Converging sensory and motor cortical inputs onto the same striatal neurons: An in vivo intracellular investigation.

The striatum is involved in the completion and optimization of sensorimotor tasks. In rodents, its dorsolateral part receives converging glutamatergic corticostriatal (CS) inputs from whisker-related primary somatosensory (S1) and motor (M1) cortical areas, which are interconnected at the cortical level. Although it has been demonstrated that the medium-spiny neurons (MSNs) from the dorsolateral striatum process sensory information from the whiskers via the S1 CS pathway, the functional impact of the corresponding M1 CS inputs onto the same striatal neurons remained unknown. Here, by combining in vivo S1 electrocorticogram with intracellular recordings from somatosensory MSNs in the rat, we first confirmed the heterogeneity of striatal responsiveness to whisker stimuli, encompassing MSNs responding exclusively by subthreshold synaptic depolarizations, MSNs exhibiting sub- and suprathreshold responses over successive stimulations, and non-responding cells. All recorded MSNs also exhibited clear-cut monosynaptic depolarizing potentials in response to electrical stimulations of the corresponding ipsilateral M1 cortex, which were efficient to fire striatal cells. Since M1-evoked responses in MSNs could result from the intra-cortical recruitment of S1 CS neurons, we performed intracellular recordings of S1 pyramidal neurons and compared their firing latency following M1 stimuli to the latency of striatal synaptic responses. We found that the onset of M1-evoked synaptic responses in MSNs significantly preceded the firing of S1 neurons, demonstrating a direct synaptic excitation of MSNs by M1. However, the firing of MSNs seemed to require the combined excitatory effects of S1 and M1 CS inputs. This study directly demonstrates that the same somatosensory MSNs can process excitatory synaptic inputs from two functionally-related sensory and motor cortical regions converging into the same striatal sector. The effectiveness of these convergent cortical inputs in eliciting action potentials in MSNs may represent a key mechanism of striatum-related sensorimotor behaviors.

Introduction

The striatum, the main input stage and largest nucleus of the basal ganglia, is involved in the production and selection of adaptive motor behaviors which require the continuous integration of sensory information [1, 2]. To achieve this function, the striatum integrates massive glutamatergic synaptic inputs from a large and diverse set of cortical areas [3, 4]. While corticostriatal (CS) projections obey a topographical organization, each striatal recipient zone receives overlapping convergent inputs from multiple, often functionally-related, cortical regions [2-6].

With respect to its role in the completion of sensorimotor tasks, the dorsolateral part of the striatum processes convergent CS inputs from somatosensory and motor cortical areas [5-10]. The whisker-related CS pathway from the primary somatosensory cortex (S1), which conveys sensory information from the immediate environment to guide rodent exploratory behaviors [11], projects bilaterally to the dorsolateral striatum and forms a clustered pattern of axonal terminals [12, 13], with a few number of postsynaptic contacts per individual afferent CS neurons [14]. S1 CS neurons are distributed uniformly across layer 5 [10, 13, 15, 16], with a significant prevalence of ipsilateral projections [10, 17]. S1 CS inputs are processed by the GABAergic striatal projection neurons—the main (∼90%) neuronal population of the striatum—which have been morphologically defined as medium-sized spiny neurons (MSNs) [18]. We have recently demonstrated in vivo that the propagation of whisker-evoked sensory flow through the S1 CS pathway resulted in a selection of sensory information within the striatum [13, 16]. Specifically, while S1 CS neurons reliably responded to the deflection of contralateral whiskers by suprathreshold depolarizing postsynaptic potentials (dPSPs), only half of the related MSNs displayed dPSPs, which were effective in eliciting action potentials (APs) in one-third of the responding neurons. The remaining striatal neurons did not exhibit any detectable responses to sensory stimuli. This relative inconstancy of striatal responses for a given sensory input was likely due to the patchy organization of S1 CS projections [12, 13] and/or to a powerful shunting synaptic conductance resulting from the feedforward activation of presynaptic striatal GABAergic interneurons by cortical inputs [13, 19–22].

The somatosensory striatal sector also receives ipsilateral synaptic inputs from the CS neurons located in the vibrissal representation of the primary motor cortex (M1), which partially overlap with the S1 CS projections [5, 6, 9, 10, 23–25]. Whisker-related M1 and S1 cortical regions form reciprocal connections [26-28] and converge into the dorsolateral striatum, making this striatal sector an input zone of the sensorimotor corticofugal channel in the basal ganglia. Anatomical studies have revealed the convergence of CS projections from M1 and S1 onto the same striatal GABAergic interneurons [24]. In addition, a recent in vitro investigation from the mouse dorsolateral striatum has demonstrated that activation of S1 and M1 CS inputs generated monosynaptic dPSPs in both MSNs and striatal interneurons, with an elevated synaptic strength for the M1-induced responses [22]. However, the functional impact of M1 CS projections onto somatosensory MSNs in vivo, an experimental condition that fully preserves CS connectivity, has not been investigated so far and we still ignore whether S1 and M1 CS inputs can synaptically-depolarize and fire the same MSN. Addressing these issues is of crucial importance to understand the mechanisms of sensorimotor integration in the striatum. This will also extend our knowledge on the basic processes of integration of cortical information by individual MSNs, which have long been considered as “coincidence detectors” of converging excitatory inputs due to their distinctive electrical membrane properties endowing them with an excessive membrane polarization, a low intrinsic excitability and a weak capacity to summate synaptic events over time from their resting membrane potential [29-32].

In the present study, by combining in vivo S1 electrocorticogram (ECoG) with intracellular recordings from somatosensory MSNs and S1 layer 5 pyramidal neurons, we first analyzed the responsiveness of these two cell types to deflection of contralateral whiskers (Fig 1A). We then examined their synaptic and firing responses to electrical stimulations of the whisker-related M1 cortex (Fig 1B). By comparing the delay of M1-evoked firing in S1 neurons and the latency of M1-evoked dPSPs in the related MSNs, we found that M1 CS pathway could generate short-latency responses in somatosensory MSNs that did not require the intra-cortical recruitment of the S1 CS network (Fig 1B). This study directly demonstrates that the same somatosensory MSNs can process, and be excited by, both somatosensory and motor cortical inputs in vivo.

Fig 1

In vivo experimental paradigm.

(A) Intracellular (Intra) recordings were obtained from the primary somatosensory cortex (S1) and from MSNs located in the S1 projection field of the dorsolateral striatum, simultaneously with an ECoG of S1. Sensory stimuli consisted in iterative air puffs delivered onto the contralateral whiskers and the evoked-responses were collected in the ECoG, S1 neurons and MSNs. (B) Activation of the deep layers of the primary orofacial motor cortex (M1), projecting to the somatosensory striatum and to S1 reciprocally, was achieved through low intensity electrical stimulations (Stim.). M1-induced responses were recorded from the somatosensory MSNs and S1 neurons. St: striatum.

Materials and methods

The care and experimental manipulation of the animals strictly followed the European Union guidelines (directive 2010/63/EU) and received approval from the French Ministry for Research (# 00773.03) and the Charles Darwin Ethical Committee on Animal Experimentation (C2EA-05). Every precaution was taken to minimize suffering (see general and local anesthesia procedures below) and the number of animals used in each experimental series.

Animal preparation

Experiments were conducted in vivo from 51 adult male Sprague-Dawley rats weighting from 260 to 430 g (Charles River, L’Arbresle, France). Animals were initially anesthetized with sodium pentobarbital (30 mg/kg, i.p.; Centravet, Plancoët, France) and ketamine (25 mg/kg, i.p.; Centravet, Plancoët, France). A cannula was inserted into the trachea and the animal was placed in a stereotaxic frame. Wounds and pressure points were repeatedly (every 2 h) infiltrated with lidocaïne (2%; Centravet, Plancoët, France). After completion of the surgical procedures, moderate doses of pentobarbital (10–15 mg/kg, i.p.) were regularly administered during the recording session and the depth of anesthesia was assessed by continuous monitoring of heart rate and ECoG activity. To obtain long-lasting stable intracellular recordings, rats were immobilized with gallamine triethiodide (40 mg/2h, i.m.; Sigma, France) and artificially ventilated. Body temperature was maintained at 37°C using a feedback-controlled heating blanket. At the end of the experiments, animals received an overdose of sodium pentobarbital (200 mg/kg, i.p.).

