Limits...
Heterogeneity and convergence of olfactory first-order neurons account for the high speed and sensitivity of second-order neurons.

Rospars JP, Grémiaux A, Jarriault D, Chaffiol A, Monsempes C, Deisig N, Anton S, Lucas P, Martinez D - PLoS Comput. Biol. (2014)

Bottom Line: We found that over all their dynamic range, PNs respond with a shorter latency and a higher firing rate than most ORNs.So, far from being detrimental to signal detection, the ORN heterogeneity is exploited by PNs, and results in two different schemes of population coding based either on the response of a few extreme neurons (latency) or on the average response of many (firing rate).Moreover, ORN-to-PN transformations are linear for latency and nonlinear for firing rate, suggesting that latency could be involved in concentration-invariant coding of the pheromone blend and that sensitivity at low concentrations is achieved at the expense of precise encoding at high concentrations.

View Article: PubMed Central - PubMed

Affiliation: Institut National de la Recherche Agronomique (INRA), Unité Mixte de Recherche 1392 Institut d'Ecologie et des Sciences de l'Environnement de Paris, Versailles, France.

ABSTRACT
In the olfactory system of male moths, a specialized subset of neurons detects and processes the main component of the sex pheromone emitted by females. It is composed of several thousand first-order olfactory receptor neurons (ORNs), all expressing the same pheromone receptor, that contact synaptically a few tens of second-order projection neurons (PNs) within a single restricted brain area. The functional simplicity of this system makes it a favorable model for studying the factors that contribute to its exquisite sensitivity and speed. Sensory information--primarily the identity and intensity of the stimulus--is encoded as the firing rate of the action potentials, and possibly as the latency of the neuron response. We found that over all their dynamic range, PNs respond with a shorter latency and a higher firing rate than most ORNs. Modelling showed that the increased sensitivity of PNs can be explained by the ORN-to-PN convergent architecture alone, whereas their faster response also requires cell-to-cell heterogeneity of the ORN population. So, far from being detrimental to signal detection, the ORN heterogeneity is exploited by PNs, and results in two different schemes of population coding based either on the response of a few extreme neurons (latency) or on the average response of many (firing rate). Moreover, ORN-to-PN transformations are linear for latency and nonlinear for firing rate, suggesting that latency could be involved in concentration-invariant coding of the pheromone blend and that sensitivity at low concentrations is achieved at the expense of precise encoding at high concentrations.

Show MeSH

Related in: MedlinePlus

Model of a large ORN population converging on a small PN population.(A, B) Example at dose C = 1 log ng of cumulated distributions of modelled firing rates (A) and latencies (B) of ORNs and comparison with experimental data. Distributions of experimental values (N = 32, same as in Fig. 5A, B) shown as staircase graphs (solid line) with 95% confidence intervals (green lines). Distributions of modelled F and L shown as smooth curves (in blue) based on N = 5 000 drawings. Firing rates in the model obey eq. 4 and latencies eq. 8 with parameters L0, ln(λ), Lm, FM, C1/2, ln(n) drawn from a multinormal distribution with their observed means, SDs and correlations (S5 Table). The modelled and experimental distributions are not significantly different (Kolmogorov–Smirnov tests at level 1%). (C) Examples of 5 simulated ORN responses with interspike interval 1/F and latency L at C = 1 log ng for 5 parameter drawings; 7000 such responses are summated to simulate the whole ORN population). (D) Simulated spontaneous activity of the whole ORN population (based on the lognormal distribution with μ = 1.23 and σ = 0.71 shown in Fig. 3E); the firing rate fluctuates around a stationary value ∼270 AP/10 ms. (E) Proportion of ORNs that respond with a shorter latency than the typical PN (with all 6 parameters equal to their median values; solid line), than a slow and insensitive PN (parameters equal to their 75% quantiles, dotted), and than a fast and sensitive PN (parameters equal to their 25% quantiles; dashed) at doses C0, C1/2 and Cs. (F) PSTH of the total number of spikes fired per 10 ms at doses from −8 to −2 log ng by a simulated population of 7000 NROs. The summated firing rate close to detection threshold (dotted line, 275 APs per 10 ms, see text) is reached for C≈−4.5 log ng at time ∼200 ms.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4256018&req=5

