12. Metabolic Integration & Regulation

The thousands of metabolic reactions in the body do not operate in isolation. Hormonal signals, allosteric effectors, and tissue-specific enzyme expression coordinate fuel storage and mobilization across the fed-fasted-starvation continuum — a symphony of metabolic regulation.

Fed vs Fasted vs Starvation States

The metabolic state of the organism shifts dramatically depending on nutrient availability. The insulin/glucagon ratio is the master switch that determines whether the body is in an anabolic (storage) or catabolic (mobilization) mode.

Fed State (0-4 h)

Insulin/glucagon ratio: HIGH (~30)

Glycolysis and glycogen synthesis active
Fatty acid synthesis and lipogenesis active
Protein synthesis stimulated
Gluconeogenesis suppressed
Lipolysis inhibited

Fasted State (4-24 h)

Insulin/glucagon ratio: LOW (~2)

Glycogenolysis in liver (depleted by ~24 h)
Gluconeogenesis activated
Lipolysis releasing fatty acids
Beta-oxidation in muscle, liver
Mild ketogenesis begins

Starvation (>3 days)

Insulin/glucagon ratio: VERY LOW (~0.5)

Massive ketogenesis (brain adapts to ketones)
Gluconeogenesis sustained (from glycerol, amino acids)
Muscle protein catabolism decreases (protein-sparing)
Basal metabolic rate falls 20-25%
Survival: 40-60 days (depending on fat stores)

The Critical Adaptation to Starvation

The brain normally consumes ~120 g glucose/day (~480 kcal), but the body's total glycogen stores provide only ~700 kcal. Without metabolic adaptation, gluconeogenesis would require ~75 g of muscle protein/day to supply the brain, leading to fatal muscle wasting within weeks. The key adaptation is the brain's shift to ketone bodies:

$$\text{Starvation brain fuel:}\;\sim 75\%\;\text{ketone bodies} + \sim 25\%\;\text{glucose}$$

This reduces glucose requirement to ~30 g/day, decreasing muscle protein catabolism from ~75 g/day to ~20 g/day — the protein-sparing effect that extends survival from weeks to months.

Derivation 1: Insulin Signaling Cascade

Insulin, discovered by Banting and Best in 1921 (Nobel Prize 1923), signals the fed state. It binds to the insulin receptor tyrosine kinase, initiating a phosphorylation cascade:

$$\text{Insulin} + \text{IR} \rightarrow \text{IR}^* \xrightarrow{\text{autophosphorylation}} \text{p-IR} \xrightarrow{\text{phosphorylation}} \text{p-IRS-1}$$

$$\text{p-IRS-1} \rightarrow \text{PI3K} \rightarrow \text{PIP}_2 \xrightarrow{\text{PI3K}} \text{PIP}_3 \rightarrow \text{PDK1} \rightarrow \text{Akt/PKB}$$

Akt/PKB: The Master Metabolic Kinase

Activated Akt phosphorylates numerous downstream targets:

AS160: phosphorylation triggers GLUT4 vesicle translocation to the plasma membrane, increasing glucose uptake 10-20 fold in muscle and adipose tissue
GSK-3: phosphorylation (inactivation) relieves inhibition of glycogen synthase, promoting glycogen synthesis
FOXO1: phosphorylation causes nuclear exclusion, suppressing gluconeogenic gene expression (PEPCK, G6Pase)
mTORC1: activation (via TSC2 phosphorylation) promotes protein synthesis and cell growth

Signal Amplification

Each step in the cascade amplifies the signal. If each kinase activates 10 downstream molecules:

$$\text{Amplification} = 10^n\;\text{where}\;n = \text{number of cascade steps}$$

With 4 kinase steps (IR $\rightarrow$ IRS $\rightarrow$ PI3K $\rightarrow$ Akt), a single insulin molecule can trigger $10^4 = 10{,}000$ downstream phosphorylation events, explaining the exquisite sensitivity of insulin signaling.

Derivation 2: Glucagon Signaling — cAMP and PKA

Glucagon, secreted by pancreatic $\alpha$-cells in response to low blood glucose, opposes insulin's effects. It acts primarily on the liver through the cAMP/PKA pathway, discovered by Earl Sutherland (Nobel Prize 1971):

$$\text{Glucagon} + \text{GPCR} \rightarrow \text{G}_s\alpha\text{-GTP} \rightarrow \text{Adenylyl Cyclase} \rightarrow \text{cAMP}\uparrow$$

$$\text{cAMP} + \text{PKA}_{\text{inactive}}\;(R_2C_2) \rightarrow 2\;\text{cAMP-R} + 2\;C_{\text{active}}$$

PKA Targets in the Liver

Active PKA catalytic subunits phosphorylate key metabolic enzymes:

Glycogen phosphorylase kinase $\rightarrow$ phosphorylase $a$ (active) $\rightarrow$ glycogenolysis ON
Glycogen synthase $\rightarrow$ synthase $b$ (inactive) $\rightarrow$ glycogen synthesis OFF
PFK-2/FBPase-2 $\rightarrow$ FBPase-2 active, PFK-2 inactive $\rightarrow$ [F2,6BP]$\downarrow$ $\rightarrow$ glycolysis OFF, gluconeogenesis ON
CREB (transcription factor) $\rightarrow$ induces PEPCK and G6Pase gene expression

cAMP Dynamics

The cAMP signal is terminated by phosphodiesterase (PDE):

$$\text{cAMP} \xrightarrow{\text{PDE}} \text{5'-AMP}\quad(\text{inactive})$$

The steady-state [cAMP] depends on the balance between synthesis (adenylyl cyclase) and degradation (PDE):

$$[\text{cAMP}]_{ss} = \frac{V_{\text{AC}}}{k_{\text{PDE}}}$$

Caffeine and theophylline inhibit PDE, raising [cAMP] and mimicking glucagon-like effects (increased lipolysis, glycogenolysis) — explaining the metabolic effects of coffee consumption.