Electrophysiological recordings and stimulations

Spontaneous ECoG activity and cortical sensory-evoked potentials (Fig 1), elicited by forced deflections of contralateral whiskers (see below), were captured with a low impedance (≈ 60 kΩ) silver electrode apposed on the dura above S1 (-2 mm anterior to the bregma; 5 mm lateral to the midline) [33]. The reference electrode was placed on a temporal muscle of the opposite side of the head.

Intracellular recordings from MSNs (n = 76) and S1 pyramidal neurons (n = 11), performed with glass micropipettes filled with 2 M KAc (50–80 MΩ), were obtained in current-clamp mode (Axoclamp-2B amplifier, Axon Instruments, Union City, CA) and in conjunction with the monitoring of S1 ECoG activity (Fig 1A). For intracellular labeling of MSNs, 1% neurobiotin (Vector Laboratories, Burlingame, CA) was added to the pipette solution. MSNs, identified on the basis of their distinctive electrophysiological and morphological properties (Fig 2), were recorded in the ipsilateral striatal projection field of S1 [10, 12, 13, 16], at the corresponding stereotaxic coordinates: -2.4 to -2.1 mm anterior to the bregma, 4.5 to 5.2 mm lateral to the midline, and 3.2 to 5.6 mm ventral to the brain surface [33]. S1 layer 5 pyramidal neurons were recorded from the vicinity (< 200 μm) of the ECoG electrode, within the cortical sector projecting to the recorded striatal region [10, 12, 13]: from -2.1 to -1.7 mm anterior to bregma; 4.8 to 5.1 mm lateral to the midline, and 1.2 to 2.1 mm under the cortical surface [33].

Fig 2

Morphological and basic electrophysiological properties of somatosensory striatal MSNs.

(A) Synthetic representation of a striatal MSN labeled by intracellular injection of neurobiotin. (B) Voltage changes and firing patterns (top records) of a MSN in response to the injection of positive (single response) and negative (averaging of 20 trials) current pulses (bottom traces). Note the high membrane polarization at rest and the slow ramp depolarization (from -60 mV, oblique arrow) induced by the depolarizing current pulse, resulting in a delayed firing. (C) Plot of voltage changes (ΔV) as a function of the injected current (Iinj). Note the membrane inward rectification in response to hyperpolarizing currents higher than -0.4 nA (arrows). The dashed line indicates the theoretical linear ohmic relation. Rm of MSNs (24 MΩ for this cell) was thus calculated from the mean voltage response obtained at -0.4 nA (crossed arrow). (D) Simultaneous recordings of S1 ECoG (top) and intracellular activity of a MSN (bottom) located in the corresponding CS projection field. The cross-correlation between intracellular and ECoG signals (inset) shows that striatal Vm fluctuations closely followed the slow (~5 Hz) cortical oscillation by about 15 ms. Data shown in (A–D) are from the same neuron. Here and in the following figures the value of Vm is indicated at the left of the intracellular records.

Sensory stimulations consisted in puffs (50 ms duration) of compressed air, delivered by a picospritzer unit (Picospritzer III, Intracel LTD, Royston Herts, UK) to the whiskers contralateral to the recording sites through a 1-mm diameter glass pipette (Fig 1A). Air-puffs were given 40 times with a low frequency (0.24 Hz) to prevent adaptation of whisker-evoked responses [34] and activity-dependent intrinsic plasticity in S1 pyramidal neurons [35]. The intensity of stimuli was set at the beginning of each experiment as the minimal air-puff pressure (20–50 psi) generating the largest ECoG response. Under these conditions, the air-puff stimuli deflected 4–8 whiskers by ~10 deg [13, 35].

Intracortical electrical stimulations (200 μs duration) were given 40 times through a bipolar stimulating electrode (1 mm tip separation) inserted (1.7 mm under the brain surface) in the orofacial part of the primary motor cortex (M1, 1.7–2.2 mm anterior to bregma; 1.8–2.2 mm lateral to the midline) [33] (Fig 1B), which is reciprocally connected with S1 [26-28] and whose CS projections overlap with S1 CS projections within the dorsolateral striatum [5, 6, 9, 10, 24]. Cortical stimulations were applied at low intensity (5–25 V) with the same frequency used for sensory stimuli (0.24 Hz). Before curarization of the animals, we ensured that electrical stimulations of M1 could induce movements of one or more whiskers (≈ 3 mm deflection).

Data acquisition and analysis

Electrophysiological recordings were digitized at 25 kHz for intracellular signals and at 3 kHz for ECoG waveforms (CED 1401plus, Spike2 software version 7.06; Cambridge Electronic Design, Cambridge, UK). Cross-correlograms between subthreshold striatal intracellular activity (1–5 s of continuous recording down-sampled at 3 kHz) and the corresponding ECoG signal (taken as the temporal reference) were computed using Spike2 software. The value of neuronal membrane potential (Vm) was calculated as the mean of the distribution of spontaneous subthreshold activity (10 s duration). The corresponding magnitude of Vm fluctuations was quantified as the standard deviation of the distribution (SD Vm). Vm values were eventually corrected when a tip potential was recorded after termination of the intracellular recording. Measurements of apparent membrane input resistance (Rm) and time constant (τm) were based on the linear electrical cable theory applied to an idealized isopotential neuron [36]. Because MSNs display prominent inward membrane rectification during injection of large amplitude negative current pulses [31, 32, 37] (Fig 2C), their Rm was assessed from the mean membrane voltage drop induced by repeated square current pulses of low intensity (-0.4 nA, n = 20, 200 ms duration, applied every 1.25 sec) (Fig 4D). τm was derived from an exponential fit applied to the initial part of the current-induced membrane hyperpolarization.

Fig 4

Properties of M1-induced responses in somatosensory MSNs.

(A–C) Intracellular responses (four superimposed records) from three somatosensory MSNs to electrical stimulations of M1 (timing and intensity of stimuli are indicated) (A1–C1), and corresponding air-puff induced responses (averaging of 15–40 successive trials) recorded from the same cells (A2–C2). The modality of sensory responsiveness in the illustrated neurons is indicated (see text for details). The bottom traces in A1 and B1 represent the averaging of at least 20 M1-evoked intracellular potentials. (D) Conductance increase upon M1-evoked depolarizations in MSNs. M1-induced potentials (averages of 20 successive responses) in a MSN, coupled with injection of current pulses (-0.4 nA, 5 ms) applied 140 ms after M1 stimulus (black record) or at the peak of the evoked response (grey record). The traces are superimposed and expanded at right. The inset at left (δVm) shows the subtraction of both records (as delimited by the dashed box), demonstrating the collapse of current-induced voltage drop at the peak of the induced depolarization (calibration bar, 5 mV). (E) Input-output relation in the M1-somatosensory MSNs circuit. (E1) Superimposed synaptic depolarizations (averages of at least 20 responses) evoked in a MSN by M1 stimulations applied with increasing intensity (5 to 20 V). The bottom trace indicates the method used to measure the latency (Lat.) and the amplitude (Ampl.) of the M1-evoked intracellular responses. (E2) Pooled data showing the changes in dPSPs amplitude in response to M1 stimuli of increasing intensity (M1 stim.). Only MSNs tested with at least two M1 stimulus intensities were computed. Data points correspond to mean ± SEM (the corresponding number of MSNs is indicated). Note the saturation of dPSP amplitude from an intensity of 15 V. The black dots in A–C and E1 indicate the electrical stimulus artifact.

The amplitude of APs was calculated as the voltage difference between their voltage threshold, measured as the membrane potential at which the dV/dt exceeded 10 V/s [35], and the peak of the waveform. Their total duration was measured as the time between their voltage threshold and the return to the same Vm value. The spontaneous firing rate of neurons was calculated from a continuous recording period ≥ 10 s.

The latency of intracellular S1- and M1-evoked responses was calculated as the time delay between the stimulus onset (air-puff or cortical electrical stimulus) and the foot of the neuronal responses (Fig 4E). Neuronal events having shape (rising and decay phases) and/or latency that did not match those of the mean synaptic response obtained after averaging of 20–40 trials were not included in the analysis. Amplitude of individual S1- or M1-evoked subthreshold potential was measured as the voltage difference between the foot and the peak of the response (Fig 4E). The firing probability of striatal and cortical neurons in response to whiskers deflection or M1 activation was calculated as the ratio between the number of suprathreshold synaptic responses and the total number of trials. Firing latency was the time between the stimulus onset and the peak of the AP waveform.

