HCN channelopathy as a key mechanism for auditory hypersensitivity in a shank3 mouse model of ASD

Universidade do Minho Escola de Medicina Ana Carolina Pinto Castro HCN channelopathy as a key mechanism for auditory hypersensitivity in a mouse model of ASD outubro de 2022 UMinho | 2022 Ana Carolina Pinto Castro HCN channelopathy as a key mechanism for auditory hypersensitivity in a mouse model of ASD Shank3 Shank3 Ana Carolina Pinto Castro HCN channelopathy as a key mechanism for auditory hypersensitivity in a mouse model of ASD Dissertação de Mestrado Mestrado em Ciências da Saúde Trabalho efetuado sob a orientação da Professora Doutora Patrícia Monteiro e do Professor Doutor Luís Jacinto Universidade do Minho Escola de Medicina outubro de 2022 Shank3 i DIREITOS DE AUTOR E CONDIÇÕES DE UTILIZAÇÃO DO TRABALHO POR TERCEIROS Este é um trabalho académico que pode ser utilizado por terceiros desde que respeitadas as regras e boas práticas internacionalmente aceites, no que concerne aos direitos de autor e direitos conexos. Assim, o presente trabalho pode ser utilizado nos termos previstos na licença abaixo indicada. Caso o utilizador necessite de permissão para poder fazer um uso do trabalho em condições não previstas no licenciamento indicado, deverá contactar o autor, através do Repositório UM da Universidade do Minho. Licença concedida aos utilizadores deste trabalho: Atribuição-NãoComercial-SemDerivações CC BY-NC-ND https://creativecommons.org/licenses/by-nc-nd/4.0/ ii AGRADECIMENTOS Começo por agradecer à minha orientadora, Professora Patrícia Monteiro, que tão bem me acolheu desde o primeiro momento na sua equipa. Obrigada pelas oportunidades e pelos ensinamentos. Ao Professor Luís Jacinto, meu coorientador, por ter sido também um enorme apoio, especialmente na altura de arranjar soluções criativas para os problemas. Sou grata também pelas minhas colegas de equipa. Agradeço à Margarida Falcão, à Patrícia Silva e à Alexandra Cruz, pessoas que me deram imensa força nesta jornada. Obrigada à Sara e ao Carlos V., companheiros de outras andanças, mas que permaneceram e permanecerão. Às novas amizades, que o ICVS me deu, Ricardo, Carlos e Inês. Deixo um agradecimento especial à Diana, que admiro muito e que tem uma paciência e dedicação inesgotáveis. Agradeço à Bruna, à Daniela, à Ana Campos-Ríos e a todas essas cientistas de garra que, tal como todas as outras que já mencionei até agora, tiveram sempre uma palavra de força para me dar. É uma enorme honra partilhar ciência com mulheres assim. Aos colegas de equipa, colegas de mestrado, colegas de laboratório, estou grata por todas as interações porque todos vocês tiveram algo para me ensinar. Finalmente, agradeço à minha família, aqueles que estiveram lá desde o início. À minha mãe, aos meus irmãos, à minha tia e aos meus queridos avós. Agradeço ainda ao João, por tornar os momentos felizes ainda mais brilhantes e os tristes mais suportáveis. FUNDING The work included in this dissertation was performed at the Life and Health Sciences Research Institute (ICVS), University of Minho. Financial support was provided by FCT project PTDC/MEDNEU/28073/2017 (POCI-01-307 0145-FEDER-028073); "La Caixa" banking foundation under grant agreement LCF/PR/HR21-00410; FEBS excellence award 2021; ICVS scientific microscopy platform, member of the national infrastructure PPBI - portuguese platform of bioimaging (PPBI-POCI-01-0145FEDER-022122); and national funds through FCTprojects UIDB/50026/2020 and UIDP/50026/2020. iii STATEMENT OF INTEGRITY I hereby declare having conducted this academic work with integrity. I confirm that I have not used plagiarism or any form of undue use of information or falsification of results along the process leading to its elaboration. I further declare that I have fully acknowledged the Code of Ethical Conduct of the University of Minho. iv RESUMO CANALOPATIA DE HCN COMO UM MECANISMO CHAVE PARA A HIPERSENSIBILIDADE AUDITIVA NUM MODELO ANIMAL DE SHANK3 PARA PEA A perturbação do espetro do autismo (PEA) é uma perturbação do neurodesenvolvimento de etiologia desconhecida. Fatores de natureza ambiental e genética estão associados a um maior risco para PEA. Até à atualidade, mais de 100 genes foram identificados como genes de risco para PEA, estando associados à regulação da expressão genética e à maturação de linhas neuronais excitatórias e inibitórias. Entre eles, o gene SHANK3 é dos mais estudados e está associado a cerca de 1-2% dos casos de PEA. As características principais da PEA incluem dificuldades na comunicação e interação social, comportamentos repetitivos e hipoou hipersensibilidade a estímulos sensoriais. Uma das alterações sensoriais mais reportadas em PEA é a sensibilidade auditiva. Diversos modelos animais para PEA também apresentam em comum esta característica, nomeadamente o modelo de murganho Shank3InsG3680+/+ , utilizado neste estudo. Neste trabalho, pretendeu-se clarificar potenciais mecanismos por detrás do fenótipo de hipersensibilidade auditiva observado no modelo de murganho Shank3-Insg3680+/+ . O principal foco foi a região cerebral do córtex auditivo primário (A1), uma vez que esta é importante para o processamento auditivo. Através de uma análise de proteómica SWATH-MS em A1, identificámos o canal HCN1 (canal regulado por hiperpolarização e por ligação a nucleótido) como um dos mais sobreexpressos em murganhos InsG3680+/+ machos adultos (6-10 semanas), mas não durante o desenvolvimento (1-5 semanas). Transcritos de HCN1 ( mHcn1 ) estão também aumentados aos 21 dias de idade (P21) em murganhos InsG3680+/+ machos. Foram também realizados registos intracelulares de correntes de hiperpolarização mediadas por HCN (Ih) por eletrofisiologia de whole-cell patch clamp . Os resultados indicam uma diminuição das correntes Ih especificamente em neurónios principais putativos em murganhos InsG3680+/+ , o que sugere uma perturbação funcional dos canais HCN em A1. Sabendo que a SHANK3 interage com HCN1 e é provavelmente responsável pela ancoragem de HCN1 na sinapse, propomos um mecanismo através do qual a falta de SHANK3 compromete a localização sub-celular de HCN1, causando redução na Ih. Como mecanismo de compensação, a célula aumenta os seus níveis de HCN1, tentando compensar os défices funcionais e levando a um fenótipo molecular de sobre-expressão de HCN1 em murganhos InsG3680+/+ adultos. Em suma, estes resultados contribuem para a melhor compreensão do funcionamento dos circuitos do córtex auditivo no modelo InsG3680+/+ e sugerem HCN1 como um potencial alvo terapêutico em PEA. Palavras-chave: Perturbação do Espetro do Autismo, Neurodesenvolvimento, Hipersensibilidade auditiva, SHANK3 , Canal regulado por hiperpolarização e por ligação a nucleótido 1. v ABSTRACT HCN CHANNELOPATHY AS A KEY MECHANISM FOR AUDITORY HYPERSENSITIVITY IN SHANK3 MOUSE MODEL OF ASD Autism spectrum disorder (ASD) is a neurodevelopmental disorder with elusive etiology. There are environmental and genetic factors underlying an increased risk for ASD. So far, over 100 genes have already been identified as ASD-risk genes, most of them associated with gene expression regulation and maturation of excitatory and inhibitory neuronal lineages. Among them, the SHANK3 human gene is one of the most studied and accounts for 1–2% of ASD cases. The core features of ASD include social-communication impairments, restricted repetitive behaviors and hypoor hyperreactivity to sensory stimuli. One of the most reported sensory-perceptual abnormalities in ASD is auditory sensitivity. Many rodent models of ASD also share auditory impairments, namely the Shank3-InsG3680+/+ mouse model, which was used in this study. Our aim was to unveil potential mechanisms underlying the auditory hypersensitivity phenotype observed in the Shank3-InsG3680+/+ mouse model of ASD. We focused on the primary auditory cortex (A1) brain region since it is an important region for auditory processing. Using SWATH-MS proteomics screening in A1, we identified HCN1 as one of the most upregulated proteins in adult InsG3680+/+ male mice (6-10 weeks), but not during development (15 weeks). HCN1 transcripts ( mHcn1 ) were also found to be increased at postnatal day 21 (P21) in InsG3680+/+ male mice. Intracellular recordings of HCN-mediated hyperpolarization currents (Ih) were also performed using whole-cell patch clamp electrophysiology. Results revealed a cell-type specific decrease of Ih in putative cortical principal neurons in InsG3680+/+ mice, suggesting a functional impairment of HCN channels in A1 brain region. Knowing that SHANK3 interacts with HCN1 and it is likely required for anchoring HCN1 at the synapse, we propose a mechanism by which lack of SHANK3 disrupts HCN1 subcellular localization, compromising HCN1 function and causing a reduction in Ih. As a compensation mechanism, the cell increases HCN1 levels trying to compensate for the lack of function, leading to a molecular phenotype of HCN1 overexpression in adult InsG3680+/+ mice. Together these results contribute to our understanding of auditory cortex circuits in the InsG3680+/+ model of ASD and reveal HCN1 as a potential therapeutic target in ASD. Keywords: Autism Spectrum Disorder, Neurodevelopment, Auditory hypersensitivity, SHANK3 , Hyperpolarization cyclic nucleotide-gated channel 1. vi INDEX DIREITOS DE AUTOR E CONDIÇÕES DE UTILIZAÇÃO DO TRABALHO POR TERCEIROS ......................... i AGRADECIMENTOS ............................................................................................................................. ii Statement of Integrity ......................................................................................................................... iii Resumo.............................................................................................................................................. iv Abstract............................................................................................................................................... v Abbreviations ...................................................................................................................................... ix List of figures ..................................................................................................................................... xii List of tables ...................................................................................................................................... xiii 1 Introduction ................................................................................................................................ 1 1.1 Autism spectrum disorder (ASD): from symptoms to etiology ................................................ 1 1.2 Genetic causes for ASD ....................................................................................................... 2 1.2.1 SHANK3 : a major risk gene for ASD ................................................................................. 3 1.2.2 Shank3 mouse models of ASD ......................................................................................... 4 1.3 Sensory processing in ASD: the auditory pathway................................................................. 6 1.4 Excitation and inhibition in primary auditory cortex circuits ................................................... 8 1.5 Auditory dysfunction in rodent models of ASD .................................................................... 10 1.5.1 Auditory processing deficits in non-genetic models of ASD .............................................. 12 1.5.1.1 Valproic acid .......................................................................................................... 13 1.5.1.2 Thalidomide ........................................................................................................... 14 1.5.2 Auditory processing deficits in monogenic models of ASD ............................................... 15 1.5.2.1 Fmr1 knockout ...................................................................................................... 15 1.5.2.2 α 7-nAChR knockout ............................................................................................... 18 1.5.2.3 Cntnap2 knockout ................................................................................................. 18 1.5.2.4 Shank3 knockout ................................................................................................... 19 1.5.2.5 Pten conditional knockout ...................................................................................... 20 xiii LIST OF TABLES Table 1. Summarized characterization of Shank3 mouse models. ........................................................ 5 Table 2. Overview of the most relevant findings regarding ASD rodent models with reported auditory dysfunctions.. ................................................................................................................................... 