Behavioural Brain Research 437 (2023) 114113 Contents lists available at ScienceDirect Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr Short communication Decreased activation of parvalbumin interneurons in the medial prefrontal cortex in intact inbred Roman rats with schizophrenia-like reduced sensorimotor gating Carles Tapias-Espinosa a, *, Ana Sánchez-González a, Toni Cañete a, Daniel Sampedro-Viana a, Maria del Mar Castillo-Ruiz b, Ignasi Oliveras a, Adolf Tobeña a, Susana Aznar c, d, **, Alberto Fernández-Teruel a, * a Department of Psychiatry & Forensic Medicine, Institute of Neurosciences, Universitat Autònoma de Barcelona, Barcelona, Spain b Histology Lab, Institute of Neurosciences, Universitat Autònoma de Barcelona, Barcelona, Spain c Research Laboratory for Stereology and Neuroscience, Bispebjerg-Frederiksberg Hospital, University Hospital of Copenhagen, Copenhagen, Denmark d Copenhagen Center for Translational Research, Bispebjerg-Frederiksberg Hospital, University Hospital of Copenhagen, Copenhagen, Denmark A R T I C L E I N F O A B S T R A C T Keywords: Prepulse inhibition (PPI) allows assessing schizophrenia-like sensorimotor gating deficits in rodents. Previous Parvalbumin studies indicate that PPI is modulated by the medial prefrontal cortex (mPFC), which is in agreement with our c-Fos findings showing that PPI differences in the Roman rats are associated with divergences in mPFC activity. Here, PSD-95 we explore whether differences in PPI and mPFC activity in male Roman rats can be explained by (i) differences Prepulse inhibition Schizophrenia in the activation (c-Fos) of inhibitory neurons (parvalbumin (PV) interneurons); and/or (ii) reduced excitatory Roman rats drive (PSD-95) to PV interneurons. Our data show that low PPI in the Roman high-avoidance (RHA) rats is associated with reduced activation of PV interneurons. Moreover, the RHA rats exhibit decreased density of both PV interneurons and PSD-95 puncta on active PV interneurons. These findings point to reduced cortical inhi­ bition as a candidate to explain the schizophrenia-like features observed in RHA rats and support the role of impaired cortical inhibition in schizophrenia. 1. Introduction show lower mPFC activity and PPI than the Roman low-avoidance (RLA) rats. Our findings raised the question of which type of neurons in the Sensorimotor gating, which is a process to filter out relevant from mPFC may be involved in the modulation of PPI. In this context, the irrelevant information, is deficient in several neuropsychiatric condi­ GABA hypothesis of schizophrenia has been postulated to explain the tions, including schizophrenia [1]. Sensorimotor gating can be opera­ etiology of cognitive symptoms [5], which are related to PPI dysfunction tionally measured by prepulse inhibition (PPI) of the startle response [6]. [2]. Impaired PPI in rodents is used as a common endophenotype to The GABA hypothesis of schizophrenia postulates that GABA- model this basic attentional schizophrenia-like deficiency and to try to mediated cortical inhibition is dysfunctional in this disease, leading to elucidate the mechanisms underlying schizophrenia symptoms. In this excessive excitatory neural transmission to subcortical brain regions [5, regard, findings from clinical and preclinical studies indicate that PPI is 7,8]. Particularly, it has been suggested that dysfunctional NMDA re­ modulated by the cortico-striato-pallido-thalamic (CSPT) circuit [1,3]. ceptors in GABAergic interneurons do not activate these inhibitory Accordingly, we found a relationship between PPI differences and di­ neurons, and this leads to overactivation of glutamatergic projections vergences in volume and activity of the medial prefrontal cortex (mPFC) [5,9]. In turn, this higher glutamatergic transmission would in both intact inbred Roman and outbred rats [4]. Particularly, our data over-activate (i) the mesolimbic dopaminergic neurons, responsible for revealed that the schizophrenia-like Roman high-avoidance (RHA) rats positive