Derivation 3: Tissue-Specific Metabolism

Different tissues have different metabolic profiles dictated by their enzyme complement, fuel access, and physiological role.

Liver: The Metabolic Hub

Unique enzymes: glucokinase (high $K_m$ for glucose, acts as glucose sensor), glucose-6-phosphatase (allows glucose release), PEPCK (gluconeogenesis). The liver is the only organ that exports glucose to maintain blood glucose homeostasis. It uses fatty acids and amino acids as its own fuel, sparing glucose for the brain. During fasting, it produces ketone bodies but cannot use them (lacks succinyl-CoA:3-oxoacid CoA-transferase).

Brain: Glucose-Dependent (Usually)

Normally depends on glucose (no significant fatty acid oxidation — fatty acids cannot cross the blood-brain barrier). Uses ~20% of body's O$_2$ and ~60% of blood glucose despite being only 2% of body mass. During prolonged fasting, adapts to use ketone bodies (up to 75% of energy needs), reducing glucose requirement from ~120 g/day to ~30 g/day.

Skeletal Muscle: Versatile Fuel Consumer

At rest: primarily fatty acids. During moderate exercise: mix of fatty acids and glucose. During intense exercise: glucose and glycogen (anaerobic glycolysis produces lactate). Contains hexokinase (not glucokinase) and lacks glucose-6-phosphatase — muscle glycogen cannot contribute to blood glucose directly. GLUT4 (insulin-sensitive) provides glucose uptake; exercise independently stimulates GLUT4 translocation via AMPK.

Adipose Tissue: Energy Storage Depot

Stores triacylglycerols (70,000-100,000 kcal in a typical adult). Lipolysis (hormone-sensitive lipase, activated by PKA/glucagon/epinephrine, inhibited by insulin) releases fatty acids and glycerol. Glycerol cannot be reused by adipose tissue (lacks glycerol kinase) — it goes to the liver for gluconeogenesis. GLUT4-dependent glucose uptake provides glycerol-3-phosphate for TAG re-esterification.

Kidney: Gluconeogenic Organ

During prolonged fasting, the kidney contributes up to 50% of gluconeogenesis (from glutamine). It also produces NH$_4^+$ from glutamine to buffer acid in urine. Renal gluconeogenesis from glutamine: glutamine $\rightarrow$ glutamate $\rightarrow$ $\alpha$-ketoglutarate $\rightarrow$ (TCA) $\rightarrow$ OAA $\rightarrow$ glucose.

AMPK: The Master Energy Sensor

AMP-activated protein kinase (AMPK) is the cell's fuel gauge — activated when energy is depleted (high AMP/ATP ratio) and orchestrating a comprehensive metabolic response to restore energy balance.

Activation Mechanism

AMPK is a heterotrimeric enzyme ($\alpha\beta\gamma$) activated by a three-pronged mechanism:

$$\text{AMP binding to }\gamma\text{ subunit:}\begin{cases} \text{1. Allosteric activation (2-5 fold)} \\ \text{2. Promotes Thr-172 phosphorylation by LKB1} \\ \text{3. Inhibits Thr-172 dephosphorylation by PP2C} \end{cases}$$

The net effect of all three mechanisms is >100-fold activation. The adenylate kinase reaction ($2\;\text{ADP} \rightleftharpoons \text{ATP} + \text{AMP}$) amplifies small changes in ATP into large changes in AMP, making AMPK exquisitely sensitive:

$$\text{If [ATP] drops 10\%:}\;[\text{AMP}]\;\text{rises} \sim 300\%\quad(\text{because } K_{eq} = \frac{[\text{ATP}][\text{AMP}]}{[\text{ADP}]^2} \approx 0.44)$$

AMPK Downstream Targets

Catabolic Pathways ON

Fatty acid oxidation (ACC phosphorylation $\rightarrow$ malonyl-CoA$\downarrow$)
Glucose uptake (GLUT4 translocation in muscle)
Autophagy (ULK1 phosphorylation)
Mitochondrial biogenesis (PGC-1$\alpha$ activation)

Anabolic Pathways OFF

Fatty acid synthesis (ACC inhibition)
Cholesterol synthesis (HMGCR inhibition)
Protein synthesis (mTORC1 inhibition via TSC2/Raptor)
Glycogen synthesis (via allosteric effects)

Pharmacological AMPK Activation

Several drugs and natural compounds activate AMPK:

Metformin: inhibits complex I $\rightarrow$ AMP/ATP ratio$\uparrow$ $\rightarrow$ AMPK activated indirectly
AICAR (5-aminoimidazole-4-carboxamide ribonucleoside): converted to ZMP, which mimics AMP and directly activates AMPK
Resveratrol: activates SIRT1 (NAD$^+$-dependent deacetylase), which deacetylates and activates LKB1, indirectly activating AMPK
Exercise: the most potent physiological AMPK activator in muscle; explains insulin-independent glucose uptake during exercise

Derivation 4: Metabolic Syndrome and Type 2 Diabetes

Insulin resistance is the hallmark of type 2 diabetes and metabolic syndrome. It represents a state where normal insulin concentrations produce subnormal biological responses. The dose-response curve shifts rightward:

$$\text{Response} = \frac{R_{\max} \cdot [\text{Insulin}]^{n_H}}{(\text{EC}_{50})^{n_H} + [\text{Insulin}]^{n_H}}$$

In insulin resistance, the EC$_{50}$ increases (rightward shift) and/or $R_{\max}$ decreases (reduced maximum response). A crucial insight is that different insulin targets have different sensitivities:

Hepatic gluconeogenesis suppression: most sensitive (EC$_{50} \approx 20$ pM)
Glycogen synthesis: intermediate (EC$_{50} \approx 30$ pM)
Glucose uptake (GLUT4): intermediate (EC$_{50} \approx 50$ pM)
Lipogenesis (ACC/FAS): least sensitive (EC$_{50} \approx 80$ pM)

This selective insulin resistance explains the metabolic paradox of type 2 diabetes: the liver fails to suppress gluconeogenesis (causing hyperglycemia) but continues to synthesize fat (causing steatosis/NAFLD). The pathways with higher EC$_{50}$ values (lipogenesis) remain responsive to the compensatory hyperinsulinemia.

Diagnostic Criteria (Metabolic Syndrome)

Three or more of the following (ATP III criteria):

Waist circumference >102 cm (men) or >88 cm (women)
Fasting triglycerides $\geq$ 150 mg/dL
HDL cholesterol <40 mg/dL (men) or <50 mg/dL (women)
Blood pressure $\geq$ 130/85 mmHg
Fasting glucose $\geq$ 100 mg/dL

Metabolic Adaptation in Disease States

Many diseases involve fundamental derangements of metabolic integration. Understanding these metabolic disruptions is essential for rational therapy.

Type 1 Diabetes: Absolute Insulin Deficiency

Autoimmune destruction of pancreatic $\beta$-cells eliminates insulin production. Without insulin, the body behaves metabolically as if in severe starvation — even in the fed state:

GLUT4 is not translocated: muscle and adipose cannot take up glucose $\rightarrow$ hyperglycemia
Unrestrained lipolysis: hormone-sensitive lipase is maximally active $\rightarrow$ massive free fatty acid release
Unrestrained hepatic ketogenesis: acetyl-CoA from beta-oxidation overwhelms TCA capacity $\rightarrow$ diabetic ketoacidosis (DKA)
Proteolysis: muscle protein breakdown releases amino acids for gluconeogenesis, worsening hyperglycemia

The paradox: cells are "starving in a sea of glucose." Blood glucose may exceed 500 mg/dL while cells cannot access it. DKA (pH <7.3, bicarbonate <15 mEq/L, ketones >10 mM) is a medical emergency treated with insulin, IV fluids, and electrolyte replacement.

Non-Alcoholic Fatty Liver Disease (NAFLD)

NAFLD (affecting ~25% of the global population) illustrates the metabolic paradox of insulin resistance. Hepatic steatosis results from:

$$\text{Fat accumulation} = \underbrace{\text{De novo lipogenesis}\uparrow}_{\text{insulin-driven}} + \underbrace{\text{FA uptake}\uparrow}_{\text{from adipose lipolysis}} - \underbrace{\text{FA oxidation}}_{\text{normal/decreased}} - \underbrace{\text{VLDL export}}_{\text{insufficient}}$$

The key insight: the liver becomes "selectively insulin-resistant" — insulin fails to suppress gluconeogenesis (FOXO1 pathway) but continues to drive lipogenesis (SREBP-1c pathway), leading to simultaneous hyperglycemia and hepatic fat accumulation.

Alcoholic Liver Disease

Ethanol metabolism profoundly disrupts hepatic redox balance:

$$\text{Ethanol} \xrightarrow{\text{ADH}} \text{Acetaldehyde} + \text{NADH} \xrightarrow{\text{ALDH}} \text{Acetate} + \text{NADH}$$

The excess NADH (high NADH/NAD$^+$ ratio) inhibits beta-oxidation (3-hydroxyacyl-CoA dehydrogenase), gluconeogenesis (shifted to lactate production), and the TCA cycle (malate accumulation), while promoting fatty acid synthesis and ketogenesis. This explains alcoholic fatty liver, hypoglycemia, lactic acidosis, and ketoacidosis seen in chronic alcohol abuse.

Derivation 5: Cancer Metabolism — The Warburg Effect

In 1924, Otto Warburg (Nobel Prize 1931) observed that cancer cells preferentially ferment glucose to lactate even in the presence of oxygen — aerobic glycolysis (the "Warburg effect"). This seems paradoxical: glycolysis produces only 2 ATP per glucose vs ~32 from complete oxidation. Why would rapidly proliferating cells choose an inefficient pathway?