22 Table 3. Primer sequence used for gene amplification in RT-qPCR. .................................................... 29 Supplementary table 1. Statistical report of Student’s t -tests and Mann-Whitney’s tests. ..................... 75 Supplementary table 2. Factorial analysis.......................................................................................... 76 1 1 INTRODUCTION Parts of this introduction have been recently published by us in a peer-reviewed journal (Castro and Monteiro, 2022) (DOI: 10.3389/FNMOL.2022.845155). 1.1 Autism spectrum disorder (ASD): from symptoms to etiology Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder with different severity levels and poorly understood etiology. It is diagnosed up to 4 times more often in males than in females (Werling and Geschwind, 2013) and it is estimated to affect around 1 in 160 children worldwide (Elsabbagh et al., 2012). Current diagnosis criteria and core features are described in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5; American Psychiatric Association, 2013) and include deficits in social communication and interaction, and repetitive restricted behaviors. Such deficits in social-emotional interaction range from maintaining a conversation to sharing interests or emotions with peers, deficits in non-verbal communication (e.g. the use and interpretation of body language and facial expressions), and difficulties in establishing and maintaining relationships. The repetitive restricted behaviors may include stereotyped motor movements, insistence on sameness and rigid patterns of behaviors and routines, strict interests towards objects or activities, and hyperor hyposensitivity to sensory stimuli inputs from vision, audition, touch, smell and taste, or unusual interest in sensory aspects of the environment (American Psychiatric Association, 2013; Robertson and Baron-Cohen, 2017). Usually, symptoms start to arise during the second year of life, although this may change with the severity of the disorder. Typically, the first noticeable symptoms are delayed development of language and lack of social interest (American Psychiatric Association, 2013). The etiology of ASD is not fully understood and is likely multifactorial. Earlier, it was though that ASD would arise mostly from environmental causes. Indeed, there are environmental risk factors that may disturb normal embryonic development and contribute to ASD, such as exposure to teratogens (certain drugs, pollutants or food contaminants), maternal infections, and even advanced parental age (Arndt et al., 2005; Herbert, 2010). Nonetheless, a growing body of evidence demonstrates that ASD has a strong genetic component (Rylaarsdam and Guemez-Gamboa, 2019). Studies in monozygotic and dizygotic twins indicate concordance rates of 60-90% and up to 10%, respectively, being the initial evidences of ASD as a genetic-linked disorder (Muhle et al., 2004). However, only around 5% of the cases are associated with syndromic forms of ASD, such as Rett syndrome, Fragile X syndrome, PTEN macroencephaly syndrome, tuberous sclerosis, among others. Even nowadays many of the diagnosed cases are still idiopathic forms of autism with unknown cause. Recent studies suggest that chromosomal 2 duplications/deletions and single nucleotide mutations in ASD-risk genes are probably behind ASD phenotypes (Sztainberg and Zoghbi, 2016). Considering all the evidence, it is clear that ASD stems from both environmental and genetic factors, having monoand polygenic causes. 1.2 Genetic causes for ASD One of the largest whole-exome sequencing studies to date used over 35,000 samples from neurotypical and ASD diagnosed individuals (Satterstrom et al., 2020). The authors identified 102 ASDassociated genes which they broadly group according to two main functions: regulation of gene expression and neuronal communication (Satterstrom et al., 2020). In accordance with the categorization of ASD as a neurodevelopmental disorder, indeed most of the candidate genes identified are expressed exactly during preand early post-natal development and display cortical enrichment in maturing excitatory and inhibitory neuronal lineages (Figure 1) (Satterstrom et al., 2020). Earlier this year, an analysis of de novo and inherited variants in ASD cases identified five new ASD-risk genes: NAV3 , ITSN1 , MARK2 , SCAF1 and HNRNPUL2 (Zhou et al., 2022). Furthermore, the study showed that some of the most highly penetrant loss of function mutations, which may be either inherited or de novo , occur in well-known ASD-risk genes such as SYNGAP1 , ADNP , FOXP1 , PTEN , POGZ , KDM5B , GIGYF1 , CHD8 , SCN2A and SHANK3 (Zhou et al., 2022). Figure 1. Large-scale exome sequencing study reveals 102 ASD-associated genes. Left) Manhattan plot obtained from the analysis of over 35’584 samples collected from neurotypical controls and ASD patients shows 102 genes identified as ASD-risk genes (top candidate genes are labeled). The red dashed line represents the threshold for a false discovery rate (FDR) of 0.05%. Right) Most of the ASD-associated genes found in this study are enriched in excitatory and inhibitory lineages of both newborn and maturing neurons. Image modified from Satterstrom et al., 2020. 3 1.2.1 SHANK3 : a major risk gene for ASD Among all the ASD candidate genes, SHANK3 is one of the most studied, being considered as a possible monogenic cause for ASD, with its variants explaining approximately 1% of ASD cases. Evidence from postmortem analysis of brains from individuals diagnosed with ASD indicates that around 15% present increased SHANK3 methylation, which may shift SHANK3 isoform expression pattern in the brain. The SHANK3 gene was initially studied in the context of Phelan-McDermid Syndrome (PMS), a disorder belonging to the autistic spectrum and caused by a terminal deletion of chromosome 22, encompassing the 22q13.3 region where SHANK3 is located. PMS main symptoms include neonatal hypotonia and delays in development and speech. In this context, SHANK3 was implicated as the major genetic cause underlying the neurological symptoms in PMS (Wilson et al., 2003). Later on, SHANK3 mutations were found in individuals diagnosed with ASD but without PMS. This provided evidence that SHANK3 , by itself, is a determinant gene in ASD (Durand et al., 2007). The SHANK3 protein belongs to a family of SHANK (SH3 and multiple ankyrin repeat domains) proteins which are scaffolding proteins located at excitatory synapses and responsible for orchestrating the organization of the post-synaptic density complex (PSD) (Monteiro and Feng, 2017) (Figure 2). Different SHANK proteins present shared domains, namely the ankyrin domain (ANK; located at the N terminus), the SRC homology 3 domain superfamily (SH3), the postsynaptic density protein 95–discs large homologue 1–zonula occludens 1 domain (PDZ), the proline-rich region (PRO domain) and the sterile alpha motif domain (SAM; located at the C terminus). These are all protein-protein interacting domains that determine the scaffolding function of SHANK proteins (Monteiro and Feng, 2017). Through these domains, SHANK proteins can interact with other proteins present at the PSD complex, such as the SAP90/PSD95-associated protein (SAPAP), the Homer and cortactin proteins (through the PRO domain), and the glutamate receptor 1 (GluR1) subunit of α-amino-3-hydroxy-5-methyl-4isoxazolepropionic acid (AMPA) receptors (through the PDZ domain). These interactions are important for the formation and maintenance of dendritic spines, synaptic communication, plasticity and cytoskeleton dynamics (Monteiro and Feng, 2017). Several studies have shown that Shank3 mutations in zebrafish (Liu et al., 2018), mice (Mei et al., 2016; Zhou et al., 2016), rats (Song et al., 2019), and even primates (Zhou et al., 2019b), lead to reduced social interaction and repetitive behaviors, which are core ASD-like features (Peça et al., 2011; Zhou et al., 2016; Delling and Boeckers, 2021). But despite these behavioral phenotypes associated with Shank3 mutations being conserved among different species, the mouse is still by far the most studied animal model. 4 Figure 2. Schematic representation of SHANK protein and its main interactors at the postsynaptic density. The SHANK protein (SH3 and multiple ankyrin repeat domains) tethers several other PSD proteins at excitatory post-synaptic sites through its multiple protein-protein interaction domains: 1) ankyrin repeat domain (ANK) - SHARPIN and SPTAN1 (spectrin alpha chain, non-erythrocytic 1) interact with the N-terminal of ANK; 2) PDZ domain (PSD protein 95 (PSD95)–discs large homologue 1-zonula occludens 1) - SHANK interacts with NMDA and AMPA receptors (NMDAR and AMPAR, respectively) through the PDZ domain. The interaction with NMDAR occurs indirectly through two other proteins: SAPAP (SAP90/PSD95-associated proteins) and PSD95. 3) PRO domain (proline rich) - SHANK protein establishes connections with the cytoskeleton (interacts with actin through cortactin) and with metabotropic glutamate receptors (mGluRs; interacts with mGluRs through homer) via PRO domain. Image adapted from Monteiro and Feng, 2017. 1.2.2 Shank3 mouse models of ASD Most Shank3 transgenic mouse models of ASD generated so far present social impairments and repetitive behavior (essentially manifested through increased self-grooming behavior) (Table 1). The severity of such features can vary depending on the transgenic mouse line studied (e.g. repetitive grooming can range from a moderately increment in time spent grooming to self-inflicted skin lesions). 5 Regarding sensory alterations, these have not been typically studied in animal models until very recently, perhaps reflecting the recent alterations in DSM-5 classification where ASD criteria now includes sensory alterations. Nonetheless, alterations in acoustic thresholds (either as increased or decreased) have been reported, as well as olfactory deficits and hyposensitivity (Zhou et al., 2016; Drapeau et al., 2018). A more detailed overview can be seen in Table 1. Table 1. Summarized characterization of Shank3 mouse models. Abbreviations: ANK (Ankyrin domain), Δex (exon deletion), GluR1 (Glutamate receptor 1), GKAP (guanylate kinase-associated protein), GluA (Glutamate AMPA receptor), NR (Nicotinamide riboside), PPI (Prepulse inhibition), SAPAP3 (SAP90/PSD95¬associated protein 3), PSD93 or PSD95 (Postsynaptic density protein 93 or 95), pGluN (phospho-NMDA receptor), SynGap (Synaptic Ras GTPase-activating protein), n.s. (not signifficant), N/A (Not applicable). Modified from (Monteiro and Feng, 2017). Protein domains Isoforms not disrupted Altered synaptic proteins Social behaviors Repetitive behaviors Sensorial alterations Refs. ANK repeat: Δex4-9Buxbaum 3 - 10 Decreased density of GluR1immunoreactive puncta Less sniffing in male-female interactions. Normal sociability in 3chamber Increased selfgrooming Hotplate: n.s.; Olfactory discrimination: n.s. (Bozdagi et al., 2010; Yang et al., 2012) ANK repeat: Δex4-9Jiang 3 - 10 Decreased GKAP, Homer1b/c, GluA1, NR2A Abnormal social interaction; Normal social novelty Increased selfgrooming. Holeboard test: increased head pokes PPI: n.s.; Olfactory discrimination: n.s. (Wang et al., 2011) ANK repeat: Δex4-7Feng 3 - 10 N/A Normal social interaction; Abnormal social novelty No increase in self-injurious grooming N/A (Peça et al., 2011) PDZ domain: Δex13-16Feng 8,9,10 Reduced Homer1, SAPAP3, NR2, NR2B, GluA2, PSD93 Abnormal social interaction and abnormal social novelty in adults and juveniles Self-injurious grooming, causing skin lesions N/A (Peça et al., 2011; Dhamne et al., 2017; Balaan et al., 2019) SH3 domain: Δex11Boeckers 6 - 10 Increased NR2B and Shank2; Reduced Homer1b/c, mGluR5 Abnormal social interaction; Abnormal social novelty Self-injurious grooming, causing skin lesions Hotplate: hyposensitivity (Schmeisser et al., 2012; Vicidomini et al., 2017) ANK repeat: Δex9Kim 3 - 10 Decreased pGluN2B Tyr1472 Inconclusive testing (even WT mice did not show preference for social target) Grooming: n.s. N/A (Lee et al., 2015) 6 Pro domain: InsG3680ex21Feng 2,5,10 Reduced Homer, SAPAP3, SynGap, NR1, NR2A, NR2B, GluR2, mGluR5 Abnormal social interaction; Abnormal social novelty Self-injurious grooming, causing skin lesions PPI: mildly impaired; Acoustic startle response: impaired. (Zhou et al., 2016) Pro domain: R1117Xex21Feng 2,5,10 Reduced Homer, PSD95, SynGap, NR1, NR2B, trend of reduced GluR2 and SAPAP3 Abnormal social interaction; Abnormal social novelty Allogrooming present in heterozygous mice only PPI: mildly impaired; Acoustic startle response: impaired (Zhou et al., 2016) ANK repeat: Δex4-9Powell 3 - 10 Decreased GluA2, GluA3, Homer1 b/c, and PSD95 Mildly impaired Increased selfgrooming PPI: n.s. (Jaramillo et al., 2016) PDZ domain: FLEx-Δex1316Feng 8,9 10 Reduced Homer, SAPA3, NR2A, NR2B, GluA2 Abnormal social interaction; Social novelty: N/A Self-injurious grooming causing skin lesions N/A (Mei et al., 2016) All domains: