Parvalbumin-Expressing Inhibitory Interneurons in Auditory Cortex Are Well-Tuned for Frequency

Parvalbumin-Expressing Inhibitory Interneurons in Auditory Cortex Are Well-Tuned for Frequency Moore Wehr

Tag layer of the DHSS Meta-Tagging Project. The article text is reproduced from the staged copy named in sourceDesc and is not altered; the tags are held separately in standOff so that overlapping and nested spans are preserved.

Parvalbumin-Expressing Inhibitory Interneurons in Auditory Cortex Are Well-Tuned for Frequency 2013 Journal of Neuroscience 10.1523/jneurosci.0663-13.2013 Moore2013-parvalbumin-expressing-inhibitory-interneuro Moore2013-parvalbumin-expressing-inhibitory-interneuro.txt, the project's staged copy 0e93cd24a63e095df8e07d3a0b1252cf33d0b3c8 not-redistributable neuroscience

the paper's own words

the tag layer; a sentence written by the tagger about the paper, always anchored to a quotation but not itself a quotation

0 spans located by searching the stored text for the quote's exact words, allowing any whitespace between them what the receptive fields of a genetically defined inhibitory cell class actually look like, as opposed to what is inferred from blocking inhibition wholesale Quotation withheld because this record is not cleared for redistribution. whether and how the response properties of PV+ neurons differ from those of PV- neurons in auditory cortex Quotation withheld because this record is not cleared for redistribution. auditory cortical PV+ neurons are well tuned for frequency, unlike their visual cortical counterparts, which is what the local-pooling account predicts given tonotopy Quotation withheld because this record is not cleared for redistribution. existence Quotation withheld because this record is not cleared for redistribution. frequency tuning of parvalbumin-positive interneurons in auditory cortex Quotation withheld because this record is not cleared for redistribution. electrophysiology electrophysiology Quotation withheld because this record is not cleared for redistribution. mechanisms PV+ interneurons pool densely and without bias from excitatory neurons in a small local radius, so their tuning inherits the feature organization of that local patch - homogeneous in a columnar or tonotopic area, heterogeneous in a salt-and-pepper one Quotation withheld because this record is not cleared for redistribution. mechanisms Greater monotonicity keeps PV+ firing rates rising over a wider span of sound levels, which stretches the level dynamic range and thereby flattens the slope of the rate-level function Quotation withheld because this record is not cleared for redistribution. mechanisms Faster membrane time constants combined with strong thalamocortical synaptic drive make PV+ cells fire early, placing them in a feedforward inhibitory loop that sharpens the timing of pyramidal responses Quotation withheld because this record is not cleared for redistribution. PV+ cells densely sample a small radius, so their tuning should mirror how the local network is organised - broad where tuning is salt-and-pepper, narrow where it is columnar Quotation withheld because this record is not cleared for redistribution. cotuned inhibition sharpens spike tuning through a thresholding or iceberg mechanism rather than by suppressing non-preferred stimuli Quotation withheld because this record is not cleared for redistribution. very short PV+ latencies suggest a feedforward inhibitory network driven rapidly by thalamocortical excitation Quotation withheld because this record is not cleared for redistribution. monotonic PV+ rate-level functions support local inhibition shaping intensity tuning in auditory cortex Quotation withheld because this record is not cleared for redistribution. the pooling and feedforward roles could be performed in sequence, or simultaneously for continuous sounds Quotation withheld because this record is not cleared for redistribution. Not every stimulus set could be run on every cell, so the sub-samples for tuning curves, rate-level functions and latency differ. what this limits: any comparison between the tuning-curve, rate-level and other sub-samples: DUE TO THE CONSTRAINTS OF RECORDING DURATION, WE WERE NOT ABLE TO PRESENT ALL STIMULUS ARRAYS FOR ALL CELLS Quotation withheld because this record is not cleared for redistribution. The PV− comparison group is only presumed to be excitatory; a small fraction could be non-PV inhibitory interneurons. what this limits: the comparison group in every PV+ versus PV- result: PRESUMABLY, THESE PV- CELLS WERE ALMOST EXCLUSIVELY EXCITATORY PYRAMIDAL CELLS, ALTHOUGH A