The Biosynthetic Advantage

The modern understanding is that the Warburg effect supports anabolic metabolism. Rapidly dividing cells need biosynthetic precursors more than ATP:

$$\text{Glucose} \rightarrow \begin{cases} \text{Ribose-5-P (nucleotides)} & \text{via PPP} \\ \text{Glycerol-3-P (lipids)} & \text{via DHAP} \\ \text{Serine, Glycine (1C units)} & \text{via 3-PG} \\ \text{Citrate (fatty acids)} & \text{via pyruvate} \rightarrow \text{acetyl-CoA} \end{cases}$$

If all glucose were completely oxidized, these biosynthetic intermediates would not be available. By maintaining high glycolytic flux (upregulating glucose transporters and glycolytic enzymes), cancer cells divert intermediates into biosynthetic pathways while using lactate production to regenerate NAD$^+$ for continued glycolysis.

Quantitative Comparison

The rate of ATP production (not just yield per glucose) matters:

$$\text{Glycolysis rate: } \sim 100\times\text{ faster than oxidative phosphorylation}$$

$$\frac{d[\text{ATP}]}{dt}\bigg|_{\text{glycolysis}} = 2 \times v_{\text{glycolysis}} \gg \frac{d[\text{ATP}]}{dt}\bigg|_{\text{OxPhos}} = 32 \times v_{\text{OxPhos}}$$

When glucose is abundant, glycolysis can actually produce ATP faster than oxidative phosphorylation (despite lower yield per molecule). This "overflow metabolism" strategy is also used by yeast (the Crabtree effect) and rapidly proliferating immune cells. The exploitation of this metabolic reprogramming enables clinical FDG-PET imaging: fluorodeoxyglucose ($^{18}$F-FDG) accumulates in tumors due to their elevated glucose uptake.

Applications in Medicine and Research

GLP-1 Receptor Agonists

Semaglutide (Ozempic/Wegovy) and other GLP-1 agonists enhance insulin secretion, suppress glucagon, slow gastric emptying, and reduce appetite. They exploit the incretin effect: oral glucose stimulates more insulin than IV glucose because GLP-1 from gut L-cells amplifies the pancreatic response.

Metformin Mechanism

Metformin (first-line for type 2 diabetes since 1957) primarily inhibits hepatic gluconeogenesis. It mildly inhibits mitochondrial complex I, raising the AMP/ATP ratio, activating AMPK, and suppressing glucose production. It also alters the gut microbiome and increases GLP-1 secretion.

SGLT2 Inhibitors

Empagliflozin and dapagliflozin block glucose reabsorption in the proximal tubule, causing glycosuria (~70 g/day glucose excreted). Beyond glucose lowering, they reduce cardiovascular events and heart failure — likely via natriuresis, ketogenesis, and improved cardiac energy metabolism.

Cancer Metabolism Therapies

IDH1/2 inhibitors (ivosidenib, enasidenib) target mutant isocitrate dehydrogenase in AML, which produces the oncometabolite 2-hydroxyglutarate. Glutaminase inhibitors and LDHA inhibitors are in clinical trials. Metabolic imaging (FDG-PET) guides treatment response assessment.

Historical Context

Claude Bernard (1855) first demonstrated that the liver stores and releases glucose (glycogenesis), establishing the concept of the milieu interieur (internal environment). Banting and Best isolated insulin in 1921, transforming type 1 diabetes from a death sentence to a manageable condition. Earl Sutherland discovered cAMP as a "second messenger" in 1958, revealing how hormones communicate with intracellular enzymes — a paradigm that earned him the 1971 Nobel Prize and revolutionized signal transduction biology.

Modern Diabetes Therapeutics: A Metabolic Perspective

Understanding metabolic integration has enabled the development of multiple drug classes for type 2 diabetes, each targeting a different node in the metabolic network:

Metformin (Biguanide)

First-line therapy since 1957. Inhibits hepatic complex I $\rightarrow$ AMPK activation $\rightarrow$ suppresses hepatic glucose production. Also improves insulin sensitivity, alters gut microbiome, increases GLP-1 secretion. Extremely safe profile (lactic acidosis is rare).

Sulfonylureas (Glyburide, Glipizide)

Block K$_{\text{ATP}}$ channels on pancreatic $\beta$-cells $\rightarrow$ depolarization $\rightarrow$ Ca$^{2+}$ influx $\rightarrow$ insulin secretion. Effective but cause hypoglycemia and weight gain. K$_{\text{ATP}}$ channel = SUR1 + Kir6.2 subunits.

DPP-4 Inhibitors (Sitagliptin)

Inhibit dipeptidyl peptidase-4, preventing degradation of GLP-1 and GIP $\rightarrow$ prolonged incretin effect $\rightarrow$ glucose-dependent insulin secretion. Weight-neutral, low hypoglycemia risk.

Thiazolidinediones (Pioglitazone)

PPAR$\gamma$ agonists $\rightarrow$ promote adipocyte differentiation and lipid storage $\rightarrow$ reduce ectopic fat in liver/muscle $\rightarrow$ improve insulin sensitivity. Cause weight gain and fluid retention. Troglitazone withdrawn (hepatotoxicity).

