17.04.06 · mol-cell-bio / energy-metabolism

Metabolic regulation: allosteric control, covalent modification, and the AMP kinase switch

stub3 tiersLean: nonepending prereqs

Anchor (Master): Hardie, D. G. — AMPK: A Key Regulator of Energy Balance (2015); Hardie, D. G. et al. — Nat. Rev. Mol. Cell Biol. 13 (2012) 251-262

Intuition Beginner

A living cell runs thousands of chemical reactions at once. If they all ran at full speed, the cell would waste energy and produce toxic byproducts. The cell needs control systems — like a thermostat in a house — to turn reactions up when energy is needed and dial them back when supplies are sufficient.

One control method is allosteric regulation. Small molecules bind to an enzyme at a site distinct from its active site. This binding changes the enzyme's shape, either increasing or decreasing its activity. For example, ATP can bind to phosphofructokinase and slow it down, telling the enzyme that the cell already has enough energy.

A second method is covalent modification. A kinase enzyme attaches a phosphate group to a target enzyme, flipping it between an active and an inactive state. A phosphatase removes the phosphate, reversing the switch. This on/off mechanism allows hormones like insulin and glucagon to rewire metabolism across the entire cell within seconds.

A third method involves a master sensor called AMP-activated protein kinase (AMPK). When cellular ATP drops and AMP rises, AMPK switches on. It activates pathways that produce ATP (fatty acid oxidation, glucose uptake) and shuts down pathways that consume ATP (fatty acid synthesis, protein synthesis). AMPK is the cell's low-fuel alarm.

Visual Beginner

The diagram above shows the three layers of metabolic regulation. Allosteric effectors (ATP, AMP, citrate, fructose-2,6-bisphosphate) modulate enzyme activity directly. Hormone-triggered phosphorylation (PKA, PFK-2/FBPase-2) flips enzymes between active and inactive states. AMPK sits at the top, sensing the AMP/ATP ratio and coordinating the cellular response to energy stress.

Worked example Beginner

How does fructose-2,6-bisphosphate regulate glycolysis and gluconeogenesis simultaneously?

After a meal, insulin signals the liver to burn glucose. Insulin activates a bifunctional enzyme called PFK-2/FBPase-2. This enzyme has two opposing activities on the same polypeptide. In its dephosphorylated state (insulin signal), the PFK-2 domain is active and synthesizes fructose-2,6-bisphosphate (F2,6BP) from fructose-6-phosphate.

F2,6BP is a potent allosteric activator of phosphofructokinase-1 (PFK-1), the committed step of glycolysis. At the same time, F2,6BP inhibits fructose-1,6-bisphosphatase, a key gluconeogenic enzyme. The net result: glycolysis is accelerated and gluconeogenesis is suppressed.

During fasting, glucagon triggers phosphorylation of PFK-2/FBPase-2 by protein kinase A. This switches off the PFK-2 domain and activates the FBPase-2 domain, which degrades F2,6BP. Without F2,6BP, PFK-1 slows down and gluconeogenesis is released from inhibition. A single molecule — F2,6BP — reciprocally regulates both pathways in opposite directions.

Check your understanding Beginner

Exercise (medium, short answer).

Explain why reciprocal regulation of glycolysis and gluconeogenesis is necessary. What would happen if both pathways ran simultaneously at full speed?

Hint

Consider the net reaction of running glycolysis (glucose to pyruvate) and gluconeogenesis (pyruvate to glucose) at the same time. What happens to ATP?

Answer

If glycolysis and gluconeogenesis ran simultaneously, they would form a futile cycle (substrate cycle). Glycolysis converts glucose to pyruvate, consuming 2 ATP and producing 4 ATP (net +2). Gluconeogenesis converts pyruvate back to glucose, consuming 6 ATP. The combined result: glucose and pyruvate cycle endlessly while 4 ATP per turn are hydrolyzed to ADP and Pi, releasing the energy as waste heat. This would drain cellular ATP without producing anything useful. Reciprocal regulation ensures that when one pathway is active, the other is suppressed, preventing this thermogenic waste under normal conditions.

