Module 4

Olfaction & Chemical Communication

Queen mandibular pheromone, alarm signaling, nestmate recognition, and odorant receptor biophysics

Chemical communication is the primary language of the honeybee colony. At least 15 distinct pheromone glands produce dozens of compounds that regulate every aspect of colony life — from queen dominance and worker reproduction to alarm responses, foraging coordination, and nestmate recognition. The bee's olfactory system, with 170 odorant receptors on ~60,000 antennal sensilla, decodes this chemical lexicon with remarkable sensitivity and specificity.

4.1 Queen Mandibular Pheromone (QMP)

The queen mandibular pheromone is the most important chemical signal in the colony, serving as the primary indicator of queen presence and health. QMP is a 5-component blend produced by the queen's mandibular glands:

Primary Components

9-ODA — (E)-9-oxo-2-decenoic acid
9-HDA — (E)-9-hydroxy-2-decenoic acid (both R and S enantiomers)
HVA — methyl p-hydroxybenzoate (homovanillyl alcohol)
HOB — 4-hydroxy-3-methoxyphenylethanol
Methyl oleate — methyl (Z)-octadec-9-enoate

Functions

Suppresses worker ovary development
Attracts drones during mating flights
Stabilizes swarm cluster cohesion
Inhibits queen cell construction (new queen rearing)
Modulates worker behavioral development

Dose-Response: Hill Equation

The biological response to QMP follows a sigmoidal dose-response relationship, well described by the Hill equation. This arises because QMP binding to odorant receptors exhibits cooperative behavior:

\[ E = \frac{E_{\max} \cdot [\text{QMP}]^n}{EC_{50}^n + [\text{QMP}]^n} \]

where:

\(E\) = observed biological effect (e.g., fraction of workers with suppressed ovaries)

\(E_{\max}\) = maximum possible effect (typically normalized to 1.0)

\([\text{QMP}]\) = concentration of QMP at receptor site

\(EC_{50}\) = concentration producing 50% of maximum effect

\(n\) = Hill coefficient (cooperativity parameter; \(n > 1\) indicates positive cooperativity)

For QMP's ovary suppression effect, experimental data suggest \(n \approx 2\text{--}3\), indicating significant cooperativity. This creates a sharp threshold effect: below a critical QMP concentration, workers rapidly begin ovary development (the "queenless response"), while above it, ovaries remain completely suppressed.

QMP Distribution Mechanism

QMP is not volatile at colony-relevant distances. Instead, it spreads by direct contact. A "retinue" of 8–12 workers surrounds the queen, licking and antennating her body surface to collect QMP. These retinue bees then distribute QMP to other workers through trophallaxis (mouth-to-mouth food exchange) and body contact.

The distribution follows a diffusion-like process through the social network:

\[ \frac{\partial C_{\text{QMP}}}{\partial t} = D_{\text{social}} \nabla^2 C_{\text{QMP}} - \lambda C_{\text{QMP}} + S(\mathbf{r}_{\text{queen}}) \]

where \(D_{\text{social}}\) is an effective diffusion coefficient through the social network, \(\lambda\) is the degradation rate, and \(S\) is the source term at the queen's location.

The QMP "signal" decays with effective distance from the queen. In a colony of 40,000+ workers, bees on the far side of the comb may receive insufficient QMP, which is one trigger for supersedure (replacement queen rearing). The critical colony size for QMP failure is approximately 30,000–50,000 bees, which correlates with the threshold for swarming behavior.

4.2 Alarm Pheromone

The primary alarm pheromone component is isopentyl acetate (IPA), also known as isoamyl acetate, released from the sting gland (Koschevnikov gland) when a bee stings. IPA has a characteristic banana-like odor detectable by humans at ~1 ppm but triggers defensive behavior in bees at nanogram levels.

Functions

Target Marking

When a bee stings, IPA released at the sting site marks the target for other defending bees. This explains why subsequent stings cluster near the first sting location.

Recruitment

Guard bees at the entrance release IPA to recruit additional defenders from inside the hive. The response is dose-dependent: low levels increase vigilance, high levels trigger mass defensive sorties.

Secondary Alarm: 2-Heptanone

The mandibular gland produces 2-heptanone, which serves as both a mild alarm pheromone and a "depleted flower" marker. Foragers deposit 2-heptanone on flowers they have already visited, deterring other foragers from wasting time.

Diffusion from Point Source

When alarm pheromone is released at a sting site, it disperses through air via molecular diffusion and convection. For an instantaneous point-source release of mass \(M\) in still air, the concentration field is described by the 3D diffusion equation:

\[ \frac{\partial C}{\partial t} = D \nabla^2 C \]

The Green's function solution (fundamental solution) for a point source at the origin at \(t = 0\) is:

\[ C(r, t) = \frac{M}{(4\pi D t)^{3/2}} \exp\!\left(-\frac{r^2}{4Dt}\right) \]

Derivation sketch: We seek a spherically symmetric solution \(C(r, t)\) to:

\[ \frac{\partial C}{\partial t} = D \left(\frac{\partial^2 C}{\partial r^2} + \frac{2}{r}\frac{\partial C}{\partial r}\right) \]

Using the substitution \(u = rC\), this reduces to the 1D diffusion equation for \(u\). The solution with \(\int_0^\infty C \cdot 4\pi r^2 \, dr = M\)(total mass conservation) gives the Gaussian profile above.

Key physical implications for alarm signaling:

Signal range: The threshold concentration for bee response is approximately \(C_{\text{thresh}} \sim 1\) ng/L. For IPA (\(D \approx 6 \times 10^{-6}\) m\(^2\)/s), the effective signaling radius reaches ~0.5 m within 10 seconds.

Temporal decay: The peak concentration at any fixed distance decays as \(t^{-3/2}\), ensuring the alarm signal is transient. This prevents false alarms from persisting.

Wind effects: In outdoor conditions, wind advection dominates over diffusion at distances beyond ~10 cm. The effective model becomes \(C(r, t) \propto \exp(-(r - vt)^2/4Dt)\) for a downwind plume.