SMALL FRACTION (<12%) COULD HAVE BEEN PV- INHIBITORY INTERNEURONS Quotation withheld because this record is not cleared for redistribution. The observed heterogeneity of PV+ response properties may partly reflect limited specificity of the PV promoter rather than true biological diversity. what this limits: the reported heterogeneity of PV+ responses: it is LIKELY DUE AT LEAST IN PART TO THE DEGREE OF SPECIFICITY OF THE PV PROMOTER, WHICH EXPRESSES IN MORE THAN ONE CELL TYPE Quotation withheld because this record is not cleared for redistribution. well tuned PV+ cells are also consistent with selective connectivity rather than unbiased pooling, though that would be a different rule from other cortical areas what this limits: the unbiased-pooling interpretation: WELL TUNED PV+ NEURONS ARE ALSO CONSISTENT WITH SELECTIVE CONNECTIVITY RATHER THAN UNBIASED POOLING, BECAUSE NARROW TUNING COULD FOLLOW FROM EITHER SCHEME Quotation withheld because this record is not cleared for redistribution. lateral inhibition could still operate here, either through a different network configuration or through other inhibitory cell types what this limits: the negative conclusion about lateral inhibition: LATERAL INHIBITION COULD STILL BE IMPLEMENTED IN AUDITORY CORTEX IN AT LEAST TWO WAYS Quotation withheld because this record is not cleared for redistribution. a small fraction of the PV- sample could have been non-PV inhibitory interneurons rather than pyramidal cells duplicate of: limitations[1] · what this limits: the same scope as limitations[1] - the PV- sample is defined by absence of a label, with a bound of under 12% on the contamination Quotation withheld because this record is not cleared for redistribution. recording duration prevented presenting every stimulus array to every cell, so the two samples are only approximately matched what this limits: the same scope as limitations[0] - stimuli PRESENTED TO OUR PV- SAMPLE WERE APPROXIMATELY MATCHED TO THE PV+ SAMPLE, not identically matched Quotation withheld because this record is not cleared for redistribution. deep-layer neurons were slightly underrepresented relative to the anatomical distribution what this limits: the laminar generality of the PV+ results: NEURONS IN DEEP LAYERS WERE SLIGHTLY UNDERREPRESENTED IN OUR SAMPLE, though the depth distributions DID NOT DIFFER SIGNIFICANTLY Quotation withheld because this record is not cleared for redistribution. the unbiased local pooling principle, carried across modalities as a prediction that reverses because auditory cortex is columnar where mouse V1 is not Quotation withheld because this record is not cleared for redistribution. the monotonicity index and rate-level function decomposition, used to give intensity tuning a single number Quotation withheld because this record is not cleared for redistribution. rate-level functions NORMALISED before comparison eliminates: the higher baseline firing rate of PV+ neurons as the source of the difference - without it, a cell type that simply fires more would appear to have a different rate-level relationship · outcome (stated by the tagger, no quotation): held Quotation withheld because this record is not cleared for redistribution. PV+ here means a cell that responded to blue light with a reliable short-latency burst, in a mouse where over 97 percent of PV-immunopositive cells carry the opsin - identity by manipulation, validated against a stain in two separate reporter lines Quotation withheld because this record is not cleared for redistribution. the FS/RS spike-width cutoff of 0.8 ms was set from THIS dataset, halfway between the two group means, and the paper states there is no consensus value in the literature because it may depend on the recording technology Quotation withheld because this record is not cleared for redistribution. UNTUNED is a category created because the measure does not apply: a sound-responsive cell with disordered frequency-intensity structure has no meaningful tuning width, so it is labelled rather than fitted Quotation withheld because this record is not cleared for redistribution. monotonicity index is a ratio of the response at the loudest tested level to the maximum response, so 1 means no intensity tuning and near zero means strong tuning - the measure carries its own null Quotation withheld because this record is not cleared for redistribution. the 80 ms response window was chosen to exclude non-evoked spikes BECAUSE PV+ cells have high spontaneous rates - an analysis choice driven by one of the very differences under study Quotation withheld because this record is not cleared for redistribution. claim_type: existence a second reading reaching the same value. Recorded: existence — note: a second reading reaching the same value from a different sentence. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. method: extracellular single-unit recording with optogenetic identification by light response, Gaussian fits to iso-intensity tuning curves and split-Gaussian fits to rate-level functions a second reading DIFFERING from the recorded value. Recorded: electrophysiology — note: a second reading, verbatim-verified, disagreeing with the value on the record. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. phenomenon: frequency tuning widths of PV+ and PV- neurons that do not differ, with very tightly tuned PV+ cells uncommon a second reading DIFFERING from the recorded value. Recorded: frequency tuning of parvalbumin-positive interneurons in auditory cortex — note: a second reading, verbatim-verified, disagreeing with the value on the record. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. contradicts → Wu et al. (2008), FS cells in rat A1 Quotation withheld because this record is not cleared for redistribution. builds-on → local-pooling account of PV+ tuning (Kerlin et al., 2010; Hofer et al., 2011) Quotation withheld because this record is not cleared for redistribution. replicates → Atencio and Schreiner (2008), FS cells in auditory cortex Quotation withheld because this record is not cleared for redistribution. contradicts → lateral inhibition models in which sharpening comes from inhibitory neurons more broadly tuned than excitatory ones Quotation withheld because this record is not cleared for redistribution. supports → the cotuning model in which excitation and inhibition share frequency tuning and inhibition sharpens by thresholding Quotation withheld because this record is not cleared for redistribution. extends → the local unbiased pooling account of PV+ tuning developed in mouse visual cortex Quotation withheld because this record is not cleared for redistribution. critiques → the use of extracellular spike waveform as a marker for PV+ interneurons in vivo Quotation withheld because this record is not cleared for redistribution. qualifies → pharmacological and voltage-clamp studies that lump all sources of synaptic inhibition together Quotation withheld because this record is not cleared for redistribution. PV+ —contrasted-with→ visual cortex PV+ visual cortex strength of the relation: 1 · note: the same cell class gives OPPOSITE tuning answers in two modalities, and that is the result rather than a discrepancy Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but auditory PV+ neurons are contrasted WITH PV+ neurons IN visual cortex; 'visual cortex' as the endpoint drops the cell type. Quotation withheld because this record is not cleared for redistribution. local pooling —causes→ tuning width local pooling tuning width strength of the relation: 1 · note: MECHANISM: dense sampling of a small radius means the cell inherits whatever tuning organisation that radius has - so a tonotopic area predicts well-tuned inhibitory cells Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the local pooling HYPOTHESIS PREDICTS similar tuning, and the data were consistent; causes is hedged. Quotation withheld because this record is not cleared for redistribution. waveform —distinct-from→ PV+ waveform PV+ strength of the relation: 1 · note: THE PROXY MEASURED AGAINST A GENETIC GROUND TRUTH AND FOUND WANTING: up to 40 percent misclassification from any single waveform measure Quotation withheld because this record is not cleared for redistribution. FS —distinct-from→ PV+ FS PV+ strength of the relation: 1 · note: the two are heterogeneous populations that OVERLAP BUT ARE NOT IDENTICAL - so results reported for one do not transfer to the other Quotation withheld because this record is not cleared for redistribution. PV+ —part-of→ basket cells PV+ basket cells strength of the relation: 0.5 — the record marks this weak · note: and only approximately half of basket cells express PV, which is the second reason the waveform proxy cannot work Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but PV is expressed in only about half of basket cells; part-of is loose and runs the other way. Quotation withheld because this record is not cleared for redistribution. untuned —distinct-from→ tuning width untuned tuning width strength of the relation: 1 · note: THE MEASURE HAS A DOMAIN: a sound-responsive cell with disordered structure gets a label rather than a number Quotation withheld because this record is not cleared for redistribution. Wilcoxon