Formal definition Intermediate+

Allosteric regulation: the MWC model

The Monod-Wyman-Changeux (MWC) model describes allosteric enzymes as oligomeric proteins that exist in an equilibrium between two conformational states: a tense state (T, low affinity for substrate) and a relaxed state (R, high affinity for substrate). All subunits undergo the concerted transition simultaneously. The fractional saturation $Y$ as a function of substrate concentration $[S]$ is

Y = \frac{L \cdot c \cdot ( 1 + c [ S ] / K _{R} ) ^{n - 1} + ( 1 + [ S ] / K _{R} ) ^{n - 1}}{L \cdot ( 1 + c [ S ] / K _{R} ) ^{n} + ( 1 + [ S ] / K _{R} ) ^{n}}

where $L = [T_{0}] / [R_{0}]$ is the equilibrium constant between the two states in the absence of ligand, $c = K_{R} / K_{T}$ is the ratio of dissociation constants, and $n$ is the number of protomers. An allosteric activator shifts the equilibrium toward the R state (effectively reducing $L$ ); an inhibitor shifts it toward the T state (increasing $L$ ).

Homotropic effectors are identical to the substrate (e.g., substrate molecules binding at both catalytic and regulatory sites, producing cooperativity). Heterotropic effectors are different molecules (e.g., ATP inhibiting PFK-1, AMP activating PFK-1, citrate inhibiting PFK-1). The MWC model handles heterotropic effectors by allowing them to bind preferentially to one state, modifying $L$ .

An alternative framework, the Koshland-Nemethy-Filmer (KNF) sequential model, allows individual subunits to change conformation independently upon ligand binding. Most real allosteric enzymes show behavior intermediate between the concerted and sequential extremes.

Feedback inhibition

In feedback inhibition, the product of a metabolic pathway inhibits an enzyme early in that pathway. The archetypal example is ATP inhibition of phosphofructokinase-1 (PFK-1). PFK-1 catalyzes the committed step of glycolysis:

Fructose-6-phosphate + ATP \to Fructose-1,6-bisphosphate + ADP

ATP is both a substrate (at the catalytic site) and an allosteric inhibitor (at a separate regulatory site). At high ATP concentrations, the regulatory site is occupied, shifting PFK-1 into the T state and reducing its affinity for fructose-6-phosphate. AMP relieves this inhibition by competing with ATP at the regulatory site and stabilizing the R state.

Citrate, an intermediate of the citric acid cycle, reinforces ATP inhibition. When both ATP and citrate are high, the cell has abundant energy and biosynthetic precursors, and glycolysis is strongly suppressed.

Covalent modification: phosphorylation and dephosphorylation

Kinases transfer a phosphate group from ATP to specific serine, threonine, or tyrosine residues on target proteins. Phosphatases remove these phosphates. The phosphorylation state determines whether the enzyme is active or inactive.

The key regulatory kinase for carbohydrate metabolism is protein kinase A (PKA), which is activated by cyclic AMP (cAMP). When glucagon binds its receptor on hepatocytes, adenylate cyclase produces cAMP, which activates PKA. PKA then phosphorylates multiple targets:

PFK-2/FBPase-2: phosphorylation activates the FBPase-2 domain (degrading F2,6BP) and inactivates the PFK-2 domain (synthesizing F2,6BP), suppressing glycolysis and promoting gluconeogenesis.
Pyruvate kinase: phosphorylation inhibits the enzyme, preventing phosphoenolpyruvate from being converted to pyruvate, which supports gluconeogenesis.
Glycogen phosphorylase kinase: phosphorylation activates glycogen phosphorylase, promoting glycogen breakdown.

Reciprocal regulation of glycolysis vs. gluconeogenesis

Three substrate cycles define the reciprocal regulation:

PFK-1 / FBPase-1: F2,6BP activates PFK-1 and inhibits FBPase-1. Insulin raises F2,6BP (glycolysis on, gluconeogenesis off). Glucagon lowers F2,6BP (reverse).
Pyruvate kinase / PEP carboxykinase + pyruvate carboxylase: PKA phosphorylation inhibits pyruvate kinase; transcriptional regulation induces PEPCK and pyruvate carboxylase during fasting.
Hexokinase / glucose-6-phosphatase: Hexokinase (and glucokinase in liver) traps glucose in the cell; glucose-6-phosphatase (expressed in liver and kidney) releases free glucose during gluconeogenesis.