symptoms of schizophrenia; and (ii) the GABAergic neurons * Corresponding authors at: Department of Psychiatry & Forensic Medicine, Institute of Neurosciences, Universitat Autònoma de Barcelona, Barcelona, Spain. ** Corresponding author at: Research Laboratory for Stereology and Neuroscience, Bispebjerg-Frederiksberg Hospital, University Hospital of Copenhagen, Copenhagen, Denmark. E-mail addresses: carles.tapias@uab.cat (C. Tapias-Espinosa), susana.aznar.kleijn@regionh.dk (S. Aznar), albert.fernandez.teruel@uab.cat (A. Fernández-Teruel). https://doi.org/10.1016/j.bbr.2022.114113 Received 27 April 2022; Received in revised form 24 August 2022; Accepted 11 September 2022 Available online 13 September 2022 0166-4328/© 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by- nc-nd/4.0/). C. Tapias-Espinosa et al. Behavioural Brain Research 437 (2023) 114113 that project to the PFC, causing negative and cognitive symptoms [5]. ◦ C for 2 h, washed twice and cryopreserved at 4 ◦ C in 30 % sucrose This hypothesis is based on the evidence that schizophrenic patients buffered solution, until frozen in isopentane. Coronal 30-μm thick sec­ exhibit poor gamma oscillation frequency in the PFC, a pattern of tions of the mPFC (Bregma: from 3.72 to 2.52 mm, infralimbic and neuronal firing critical for cognitive abilities, such as attention and prelimbic cortices) were cut in a cryostat (Leica CM3050S), collected working memory [9–11]. Specifically, a subpopulation of GABA in­ free-floating, and stored at − 20 ◦ C until immunostaining. After washes terneurons that expresses the calcium-binding protein parvalbumin (PV) in PBS, antigen retrieval (citrate buffer pH 6) and blocking were per­ seems to be essential to provide inhibitory inputs and to drive cortical formed (10 % normal horse serum (Sigma H0146). Slices were then gamma oscillations. Importantly, it has been recently demonstrated that incubated overnight at 4 ◦ C with the three primary antibodies (goat anti- excitatory synapses (measured by PSD-95) are selectively decreased on c-Fos IgG, Santa Cruz, SC52G, 1:500; mouse anti-PV, 1:2000, sigma; PV interneurons in the PFC of schizophrenic patients and this predicts Rabbit anti-PSD-95, 1:1000, Abcam) diluted in “antibody diluent” (PBS- the activity-dependent downregulation of PV [11]. In this sense, studies T with 1 % (w/v) BSA (Sigma A9647). After several washes in PBS-T have shown reduced density of PV interneurons and/or expression of PV (Tween 0.05 %), the slices were incubated with biotinylated horse in both schizophrenic patients [12–14] and animal models of schizo­ anti-goat IgG (Vector Laboratories, 1:400) 40 min at RT. After washes in phrenia [2,15–17]. Regarding the role of PV interneurons in PPI, pre­ PBS-T, the slices were incubated with Alexa Flour 555 streptavidin, anti- vious findings indicate that prenatal phencyclidine or mouse Alexa Flour 488, and anti-rabbit Alexa Flour 647 (ThermoFisher lipopolysaccharide treatments reduce the density of PV-positive (PV+) Scientific, 1:750 in antibody diluent, 45 min at RT). Finally, counter­ cells and the c-Fos activity within the mPFC in parallel to PPI impair­ staining (Hoechst, 1:10,000) was performed, and the slices were ments [18,19]. Similarly, the PV knockout (PV–/–) mice show lower PPI mounted on slides, air dried and coverslipped with aqueous medium. than the PV+/+ control group [20]. Brain images were taken on a confocal microscope (Leica SP5) using The inbred RHA and RLA rats were bidirectionally selected for rapid a 63x/1.50 NA SC oil immersion objective by an experimenter blinded to vs. non-acquisition of the two-way active avoidance task, respectively. treatment condition. For each animal, in three sections of the mPFC, Interestingly, compared with their RLA counterparts, RHA rats exhibit fifteen image stacks (512 × 512; 0.7 µm) were randomly selected. We reduced PPI, working memory, and latent inhibition. These behavioral used the IMARIS 9.5 software to automatically identify and count the phenotypes are