Δex4-22Jiang All disrupted Reduced PanSAPAP, SAPAP3, and Homer1b/c Normal social interaction and normal social novelty Self-injurious grooming, causing skin lesions PPI: impaired startle; Olfactory discrimination: n.s. (Wang et al., 2016) PDZ domain: Δex13-19 10 Reduced Homer-1 b/c and PSD95 Abnormal social interaction and social novelty Increased selfgrooming PPI and startle response: n.s. (Jaramillo et al., 2017) All domains: Δex4-22Buxbaum All disrupted N/A Increased latency for first male-female interaction, normal social novelty. Increased selfgrooming Cornea reflex, toe pinch retraction, pinna reflex, tail flick: n.s.; Preyer reflex: n.s., Startle response decreased, PPI: n.s.; Visual placing: n.s.; buried food test (olfactory): impaired (Drapeau et al., 2018) 1.3 Sensory processing in ASD: the auditory pathway Sensory-perceptual abnormalities are present in approximately 90% of individuals with ASD (Leekam et al., 2007; Tomchek and Dunn, 2007; Crane et al., 2009), being auditory hypersensitivity the most common sensory-perceptual abnormality (Gomes et al., 2008). By perceiving auditory inputs as noxious or unpleasant, patients may instinctively learn to avoid them (Marco et al., 2011), which could potentially be the root for the communication, socialization and learning impairments observed in ASD. Upon an auditory stimulus, the nervous impulse travels through the auditory nerve (AN) until a relay center in the brainstem, the cochlear nucleus (CN), which is mainly divided in dorsal (DCN) and ventral (VCN) regions. The superior olivary complex (SOC) receives input from the CN and has three nuclei involved in auditory input processing: the lateral superior olive (LSO), the medial superior olive (MSO) and the medial nucleus of the trapezoid body (MNTB). The SOC then projects to the inferior colliculus (IC) through the fibers of the lateral lemniscus (LL), synapsing in the LL nucleus (LLN). From the IC, 7 information travels to the medial geniculate body (MGB), which lies in the thalamus and is the last auditory center before reaching the auditory cortex (AC), conveying information from several regions of the auditory system (Figure 3) (Malmierca, 2003; Knipper et al., 2013; Di Bonito and Studer, 2017). Figure 3. Monoaural ascending auditory pathway. Auditory input arriving to the cochlea is transmitted through the auditory nerve (AN) to the cochlear nucleus (CN) in the brainstem. In this nucleus, AN bifurcates to the ventral cochlear nucleus (VCN) and the dorsal cochlear nucleus (DCN). Information is directed to the ipsilateral (orange arrows) and contralateral (red arrows) SOC (superior olivary complex), travelling through several of its nuclei, namely the lateral superior olive (LSO), medial superior olive (MSO) and medial nucleus of the trapezoid body AN Brainstem Inner ear Midbrain Brain Cochlea Auditory input Acoustic stimulus AC Thalamus AC MNTB VCN DCN LSO MSO LLNLLN 8 (MNTB). Both contralateral and ipsilateral input from the brainstem reach the inferior colliculus (IC) in the midbrain through fibers of the lateral lemniscus (LL), synapsing in the LL nucleus (LLN). From the IC, fibers project ipsilaterally and contralaterally to the medial geniculate body (MGB). The MGB is located in the thalamus and is the brain region that precedes the auditory cortex (AC) in the information flow of ascending auditory pathway. Image adapted from Castro and Monteiro, 2022. 1.4 Excitation and inhibition in primary auditory cortex circuits The primary auditory cortex (A1) is the final stop of the ascending auditory pathway. By receiving and filtering subcortical inputs, the A1 region conveys a cortical representation of the sensory information. Here, sensory inputs are processed by an intricate but highly efficient network of excitatory and inhibitory connections that are perfectly shaped and tightly pruned during development (Sanes and Kotak, 2011). The primary auditory cortex (A1), like the rest of the neocortex, is organized in cortical layers, typically 6 in the rodent brain (L1-6, Figure 4). The balance between excitation and inhibition in sensory regions is an important mechanism that allows the definition of fine receptive fields and ensures temporal precision of the cortical output. This excitation/inhibition (E/I) balance can vary with the cortical layer and is important to filter relevant information within each neuronal circuit (D’Souza and Burkhalter, 2017). Figure 4. Representation of the layered organization of the primary auditory cortex (A1). Representation of a horizontal section of the mouse brain with the primary auditory cortex (A1) region 9 highlighted. In a zoom-in of the A1, it is visible the layered organization of this region, which is typically divided into 6 layers (L1-6). The principal neurons are represented in blue and are typically pyramidal neurons whose dendrites may span from L5 until as far as L1. In pink are represented the local GABAergic interneurons, responsible for modulating the activity of principal cells and other interneurons. Created with Biorender.com. Such excitation/inhibition interplay is supported by two major cellular populations: cortical excitatory neurons (mainly pyramidal cells) that receive and provide sensory input, and inhibitory cells (mainly GABAergic local interneurons) that modulate the activity of pyramidal cells by releasing gammaaminobutyric acid (GABA) (DeFelipe et al., 2002). These cells are organized in a fine microcircuit where the pyramidal cells present input-output connections with either other brain regions or other cortical neurons (DeFelipe et al., 2002). Not only the direct inhibitory effect of interneurons on pyramidal neurons regulates their activity, but also the inhibitory effect of interneurons over other interneurons has also a disinhibitory effect over pyramidal cells (D’Souza and Burkhalter, 2017). But besides this circuit-level regulation, the intrinsic excitability of each neuron also defines the way it will respond to inputs (Chen et al., 2022), being also an important mechanism to maintain E/I balance. The main determinants for intrinsic neuronal excitability are ionic channels. They may be classified according to their gating: voltage or ligand gated channels, or by the cations they allow to flow: Na+, K+, Ca2+ or Cl-. The channel composition of a cell determines its intrinsic properties such as resting membrane potential and input resistance, which closely relate to its excitability. Briefly, when the depolarizing signal (the action potential) reaches an axon terminal, voltage-sensing Ca2+ channels open and the Ca2+ intake leads to neurotransmitters release. These neurotransmitters then bind to their respective receptors on the post-synaptic neuron, causing the opening of ligand-gated ionic channels that will allow the inward flow of cations and the depolarization of the post-synaptic neuron. The magnitude of this process depends on the frequency by which action potentials arrive to the synapse but also on the channel composition of the synapse. The depolarization signal of the post-synaptic membrane travels to the soma of the neuron in the form of a graded potential. This means that, depending on timing and frequency by which other graded potentials reach the soma, even a small membrane depolarization may sum and allow the generation of another action potential in a phenomenon called dendritic integration (Stanfield, 2013). In this process, the presence of other voltage-gated channels such as the hyperpolarization cyclic nucleotidegated channels (HCNs) is determinant. These are channels activated upon hyperpolarizing voltages and are organized in a gradient throughout the dendrites of L5 pyramidal neurons. The great density of HCN 16 2013). When KO mice are exposed to trains of noise, the obtained responses differ between the first and subsequent stimulus. Initially, weaker responses are recorded at the AAF and ventral auditory field (VAF), but after several trains of noise, VAF presents weaker response, and the posterior auditory field (PAF) becomes stronger. The overall most affected area seems to be VAF (Engineer et al., 2014b). Fmr1 -KO also present less selectivity to frequency modulated (FM) sweep rates and higher response to fast sweep rates, without differences regarding direction selectivity (upor downward sweeps) (Rotschafer and Razak, 2013). Regarding AC response to speech sounds, evoked LFPs display lower peak amplitudes (N1, P2, N2 and P3 components) in Fmr1 -KO. When splitting AC data into four fields (AAF, A1, VAF, and PAF), A1 and VAF show the weakest amplitudes for all components (Engineer et al., 2014b). Furthermore, onset latency in PAF and VAF is increased but temporal precision and strength to speech sounds are reduced (Engineer et al., 2014b). Fmr1 -KO mice are reported to have higher ABR thresholds for click sounds and pure tone frequencies, and reduced peak I and III amplitudes, suggesting defects in AN and brainstem. Response latency to sound stimulation and inter-peak latency are not altered (Rotschafer et al., 2015). Histological assessments in the auditory brainstem of adult Fmr1 -KO mice further reveal that neuronal cell size is reduced in VCN, MSO and MNTB (not in LSO) (Rotschafer et al., 2015; Ruby et al., 2015). Interestingly, although several cellular defects can be detected since birth at MNTB, these only emerge at VCN after hearing onset (Ruby et al., 2015; Rotschafer and Cramer, 2017). Of note, FMRP is expressed across the auditory brainstem in a specific pattern. In MNTB it has a tonotopical expression similar to potassium channel Kv3.1b, a channel that seems crucial for normal sound processing specifically in the binaural sound localization circuit (Strumbos et al., 2010; Ruby et al., 2015). In Fmr1 - KO, however, the typical expression pattern of Kv3.1b in MNTB is not observed, namely its tonotopical gradient or increased expression upon sound stimulation (Strumbos et al., 2010; Ruby et al., 2015). Additionally, Fmr1 -KO mice present reduced axonal profiles and CB immunoreactive terminals in MNTB somata which may indicate abnormal Ca2+ signaling regulation (Ruby et al., 2015). Focusing on two important auditory nuclei of the SOC (MNTB and LSO), one study found a greater strengthening of excitatory input from the VCN to the LSO in Fmr1 -KO, likely due to an increased number of synaptic connections (Garcia-Pino et al., 2017). Besides receiving excitatory input from the VCN, the LSO receives inhibitory input from the MNTB, which seems unaffected in Fmr1 -KO mice. Therefore, the net effect is a potential increase in excitation. Such changes in E/I balance could explain the increased firing rates and broadened tuning curves observed in the LSO of Fmr1 -KO mice and might contribute to their reported acoustic hypersensitivity (Garcia-Pino et al., 2017). 17 Auditory stimulation with tone bursts and amplitude-modulated tones leads to increased activation of IC in Fmr1 -KO mice, especially in neurons that respond to lower frequencies (<20kHz). Broader tuning frequency is generally observed in individual neurons (Nguyen et al., 2020). Although the development of connections within the auditory circuitry seems to occur normally in Fmr1 -KO, sound-driven refinements of excitatory inputs seem affected by the time of hearing onset (~P10-P11) (Nguyen et al., 2020). Because these refinements depend on precisely timed inhibitory inputs, it is thought that impairments in GABAergic signaling may cause E/I defects in IC. Of note, neurons in the dorsal region of the IC present a CF <20kHz, being the most responsive in Fmr1 -KO and more hypersensitive. Given the prominent role of GABA in the dorsal region of the IC, it seems that GABAergic dysfunction may underlie abnormal responsiveness in this area (Nguyen et al., 2020). In a broader picture, there is evidence consistent with both an increase in excitation mediated by strengthening of excitatory inputs and decrease in inhibition due to impairments on GABA-related neurotransmission. Thus, the E/I imbalance often described as a hallmark of ASD seems to hold true in the Fmr1 -KO model. At the molecular level, one of the possible candidates underlying this imbalance and contributing to auditory hypersensitivity in Fmr1 -KO is matrix metalloproteinase-9 (MMP-9). This enzyme is upregulated in the AC of Fmr1 -KO mice and has been associated with its reduced ERP habituation (event related potentials) (Lovelace et al., 2016, 2018, 2020b). Furthermore, MMP-9 upregulation likely hinders the formation of perineuronal nets (PNNs) around PV+ cortical interneurons (GABAergic neurons), leading to reduced cortical network inhibition and E/I imbalance in the AC (Wen et al., 2018). Genetic or pharmacological strategies for reducing MMP-9 levels seem to rescue the following phenotypes: 1) audiogenic seizure susceptibility, 2) AC activity (spontaneous and evoked), 3) auditory ERP habituation, 4) formation of PNNs around PV+ cells, 5) anxiety-like and hyperactive behaviors, 6) ASR (Gkogkas et al., 2014; Lovelace et al., 2016, 2020a; Wen et al., 2018; Kokash et al., 2019; Pirbhoy et al., 2020). Enhancement of endocannabinoid production (2-arachidonoyl-sn-glycerol) seems to ameliorate synchrony in the cortical response to auditory stimuli, and rescue anxiety-like and hyperactive behaviors (Pirbhoy et al., 2021). Inhibiting phosphodiesterase 10A seems an additional therapeutic approach since it was shown to improve auditory processing in Fmr1 -KO mice (Jonak et al., 2021). Interestingly, sound exposure during postnatal development seems to rescue/normalize ERP, PV+ cell density and dendritic spine density in the AC (Kulinich et al., 2020). 