AMPK: structure and activation

AMP-activated protein kinase is a heterotrimeric complex: alpha (catalytic), beta (scaffold, binds glycogen), and gamma (regulatory, contains four CBS domains that bind AMP, ADP, and ATP competitively).

AMPK is activated by three mechanisms:

AMP binding to the gamma subunit promotes a conformational change that facilitates phosphorylation at Thr172 on the alpha subunit by upstream kinases and inhibits dephosphorylation by protein phosphatases.
Direct phosphorylation at Thr172 by upstream kinases: LKB1 (constitutively active, the primary upstream kinase) and CaMKKbeta (activated by elevated intracellular Ca $^{2 +}$ ).
ADP binding provides a weaker activation than AMP, protecting Thr172 from dephosphorylation without promoting phosphorylation.

The net effect: AMPK activity is ultrasensitive to the cellular AMP/ATP and ADP/ATP ratios. Even a small drop in ATP produces a large increase in AMPK activity because AMP concentration is amplified by the adenylate kinase reaction ( $2 ADP \leftrightarrow ATP + AMP$ ).

AMPK downstream targets:

Acetyl-CoA carboxylase (ACC): AMPK phosphorylates and inhibits ACC, reducing malonyl-CoA. This relieves CPT1 inhibition, promoting fatty acid oxidation (energy-producing) and blocking fatty acid synthesis (energy-consuming).
mTORC1: AMPK phosphorylates TSC2 and Raptor, inhibiting mTORC1 and suppressing protein synthesis and cell growth.
GLUT4 translocation: AMPK promotes GLUT4 insertion into the plasma membrane, increasing glucose uptake in skeletal muscle.
SREBP-1c: AMPK suppresses SREBP-1c activity, reducing expression of lipogenic genes.

Core model Intermediate+

The Goldbeter-Koshland ultrasensitivity model

Covalent modification cycles (kinase/phosphatase pairs) can produce switch-like responses rather than gradual ones. The Goldbeter-Koshland (1981) model shows that when the modifying enzymes operate near saturation (zero-order kinetics with respect to substrate), the fraction of phosphorylated protein $W$ changes ultrasensitively with the ratio of kinase to phosphatase activity:

W = \frac{2 V _{1} K _{2}}{B + B ^{2} - 4 ( V _{1} - V _{2} ) V _{2} K _{2} / K _{1}}

where $V_{1}, V_{2}$ are the maximal velocities of the kinase and phosphatase, and $K_{1}, K_{2}$ are their Michaelis constants. The key insight: when $K_{1}$ and $K_{2}$ are small relative to total substrate concentration (zero-order regime), the response becomes highly ultrasensitive — a small change in kinase activity produces a large change in phosphorylation state. This zero-order ultrasensitivity converts a graded hormonal signal into an all-or-nothing metabolic switch.

Fructose-2,6-bisphosphate as the key signal

F2,6BP concentration in hepatocytes is determined by the net activity of the bifunctional PFK-2/FBPase-2 enzyme. The concentration of F2,6BP changes by more than 100-fold between fed and fasted states (from ~10 $μ$ M to <0.1 $μ$ M), making it one of the most dynamically regulated metabolites in the cell. This large dynamic range arises from the covalent modification switch described above: PKA phosphorylation flips the enzyme from net PFK-2 activity to net FBPase-2 activity, and the Goldbeter-Koshland mechanism makes the switch sharp.

Substrate cycles and thermogenic waste

A substrate cycle (futile cycle) occurs when two opposing metabolic reactions run simultaneously. The glycolysis/gluconeogenesis pair is the most prominent example. At the PFK-1/FBPase-1 junction, the net cost is 1 ATP per cycle (PFK-1 uses 1 ATP; FBPase-1 hydrolyzes water without recovering ATP). Under most conditions the cycle rate is kept near zero by reciprocal regulation. However, some thermogenic substrate cycling is physiologically important: bumblebees use substrate cycling between fructose-6-phosphate and fructose-1,6-bisphosphate to generate heat and maintain flight muscle temperature in cold environments.