rank-sum test Wilcoxon rank-sum test Quotation withheld because this record is not cleared for redistribution. standard deviation standard deviation Quotation withheld because this record is not cleared for redistribution. Kolmogorov-Smirnov test Kolmogorov-Smirnov test Quotation withheld because this record is not cleared for redistribution. Fisher's exact test Fisher's exact test Quotation withheld because this record is not cleared for redistribution. ANOVA ANOVA Quotation withheld because this record is not cleared for redistribution. curve fitting (Gaussian) curve fitting (Gaussian) Quotation withheld because this record is not cleared for redistribution. goodness of fit / coefficient of determination goodness of fit / coefficient of determination Quotation withheld because this record is not cleared for redistribution. correlation correlation Quotation withheld because this record is not cleared for redistribution. resampling (random subsampling of trials) resampling (random subsampling of trials) Quotation withheld because this record is not cleared for redistribution. parametric model fitting parametric model fitting Quotation withheld because this record is not cleared for redistribution. Wilcoxon rank sum test Wilcoxon rank sum test Quotation withheld because this record is not cleared for redistribution. two-sample Kolmogorov–Smirnov test two-sample Kolmogorov–Smirnov test Quotation withheld because this record is not cleared for redistribution. Frequency tuning width did not differ significantly between optogenetically identified PV+ interneurons and PV- neurons in mouse auditory cortex (0.84 vs 0.62 octaves). Quotation withheld because this record is not cleared for redistribution. Very tightly tuned cells (below 0.5 octaves) were about half as common among PV+ cells (29%) as among PV- cells (58%), even though the two distributions largely overlapped. Quotation withheld because this record is not cleared for redistribution. The best frequency of every tuned PV+ neuron was within half an octave of its recording site's best frequency, so PV+ tuning matched the local population. Quotation withheld because this record is not cleared for redistribution. PV+ cells responded to best-frequency tones roughly 11 ms earlier than PV- cells (17.9 vs 29.1 ms). Quotation withheld because this record is not cleared for redistribution. The latency advantage was stimulus-specific: for white noise, PV+ and PV- latencies did not differ significantly. Quotation withheld because this record is not cleared for redistribution. PV+ cells had twofold higher spontaneous firing rates than PV- cells (5.5 vs 2.0 Hz), and higher evoked rates. Quotation withheld because this record is not cleared for redistribution. After normalizing for their higher firing rates, PV+ cells had shallower response gain to characteristic-frequency tones than PV- cells (0.04 vs 0.15). Quotation withheld because this record is not cleared for redistribution. PV+ cells were markedly less intensity-tuned than PV- cells for tones at characteristic frequency: only 14% of PV+ cells were nonmonotonic compared with 55% of PV- cells. Quotation withheld because this record is not cleared for redistribution. For white noise, PV+ and PV- neurons did not differ significantly in monotonicity. Quotation withheld because this record is not cleared for redistribution. Classifying cells by extracellular spike waveform instead of by PV expression would misclassify as many as 40% of them, so waveform is an unreliable marker of PV+ identity. Quotation withheld because this record is not cleared for redistribution. The classification method changes the scientific conclusion: sorting the same cells by spike width rather than PV expression made narrow-spike cells come out significantly more broadly tuned than broad-spike cells (0.87 vs 0.60 octaves). Quotation withheld because this record is not cleared for redistribution. The PV+/PV- differences reported across all layers held within each laminar group as well. Quotation withheld because this record is not cleared for redistribution. frequency tuning widths of PV+ and PV- cells were not significantly different, as the local pooling account predicts for a tonotopic area Quotation withheld because this record is not cleared for redistribution. very tightly tuned PV+ cells were comparatively rare, which fits pooling over locally diverse tuning Quotation withheld because this record is not cleared for redistribution. THE PREDICTION WAS DERIVED BEFORE THE TEST: if unbiased pooling generalises across modalities, PV+ cells should be well tuned here because auditory