4.3 Nasonov Pheromone & Nestmate Recognition

Nasonov Gland

The Nasonov gland, located on the dorsal surface of the 7th abdominal tergite, produces a blend of terpenoid compounds used for orientation and recruitment:

Geraniol

C\(_{10}\)H\(_{18}\)O

Predominant component

Citral

C\(_{10}\)H\(_{16}\)O

Mixture of geranial + neral

Geranic acid

C\(_{10}\)H\(_{16}\)O\(_2\)

Synergistic with geraniol

Bees expose the Nasonov gland by raising the abdomen and fanning wings to disperse the volatile signal. This "scenting" behavior is used to mark the hive entrance for returning foragers, to guide swarm bees to the new nest site, and to mark reliable water sources.

Cuticular Hydrocarbon (CHC) Profile

Colony identity is encoded in the cuticular hydrocarbon profile — a unique blend of long-chain hydrocarbons (C\(_{25}\)–C\(_{33}\)) on each bee's exoskeleton. These compounds are synthesized in oenocytes and transported to the cuticle. Within a colony, CHC profiles are homogenized through trophallaxis and body contact, creating a "colony odor" that is distinct from other colonies.

Guard bees stationed at the hive entrance make nestmate/non-nestmate discrimination decisions within approximately 0.5 seconds of antennal contact. This remarkably fast and accurate discrimination can be analyzed using signal detection theory.

Signal Detection Theory for Nestmate Recognition

Each incoming bee presents a CHC profile that the guard compares against a learned template of the colony odor. Due to natural variation, both nestmate and foreign bee CHC profiles form overlapping distributions:

\[ d' = \frac{\mu_{\text{nest}} - \mu_{\text{foreign}}}{\sigma} \]

where:

\(d'\) = discriminability index (sensitivity)

\(\mu_{\text{nest}}\) = mean similarity score for nestmate CHC profiles

\(\mu_{\text{foreign}}\) = mean similarity score for non-nestmate CHC profiles

\(\sigma\) = standard deviation (assumed equal for both distributions)

The guard sets a decision criterion \(\beta\) (threshold): if the perceived similarity exceeds \(\beta\), accept as nestmate; otherwise, reject. The optimal \(\beta\) depends on the relative costs of errors:

\[ \beta_{\text{opt}} = \frac{C_{\text{FA}}}{C_{\text{miss}}} \cdot \frac{P(\text{foreign})}{P(\text{nest})} \]

where \(C_{\text{FA}}\) = cost of false alarm (rejecting a nestmate), \(C_{\text{miss}}\) = cost of miss (admitting a robber bee).

In practice, guards set a liberal criterion during nectar flow (when robbing is rare and the cost of rejecting nestmates is high) and a strict criterion during dearth (when robbing is common). Experimental measurements give\(d' \approx 2\text{--}4\), indicating good but imperfect discrimination.

4.4 Odorant Receptor Machinery

The honeybee antenna houses approximately 60,000 olfactory sensilla, each containing 5–35 olfactory receptor neurons (ORNs). The Apis mellifera genome encodes 170 odorant receptors (ORs) — significantly more than Drosophila melanogaster(62 ORs), reflecting the honeybee's greater reliance on chemical communication.

Insect OR Structure: OR + Orco Complex

Insect odorant receptors have a fundamentally different architecture from vertebrate ORs:

Vertebrate ORs

G-protein coupled receptors (GPCRs). 7-transmembrane domains with extracellular N-terminus. Signal via G-protein cascade (G\(_{\text{olf}}\)\(\rightarrow\) adenylyl cyclase \(\rightarrow\) cAMP\(\rightarrow\) CNG channels). Slow (~100 ms latency).

Insect ORs

Ligand-gated ion channels. 7-transmembrane domains with intracellularN-terminus (inverted topology). OR + Orco (odorant receptor co-receptor) form heteromeric channel. Fast (~10 ms latency). Direct ionotropic signaling.

The Orco co-receptor is highly conserved across all insects and is required for proper OR trafficking to the dendritic membrane. Each ORN typically expresses one specific OR gene plus Orco. When the cognate odorant binds, the OR-Orco complex opens, allowing cation influx (mainly Ca\(^{2+}\) and Na\(^+\)) that depolarizes the neuron.

Combinatorial Coding Capacity

If each of the 170 ORs can be either activated or not by a given odorant, the theoretical coding capacity is:

\[ N_{\text{odors}}^{\text{theoretical}} = 2^{170} \approx 1.5 \times 10^{51} \]

This astronomical number far exceeds the number of distinct molecules in nature. In practice, the effective coding capacity is much smaller due to several constraints:

Cross-reactivity: Each OR responds to multiple structurally similar odorants, and each odorant activates multiple ORs. The typical "tuning breadth" is 5–20 odorants per OR.

Concentration dependence: The pattern of activated ORs changes with concentration, complicating identity coding.

Neural noise: Spontaneous firing rates (~1–5 Hz) and response variability limit the effective number of distinguishable activation patterns.

A more realistic estimate considers mutual information:

\[ I = \sum_{i=1}^{170} H(R_i) - H(R_i | S) \]

where \(H(R_i)\) is the entropy of receptor \(i\)'s response and \(H(R_i|S)\) is the noise entropy.

With realistic noise levels and cross-reactivity, the practical number of discriminable odors is estimated at approximately ~1,000 distinct odor identities, which is consistent with behavioral conditioning experiments showing bees can learn to discriminate at least several hundred distinct odorants.

Glomerular Processing

All ORNs expressing the same OR converge onto one of ~160 glomeruli in the antennal lobe (first olfactory processing center). Local interneurons perform lateral inhibition, enhancing contrast between similar odors. Projection neurons then convey the processed signal to the mushroom bodies (learning/memory) and lateral horn (innate responses).

The mushroom bodies of honeybees are exceptionally large (~170,000 Kenyon cells per hemisphere), reflecting the demands of olfactory learning in foraging. A forager bee learns floral odor associations in as few as 1–3 trials, forming long-term memories that persist for days.

4.5 Pheromone Communication Network

Schematic of the major pheromone-mediated communication channels in the honeybee colony, showing the queen's QMP signaling, alarm pheromone at the entrance, Nasonov scenting, and CHC-based nestmate recognition.

4.6 Simulation: Pheromone Diffusion & Signal Detection

This simulation models alarm pheromone diffusion from a point source in 2D, QMP dose-response curves with varying Hill coefficients, and cuticular hydrocarbon profile discriminant analysis for nestmate recognition.