Metabolic control analysis and the AMPK switch in disease Master

Metabolic control analysis: flux control and elasticity

Metabolic control analysis (MCA), introduced in unit 17.04.01, provides the quantitative framework for understanding how control is distributed across a pathway. For a pathway with steady-state flux $J$ through enzymes $E_{1}, \dots, E_{n}$ , the flux control coefficient of enzyme $i$ is

C_{J}^{E_{i}} = \frac{\partial ln J}{\partial ln [ E _{i} ]}

and the summation theorem requires $\sum_{i} C_{J}^{E_{i}} = 1$ . The elasticity $ε_{S}^{v_{i}} = \partial ln v_{i} / \partial ln [S]$ measures how enzyme velocity $v_{i}$ responds to changes in metabolite concentration $[S]$ , and the connectivity theorem $\sum_{i} C_{J}^{E_{i}} ε_{S}^{v_{i}} = 0$ couples control coefficients to the kinetic properties of individual enzymes.

Applied to glycolytic regulation, MCA reveals that the allosteric regulation of PFK-1 by ATP, AMP, and F2,6BP does not give PFK-1 a flux control coefficient of 1. Experimental measurements place PFK-1's coefficient at 0.2–0.3 under most conditions. The regulatory significance of PFK-1 lies not in carrying the majority of flux control but in its gain — the large change in enzyme activity per unit change in effector concentration. An enzyme can have high regulatory gain and low flux control coefficient simultaneously; the two concepts are distinct.

The allosteric architecture of PFK-1 creates high gain through cooperative substrate binding (Hill coefficient ~1.8 for fructose-6-phosphate in the absence of effectors, rising to ~2.5 with AMP) and through the amplification provided by the MWC concerted transition. Adding F2,6BP shifts $L$ by more than two orders of magnitude, producing a 20-fold change in enzyme activity with only a 10-fold change in F2,6BP concentration. This is the kinetic substrate for the Goldbeter-Koshland ultrasensitivity at the PFK-2/FBPase-2 junction: the bifunctional enzyme generates a large F2,6BP transient, and PFK-1 converts that transient into a large flux change.

AMPK in diabetes: the metformin mechanism

Metformin, the first-line drug for Type 2 diabetes, acts primarily through AMPK activation. The mechanism involves inhibition of mitochondrial Complex I, which reduces ATP production and raises the AMP/ATP ratio. AMPK activation in the liver suppresses gluconeogenesis through multiple downstream effects: inhibition of ACC (reducing the acetyl-CoA and malonyl-CoA that support gluconeogenic gene expression), suppression of PEPCK and G6Pase transcription through CREB and FoxO1 inhibition, and reduced hepatic fat accumulation (which itself impairs insulin sensitivity).

Recent work has shown that metformin also has AMPK-independent effects at therapeutic concentrations, including direct inhibition of mitochondrial glycerophosphate dehydrogenase and effects on the gut microbiome. The relative contribution of AMPK-dependent and AMPK-independent mechanisms remains an active research area, but the AMPK-dependent pathway is the best-characterized and the one most directly relevant to the metabolic regulation framework of this unit.

AMPK in cancer: the Warburg effect and metabolic reprogramming

The Warburg effect — aerobic glycolysis in cancer cells — represents a deliberate metabolic reprogramming that bypasses normal regulatory controls. Oncogenes (MYC, HIF-1 $α$ , PI3K/Akt) upregulate glucose transporters and glycolytic enzymes while downregulating oxidative phosphorylation. AMPK functions as a tumor suppressor in this context: it inhibits mTORC1 (reducing the biosynthetic flux that supports rapid proliferation), phosphorylates p53 (activating cell-cycle checkpoints), and suppresses the lipogenic program (reducing the membrane lipid synthesis required for cell division).

Loss-of-function mutations in LKB1 (the primary upstream kinase for AMPK) are found in lung adenocarcinomas and Peutz-Jeghers syndrome. LKB1 loss means AMPK cannot be activated by energy stress, removing a brake on mTORC1-driven growth. The therapeutic implication is that AMPK activators (metformin, AICAR) may have anti-tumor effects in LKB1-competent cancers, though in LKB1-null tumors AMPK activation is impossible and alternative strategies are needed.

The metabolic profile of AMPK-null cells illustrates the enzyme's integrative role: without AMPK, cells cannot downregulate fatty acid synthesis or protein synthesis in response to energy stress. They continue to consume ATP at high rates even as ATP levels fall, eventually triggering necrosis or apoptosis. AMPK is not merely a sensor — it is the decision engine that determines whether a cell invests its limited resources in growth or in survival.