cortex has frequency columns Quotation withheld because this record is not cleared for redistribution. and the same principle explains why PV+ cells are broadly tuned in mouse V1, which lacks orientation columns Quotation withheld because this record is not cleared for redistribution. the tag was quantified in both directions and in two reporter lines: over 97 percent of PV-stained cells carried the opsin, and over 97 percent of reporter cells stained for PV Quotation withheld because this record is not cleared for redistribution. A CONTAMINATION BOUND COMPUTED RATHER THAN ASSERTED: from under 1 percent pyramidal-morphology reporter cells and 24 of 68 recordings in layer V, the sample is estimated to contain fewer than 0.24 pyramidal neurons Quotation withheld because this record is not cleared for redistribution. PV+ cells did have narrower spikes, 0.75 against 0.93 ms Quotation withheld because this record is not cleared for redistribution. but the distributions overlapped on every waveform measure: 27 of 72 PV- cells had narrow spikes, 29 had negative endslopes, 17 had low peak/trough ratios Quotation withheld because this record is not cleared for redistribution. THE PROXY QUANTIFIED AND REJECTED: any single waveform measure could misclassify up to 40 percent, and PV+ and FS are overlapping but non-identical populations Quotation withheld because this record is not cleared for redistribution. and there is no consensus spike-width cutoff in the literature, because it may depend on the recording technology Quotation withheld because this record is not cleared for redistribution. two reasons the proxy cannot work: extracellular duration depends on cell size and electrode position, and even intracellularly the fast-spiking to PV relation is not one to one Quotation withheld because this record is not cleared for redistribution. SOME CELLS GOT A LABEL INSTEAD OF A NUMBER: a sound-responsive neuron whose Gaussian fit was poor was categorized as untuned rather than given a tuning width Quotation withheld because this record is not cleared for redistribution. the untuned criterion was declared arbitrary and varying it by plus or minus 0.2 did not change the result Quotation withheld because this record is not cleared for redistribution. further cells were dropped when best frequency exceeded the speaker range, leaving 31 PV+ and 34 PV- for tuning width Quotation withheld because this record is not cleared for redistribution. NO A PRIORI LATENCY CONSTRAINTS, and imposing several exclusion criteria afterwards did not change the results Quotation withheld because this record is not cleared for redistribution. and the latency result was confirmed with an alternative estimation method Quotation withheld because this record is not cleared for redistribution. PV+ neurons had markedly faster response latencies, read as feedforward inhibition driven by thalamocortical excitation Quotation withheld because this record is not cleared for redistribution. PV+ rate-level functions were more monotonic at characteristic frequency, supporting a role in shaping intensity tuning Quotation withheld because this record is not cleared for redistribution. STIMULUS-DEPENDENT: for white noise, gain, dynamic range and monotonicity did not separate the two classes at all Quotation withheld because this record is not cleared for redistribution. the broadly-tuned-inhibition version of lateral inhibition is inconsistent with equal tuning widths Quotation withheld because this record is not cleared for redistribution. and the far-projecting version is inconsistent with PV+ tuning matching the local field potential and with equal connectivity spread Quotation withheld because this record is not cleared for redistribution. so the results favour cotuning with only moderate impact on frequency tuning Quotation withheld because this record is not cleared for redistribution. though lateral inhibition remains possible through a different network configuration or through other inhibitory cell types Quotation withheld because this record is not cleared for redistribution. A DISSOCIATION BETWEEN THE GENETIC AND PHYSIOLOGICAL CLASSES: tuning was equal for PV+ and PV- but slightly broader for FS cells, hinting at two distinct inhibitory networks Quotation withheld because this record is not cleared for redistribution. well tuned PV+ cells would also follow from selective connectivity, which the data cannot exclude Quotation withheld because this record is not cleared for redistribution. the recorded depth distribution matched the anatomical one, with deep layers slightly underrepresented Quotation