Part 2 · Chapter 2.5

Metabolic Integration

Glycolysis, the Krebs cycle, and oxidative phosphorylation do not operate in isolation. They are embedded in a web of interconnected pathways—pentose phosphate, fatty acid oxidation, ketogenesis, amino acid catabolism—coordinated across organs by insulin, glucagon, AMPK, and epinephrine to maintain homeostasis from the fed to the starving state.

Learning Objectives

▸Compute the full ATP yield of complete glucose oxidation (30–32 ATP)
▸Distinguish the Cori cycle, glucose–alanine cycle, and ketogenesis
▸Describe the pentose phosphate pathway’s role in NADPH supply and nucleotide precursors
▸Explain the hormonal logic of insulin, glucagon, and epinephrine
▸Identify AMPK as the cellular energy sensor and its regulatory targets

◆Total ATP Yield per Glucose

The complete oxidation of one glucose molecule to CO₂ and H₂O yields approximately 30–32 ATP, with the exact number depending on the shuttle that exports cytosolic NADH into mitochondria:

\[ \text{C}_6\text{H}_{12}\text{O}_6 + 6\,\text{O}_2 \longrightarrow 6\,\text{CO}_2 + 6\,\text{H}_2\text{O},\quad \Delta G^{\circ\prime} = -2870\;\text{kJ/mol} \]

Glycolysis

+2 ATP (net, substrate-level)
+2 NADH (cytosolic → shuttle)

Pyruvate DH + TCA

+8 NADH (2 PDH + 6 TCA)
+2 FADH₂ (TCA)
+2 GTP

OxPhos Output

10 NADH × 2.5 = 25 ATP
2 FADH₂ × 1.5 = 3 ATP
+2 ATP + 2 GTP = 4 ATP direct

The malate-aspartate shuttle (liver, kidney, heart) delivers cytosolic NADH as matrix NADH (yield 2.5 ATP each); the glycerol-3-phosphate shuttle (muscle, brain) converts NADH to FADH₂ in the matrix (yield 1.5 ATP each). Hence the 30–32 range.

◆Simulation 1: Full ATP Budget Visualization

This code builds a stacked-bar accounting of ATP from each pathway and compares different metabolic modes (aerobic with either shuttle, anaerobic, Warburg cancer cell).

Python

script.py94 lines

import numpy as np
import matplotlib.pyplot as plt

# Total ATP yield per glucose: pathway-by-pathway accounting
# Using P/O = 2.5 for NADH, 1.5 for FADH2

stages = ['Glycolysis', 'Pyruvate DH', 'TCA cycle', 'Total ETC']
cat = ['ATP (substrate)', 'NADH -> ATP', 'FADH2 -> ATP']

# Per glucose (2 pyruvate):
# Glycolysis:      +2 ATP (net), +2 NADH (cytosolic -> 1.5-2.5 depending on shuttle)
# PDH:             +0 ATP, +2 NADH
# TCA (2 turns):   +2 GTP, +6 NADH, +2 FADH2

glycolysis_direct = 2
glycolysis_nadh   = 2 * 2.5   # assume malate-aspartate shuttle
pdh_nadh          = 2 * 2.5
tca_direct        = 2
tca_nadh          = 6 * 2.5
tca_fadh          = 2 * 1.5

# Assemble totals
matrix = np.array([
    [glycolysis_direct, glycolysis_nadh, 0],
    [0,                 pdh_nadh,        0],
    [tca_direct,        tca_nadh,        tca_fadh],
])

totals = matrix.sum(axis=0)
grand_total = matrix.sum()

fig, axes = plt.subplots(1, 2, figsize=(14, 5.5))

# Stacked bar by stage
ax = axes[0]
bot = np.zeros(len(stages)-1)
colors_c = ['#fbbf24','#f87171','#fdba74']
for i, (label, row) in enumerate(zip(cat, matrix.T)):
    ax.bar(stages[:3], matrix[:, i], bottom=bot, color=colors_c[i], label=label, edgecolor='white')
    bot += matrix[:, i]
# Annotate totals
for i, s in enumerate(stages[:3]):
    total = matrix[i,:].sum()
    ax.text(i, total+0.5, f'{total:.1f}', ha='center', color='white', fontweight='bold', fontsize=12)

ax.set_ylabel('ATP equivalents per glucose', color='white')
ax.set_title(f'ATP yield breakdown: TOTAL = {grand_total:.1f} ATP / glucose',
             color='white', fontsize=13, fontweight='bold')
ax.legend(facecolor='#0a0a1a', edgecolor='#fbbf24', labelcolor='white')
ax.set_facecolor('#0a0a1a'); ax.tick_params(colors='white')
ax.grid(alpha=0.2, color='#fbbf24', axis='y')
for s in ax.spines.values(): s.set_color('#fbbf24')

