Control of sustained attention and impulsivity by Gq-protein signalling in parvalbumin interneurons of the anterior cingulate cortex

The anterior cingulate cortex (ACC) has been implicated in attention deficit hyperactivity disorder (ADHD). More specifically, an appropriate balance of excitatory and inhibitory activity in the ACC may be critical for the control of impulsivity, hyperactivity, and sustained attention which are centrally affected in ADHD. Hence, pharmacological augmentation of parvalbumin- (PV) or somatostatin-positive (Sst) inhibitory ACC interneurons could be a potential treatment strategy. We, therefore, tested whether stimulation of Gq-protein-coupled receptors (GqPCRs) in these interneurons could improve attention or impulsivity assessed with the 5-choice-serial reaction-time task in male mice. When challenging impulse control behaviourally or pharmacologically, activation of the chemogenetic GqPCR hM3Dq in ACC PV-cells caused a selective decrease of active erroneous—i.e. incorrect and premature—responses, indicating improved attentional and impulse control. When challenging attention, in contrast, omissions were increased, albeit without extension of reward latencies or decreases of attentional accuracy. These effects largely resembled those of the ADHD medication atomoxetine. Additionally, they were mostly independent of each other within individual animals. GqPCR activation in ACC PV-cells also reduced hyperactivity. In contrast, if hM3Dq was activated in Sst-interneurons, no improvement of impulse control was observed, and a reduction of incorrect responses was only induced at high agonist levels and accompanied by reduced motivational drive. These results suggest that the activation of GqPCRs expressed specifically in PV-cells of the ACC may be a viable strategy to improve certain aspects of sustained attention, impulsivity and hyperactivity in ADHD.


Surgery for viral transduction
Once mice had reached at least stage 3 of 5-choice-serial-reaction-time task training (see below), they were assigned to the control or the DREADD group, based on their performance over the first 3 days of this stage as a measure of counter-balancing. Animals were anaesthetized using isoflurane (AbbVie, DE), received s.c. injections of analgesics Surgery with chronic electrophysiological implantations and awake-state recordings during the 5-CSRTT and LMA testing Implantation surgeries were timed and conducted like normal surgeries for viral transduction (see above), except that 6 single polyimide-insulated tungsten wires of 50 µm diameter (WireTronic Inc., CA, US) and two surface electrodes (scull screws; 1.2 mm diameter, Precision Technologies, UK) were implanted as chronic field electrodes immediately after the last AAV infusion into the ACC (Cg1/2). Coordinates for the various regions were as follows with reference to Bregma (in mm) and DV measured from pia: Cg1-anterior (Cg1a; AP +1.8- After recovery from surgery, training was continued in the 5-CSRTT until the final baseline stage (5

Surgery for tetrode-recordings under anaesthesia
Male PV-Cre mice that had undergone surgery for viral transduction (see above) and were transduced with either an hM4Di-expressing AAV (N = 5) or no AAV (N = 3) in the ACC. They were used for tetrode recordings under terminal anaesthesia several months later. The mice were anaesthetized with isoflurane as described above (see Surgery) to temporally implant a microdrive (Axona Ltd., UK) holding four movable tetrodes for extracellular recordings, whereby two tetrodes were placed in each hemisphere. One tetrode was constructed from four 12 µm-diameter tungsten wires (California FineWire Company, US), which were twisted and fused together by heating the insulation. The tetrodes were held in a microdrive assembly (Axona Ltd) that allowed them to be lowered or raised individually. The impedance of the tetrodes was reduced to 300-500 kΩ by gold-plating of the wire tips before implantation. The craniotomy was made bilaterally either over the anterior (AP +1.8, ML 0.25) or the posterior part of the ACC (AP +0.7, ML 0.3). A stainless-steel screw was implanted into the bone above the cerebellum serving as reference and ground signals. The dura mater above the regions of interest was removed, and the tetrodes were inserted at the centre of the craniotomies down to 1.3 mm DV into the brain. The ground wires of the microdrive were connected to the ground screw.
The recording was started when the animal was stably breathing without any signs of pain perception at an isoflurane concentration of 0.8-1.0% and spiking activity was visible at the chosen recording site in Cg1. Firstly, a baseline of approx. 10 min was recorded, followed by an i.v. injection of saline vehicle into the tail vein and a further 10 min of recording.
Subsequently, 10 mg/kg CNO was applied i.v. and the signal was recorded for another 120 min (or shorter in one control mouse). The neural signals were recorded using the Omniplex Neural Data Acquisition System (Plexon Inc., US) at a sampling rate of 40 kHz and a gain of 5000.