Numerical values are given as means ± standard error of the mean (SEM). Statistical significance was assessed using, appropriately, unpaired Student’s t tests, one-way analysis of variance or Mann-Whitney Rank Sum test. Statistical analyses were performed with SigmaStat 3.5 (SPSS Inc., Chicago, IL, USA).

Results

Morpho-functional properties of somatosensory MSNs and sensory-evoked responses

In vivo intracellular recordings were obtained from 76 MSNs located in the striatal region processing somatosensory information originating from the whisker-related S1 (Fig 1). Labeled cells (n = 4) displayed the distinctive morphological features of MSNs [13, 18], including somata with diameters between 10 and 20 μm which gave rise to 4–6 ramified primary dendrites densely covered with spines apart from their most proximal region (Fig 2A). Consistently, all recorded striatal neurons also displayed the characteristic electrophysiological properties of MSNs [13, 16, 31, 32, 37]. They exhibited a highly polarized mean Vm (-79.2 ± 0.5 mV, n = 76 MSNs), as compared to pyramidal cortical neurons (see Fig 5A and 5B), associated with a relatively low Rm (27.1 ± 0.8 MΩ) and a short τm (6.7 ± 0.2 ms) (Fig 2B). Threshold positive current pulses evoked in these cells a ramp-like membrane depolarization, starting at about -60 mV, which led to a long delay in the first AP discharge (Fig 2B). APs, measured on current-induced depolarizations, had an amplitude of 59.7 mV (± 0.8 mV, n = 73 MSNs), a total duration of 1.30 ms (± 0.03 ms) and a voltage threshold of -49.8 mV (± 0.2 mV). The membrane current-voltage relationship of recorded neurons consistently displayed a marked inward rectification in response to negative current pulses of increasing intensity (Fig 2C). The typical spontaneous activity of MSNs consisted in a continuous barrage of synaptic depolarizations (Figs 2D and 3, bottom records), responsible for large Vm fluctuations (SD Vm = 4.5 ± 0.2 mV, n = 76 MSNs), causing, however, a low firing rate (0.1–2 Hz) in a restricted number of neurons (15 out of 76). This ongoing rhythmic synaptic activity was tightly correlated with the oscillatory waveforms simultaneously recorded in S1 ECoG, with a negative temporal shift (10–17 ms) (Fig 2D, inset) consistent with the unidirectional and monosynaptic excitation of somatosensory MSNs by S1 [13].

Fig 5

Functional properties of S1 neurons and their role in the M1-evoked striatal responses.

(A–B) Spontaneous activity and sensory-induced responses in the S1 CS pathway. (A) Ongoing synaptic and firing activity in a S1 pyramidal neuron (A1) and a MSN (A2) recorded separately from the same experiment. (B) Superimposed (n = 5) whisker-induced (top trace, 40 psi) subthreshold (grey records) and suprathreshold (black records) synaptic responses captured from the S1 cell (B1) and the MSN (B2) shown in (A). The arrows indicate the firing (APs are truncated for commodity) caused by sensory stimuli. The insets show the cell responses to intracellular application of a current pulse of +0.4 nA (bottom trace). The lowest trace in (B2) is the average of 20 successive sensory responses recorded in the MSN. The vertical dashed line indicates the onset of sensory stimuli. (C) Superimposed intracellular responses, obtained in a S1 pyramidal neuron (top records) and in a somatosensory MSN (bottom records) from the same rat, to electrical stimulations of M1 with the indicated intensity. The inset is an expanded average (n = 10 trials) response recorded from the MSN. The grey boxes cover the initial portion of the M1-induced striatal dPSPs that preceded the activation of the S1 neuron. (D) Pooled data showing the latency of striatal dPSPs foot, striatal firing (MSN APs) and firing of S1 neurons (S1 APs) induced by electrical stimulations (20 V) of M1. The number of computed neurons is indicated. * P < 0.05.

Fig 3

Characterization of somatosensory MSNs as a function of their responsiveness to a given whisker stimulation.

(A–C) Air-puff stimuli of same intensity (optimal stimuli, top traces) induced an early negative deflection in the ECoG (black dots), associated with either subthreshold synaptic responses (A, dPSP-responding), suprathreshold synaptic responses (B, AP-responding) or a lack of synaptic response (C, Non-responding) in the simultaneously recorded MSNs (Intra MSN). Three to 5 five successive responses are superimposed. The vertical dashed lines mark the onset of whisker stimuli. The average traces (excluding firing responses) were computed from at least 20 trials. The oblique arrow in (C) indicates the occurrence of occasional temporally-matching sensory responses, which was largely curtailed after averaging of 40 successive trials (crossed arrow). The lowest records are segments of spontaneous activity recorded in the corresponding MSNs. Results shown in (A), (B) and (C) are from three distinct experiments.

We further characterized the functional properties of recorded MSNs on the basis of their responsiveness to sensory stimulations (Fig 1A). The vast majority of MSNs (n = 67 of the 76 tested neurons; ∼87%) responded to air-puff deflections (Fig 3), as attested by the clear-cut average sensory responses which exhibited the typical profile of whisker-evoked depolarizing post-synaptic potentials (dPSPs) (Fig 3A and 3B) [13]. Among these responding neurons (n = 67), 56 displayed only subthreshold dPSPs (dPSP-responding MSNs) regardless of the intensity of the stimulation (Fig 3A). These subthreshold synaptic responses had a mean latency of 22.5 ± 0.6 ms (n = 56 MSNs) and a maximal amplitude of 7.6 ± 0.4 mV. In this group of striatal cells, whisker stimuli were able to generate definite synaptic responses with a probability as high as 0.75 ± 0.4 (n = 56 MSNs), demonstrating the robustness of sensory information propagation from the periphery to the dorsal striatum.

Eleven of the 67 responding MSNs could elicit an AP on the sensory-induced dPSP (AP-responding MSNs), with however a relatively low probability of evoked discharge, from 0.03 to 0.26 (Pfiring = 0.11 ± 0.02, n = 11 MSNs) (Fig 3B). The latency of APs on whisker-evoked sensory responses was 29.5 ± 1.5 ms (n = 11 MSNs). The amplitude of subthreshold responses in these AP-responding MSNs (12.1 ± 1.9 mV, n = 11 MSNs) was larger than that measured in dPSP-responding neurons (P < 0.001) and their probability of occurrence over repeated sensory stimulations was increased (P = 0.89 ± 0.02, n = 11 MSNs; P < 0.05). Voltage firing threshold, as well as Rm values, calculated from dPSPs-responding (-49.7 ± 0.2 mV and 26.8 ± 0.95 MΩ, respectively, n = 56) and AP-responding (-50.5 ± 0.2 mV and 25.4 ± 2.0 MΩ, respectively, n = 11) MSNs, were not significantly different (P > 0.1 for both parameters), indicating that the differences in sensory responsiveness did not result from a dissimilar intrinsic excitability between the two striatal cell populations. However, we found that AP-responding neurons had a more depolarized Vm (-75.4 ± 4 mV, n = 11 AP-responding MSNs versus -79.6 ± 0.6 mV, n = 56 dPSP-responding MSNs; P < 0.01) and larger baseline Vm fluctuations (SD Vm = 4.5 ± 0.6 mV, n = 11 AP-responding MSNs versus 3.4 ± 0.2 mV, n = 56 dPSP-responding MSNs; P < 0.05) (Fig 3A and 3B), suggesting a more intense background synaptic activity in the MSNs that fired AP in response to sensory stimulations.

In the 9 remaining MSNs, no sensory-evoked dPSP could be detected (Non-responding MSNs) (Fig 3C). Although occasional large amplitude synaptic events could occur within the appropriate temporal window after air-puff stimulations (Fig 3C, arrow), they were likely spontaneously generated since they were barely detectable after averaging of a large number of trials (Fig 3C, average trace). The Rm of non-responding neurons (31.1 ± 2.8 MΩ, n = 9) was similar to that of dPSP- and AP-responding cells (P > 0.1). Their Vm (-81.0 ± 1.6 mV) was comparable to that measured from MSNs exhibiting subthreshold responses (P > 0.05), but was more polarized (P < 0.05) and less fluctuating (SD Vm = 3.8 ± 0.4 mV, n = 9 non-responding MSNs; P < 0.01) compared to AP-responding cells (Fig 3C, bottom trace).