Python Simulations

Metabolic Fuel Selection During Fasting

Python

Visualize the shift in fuel sources and hormone levels across the fed-fasted-starvation continuum.

script.py77 lines

import numpy as np
import matplotlib.pyplot as plt

# Metabolic fuel selection across fed/fasted/starvation states
fig, axes = plt.subplots(1, 2, figsize=(14, 6))

# Panel 1: Fuel utilization over time during fasting
ax1 = axes[0]
hours = np.array([0, 4, 8, 12, 24, 48, 72, 168, 336])  # hours of fasting
labels_h = ['0', '4h', '8h', '12h', '1d', '2d', '3d', '1wk', '2wk']

# Percentage of energy from each fuel source
glucose_pct = np.array([70, 55, 40, 30, 15, 8, 5, 3, 2])
fat_pct = np.array([25, 35, 45, 50, 60, 55, 50, 45, 40])
ketone_pct = np.array([0, 2, 5, 10, 15, 30, 40, 48, 55])
protein_pct = np.array([5, 8, 10, 10, 10, 7, 5, 4, 3])

ax1.stackplot(range(len(hours)), glucose_pct, fat_pct, ketone_pct, protein_pct,
              labels=['Glucose', 'Fatty Acids', 'Ketone Bodies', 'Amino Acids'],
              colors=['#34d399', '#fbbf24', '#f87171', '#a78bfa'], alpha=0.85)
ax1.set_xticks(range(len(hours)))
ax1.set_xticklabels(labels_h, fontsize=9, color='white')
ax1.set_xlabel('Duration of Fasting', fontsize=12, color='white')
ax1.set_ylabel('% Energy Contribution', fontsize=12, color='white')
ax1.set_title('Fuel Selection During Fasting', fontsize=14, color='white', fontweight='bold')
ax1.legend(loc='center right', fontsize=9, facecolor='#1a1a2e', edgecolor='#34d399', labelcolor='white')
ax1.set_facecolor('#0a0a1a')
ax1.tick_params(colors='white')
ax1.grid(True, alpha=0.15, color='#34d399')
for spine in ax1.spines.values():
    spine.set_color('#34d399')

# Panel 2: Hormone levels and key metabolites during fasting
ax2 = axes[1]
time_pts = [0, 4, 12, 24, 72, 168]
time_labels = ['Fed', '4h', '12h', '1d', '3d', '1wk']

insulin = [100, 60, 30, 15, 10, 8]
glucagon = [30, 50, 70, 85, 90, 92]
blood_glucose = [90, 85, 75, 65, 60, 55]
blood_ketones = [0.1, 0.2, 0.5, 1.5, 5.0, 7.0]

ax2_twin = ax2.twinx()
ax2.plot(range(len(time_pts)), insulin, 's-', color='#34d399', linewidth=2.5, markersize=7, label='Insulin (rel.)')
ax2.plot(range(len(time_pts)), glucagon, 'o-', color='#f87171', linewidth=2.5, markersize=7, label='Glucagon (rel.)')
ax2.plot(range(len(time_pts)), blood_glucose, '^-', color='#fbbf24', linewidth=2.5, markersize=7, label='Glucose (mg/dL)')
ax2_twin.plot(range(len(time_pts)), blood_ketones, 'D--', color='#a78bfa', linewidth=2.5, markersize=7, label='Ketones (mM)')

ax2.set_xticks(range(len(time_pts)))
ax2.set_xticklabels(time_labels, fontsize=10, color='white')
ax2.set_xlabel('Time', fontsize=12, color='white')
ax2.set_ylabel('Relative Level / mg/dL', fontsize=12, color='#34d399')
ax2_twin.set_ylabel('Blood Ketones (mM)', fontsize=12, color='#a78bfa')
ax2.set_title('Hormones & Metabolites During Fasting', fontsize=14, color='white', fontweight='bold')
ax2.set_facecolor('#0a0a1a')
ax2.tick_params(colors='#34d399')
ax2_twin.tick_params(colors='#a78bfa')
ax2.grid(True, alpha=0.15, color='#34d399')
for spine in ax2.spines.values():
    spine.set_color('#34d399')
for spine in ax2_twin.spines.values():
    spine.set_color('#34d399')

lines1 = ax2.get_legend_handles_labels()
lines2 = ax2_twin.get_legend_handles_labels()
ax2.legend(lines1[0] + lines2[0], lines1[1] + lines2[1],
           fontsize=8, facecolor='#1a1a2e', edgecolor='#34d399', labelcolor='white', loc='center right')

fig.patch.set_facecolor('#0a0a1a')
plt.tight_layout()
plt.savefig('output.png', dpi=150, bbox_inches='tight', facecolor='#0a0a1a')
plt.show()
print("Fed state: insulin high, glucose primary fuel, fat synthesis active")
print("Early fasting (12-24h): glucagon rises, glycogenolysis + gluconeogenesis + lipolysis")
print("Prolonged fasting (>3d): ketone bodies become brain's major fuel, sparing muscle protein")

Click Run to execute the Python code

Code will be executed with Python 3 on the server

Tissue-Specific Metabolism and Insulin Dose-Response

Python

Compare fuel preferences across tissues and explore the differential sensitivity of insulin signaling targets.