accompanied by different neural abnormalities that number of c-Fos and PV nuclei and PSD-95 puncta per mm2 and calcu­ indicate that the RHA rats might be a valid animal model to study the late the average for each animal. Based on the ability of antibodies to underlying mechanisms of schizophrenia-related symptoms [21]. penetrate tissue, we obtained a representative image of objects that were In this study, we investigated, by triple fluorescence immunostain­ in the middle of the z-planes (see a representative example in Fig. 1). No ing, the degree of the colocalization of c-Fos (as indirect marker of ac­ group differences were observed in the mean volume of tissue sampled. tivity) with PV interneurons and PSD-95 puncta on PV cell bodies in the Colocalization of c-Fos and PV cell bodies was calculated to deter­ mPFC of Roman rats after a PPI session. We hypothesize that the mine the density of active PV interneurons. The percentage of PV in­ schizophrenia-like RHA rats, compared to the RLAs, would show fewer terneurons colocalized with c-Fos (%PV+/c-Fos) was determined by the number of active PV interneurons and lower PSD-95 puncta on PV following formula: PV/c-Fos density = [number of “PV+c-Fos+” cells interneurons. /“PV+ total” cells x 100]. The number of PSD-95 puncta per surface area of “PV+ total” or “PV+c-Fos+” cells was calculated to determine the 2. Materials and methods density of excitatory synapses on PV interneurons or active PV in­ terneurons, respectively. We used 42 inbred male Roman rats (RHA = 21; RLA = 21) from our All the analyses were performed using the SPSS software. Student’s t- permanent colony at the Autonomous University of Barcelona. Rats test was used to test for significant PPI differences between RHA and were 2.5–3 months of age and weighted 220–290 g. They were housed in RLA rats. ANOVA (‘2 strains x 4 conditions’) was used to explore dif­ pairs in macrolon cages (50 × 25 × 14 cm) and maintained with food ferences in the neural markers and behavioral conditions between the and water ad libitum, under a 12:12 light-dark cycle (lights on at 08:00 RHA and RLA rats. Significance level was set at p < 0.05. am) and controlled temperature (22 ± 2 ◦ C) and humidity (50–60 %). All procedures were carried out in accordance with the Spanish Legis­ 3. Results lation (Royal Decree 53/2013, 1st February 2013) and the European regulation for “Protection of Animals used for Scientific Purposes” Behavioral data showed lower PPI in the RHA rats than RLAs, but no (2010/63/UE, 22 September 2010). differences in startle response (Table 1). PPI was performed as previously described in [4]. In short, rats were In relation to neuronal activity, ANOVA revealed ‘Strain’ individually located in an acrylic cylinder within a sound-attenuated (F(1,34)= 23.711; p = <0.001), ‘Condition’ (F(3,34)= 3.863; p = 0.018), box. The PPI session consisted of: (i) 5-min acclimation period; (ii) 10 and ‘Strain x Condition’ (F(3,34)= 2.985; p = 0.045) effects in c-Fos ac­ pulse-alone trials (105 dB(A), SPL, 40 ms); (iii) 60 random trials of: 10 tivity in the mPFC. Importantly, Duncan’s post hoc test confirmed that pulse-alone (used to calculate the %PPI), 10 prepulses of each intensity the RLA rats under the PPI conditions showed higher c-Fos activity than 65/70/75/80 dB(A, SPL, 20 ms) followed by pulse stimulus with an the NP, Pre, and Pulse RLA groups and all the RHA groups, while there inter-stimulus interval of 100 ms, or non-stimulus trials (background were no differences among RHA groups (Fig. 2b). noise of 55 dB). The %PPI for each prepulse intensity was obtained by Regarding PV interneurons, 2 × 4 ANOVA revealed a significant applying the following formula: %PPI = [100 – (startle amplitude on ‘Strain’ effect (F(1,34) = 15.891; p = <0.001; Fig. 2a), as the RLA rats prepulse trials/startle amplitude on pulse-alone trials x 100)]. The showed a higher number of PV+ cells in the mPFC than the RHA rats, No-Pulse (NP), Prepulse-alone (Pre), and the Pulse-alone (Pulse) control