withheld because this record is not cleared for redistribution. monotonicity index (MI) monotonicity index (MI) Quotation withheld because this record is not cleared for redistribution. characteristic frequency characteristic frequency Quotation withheld because this record is not cleared for redistribution. response gain response gain Quotation withheld because this record is not cleared for redistribution. endslope endslope Quotation withheld because this record is not cleared for redistribution. monotonicity index monotonicity index Quotation withheld because this record is not cleared for redistribution. untuned untuned Quotation withheld because this record is not cleared for redistribution. FS FS Quotation withheld because this record is not cleared for redistribution. cotuning cotuning Quotation withheld because this record is not cleared for redistribution. frequency tuning width frequency tuning width Quotation withheld because this record is not cleared for redistribution. response latency response latency Quotation withheld because this record is not cleared for redistribution. sound-evoked firing rate sound-evoked firing rate Quotation withheld because this record is not cleared for redistribution. maximal evoked firing rate maximal evoked firing rate Quotation withheld because this record is not cleared for redistribution. depth distributions depth distributions Quotation withheld because this record is not cleared for redistribution. tuning width tuning width Quotation withheld because this record is not cleared for redistribution. spike width spike width Quotation withheld because this record is not cleared for redistribution. latency latency Quotation withheld because this record is not cleared for redistribution. n = 68 — PV+ interneurons identified in vivo by their short-latency responses to blue light Quotation withheld because this record is not cleared for redistribution. More than 97% — proportion of PV-immunostained cells that expressed the ChR2-eYFP optogenetic tag (expression specificity) Quotation withheld because this record is not cleared for redistribution. 31 PV+ and 34 PV− — cells surviving the tuning-curve exclusion criteria and entering the tuning-width analysis Quotation withheld because this record is not cleared for redistribution. 0.75 ± 0.18 ms / 0.93 ± 0.29 ms — mean extracellular spike width of PV+ versus PV- cells Quotation withheld because this record is not cleared for redistribution. up to 40% — misclassification rate incurred by identifying PV+ cells from any single extracellular waveform measure Quotation withheld because this record is not cleared for redistribution. σ = 0.17 octaves to σ = 2.3 octaves — full range of measured frequency tuning widths across the 65 analysed cells Quotation withheld because this record is not cleared for redistribution. 0.76 ± 0.22 / 0.55 ± 0.28 — mean monotonicity index at characteristic frequency for PV+ versus PV- cells Quotation withheld because this record is not cleared for redistribution. 21.1 ± 18.3 Hz / 8.6 ± 6.5 Hz — firing-rate dynamic range of PV+ versus PV- rate-level functions Quotation withheld because this record is not cleared for redistribution. 32 ± 18 dB / 19 ± 15 dB — sound-level dynamic range over which firing rate kept increasing, PV+ versus PV- Quotation withheld because this record is not cleared for redistribution. up to 40% — misclassification rate from separating cells by any single waveform measure Quotation withheld because this record is not cleared for redistribution. 0.8 ms — the fast-spiking cutoff, set from this dataset because the literature has no consensus value Quotation withheld because this record is not cleared for redistribution. <0.24 pyramidal neurons — computed expected contamination of the 68-cell PV+ sample Quotation withheld because this record is not cleared for redistribution. 0.75 ± 0.18 ms and 0.93 ± 0.29 ms — PV+ and PV- spike widths; PV+ cells indeed had narrower spikes Quotation withheld because this record is not cleared for redistribution. 27/72 — PV- cells that would have been called fast-spiking on spike width alone Quotation withheld because this record is not cleared for redistribution. 31 PV+ and 34 PV− — cells surviving both exclusions for the tuning width comparison, from 68 and 72 recorded Quotation withheld because this record is not cleared for redistribution.

The full text of this article is not redistributable, so it is not included here. The evidence sentences are withheld with it; the tag layer and its grounding status remain. The article itself is at https://doi.org/10.1523/jneurosci.0663-13.2013.