script.py79 lines

import numpy as np
import matplotlib.pyplot as plt

# Tissue-specific metabolism: metabolic profiles of major organs
fig, axes = plt.subplots(1, 2, figsize=(14, 6))

# Panel 1: Tissue fuel preferences (fed vs fasted)
ax1 = axes[0]
tissues = ['Brain', 'Muscle\n(rest)', 'Muscle\n(exercise)', 'Liver', 'Adipose', 'Kidney']
# Fuel preference scores (0-100) for each fuel type in FASTED state
glucose_use = [60, 10, 30, 5, 5, 10]
fa_use = [0, 70, 20, 10, 5, 60]
ketone_use = [35, 15, 5, 0, 0, 15]
aa_use = [0, 5, 10, 15, 0, 15]
lactate_use = [5, 0, 35, 70, 90, 0]

x = np.arange(len(tissues))
width = 0.15

ax1.bar(x - 2*width, glucose_use, width, label='Glucose', color='#34d399', alpha=0.85)
ax1.bar(x - width, fa_use, width, label='Fatty Acids', color='#fbbf24', alpha=0.85)
ax1.bar(x, ketone_use, width, label='Ketone Bodies', color='#f87171', alpha=0.85)
ax1.bar(x + width, aa_use, width, label='Amino Acids', color='#a78bfa', alpha=0.85)
ax1.bar(x + 2*width, lactate_use, width, label='Lactate/Other', color='#38bdf8', alpha=0.85)

ax1.set_xticks(x)
ax1.set_xticklabels(tissues, fontsize=9, color='white')
ax1.set_ylabel('Relative Fuel Utilization (%)', fontsize=12, color='white')
ax1.set_title('Tissue Fuel Preferences (Fasted State)', fontsize=14, color='white', fontweight='bold')
ax1.legend(fontsize=8, facecolor='#1a1a2e', edgecolor='#34d399', labelcolor='white', ncol=3, loc='upper right')
ax1.set_facecolor('#0a0a1a')
ax1.tick_params(colors='white')
ax1.grid(True, alpha=0.2, color='#34d399', axis='y')
for spine in ax1.spines.values():
    spine.set_color('#34d399')

# Panel 2: Insulin signaling cascade - response curves
ax2 = axes[1]
insulin_conc = np.linspace(0, 200, 500)  # pM

# Different downstream effects with different dose-response sensitivities
# Using Hill equation: response = Rmax * [I]^n / (EC50^n + [I]^n)
def hill_response(conc, ec50, n, rmax=100):
    return rmax * conc**n / (ec50**n + conc**n)

glucose_uptake = hill_response(insulin_conc, 50, 2.0)   # GLUT4 translocation
lipogenesis = hill_response(insulin_conc, 80, 1.5)       # ACC activation
glycogen_synth = hill_response(insulin_conc, 30, 2.5)   # GSK-3 inhibition
gluconeogenesis_inhib = hill_response(insulin_conc, 20, 3.0)  # FOXO inactivation

ax2.plot(insulin_conc, glucose_uptake, linewidth=2.5, color='#34d399', label='Glucose uptake (GLUT4)')
ax2.plot(insulin_conc, lipogenesis, linewidth=2.5, color='#fbbf24', label='Lipogenesis (ACC)')
ax2.plot(insulin_conc, glycogen_synth, linewidth=2.5, color='#38bdf8', label='Glycogen synthesis')
ax2.plot(insulin_conc, gluconeogenesis_inhib, linewidth=2.5, color='#f87171', label='GNG suppression (FOXO)')

ax2.set_xlabel('Insulin Concentration (pM)', fontsize=12, color='white')
ax2.set_ylabel('Response (% Maximum)', fontsize=12, color='white')
ax2.set_title('Insulin Dose-Response: Different Targets', fontsize=14, color='white', fontweight='bold')
ax2.legend(fontsize=9, facecolor='#1a1a2e', edgecolor='#34d399', labelcolor='white')
ax2.set_facecolor('#0a0a1a')
ax2.tick_params(colors='white')
ax2.grid(True, alpha=0.2, color='#34d399')
for spine in ax2.spines.values():
    spine.set_color('#34d399')
ax2.axvline(x=50, color='gray', linestyle=':', alpha=0.4)
ax2.text(55, 5, 'Fasting\ninsulin', fontsize=8, color='gray')
ax2.axvline(x=150, color='gray', linestyle=':', alpha=0.4)
ax2.text(155, 5, 'Fed\ninsulin', fontsize=8, color='gray')

fig.patch.set_facecolor('#0a0a1a')
plt.tight_layout()
plt.savefig('output.png', dpi=150, bbox_inches='tight', facecolor='#0a0a1a')
plt.show()
print("Key insight: different insulin targets have different dose-response sensitivities")
print("Gluconeogenesis suppression is most sensitive (EC50 ~20 pM)")
print("Lipogenesis is least sensitive (EC50 ~80 pM)")
print("This explains why insulin resistance first affects glucose disposal, while lipogenesis persists")

Click Run to execute the Python code

Code will be executed with Python 3 on the server

The Warburg Effect and FDG-PET Imaging

Python

Compare ATP production rates between glycolysis and oxidative phosphorylation, and visualize the FDG-PET principle in cancer diagnosis.