both in cell bodies and surface area. There were no ‘Condition’ (F(3,34) groups for each strain underwent sessions of non-stimulus trials, = 0.256; p = 0.857) or ‘Strain × Condition’ (F(3,34) = 1.172; p = 0.335) prepulse-alone trials, and pulse-alone trials of the same duration as the effects. PPI test. The distribution of groups was as follows: RHA-NP = 4; On the other hand, concerning the percentage of c-Fos positive PV RHA-Pre = 4; RHA-Pulse = 5; RHA-PPI = 8; RLA-NP = 4; RLA-Pre = 4; interneurons, 2 × 4 ANOVA showed a significant ‘Strain’ effect (F(1,34) RLA-Pulse = 6; RLA-PPI = 7. = 4.864; p = 0.034), as it was globally higher in the RLA than in the Two hours after the described behavioral conditions, rats were RHA rats, and a “Strain x Condition” effect (F(3,34) = 2.894; p = 0.049). euthanized with a lethal dose of pentobarbital and transcardially Post-hoc Duncan’s test confirmed that the RLA under the PPI condition perfused with a solution of 0.1 M PBS followed by 3.7–4 % buffered showed higher percentage of active PV interneurons than the RLA formaldehyde. Removed brains were immersed in the same fixative at 4 control groups and the RHA groups (p < 0.05; Fig. 2d). In fact, the RHA- 2 C. Tapias-Espinosa et al. Behavioural Brain Research 437 (2023) 114113 Fig. 1. Triple immunofluorescence staining for parvalbumin (PV)-containing interneurons, c-Fos expression and PSD-95 puncta. Representative images of PV (A), c- Fos (B), PSD-95 puncta (C) markers and merged images (D). number of PV interneurons in schizophrenia [12–14] and animal models Table 1 of neuropsychiatric disorders [2,15–17]. For example, prenatal and Mean ± SEM of startle response amplitude and prepulse inhibition in RHA and postnatal neurodevelopmental models of schizophrenia have shown RLA rats. Student’s t-test showed Strain effects on “PPI 65_70dB” (average of the two lower prepulse intensities) [t(1,13)= 2.885; p = 0.013] and “PPI Total” impaired PPI associated to reduced density of PV interneurons and c-Fos (average of all prepulse intensities, i.e. 65, 70, 75, 80 dB) [t(1,13)= 2.395; activity in the PFC after PPI [18,19], while loss of PV gene involves p = 0.032]. No differences were found in Startle response (average of pulse trials impaired PPI [20]. In contrast, other animal models of schizophrenia, to calculate PPI) [t(1,13)= 0.776; p = 0.452]. *, significant difference vs the such as maternal immune activation with polyI:C, report no changes in corresponding RHA group. PV interneurons, but differences in perineuronal nets surrounding PV RHA-PPI (n = 8) RLA-PPI (n = 7) interneurons in the medial prelimbic cortex [22]. Regarding PV in­ terneurons density, we have shown that the schizophrenia-like RHA rats Startle response 922.1 ± 174.1 1190.7 ± 199.8 PPI 65_70dB 43.9 ± 7.2 67.6 ± 3.0* exhibit lower density of inhibitory PV interneurons in the mPFC than the PPI Total 57.1 ± 6.1 73.8 ± 2.5* RLAs. However, as noted in Refs. [16,18], we may consider that the use of fluorescent immunohistochemistry may not allow us to determine whether the RHA rats had lower PV expression (below the detection PPI rats, compared to their RLA-PPI counterparts, exhibit an approxi­ limit) rather than loss of PV interneurons, as it happens in the mately 60 % reduced activation of PV interneurons. NR1-deficient mice model of schizophrenia [23]. Indeed, in a previous Finally, regarding PSD-95, no significant effects were observed in study with the Roman rats using the more sensitive either total number of PSD-95 puncta or PSD-95 puncta on PV+ cells HRP-immunocytochemistry and stereology, we found no between-strain (Fig. 2c and e, respectively; all p > 0.498). However, we found a ‘Strain’ differences in PV numbers in the frontal cortex in basal conditions [24]. effect in the density of PSD-95 puncta on active PV+ cells (PSD-95 Nevertheless, the possible differences in PV expression between the puncta on “PV+c-Fos+” cells; F(1,34) = 6.447; p = 0.016), as it was Roman rats are in line with differences observed in PV