18 1.5.2.2 α 7-nAChR knockout CHRNA7 is a gene encoding for α7-nicotinic acetylcholine receptor (α7-nAChR), an homopentameric transmembranar protein highly expressed in the brain (Schaaf, 2014). α7-nAChR expression starts during prenatal development and peaks during the first synaptogenesis events. Changes in channel expression negatively influence neurogenesis, synaptogenesis and neuroblasts’ migratory events (De Jaco et al., 2016). In humans, microduplications in this gene are associated with a wide range of neurobehavioral disorders, including ASD (Dennis et al., 2012). In rodents, α7-nAChR loss is associated with developmental impairments that affect auditory processing. ABR hearing thresholds and peak amplitudes are unaffected, but peak IV latency is increased, suggestive of impairments in the midbrain. Accordingly, single unit responses recorded in the IC revealed that KO animals have a subset of neurons with an atypical response to pure tones, presenting also deficits in spike timing, forward masking, and silent gap detection (Felix et al., 2019). In more detail, evoked responses in the IC are typically transient (1to 40-ms duration) and sustained (80to 120-ms). In the KOs, however, a third type of response can be detected with intermediate response (40to 80-ms) (Felix et al., 2019). Degraded spike timing was also found in the VNLL and SPON, which are primary targets of octopus cells (highly temporal precise cells likely responsible for shaping temporal responses in the midbrain) (Felix et al., 2019). 1.5.2.3 Cntnap2 knockout Contactin-associated protein-like 2 (CNTNAP2) is encoded by an ASD-related gene implicated in language impairments (Rodenas-Cuadrado et al., 2014). The CNTNAP2 is responsible for tethering potassium channels in myelinated axons, being crucial for action potential propagation (Truong et al., 2015). Loss of CNTNAP2 in rodents leads to alterations in social behavior, reduced USVs and hyperactivity (Peñagarikano et al., 2011; Brunner et al., 2015). Conclusions regarding sensory-motor gating ability of Cntnap2 -KO mice are not consensual. Different studies report either unchanged (Peñagarikano et al., 2011), impaired (Scott et al., 2018), or more efficient PPI (Brunner et al., 2015; Truong et al., 2015). The same is observed regarding ASR whose changes may impact PPI and explain results disparity (probably also age factor). A battery of tests based on PPI paradigm have been performed in Cntnap2 -KO. By changing the type of prepulse cue, it is possible to detect impaired ability to perceive short SGs in a continuous broadband white noise background, but increased ability to discriminate slight changes in pitch (Truong et al., 2015). In order to elicit a startle inhibition with SGs, these must be much longer, suggesting that Cntnap2 -KO have impairments in 19 temporal sound processing. Ability to attenuate a startle response when a pitch cue was presented was, however, independent of pitch tone duration. The auditory alterations reported in this model seem to be linked to MGB neuronal changes (Truong et al., 2015). Analysis of ABR to click sounds and pure tones in Cntnap2 -KO rats shows altered peak amplitudes and latencies but overall unaffected hearing thresholds. In more detail, peaks II, III and IV latencies were observed to be consistently increased in juvenile rats, although this trait was recovered in adulthood. The amplitude of peak IV was decreased across development and in adulthood. Interpeak latencies were also affected, mostly the latency between peaks I-II which was decreased both during development and adulthood (Scott et al., 2018). Histological data shows reduced neuronal count and size in the MGB, which may help explain sound processing abnormalities. Given the results from ABR testing, further histological data would be useful to clarify the extent of changes in this model, similarly to other ASD models presented in this review (Truong et al., 2015; Scott et al., 2018). 1.5.2.4 Shank3 knockout Being SHANK3 haploinsufficiency a clear monogenic cause of ASD, dozens of mouse lines carrying different Shank3 mutations have been generated so far (Monteiro and Feng, 2017). One in particular, carries a human genetic mutation from an ASD patient - the InsG3680-Shank3 mouse line – making it a valuable translational model in preclinical research. This rodent model displays increased ASR, which indicates a possible impairment in sound processing and potential auditory hypersensitivity. PPI seems to be decreased in this model, although the baseline acoustic reactivity may portrait a confounding factor to this test, as discussed by the authors (Zhou et al., 2016). In a different model, the Shank3B -KO, no differences were found in PPI or silent gap (SG) test (Figure 5). On the other hand, Shank3B -KO seem to have increased ability to discriminate pitch changes (Rendall et al., 2019). Impairments in cortical sound processing are reported for heterozygous Shank3 -deficient rats carrying a 68 bp deletion in exon six with a premature stop codon (Engineer et al., 2018). Cortical response to tone stimulus, noise bursts, and speech sounds, are overall weaker in these rats, but temporal properties seem to be unaltered in response to tones. The same is not true for response to speech sounds, that besides being weaker, is also delayed. Of note, although neural discrimination accuracy does not seem to be impaired when sounds are presented in an isolated manner, neural responses are degraded when sounds are presented at increased rates, such as human speech rate (Engineer et al., 2018). 20 1.5.2.5 Pten conditional knockout PTEN mutations have been identified in individuals diagnosed with ASD and also displaying macrocephaly. Accordingly, PTEN -KO mice present several alterations namely increased neuronal soma size, hypertrophic dendrites, higher excitatory spontaneous activity, and hypertrophic and ectopic dendrites (Xiong et al., 2012). Such alterations are also consistent with the known role of this gene, which is required for normal brain wiring and development. Pten -KO pups tend to present increased frequency of USVs when separated from their mothers, a result interpreted as evidence of higher anxiety (Sarn et al., 2021). Conditional KO of Pten in the left AC of mice increases the strength of callosal inputs to that region and the efficiency of excitatory long-range synaptic inputs from contralateral AC and thalamus. An increase in dendritic branch number and spine density together with increased amplitude of miniature excitatory postsynaptic currents is consistent with overall strengthening of synaptic connections in this region (Xiong et al., 2012). 1.5.2.6 Mecp2 transgenic mouse The MECP2 gene encodes methyl-CpG binding protein (MeCP2) that acts as a regulator of gene expression, playing an important role in prenatal neurogenesis and postnatal synaptic development, function, and plasticity (Brand et al., 2021). MECP2 genetic loss of function is associated with intellectual disability and Rett syndrome (Brand et al., 2021), whereas its duplication is characterized by motor and cognitive impairments, delayed or absent speech, seizures, and ataxia (D’Mello, 2021). In both situations, ASD-related phenotypes are often times present. Altered sound-evoked cortical responses have been reported in Mecp2overexpressing transgenic mice ( Mecp2 -TG). Although displaying normal CF distribution, thresholds to trigger tone-evoked cortical responses are increased. In contrast, cortical responses to noise are stronger, but delayed. Such abnormalities indicate a noise sensitivity phenotype associated to fast-spiking neurons that might be due to lack of cortical inhibition (Zhou et al., 2019a). Unlike many other ASD models, ABR is normal in Mecp2 overexpressing mice and cortical tonotopy does not seem to be affected (Zhou et al., 2019a). 1.5.2.7 Other ASD models displaying auditory related impairments Although far less explored, there are other animal models that potentially display auditory processing abnormalities. The Cyfip1+/- mouse model presents a lower PPI, evidencing sensory-motor gating impairments (Domínguez-Iturza et al., 2019) and the Nrxn1 α-KO presents increased ASR without 21 changes in PPI, perhaps due to acoustic hypersensitivity (Esclassan et al., 2015). However, PPI results in Nrxn1 α-KO rats may be confounded by their decreased ASR (Esclassan et al., 2015). Adnp+/- mice display increased ABR thresholds and latency, as well as decreased number of USV calls. Such findings might be a consequence of the reported significant hearing loss displayed by Adnp+/- mice (HacohenKleiman et al., 2019). 1.6 Open questions and future perspectives Increasing evidence from animal models demonstrates that auditory-perceptual alterations found in ASD patients can be recapitulated in several animal models. Such neurodivergent processing of auditory inputs, translated into hypoor hypersensitivity to sensory stimulation, may have a strong impact on behavior and impose limitations in the quality of life for individuals diagnosed with ASD. In particular, auditory processing abnormalities may underlie deficits in communication and social interaction. Tackling the neurobiological mechanisms causing such alterations becomes of utmost importance to design strategies to attenuate or prevent sensory impairments. Rodent models are a powerful resource to better understand behavioral and neurobiological alterations in ASD, holding tremendous translational potential. Despite the great diversity of ASD models, mirroring the great heterogeneity in the etiology of ASD, it is possible to identify shared features across models. Auditory impairments seem to arise from deficits in central processing rather than from periphery. Along the auditory pathway, multiple defects are observed, such as decreased tonotopicity, altered thresholds to sound stimuli, and abnormal spectral and temporal processing (especially in the auditory regions of the brainstem and cortex). The origin of these differences is still not fully understood, but E/I imbalances during postnatal development seem to be contributing to these defects, mainly due to impairments in GABAergic signaling. Future work is needed to better support these observations and unveil specific regions, neuronal circuits, and molecular players that are determinant for the auditory phenotype. Since ASD is a neurodevelopmental disorder, clarifications on critical developmental stages will be crucial, together with the establishment of novel molecular targets that might be particularly effective during those developmental windows. Together, this knowledge will hopefully help defining efficient therapeutic approaches in the near future. 