Quantitative modeling of the AMPK activation curve

The AMPK activation curve as a function of the AMP/ATP ratio can be approximated by a Hill-type function:

AMPK activity = \frac{V _{m a x} \cdot ([ AMP ] / [ ATP ] ) ^{n}}{K _{0.5}^{n} + ([ AMP ] / [ ATP ] ) ^{n}}

with an apparent Hill coefficient $n \approx 2$ – $4$ , reflecting cooperative binding at the four CBS domains on the gamma subunit and the dual mechanism of AMP action (promoting Thr172 phosphorylation and inhibiting its dephosphorylation). The adenylate kinase reaction amplifies the signal: because $[ATP] ≫ [ADP] ≫ [AMP]$ under normal conditions (typical ratios in muscle: ATP ~5 mM, ADP ~0.5 mM, AMP ~0.05 mM), a 10% drop in ATP produces a ~400% increase in AMP, making AMPK exquisitely sensitive to small changes in energy charge.

The energy charge of the cell, defined as $([A T P] + 0.5 [A D P]) / ([A T P] + [A D P] + [A M P])$ , is normally maintained between 0.8 and 0.95 in healthy cells. AMPK activates sharply as energy charge drops below 0.8, providing a threshold detector that triggers the catabolic switch.

Substrate cycles as amplifiers and thermogenic devices

Substrate cycles are not merely wasteful. At low cycle rates they serve as amplifiers: a small change in the forward reaction rate (glycolysis) combined with an opposing small change in the reverse reaction rate (gluconeogenesis) produces a large change in net flux. The amplification factor equals the ratio of gross cycling flux to net flux. At the PFK-1/FBPase-1 junction, the cycling flux can be 10–20% of net flux, providing a 5–10 fold amplification of the regulatory signal from F2,6BP.

At high cycle rates, substrate cycles become thermogenic devices. The bumblebee example (cycling fructose-6-phosphate and fructose-1,6-bisphosphate at high rates to warm flight muscles before takeoff in cold weather) demonstrates that futile cycles have adaptive significance. In mammals, brown adipose tissue uses a different thermogenic strategy (UCP1 uncoupling), but substrate cycling contributes to diet-induced thermogenesis and fever responses. The cost is ATP consumption without productive work, but the benefit is rapid temperature regulation.

Connections Master

Metabolic regulation connects to every other pathway in the cell through shared metabolites and signaling molecules:

Glycolysis and gluconeogenesis are reciprocally regulated at three substrate cycles (hexokinase/G6Pase, PFK-1/FBPase-1, pyruvate kinase/PEP carboxykinase), ensuring that carbon flows in only one direction at a time. The control logic was developed in unit 17.04.01.
Fatty acid synthesis and beta-oxidation are reciprocally regulated by malonyl-CoA (synthesized by ACC, which is phosphorylated and inhibited by AMPK). This prevents the simultaneous operation of fat-building and fat-burning pathways, as described in unit 17.04.04.
The citric acid cycle is regulated by the NADH/NAD $^{+}$ and ATP/ADP ratios at three allosteric enzymes (citrate synthase, isocitrate dehydrogenase, alpha-ketoglutarate dehydrogenase). These ratios integrate cycle flux with the energy state of the entire cell.
Signal transduction cascades (G-protein-coupled receptors, receptor tyrosine kinases) feed into metabolic regulation through PKA, PKB/Akt, and AMPK. Hormonal signals from insulin, glucagon, epinephrine, and adiponectin converge on the same metabolic enzymes through different signaling pathways, as described in units 17.07.01 and 17.07.02.
Transcriptional regulation complements acute metabolic control on longer timescales. AMPK and insulin signaling both regulate gene expression through transcription factors (SREBP-1c, ChREBP, FoxO1, PGC-1 $α$ ), reprogramming the enzyme portfolio of the cell over hours to days.
The pentose phosphate pathway competes with glycolysis for glucose-6-phosphate. Its flux is regulated by NADP $^{+}$ /NADPH ratio and by the metabolic demand for ribose-5-phosphate (nucleotide synthesis) and NADPH (fatty acid synthesis, antioxidant defense).