# Comparison: aerobic vs anaerobic
ax = axes[1]
conditions = ['Aerobic\n(glycerol-P shuttle)',
              'Aerobic\n(malate-Asp shuttle)',
              'Anaerobic\n(lactate)',
              'Warburg\ncancer cell']
aerobic_glyc_ps = 2 + 2*1.5 + 2*2.5 + 6*2.5 + 2*1.5 + 2   # ~30
aerobic_mas     = 2 + 2*2.5 + 2*2.5 + 6*2.5 + 2*1.5 + 2   # ~32
anaerobic       = 2
warburg         = 2 + 2*2.5 + 0.5*pdh_nadh  # partial OxPhos
values = [aerobic_glyc_ps, aerobic_mas, anaerobic, warburg]
colors_b = ['#67e8f9','#86efac','#f87171','#fbbf24']
bars = ax.bar(conditions, values, color=colors_b, edgecolor='white')
for b, v in zip(bars, values):
    ax.text(b.get_x() + b.get_width()/2, v+0.5, f'{v:.1f}',
            ha='center', color='white', fontweight='bold', fontsize=12)
ax.set_ylabel('ATP per glucose', color='white')
ax.set_title('ATP yield under different metabolic modes',
             color='white', fontsize=13, fontweight='bold')
ax.set_facecolor('#0a0a1a'); ax.tick_params(colors='white')
ax.grid(alpha=0.2, color='#67e8f9', axis='y')
for s in ax.spines.values(): s.set_color('#67e8f9')

fig.patch.set_facecolor('#0a0a1a')
plt.tight_layout()
plt.savefig('output.png', dpi=120, bbox_inches='tight', facecolor='#0a0a1a')

# Energy efficiency
dG_glu_ox = 2870  # kJ/mol (complete combustion of glucose)
dG_atp_cell = 50  # kJ/mol
efficiency = grand_total * dG_atp_cell / dG_glu_ox * 100

print(f"Total ATP (aerobic, MAS shuttle):       {grand_total:.1f}")
print(f"Complete glucose oxidation \u0394G:       {dG_glu_ox} kJ/mol")
print(f"Energy captured as ATP (cellular):     {grand_total*dG_atp_cell:.0f} kJ/mol")
print(f"Overall thermodynamic efficiency:      {efficiency:.1f}%")
print()
print("Anaerobic (lactate only) = 2 ATP/glucose -> 15x less efficient than OxPhos")
print("But anaerobic can be 100x faster per unit enzyme -> Warburg effect")

Click Run to execute the Python code

Code will be executed with Python 3 on the server

◆Inter-Organ Metabolic Cycles

The Cori Cycle (Lactate Shuttle)

During sustained exercise, skeletal muscle produces lactate from glucose (glycolysis outstrips OxPhos). Lactate diffuses to the liver, where gluconeogenesis regenerates glucose that is released back to blood. Net effect: muscle shifts its ATP debt to the liver (which burns ATP for gluconeogenesis).

\[ \text{Muscle: glucose} \to 2\,\text{lactate}\;(+2\,\text{ATP}) \]\[ \text{Liver: 2 lactate} \to \text{glucose}\;(-6\,\text{ATP}) \]

Glucose–Alanine Cycle

Similar logic, but with the addition of nitrogen transport. Muscle transaminates pyruvate to alanine (disposing of NH₃ from amino acid catabolism); alanine travels to liver, where it is converted back to pyruvate for gluconeogenesis while the amino group enters the urea cycle.

\[ \text{Pyruvate} + \text{Glu} \;\rightleftharpoons\; \text{Alanine} + \alpha\text{-KG} \]

Ketogenesis (Fasting / Diabetes)

When carbohydrate is scarce, liver oxidizes fatty acids to acetyl-CoA faster than the TCA cycle can consume it (OAA is depleted for gluconeogenesis). Excess acetyl-CoA condenses to form acetoacetate and \(\beta\)-hydroxybutyrate—the ketone bodies—which are exported to peripheral tissues (brain, heart, muscle) as an alternative fuel.