22 Table 2. Overview of the most relevant findings regarding ASD rodent models with reported auditory dysfunctions. Abbreviations: acoustic startle response (ASR), prepulse inhibition (PPI), phosphatidylinositol-3-kinase (PI3K), auditory cortex (AC), auditory brainstem response (ABR), medial geniculate body (MGB), inferior colliculus (IC), ventral nucleus of the lateral lemniscus (VNLL), superior paraolivary nucleus (SPON), medial nucleus of the trapezoid body (MNTB), valproic acid (VPA). Model Gene Mouse locus Human locus Protein Protein function Main findings Shank3 -KO Shank3 15; 15 E3 22q13.3 SHANK3 (SH3 and multiple ankyrin repeat domains protein 3) Scaffold protein important for organization of excitatory synapses and integrity of the post-synaptic density (Monteiro and Feng, 2017). Increased ASR, decreased PPI, increased ability for pitch discrimination (Zhou et al., 2016; Rendall et al., 2019). Weaker cortical responses to sounds, temporal processing is impaired when sounds are presented at high rate, resembling speech rate (Engineer et al., 2018). Pten conditional KO Pten 19 C1; 19 28.14 cM 10q23.31 PTEN (phosphatase and tensin homolog) Negative regulator of the PI3K/AKT signaling, important for cell growth, proliferation and survival, and axonal growth (Zhou and Parada, 2012; Sarn et al., 2021). Strengthened synaptic connectivity: stronger callosal inputs to the AC from contralateral AC and thalamus (Xiong et al., 2012). Cntnap2KO Cntnap2 6; 6 B2.2B2.3 7q35-q36.1 CNTNAP2 (Contactin associated protein 2) Maintenance of normal action potential propagation through tethering of potassium channels in myelinated axons (Truong et al., 2015). Increased pitch discrimination, but impaired temporal processing of sounds. Altered ABRs (peak amplitudes and latencies), histological alterations in MGB (Truong et al., 2015; Scott et al., 2018). α 7-nAChR -KO Chrna7 7 C; 7 34.47 cM 15q13.3 α7-AChR (α7-nicotinic acetylcholine receptor) Ligand-gated cationic channel that interacts with acetylcholine and mediates synaptic transmission (Schaaf, 2014; De Jaco et al., 2016). Increased latency of ABR peak IV, atypical spectral and temporal response of IC neurons. Degraded spike timing in VNLL and SPON (Felix et al., 2019). Fmr1 -KO Fmr1 X A7.1; X 34.83 cM Xq27.3 FMRP (Fragile X Mental Retardation Protein) Actin remodeling, regulation of cytoskeleton-related gene expression (Braun and Segal, 2000). Behavioral evidence of deficits in sound processing and hyperacusis (Reinhard et al., 2019; Auerbach et al., 2021). Altered cortical responses to different types of stimuli (Rotschafer and Razak, 2013), higher ABR thresholds (Rotschafer et al., 2015) and lack of tonotopicity in MNTB (Strumbos et al., 2010; Ruby et al., 2015), IC (Nguyen et al., 2020) and AC (Rotschafer and Razak, 2013). Mecp2 -TG Mecp2 X A7.3; X 37.63 cM 2p16.3 MECP2 (methyl-CpG binding protein 2) Mainly expressed in the brain. Gene expression regulator, important for prenatal neurogenesis and synaptic development (Brand et al., 2021). Normal ABR and cortical tonotopy. Increased thresholds for toneevoked cortical responses, heightened and delayed cortical responses to noise (Zhou et al., 2019a). VPA N/A N/A N/A N/A N/A Reduced USVs throughout life, normal ASR but reduced PPI suggesting sensory-motor gating impairments (Gandal et al., 2010). Cortical alterations on LFPs, temporal and spectral processing, and tonotopicity (Engineer et al., 2014a; Anomal et al., 2015). Auditory brainstem overactivation upon sound stimulation (Dubiel and Kulesza, 2016). Impaired connectivity between the brainstem and the midbrain, which is one of the most susceptible regions to VPA exposure (Zimmerman et al., 2020). Thalidomide N/A N/A N/A N/A N/A Impairments in MNTB: diminished in size, fewer projections, expanded responsive area to sound stimulation (Ida-Eto et al., 2017; Tsugiyama et al., 2020). 23 2 RESEARCH OBJECTIVES Sensory-perceptual abnormalities are present in approximately 90% of individuals with ASD being auditory hypersensitivity one of the most commonly reported sensory-perceptual abnormalities. This translates into a huge impact on patients’ life quality and may even underlie other ASD-related traits such as communication difficulties. Several animal models used in ASD studies share traits of auditory dysfunction, namely hypersensitive auditory phenotypes, which could suggest a shared upstream mechanism among different animal models of ASD. Accordingly, our research group has been studying auditory sensory function in a transgenic mouse line that carries an ASD-patient derived mutation – the Shank3-InsG3680+/+ mice. In line with the literature, we found that these mice present not only typical ASD-like behaviors such as social interaction deficits and repetitive behaviors, but also behavioral auditory hypersensitivity (unpublished). In order to begin unveiling potential molecular mechanisms that might underlie auditory sensory dysfunction in the Shank3-InsG3680+/+ mouse model of ASD, we decided to focus this dissertation work on the A1 brain region (primary auditory cortex). It is our aim to: 1. Identify unique molecular signatures in the primary auditory cortex of Shank3-InsG3680+/+ mutant mice; 2. Evaluate the emergence of such molecular signatures along development; 3. Study those alterations at the molecular and functional levels and evaluate potential cell-type specific profiles. This work will contribute to a better understanding of the molecular and cellular changes occurring in the developing ASD brain. Results may potentially uncover novel molecular targets which may hold future therapeutic potential in ASD. 24 3 METHODS 3.1 Animals Animals used in this study (C57/BL6 male Shank3-InsG3680+/+ and WT controls) were housed in an animal facility approved by Portuguese national authorities (Direção Geral de Alimentação e Veterinária - DGAV). Housing was performed at standard conditions (temperature 20-24 ºC, humidity 55 ± 10%), with a light/dark cycle of 12/12 h. Food (Mucedola: 4RF25 during gestation and before weaning, 4RF21 post weaning) and water were available ad libitum . All animals were housed in polysulfone cages (1264C EUROSTANDARD type II or 1284L EUROSTANDARD type II L, Tecniplast) in groups of 5 to 6 mice respectively. The floor of the cages was covered by corncob bedding and enriched with tissue paper before weaning and with a combination of tissue paper and paper strips after weaning, both to enable naturalistic nesting behavior. All animal procedures were approved by national authorities Direção Geral de Alimentação e Veterinária (ID: DGAV 8519), the Ethics Subcommittee for Life Sciences and Health (SECVS) of University of Minho (ID: SECVS 01/18) and were performed in accordance with European Community Council Directives (2010/63/EU) and the Portuguese law DL Nº 113/2013 for the care and use of laboratory animals. 3.2 Genotyping Genotype of the animals used in this study was confirmed through polymerase chain reaction (PCR) analysis. Between postnatal days 10 and 14, a tail sample was collected and preserved at -20 ºC until further processing. For DNA extraction, a strong alkaline solution (25 mM NaOH, 0.2 mM EDTA, pH 12) was added to each tail sample. After 30 minutes incubation at 95 ºC, a neutralization buffer (40 mM TrisHCl, pH 5) was added in 1:1 ratio. The PCR for Shank3 fragment amplification was performed with MyTaq PCR Mix (Bioline, BIO-25041: Taq buffer containing MgCl, 10 nM dNTPs (dCTP, dGTP, dATP, dTTP), Taq polymerase enzyme, template DNA from each tail sample, forward and reverse primers: Primer forward - 5’ AGCTGGCCTCATTGTTGTG 3’ Primer reverse - 5’ CAGGTTCACAGCGAATACCA 3’ Samples from known WT, Shank3 - InsG3680+/+ and Shank3 - InsG3680+/- animals were used as positive controls together with a negative control without any DNA sample. The initial denaturation step lasted for 1 minute at 95 ºC followed by 35 repeated cycles of: denaturation at 95 ºC, primer annealing at 58 ºC and polymerization at 72 ºC. All these three steps lasted for 15 seconds each, at every cycle. After this, samples were kept at 8 ºC. The PCR product was mixed with bromophenol blue (1:1 ratio) 25 and loaded to a 1.5% agarose gel containing GreenSafe and submerged in Tris/Borate/EDTA buffer (TBE). An electric current of 100 V was applied during ≈1.5 h for DNA migration and band resolution. The gel image was obtained with Gel-Doc (BIO RAD) and the genotype was determined by comparison of the obtained bands with the above-mentioned positive controls. 3.3 Behavioral tests All behavioral tests were performed at nighttime, which corresponds to the active period of mice. The mice used were 7 weeks-old (Figure 8A). Before testing, animals were habituated to the room for at least 1 hour under the same lighting conditions as the experimental test. 3.3.1 Splash test Each animal was placed in an individual polysulfone box (42,5×26,6×18,5 cm; clean housing cage) containing corncob bedding covering the entire floor and evenly exposed to dim light (100 – 110 lux). At beginning of the test, animals were sprayed on their dorsal coat with a 30% sucrose solution and a transparent acrylic cover was placed on top of the cage to allow animal visualization and videotaping for 10 minutes (Figure 6). The number and duration of grooming bouts, rears and digs were manually counted from the resulting videos. Grooming bouts comprise nose and face grooms (circular movements around these regions), head grooms (circular movements encompassing the top head of the animal and the ears) and body grooms (body fur scratching or licking) (Yalcin et al., 2005). Figure 6. Schematics of the splash test paradigm. The animal is placed in a clean homecage containing bedding on the floor. The animal’s back is then sprayed with a 30 % sucrose solution and the resulting behavior is videotaped for 10 minutes. Created with Biorender.com. 32 The level of confidence used for all statistical tests was 95% unless stated otherwise. Results are reported with mean and standard deviation (SD) (see supplementary tables in annexes for detailed statistical reports) and plotted with mean and standard error of the mean (SEM). Statistical significance was considered when the p-value ( p ) ≤ 0.05, unless stated otherwise. Figure 8. Schematics of the study design used for molecular and functional experiments. A B 33 A) Experimental timeline indicating the timepoints at which the different procedures were performed. B) Schematic representation of the study design. For molecular biology and ex vivo electrophysiology experiments, acute horizontal brain sections were obtained to extract the primary auditory cortex (A1). Mice were deeply anesthetized with avertin and transcardially perfused with NMDG-based solution. Brain was quickly extracted and sectioned in 300 µm-thick slices using a vibratome. A1 region was macrodissected and used for protein or RNA extraction. After protein quantification, protein homogenates were used for proteomics screening or western blot analysis. For ex vivo electrophysiology, the brain slices were recovered in NMDG-based solution at 32 ºC for 11 minutes and incubated in regular aCSF at room temperature (RT) for 1 h and afterwards recordings were performed. Created with Biorender.com 34 4. RESULTS 4.1 Genotyping protocol allows identification of mutant mice All mice genotypes were confirmed by PCR analysis where a fragment of the Shank3 gene was amplified. This fragment comprised the guanine insertion locus ( InsG3680 ), allowing the distinction between all possible genotypes (Figure 9): - InsG3680+/+ mutants: longer PCR fragment, producing a band of 310 base pairs (bp) due to the presence of the guanine insertion with leftover cassette from the original targeting vector; - Wild type (WT) / InsG3680-/- mice: shorter PCR fragment, producing a band of 235 base pairs (bp) due to the absence of the extra guanine and its associated leftover transgenic cassette; - InsG3680+/- mice: presence of both fragments (310 and 235 bp), indicating the presence of a WT allele as well as a mutant allele. Figure 9. Representative genotyping gel. Image shows an agarose gel containing several gene amplification products. The bands obtained in the testing samples are identified by comparison to the positive controls. The fragments containing the guanine insertion mutation correspond to the higher molecular weight bands. Hence, samples containing a single band of higher molecular weight represent transgenic homozygous mice (Shank3-InsG3680+/+ ). Samples displaying only a lower molecular weight band represent wild type mice ( InsG3680-/- ). Samples contain both alleles ( InsG3680+/- ) represent heterozygous mice which were not used in this study. Negative control does not contain any DNA sample. 