Pheromone Diffusion, Dose-Response & Nestmate Recognition

Python

script.py175 lines

import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
from matplotlib.colors import LogNorm

fig = plt.figure(figsize=(16, 12))

# ======= Panel 1: 2D Alarm Pheromone Diffusion =======
ax1 = fig.add_subplot(2, 2, 1)

D = 6e-6    # Diffusion coefficient for IPA in air (m^2/s)
M = 1e-9    # Mass released (1 ng)

# 2D slice at z=0 of 3D diffusion
x = np.linspace(-0.5, 0.5, 200)  # meters
y = np.linspace(-0.5, 0.5, 200)
X, Y = np.meshgrid(x, y)
R2 = X**2 + Y**2

t_snapshot = 5.0  # 5 seconds after release
C_2d = (M / (4 * np.pi * D * t_snapshot)**(1.5)) * np.exp(-R2 / (4 * D * t_snapshot))

# Convert to ng/L
C_2d_ngL = C_2d * 1e9 * 1e3  # ng/m^3 -> ng/L

im = ax1.pcolormesh(X * 100, Y * 100, C_2d_ngL,
                     cmap='YlOrRd', shading='auto',
                     norm=LogNorm(vmin=0.01, vmax=C_2d_ngL.max()))
cbar = plt.colorbar(im, ax=ax1, label='Concentration (ng/L)')
cbar.ax.yaxis.label.set_color('white')
cbar.ax.tick_params(colors='white')

# Threshold circle
theta = np.linspace(0, 2*np.pi, 100)
# Find radius where C = 1 ng/L threshold
C_thresh = 1.0  # ng/L
r_thresh = np.sqrt(-4 * D * t_snapshot * np.log(C_thresh / (M / (4*np.pi*D*t_snapshot)**1.5 * 1e9 * 1e3)))
if not np.isnan(r_thresh):
    ax1.plot(r_thresh*100*np.cos(theta), r_thresh*100*np.sin(theta), 'w--', linewidth=1.5, alpha=0.7)
    ax1.text(r_thresh*100*0.7, r_thresh*100*0.7 + 3, 'Threshold', color='white', fontsize=8)

ax1.set_xlabel('X (cm)', fontsize=11, color='white')
ax1.set_ylabel('Y (cm)', fontsize=11, color='white')
ax1.set_title('Alarm Pheromone Diffusion (t = 5s)', fontsize=13, color='#f59e0b', fontweight='bold')
ax1.set_facecolor('#0a0a0a')
ax1.tick_params(colors='white')
ax1.set_aspect('equal')

# ======= Panel 2: QMP Dose-Response (Hill Equation) =======
ax2 = fig.add_subplot(2, 2, 2)

conc = np.logspace(-2, 2, 500)  # Relative QMP concentration
EC50 = 1.0

hill_coeffs = [0.5, 1.0, 2.0, 3.0, 5.0]
colors_hill = ['#60a5fa', '#22c55e', '#f59e0b', '#ef4444', '#a78bfa']

for n, color in zip(hill_coeffs, colors_hill):
    E = conc**n / (EC50**n + conc**n)
    ax2.semilogx(conc, E, color=color, linewidth=2, label=f'n = {n}')

ax2.axhline(0.5, color='white', linestyle=':', alpha=0.3)
ax2.axvline(EC50, color='white', linestyle=':', alpha=0.3)
ax2.text(EC50 * 1.2, 0.05, 'EC50', color='white', fontsize=9, alpha=0.6)

ax2.set_xlabel('[QMP] / EC50', fontsize=11, color='white')
ax2.set_ylabel('Effect (fraction)', fontsize=11, color='white')
ax2.set_title('QMP Dose-Response (Hill Equation)', fontsize=13, color='#f59e0b', fontweight='bold')
ax2.legend(title='Hill coeff.', fontsize=9, facecolor='#1a1a1a', edgecolor='#f59e0b',
           labelcolor='white', title_fontsize=9)
ax2.legend_.get_title().set_color('white')
ax2.set_facecolor('#0a0a0a')
ax2.tick_params(colors='white')
ax2.grid(True, alpha=0.15, color='white')
ax2.set_ylim(-0.05, 1.05)

# ======= Panel 3: CHC Profile - Nestmate vs Foreign =======
ax3 = fig.add_subplot(2, 2, 3)

np.random.seed(42)
n_compounds = 15  # Number of CHC peaks
compound_names = [f'C{i}' for i in range(23, 23 + n_compounds)]

# Nestmate colony profile (mean)
nestmate_mean = np.array([5, 12, 8, 3, 15, 20, 7, 10, 2, 18, 6, 9, 14, 4, 11], dtype=float)
nestmate_mean = nestmate_mean / nestmate_mean.sum() * 100  # percentages

# Foreign colony profile (shifted)
foreign_mean = nestmate_mean.copy()
# Shift some peaks
foreign_mean[1] *= 0.5
foreign_mean[4] *= 1.8
foreign_mean[7] *= 0.3
foreign_mean[9] *= 1.5
foreign_mean[12] *= 2.0
foreign_mean = foreign_mean / foreign_mean.sum() * 100

x_pos = np.arange(n_compounds)
width = 0.35

bars1 = ax3.bar(x_pos - width/2, nestmate_mean, width, color='#f59e0b', alpha=0.8, label='Nestmate')
bars2 = ax3.bar(x_pos + width/2, foreign_mean, width, color='#3b82f6', alpha=0.8, label='Foreign')

ax3.set_xlabel('Cuticular Hydrocarbon', fontsize=11, color='white')
ax3.set_ylabel('Relative Abundance (%)', fontsize=11, color='white')
ax3.set_title('CHC Profile: Nestmate vs. Foreign Colony', fontsize=13, color='#f59e0b', fontweight='bold')
ax3.set_xticks(x_pos)
ax3.set_xticklabels(compound_names, fontsize=7, rotation=45)
ax3.legend(fontsize=9, facecolor='#1a1a1a', edgecolor='#f59e0b', labelcolor='white')
ax3.set_facecolor('#0a0a0a')
ax3.tick_params(colors='white')
ax3.grid(True, alpha=0.1, color='white', axis='y')