Pentose Phosphate Pathway

Branches from G6P in the cytosol. Oxidative branch produces 2 NADPH (for reductive biosynthesis and glutathione) and ribose-5-phosphate (for nucleotides). Non-oxidative branch interconverts C3, C4, C5, C6, C7 sugars through transketolase and transaldolase. No ATP produced or consumed.

Glyoxylate Cycle (Plants, Microbes, Not Animals)

A modified TCA found in plants, fungi, bacteria, and nematodes. Isocitrate lyase cleaves isocitrate to succinate + glyoxylate, bypassing the two decarboxylations. This allows net synthesis of carbohydrates from acetyl-CoA (fatty acids)—impossible in mammals. Enables germinating seeds to convert seed-oil triglycerides into sucrose.

◆Hormonal Regulation

◆AMPK: The Cellular Energy Sensor

AMP-activated protein kinase (AMPK) is a heterotrimeric \(\alpha\beta\gamma\) Ser/Thr kinase that senses the AMP:ATP ratio. When ATP falls (e.g., exercise, ischemia, glucose deprivation), AMP binds the \(\gamma\)-subunit, promoting phosphorylation of AMPK at T172 by LKB1 (tumor suppressor) or CaMKK\(\beta\). Activated AMPK broadly shifts metabolism from anabolic to catabolic:

Activated (ATP-generating)

↑ Fatty acid oxidation (phosphorylates ACC)
↑ Glucose uptake (GLUT4 translocation)
↑ Autophagy (ULK1 phosphorylation)
↑ Mitochondrial biogenesis (PGC-1\(\alpha\))

Suppressed (ATP-consuming)

↓ Fatty acid synthesis (ACC, SREBP-1c)
↓ Cholesterol synthesis (HMG-CoA reductase)
↓ Protein synthesis (mTORC1, TSC2)
↓ Gluconeogenesis in liver

Metformin, the first-line drug for type 2 diabetes, works partly by weakly inhibiting Complex I, raising the AMP:ATP ratio, and activating AMPK. This suppresses hepatic gluconeogenesis and sensitizes tissues to insulin. The same mechanism is being studied for its anti-cancer and anti-aging effects.

◆Simulation 2: Fed vs Fasted Whole-Body Model

A six-compartment model (blood glucose, liver glycogen, muscle glycogen, adipose TG, ketones, FFA) shows the dramatic divergence of metabolic state between fed (high insulin→storage) and fasted (high glucagon→mobilization) conditions over 24 hours.

Python

script.py97 lines

import numpy as np
import matplotlib.pyplot as plt
from scipy.integrate import odeint

# Compartmental model comparing fed vs. fasted whole-body metabolism
# Compartments: blood_glucose, liver_glycogen, muscle_glycogen,
#               adipose_TG, blood_ketones, blood_FFA
#
# Fluxes (arbitrary units representing mmol/h):
#   absorption (fed only), glycolysis, gluconeogenesis, glycogenolysis,
#   glycogen synthesis, lipolysis, ketogenesis
#
# Hormonal control: insulin (fed) vs. glucagon + epinephrine (fasted)

def metab_model(y, t, state):
    glu_blood, liv_gly, mus_gly, adp_tg, ket_blood, ffa_blood = y

if state == 'fed':
        insulin, glucagon = 1.0, 0.1
        absorption = 8.0 if t < 4 else 0.0
    else:  # fasted
        insulin, glucagon = 0.1, 1.0
        absorption = 0.0

hormonal = glucagon/(insulin+0.05)

# Fluxes
    f_uptake_muscle  = 3.0*insulin * glu_blood/(glu_blood + 5.0)      # GLUT4
    f_uptake_liver   = 2.0 * glu_blood/(glu_blood + 10.0)             # GLUT2
    f_glycolysis_liv = 2.5*insulin * glu_blood/(glu_blood + 5.0)
    f_glycogensyn_liv = 2.0*insulin * glu_blood/(glu_blood + 5.0)
    f_glycogenolysis_liv = 3.0*hormonal * liv_gly/(liv_gly + 50.0)
    f_gluconeogenesis = 2.5*hormonal
    f_glycogensyn_mus = 2.0*insulin * glu_blood/(glu_blood + 5.0)
    f_glycolysis_mus  = 1.5*insulin * mus_gly/(mus_gly + 50.0)
    f_lipolysis       = 1.5*hormonal * adp_tg/(adp_tg + 500.0)
    f_ketogenesis     = 0.8*hormonal * ffa_blood/(ffa_blood + 0.5)
    f_ketone_use      = 2.0 * ket_blood/(ket_blood + 0.2)
    f_peripheral_glu  = 0.5 * glu_blood/(glu_blood + 4.0)
    f_lipogenesis     = 1.5*insulin * glu_blood/(glu_blood + 5.0)