Behavioural testing
Behavioural testing was done blind to the subgroup identity of the mice. Mice started training in the 5-choice-serial-reaction-time task (5-CSRTT) at 2-3 months of age and were kept under food-restriction at 85-95% of their average free-feeding weight which was measured (1:1000, Thermo Fisher Scientific, US) in PBS/ 0.3% Triton™-X100/ 1% NGS for 2 h in the dark followed by washing with PBS for 10 min. Finally, brain sections were stained with DAPI, mounted on object slides and stored at 4°C until further use. Images were taken using the Leica DM6B epifluorescence microscope with a 20x objective. The co-localization of the DREADD marker mCherry and the PV or SST antibody was quantified manually using the Image J distribution Fiji 6 . For the determination of cFos-expression in hM3Dq-mCherrypositive and -negative ACC cells in Sst-Cre mice, a similar protocol was used, except for the following deviations: blocking solution contained 0.3% Triton™-X100 and 5% NGS in PBS, carrier solution for the primary antibody contained 1% normal horse serum (NHS) and a polyclonal rabbit anti-cFos antibody (9F6; Cell Signaling Technology #2250; 1:1000) in addition to 0.3% Triton in PBS, and the carrier solution for the secondary antibody contained 1% NHS and 0.1% Triton in PBS. Images were acquired with a 5x objective at a Leica DM6 epifluorescence microscope. Counting was done blind to prior treatment.

Analysis of tetrode recordings in PV-hM4Di mice
The recorded signal was high-pass filtered with a 4-pole Butterworth filter and a cut-off frequency of 400 Hz to receive the multi-unit spiking activity (MUA). Peaks were detected as spikes when crossing a negative threshold of four standard deviations below the mean signal of the total recording time. The number of spikes was analysed in 10 min bins consisting of the first (baseline) bin, the second bin (activity after vehicle injection) and 10-12 further bins following CNO-injection.

Analysis of LFP recordings in PV-Gq mice
Local power of field potential oscillations was computed from LFP recordings as described by us before for the data from the LMA test 7 and for operant testing 8 , respectively, using MatLab (MathWorks, US). In brief, datathat had been band-pass filtered during acquisition between 0.1-300 Hz using Open-EPhyswas downsampled from 20 to 1 kHz, detrended using the   Note that, in the PV-Gq (behaviour) and Sst-Gq cohorts, some further chemogenetic 5-CSRTT experiments have been tried towards the end of the experimental schedule, mostly only in a subset of animals, but have not been further pursued due to technical failures or redundancy with previous experiments. Also, in the PV-Gq(electrophysiology) cohort, two challenges without chemogenetic modulation were conducted before the series of chemogenetic experiments started. Remarks: * Two values are stated referring to two parts of the cohort that were run separately due to logistic limitations. ** A Ro-challenge was conducted already 7d before the actual 3-session sequence of this experiment to avoid increases in variance between the first and second Ro-application of the actual experiment. *** Note, that this 9s-ITI challenge was conducted either 1 or 2 wks after the last previous 7s-ITI challenge, which is why the ITI was increased.
0.8s-SD challenge with 0.7 mg/kg CNO because they did not participate in the task (1-2 number corrects) and from one further Gq-mouse data was lost due to a technical error. The

Supplementary Table 5. CNO-induced 5-CSRTT behaviour in the PV-hM4Di cohort.
Results of repeated-measures ANOVA (left), pairwise between-subject and paired withinsubject Sidak-adjusted simple-main effects post-hoc tests (middle) for the experiments shown in Fig. 4 and Supplementary Fig. 11 conducted in PV-Cre mice transfected with hM4Di and their mCherry-transfected controls. Reasons for varying N-numbers: In the 0.8s-SD challenge (CLZ), one hM4Di-mouse did not perform on one of the experimental days (>90% omissions) and was therefore excluded. One mCh-mouse had to be killed prematurely due to ill health (unrelated to procedures) and therefore did not participate in the two combined 7s-ITI/0. baselineeach with 2.1 mg/kg CNO) due to unrelated health issues. A further Gq mouse did not participate in the first (7s ITI) challenge because it was delayed in training. One mCh-Control was excluded from the second 7s-ITI challenge with 2.1 mg/kg CNO, due to an error in the operant box which led to lack of participation on one of the test days (78% omissions); one Gq-mouse did not run in the 9s-ITI challenge with 10 mg/kg CNO, another one was excluded because it did not participate in the task (sedated appearance); two Gq-mice did not contribute data in the 0.8s-SD challenge with 10 mg/kg due to