# ======= Panel 4: Signal Detection Theory =======
ax4 = fig.add_subplot(2, 2, 4)

# Similarity score distributions
sigma = 1.0
d_prime_vals = [1.0, 2.0, 3.0, 4.0]
colors_sdt = ['#ef4444', '#f59e0b', '#22c55e', '#3b82f6']

s = np.linspace(-3, 8, 500)

for d_prime, color in zip(d_prime_vals, colors_sdt):
    mu_nest = d_prime
    mu_foreign = 0.0

# Nestmate distribution
    p_nest = np.exp(-0.5 * ((s - mu_nest) / sigma)**2) / (sigma * np.sqrt(2 * np.pi))
    p_foreign = np.exp(-0.5 * ((s - mu_foreign) / sigma)**2) / (sigma * np.sqrt(2 * np.pi))

ax4.plot(s, p_nest, color=color, linewidth=2, label=f"d' = {d_prime}")
    ax4.plot(s, p_foreign, color=color, linewidth=1, linestyle='--', alpha=0.5)

# Label distributions for d'=3 example
ax4.annotate('Nestmate', xy=(3.0, 0.35), fontsize=9, color='#86efac', ha='center')
ax4.annotate('Foreign', xy=(0.0, 0.35), fontsize=9, color='#86efac', ha='center', alpha=0.6)

# Decision criterion line
ax4.axvline(1.5, color='white', linestyle=':', alpha=0.5, linewidth=1)
ax4.text(1.7, 0.38, 'Decision\ncriterion', fontsize=8, color='white', alpha=0.7)

ax4.set_xlabel('CHC Similarity Score', fontsize=11, color='white')
ax4.set_ylabel('Probability Density', fontsize=11, color='white')
ax4.set_title("Signal Detection: Nestmate Recognition (d')", fontsize=13, color='#f59e0b', fontweight='bold')
ax4.legend(fontsize=9, facecolor='#1a1a1a', edgecolor='#f59e0b', labelcolor='white')
ax4.set_facecolor('#0a0a0a')
ax4.tick_params(colors='white')
ax4.grid(True, alpha=0.15, color='white')

fig.patch.set_facecolor('#0a0a0a')
plt.tight_layout()
plt.savefig('output.png', dpi=150, bbox_inches='tight', facecolor='#0a0a0a')
plt.close()

print("=== Pheromone Communication Simulation ===")
print(f"\nAlarm Pheromone (IPA):")
print(f"  Diffusion coefficient: D = {D:.1e} m^2/s")
print(f"  Mass released: {M*1e9:.1f} ng")
print(f"  At t = {t_snapshot}s:")
if not np.isnan(r_thresh):
    print(f"    Signal radius (1 ng/L threshold): {r_thresh*100:.1f} cm")
print(f"    Peak concentration: {C_2d_ngL.max():.1f} ng/L")
print(f"\nQMP Dose-Response:")
print(f"  EC50 = {EC50} (normalized)")
print(f"  Hill coefficient n = 2-3 (biological estimate)")
print(f"  At [QMP] = 0.5*EC50, n=3: E = {0.5**3/(1+0.5**3):.3f}")
print(f"  At [QMP] = 2*EC50,   n=3: E = {2**3/(1+2**3):.3f}")
print(f"\nCHC Nestmate Recognition:")
print(f"  Number of CHC compounds analyzed: {n_compounds}")
cos_sim = np.dot(nestmate_mean, foreign_mean) / (np.linalg.norm(nestmate_mean) * np.linalg.norm(foreign_mean))
print(f"  Cosine similarity (nest vs foreign): {cos_sim:.3f}")
print(f"  Typical d' for guard discrimination: 2-4")
print(f"  At d' = 3: hit rate ~93%, false alarm rate ~7%")

Click Run to execute the Python code

Code will be executed with Python 3 on the server

Pheromone Biochemistry: Biosynthesis Pathways

Honeybee pheromones are synthesized in specialized exocrine glands from core metabolic precursors. The biochemistry reveals how evolution has repurposed fatty acid synthesis, terpene biosynthesis, and amino acid catabolism to produce the colony's chemical language. Below we trace the biosynthetic pathways for the four major pheromone/secretion classes, derive enzyme kinetics for rate-limiting steps, and quantify the energetic costs of chemical communication.

Cross-reference: For comparative insect pheromone biochemistry including trail pheromone biosynthesis from proline, solenopsin alkaloid synthesis from lysine, and formic acid production via the THF pathway, see the Ant Biophysics — Pheromone Biochemistry section. Many enzymatic mechanisms (CYP450 hydroxylases, elongases, MVA pathway enzymes) are conserved across Hymenoptera.

1. Queen Mandibular Pheromone (QMP) Biosynthesis

9-ODA: (E)-9-Oxo-2-decenoic Acid

9-ODA is the primary active component of QMP, responsible for retinue attraction, ovary suppression, and mating flight signaling. It is synthesized from dietary stearic acid (C18:0) through a multi-step pathway in the queen's mandibular gland:

\[ \text{Stearic acid (C18:0)} \xrightarrow{\Delta 9\text{-desaturase}} \text{Oleic acid (C18:1)} \xrightarrow{\omega\text{-hydroxylase (CYP450)}} \text{18-OH oleic acid} \]

\[ \text{18-OH oleic acid} \xrightarrow{\beta\text{-oxidation} \times 4} \text{10-Hydroxy-2(E)-decenoic acid (10-HDA)} \xrightarrow{\text{oxidation}} \text{9-ODA} \]

Each cycle of \(\beta\)-oxidation shortens the chain by 2 carbons (C18 \(\rightarrow\) C16 \(\rightarrow\) C14\(\rightarrow\) C12 \(\rightarrow\) C10). The final oxidation of the C-9 hydroxyl to a keto group yields 9-ODA. The double bond at position 2 (trans configuration) is preserved from the oleic acid precursor through enoyl-CoA isomerase activity during \(\beta\)-oxidation.