script.py67 lines

import numpy as np
import matplotlib.pyplot as plt

# Warburg effect and cancer metabolism
fig, axes = plt.subplots(1, 2, figsize=(14, 6))

# Panel 1: ATP production rate comparison
ax1 = axes[0]
pathways = ['Glycolysis\n(anaerobic)', 'Glycolysis\n(aerobic, cancer)', 'OxPhos\n(normal cell)']
atp_per_glucose = [2, 2, 32]
flux_rate = [100, 100, 1]  # relative glucose consumption rate
atp_production_rate = [a * f for a, f in zip(atp_per_glucose, flux_rate)]

x = np.arange(len(pathways))
width = 0.3
bars1 = ax1.bar(x - width/2, atp_per_glucose, width, label='ATP/Glucose', color='#34d399', alpha=0.85)
bars2 = ax1.bar(x + width/2, atp_production_rate, width, label='ATP Rate (relative)', color='#fbbf24', alpha=0.85)

ax1.set_xticks(x)
ax1.set_xticklabels(pathways, fontsize=9, color='white')
ax1.set_ylabel('ATP Units', fontsize=12, color='white')
ax1.set_title('Warburg Effect: Yield vs Rate', fontsize=14, color='white', fontweight='bold')
ax1.legend(fontsize=10, facecolor='#1a1a2e', edgecolor='#34d399', labelcolor='white')
ax1.set_facecolor('#0a0a1a')
ax1.tick_params(colors='white')
ax1.grid(True, alpha=0.2, color='#34d399', axis='y')
for spine in ax1.spines.values():
    spine.set_color('#34d399')

for bar, val in zip(bars1, atp_per_glucose):
    ax1.text(bar.get_x() + bar.get_width()/2, val + 0.5, str(val),
             ha='center', fontsize=10, color='white', fontweight='bold')
for bar, val in zip(bars2, atp_production_rate):
    ax1.text(bar.get_x() + bar.get_width()/2, val + 2, str(val),
             ha='center', fontsize=10, color='white', fontweight='bold')

# Panel 2: FDG-PET principle - glucose uptake in tumors
ax2 = axes[1]
tissue_types = ['Normal\nbrain', 'Normal\nmuscle', 'Normal\nliver', 'Low-grade\ntumor',
                'High-grade\ntumor', 'Metastatic\ncancer']
fdg_uptake = [8, 1, 3, 6, 15, 20]  # SUV (standardized uptake value)
colors_fdg = ['#34d399', '#34d399', '#34d399', '#fbbf24', '#f87171', '#dc2626']

bars3 = ax2.bar(range(len(tissue_types)), fdg_uptake, color=colors_fdg, alpha=0.85,
                edgecolor='white', linewidth=0.5, width=0.6)
ax2.axhline(y=2.5, color='#fbbf24', linestyle='--', alpha=0.6, label='SUV threshold for malignancy')
ax2.set_xticks(range(len(tissue_types)))
ax2.set_xticklabels(tissue_types, fontsize=9, color='white')
ax2.set_ylabel('FDG Uptake (SUV)', fontsize=12, color='white')
ax2.set_title('FDG-PET: Glucose Uptake by Tissue', fontsize=14, color='white', fontweight='bold')
ax2.legend(fontsize=10, facecolor='#1a1a2e', edgecolor='#34d399', labelcolor='white')
ax2.set_facecolor('#0a0a1a')
ax2.tick_params(colors='white')
ax2.grid(True, alpha=0.2, color='#34d399', axis='y')
for spine in ax2.spines.values():
    spine.set_color('#34d399')

fig.patch.set_facecolor('#0a0a1a')
plt.tight_layout()
plt.savefig('output.png', dpi=150, bbox_inches='tight', facecolor='#0a0a1a')
plt.show()
print("Warburg effect: cancer cells can produce ATP FASTER via glycolysis despite lower yield")
print("FDG-PET exploits elevated glucose uptake in tumors (GLUT1/GLUT3 overexpression + hexokinase II)")
print("FDG is phosphorylated to FDG-6-P by hexokinase but cannot be further metabolized -> trapped in cell")
print("SUV > 2.5 generally suggests malignancy; brain has high baseline FDG uptake (obligate glucose user)")

Click Run to execute the Python code

Code will be executed with Python 3 on the server

Inter-Organ Metabolic Cycles

Metabolism is not cell-autonomous — organs communicate metabolically through shared intermediates in the blood. Three major inter-organ cycles coordinate fuel metabolism:

The Cori Cycle (Muscle $\leftrightarrow$ Liver)

During intense exercise, muscle produces lactate from anaerobic glycolysis. Lactate travels to the liver, where it is reconverted to glucose via gluconeogenesis. The glucose returns to muscle, completing the cycle.

$$\text{Muscle: Glucose} \xrightarrow{+2\;\text{ATP}} 2\;\text{Lactate} \quad\xrightarrow{\text{blood}}\quad \text{Liver: 2 Lactate} \xrightarrow{-6\;\text{ATP}} \text{Glucose}$$

Net cost: 4 ATP per cycle, borne by the liver. This shifts the metabolic burden of anaerobic exercise from muscle to liver.

The Glucose-Alanine Cycle (Muscle $\leftrightarrow$ Liver)

Muscle transaminates pyruvate to alanine (carrying the amino group from protein catabolism), which travels to the liver. The liver regenerates pyruvate (for gluconeogenesis) and feeds the amino group into the urea cycle.

$$\text{Muscle: Pyruvate + Glutamate} \xrightarrow{\text{ALT}} \text{Alanine + }\alpha\text{-KG} \quad\xrightarrow{\text{blood}}\quad \text{Liver: reverse ALT + urea cycle}$$

Dual function: transfers both carbon (for gluconeogenesis) and nitrogen (for urea synthesis) from muscle to liver.