interneurons globally higher in RLA rats than in RHAs (Duncan’s post hoc test, activation, as lower PV expression would also indicate lower activity p < 0.05; Fig. 2f). No ‘Condition’ (F(3,34) = 0.227; p = 0.841) or ‘Strain [11]. This is consistent with previous reports showing that schizophrenic × Condition’ (F(3,34) = 0.159; p = 0.923) effects were found in the patients display lower GAD67 expression in PV interneurons [25,26], as density of PSD-95 puncta on active PV+ cells. well as reduced density of PV interneurons [2,13,14]. In this sense, a reduced activation of PV interneurons could drive to the disinhibition of 4. Discussion pyramidal neurons that causes increased mesolimbic dopamine and schizophrenia-like symptoms in the RHA rats [21]. Accordingly, exog­ This study aimed to elucidate specific mechanisms associated with enous oxytocin administration, which shows an inhibitory profile on reduced mPFC activity during PPI in the RHA rats compared to the RLAs. GABA/Glutamate function, attenuates natural PPI deficits in the RHA Our results indicate that it might be explained by differences in the rats [27]. activation of PV interneurons. Specifically, we show that the RHA rats On the other hand, we did not find differences either between strains display both lower PPI and activation of PV interneurons than the RLAs. or the PPI groups in PSD-95 puncta density, which contrasts with a Our result showing a relationship between reduced activation of PV previous study in humans reporting an 18 % reduction in mean density interneurons and low PPI agrees with several studies reporting reduced of PSD-95 puncta on PV+ interneurons in schizophrenic patients [11]. 3 C. Tapias-Espinosa et al. Behavioural Brain Research 437 (2023) 114113 Fig. 2. Reduced PPI in Roman rats is related to decreased activation of parvalbumin-positive (PV+) cells in the medial prefrontal cortex. A) The RHA rats show globally lower PV+ density than the RLAs. B) The RLA rats in the PPI condition showed higher c-Fos activity in the mPFC than the RLA control groups (NP, no-pulse; Pre, prepulse-alone; Pulse, pulse-alone) and all the RHA groups. C) No differences in the density of PSD-95 puncta were observed. D) The RLA rats in the PPI condition show a higher percentage of active PV+ interneurons than the RLA control groups (NP, no-pulse; Pre, prepulse-alone; Pulse, pulse-alone) and all the RHA groups. E) No differences in the density of PSD-95 puncta on PV+ cells were observed. F) The RHA rats show globally lower density of PSD-95 puncta on active PV+ interneurons than the RLAs. Values are mean ± SEM. See “n”/groups in Methods. *p < 0.05; ***p < 0.001; ns, non-significant (Duncan’s multiple range test). However, interestingly, we found strain differences in the mean density Data Availability of PSD95 puncta on “PV+c-Fos+” interneurons, as the schizophrenia-like RHA rats showed globally lower PSD-95 density on The authors do not have permission to share data. active PV+ cells than the RLA strain. Even though PSD-95 seems not to be related to PPI performance, this finding is consistent with the idea References that the excitatory drive plays a role in the activity of PV interneurons. For future studies, it would be interesting to analyze other GABAergic [1] S. Kohl, K. Heekeren, J. Klosterkötter, J. Kuhn, Prepulse inhibition in psychiatric disorders–apart from schizophrenia. J. Psychiatr. Res. 47 (2013) 445–452, https:// interneurons or glutamatergic neurons, as well as the broad CSPT cir­ doi.org/10.1016/j.jpsychires.2012.11.018. cuit, to determine the specificity of the present findings. [2] S.B. Powell, M. Weber, M.A. Geyer, Genetic models of sensorimotor gating: Our data indicate that the lower neuronal activity in the mPFC relevance to neuropsychiatric disorders, Curr. Top. Behav. Neurosci. 