Historical notes Master

The concept of allosteric regulation was formalized by Jacques Monod, Jeffries Wyman, and Jean-Pierre Changeux in their landmark 1965 paper proposing the MWC model. Their work built on earlier observations by Lwoff and Monod on enzyme induction in bacteria and by Perutz on hemoglobin cooperativity. The MWC paper introduced the idea that proteins could exist in discrete conformational states and that ligand binding could shift the equilibrium between them — a concept that now underpins all of molecular pharmacology and metabolic regulation.

The sequential model (KNF) was proposed by Daniel Koshland, George Nemethy, and D. Filmer in 1966 as an alternative to the concerted MWC mechanism. The debate between concerted and sequential models drove much of the experimental work on allosteric enzymes in the 1970s and 1980s, with the eventual recognition that real enzymes can display features of both.

The discovery of reversible protein phosphorylation as a regulatory mechanism is credited to Edmond Fischer and Edwin Krebs, who showed in 1955 that phosphorylase kinase activates glycogen phosphorylase through phosphorylation. This work, which earned them the 1992 Nobel Prize, established that covalent modification is a general regulatory strategy. The subsequent discovery of hundreds of kinases and phosphatases in eukaryotic genomes confirmed that phosphorylation is the most widespread form of post-translational regulation.

AMPK was discovered independently by several groups in the late 1970s and early 1980s. The key figure in developing the modern understanding of AMPK is D. Grahame Hardie at the University of Dundee, who showed in the 1990s and 2000s that AMPK is the central energy sensor in eukaryotic cells, activated by AMP and ADP, and regulating a vast network of downstream targets. The connection between AMPK and the tumor suppressor LKB1 was established in 2004 by three groups simultaneously, revealing that a kinase originally identified in the context of a rare inherited cancer syndrome (Peutz-Jeghers) is the primary upstream activator of the cell's energy sensor.

The Goldbeter-Koshland model of zero-order ultrasensitivity was published by Albert Goldbeter and Daniel Koshland in 1981, showing that covalent modification cycles can produce sharp, switch-like responses without requiring cooperative binding. This result was initially counterintuitive — a single-substrate, single-enzyme modification reaction producing cooperativity — and it took several years to be widely appreciated. It is now recognized as a fundamental design principle of signaling networks.

Fructose-2,6-bisphosphate was discovered in 1980 by Henri-Gery Hers and Emile Van Schaftingen in Louvain, who identified it as the mysterious activator of liver PFK-1 that appeared after glucagon treatment. The identification of the bifunctional PFK-2/FBPase-2 enzyme followed in 1981, establishing the molecular basis for reciprocal regulation of glycolysis and gluconeogenesis.

Exercises Intermediate+

Exercise 1 (medium, short answer).

In the MWC model, what is the effect of an allosteric activator on the parameter $L$ ? Explain how this changes the substrate saturation curve.

Hint

$L$ is the ratio of enzyme in the T state to the R state in the absence of ligand. An activator should shift the equilibrium toward the high-affinity state.

Answer

An allosteric activator binds preferentially to the R state, stabilizing it and shifting the T $\leftrightarrow$ R equilibrium toward R. This effectively reduces the value of $L$ (the T/R ratio). In the substrate saturation curve ( $Y$ vs. $[S]$ ), reducing $L$ shifts the curve to the left (lower $[S]$ required for half-saturation) and makes it more hyperbolic (less sigmoidal), because the enzyme spends more time in the high-affinity R state even without substrate bound. The opposite holds for inhibitors, which increase $L$ and shift the curve rightward.

Exercise 2 (medium, short answer).

Explain the dual mechanism by which AMP activates AMPK. Why is this dual mechanism more effective than a single mechanism?

Hint

AMP binding to the gamma subunit affects AMPK through both phosphorylation of Thr172 and dephosphorylation at the same site.

Answer

AMP activates AMPK through two synergistic mechanisms: (1) it allosterically promotes phosphorylation of Thr172 by upstream kinases LKB1 and CaMKKbeta, and (2) it inhibits dephosphorylation of Thr172 by protein phosphatases. The dual mechanism creates an ultrasensitive response: a small increase in AMP both accelerates the "on" reaction and decelerates the "off" reaction at the same site. The net effect is much larger than either mechanism alone, because the activation is multiplicative rather than additive. This is an example of the Goldbeter-Koshland principle applied to a biological circuit.

Exercise 3 (medium, short answer).

A patient takes metformin for Type 2 diabetes. Explain the chain of events from metformin administration to reduced hepatic glucose output.

