Why Transmitter Concentration Does Not Identify Neural Activity

Concentration Is a Balance, Not a Direct Readout of Activity

An extracellular transmitter concentration is the result of competing processes. Release adds transmitter to the measured space; uptake, enzymatic degradation, diffusion, and other losses remove it. A concentration measurement therefore describes the net state of that balance, not the rate of firing or release by itself. Two neural systems can have the same measured concentration while operating at very different rates.

This distinction matters in quantitative systems pharmacology because a measured biomarker is often used to constrain a mechanistic model. If concentration is treated as a direct proxy for neural activity, a model may assign a change to firing when it actually reflects clearance, or infer unchanged activity when release and clearance have both changed. The earlier material on neurotransmitter turnover and presynaptic feedback provides useful model components; the key issue here is which combinations of those components a concentration observation can identify.

Why a Single Concentration Does Not Identify Release and Clearance

At steady state, concentration is constant, so \(dC/dt=0\). The model then implies \(R=kC_{\mathrm{ss}}\), or \(C_{\mathrm{ss}}=R/k\). A measured steady-state concentration supplies one constraint on two unknown rates. Unless one rate is known independently or additional informative observations are available, the individual rates cannot be recovered from that concentration.

The non-identifiability can be shown directly. For any positive multiplier \(a\), the rates \(R'=aR\) and \(k'=ak\) give \(R'/k'=R/k\). They therefore produce exactly the same steady-state concentration. This is a structural limitation of the observation under the model, not a problem that can be solved merely by measuring the same steady-state concentration more precisely.

Firing rate is not the same quantity as \(R\). A higher firing rate may increase release, but release per spike can vary with release probability, available vesicles, transmitter content, autoreceptor feedback, and other factors. Conversely, changes to clearance can alter \(C\) without changing firing. A model that links firing \(F\) to release must state that link, for example \(R=\beta F\) over a specified range, and must treat \(\beta\) as an assumption or separately estimated parameter rather than as a universal constant.

Time Courses and Perturbations Add Information

With constant \(R\) and \(k\), the solution from an initial concentration \(C_0\) is

The steady-state value constrains \(R/k\), while the rate of approach also depends on \(k\). In a particularly informative experiment, release is abruptly stopped, so \(R=0\) during the observed decay. The model becomes \(dC/dt=-kC\), with solution \(C(t)=C_0e^{-kt}\). Taking logarithms gives \(\ln(C(t)/C_0)=-kt\). If the assumptions hold and the decay is measured over a suitable range, the slope of log concentration against time estimates \(-k\); the initial concentration and the steady-state balance can then help constrain \(R\).

This argument depends on the perturbation actually suppressing the relevant release input, on clearance remaining approximately first-order and stable during the measurement, and on the observed signal tracking the same compartment over time. If release continues, multiple compartments contribute, or the probe averages over a changing region, a simple exponential decay may not apply. A good fit to an exponential is not, by itself, proof that these assumptions are true.

What the Three Measurement Types Contribute

Microdialysis, PET displacement, and electrophysiology do not measure interchangeable versions of “neural activity.” They sample different aspects of the system, often at different spatial and temporal scales. Their joint value comes from combining distinct constraints while respecting each method’s limitations.

Microdialysis commonly collects transmitter over a finite interval, and the dialysate concentration is not automatically equal to the extracellular concentration. A simplified calibration may represent dialysate concentration as \(C_{\mathrm{dial}}=rC_{\mathrm{ext}}\), where \(r\) is probe recovery and \(C_{\mathrm{ext}}\) is the relevant extracellular concentration. Recovery can depend on the probe and experimental conditions. If \(r\) is unknown or changes across conditions, a difference in dialysate values need not be a proportional difference in extracellular concentration.

PET displacement requires particular care. In a simplified competitive binding picture, more endogenous transmitter can reduce radioligand binding, but the observed signal also depends on the number and properties of available receptors and on tracer kinetics. A change in PET signal is therefore evidence to interpret within a binding and measurement model, not a direct numerical readout of transmitter concentration. The same observed signal can have more than one explanation.

Electrophysiology provides a more direct observation of electrical activity in the sampled cells than either concentration or receptor binding does. However, a firing-rate estimate is still not a release-rate estimate. For example, equal firing rates need not imply equal release if release probability or presynaptic state differs. Similarly, different firing rates can coexist with equal extracellular concentrations if clearance differs.

Joint Interpretation, Diagnostics, and Alternatives

A useful analysis begins by writing down what each observation constrains. Microdialysis constrains a local, time-averaged extracellular concentration after accounting for recovery. Electrophysiology constrains activity in sampled cells. PET constrains radioligand binding under a receptor and tracer model. The turnover model connects concentration to release and clearance; additional links are required to connect firing to release and receptor binding to transmitter. A conclusion about one unmeasured process should not be presented as if it were directly observed.

Diagnostics include checking whether concentration follows a plausible time course, whether a log-linear decay is supported over the relevant range after release suppression, and whether conclusions change materially when probe recovery or receptor availability is varied within plausible bounds. A model that fits concentration but predicts the wrong decay or fails to match the electrophysiology is not rescued by its concentration fit. Likewise, an apparent PET displacement should be examined for alternative changes in receptor availability, tracer delivery, or binding assumptions.

The main alternative to interpreting a concentration snapshot alone is not simply “use more measurements”; it is to use measurements or interventions that constrain different parts of the mechanism. Repeated concentration measurements may reveal dynamics, but if they all occur at steady state under unchanged conditions they may still constrain only a rate ratio. A release-suppression experiment can inform clearance under its assumptions. Electrophysiology can characterize sampled activity, while PET can add receptor-level information. Each approach has limitations, so joint mechanistic inference should preserve uncertainty rather than claim unique identification when several parameter combinations remain compatible with the data.

A common mistake is: “The extracellular concentration is unchanged, so neural activity and transmitter release are unchanged.” The correction is that unchanged concentration establishes only that the measured balance is unchanged at the sampled scale. At steady state, \(R=kC_{\mathrm{ss}}\); an increase in release can be offset by faster clearance, and firing can change without a proportional change in release. Conversely, the same concentration can coexist with different receptor availability and electrical activity.