12 (2012) 251–318, https://doi.org/10.1007/7854_2011_195. during PPI, showed by the RHA rats relative to the RLAs, might be [3] N.R. Swerdlow, G.A. Light, Animal models of deficient sensorimotor gating in explained by differences in PV interneurons activation. This result points schizophrenia: Are they still relevant? Curr. Top. Behav. Neurosci. (2016) to reduced cortical inhibition as a likely candidate to explain the 305–325, https://doi.org/10.1007/7854_2015_5012. [4] C. Tapias-Espinosa, C. Río-Álamos, A. Sánchez-González, I. Oliveras, D. Sampedro- schizophrenia-like features observed in the RHA rats and supports the Viana, M. del, M. Castillo-Ruiz, T. Cañete, A. Tobeña, A. Fernández-Teruel, role of impaired cortical inhibition in schizophrenia. Moreover, our Schizophrenia-like reduced sensorimotor gating in intact inbred and outbred rats is findings could also apply to other neuropsychiatric conditions, such as associated with decreased medial prefrontal cortex activity and volume, Neuropsychopharmacology 44 (2019) 1975–1984, https://doi.org/10.1038/ obsessive compulsive disorder and Gilles de la Tourette’s syndrome, s41386-019-0392-x. which are accompanied by impaired PPI [1]. [5] E. Elert, Aetiology: Searching for schizophrenia’s roots, Nature 508 (2014) S2–S3, https://doi.org/10.1038/508S2a. Funding and disclosure [6] C.M. Powell, T. Miyakawa, Schizophrenia-relevant behavioral testing in rodent models: a uniquely human disorder, Biol. Psychiatry 59 (2006) 1198–1207, https://doi.org/10.1016/j.biopsych.2006.05.008. This study was supported by grants from the Spanish Government [7] B.P. Blum, J.J. Mann, The GABAergic system in schizophrenia. Int. J. (PSI2017-82257-P and PID2020-114697GB-I00) and the Generalitat de Neuropsychopharmacol. 5 (2002) 159–179, https://doi.org/10.1017/ S1461145702002894. Catalunya (2017SGR-1586 and ICREA-Academia 2013) [to AFT] and [8] J.E. Lisman, J.T. Coyle, R.W. Green, D.C. Javitt, F.M. Benes, S. Heckers, A.A. Grace, “Bispebjerg Research Funds” [to SA]; and by fellowships from the Circuit-based framework for understanding neurotransmitter and risk gene Spanish Government (FPU15/06307 [to CTE] and FJC2018-038808-I interactions in schizophrenia, Trends Neurosci. 31 (2008) 234–242, https://doi. org/10.1016/j.tins.2008.02.005. [to IO]). [9] B. Moghaddam, D. Javitt, From revolution to evolution: the glutamate hypothesis of schizophrenia and its implication for treatment, Neuropsychopharmacology 37 Conflict of interest (2012) 4–15, https://doi.org/10.1038/npp.2011.181. [10] G. Gonzalez-Burgos, R.Y. Cho, D.A. Lewis, Alterations in cortical network oscillations and parvalbumin neurons in schizophrenia, Biol. Psychiatry 77 (2015) The authors declare no competing interests. 1031–1040, https://doi.org/10.1016/j.biopsych.2015.03.010. [11] D.W. Chung, K.N. Fish, D.A. Lewis, in: J. Am (Ed.), Pathological basis for deficient excitatory drive to cortical parvalbumin interneurons in schizophrenia, Psychiatry, 4 C. Tapias-Espinosa et al. Behavioural Brain Research 437 (2023) 114113 American Psychiatric Association, 2016, pp. 1131–1139, https://doi.org/10.1176/ mice, Neurosci. Lett. 553 (2013) 216–220, https://doi.org/10.1016/j. appi.ajp.2016.16010025. neulet.2013.08.042. [12] D.A. Lewis, D.A. Cruz, D.S. Melchitzky, J.N. Pierri, Lamina-specific deficits in [21] A. Fernández-Teruel, I. Oliveras, T. Cañete, C. Rio-Álamos, C. Tapias-Espinosa, parvalbumin-immunoreactive varicosities in the prefrontal cortex of subjects with D. Sampedro-Viana, A. Sánchez-González, F. Sanna, R. Torrubia, J. González- schizophrenia: evidence for fewer projections from the thalamus, Am. J. Psychiatry Maeso, P. Driscoll, I. Morón, C. Torres, S. Aznar, A. Tobeña, M.G. Corda, O. Giorgi, 158 (2001) 1411–1422, https://doi.org/10.1176/appi.ajp.158.9.1411. Neurobehavioral and neurodevelopmental profiles of a heuristic genetic model of [13] C.L. Beasley, G.P. Reynolds, Parvalbumin-immunoreactive neurons are reduced in differential schizophrenia- and addiction-relevant features: the RHA vs. RLA rats, the prefrontal cortex of schizophrenics, Schizophr. Res. 24 (1997) 349–355. Neurosci. Biobehav. Rev. 131 (2021) 597–617, https://doi.org/10.1016/J. 