4.2 InsG3680+/+ mice present repetitive grooming behavior After successful establishment of a transgenic mouse colony of Shank3-InsG3680+/+ mice, we proceed to evaluate their ASD-like behavioral phenotypes. Results show that InsG3680+/+ mice have Positive controls Testing samples Negative control 35 reduced social interaction as well as auditory hypersensitivity (data obtained by other lab members; data not shown), hence, we proceed to evaluate their grooming behavior. Repetitive and stereotyped behaviors are one of the core features of ASD (American Psychiatric Association, 2013). This feature is also quite extensively reported for animal models of ASD (Gandhi and Lee, 2021), including different Shank3 mutant mouse lines (Peça et al., 2011; Yoo et al., 2019; Liu et al., 2021). Increased spontaneous grooming has been previously reported in InsG3680+/+ mice (Zhou et al., 2016). Here, we used the splash test to investigate induced grooming behavior. The splash test begins by spraying mice’s fur coat with a sucrose solution, followed by behavioral videotaping for 10 minutes. The solution triggers a self-care grooming behavior in order to remove the sticky sucrose solution from the fur (Pitzer et al., 2022). Results revealed that InsG3680+/+ mice spent significantly more time grooming compared to WT control group (Figure 10, A). Besides total time, the number of grooming bouts was also significantly increased in the mutant group (Figure 10, B). When analyzing the videos, we noticed that other behavioral patterns also seemed altered. Hence, exploratory and naturalistic behaviors such as rearing and digging were also quantified. Data revealed that InsG3680+/+ mice tended to spend less time rearing compared to WT controls, although the difference did not reach statistical significance. The same tendency was also observed in the number of rearing bouts (Figure 10, C and D). Strikingly, InsG3680+/+ did not display any digging behavior during the entire duration of the test while WT controls spent approximately 10% of their time digging (Figure 10 E and F). Together, these findings show that InsG3680+/+ mice present repetitive grooming and reduced exploratory behavior (Figure 10G and H), perhaps due to high anxiety levels. Such behavioral alterations add face validity to their construct validity as a rodent model of ASD. Given our hypothesis that sensory alterations in the auditory pathway, namely auditory hypersensitivity, may lead to social withdrawal, acoustic avoidance and promote self-centered behaviors, we proceeded to explore the molecular landscape of primary auditory cortex brain region in InsG3680+/+ mice. 36 Figure 10. Behavioral results from the splash test: InsG3680+/+ mice spend more time grooming and less time digging and rearing. Splash test successfully induced grooming behavior in both genotypes. However, InsG3680+/+ mice spent significantly more time grooming (A) and had a higher number of grooming bouts (B) when compared to wild type (WT) controls. Although not statistically significant, the time spent rearing (C) and the number of rearing events (D) display a trend towards reduction in InsG3680+/+ mice. Digging behavior was absent Wild type InsG3680 +/+ 0 100 200 300 400 500 Time grooming (s) * Wild type InsG3680 +/+ 0 100 200 300 400 500 Time rearing (sec) n.s. Wild type InsG3680 +/+ 0 100 200 300 400 500 Time digging (sec) Wild type InsG3680 +/+ 0 20 40 60 Grooming bouts * Wild type InsG3680 +/+ 0 20 40 60 Rearing events n.s. Wild type InsG3680 +/+ 0 20 40 60 Digging events A B C D E F GWild type H InsG3680+/+ 37 in InsG3680+/+ mice during the splash test (E, F). Illustrative timestamps for one WT (G) and one InsG3680+/+ mouse (H) showing the distribution of grooming, digging and rearing events during the entire duration of the splash test (G,H). Two-tailed unpaired Student’s t -test, n = 5 mice per group (male, 7 weeks old); Data are mean ± SEM; *p < 0.05, n.s. not significant. 4.3 InsG3680+/+ mice have HCN1 overexpression in A1 region To begin unveiling potential molecular underpinnings of auditory hypersensitivity, we performed a shotgun proteomics analysis of the primary auditory cortex (A1) of adult (6and 10-week-old, 6W and 10 W) InsG3680+/+ and WT control mice. From 4144 proteins identified, 2862 were confidently quantified by SWATH-MS. Then, using the online open-access tool REACTOME, we restricted our analysis to proteins involved in neuronal pathways (139 proteins). For each of those, the InsG3680+/+ relative protein level was normalized to its matching WT sample (Figure 11A). Several proteins were found to be significantly altered in InsG3680+/+ mice (Figure 11B). The top 10 upregulated and downregulated proteins are represented in Figure 11C and 11D. As expected, SHANK3 was the top downregulated protein, consistent with the known effects of InsG3680 mutation where the insertion of a single guanine nucleotide (G) at cDNA position 3680 leads to a frameshift mutation and a downstream stop codon. One of the most upregulated proteins found in adult InsG3680+/+ mice was the hyperpolarization cyclic nucleotide gated channel 1 (HCN1), a cationic channel whose activity is modulated by, among others, cyclic adenosine monophosphate (cAMP) (Biel et al., 2009). 38 Figure 11. Proteomics screening of primary auditory cortex (A1) from adult InsG3680+/+ mice. (A) Heatmap showing all A1 proteins (y-axis) from the “neuronal system” family according to the REACTOME database. Each column represents InsG3680+/+ protein levels normalized to the WT levels. -2 -1 0 1 2 0 1 2 3 4 log2(fold change) -log10( p -value) SHAN3 HCN1 SNP25 GBG3 ADCY1 NRX2A ADCY5 ADCY9 GBB4 PTPRSGBB1 p = 0.10 A B C D ADCY9 LRC4B KCNC2 GNAI3 SLIK1 NRX2A PTPRS HCN1 ADCY5 ADCY1 0.0 0.5 1.0 1.5 2.0 Fold change SHAN3 GBG3 SNP25 GBB1 GBB4 VAMP2 TSN7 GBB2 GBRB2 GNB5 0.0 0.5 1.0 1.5 2.0 Fold change KCNA6 ADCY1 LIPA2 AOFA ADCY5 GBG4 HCN1 KKCC2 SYN3 PTPRS NRX2A KCC2D FLOT2 DLG4 DCE1 GBRA3 SLIK1 FLOT1 GNAI3 KAP1 KCNC2 KCC4 CACB3 LRC4B ADCY9 GBRA1 ADCY2 SYN1 ACTN2 GRIA4 SYT1 NLGN2 SYN2 NCALD GBRB1 GBRA4 KCC1A AP2M1 KCJ10 DLG1 APBA1 NLGN3 KAPCA KCND2 DLGP4 LRRC7 KAP2 KCC2G HCN2 RTN3 LIPA3 E41L2 GBRB3 NMDE2 KCC2B PLCB1 STXB1 CACB4 CAC1A DLG3 SHAN1 AP2A2 GBRA2 DCE2 VIAAT GRM5 CA2D3 KCMA1 DBNL PRAF3 S6A11 DNJC5 GABR1 KPCG VGLU1 KCNA2 NRX3A PTPRD KAP0 NSF GNAI1 GRIA2 GBRG2 KAPCB GRIA3 AP2S1 PICK1 KCC2A NMDZ1 E41L3 STX1A KCNA1 DLGP3 DLGP2 COMT GRIA1 GABR2 ARHG9 SSDH GBG2 HSP7C CPLX1 HOME1 SC6A1 CCG2 KAP3 CSKP RAB3A GLSK GABT GNAI2 AP2B1 KCAB2 KCNQ2 AP2A1 KPCB NRX1A NFL EAA1 GBG10 RIMS1 LRRT2 ALDH2 GLNA LIN7A EAA2 CAC1E GNB5 GBRB2 GBB2 TSN7 VAMP2 GBB4 GBB1 SNP25 GBG3 SHAN3 6 weeks 10 weeks 0.5 1.0 1.5 2.0 39 Protein fold-enrichment is color coded (blue: decreased expression; red: increased expression relative to control). (B) Volcano plot representing neuronal proteins identified in the proteomics screening. For a level of significance of 0.10, the ten proteins with the biggest fold change difference are depicted in the violin plots (C,D). One-sample Student’s t -test, n = 4 per genotype (each sample corresponds to A1 tissue combined from 3 mice, aged 6 or 10 weeks). Interestingly, three different isoforms of adenylate cyclase (ADCY1, ADCY5 and ADCY9) are also among the most upregulated proteins. Knowing that ADCYs catalyze the conversion of ATP (adenosine triphosphate) to cAMP upon activation of G-coupled proteins (Figure 12), these results suggest that the upregulation of this set of proteins might increase the intracellular concentration of cAMP and consequently have an effect on HCN activity, suggesting a strong convergence on this pathway. Figure 12. ADCY indirect modulation of HCN activity through cAMP. The activation of G-coupled protein receptors (GPCRs) leads to the subsequent activation of adenylate cyclase (ADCY) through GPCR subunit α. The ADCY promotes the conversion of ATP to cAMP which is an important secondary messenger in the cell. cAMP also binds to the cyclic nucleotide binding domains of HCN receptors, modulating their activity by accelerating their opening kinetics and producing a positive shift on their half-maximum opening voltage (V0.5). Created with Biorender.com 40 4.4 HCN1 overexpression seems to emerge after P21 Given that ASD is a neurodevelopmental disorder, we wondered whether this HCN1 upregulation could be noticeable before adulthood. During postnatal development, there are time windows known as critical periods of plasticity in which sensory regions (such as A1) present increased plasticity. During these periods, the brain regions associated to sensory processing are very immature and particularly sensitive to any incoming sensory inputs. After that, critical windows close and sensory circuits become less plastic, but more stable, in adulthood. The development/maturation of these regions occurs sequentially in the brain and is highly influenced by the environmental stimulation they receive. In mouse A1 specifically, there are two critical periods of plasticity: one where A1 refines its response to pure tones (usually occurring from postnatal day P12 to P15), and another where A1 refines its response to frequency modulated sweeps (FMS, occurring from P31 to P38) (Nakamura et al., 2020). Having in mind these different critical periods in mice, we collected A1 samples at four different stages for proteomic analysis: 1) P10, before any critical period begins in A1; 2) P14, during A1 critical period for pure tones; 3) P21, in between A1 critical periods; and 4) P35, during A1 critical period for FMS. From 4132 proteins identified, 2739 were confidently quantified by SWATH-MS. Among those, 155 belonged to neuronal pathways according to the REACTOME database. Relative protein abundance was then normalized to WT levels for all developmental timepoints (Figure 13A) and then the average expression fold-change was statistically compared between WT and InsG3680+/+ mice (Figure 13B). The five most up and downregulated proteins are represented in Figure 13C and D, respectively. Once again, data revealed that SHANK3 was the top downregulated protein as previously observed in our proteomics dataset from adult animals. HCN1, however, was not significantly upregulated, suggesting that HCN1 overexpression in A1 occurs only in the adult stage rather than during development. 41 Figure 13. Proteomics screening of primary auditory cortex (A1) from young InsG3680+/+ mice. A) Heatmap showing all A1 proteins (y-axis) from the “neuronal system” family according to the REACTOME database. Each column represents InsG3680+/+ protein levels normalized to the P10P14P21P35 GBRB2 SHAN1 DLG4 LRC4B DCE2 LRFN1 SHAN2 ADCY2 HCN2 NLGN3 CCG2 DLGP2 KAPCB NLGN2 GBG2 LRRC7 GBG3 NEB2 NRX1A SYN3 ADCY9 DNJC5 VAMP2 PRAF3 FLOT1 GBRA1 GNAI3 GBG7 GBRB1 KCC2B GNAI2 ACTN2 S6A11 KCC4 KKCC2 KPCG RIMS1 GABR2 NCALD GBB4 SNP25 TSN7 GRM5 SYN1 KCC2G GABR1 PICK1 AP2B1 DBNL AP2M1 KCC1A STX1A LIPA2 GBG4 MYEF2 SC6A1 DLG3 CSKP FLOT2 EAA1 KCJ10 GRIN1 PTPRD NCAM2 CYRIB GEPH KAP0 NCAM1 AP2A1 GNB5 KAP3 AP2A2 LIPA3 NRX3A E41L3 KCC2D SNP47 STXB1 RTN3 KAP2 SGIP1 PLCB1 SYT1 NRCAM NCDN ANK3 LIN7A GBRA3 NFASC HSP7C NCAN PTPRS GRIA3 SSDH DLGP4 ANK2 HOME1 DLGP3 NMDE2 KKCC1 NPTX1 CAC1E VGLU1 EAA2 SYN2 ADCY1 DLG2 GRIA2 COMT CA2D1 HIP1R KAP1 CAP2 GRIA1 SV2A GNAI1 GLSK NPTN GRM1 NSF GABT GLNA NMDZ1 CACB3 KCC2A CACB1 AOFA ALDH2 HCN1 KCAB2 GRIK2 GBRA2 GBB2 E41L2 RAB3A GBRB3 APBA1 CACB4 DLG1 KPCB CA2D3 GBB1 SLIK1 DCE1 LRRT4 ARHG9 ADCY3 NFL KAPCA SHAN3 1 2 SHAN3 KAPCA NFL LRRT4 CA2D3 0.0 0.5 1.0 1.5 2.0 Fold change NEB2 GBG3 GBG2 DLGP2 LRFN1 0.0 0.5 1.0 1.5 2.0 Fold change AB C D -2 -1 0 1 2 0 1 2 3 log2(fold change) -log10( p -value) SHAN3 LRRT4 KAPCA NFL GBG3 CA2D3 GBG2 LRFN1 DLGP2 p = 0.10 48 neurons. G, H) Current-voltage (I-V) curves representing Iss (G) and Ih (H) current densities in WT and InsG3680+/+ neurons. No significant differences were found between groups. I) The activation time constant of InsG3680+/+ neurons seems slightly faster in comparison to WT neurons, although not statistically significant. (F) Student’s paired t -test, WT: n=11, InsG3680+/+ : n= 9; (G) Mixed-effects ANOVA, WT: n = 22, InsG3680+/+ = 16; (H) Mixed-effects ANOVA, WT: n = 13, InsG3680+/+ = 10; (I) Student’s unpaired t -test, WT: n = 9, InsG3680+/+ = 6. Data are mean ± SEM; ** p <0.01, *** p <0.001, n.s.: not significant. WT: N = 6, InsG3680+/+ : N = 4 mice (males, 7-10 weeks old). Next, we asked if there were any differences regarding current density values between genotypes. Since there is an HCN1 upregulation in the A1 of InsG3680+/+ mice, we would expect to observe a greater current density in response to hyperpolarizing voltage steps in this group. However, there was no significant effect of genotype explaining the variance observed in the I-V curve for Iss current density (baseline conditions; Figure 17G) nor at Ih current density (Figure 17H). Finally, to assess Ih decay time constant (τact), we fitted a standard mono-exponential function to the initial segment of the -130 mV step (Figure 17I). No significant differences were found between genotypes, suggesting that Ih activation kinetics is similar between genotypes. There is a great variety of neuronal cell types in the neocortex with very different biophysical properties. These cells can be mainly classified into principal cells (cortical pyramidal neurons) and interneurons (Figure 18A). The principal cells are large glutamatergic neurons that send excitatory signals to different brain regions, whereas the interneurons are smaller GABAergic cells that inhibit the activity of local principal cells (Tremblay et al., 2016). Although interneurons can be divided into several classes depending on biophysical, morphological, and molecular characteristics, all interneurons tend to be very distinct from principal cells due to their smaller size and typically faster membrane time constant (τmemb) (Tremblay et al., 2016). It is already reported that Ih dynamics between principal neurons and PV+- interneurons shifts throughout mouse brain development (Yang et al., 2018), so we next asked whether this dynamics could be altered in InsG3680+/+ mice. 49 Figure 18. Clustering with Gaussian Mixture Model (GMM) of all the cells recorded in A1. A) Schematic representation of cortical layers (L1 to L6) in A1 region and layer-specific distribution of neuronal cell types. Principal neurons (mainly cortical pyramidal cells) are represented in blue and cortical interneurons are represented in pink. B) Gaussian Mixture Model (GMM) was used for unsupervised clustering of all recorded cells according to their membrane time constant (τmemb) and capacitance (Cp) values. The lines represent the contour of the probability density function from the GMM and are colorcoded (blue and