Altogether, these findings confirm and extend our previous categorization of somatosensory MSNs [13] as dPSP-, AP- and non-responding neurons for a given sensory stimulus.

Motor cortex-induced synaptic responses in somatosensory MSNs

We next explored, for the first time, the responsiveness of somatosensory MSNs to electrical stimulations (5–25 V, 0.24 Hz) of the orofacial part of M1 known to send CS projections into the somatosensory striatal sector [5, 6, 9, 10, 24]. All tested MSNs (n = 75), previously challenged with sensory stimulations, displayed clear-cut sub- or suprathreshold synaptic depolarizations in response to electrical stimulations of M1.

Striatal cells which responded to M1 electrical stimulations by subthreshold dPSPs (n = 65), regardless of the intensity tested, encompassed the three categories of MSNs previously classified by their responsiveness to whisker stimulations: AP-responding (n = 8), dPSP-responding (n = 49) and non-responding MSNs (n = 8) (Fig 4A and 4B). These cells had a mean Vm and Rm of -79.8 ± 0.5 mV and 26.7 ± 0.9 MΩ (n = 65), respectively, and a voltage firing threshold of -49.6 ± 0.3 mV (n = 62). The onset latency of subthreshold dPSPs, calculated from the maximal stimulus intensity used in each tested cell (15–25V), was 6.7 ± 0.3 ms (n = 65), a value compatible with a monosynaptic CS-induced response in MSNs (see Discussion). The amplitude of M1-induced synaptic responses, which was on average of 15.6 ± 0.8 mV (n = 65) (Fig 4E1 and 4E2), exhibited a relatively large trial-to-trial variability (SD dPSPs = 3.5 ± 0.2 mV, n = 65).

In a smaller subset of MSNs (n = 10 out 75), including dPSP-responding (n = 6) (Fig 4C), AP-responding (n = 3) and non-responding (n = 1) MSNs to sensory stimuli, electrical stimulation of M1 (15–25V) elicited dPSPs that could reach threshold for AP firing (Fig 4C1). The mean Vm of these cells was -78.1 ± 2.2 mV (n = 10), a value similar to that measured for MSNs responding to M1 stimuli by merely subthreshold dPSPs (P > 0.2). Their Rm (28.9 ± 3.1 MΩ), as well as their firing threshold (-50.0 ± 0.6 mV), were also comparable (P > 0.2 for both parameters), indicating that the differential responsiveness of MSNs to M1 stimuli was not caused by a distinct intrinsic excitability. In these MSNs, the suprathreshold responses occurred with a probability of 0.5 ± 0.1 (from 0.005 to 1, n = 10) and were interposed with large amplitude dPSPs (18.9 ± 3.1 mV, n = 9). APs latency on M1-induced synaptic depolarizations was 14.9 ± 1.1 ms (n = 10), a value considerably shorter compared to that measured in response to sensory stimulations (29.5 ±1.5, n = 11 MSNs; P < 0.001).

The M1-evoked responses in MSNs could originate from the activation of postsynaptic glutamatergic receptors through the activation of CS neurons, and/or from feedforward GABAergic synaptic inputs caused by the excitation of striatal interneurons by M1 CS projections [20–22, 24]. It is expected that such synaptic inputs would considerably increase MSNs membrane conductance as the M1-induced response develops. We thus compared in 8 MSNs the current-induced (-0.4 nA, 5–20 ms of duration, n = 20 trials) voltage changes generated at the summit of the cell responses (Fig 4D, grey records) to those obtained 120–160 ms after M1 stimulations, i.e. after termination of synaptic potentials (Fig 4D, black records). While the mean voltage drop obtained in between M1-evoked responses was of 8.1 ± 1.2 mV (from 4.6 to 12.4 mV, n = 8 MSNs) (Fig 4D, black records), it was entirely cancelled in 4 MSNs at the peak of the evoked depolarization or had a net amplitude, after subtracting the M1-induced potential free of current pulses, less than 1.5 mV (n = 4 MSNs) (Fig 4D, insets). These findings demonstrate a dramatic increase in membrane conductance at the time of M1-evoked response, as expected from a mechanism involving the activation of postsynaptic receptors.

Pooling the measurements obtained from MSNs for which at least two intensities of stimulation were tested (n = 56), we found that dPSPs increased in amplitude as the intensity of M1 electrical stimuli was augmented (Fig 4E1), indicating a progressive recruitment of excitatory inputs onto the recorded cells. The amplitude of synaptic responses reached saturation (~16 mV) for intensities ≥ 15 V (Fig 4E2), suggesting that nearly all CS neurons that could be stimulated were already activated from this stimulation intensity (see Discussion).

Activation of S1 pyramidal neurons by M1 stimulations and its relation to striatal responses

The synaptic responses recorded in somatosensory MSNs following electrical stimulation of M1 could primarily result from the intra-cortical synaptic activation of S1 CS neurons, via the reciprocal excitatory connections between the two cortical areas [26-28] and/or from an antidromic activation of S1 neurons projecting to M1. To test for these possibilities, we made intracellular recordings from S1 layer 5 pyramidal neurons (n = 12) (Figs 1B, 5A1 and 5B1), known to provide the main CS source to the corresponding ipsilateral striatal sector [10, 13], and sought to determine whether: 1) S1 neurons were synaptically-excited by whiskers deflection (Figs 1A and 5B1), 2) S1 neurons could be synaptically- and/or antidromically-activated by M1 stimuli (Figs 1B and 5C, upper traces) and, if so, 3) their firing upon M1 stimuli could fully account, or not, for the postsynaptic responses recorded in somatosensory MSNs (Fig 5C and 5D).

S1 layer 5 neurons exhibited oscillatory synaptic depolarizations around a mean Vm of -65.9 ± 1.5 mV (n = 12). This sustained background synaptic activity resulted in spontaneous firing activity in 10 out of the 12 recorded cells (2.4 ± 0.7 Hz) (Fig 5A1), contrasting with the subthreshold synaptic barrage recorded from MSNs in the same experiments (Fig 5A2). S1 neurons were identified as pyramidal cells on the basis of their intrinsic firing profile in response to suprathreshold direct stimulations, classified as either regular spiking (n = 8) (Fig 5B1, inset) or intrinsic bursting (n = 4). In response to sensory stimuli, 11 cortical neurons displayed short latency (15 ± 0.5 ms) synaptic depolarizations (Fig 5B1), which could elicit an AP in the majority of cells with a mean probability of 0.4 ± 0.1 (n = 8) (Fig 5B1). One of the recorded cells exhibited a pure hyperpolarizing postsynaptic potential (15.7 ms of latency) following whiskers deflection. Altogether, these properties are consistent with those already described for whisker-related synaptic responses in layer 5 S1 neurons [35, 38, 39], including pyramidal cells identified as CS neurons [13].

Eleven of the 12 tested S1 pyramidal cells displayed synaptic depolarizations in response to optimal M1 electrical stimulations (15–20 V), with a foot latency of 3.4 ± 0.4 ms. In 3 neurons, the M1-evoked synaptic depolarizations remained subthreshold, with a maximum amplitude of 5.1 ± 1 mV. In the majority of cells (8 out of 12), the M1-evoked synaptic depolarizations reached a maximal amplitude of 8.7 ± 0.9 mV (Fig 5C) and could cause firing with a mean probability of 0.5 ± 0.1 and an AP latency of 9.5 ± 1.3 ms (n = 8 S1 neurons) (Fig 5C and 5D). None of the recorded neurons were antidromically activated, or directly fired, by M1 electrical stimulations in the range of stimuli tested in this study.