\(\omega\)-Hydroxylase Enzyme Kinetics (CYP450 Family)

The \(\omega\)-hydroxylation step is catalyzed by a cytochrome P450 monooxygenase (CYP4 subfamily). This is the committed and rate-limiting step in 9-ODA biosynthesis. The enzyme follows standard Michaelis-Menten kinetics:

\[ v = \frac{V_{\max}[\text{oleic acid}]}{K_m + [\text{oleic acid}]} = \frac{k_{\text{cat}}[E]_0[\text{oleic acid}]}{K_m + [\text{oleic acid}]} \]

Measured kinetic parameters for insect CYP4 \(\omega\)-hydroxylases:

\(K_m\) \(\approx 15\,\mu\text{M}\) for oleic acid

\(k_{\text{cat}}\) \(\approx 2.5\,\text{s}^{-1}\)

\(k_{\text{cat}}/K_m\) \(\approx 1.7 \times 10^5\,\text{M}^{-1}\text{s}^{-1}\)

\([E]_0\) \(\approx 0.5\,\mu\text{M}\) in mandibular gland

Derivation of production rate: At saturating substrate (\([\text{oleic acid}] \gg K_m\)):

\[ v_{\max} = k_{\text{cat}} \cdot [E]_0 = 2.5\,\text{s}^{-1} \times 0.5\,\mu\text{M} = 1.25\,\mu\text{M/s} \]

Converting to mass production rate for the mandibular gland (volume \(\approx 0.5\,\mu\text{L}\)):

\[ \dot{m}_{\text{9-ODA}} = v_{\max} \cdot V_{\text{gland}} \cdot MW_{\text{9-ODA}} = 1.25 \times 10^{-6}\,\frac{\text{mol}}{\text{L}\cdot\text{s}} \times 0.5 \times 10^{-6}\,\text{L} \times 168\,\frac{\text{g}}{\text{mol}} \]

\[ \dot{m}_{\text{9-ODA}} \approx 1.05 \times 10^{-10}\,\text{g/s} \approx 9.1\,\mu\text{g/day} \]

This is consistent with measured QMP production rates of ~5–15 \(\mu\)g/day in mated, laying queens.

HVA: Homovanillyl Alcohol

HVA (4-hydroxy-3-methoxyphenylethanol) is an aromatic QMP component derived from tyrosine catabolism via the dopamine pathway:

\[ \text{L-Tyrosine} \xrightarrow{\text{TH}} \text{L-DOPA} \xrightarrow{\text{DDC}} \text{Dopamine} \xrightarrow{\text{MAO}} \text{DOPAC} \xrightarrow{\text{COMT}} \text{HVA} \]

TH = tyrosine hydroxylase; DDC = DOPA decarboxylase; MAO = monoamine oxidase; COMT = catechol-O-methyltransferase. This is the same catabolic pathway found in mammalian dopamine metabolism. HVA acts synergistically with 9-ODA, enhancing retinue behavior at concentrations 10-fold lower than 9-ODA alone.

2. Alarm Pheromone Biosynthesis — Isopentyl Acetate (IPA)

Isopentyl acetate (3-methylbutyl acetate, "banana oil") is synthesized in the sting gland (Koschevnikov gland) from the mevalonate (MVA) pathway intermediate isopentenol:

\[ \text{Acetyl-CoA} \xrightarrow{\text{MVA pathway}} \text{IPP} \xrightarrow{\text{phosphatase}} \text{Isopentenol (3-methyl-3-buten-1-ol)} \]

\[ \text{Isopentenol} \xrightarrow{\text{reductase}} \text{Isopentanol (3-methyl-1-butanol)} \xrightarrow{\text{acetyltransferase}} \text{Isopentyl acetate (IPA)} \]

The final step is an ester bond formation catalyzed by an alcohol acetyltransferase using acetyl-CoA as the acyl donor:

\[ \text{R-OH} + \text{CH}_3\text{CO-SCoA} \xrightarrow{\text{AAT}} \text{R-OOCCH}_3 + \text{CoA-SH} \]

Physical Properties of IPA

130 Da

Molecular weight

142°C

Boiling point

~30 s

Half-life in air

The high volatility (low MW, low boiling point) ensures rapid signal propagation and equally rapid signal decay. The ~30 s half-life in open air means the alarm signal self-erases, preventing prolonged false alarms. The diffusion coefficient in air is \(D \approx 6 \times 10^{-6}\,\text{m}^2/\text{s}\).

The exponential decay of IPA concentration at a fixed distance from the sting site:

\[ C(t) = C_0 \exp\!\left(-\frac{t}{\tau}\right), \quad \tau = \frac{1}{\lambda} \approx 43\,\text{s} \quad (\text{where } t_{1/2} = \tau \ln 2 \approx 30\,\text{s}) \]

3. Nasonov Pheromone Biosynthesis — Geraniol & Citral

The Nasonov pheromone components (geraniol, citral, geranic acid, nerol, nerolic acid) are monoterpenoids synthesized via the mevalonate (MVA) pathway in the Nasonov gland epithelium:

\[ \text{3 Acetyl-CoA} \xrightarrow{\text{MVA pathway}} \text{IPP} \rightleftharpoons \text{DMAPP} \]

\[ \text{IPP} + \text{DMAPP} \xrightarrow{\text{GPP synthase}} \text{GPP (geranyl pyrophosphate)} \xrightarrow{\text{phosphatase}} \text{Geraniol} \]

\[ \text{Geraniol} \xrightarrow{\text{ADH (alcohol dehydrogenase)}} \text{Geranial (trans-citral)} \]

Citral is a mixture of two geometric isomers:

Geranial (citral A, trans)

C\(_{10}\)H\(_{16}\)O, MW = 152 Da. The trans isomer around the C2=C3 double bond. Predominant form from direct ADH oxidation of geraniol.

Neral (citral B, cis)

C\(_{10}\)H\(_{16}\)O, MW = 152 Da. The cis isomer, formed by isomerization of geranial or by oxidation of nerol (the cis-isomer of geraniol).

GPP synthase catalyzes the head-to-tail condensation of IPP and DMAPP. The reaction proceeds via an ionization-condensation mechanism with an allylic carbocation intermediate:

\[ \text{DMAPP} \xrightarrow{-\text{PP}_i} \text{Allylic cation} \xrightarrow{+\text{IPP}} \text{GPP} + \text{H}^+ \]

Kinetic parameters for GPP synthase: \(K_m^{\text{DMAPP}} \approx 3\,\mu\text{M}\),\(K_m^{\text{IPP}} \approx 8\,\mu\text{M}\),\(k_{\text{cat}} \approx 0.8\,\text{s}^{-1}\). The enzyme is a homodimer with Mg\(^{2+}\)-dependent activity.