pcbi-1003975-g009: Model of a large ORN population converging on a small PN population.(A, B) Example at dose C = 1 log ng of cumulated distributions of modelled firing rates (A) and latencies (B) of ORNs and comparison with experimental data. Distributions of experimental values (N = 32, same as in Fig. 5A, B) shown as staircase graphs (solid line) with 95% confidence intervals (green lines). Distributions of modelled F and L shown as smooth curves (in blue) based on N = 5 000 drawings. Firing rates in the model obey eq. 4 and latencies eq. 8 with parameters L0, ln(λ), Lm, FM, C1/2, ln(n) drawn from a multinormal distribution with their observed means, SDs and correlations (S5 Table). The modelled and experimental distributions are not significantly different (Kolmogorov–Smirnov tests at level 1%). (C) Examples of 5 simulated ORN responses with interspike interval 1/F and latency L at C = 1 log ng for 5 parameter drawings; 7000 such responses are summated to simulate the whole ORN population). (D) Simulated spontaneous activity of the whole ORN population (based on the lognormal distribution with μ = 1.23 and σ = 0.71 shown in Fig. 3E); the firing rate fluctuates around a stationary value ∼270 AP/10 ms. (E) Proportion of ORNs that respond with a shorter latency than the typical PN (with all 6 parameters equal to their median values; solid line), than a slow and insensitive PN (parameters equal to their 75% quantiles, dotted), and than a fast and sensitive PN (parameters equal to their 25% quantiles; dashed) at doses C0, C1/2 and Cs. (F) PSTH of the total number of spikes fired per 10 ms at doses from −8 to −2 log ng by a simulated population of 7000 NROs. The summated firing rate close to detection threshold (dotted line, 275 APs per 10 ms, see text) is reached for C≈−4.5 log ng at time ∼200 ms.

Mentions: Not all ORNs in the population contribute equally to the PN response. The major contribution comes from the ORNs whose latency is shorter or equal to the PN latency, since no PN can respond faster than its presynaptic ORNs. In order to determine the fraction of contributing ORNs we relied on a model based on the C-F and C-L curves and the distributions of their parameters established above (see last section “Model of the signal delivered by the ORN population” in Materials and Methods). The model predicts the firing rates and latencies observed in the ORN population (Fig. 9A, B) and allows us to simulate the spike trains fired by this population when stimulated (Fig. 9C) and in the absence of stimulation (Fig. 9D). From these simulations, we calculated at any dose C the proportion of ORNs that respond with a given latency L1 or shorter. This proportion, as shown in Fig. 9E when L1 is the latency of the typical PN reconstructed from the median values of the fitted parameters, decreases with the dose and only 5±2% of the ORNs are enough to activate the typical PN. The proportion is greater (16±7%) for the slow PNs and smaller (2±0.3%) for the fast ones.


Heterogeneity and convergence of olfactory first-order neurons account for the high speed and sensitivity of second-order neurons.

Rospars JP, Grémiaux A, Jarriault D, Chaffiol A, Monsempes C, Deisig N, Anton S, Lucas P, Martinez D - PLoS Comput. Biol. (2014)

Model of a large ORN population converging on a small PN population.(A, B) Example at dose C = 1 log ng of cumulated distributions of modelled firing rates (A) and latencies (B) of ORNs and comparison with experimental data. Distributions of experimental values (N = 32, same as in Fig. 5A, B) shown as staircase graphs (solid line) with 95% confidence intervals (green lines). Distributions of modelled F and L shown as smooth curves (in blue) based on N = 5 000 drawings. Firing rates in the model obey eq. 4 and latencies eq. 8 with parameters L0, ln(λ), Lm, FM, C1/2, ln(n) drawn from a multinormal distribution with their observed means, SDs and correlations (S5 Table). The modelled and experimental distributions are not significantly different (Kolmogorov–Smirnov tests at level 1%). (C) Examples of 5 simulated ORN responses with interspike interval 1/F and latency L at C = 1 log ng for 5 parameter drawings; 7000 such responses are summated to simulate the whole ORN population). (D) Simulated spontaneous activity of the whole ORN population (based on the lognormal distribution with μ = 1.23 and σ = 0.71 shown in Fig. 3E); the firing rate fluctuates around a stationary value ∼270 AP/10 ms. (E) Proportion of ORNs that respond with a shorter latency than the typical PN (with all 6 parameters equal to their median values; solid line), than a slow and insensitive PN (parameters equal to their 75% quantiles, dotted), and than a fast and sensitive PN (parameters equal to their 25% quantiles; dashed) at doses C0, C1/2 and Cs. (F) PSTH of the total number of spikes fired per 10 ms at doses from −8 to −2 log ng by a simulated population of 7000 NROs. The summated firing rate close to detection threshold (dotted line, 275 APs per 10 ms, see text) is reached for C≈−4.5 log ng at time ∼200 ms.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4256018&req=5