As exemplified by the experiment shown in Fig 5C, in which a S1 pyramidal cell (top records) and a somatosensory MSN (bottom records) were recorded in the same animal, the onset of dPSPs evoked in the striatal cell by electrical stimulations of M1 preceded by 4 ms (Fig 5C, inset) the initial discharge of S1 neurons induced by the same stimuli. This temporal advance of M1-induced synaptic responses in MSNs relative to S1 neurons firing was significant over the whole set of tested neurons (MSNs dPSPs: 6.7 ± 0.3 ms, n = 75 versus AP S1 neurons: 9.5 ± 1.3 ms, n = 8 neurons; P < 0.05) (Fig 5D). These findings indicate that M1 stimulations induced a synaptically-mediated activation of S1 pyramidal neurons, instead of direct/antidromic excitation, but that the early synaptic effect of M1 onto somatosensory MSNs was not the result of an intra-cortical recruitment of S1 CS neurons (see Discussion).

Discussion

Properties of responses evoked in somatosensory MSNs by M1 stimulations

It is well established that CS projections from M1 and S1 whisker regions converge with a substantial overlap into the dorsolateral striatum [5, 6, 9, 10, 23]. However, the functional impact of excitatory M1 CS inputs onto the MSNs that process sensory information from the related S1 region remained to be characterized. We demonstrated in the present in vivo work that electrical stimulation of M1 caused depolarizing synaptic responses in all recorded MSNs, including those that did not display a detectable sensory response. MSNs were here recorded without separating them into direct (striatonigral) and indirect (striatopallidal) pathway neurons [40]. Previous works have shown that both types of striatal neurons could respond to the activation of S1 CS inputs, with however possible differences in their response properties [19, 22, 41]. Given the importance of these two projection pathways in basal ganglia function [42], it will be crucial to investigate in the future whether sensorimotor integration differs between these subpopulations of MSNs.

The M1-evoked potentials in somatosensory MSNs had the properties of monosynaptic dPSPs, demonstrating direct and functional excitatory connections from M1 onto ipsilateral striatal neurons. First, striatal responses exhibited the classical shape and duration of postsynaptic potentials, with an onset latency (about 6 ms for optimal stimuli) consistent with a CS monosynaptic delay [43-45] and the conduction velocity of APs in M1 CS axons [46]. Second, the M1-induced responses in MSNs fluctuated in amplitude from trial-to-trial, in agreement with the probabilistic nature of synaptic transmission. Third, the evoked-depolarizations were associated with a dramatic increase in membrane conductance as expected for cortically-generated excitatory synaptic potentials in MSNs [47]. This increase in synaptic conductance was likely aggravated by an additional feedforward GABAergic inhibition of MSNs caused by the activation of striatal interneurons by CS M1 projections [20, 22, 24, 48]. The initial part (the first 3 ms) of M1-evoked dPSPs in MSNs probably resulted from the activation of M1 CS neurons and not from the secondary intra-cortical recruitment of S1 CS neurons by electrical stimulation of M1. This conclusion is supported by: 1) the monosynaptic nature of the synaptic responses (see above), 2) the fact that S1 CS neurons were never antidromically-activated in our experimental conditions, a process that could potentially generate a fast response in MSNs via an axon reflex in S1 neurons projecting to the striatum and, 3) the M1-induced firing latency of CS S1 pyramidal neurons that consistently, and significantly, followed the early part of the synaptic depolarization in MSNs. It is reasonable to think that the subsequent part of the M1-evoked synaptic depolarization in MSNs represents the successive, and overlapping, effects of M1 and S1 CS neurons, the latter being monosynaptically-activated by the intra-cortical excitatory projections from the whisker region of M1 [26]. Such a process is in agreement with the short-latency (~3 ms) synaptic depolarizations recorded in S1 neurons upon M1 stimulations, which were effective in triggering APs in most of the cells with a high probability. Notably, firing latency of S1 neurons in response to M1 stimulations (~9 ms) shortly preceded the peak depolarization in MSNs, suggesting that a cumulative effect of converging, and temporally correlated, M1 and S1 CS inputs could intensify the fire rate of MSNs.

We found that striatal responsiveness to ipsilateral M1 stimulations was more reliable and robust compared to that consecutive to sensory stimuli. This could be due to multiple, and non-exclusive, mechanisms. First, intra-cortical electrical stimulations are likely to produce a massive recruitment of CS neurons, including a direct activation of M1 CS neurons and the subsequent synaptic excitation of S1 CS neurons, whereas sensory stimuli are conveyed to S1 CS neurons via a polysynaptic, and thus less reliable, network. Second, the ipsilateral CS projections from M1 into the dorsolateral striatum are more extensive than those originating from S1 [9], and thus more prone to contact a larger number of MSNs. And third, the synaptic efficacy of ipsilateral M1 CS neurons onto somatosensory MSNs seems to be greater compared to S1 CS connections, at least in the mouse dorsolateral striatum in vitro [22]. It has been previously reported that contralateral CS projections from S1 are much sparser than those originating from M1 [17, 49]. One can thus expect that the differences we observed between striatal responsiveness to ipsilateral S1 and M1 inputs be even more pronounced during activation of the contralateral cortices.

Functional considerations

We demonstrated that MSNs located in the dorsolateral part of the striatum integrate converging excitatory synaptic inputs originating from whisker-related sensory and motor cortical areas, which are anatomically, and functionally, interconnected. These cortical areas are nearly co-activated during a sensorimotor task involving the whisker system [50], providing an optimal pattern of synaptic drive for MSNs which are known to operate as coincidence detector of excitatory inputs [30, 31] because of their peculiar intrinsic electrical properties. Indeed, MSNs exhibit a powerful inwardly rectifying potassium selective current, responsible for most of their resting membrane conductance and highly polarized Vm, and for a pronounced inward rectification during membrane hyperpolarization [31, 32, 37, 45]. As a consequence, the voltage-dependent activation of this potassium current is mainly accountable for the relatively low Rm and τm of MSNs when measured from the resting membrane potential. In addition, the presence of this potassium conductance in the dendritic tree of striatal neurons [51] likely causes a dramatic increase of the dendritic electrotonic length and, consequently, negatively alters the amplitude and shape of propagated synaptic depolarizations. This intrinsic control of membrane excitability and synaptic responses of MSNs acts in synergy with a feedforward inhibition caused by the activation of striatal fast-spiking GABAergic interneurons by CS afferents [19-22]. The firing of striatal interneurons produces a robust chloride-dependent post-synaptic conductance able to block the generation of APs in MSNs [52] and to shunt incoming excitatory synaptic inputs [13, 20].

According to these multiple mechanisms in the inhibitory control of MSNs excitability and responsiveness, it is hypothesized that only a large number of correlated excitatory inputs may be able to significantly depolarize the striatal membrane. Indeed, cortical synchronization is required to impede the feedforward inhibition of MSNs and to deactivate the inwardly rectifying potassium current, leading to an explosive change in the electrotonic structure of MSNs and an increased probability for dPSPs to reach voltage firing threshold [30, 31]. Our findings support such a dynamic scenario for striatal activation upon stimulation of M1. Hence, we have shown that electrical activation of M1 CS neurons engendered an initial membrane depolarization in MSNs, which is further, and rapidly, amplified by the fast intra-cortical recruitment of S1 CS neurons [26], which have dense overlapping inputs with M1 CS projections within the same striatal sector [5, 6, 9] and, as attested by the present study, onto the same MSNs. This augmented excitatory synaptic drive onto MSNs, which could also result from a fast sequential activation from S1 to M1 during sensory stimulation [53], may thus become sufficient to overcome the shunting effect caused by intrinsic membrane properties and the recruitment of GABAergic synaptic inputs, allowing MSNs to depolarize up to their firing threshold. This process is expected to induce a sustained synaptic activity and an increased firing rate in MSNs when the two converging cortical areas are concomitantly and repeatedly co-activated [41], ultimately leading to the optimization of striatum-related sensorimotor behaviors.

19 Nov 2019

PONE-D-19-30235

Converging sensory and motor cortical inputs onto the same striatal neurons: an in vivo intracellular investigation

PLOS ONE

Dear Pr. Charpier,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by Jan 03 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Jean-Pierre Mothet, Ph.D

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The paper by Charpier et al describes intracellular recordings from neurons in the dorsolateral striatum of anesthetized rats. The authors show that striatal projection neurons (MSNs) responding to contralateral whisker deflections also respond to electrical stimulation of the ipsilateral motor cortex, thus suggesting that they may be involved in sensorimotor processes. The authors also confirm that these responses to electrical stimulation are monosynaptic, based on the onset latencies in striatal and cortical neurons. The manuscript is interesting, timely, and well-written, and the data presented are convincing. I have just a few comments regarding additional information to be added and the relationship to previously published studies.