4. Beeswax Biosynthesis — Wax Esters from Fatty Acids

Beeswax is secreted by the wax glands (four pairs on the ventral abdomen, sternites 4–7) of worker bees aged 12–18 days. It is a complex mixture of hydrocarbons (~14%), monoesters (~35%), diesters (~14%), hydroxy monoesters (~4%), fatty acids (~12%), and fatty alcohols (~1%).

\[ \text{Acetyl-CoA} \xrightarrow[\text{malonyl-CoA}]{\text{FAS}} \text{Palmitic acid (C16:0)} \xrightarrow{\text{elongases}} \text{C24--C34 fatty acids} \]

\[ \text{C24--C34 acids} \xrightarrow{\text{reductase}} \text{Fatty alcohols (C24--C34)} \xrightarrow{\text{wax synthase}} \text{Wax esters} \]

The dominant wax ester is triacontyl palmitate(C\(_{30}\) alcohol + C\(_{16}\) acid = C\(_{46}\)ester, MW \(\approx\) 677 Da). The ester bond is formed by a wax synthase (acyltransferase).

Energy Cost of Wax Production

The conversion ratio is remarkably expensive: 8 kg honey \(\rightarrow\) 1 kg beeswax. This can be derived from the ATP cost of fatty acid synthesis:

\[ \text{Palmitate (C16): } 8\,\text{Acetyl-CoA} + 7\,\text{ATP} + 14\,\text{NADPH} \rightarrow \text{C16:0} + 7\,\text{ADP} + 14\,\text{NADP}^+ \]

\[ \text{Elongation to C30: additional } 7 \times (1\,\text{ATP} + 2\,\text{NADPH}) = 7\,\text{ATP} + 14\,\text{NADPH} \]

Total ATP equivalents per C30 fatty acid (counting NADPH as 3 ATP):

\[ \text{ATP}_{\text{total}} = (7 + 7) + 3 \times (14 + 14) = 14 + 84 = 98\,\text{ATP equivalents per C30 chain} \]

A wax ester requires one acid chain + one alcohol chain. With glucose yielding ~30 ATP via oxidative phosphorylation, and honey being ~80% sugars:

\[ \frac{m_{\text{honey}}}{m_{\text{wax}}} = \frac{2 \times 98\,\text{ATP} \times 180\,\text{g/mol glucose}}{30\,\text{ATP/glucose} \times 677\,\text{g/mol wax} \times 0.80} \approx 6.5\text{--}8 \]

The measured ratio (6.6–8.4 kg honey per kg wax) matches this metabolic calculation, confirming that wax production is an enormous energetic investment for the colony.

Enzyme Kinetics: Substrate Inhibition in CYP450 \(\omega\)-Hydroxylase

At high oleic acid concentrations, the CYP450 \(\omega\)-hydroxylase exhibits substrate inhibition, where excess substrate binds to the enzyme-substrate complex, forming a dead-end ternary complex. The rate equation becomes:

\[ v = \frac{V_{\max}[S]}{K_m + [S] + \frac{[S]^2}{K_i}} \]

Derivation: Consider the enzyme mechanism:

\[ E + S \underset{k_{-1}}{\overset{k_1}{\rightleftharpoons}} ES \overset{k_2}{\rightarrow} E + P \]

\[ ES + S \underset{k_{-3}}{\overset{k_3}{\rightleftharpoons}} SES \quad (\text{dead-end complex, no product formed}) \]

Applying steady-state to [ES] and [SES]:

\[ [E]_0 = [E] + [ES] + [SES] \]

\[ K_m = \frac{k_{-1} + k_2}{k_1}, \quad K_i = \frac{k_{-3}}{k_3} \]

\[ [ES] = \frac{[E]_0 [S]}{K_m + [S] + [S]^2/K_i}, \quad v = k_2[ES] = \frac{k_2[E]_0[S]}{K_m + [S] + [S]^2/K_i} \]

The velocity peaks at the optimal substrate concentration:

\[ [S]_{\text{opt}} = \sqrt{K_m \cdot K_i} \]

\[ v_{\text{peak}} = \frac{V_{\max}}{1 + 2\sqrt{K_m / K_i}} \]

For the CYP450 \(\omega\)-hydroxylase: \(K_i \approx 200\,\mu\text{M}\), so \([S]_{\text{opt}} = \sqrt{15 \times 200} \approx 55\,\mu\text{M}\). This substrate inhibition may serve as a self-regulatory mechanism, preventing excessive 9-ODA production that could be deleterious.

Pheromone Biochemistry: Biosynthesis Kinetics & Energetics

Python

script.py189 lines

import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

fig, axes = plt.subplots(2, 2, figsize=(16, 14))
fig.patch.set_facecolor('#0a0a00')
fig.suptitle("Bee Pheromone Biochemistry: Biosynthesis & Energetics", fontsize=16, color='white', fontweight='bold', y=0.98)

# ======= Panel 1: QMP 9-ODA Biosynthesis Rate (Substrate Inhibition) =======
ax1 = axes[0, 0]
ax1.set_facecolor('#1a1500')