Hint

Metformin inhibits mitochondrial Complex I. Trace the consequences: ATP production, AMP/ATP ratio, AMPK activation, and downstream effects on gluconeogenesis.

Answer

Metformin inhibits mitochondrial Complex I, reducing electron transport and oxidative phosphorylation. Hepatic ATP production decreases, raising the AMP/ATP ratio. AMP activates AMPK. Activated AMPK phosphorylates and inhibits acetyl-CoA carboxylase (reducing malonyl-CoA and thereby promoting fatty acid oxidation over synthesis), suppresses mTORC1 (reducing the translational capacity for gluconeogenic enzymes), and inhibits the transcription factors CREB and FoxO1 that drive PEPCK and G6Pase expression. The combined effect is reduced hepatic gluconeogenesis and lower blood glucose. AMPK also promotes GLUT4 translocation in muscle, increasing peripheral glucose uptake.

Exercise 4 (hard, short answer).

A cell has [ATP] = 5.0 mM, [ADP] = 0.5 mM, and [AMP] = 0.05 mM. Calculate the energy charge. If ATP drops by 10% and adenylate kinase is at equilibrium ( $K_{e q} \approx 1$ ), estimate the new [AMP] and the fold-increase in the AMP/ATP ratio.

Hint

Energy charge = $([A T P] + 0.5 [A D P]) / ([A T P] + [A D P] + [A M P])$ . Use the adenylate kinase equilibrium: $[A T P] [A M P] / [A D P]^{2} \approx 1$ . If ATP drops 10% to 4.5 mM and ADP rises correspondingly, recalculate AMP from the equilibrium expression.

Answer

Initial energy charge = $(5.0 + 0.5 \times 0.5) / (5.0 + 0.5 + 0.05) = 5.25/5.55 = 0.946$ .

If ATP drops 10%: new [ATP] = 4.5 mM. By mass balance, the 0.5 mM drop in ATP increases ADP (and potentially AMP through adenylate kinase). Assuming the 0.5 mM transfers to ADP first: new [ADP] $\approx$ 1.0 mM. From the adenylate kinase equilibrium: $[AMP] = [ADP]^{2} / [ATP] = (1.0)^{2} /4.5 = 0.222$ mM.

Original AMP/ATP = 0.05/5.0 = 0.01. New AMP/ATP = 0.222/4.5 = 0.049. Fold increase = $0.049/0.01 = 4.9$ . A 10% drop in ATP produces nearly a 5-fold increase in the AMP/ATP ratio. This amplification is why AMPK can detect small changes in energy charge with high sensitivity.

Exercise 5 (hard, short answer).

Explain how zero-order ultrasensitivity arises in a covalent modification cycle. What condition must the kinase and phosphatase satisfy for ultrasensitivity to occur?

Hint

Consider the Michaelis-Menten kinetics of both enzymes. What happens when the substrate concentration (the target protein) is much larger than $K_{m}$ ?

Answer

In a covalent modification cycle, a kinase converts substrate $W$ to phosphorylated product $W^{*}$ , while a phosphatase converts $W^{*}$ back to $W$ . When both enzymes operate near saturation — meaning the total substrate concentration $[W]_{T}$ is much greater than both $K_{m}$ values — the velocity of each reaction becomes approximately independent of substrate concentration (zero-order kinetics). In this regime, the steady-state fraction of $W^{*}$ is determined by the balance of the two maximal velocities ( $V_{1}$ for kinase, $V_{2}$ for phosphatase). Because both reactions are nearly flat functions of substrate concentration, a small shift in the $V_{1} / V_{2}$ ratio produces a large shift in the steady-state — the system "snaps" from one state to the other. The ultrasensitivity increases as the ratio $[W]_{T} / K_{m}$ increases, approaching an ideal switch in the limit of infinite saturation.

Exercise 6 (hard, short answer).

Cancer cells with loss-of-function mutations in LKB1 cannot activate AMPK in response to energy stress. Predict the metabolic consequences for (a) mTORC1 activity, (b) fatty acid synthesis, and (c) cell survival under glucose deprivation.

Hint

Without LKB1, AMPK cannot be phosphorylated at Thr172. Consider each AMPK downstream target and what happens when the brake is removed.