Sensorimotor targeting: The targeting of MSNs by sensory and motor cortices, even in overlapping regions, was shown in earlier papers using other methods: (Wall…Kreitzer, Neuron (2013), Reig & Silberberg, Cerebral cortex (2016), ). Moreover, it was shown that both types of MSNs (direct and indirect pathway neurons) receive these inputs. The issue of selective targeting of direct and indirect pathways should be addressed, especially as the data in the paper showing that a vast majority of recorded MSNs respond to both types of cortical inputs. These data strongly suggest that both MSN types respond to both types of cortical stimulation.

Was there any attempt to explain the diversity in response by staining for D1/D2/SP/ENK to determine dMSN or iMSN identity of recorded neurons?

Another comparison would be to assess the difference between S1 and M1 responses under the same conditions. Currently, the stimulation of S1 is done indirectly by whisker deflection while M1 is activated electrically. Do the authors have comparable data showing MSN responses to electrical stimulation in S1 compared to M1? If such electrical stimuli are given with same parameters, one could conclude whether MSNs in dorsolateral striatum receive stronger or weaker inputs from these regions.

Bilateral projections: previous studies (Brown et al., Neuroscience (1996); Lei et al, J.Neuroscience (2004); Reig and Silberberg, cerebral cortex (2016)) have shown that one of the main differences between corticostriatal projections from primary sensory and motor cortices is the density of contralateral innervation. Projections to contralateral striatum from primary sensory cortex are much sparser than those from motor cortex and other more frontal regions. Do the authors have any data regarding bilateral whisker and/or motor cortex stimulation? One interesting question that could be answered using the experimental setup is the comparison between ipsilateral and contralateral electrical and sensory stimulation. The prediction, based on previous data is that such differences will be more pronounced in the sensory responses than motor stimulation.

Feedforward inhibition: the role of inhibitory connections within the striatum (mainly striatal interneurons but also MSN collaterals) was shown to be instrumental in shaping both spontaneous activity and whisker evoked responses (Reig & Silberberg, neuron 2014). Moreover, inputs to striatal interneurons (in particular FS interneurons) were shown to arise from various cortical regions, including sensory, motor and others. This suggests that the responses described in this study (both sensory and electrical stimuli) are shaped by an inhibitory component as well. This issue should be addressed, at least in discussing the cortical inputs, and if possible also in presentation of experimental data. Such an inhibitory component could be revealed by depolarizing the MSNs recorded intracellularly.

Minor:

The authors show that 87% of recorded neurons respond to whisker stimulation. This is significantly higher than their earlier paper (2011) that showed 55% of responding neurons. What could be the reason for this difference in responsiveness? Was there any difference in the recording region or stimulus presentation?

The order of authors is different in the paper heading and in the cover pages (2nd and 3rd authors).

Results page 8: “…relatively weak Rm…”. Better to use high/low Rm.

Reviewer #2: In this manuscript, Charpier et al., investigated if projection neurons in the dorsolateral region of the striatum respond to both cortical input from functionally-related sensory and cortical regions. A key aspect of this study is that it is conducted in vivo, using intracellular electrophysiological recordings of striatal projection neurons (76 in total, which is a large number, considering the technical challenge of such approach), thus allowing to monitor supra et subthreshold changes in the membrane potential, in response to cortical/whisker stimulation. First, the authors reported that a large proportion of the striatal neurons recorded respond to whisker stimulation either through sub-threshold or supra-threshold depolarizing potentials. EcoG and intracell recordings in the barrel cortex provide strong evidence that whisker-evoked responses in the striatum were cortically driven. Then the authors showed that the same neurons also responded to electrical stimulation of M1 and provided evidence that these responses were mono-synaptically driven (rather than secondary to intra-cortical stimulation of S1). A key experiment was to record simultaneously S1 and striatal neurons while stimulating M1. The authors reported that striatal neurons responded with a shorter latency than S1 to M1 stimulation.

Altogether the authors provide excellent and rare in vivo electrophysiolgical data that demonstrate that the same striatal projection neurons can respond to sensory and motor cortical input from a functionally-related brain region. This knowledge is critical to understand the function of this brain region.

The study was carefully and rigorously conducted and the results which are in agreement with previous anatomical studies, obtained in vivo, constitute an important piece of information that should be shared with the community.

I have a few minor suggestions that the author may consider to address to improve their manuscript.

1) I was surprised by the opening sentence of the manuscript and the references cited. The authors mentioned the role of the striatum production and optimization of context-dependant behavior (isn't this the purpose of the whole nervous system?) and cite relatively “old” reviews of Ann Graybiel, which postulate a role of the striatum in motor control akin to as sort of traffic light signaling the beginning and end of automatized motor sequences, and thus overlook the moment to moment sensory representation associated with any behavior.

In my opinion, there are studies from the Silberberg and West labs that could have been provided a better context for this study to emphasis the importance of sensorimotor integration in the striatum (see also Robbe 2018 for a recent review on this topic)

2) A recent detailed anatomical tracing study nicely illustrate that vibrissal M1 and S1 send convergent input into specific region of the striatum (Hooks et al., 2018) and that this convergence does not occur for functionnaly distinct cortical area. This work should be cited

3) In the introduction, the authors suggested that sensory information is refined in the striatum. But from the description provided one could very well conclude that the information is degraded in the striatum. In fact both interpretation (refinement vs degradation or even filtering) could be correct. Maybe the author should use a more neutral word than “refinement” ?

4) When reading the result section for the first time it was not clear to me if the neurons recorded while stimulating M1 where the same one than those for which whisker responsiveness had been tested. The authors should make it very clear in the text (p10, after subsection title) that these are the same neurons.

5) The authors recorded 76 projection neurons. Most likely half of them where striatopallidal neurons while the other half. There has been anatomical evidence that striatonigral (direct-pathway) neurons are more strongly connected with the sensory cortex while striatopallidal (indirect-pathway) neurons are more connected with the motor cortex. It would have been nice if the authors would have speculated on the possible dichotomy of responses (or absence of) between these two populations.

6) I am not sure the authors should use a t-test to compare latencies between responses of striatal and S1 neurons (figure 5). The number of S1 is small (n=8) and I doubt normality can be evaluate. A non parametric test would be more appropriate, even if I have no doubt the result will be the same.

David Robbe

Reviewer #3: In this manuscript, the authors examine and characterize the cortico-striatal inputs from M1 and S1 to MSNs of the dorsolateral striatum (DLS) during whisker-induced sensory processing. Their major aim is to evaluate whether M1 and S1 inputs may converge on the very same MSN in this striatal area. The authors used an approach combining S1 and MSN intracellular recording and S1 electrocorticogram in vivo in anaesthetized rats subjected to whisker stimulation. The data showed an heterogenous response of DLS-somatosensory MSN to whisker sensory stimuli, and indirectly demonstrated a short-lasting direct monosynaptic M1 to MSN DLS connection. The data also demonstrated that sensory evoked whisker stimuli elicit responses in MSNs which are also responsive to M1 input, confirming their hypothesis of a M1 - S1 input convergence on individual MSNs in the DLS.

The question addressed in this paper is of high interest in the field to more deeply understand the motor and somatosensory input processing in the striatum. The experimental design providing in vivo recordings in DLS and the S1 cortex in a paradigm of sensory stimulation is highly adequate and perfectly conducted, and, the manuscript is clearly written with figures and text describing the results accurately. It delivers data that reinforce the knowledge on striatal computation of diverse cortical inputs. The following minor concerns must be considered.

The authors acknowledged the fact that CS inputs activation in their paradigms may or should result in both direct excitatory activation and feedforward inhibition of MSN, through activation of interneurons (see the discussion on the increase in synaptic conductance). However, no experiments were performed to evaluate the contribution of this feedforward inhibition in the protocols used in this study. Intrastriatal pharmacological GABAergic blockade could help to identify this contribution and this could be therefore clarified.

The data examining the membrane conductance of MSNs during the M1-induced response as illustrated in fig 4D should be more clearly presented. Indeed, the negative current pulse has a duration of 5ms in case of conjunction with the M1-induced responses (bottom traces) whilst it is of hundreds of ms in absence of stimulation (top trace). Moreover, calculation and comparison of conductances in both conditions should be provided.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: David Robbe

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: reviewer comments.pdf

Click here for additional data file.