S = np.linspace(0, 500, 500)  # oleic acid concentration (uM)
Km = 15.0    # uM
Vmax = 1.25  # uM/s
Ki_vals = [50, 100, 200, 500, 1e6]  # uM (1e6 = effectively no inhibition)
colors1 = ['#ef4444', '#f59e0b', '#fde047', '#22c55e', '#60a5fa']
labels1 = ['Ki = 50 uM', 'Ki = 100 uM', 'Ki = 200 uM', 'Ki = 500 uM', 'No inhibition']

for Ki, c, lab in zip(Ki_vals, colors1, labels1):
    v = Vmax * S / (Km + S + S**2 / Ki)
    ax1.plot(S, v, color=c, linewidth=2.5, label=lab)
    # Mark S_opt
    S_opt = np.sqrt(Km * Ki)
    if S_opt < 500:
        v_opt = Vmax * S_opt / (Km + S_opt + S_opt**2 / Ki)
        ax1.plot(S_opt, v_opt, 'o', color=c, markersize=6, markeredgecolor='white', markeredgewidth=0.8)

ax1.axhline(Vmax, color='white', linewidth=1, linestyle=':', alpha=0.4)
ax1.text(450, Vmax + 0.02, 'Vmax', color='white', fontsize=9, alpha=0.6)
ax1.set_xlabel('[Oleic acid] (uM)', color='white', fontsize=12)
ax1.set_ylabel('v (uM/s)', color='white', fontsize=12)
ax1.set_title('QMP 9-ODA Biosynthesis Rate\nCYP450 omega-Hydroxylase with Substrate Inhibition', color='#fde047', fontsize=13, fontweight='bold')
ax1.legend(fontsize=9, facecolor='#1a1500', edgecolor='#f59e0b', labelcolor='white')
ax1.tick_params(colors='white')
ax1.grid(True, alpha=0.15, color='#f59e0b')
for sp in ax1.spines.values(): sp.set_color('#f59e0b')

# ======= Panel 2: IPA Volatility Decay =======
ax2 = axes[0, 1]
ax2.set_facecolor('#1a1500')

t = np.linspace(0, 300, 500)  # seconds
half_life = 30.0  # seconds
tau = half_life / np.log(2)
lambda_decay = 1.0 / tau

# Different initial amounts (ng released)
M_vals = [100, 50, 20, 10, 5]
colors2 = ['#fde047', '#f59e0b', '#ef4444', '#22c55e', '#60a5fa']

for M0, c in zip(M_vals, colors2):
    C = M0 * np.exp(-lambda_decay * t)
    ax2.plot(t, C, color=c, linewidth=2.5, label=f'M0 = {M0} ng')

# Threshold line
C_thresh = 1.0  # ng (bee detection threshold)
ax2.axhline(C_thresh, color='white', linewidth=1.5, linestyle='--', alpha=0.6)
ax2.text(220, C_thresh + 1.5, 'Detection threshold (~1 ng)', color='white', fontsize=9, alpha=0.7)

# Mark half-life
ax2.axvline(half_life, color='#f87171', linewidth=1, linestyle=':', alpha=0.5)
ax2.text(half_life + 3, 80, 't1/2 = 30s', color='#f87171', fontsize=9, rotation=90)

ax2.set_xlabel('Time (s)', color='white', fontsize=12)
ax2.set_ylabel('IPA remaining (ng)', color='white', fontsize=12)
ax2.set_title('Alarm Pheromone (IPA) Volatility Decay\nExponential decay, t1/2 = 30s', color='#fde047', fontsize=13, fontweight='bold')
ax2.legend(fontsize=9, facecolor='#1a1500', edgecolor='#f59e0b', labelcolor='white')
ax2.tick_params(colors='white')
ax2.grid(True, alpha=0.15, color='#f59e0b')
ax2.set_ylim(0, 110)
for sp in ax2.spines.values(): sp.set_color('#f59e0b')

# ======= Panel 3: Geraniol Dose-Response (Nasonov Scenting Recruitment) =======
ax3 = axes[1, 0]
ax3.set_facecolor('#1a1500')

conc = np.logspace(-2, 3, 500)  # geraniol concentration (ng/L)
EC50_vals = [1.0, 5.0, 20.0, 100.0]
n_hill = 2.5  # Hill coefficient for Nasonov response
colors3 = ['#22c55e', '#4ade80', '#86efac', '#bbf7d0']
labels3 = ['EC50 = 1 ng/L', 'EC50 = 5 ng/L', 'EC50 = 20 ng/L', 'EC50 = 100 ng/L']

for EC50, c, lab in zip(EC50_vals, colors3, labels3):
    response = conc**n_hill / (EC50**n_hill + conc**n_hill)
    ax3.semilogx(conc, response * 100, color=c, linewidth=2.5, label=lab)

ax3.axhline(50, color='white', linewidth=1, linestyle=':', alpha=0.3)
ax3.text(0.015, 52, '50% response', color='white', fontsize=9, alpha=0.5)

# Mark natural concentration range
ax3.axvspan(5, 200, alpha=0.08, color='#22c55e')
ax3.text(30, 10, 'Natural\nconc. range', color='#22c55e', fontsize=10, ha='center', fontweight='bold')

ax3.set_xlabel('Geraniol concentration (ng/L)', color='white', fontsize=12)
ax3.set_ylabel('Recruitment response (%)', color='white', fontsize=12)
ax3.set_title('Nasonov Geraniol Dose-Response\n(Hill equation, n = 2.5)', color='#fde047', fontsize=13, fontweight='bold')
ax3.legend(fontsize=9, facecolor='#1a1500', edgecolor='#22c55e', labelcolor='white')
ax3.tick_params(colors='white')
ax3.grid(True, alpha=0.15, color='#f59e0b')
ax3.set_ylim(-5, 105)
for sp in ax3.spines.values(): sp.set_color('#f59e0b')

# ======= Panel 4: Wax Production Energy Budget =======
ax4 = axes[1, 1]
ax4.set_facecolor('#1a1500')

# Wax production over worker age
age_days = np.linspace(0, 45, 500)

# Wax gland activity peaks at day 12-18
wax_rate = 3.5 * np.exp(-0.5 * ((age_days - 15) / 4)**2)  # mg wax / day
honey_consumed = wax_rate * 8.0  # mg honey / day (8:1 ratio)

# Cumulative
dt = age_days[1] - age_days[0]
cum_wax = np.cumsum(wax_rate) * dt
cum_honey = np.cumsum(honey_consumed) * dt

ax4_twin = ax4.twinx()

# Bar-like representation of energy budget
ax4.fill_between(age_days, honey_consumed, alpha=0.25, color='#f59e0b', label='Honey consumed (mg/day)')
ax4.plot(age_days, honey_consumed, color='#f59e0b', linewidth=2.5)
ax4.fill_between(age_days, wax_rate, alpha=0.25, color='#fde047', label='Wax produced (mg/day)')
ax4.plot(age_days, wax_rate, color='#fde047', linewidth=2.5)

ax4_twin.plot(age_days, cum_wax, color='#22c55e', linewidth=2, linestyle='--', label='Cumulative wax (mg)')
ax4_twin.plot(age_days, cum_honey, color='#ef4444', linewidth=2, linestyle='--', label='Cumulative honey (mg)')