dglu   = absorption - f_uptake_muscle - f_uptake_liver              + f_glycogenolysis_liv + f_gluconeogenesis - f_peripheral_glu
    dliv   = f_glycogensyn_liv - f_glycogenolysis_liv
    dmus   = f_glycogensyn_mus - f_glycolysis_mus
    dtg    = -f_lipolysis + f_lipogenesis
    dket   = f_ketogenesis - f_ketone_use
    dffa   = f_lipolysis - f_ketogenesis

return [dglu, dliv, dmus, dtg, dket, dffa]

# Initial conditions: typical post-meal, units roughly mmol per compartment
y0_fed = [5.0, 300, 400, 1500, 0.1, 0.5]

# Simulate 24 h
t = np.linspace(0, 24, 600)
sol_fed = odeint(metab_model, y0_fed, t, args=('fed',))
sol_fst = odeint(metab_model, y0_fed, t, args=('fasted',))

fig, axes = plt.subplots(2, 3, figsize=(15, 9))
titles = ['Blood glucose (mM)', 'Liver glycogen', 'Muscle glycogen',
          'Adipose TG', 'Blood ketones', 'Blood FFA']
colors_a = ['#67e8f9','#fbbf24','#86efac','#c084fc','#f87171','#fdba74']

for i, ax in enumerate(axes.flatten()):
    ax.plot(t, sol_fed[:,i], color='#67e8f9', lw=2.5, label='Fed state')
    ax.plot(t, sol_fst[:,i], color='#f87171', lw=2.5, label='Fasted state')
    ax.set_xlabel('Time (h)', color='white')
    ax.set_title(titles[i], color='white', fontweight='bold')
    ax.legend(facecolor='#0a0a1a', edgecolor=colors_a[i], labelcolor='white', fontsize=9)
    ax.set_facecolor('#0a0a1a'); ax.tick_params(colors='white')
    ax.grid(alpha=0.2, color=colors_a[i])
    for s in ax.spines.values(): s.set_color(colors_a[i])

fig.suptitle('Whole-body metabolism: fed vs. fasted 24-hour simulation',
             color='white', fontsize=15, fontweight='bold')
fig.patch.set_facecolor('#0a0a1a')
plt.tight_layout()
plt.savefig('output.png', dpi=120, bbox_inches='tight', facecolor='#0a0a1a')

# Integrated quantities using trapezoid
auc_glu_fed = np.trapezoid(sol_fed[:,0], t)
auc_glu_fst = np.trapezoid(sol_fst[:,0], t)
delta_gly_fed = sol_fed[-1,1] - y0_fed[1]
delta_gly_fst = sol_fst[-1,1] - y0_fed[1]

print("=== 24-h metabolic integrals ===")
print(f"AUC blood glucose, fed:    {auc_glu_fed:.1f} mM*h")
print(f"AUC blood glucose, fasted: {auc_glu_fst:.1f} mM*h")
print(f"Liver glycogen change, fed:    {delta_gly_fed:+.1f}")
print(f"Liver glycogen change, fasted: {delta_gly_fst:+.1f}")
print(f"Final ketone level, fed:    {sol_fed[-1,4]:.3f}  vs. fasted: {sol_fst[-1,4]:.3f}")
print()
print("Hormonal drivers:")
print("  Fed:    insulin high -> glycogen synthesis, glycolysis, lipogenesis")
print("  Fasted: glucagon+epi -> glycogenolysis, gluconeogenesis, lipolysis, ketogenesis")

Click Run to execute the Python code

Code will be executed with Python 3 on the server

◆The Warburg Effect Reconsidered

Otto Warburg observed in the 1920s that tumor cells favor glycolysis even in the presence of oxygen. For decades this was viewed as a bizarre metabolic quirk. Modern understanding:

Why the Warburg effect?