19 Dec 2019

Reviewer # 1

We thank the reviewer for his/her positive assessment of this work and appreciate the suggestions, which have been carefully taken into account.

Major points

1. “Sensorimotor targeting: The targeting of MSNs by sensory and motor cortices, even in overlapping regions, was shown in earlier papers using other methods: (Wall…Kreitzer, Neuron (2013), Reig & Silberberg, Cerebral cortex (2016)). Moreover, it was shown that both types of MSNs (direct and indirect pathway neurons) receive these inputs. The issue of selective targeting of direct and indirect pathways should be addressed, especially as the data in the paper showing that a vast majority of recorded MSNs respond to both types of cortical inputs. These data strongly suggest that both MSN types respond to both types of cortical stimulation.” “Was there any attempt to explain the diversity in response by staining for D1/D2/SP/ENK to determine dMSN or iMSN identity of recorded neurons?”

Answer: Our study, at this stage, was specifically designed to examine the CS sensorimotor convergences onto single MSNs, regardless their biochemical content and output connectivity. As pointed out by the Reviewer, it is very likely that the two subpopulations of MSNs (striatonigral and striatopallidal) were indeed included in the data set given the elevated number of recorded MSNs (n = 75). Our data showing that all recorded cells could respond to M1 stimuli and that the three classes of somatosensory MSNs, differentially responding to S1 inputs, did not exhibit specific intrinsic membrane properties (as it is expected for direct and indirect pathway striatal neurons) do not permit to conclude on a possible distinct processing of S1 and M1 inputs by the two subpopulations of MSNs. However, to follow the Referee’s suggestion, this crucial issue is now discussed (page 14, lines 9-14).

2. “Currently, the stimulation of S1 is done indirectly by whisker stimulation while M1 is activated electrically. Do the authors have comparable data showing MSN responses to electrical stimulation in S1 compared to M1? If such electrical stimuli are given with same parameters, one could conclude whether MSNs in dorsolateral striatum receive stronger or weaker inputs from these regions.”

Answer: Contralateral air-puff stimulations were here preferred to electrical stimulations of the S1 cortex to activate somatosensory CS projections in a more physiological way. Moreover, this approach allowed assessing the impact of M1 CS inputs on the three classes of somatosensory MSNs we previously identified using natural sensory stimuli, i.e. dPSP-responding, AP-responding and non-responding MSNs. It is likely that electrical stimulations of S1, causing a massive non-specific activation of S1 CS neurons, would have precluded the distinction between the three types of somatosensory MSNs.

Although our experimental design did not permit to compare the strength of S1 and M1 inputs onto striatal cells (a point which is discussed on page 15, lines 3-11), it led us to demonstrate that all the functional subpopulations of somatosensory MSNs could indeed be excited by CS motor inputs.

3. “Bilateral projections: previous studies (Brown et al., Neuroscience (1996); Lei et al, J.Neuroscience (2004); Reig and Silberberg, cerebral cortex (2016)) have shown that one of the main differences between corticostriatal projections from primary sensory and motor cortices is the density of contralateral innervation. Projections to contralateral striatum from primary sensory cortex are much sparser than those from motor cortex and other more frontal regions. Do the authors have any data regarding bilateral whisker and/or motor cortex stimulation? One interesting question that could be answered using the experimental setup is the comparison between ipsilateral and contralateral electrical and sensory stimulation. The prediction, based on previous data is that such differences will be more pronounced in the sensory responses than motor stimulation.”

Answer: The differential impact of sensory and motor contralateral CS innervation onto striatal responsiveness is indeed a very interesting question, although not directly addressed in the present study. Comparing bilateral sensory and motor integration at the level of individual MSNs would require an extensive set of additional experiments, which are beyond the scope of this work. This important issue is now exposed as a forward-looking view point in the Discussion and appropriate literature is cited (page 15, lines 11-14).

4. “Feedforward inhibition: the role of inhibitory connections within the striatum (mainly striatal interneurons but also MSN collaterals) was shown to be instrumental in shaping both spontaneous activity and whisker evoked responses (Reig & Silberberg, neuron 2014). Moreover, inputs to striatal interneurons (in particular FS interneurons) were shown to arise from various cortical regions, including sensory, motor and others. This suggests that the responses described in this study (both sensory and electrical stimuli) are shaped by an inhibitory component as well. This issue should be addressed, at least in discussing the cortical inputs, and if possible also in presentation of experimental data. Such an inhibitory component could be revealed by depolarizing the MSNs recorded intracellularly.”

Answer: We agree with the Referee that feedforward and collateral recurrent inhibition of MSNs, stimulated by CS inputs, is of crucial importance to shape striatal responsiveness, as it may both participate to a conductance increase in MSNs and tune their synaptic and firing responses. We have explored this issue in our previous study characterizing the sensory-evoked responses in somatosensory MSNs (Pidoux et al., 2011). By the use of KCl-filled microelectrodes, displacing the equilibrium potential of chloride in MSNs, we found that sensory stimuli, in such condition, could evoke larger synaptic depolarizations in MSNs resulting in an increased relative number of AP-responding neurons, indicating the presence of a powerful Cl--dependent conductance. We also showed, by means of extracellular recordings of striatal GABAergic interneurons, that contralateral whisker sensory stimuli could induce repetitive discharges in striatal interneurons. Overall, this demonstrates that CS neurons activation by sensory stimuli resulted in a potent feedforward inhibition of MSNs intermingled with the direct CS-induced glutamatergic synaptic excitation. It is very likely that a similar process occurs in the case of M1 stimulations through the activation of S1 CS neurons (as previously demonstrated) but also directly from the activation of M1 CS neurons, which are known to connect striatal interneurons in the somatosensory striatum (Ramanathan et al., 2002; see also Lee et al., 2019). Thus, as suggested by the Referee, this hypothesis is now developed in the Discussion (page 15, lines 29-37 and page 16, lines 1-2).

Minor points

1. “The authors show that 87% of recorded neurons respond to whisker stimulation. This is significantly higher than their earlier paper (2011) that showed 55% of responding neurons. What could be the reason for this difference in responsiveness? Was there any difference in the recording region or stimulus presentation?”

Answer: The proportion of MSNs responding to sensory stimulations is indeed more elevated in this study compared to our previous investigation. However, we confirmed here the three types of responses and the percentage of AP-responding cells was found to be similar between the two studies, further demonstrating a cell-dependent filtering of sensory information in the striatum. The partial discrepancy between the two investigations does not reflect different experimental conditions or neuronal properties, including anesthesia procedures, sites of recordings, stimulus features and intrinsic membrane properties. Instead, it may result from the higher number of neurons recorded in the present study: 76 vs 49 in the earlier one, which could have refined the statistical assessment of the three neuronal populations. Moreover, because CS glutamatergic connections display various forms of activity-dependent plasticity, it is plausible that the experimental groups used in these two studies were subject to various levels of whisker-related behavioral learning, differentially affecting the CS connectivity. Although speculative, this hypothesis deserves to be tested in the future.

2. “The order of authors is different in the paper heading and in the cover pages (2nd and 3rd authors).”

Answer: We thank the Reviewer for having warned us on this. Séverine Mahon is the last author; this has been corrected.

3. “Results page 8: “…relatively weak Rm…”. Better to use high/low Rm.”

Answer: “Weak” has been replaced by “low” (page 8, line 11).

Reviewer # 2

We are grateful to Dr David Robbe for his constructive and thoughtful comments. The revised version has been amended following his suggestions.

Minor suggestions

1. “I was surprised by the opening sentence of the manuscript and the references cited. The authors mentioned the role of the striatum production and optimization of context-dependant behavior (isn't this the purpose of the whole nervous system?) and cite relatively “old” reviews of Ann Graybiel, which postulate a role of the striatum in motor control akin to as sort of traffic light signaling the beginning and end of automatized motor sequences, and thus overlook the moment to moment sensory representation associated with any behavior. In my opinion, there are studies from the Silberberg and West labs that could have been provided a better context for this study to emphasis the importance of sensorimotor integration in the striatum (see also Robbe 2018 for a recent review on this topic)”.