Answer

(a) mTORC1 activity increases constitutively. AMPK normally inhibits mTORC1 by phosphorylating TSC2 and Raptor. Without AMPK activation, mTORC1 drives protein synthesis, ribosome biogenesis, and cell growth regardless of energy availability, contributing to uncontrolled proliferation.

(b) Fatty acid synthesis increases. AMPK normally phosphorylates and inhibits ACC, limiting malonyl-CoA production. Without this brake, ACC runs unchecked, producing abundant malonyl-CoA for fatty acid synthesis and simultaneously inhibiting CPT1 to block fatty acid oxidation. This is consistent with the lipogenic phenotype of many cancers.

(c) Cell survival under glucose deprivation drops sharply. Without AMPK, the cell cannot activate catabolic pathways (fatty acid oxidation, autophagy, GLUT4 translocation) in response to falling ATP. The cell continues to consume ATP through biosynthetic pathways while failing to increase energy production, leading to ATP depletion and necrotic cell death. This synthetic lethality is the basis for therapeutic strategies that exploit LKB1-null tumors' inability to cope with energy stress.

Bibliography Master

Alberts, B., Hopkin, K., Johnson, A., Morgan, D., Roberts, K., & Walter, P. — Molecular Biology of the Cell, 7th ed. (Garland Science, 2022). Chapter 15: Cell Signaling — Metabolic regulation.
Berg, J. M., Tymoczko, J. L., & Stryer, L. — Biochemistry, 9th ed. (W. H. Freeman, 2019). Chapter 16: Glycolysis and Gluconeogenesis — Regulation.
Hardie, D. G., Ross, F. A., & Hawley, S. A. — "AMPK: a nutrient and energy sensor that maintains energy homeostasis." Nat. Rev. Mol. Cell Biol. 13 (2012): 251–262.
Monod, J., Wyman, J., & Changeux, J. P. — "On the nature of allosteric transitions: a plausible model." J. Mol. Biol. 12 (1965): 88–118.
Koshland, D. E., Nemethy, G., & Filmer, D. — "Comparison of experimental binding data and theoretical models in proteins containing subunits." Biochemistry 5 (1966): 365–385.
Goldbeter, A. & Koshland, D. E. — "An amplified sensitivity arising from covalent modification in biological systems." Proc. Natl. Acad. Sci. USA 78 (1981): 6840–6844.
Hardie, D. G. — "AMPK: a target for drugs both ancient and modern." Cell Metab. 22 (2015): 331–333.
Hers, H. G. & Van Schaftingen, E. — "Fructose 2,6-bisphosphate 2 years after its discovery." Biochem. J. 206 (1982): 1–12.
Kacser, H. & Burns, J. A. — "The control of flux." Symp. Soc. Exp. Biol. 27 (1973): 65–104.
Hawley, S. A., et al. — "Complexes between the LKB1 tumor suppressor, STRAD alpha/beta and MO25 alpha/beta are upstream kinases in the AMP-activated protein kinase cascade." J. Biol. 2 (2003): 28.

Prerequisites

17.04.01 pending
17.04.04 pending

Tier anchors

beginner: Alberts et al., Molecular Biology of the Cell, 7th ed. (2022), Ch. 15
intermediate: Berg, Tymoczko & Stryer, Biochemistry, 9th ed. (2019), Ch. 16
master: Hardie, D. G. — AMPK: A Key Regulator of Energy Balance (2015); Hardie, D. G. et al. — Nat. Rev. Mol. Cell Biol. 13 (2012) 251-262

References

Alberts et al. — Molecular Biology of the Cell, 7th ed. (Garland Science, 2022) · Ch. 15 Cell Signaling — Metabolic regulation
Berg, Tymoczko & Stryer — Biochemistry, 9th ed. (W. H. Freeman, 2019) · Ch. 16 Glycolysis and Gluconeogenesis — Regulation
Hardie, D. G., Ross, F. A. & Hawley, S. A. — AMPK: a nutrient and energy sensor that maintains energy homeostasis · Nat. Rev. Mol. Cell Biol. 13 (2012) 251-262
Monod, J., Wyman, J. & Changeux, J. P. — On the nature of allosteric transitions: a plausible model · J. Mol. Biol. 12 (1965) 88-118

Estimated time

beginner: 15m
intermediate: 30m
master: 45m