The Lipid Transport Cycle (Liver $\leftrightarrow$ Adipose $\leftrightarrow$ Muscle)

In the fed state, the liver packages triacylglycerols into VLDL particles for export. Lipoprotein lipase (LPL) on capillary endothelium in adipose tissue and muscle hydrolyzes TG, releasing fatty acids for storage or oxidation. During fasting, adipose hormone-sensitive lipase releases stored fatty acids (bound to albumin) for delivery to muscle and liver.

The "Randle Cycle" — Glucose-Fatty Acid Competition

Philip Randle (1963) discovered that fatty acid oxidation inhibits glucose utilization in muscle (and vice versa). The mechanism:

$$\text{FA oxidation}\uparrow \rightarrow \text{[Acetyl-CoA]}\uparrow \rightarrow \text{Citrate}\uparrow \rightarrow \text{PFK-1 inhibited} \rightarrow \text{Glycolysis}\downarrow$$

$$\text{FA oxidation}\uparrow \rightarrow \text{[Acetyl-CoA]}\uparrow \rightarrow \text{PDH inhibited (product inhibition)} \rightarrow \text{Glucose oxidation}\downarrow$$

This substrate competition is physiologically important: during fasting, elevated fatty acid oxidation spares glucose for the brain. However, chronic Randle cycle activation (obesity, elevated free fatty acids) contributes to muscle insulin resistance by impairing insulin-stimulated glucose uptake — a key mechanism linking obesity to type 2 diabetes.

Exercise Metabolism: From Sprint to Marathon

The metabolic response to exercise depends critically on intensity and duration:

Sprint (0-10 seconds): ATP and Creatine Phosphate

Immediate energy from pre-formed ATP (~2 seconds) and creatine phosphate (~8 seconds). The creatine kinase reaction:

$$\text{Phosphocreatine} + \text{ADP} \xrightleftharpoons{\text{Creatine Kinase}} \text{Creatine} + \text{ATP}$$

High Intensity (10 sec - 2 min): Anaerobic Glycolysis

Muscle glycogen is rapidly converted to lactate (2 ATP per glucose). Lactate accumulation causes acidosis, contributing to fatigue. Lactate threshold (~60-80% VO$_2$ max) defines the transition from aerobic to anaerobic metabolism.

Moderate Intensity (2 min - 2 hours): Aerobic Oxidation

Mix of muscle glycogen and fatty acids. The crossover concept: at ~40-65% VO$_2$ max, fat oxidation peaks. Above this intensity, carbohydrate oxidation predominates. AMPK activation in muscle promotes GLUT4 translocation and fatty acid oxidation independent of insulin.

Endurance (>2 hours): "Hitting the Wall"

When muscle glycogen is depleted (~2 hours at marathon pace), the athlete "hits the wall" (bonking). Performance drops dramatically because fatty acid oxidation cannot sustain the high ATP production rate needed for intense exercise. Carbohydrate loading (supercompensation) before events and glucose/fructose ingestion during exercise extend glycogen availability.

The maximum power output from each fuel system differs dramatically: creatine phosphate supports ~36 kcal/min (highest power, shortest duration), anaerobic glycolysis ~16 kcal/min, aerobic carbohydrate oxidation ~10 kcal/min, and fatty acid oxidation ~6 kcal/min (lowest power, longest duration). This hierarchy explains why sprinters are muscular (high PCr/glycolytic capacity) while marathon runners are lean (optimized fat oxidation).

Key Takeaways

The insulin/glucagon ratio is the master switch between anabolic (fed) and catabolic (fasted) metabolism.
Insulin signals via IRS $\rightarrow$ PI3K $\rightarrow$ Akt; glucagon signals via cAMP $\rightarrow$ PKA. Both cascades amplify the hormonal signal ~10,000-fold.
During starvation, the brain's adaptation to ketone bodies reduces muscle protein catabolism (protein-sparing), extending survival.
Each tissue has a unique metabolic profile: liver exports glucose, brain demands glucose/ketones, muscle uses fatty acids at rest.
Selective insulin resistance in type 2 diabetes: hepatic gluconeogenesis escapes suppression while lipogenesis continues — the metabolic paradox.
The Warburg effect in cancer cells supports biosynthetic precursor supply, not just ATP production; exploited clinically via FDG-PET imaging.
AMPK is the central energy sensor: activated by AMP (amplified ~30-fold by adenylate kinase), it switches on catabolism and off anabolism.
The Randle cycle (glucose-fatty acid cycle) explains substrate competition and contributes to insulin resistance in obesity.
Inter-organ metabolic cycles (Cori, glucose-alanine, lipid transport) coordinate whole-body fuel homeostasis.
Exercise fuel hierarchy: creatine phosphate (0-10 s) $\rightarrow$ anaerobic glycolysis (10 s-2 min) $\rightarrow$ aerobic oxidation (2 min+) $\rightarrow$ fat oxidation (endurance).

← Nucleotide Metabolism DNA Replication →