Answer: We agree with the Reviewer that the role of the striatum in sensory integration was not sufficiently emphasized in the previous Introduction. This part of the manuscript has been rewritten and appropriate literature has been cited (page 3, lines 2-9).

2. “A recent detailed anatomical tracing study nicely illustrate that vibrissal M1 and S1 send convergent input into specific region of the striatum (Hooks et al., 2018) and that this convergence does not occur for functionally distinct cortical area. This work should be cited”

Answer: We thank the Referee for having pointed out this important piece of recent literature, which is now cited in the Introduction (page 3, lines 7-9) and throughout the manuscript.

3. “In the introduction, the authors suggested that sensory information is refined in the striatum. But from the description provided one could very well conclude that the information is degraded in the striatum. In fact both interpretation (refinement vs degradation or even filtering) could be correct. Maybe the author should use a more neutral word than “refinement” ?”

Answer: The term “refinement”, used in our previous article [Pidoux et al., J Physiol. (London) 2011] to qualify the variability of sensory responsiveness among MSNs, was chosen to stress that only a fraction of striatal neurons could be excited by a given whisker stimulus, a process that may allow selecting pertinent cortical information at the striatal level. As suggested by the referee, and to stay close to the original concept, the word “selection” is now used (page 3, line 18).

4. “When reading the result section for the first time it was not clear to me if the neurons recorded while stimulating M1 where the same one than those for which whisker responsiveness had been tested. The authors should make it very clear in the text (p10, after subsection title) that these are the same neurons.”

Answer: Indeed, out of the 76 neurons intracellularly recorded in this study, 75 were challenged by both sensory and motor cortex stimulations (as shown in Fig. 4). To avoid any ambiguity, this is now clearly stated in the Results section (page 10, line 21-22).

5. “The authors recorded 76 projection neurons. Most likely half of them where striatopallidal neurons while the other half. There has been anatomical evidence that striatonigral (direct-pathway) neurons are more strongly connected with the sensory cortex while striatopallidal (indirect-pathway) neurons are more connected with the motor cortex. It would have been nice if the authors would have speculated on the possible dichotomy of responses (or absence of) between these two populations.”

Answer: Our study, at this stage, was specifically designed to examine the CS sensorimotor convergences onto single MSNs, regardless their biochemical content and output connectivity. As pointed out by the Reviewer, it is very likely that the two subpopulations of MSNs (striatonigral and striatopallidal) were indeed included in the data set given the elevated number of recorded MSNs (n = 75). Our data showing that all recorded cells could respond to M1 stimuli and that the three classes of somatosensory MSNs, differentially responding to S1 inputs, did not exhibit specific intrinsic membrane properties (as it is expected for direct and indirect pathway striatal neurons) do not permit to conclude on a possible distinct processing of S1 and M1 inputs by the two subpopulations of MSNs. However, to follow the Referee’s suggestion, this crucial issue is now discussed (page 14, lines 9-14).

6. “I am not sure the authors should use a t-test to compare latencies between responses of striatal and S1 neurons (figure 5). The number of S1 is small (n=8) and I doubt normality can be evaluate. A non-parametric test would be more appropriate, even if I have no doubt the result will be the same.”

Answer: To follow the pertinent suggestion of the Referee, we have now applied a non-parametric test (Mann-Whitney Rank Sum test) and confirmed the significant difference between the response latency of MSNs and S1 neurons to M1 stimulations (P = 0.03). The new test and corresponding statistical value are indicated (page 7, lines 21-22 and page 13, lines 4 and 30), and the figure 5 has been modified accordingly.

Reviewer # 3

Referee’ statement. “The question addressed in this paper is of high interest in the field to more deeply understand the motor and somatosensory input processing in the striatum. The experimental design providing in vivo recordings in DLS and the S1 cortex in a paradigm of sensory stimulation is highly adequate and perfectly conducted, and, the manuscript is clearly written with figures and text describing the results accurately. It delivers data that reinforce the knowledge on striatal computation of diverse cortical inputs.”

Comment: We deeply thank the Referee for his/her very positive comments on both our experimental strategy and findings. We are also grateful for his/her pertinent suggestions, which have all been taken into account. We have indeed done our best to design an in vivo paradigm allowing both the assessment of M1-induced responses in somatosensory MSNs and the estimate of the relative impact of M1 cortex and the related S1 region on the same striatal neurons. We hope that this study will stimulate further studies devoted to the elucidation of the functional effects of converging CS inputs onto single MSNs and output networks of basal ganglia.

Minor concerns

1. “The authors acknowledged the fact that CS inputs activation in their paradigms may or should result in both direct excitatory activation and feedforward inhibition of MSN, through activation of interneurons (see the discussion on the increase in synaptic conductance). However, no experiments were performed to evaluate the contribution of this feedforward inhibition in the protocols used in this study. Intrastriatal pharmacological GABAergic blockade could help to identify this contribution and this could be therefore clarified.”

Answer: We fully agree with the Referee that the role of feedforward inhibition of MSNs through CS-induced excitation of striatal interneurons is of crucial importance. It may both participate to a conductance increase in MSNs and tune their synaptic and firing responses. We have explored this issue in our previous study characterizing the sensory-evoked responses in somatosensory MSNs (Pidoux et al., 2011). By the use of KCl-filled microelectrodes, displacing the equilibrium potential of chloride in MSNs, we found that sensory stimuli, in such condition, could evoke larger synaptic depolarizations in MSNs resulting in an increased relative number of AP-responding neurons, indicating the presence of a powerful Cl--dependent conductance. We also showed, by means of extracellular recordings of striatal GABAergic interneurons, that contralateral whisker sensory stimuli could induce repetitive discharges in striatal interneurons. Overall, this demonstrates that CS neurons activation by sensory stimuli resulted in a potent feedforward inhibition of MSNs intermingled with the direct CS-induced glutamatergic synaptic excitation. It is very likely that a similar process occurs in the case of M1 stimulations through the activation of S1 CS neurons (as previously demonstrated) but also directly from the activation of M1 CS neurons, which are known to connect striatal interneurons in the somatosensory striatum (Ramanathan et al., 2002; see also Lee et al., 2019). Thus, as suggested by the Referee, this hypothesis is now developed in the Discussion (page 15, lines 29-37 and page 16, lines 1-2).

2. “The data examining the membrane conductance of MSNs during the M1-induced response as illustrated in fig 4D should be more clearly presented. Indeed, the negative current pulse has a duration of 5ms in case of conjunction with the M1-induced responses (bottom traces) whilst it is of hundreds of ms in absence of stimulation (top trace). Moreover, calculation and comparison of conductances in both conditions should be provided”

Answer: The Referee is right. It could indeed be problematic to compare conductance changes from current pulses having different duration, even if in the present case the collapse of current-induced Vm drop (�Vm) demonstrates a dramatic increase in membrane conductance at the peak of synaptic depolarizations. However, to follow the pertinent remark of the Referee, we have now compared the �Vm evoked by the same current pulses (5-20 ms of duration; -0.4 nA, n = 8 cells) at the summit of synaptic potentials and 120-160 ms after M1 stimulations, i.e. after termination of synaptic responses. However, since the duration of current pulses was mostly shorter than the capacitive charge time of the MSN membrane, we have calculated the �Vm instead of Rm values. This is now illustrated in the new panel D of Figure 4 and the corresponding data, demonstrating a breakdown in the membrane resistivity during M1 stimulations, are given in the Results section (page 11, lines 34-37 and page 12, lines 1-5). The figure legend has been modified accordingly (page 11, lines 5-9).

Submitted filename: Reviewers responses-sm-sc-17-12-sm.docx

Click here for additional data file.

13 Jan 2020

Converging sensory and motor cortical inputs onto the same striatal neurons: an in vivo intracellular investigation

PONE-D-19-30235R1

Dear Dr. Charpier,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Jean-Pierre Mothet, Ph.D

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors have addressed my comments in the discussion section of the paper. No additional data has been added but I hope will be done in future studies.

Reviewer #2: (No Response)

Reviewer #3: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: David Robbe

Reviewer #3: No

21 Jan 2020

PONE-D-19-30235R1

Converging sensory and motor cortical inputs onto the same striatal neurons: an in vivo intracellular investigation

Dear Dr. Charpier:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr Jean-Pierre Mothet

Academic Editor

PLOS ONE