8.8 KiB
Table of Contents
include_toc
| include_toc |
|---|
| true |
Intro
The synapse uses three interlocking signal systems to translate present activity into future behavioral bias. Ca²⁺ is the universal event recorder — each compartment reads its concentration dynamics differently (amplitude and speed of rise in the postsynapse, residual accumulation in the presynapse, IP3-triggered waves in the astrocyte), so the same ion encodes distinct instructions depending on where and how it appears. cAMP/PKA is the contextual gate: driven by neuromodulatory broadcast (dopamine, norepinephrine), it doesn't write changes itself but determines whether the Ca²⁺ signal gets committed to permanent structure — by priming AMPA receptor insertion, silencing the LTD phosphatase machinery via DARPP-32, and activating CREB-driven gene expression for structural proteins. mGluRs provide the overflow sensing layer: when glutamate spills beyond the cleft, group II/III mGluRs on the presynapse activate a Gi-mediated autoinhibitory brake, while group I mGluRs on the astrocyte trigger the IP3→Ca²⁺→D-serine cascade that amplifies NMDA coincidence detection — a push-pull architecture that simultaneously throttles excessive release and widens the postsynaptic learning window.
Together these three systems form a hierarchical filter: Ca²⁺ asks did something happen?, mGluRs ask was it excessive?, and cAMP/PKA asks was it worth saving? — and only when all three align does the synapse commit to rewriting its future response.
signal state variables
// ── Ca²⁺ : event recorder ──────────────────────────────────────
pre_Ca_residual // leftover Ca²⁺ between spikes — encodes recent history
post_Ca_amplitude // peak rise magnitude in spine
post_Ca_rise_speed // rate of rise — fast=LTP, slow=LTD
astro_Ca_local // IP3-triggered local rise near synapse
astro_Ca_global // soma-wide wave — network overload flag
// ── cAMP/PKA : context gate ────────────────────────────────────
cAMP_level // set by dopamine/NE via Gs → adenylyl cyclase
PKA_activity // downstream of cAMP
GluA1_Ser845_primed // bool — AMPA insertion threshold lowered
DARPP32_phospho // bool — PP1 (LTD phosphatase) silenced
CREB_active // bool — structural gene expression enabled
// ── mGluRs : overflow sensor ───────────────────────────────────
glutamate_spillover // extrasynaptic [glu] — only high when cleft saturated
mGluR2_3_activation // presynaptic Gi — autoinhibitory brake
mGluR5_activation // astrocytic Gq — IP3 → Ca²⁺ → D-serine cascade
layer 1 — Ca²⁺: did something happen?
function Ca_event_recorder(spike_history, input_freq):
// Presynapse: residual Ca²⁺ = trace of recent firing
pre_Ca_residual += spike_influx(input_freq)
pre_Ca_residual *= decay(τ ≈ 100ms) // fades unless spikes keep arriving
vesicle_release_prob *= facilitation(pre_Ca_residual)
// Postsynapse: amplitude + speed encode the instruction
post_Ca_amplitude = NMDA_influx(glutamate_cleft, membrane_potential)
post_Ca_rise_speed = d(post_Ca_amplitude) / dt
if post_Ca_amplitude > Ca_HIGH and post_Ca_rise_speed > fast_threshold:
activate(CaMKII) // → LTP kinase pathway
elif post_Ca_amplitude > Ca_LOW and post_Ca_rise_speed < slow_threshold:
activate(PP1, PP2B) // → LTD phosphatase pathway
else:
pass // sub-threshold — no instruction encoded
// Astrocyte: local vs global Ca²⁺ = two different alarms
astro_Ca_local = IP3_release(mGluR5_activation) // activity-proportional
astro_Ca_global = soma_wave(astro_Ca_local > OVERLOAD_threshold)
if astro_Ca_local > local_threshold:
D_serine_release += gliotransmitter_pulse() // widens NMDA window
if astro_Ca_global:
trigger(shockwave_lockdown) // circuit-breaker
layer 2 — mGluRs: was it excessive?