〈http://www.ncbi.nlm.nih.gov/pubmed/9134596〉. accessed January 28, 2019. NEUBIOREV.2021.09.042. [14] C.L. Beasley, Z.J. Zhang, I. Patten, G.P. Reynolds, Selective deficits in prefrontal [22] J.W. Paylor, B.R. Lins, Q. Greba, N. Moen, R.S. De Moraes, J.G. Howland, I. cortical GABAergic neurons in schizophrenia defined by the presence of calcium- R. Winship, Developmental disruption of perineuronal nets in the medial prefrontal binding proteins, Biol. Psychiatry 52 (2002) 708–715. 〈http://www.ncbi.nlm.nih. cortex after maternal immune activation, Sci. Rep. 6 (2016), https://doi.org/ gov/pubmed/12372661〉. accessed January 28, 2019. 10.1038/SREP37580. [15] G.B. Bissonette, M.H. Bae, T. Suresh, D.E. Jaffe, E.M. Powell, Prefrontal cognitive [23] M.J. Gandal, J. Sisti, K. Klook, P.I. Ortinski, V. Leitman, Y. Liang, T. Thieu, deficits in mice with altered cerebral cortical GABAergic interneurons, Behav. R. Anderson, R.C. Pierce, G. Jonak, R.E. Gur, G. Carlson, S.J. Siegel, GABA B Brain Res. 259 (2014) 143–151, https://doi.org/10.1016/j.bbr.2013.10.051. -mediated rescue of altered excitatory-inhibitory balance, gamma synchrony and [16] S.S. Kaalund, J. Riise, B.V. Broberg, K. Fabricius, A.S. Karlsen, T. Secher, N. Plath, behavioral deficits following constitutive NMDAR-hypofunction, Transl. Psychiatry B. Pakkenberg, Differential expression of parvalbumin in neonatal phencyclidine- 2 (2012), https://doi.org/10.1038/tp.2012.69. treated rats and socially isolated rats, J. Neurochem. 124 (2013) 548–557, https:// [24] A. Sánchez-González, E. Thougaard, C. Tapias-Espinosa, T. Cañete, D. Sampedro- doi.org/10.1111/jnc.12061. Viana, J.M. Saunders, R. Toneatti, A. Tobeña, J. Gónzalez-Maeso, S. Aznar, [17] J. Gilabert-Juan, M. Belles, A.R. Saez, H. Carceller, S. Zamarbide-Fores, M. A. Fernández-Teruel, Increased thin-spine density in frontal cortex pyramidal D. Moltó, J. Nacher, A “double hit” murine model for schizophrenia shows neurons in a genetic rat model of schizophrenia-relevant features, Eur. alterations in the structure and neurochemistry of the medial prefrontal cortex and Neuropsychopharmacol. 44 (2021) 79–91, https://doi.org/10.1016/j. the hippocampus, Neurobiol. Dis. 59 (2013) 126–140, https://doi.org/10.1016/j. euroneuro.2021.01.006. nbd.2013.07.008. [25] S. Akbarian, J.J. Kim, S.G. Potkin, J.O. Hagman, A. Tafazzoli, W.E. Bunney, E. [18] K. Toriumi, M. Oki, E. Muto, J. Tanaka, A. Mouri, T. Mamiya, H.-C. Kim, G. Jones, Gene expression for glutamic acid decarboxylase is reduced without loss T. Nabeshima, Prenatal phencyclidine treatment induces behavioral deficits of neurons in prefrontal cortex of schizophrenics, Arch. Gen. Psychiatry 52 (1995) through impairment of GABAergic interneurons in the prefrontal cortex, 258–266, https://doi.org/10.1001/archpsyc.1995.03950160008002. Psychopharmacology 233 (2016) 2373–2381, https://doi.org/10.1007/s00213- [26] T. Hashimoto, D.W. Volk, S.M. Eggan, K. Mirnics, J.N. Pierri, Z. Sun, A.R. Sampson, 016-4288-8. D.A. Lewis, Gene expression deficits in a subclass of GABA neurons in the [19] L. Wischhof, E. Irrsack, C. Osorio, M. Koch, Prenatal LPS-exposure - a prefrontal cortex of subjects with schizophrenia, J. Neurosci. 23 (2003) neurodevelopmental rat model of schizophrenia - differentially affects cognitive 6315–6326, https://doi.org/10.1523/jneurosci.23-15-06315.2003. functions, myelination and parvalbumin expression in male and female offspring, [27] C. Tapias-Espinosa, T. Cañete, D. Sampedro-Viana, T. Brudek, A. Kaihøj, I. Oliveras, Prog. Neuro-Psychopharmacology, Biol. Psychiatry 57 (2015) 17–30, https://doi. A. Tobeña, S. Aznar, A. Fernández-Teruel, Oxytocin attenuates schizophrenia-like org/10.1016/j.pnpbp.2014.10.004. reduced sensorimotor gating in outbred and inbred rats in line with strain [20] J. Popelář, N. Rybalko, J. Burianová, B. Schwaller, J. Syka, The effect of differences in CD38 gene expression, Physiol. Behav. 240 (2021), 113547, https:// parvalbumin deficiency on the acoustic startle response and prepulse inhibition in doi.org/10.1016/j.physbeh.2021.113547. 5