can be operationalised as a measurable experimental endpoint remains an open question; at present it is a theoretical construct whose components can be approximated but not directly measured as a composite.
This paper presents a mechanistic framework grounded in available biology. It is not a validated clinical protocol. The distinction is important and is not a disclaimer. It reflects the honest position of a framework that is further along in its reasoning than the evidence base that directly supports it, and that is accountable to that gap. The predictions in Section 7 are the bridge between the two.
The Quiet Biology framework proposes a shift in how biological interventions are conceived, not from the question of which pathway to suppress or activate, but from the question of what conditions might allow the cell's own quality-control machinery to function more accurately.
The four-signal architecture, stress, inspection, stabilisation, reconstruction, is the operational expression of that shift. Each signal has a proposed role, each role has a timing, and the timing is not incidental. It is the mechanism by which four individually incomplete interventions may become a coherent cycle of cellular quality-control and renewal.
Signal fidelity, the ratio of genuine damage signal to metabolic background noise, is the theoretical construct that unifies the framework. The framework's claim is that improving this ratio across repeated cycles may allow cellular quality-control machinery to make progressively more accurate decisions. That claim generates testable predictions. Those predictions are the framework's accountability to the evidence it does not yet have.
Organ-level improvements that may emerge from this framework are not evidence that organs are being targeted. They are evidence that the systemic metabolic field organs depend on may be responding. The field is the target. The organs are a measure of whether the field is improving.
The goal is not to force the system toward a predetermined outcome. It is to create the conditions in which it may make better decisions. Biology, given the right environment, tends to do this well. It has been doing it, after all, for more than a billion years.
The sequence is the intervention.
Improve the signal environment and the biology may improve its own decisions.
- Laplante M, Sabatini DM. mTOR signaling in growth control and disease. Cell. 2012;149(2):274–293. doi:10.1016/j.cell.2012.03.017. For the oscillation argument: Purvis JE, Lahav G. Encoding and decoding cellular information through signaling dynamics. Cell. 2013;152(5):945–956.
- Cantó C, Jiang LQ, Deshmukh AS, et al. Interdependence of AMPK and SIRT1 for metabolic adaptation to fasting and exercise in skeletal muscle. Cell Metabolism. 2010;11(3):213–219. doi:10.1016/j.cmet.2010.02.006
- Lamb R, Ozsvari B, Lisanti CL, et al. Antibiotics that target mitochondria effectively eradicate cancer stem cells, across cell line and patient-derived cancer stem cell models. Oncotarget. 2015;6(7):4569–4584. doi:10.18632/oncotarget.3174
- Dankó T, Petrányi Á, Szábó I, et al. Rapamycin plus doxycycline combination affects growth arrest and selective autophagy-dependent cell death in melanoma cells. International Journal of Molecular Sciences. 2021;22(10):5193. doi:10.3390/ijms22105193
- Kapuy O, Holczer M, Csabai L, Korcsmáros T. Oscillatory autophagy induction is enabled by the regulatory architecture of the mTORC1–AMPK–ULK1 network. PLOS ONE. 2024. doi:10.1371/journal.pone.0313302
- Ding Y, Xia B, Zhang C, Zhuo G. Sodium butyrate induces mitophagy and apoptosis through the mTOR signalling pathway. International Journal of Molecular Sciences. 2023;24(3):2420. Also: Frontiers in Nutrition. Butyrate's inhibition of HDACs activates PPAR-δ and PGC-1α, raising AMP:ATP, activating AMPK, and through AMPK inhibiting mTOR. 2025. doi:10.3389/fnut.2025.1603490
- Haigis MC, Sinclair DA. Mammalian sirtuins: biological insights and disease relevance. Annual Review of Pathology. 2010;5:253–295. doi:10.1146/annurev.pathol.4.110807.092250. For PGC-1α and mitochondrial biogenesis: Cantó C, Auwerx J. Caloric restriction, SIRT1 and longevity. Trends in Endocrinology and Metabolism. 2009;20(7):325–331.
- Harrison SA, Taub R, Neff GW, et al. Triple hormone receptor agonist retatrutide for metabolic dysfunction-associated steatotic liver disease: a randomised phase 2a trial. Nature Medicine. 2024;30:2037–2048. doi:10.1038/s41591-024-03018-2
- Lahav G, Rosenfeld N, Sigal A, et al. Dynamics of the p53–Mdm2 feedback loop in individual cells. Nature Genetics. 2004;36(2):147–150. Also: Purvis JE, Karhohs KW, Mock C, et al. p53 dynamics control cell fate. Science. 2012;336(6087):1440–1444.
- Kim J, Guan K-L. AMPK connects energy sensing to PINK1/Parkin-dependent mitophagy and mitochondrial biogenesis. EMBO Reports. 2019. Also: Kazyken D, Dame SG, Wang C, et al. Unexpected roles for AMPK in the suppression of autophagy and the reactivation of mTORC1 during prolonged amino acid deprivation. Autophagy. 2024;20(9):2017–2040.
- IndolentRead phase →
- InflectionRead phase →
- ActiveRead phase →
- RefractoryRead phase →