yellow represent, respectively, low or high probability of belonging to each cluster). C) The same data as represented in B with the final cluster classification: blue dots represent putative interneurons and red dots represent putative principal neurons. Panel A created with Biorender.com. B C A τmemb (ms)τmemb (ms) Cp(pF) Cp(pF) 50 Given that principal neurons present higher τmemb and Cp and the opposite is true for interneurons, we used these two properties to plot all our recorded cells and then used a Gaussian Mixture Model (GMM) to perform clustering analysis. GMM is a clustering method based on the probability of each point belonging to a certain gaussian distribution. As expected, GMM analysis revealed two main clusters of cells: one with faster τmemb and smaller Cp (consistent with characteristics described for interneurons and, hence, further classified by us as putative interneurons), and another with higher τmemb and Cp (further classified as putative principal neurons (Figure 18B)). The contour of the probability density function for the GMM is represented in Figure 18C and gives the relative likelihood for each datapoint to belong to the cluster. Next, we repeated all electrophysiological analysis as before, but now considering the two putative cell types as classified by GMM clustering. In baseline conditions, there were no statistically significant differences in the Rm explained by the genotype for any of the two cell types (Figure 19A). Nonetheless, InsG3680+/+ putative interneurons seemed to have decreased Rm in comparison to WT and the opposite appeared to occur for the putative principal neurons. The effect of HCN blockade by ZD7288 in the Rm also seemed more pronounced in InsG3680+/+ putative interneurons, but again without any statistical differences (Figure 19B). Such differences, if significant, could indicate that HCN has a cell-type specific impact on Rm. Figure 19. Membrane resistance from WT and InsG3680+/+ A1 neurons according to celltype. A. B. Wild type InsG3680 +/+ Wild type InsG3680 +/+ 0 100 200 300 Rm (M) Interneurons Principal n. n.s. n.s. Wild type InsG3680 +/+ Wild type InsG3680 +/+ -150 -100 -50 0 50 100 Rm (M) Interneurons Principal n. n.s. n.s. 51 A) Average membrane resistance (Rm) is not different between wild type (WT) and InsG3680+/+ A1 neurons. B) Blocking HCN channels with ZD7288 incubation for 15 minutes seems to have greater impact on the Rm of InsG3680+/+ putative interneurons but it is not significantly different from WT interneurons. There is no significant effect of the cell type or the genotype on the Rm or ΔRm. (A) Mixed-effects ANOVA, Interneurons WT: n = 11, InsG3680+/+ : n = 8; Principal neurons WT and InsG3680+/+ : n=12 per group; (B) Mixed-effects ANOVA, Interneurons WT and InsG3680+/+ : n = 5 cells per group, Principal neurons WT: n = 4, InsG3680+/+ : n = 5; Data are mean ± SEM; n.s.: not significant. WT: N = 6, InsG3680+/+ : N = 4 mice (males, 7-10 weeks old). To assess cell-type specific impact of HCN blockade on Iss current amplitude, we compared baseline currents with currents recorded 15 minutes after ZD7288 incubation. Regarding putative interneurons, we observed a significant decrease in Iss current density for both genotypes, suggesting these cells were sensitive to ZD7288 application. Interestingly, the same did not occur in putative principal neurons. Whereas in the WT group there was a significant decrease in Iss amplitude, the same did not occur for the InsG3680+/+ group, suggesting a poor modulation of HCN channels in principal neurons after ZD7288 incubation (figure 20A). Notably, Ih currents in InsG3680+/+ putative principal neurons were significantly decreased as it can be observed in figure 20D (I-V curve). At baseline, neither putative principal neurons (Figure 20C) or interneurons (Figure 20D) showed significant differences associated with the genotype in Iss current density and the variance of Ih in putative interneurons was not explained by the genotype. Overall, this suggests that Ih differences between WT and InsG3680+/+ mice are cell-type specific, with decreased Ih in A1 principal neurons of InsG3680+/+ mice. 52 Figure 20. Cell-type specific HCN-mediated currents from WT and InsG3680+/+ neurons in A1 region. A) ZD7288 incubation for 15 minutes fails to significantly decrease baseline steady state current (Iss) density in InsG3680+/+ putative principal neurons. B) Activation time constant (τact) of InsG3680+/+ putative interneurons seems faster in comparison to WT, although no statistically significant differences were found for genotype or cell type. C, D) Current-voltage (I-V) curves representing Iss (C) and hyperpolarization current (Ih) (D) densities in WT and InsG3680+/+ putative principal neurons. The Ih of InsG3680+/+ putative principal neurons is significantly reduced (D) but no differences were found for the Iss in comparison to WT (C). E, F) Current-voltage (I-V) curves representing Iss (E) and Ih (F) current densities in WT and InsG3680+/+ putative interneurons. No differences were found between genotypes for Iss or Ih current densities. (A) Paired samples Student’s t -test; Principal neurons WT: n = 6, InsG3680+/+ : n = 5; Interneurons WT: n = 5, InsG3680+/+ : n = 4; (B) Unpaired Student’s t -test; Principal neurons WT: n = 5, InsG3680+/+ : n = 3, Interneurons WT: n = 4, InsG3680+/+ : n = 3; (C-F) Mixed-effects ANOVA with Wild type InsG3680 +/+ Wild type InsG3680 +/+ -15 -10 -5 0 Current density (pA/pF) Control ZD7288 Interneurons Principal n. ** ** n.s. Wild type InsG3680 +/+ Wild type InsG3680 +/+ 0 5 10 15 20 act (ms) Interneurons Principal n. n.s. n.s. -140-120-100-80-60 -20 -15 -10 -5 0 Step (mV) Current density (pA/pF) n.s. -140-120-100-80-60 -20 -15 -10 -5 0 5 Step (mV) Current density (pA/pF) n.s. A. B. C. D. E. F. -140-120-100-80-60 -4 -2 0 Step (mV) Current density (pA/pF) Wild type InsG3680+/+ * -140-120-100-80-60 -4 -2 0 Step (mV) Current density (pA/pF) Wild type InsG3680+/+ n.s. Iss Principal neurons Ih Principal neurons Ih Interneurons Iss Interneurons 53 Bonferroni’s post-hoc test for multiple comparisons; (C) Principal neurons WT: n = 11, InsG3680+/+ : n = 9; (D) Principal neurons WT: n = 7, InsG3680+/+ n = 4; (E) Interneurons WT: n = 11, InsG3680+/+ n = 8; (F) Interneurons WT: n = 6, InsG3680+/+ n = 5. Data are mean ± SEM, * p <0.05, ** p <0.01, n.s. not significant. WT: N = 6, InsG3680+/+ : N = 4 mice (males, 7-10 weeks old). τact was not found to be significantly different between genotypes for neither cell types, although InsG3680+/+ putative interneurons seemed to present a slightly faster τact in comparison to WT (Figure 20B). In putative principal neurons that putative difference was not so prominent. 54 5. DISCUSSION ASD is a complex and multifactorial disorder that manifests itself in several degrees of severity and variety of symptoms. Nonetheless, shared manifestations such as social-communication impairments and repetitive restricted behaviors may share a common pathological mechanism upstream (American Psychiatric Association, 2013). Sensory abnormalities are present in approximately 90% of individuals with autism (Balasco et al., 2020), being auditory hypersensitivity the most common sensory-perceptual abnormality. One hypothesis in the ASD-field is that ASD impairments in communication might be a consequence of the auditory hypersensitivity observed in patients, where atypical neuronal activity in brain auditory cortex region could underlie the noxious or unpleasant perception of auditory stimuli. For this reason, patients would avoid auditory stimulation, ultimately affecting their learning, communication, and quality of life. As such, in this dissertation, it was our aim to explore potential molecular and functional alterations in the primary auditory cortex region, using a rodent model of ASD. Given the central translational assumption between human and rodents, it was crucial for us to use the best possible animal model that could mimic the human disorder both in terms of its biological origin (construct validity) and behavioral phenotype (face validity). For this reason, we used a transgenic animal model of ASD that carries a patient derived mutation in the Shank3 gene, the Shank3-InsG3680+/+ mice. This mutation inserts a single guanine nucleotide (G) at cDNA position 3680, inducing a frameshift mutation that leads to a subsequent stop codon downstream. In order words, this knock-in mouse line leads to a knock-out (KO) of Shank3 . Shank3-InsG3680+/+ mice exhibit impaired social interaction, repetitive spontaneous self-grooming behavior, anxiety, and behavioral sensory alterations, being a good animal model of ASD (Zhou et al., 2016). In order to further study the auditory cortex in this animal model, we began by validating their genotype and phenotype in our hands. Unpublished results from our group, and not described on this dissertation, confirm impaired social interaction in the 3-chamber social test as well as auditory hypersensitivity in InsG3680+/+ mice. Furthermore, the splash test performed in this dissertation revealed increased self-grooming time in InsG3680+/+ mice upon spraying a 30% sucrose solution on their fur coat, reflecting perhaps an increased susceptibility to enroll in repetitive behavioral patterns upon triggering. Although the splash test has not been typically used in the autism field to evaluate repetitive behaviors, it is a well-established test to evaluate grooming behavior and to assess depressive-like states and motivation for self-care (Yalcin et al., 2005; Pitzer et al., 2022). Here, we use this approach with the purpose of understanding the impact of introducing an external trigger on mice behavior. The striking differences observed, not only in the time spent grooming but also in the number of grooming bouts, support a phenotype of repetitive behavior 55 commonly observed in animal models of ASD (as reviewed in the introduction of this dissertation). Moreover, the fact that we introduced here an external trigger for grooming shows that, besides having increased spontaneous grooming behavior (as previously described in the literature), these mice also have increased evoked grooming behavior. When analyzing behavioral videos collected during the splash test, we noticed that some exploratory and naturalistic behaviors seemed altered in InsG3680+/+ mice. Upon quantification, we found that rearing behavior, which could be considered as an exploratory behavior (Krüttner et al., 2022) and as an indirect measure of anxiety in rodents (Sturman et al., 2018), was reduced in InsG3680+/+ mice. This suggests the existence of an anxious phenotype in InsG3680+/+ mice, corroborating the findings previously published by Zhou et al (Zhou et al., 2016). Furthermore, we also detected reduced / absent digging behavior in InsG3680+/+ mice. Digging behavior is usually tested through the marble burying test where it is considered a measure of repetitive behavior (Angoa-Pérez et al., 2013). But given that the InsG3680+/+ mice have increased repetitive behavior (grooming), why not repetitive digging? One possibility is that the splash test has a bigger driving force behind: the sticky sucrose solution makes mice focus on selfgrooming for cleaning their fur coat, reducing the overall richness of their behavioral repertoire such as dig, rear, etc. Moreover, the marble burying test uses marbles to evoke digging behavior in rodents, a triggering factor that was absent during the splash test. After validating the InsG3680+/+ mouse model of ASD in our hands, we moved on towards our main aim which was to explore potential molecular and functional alterations in the primary auditory cortex region (A1). Despite not being reported in this dissertation, preliminary data from our lab indicates that InsG3680+/+ mice display an auditory hypersensitivity behavioral phenotype, which is already a strong indication that molecular and/or functional alterations might be present in their auditory cortex. This was actually first suggested in previous published work with the same animal model where the acoustic thresholds for a startle response were greatly diminished in InsG3680+/+ mice in comparison to WT controls (Zhou et al., 2016), suggesting an hyperreactive behavior in response to sounds. To start unveiling potential molecular underpinnings for auditory hypersensitivity in this model, we decided to study molecular signatures in A1, a brain region that is crucial to the processing of auditory stimuli. Using an unbiased proteomics approach, we found a large set of neuronal proteins that displayed altered expression levels in adult InsG3680+/+ male mice. Notably, there was a strong functional convergence between some of the most upregulated proteins: the HCN1 channel and ADCY1, 5 and 9 proteins. ADCYs are catalysts for the conversion of ATP to cAMP (Visel et al., 2006), while HCN1 is a 56 cationic channel whose activity can be modulated by cAMP (Wahl-Schott