# Annotations
ax4.annotate('Peak wax\nproduction', xy=(15, 3.5), xytext=(25, 3.2),
            color='#fde047', fontsize=10, fontweight='bold',
            arrowprops=dict(arrowstyle='->', color='#fde047', lw=1.5))

# 8:1 ratio annotation
ax4.annotate('8:1 ratio\n(honey:wax)', xy=(15, 28), xytext=(30, 25),
            color='#f59e0b', fontsize=10,
            arrowprops=dict(arrowstyle='->', color='#f59e0b', lw=1.5))

ax4.set_xlabel('Worker age (days)', color='white', fontsize=12)
ax4.set_ylabel('Daily rate (mg/day)', color='#fde047', fontsize=12)
ax4_twin.set_ylabel('Cumulative (mg)', color='#22c55e', fontsize=12)
ax4.set_title('Beeswax Production Energy Budget\n8 kg honey per 1 kg wax', color='#fde047', fontsize=13, fontweight='bold')

# Phase labels
ax4.axvspan(0, 10, alpha=0.05, color='#60a5fa')
ax4.text(5, max(honey_consumed) * 0.85, 'Nurse\nphase', color='#60a5fa', fontsize=9, ha='center')
ax4.axvspan(10, 20, alpha=0.05, color='#fde047')
ax4.text(15, max(honey_consumed) * 0.85, 'Wax\nbuilder', color='#fde047', fontsize=9, ha='center')
ax4.axvspan(20, 45, alpha=0.05, color='#22c55e')
ax4.text(32, max(honey_consumed) * 0.85, 'Forager\nphase', color='#22c55e', fontsize=9, ha='center')

ax4.tick_params(colors='#fde047', axis='y')
ax4.tick_params(colors='white', axis='x')
ax4_twin.tick_params(colors='#22c55e')
ax4.grid(True, alpha=0.15, color='#f59e0b')
for sp in ax4.spines.values(): sp.set_color('#f59e0b')
for sp in ax4_twin.spines.values(): sp.set_color('#f59e0b')

lines1, labels1 = ax4.get_legend_handles_labels()
lines2, labels2 = ax4_twin.get_legend_handles_labels()
ax4.legend(lines1 + lines2, labels1 + labels2, fontsize=8, facecolor='#1a1500', edgecolor='#f59e0b', labelcolor='white', loc='upper right')

plt.tight_layout()
plt.savefig('output.png', dpi=150, bbox_inches='tight', facecolor='#0a0a00')
plt.close()

print("=== Bee Pheromone Biochemistry Simulation ===")
print(f"\nPanel 1: QMP 9-ODA Biosynthesis")
print(f"  Km = {Km} uM, Vmax = {Vmax} uM/s")
print(f"  S_opt (Ki=200): {np.sqrt(Km * 200):.1f} uM")
v_peak = Vmax * np.sqrt(Km*200) / (Km + np.sqrt(Km*200) + Km*200/200)
print(f"  v_peak (Ki=200): {v_peak:.3f} uM/s")
print(f"  Daily 9-ODA production: ~9.1 ug/day")
print(f"\nPanel 2: IPA Volatility Decay")
print(f"  Half-life: {half_life} s")
print(f"  Decay constant: {lambda_decay:.4f} s^-1")
print(f"  Time to reach threshold (100ng -> 1ng): {-np.log(1/100)/lambda_decay:.0f} s")
print(f"  MW = 130 Da, BP = 142 C")
print(f"\nPanel 3: Geraniol Dose-Response")
print(f"  Hill coefficient: n = {n_hill}")
print(f"  At 10 ng/L (EC50=5): {10**n_hill / (5**n_hill + 10**n_hill) * 100:.1f}% response")
print(f"\nPanel 4: Wax Production Energy Budget")
print(f"  Peak wax rate: {max(wax_rate):.1f} mg/day at day 15")
print(f"  Total lifetime wax: {cum_wax[-1]:.1f} mg")
print(f"  Total honey consumed: {cum_honey[-1]:.1f} mg")
print(f"  Honey:wax ratio: {cum_honey[-1]/cum_wax[-1]:.1f}:1")

Click Run to execute the Python code

Code will be executed with Python 3 on the server

References

Slessor, K.N., Winston, M.L. & Le Conte, Y. (2005). Pheromone communication in the honeybee (Apis mellifera L.). Journal of Chemical Ecology, 31(11), 2731–2745.
Hoover, S.E.R., Keeling, C.I., Winston, M.L. & Slessor, K.N. (2003). The effect of queen pheromones on worker honey bee ovary development. Naturwissenschaften, 90(10), 477–480.
Boch, R., Shearer, D.A. & Stone, B.C. (1962). Identification of isoamyl acetate as an active component in the sting pheromone of the honey bee. Nature, 195, 1018–1020.
Breed, M.D. (1998). Recognition pheromones of the honey bee. BioScience, 48(6), 463–470.
Robertson, H.M. & Wanner, K.W. (2006). The chemoreceptor superfamily in the honey bee, Apis mellifera: expansion of the odorant, but not gustatory, receptor family. Genome Research, 16(11), 1395–1403.
Sato, K., Pellegrino, M., Nakagawa, T., Vosshall, L.B. & Touhara, K. (2008). Insect olfactory receptors are heteromeric ligand-gated ion channels. Nature, 452(7190), 1002–1006.
Dani, F.R., Jones, G.R., Corsi, S., Beard, R., Pradella, D. & Turillazzi, S. (2005). Nestmate recognition cues in the honey bee: differential importance of cuticular alkanes and alkenes. Behavioral Ecology and Sociobiology, 59(1), 1–13.
Free, J.B. (1987). Pheromones of Social Bees. Chapman and Hall.
Sandoz, J.C. (2011). Behavioral and neurophysiological study of olfactory perception and learning in honeybees. Frontiers in Systems Neuroscience, 5, 98.
Le Conte, Y. & Hefetz, A. (2008). Primer pheromones in social Hymenoptera. Annual Review of Entomology, 53, 523–542.

M3. Thermoregulation & Energetics M5. Honey & Wax Biochemistry