Rapid ATP generation (per unit time) even at low yield per glucose
Biosynthetic precursors: ribose (PPP), serine, glycine, amino acids from pyruvate/lactate
NADPH via PPP: reductive biosynthesis + antioxidant defense
Acidification of tumor microenvironment: promotes invasion, immune escape
HIF-1\(\alpha\) stabilization under pseudo-hypoxia

Clinical implications

FDG-PET imaging leverages elevated hexokinase activity in tumors
Drug targets: PKM2, LDH-A, GLUT1, hexokinase II
Ketogenic diet as adjunct therapy in some cancers (glucose deprivation)
Lactate is now recognized as an inter-tissue fuel and signaling molecule, not just waste

Summary: The Metabolic Symphony

Cellular energetics is not a set of isolated biochemical pathways but a tightly coordinated symphony. Glycolysis provides rapid ATP and carbon skeletons; the TCA cycle is the universal oxidative hub; the electron transport chain builds the proton-motive force; ATP synthase converts it into the universal currency. At the whole-body level, liver, muscle, adipose, brain, and kidney specialize metabolically and communicate via lactate, glucose, fatty acids, ketones, and amino acids, orchestrated by insulin, glucagon, epinephrine, and the master cellular sensor AMPK.

Understanding these integrated pathways is fundamental not only to cell biology but to diabetes, cancer, cardiovascular disease, neurodegeneration, and the biology of aging—all of which can be viewed as disorders of metabolic homeostasis.

◆Phases of Fasting: A Metabolic Timeline

As the duration of fasting increases, different fuel sources are mobilized in sequence. The brain, glucose-dependent under normal conditions, progressively adapts to using ketone bodies over days.

Hours 0–4 (Post-absorptive)

Insulin falls, glucagon rises. Liver glycogenolysis provides blood glucose. Peripheral glucose use shifts toward FFAs.

Hours 4–16 (Early fasting)

Liver glycogen depletes (~100 g total). Gluconeogenesis begins: lactate, alanine, glycerol are substrates. Lipolysis supplies FFAs to muscle.

Days 1–3 (Early starvation)

Gluconeogenesis peaks; skeletal muscle is catabolized to provide amino acid carbons (especially alanine). Ketone bodies rise.

Days 3+ (Prolonged starvation)

Brain shifts to ~65% ketones for ATP. Muscle proteolysis slows; protein conservation. Survival extends to weeks.

◆Fuel Preferences of Key Tissues

Tissue	Preferred Fuel	Alternatives	Special Features
Brain	Glucose (120 g/d)	Ketones (starvation)	No FFA oxidation; no fuel storage
Liver	FFAs, amino acids	Glucose (post-meal)	Makes glucose + ketones for others
Resting muscle	FFAs	Glucose, ketones	Stores glycogen for own use
Active muscle	Glucose, glycogen	FFAs at low intensity	Generates lactate (Cori cycle)
Heart	FFAs (70%)	Lactate, glucose, ketones	Extreme OxPhos capacity
Erythrocytes	Glucose (only)	None	No mitochondria; strict glycolysis + lactate
Adipose	Glucose (fed)	FFAs (fasted)	Stores triglycerides; secretes leptin
Kidney	FFAs, glutamine	Glucose	Minor gluconeogenesis; acid-base

Clinical Relevance

Type 2 Diabetes

Insulin resistance in muscle/adipose + \(\beta\)-cell failure → hyperglycemia. Metformin activates AMPK; GLP-1 agonists, SGLT2 inhibitors also modulate glucose homeostasis.

Diabetic Ketoacidosis

Absolute insulin deficiency → unrestrained lipolysis → overwhelming ketogenesis; blood pH drops, anion gap rises. Medical emergency.

G6PD Deficiency

X-linked. Impaired PPP → NADPH ↓ → reduced glutathione cannot regenerate → oxidative hemolysis on exposure to oxidants (fava beans, primaquine, sulfa drugs).

MCAD Deficiency

Most common fatty acid oxidation defect. Fasting → hypoketotic hypoglycemia because ketogenesis is blocked; rapid seizures and death unless carbohydrates given.

von Gierke Disease

Glucose-6-phosphatase deficiency (GSD I). Liver cannot release free glucose → severe fasting hypoglycemia, hepatomegaly, lactic acidosis, hyperuricemia.

McArdle Disease

Muscle phosphorylase deficiency (GSD V). Exercise intolerance, cramps, second-wind phenomenon as the body shifts to fatty acid oxidation.

←Prev: 2.4 ATP Synthase Next: Part 3 Cell Signaling→