pcbi-1003975-g009: Model of a large ORN population converging on a small PN population.(A, B) Example at dose C = 1 log ng of cumulated distributions of modelled firing rates (A) and latencies (B) of ORNs and comparison with experimental data. Distributions of experimental values (N = 32, same as in Fig. 5A, B) shown as staircase graphs (solid line) with 95% confidence intervals (green lines). Distributions of modelled F and L shown as smooth curves (in blue) based on N = 5 000 drawings. Firing rates in the model obey eq. 4 and latencies eq. 8 with parameters L0, ln(λ), Lm, FM, C1/2, ln(n) drawn from a multinormal distribution with their observed means, SDs and correlations (S5 Table). The modelled and experimental distributions are not significantly different (Kolmogorov–Smirnov tests at level 1%). (C) Examples of 5 simulated ORN responses with interspike interval 1/F and latency L at C = 1 log ng for 5 parameter drawings; 7000 such responses are summated to simulate the whole ORN population). (D) Simulated spontaneous activity of the whole ORN population (based on the lognormal distribution with μ = 1.23 and σ = 0.71 shown in Fig. 3E); the firing rate fluctuates around a stationary value ∼270 AP/10 ms. (E) Proportion of ORNs that respond with a shorter latency than the typical PN (with all 6 parameters equal to their median values; solid line), than a slow and insensitive PN (parameters equal to their 75% quantiles, dotted), and than a fast and sensitive PN (parameters equal to their 25% quantiles; dashed) at doses C0, C1/2 and Cs. (F) PSTH of the total number of spikes fired per 10 ms at doses from −8 to −2 log ng by a simulated population of 7000 NROs. The summated firing rate close to detection threshold (dotted line, 275 APs per 10 ms, see text) is reached for C≈−4.5 log ng at time ∼200 ms.
Mentions: Not all ORNs in the population contribute equally to the PN response. The major contribution comes from the ORNs whose latency is shorter or equal to the PN latency, since no PN can respond faster than its presynaptic ORNs. In order to determine the fraction of contributing ORNs we relied on a model based on the C-F and C-L curves and the distributions of their parameters established above (see last section “Model of the signal delivered by the ORN population” in Materials and Methods). The model predicts the firing rates and latencies observed in the ORN population (Fig. 9A, B) and allows us to simulate the spike trains fired by this population when stimulated (Fig. 9C) and in the absence of stimulation (Fig. 9D). From these simulations, we calculated at any dose C the proportion of ORNs that respond with a given latency L1 or shorter. This proportion, as shown in Fig. 9E when L1 is the latency of the typical PN reconstructed from the median values of the fitted parameters, decreases with the dose and only 5±2% of the ORNs are enough to activate the typical PN. The proportion is greater (16±7%) for the slow PNs and smaller (2±0.3%) for the fast ones.

Bottom Line: We found that over all their dynamic range, PNs respond with a shorter latency and a higher firing rate than most ORNs.So, far from being detrimental to signal detection, the ORN heterogeneity is exploited by PNs, and results in two different schemes of population coding based either on the response of a few extreme neurons (latency) or on the average response of many (firing rate).Moreover, ORN-to-PN transformations are linear for latency and nonlinear for firing rate, suggesting that latency could be involved in concentration-invariant coding of the pheromone blend and that sensitivity at low concentrations is achieved at the expense of precise encoding at high concentrations.

View Article: PubMed Central - PubMed

Affiliation: Institut National de la Recherche Agronomique (INRA), Unité Mixte de Recherche 1392 Institut d'Ecologie et des Sciences de l'Environnement de Paris, Versailles, France.

ABSTRACT
In the olfactory system of male moths, a specialized subset of neurons detects and processes the main component of the sex pheromone emitted by females. It is composed of several thousand first-order olfactory receptor neurons (ORNs), all expressing the same pheromone receptor, that contact synaptically a few tens of second-order projection neurons (PNs) within a single restricted brain area. The functional simplicity of this system makes it a favorable model for studying the factors that contribute to its exquisite sensitivity and speed. Sensory information--primarily the identity and intensity of the stimulus--is encoded as the firing rate of the action potentials, and possibly as the latency of the neuron response. We found that over all their dynamic range, PNs respond with a shorter latency and a higher firing rate than most ORNs. Modelling showed that the increased sensitivity of PNs can be explained by the ORN-to-PN convergent architecture alone, whereas their faster response also requires cell-to-cell heterogeneity of the ORN population. So, far from being detrimental to signal detection, the ORN heterogeneity is exploited by PNs, and results in two different schemes of population coding based either on the response of a few extreme neurons (latency) or on the average response of many (firing rate). Moreover, ORN-to-PN transformations are linear for latency and nonlinear for firing rate, suggesting that latency could be involved in concentration-invariant coding of the pheromone blend and that sensitivity at low concentrations is achieved at the expense of precise encoding at high concentrations.

Show MeSH
Related in: MedlinePlus