function mGluR_overflow_sensor():
// Only fires when cleft is genuinely saturated (low-affinity receptors)
glutamate_spillover = extrasynaptic_diffusion(glutamate_cleft)
if glutamate_spillover > spillover_threshold:
// Presynapse arm: Gi → brake
mGluR2_3_activation = True
cAMP_level -= Gi_inhibition(adenylyl_cyclase) // suppress PKA locally
vesicle_release_prob -= VGCC_suppression() // autoinhibitory brake
// Astrocyte arm: Gq → amplify (push-pull)
mGluR5_activation = True
astro_Ca_local += IP3_cascade(PLC_activation) // feeds back into layer 1
D_serine_release += proportional_to(astro_Ca_local)
// Net: same overflow signal brakes pre, amplifies post-learning window
return (mGluR2_3_activation, mGluR5_activation)
layer 3 — cAMP/PKA: was it worth saving?
function PKA_context_gate():
// Neuromodulators set the gate via Gs protein
if dopamine_level > D1_threshold or norepinephrine_level > β_threshold:
cAMP_level += Gs_activation(adenylyl_cyclase)
PKA_activity = proportional_to(cAMP_level)
// Target 1: prime AMPA insertion
phosphorylate(GluA1, site=Ser845)
GluA1_Ser845_primed = True // lowers threshold for CaMKII to anchor receptors
// Target 2: silence the forgetting machinery
phosphorylate(DARPP32)
DARPP32_phospho = True // inhibits PP1 → LTD pathway blocked
// Target 3: enable structural gene expression
translocate(PKA → nucleus)
phosphorylate(CREB)
CREB_active = True // new receptors, cytoskeleton, scaffolding
hierarchical filter — commit decision
function commit_to_structural_change():
// All three layers must align
event_detected = post_Ca_amplitude > Ca_HIGH // layer 1: did something happen?
overflow_sensed = mGluR5_activation == True // layer 2: was it excessive?
context_validated = DARPP32_phospho and GluA1_Ser845_primed // layer 3: worth saving?
if event_detected and overflow_sensed and context_validated:
activate(CaMKII) // Ca²⁺ signal now gets converted
AMPA_count += receptor_insertion(CaMKII, GluA1_Ser845_primed)
active_zone_size += structural_expansion(CREB_active)
ECM_integrity += astrocyte_sealing(astro_Ca_local)
return "potentiated"
elif event_detected and not context_validated:
return "temporary facilitation only" // Ca²⁺ rose but no save signal
elif not event_detected and overflow_sensed:
activate(PP1) // phosphatase wins — LTD
AMPA_count -= receptor_internalization(PP1)
return "depressed"
else:
return "baseline — no change"
Conclusion
The key architectural decision in this pseudocode is the separation into three explicit layers that feed into a single commit_to_structural_change function. Each layer answers one question independently before the final AND-gate runs — Ca²⁺ detects the event, mGluRs assess its magnitude, and cAMP/PKA validates its context. Notice also that mGluR layer has a push-pull side effect that feeds back into the Ca²⁺ layer (astro_Ca_local is updated by mGluR5_activation), making the system not a strict pipeline but a loop — the overflow sensor actively reshapes what the event recorder sees next.
Neuromodulators
These are produced by small, anatomically concentrated nuclei that broadcast widely across the brain:
dopamine_level // "save button" — validates LTP norepinephrine_level // arousal / signal-to-noise gain acetylcholine_level // attention — lowers LTP threshold
Dopamine
Dopamine is produced primarily by neurons in the Substantia Nigra pars compacta (projecting to the striatum, relevant for motor learning and habit formation) and the Ventral Tegmental Area (VTA) (projecting to the prefrontal cortex and limbic system via the mesolimbic and mesocortical pathways, relevant for reward, motivation, and the "save button" function in your model).
Norepinephrine
Norepinephrine is produced almost exclusively by the Locus Coeruleus, a tiny nucleus in the brainstem pons. Despite its small size it projects diffusely across virtually the entire brain — cortex, hippocampus, cerebellum, spinal cord. It's essentially the brain's arousal and signal-to-noise broadcaster, firing tonically at low rates during calm wakefulness and phasically during novel or stressful events.
Acetylcholine
Acetylcholine has two main sources: the basal forebrain nuclei (including the nucleus basalis of Meynert) projecting to the cortex and hippocampus — relevant for attention and learning gating — and the medial septum projecting specifically to the hippocampus, where it strongly modulates theta rhythms and memory encoding.
What's striking in the context of your model is that all three systems share the same architectural logic: a tiny, localized cell population broadcasts a global contextual signal that shifts the operational threshold of millions of synapses simultaneously — none of them carrying specific content, all of them modulating how content gets written.