and Biel, 2009). Such finding indicates that SHANK3 absence, somehow, affects this pathway. Surprisingly, the SHANK3 protein has been shown to directly bind to HCN channels and to have a general function in scaffolding HCN channels at the cell surface (Yi et al., 2016). One possibility is that lack of SHANK3 leads to unstable tethering of HCN channels, compromising their function and hence triggering a compensation mechanism of HCN upregulation, both by its overexpression as well as upregulation of the cAMP pathway, in an attempt to restore HCN function. Given that ASD is a neurodevelopmental disorder, we also investigated A1 proteome during postnatal development but found fewer proteins significantly altered in young versus adult InsG3680+/+ mice. Of note, and contrarily to the adulthood dataset, HCN1 expression was not found to be increased in young InsG3680+/+ mice. Such observation could in fact support our hypothesis that HCN1 upregulation may be a compensation mechanism due to the loss of SHANK3. Given that SHANK3 expression is known to increase after P21 and peak only after P35 (Wang et al., 2014), it is perhaps not surprising that HCN1 upregulation only occurs after that timepoint because it is being triggered by the absence of SHANK3 at P35, a timepoint where SHANK3 presence was expected at the synapse. In fact, our proteomic results show a sudden and abrupt decrease of HCN1 levels at P35 (Figure 14A). Although one possible explanation is an experimental error that can only be ruled out by replicating such experiment, another possibility is that the sudden absence of SHANK3 at P35 (a timepoint where its expression should peak), leads to an abrupt reduction of HCN1 levels, hence triggering a long-term compensation mechanism of HCN1 upregulation. Repeating the quantification of HCN1 expression levels at several neurodevelopmental timepoints would be crucial to solve this question. Given that proteomics is a powerful but expensive technique to apply in this context, we pursued other molecular biology techniques to tackle the same question: how does the expression of HCN1 evolve across time? The first approach tested was western blot analysis which unfortunately failed, probably due to poor affinity of the primary antibody. Alternatively, we looked to mHcn1 transcript levels along development using real-time quantitative PCR (RT-qPCR). Although not answering exactly the same question (changes in RNA level do not necessarily translate into changes at the protein level), this experiment allowed us to assess whether the overexpression of HCN1 protein was also present at the RNA level. Somehow in agreement with the idea that subtle genotype differences in HCN1 expression may start around P21, our RT-qPCR results clearly indicate increased levels of mHcn1 transcript in the A1 at P21. In fact, the remaining timepoints tested also show a clear tendency (although not statistically significant) for increased mHcn1 , although with higher experimental variability. The same phenomenon 57 was observed for mHcn2 , the transcript for HCN2 channel that is also expressed in the neocortex. However, unlike HCN1 protein, HCN2 expression does not seem to follow the profile of increased transcript levels. Together this data suggests P21 as a crucial age for HCN dynamics in A1 and indicate that post-transcriptional regulation mechanisms likely play a key role in regulating the expression of different HCNs. HCN channels are responsible for mediating hyperpolarization currents (Ih) which were first observed in motorneurons and later in the heart by Noma and Irisawa (Noma and Irisawa, 1976; Bender and Baram, 2008). HCN1 channels are, together with HCN2, the main Ih currents facilitators in the neocortex (Shah, 2014). Given the increase in HCN1 expression in InsG3680+/+ adult mice, we expected to observe also a change in Ih currents in the neurons of A1. However, upon application of hyperpolarizing voltage steps in A1 neurons, no differences were observed between genotypes (neither in baseline Iss or in Ih). However, several explanations are possible for this observation. Our recordings were performed in neurons of deeper layers (around layer 4 to 6) of the A1, where we would expect greater HCN1 expression, especially in pyramidal neurons from layer 5 (Santoro et al., 2000). However, the expression pattern of HCN1 does not only vary depending on the cortical layer, but also at the cellular sub compartmental level. Observations both in rats and mice show that, in pyramidal neurons, HCN1 is mostly expressed at distal apical dendrites that expand from layer 5 until layer 1, where they co-localize with HCN2. Their abundance reduces gradually near the soma, where they are much less expressed (Lörincz et al., 2002; Notomi and Shigemoto, 2004). In other cell types, such as GABAergic inhibitory interneurons, HCN1, 2 and 4 seem to be further expressed in pre-synaptic terminals, especially in parvalbumin-expressing interneurons from layers 5-6. Here, HCN channels play an important regulatory role for GABA release from the GABAergic terminals through increasing Ca2+ influx by T-type Ca2+ channels (Cai et al., 2022). Since all our recordings were performed solely at the soma, we are probably not fully detecting Ih currents in more distal dendritic regions due to filtering effects and spatial clamping limitations, especially in pyramidal neurons. Hence, not only Ih currents at the soma may differ significantly depending on cell type but potential differences in HCN1 function may not be readily detectable among genotypes due to its distal expression at dendrites. Accordingly, and since the properties of the recorded cells varied considerably, we performed unsupervised clustering of our recorded cells based on their Cp and τmemb values into two main classes: putative principal neurons and putative interneurons. Interestingly, certain differences emerged, especially regarding putative principal neurons in InsG3680+/+ mice. Ih current density was decreased in this group, suggesting a decrease in functional HCN1 expression in these cells. Consistently, putative 64 Delling, J. P., and Boeckers, T. M. (2021). Comparison of SHANK3 deficiency in animal models: phenotypes, treatment strategies, and translational implications. J. Neurodev. Disord. 13, 55. doi:10.1186/s11689-021-09397-8. Dennis, M. K., Field, A. S., Burai, R., Ramesh, C., Whitney, K., Bologa, C. G., et al. (2012). Structures and Molecular Mechanisms for Common 15q13.3 Microduplications Involving CHRNA7: Benign or Pathological? Hum Mutat 127, 358–366. doi:10.1002/humu.21284.Structures. Dhamne, S. C., Silverman, J. L., Super, C. E., Lammers, S. H. T., Hameed, M. Q., Modi, M. E., et al. (2017). Replicable in vivo physiological and behavioral phenotypes of the Shank3B null mutant mouse model of autism. Mol. Autism 8, 26. doi:10.1186/s13229-017-0142-z. Di Bonito, M., and Studer, M. (2017). Cellular and molecular underpinnings of neuronal assembly in the central auditory system during mouse development. Front. Neural Circuits 11, 1–25. doi:10.3389/fncir.2017.00018. Domínguez-Iturza, N., Lo, A. C., Shah, D., Armendáriz, M., Vannelli, A., Mercaldo, V., et al. (2019). The autismand schizophrenia-associated protein CYFIP1 regulates bilateral brain connectivity and behaviour. Nat. Commun. 10, 1–13. doi:10.1038/s41467-019-11203-y. Drapeau, E., Riad, M., Kajiwara, Y., and Buxbaum, J. D. (2018). Behavioral Phenotyping of an Improved Mouse Model of Phelan–McDermid Syndrome with a Complete Deletion of the Shank3 Gene. eneuro 5, ENEURO.0046-18.2018. doi:10.1523/ENEURO.0046-18.2018. Dubiel, A., and Kulesza, R. J. (2016). Prenatal valproic acid exposure disrupts tonotopic c-Fos expression in the rat brainstem. Neuroscience 324, 511–523. doi:10.1016/j.neuroscience.2016.01.030. Durand, C. M., Betancur, C., Boeckers, T. M., Bockmann, J., Chaste, P., Fauchereau, F., et al. (2007). Mutations in the gene encoding the synaptic scaffolding protein SHANK3 are associated with autism spectrum disorders. Nat. Genet. 39, 25–27. doi:10.1038/ng1933. Elsabbagh, M., Divan, G., Koh, Y.-J., Kim, Y. S., Kauchali, S., Marcín, C., et al. (2012). Global Prevalence of Autism and Other Pervasive Developmental Disorders. Autism Res. 5, 160–179. doi:10.1002/aur.239. Engineer, C. T., Centanni, T. M., Im, K. W., Borland, M. S., Moreno, N. A., Carraway, R. S., et al. (2014a). Degraded auditory processing in a rat model of autism limits the speech representation in nonprimary auditory cortex. Dev. Neurobiol. 74, 972–986. doi:10.1002/dneu.22175. Engineer, C. T., Centanni, T. M., Im, K. W., Rahebi, K. C., Buell, E. P., and Kilgard, M. P. (2014b). Degraded speech sound processing in a rat model of fragile X syndrome. Brain Res. 1564, 72–84. doi:10.1016/j.brainres.2014.03.049. 65 Engineer, C. T., Rahebi, K. C., Borland, M. S., Buell, E. P., Im, K. W., Wilson, L. G., et al. (2018). Shank3deficient rats exhibit degraded cortical responses to sound. Autism Res. 11, 59–68. doi:10.1002/aur.1883. Esclassan, F., Francois, J., Phillips, K. G., Loomis, S., and Gilmour, G. (2015). Phenotypic characterization of nonsocial behavioral impairment in neurexin 1α knockout rats. Behav. Neurosci. 129, 74–85. doi:10.1037/bne0000024. Felix, R. A., Chavez, V. A., Novicio, D. M., Morley, B. J., and Portfors, C. V. (2019). Nicotinic acetylcholine receptor subunit α7-knockout mice exhibit degraded auditory temporal processing. J. Neurophysiol. 122, 451–465. doi:10.1152/jn.00170.2019. Gandal, M. J., Edgar, J. C., Ehrlichman, R. S., Mehta, M., Roberts, T. P. L., and Siegel, S. J. (2010). Validating γ oscillations and delayed auditory responses as translational biomarkers of autism. Biol. Psychiatry 68, 1100–1106. doi:10.1016/j.biopsych.2010.09.031. Gandhi, T., and Lee, C. C. (2021). Neural Mechanisms Underlying Repetitive Behaviors in Rodent Models of Autism Spectrum Disorders. Front. Cell. Neurosci. 14. doi:10.3389/fncel.2020.592710. Garcia-Pino, E., Gessele, N., and Koch, U. (2017). Enhanced excitatory connectivity and disturbed sound processing in the auditory brainstem of fragile X mice. J. Neurosci. 37, 7403–7419. doi:10.1523/JNEUROSCI.2310-16.2017. Gkogkas, C. G., Khoutorsky, A., Cao, R., Jafarnejad, S. M., Prager-Khoutorsky, M., Giannakas, N., et al. (2014). Pharmacogenetic Inhibition of eIF4E-Dependent Mmp9 mRNA Translation Reverses Fragile X Syndrome-like Phenotypes. Cell Rep. 9, 1742–1755. doi:10.1016/j.celrep.2014.10.064. Gogolla, N., Takesian, A. E., Feng, G., Fagiolini, M., and Hensch, T. K. (2014). Sensory Integration in Mouse Insular Cortex Reflects GABA Circuit Maturation. Neuron 83, 894–905. doi:10.1016/j.neuron.2014.06.033. Gomes, E., Pedroso, F. S., and Wagner, M. B. (2008). Hipersensibilidade auditiva no transtorno do espectro autístico. Pró-Fono Rev. Atualização Científica 20, 279–284. doi:10.1590/S010456872008000400013. Hacohen-Kleiman, G., Yizhar-Barnea, O., Touloumi, O., Lagoudaki, R., Avraham, K. B., Grigoriadis, N., et al. (2019). Atypical Auditory Brainstem Response and Protein Expression Aberrations Related to ASD and Hearing Loss in the Adnp Haploinsufficient Mouse Brain. Neurochem. Res. 44, 1494– 1507. doi:10.1007/s11064-019-02723-6. Herbert, M. R. (2010). Contributions of the environment and environmentally vulnerable physiology to autism spectrum disorders. Curr. Opin. Neurol. 23, 103–110. 66 doi:10.1097/WCO.0b013e328336a01f. Ida-Eto, M., Hara, N., Ohkawara, T., and Narita, M. (2017). Mechanism of auditory hypersensitivity in human autism using autism model rats. Pediatr. Int. 59, 404–407. doi:10.1111/ped.13186. Jaramillo, T. C., Speed, H. E., Xuan, Z., Reimers, J. M., Escamilla, C. O., Weaver, T. P., et al. (2017). Novel Shank3 mutant exhibits behaviors with face validity for autism and altered striatal and hippocampal function. Autism Res. 10, 42–65. doi:10.1002/aur.1664. Jaramillo, T. C., Speed, H. E., Xuan, Z., Reimers, J. M., Liu, S., and Powell, C. M. (2016). Altered Striatal Synaptic Function and Abnormal Behaviour in Shank3 Exon4-9 Deletion Mouse Model of Autism. Autism Res. 9, 350–375. doi:10.1002/aur.1529. Jonak, C. R., Sandhu, M. S., Assad, S. A., Barbosa, J. A., Makhija, M., and Binder, D. K. (2021). The PDE10A Inhibitor TAK-063 Reverses Sound-Evoked EEG Abnormalities in a Mouse Model of Fragile X Syndrome. Neurotherapeutics 18, 1175–1187. doi:10.1007/s13311-021-01005-w. Knipper, M., Van Dijk, P., Nunes, I., Rüttiger, L., and Zimmermann, U. (2013). Advances in the neurobiology of hearing disorders: Recent developments regarding the basis of tinnitus and hyperacusis. Prog. Neurobiol. 111, 17–33. doi:10.1016/j.pneurobio.2013.08.002. Kokash, J., Alderson, E. M., Reinhard, S. M., Crawford, C. A., Binder, D. K., Ethell, I. M., et al. (2019). Genetic reduction of MMP-9 in the Fmr1 KO mouse partially rescues prepulse inhibition of acoustic startle response. Brain Res. 1719, 24–29. doi:10.1016/j.brainres.2019.05.029. Krüttner, S., Falasconi, A., Valbuena, S., Galimberti, I., Bouwmeester, T., Arber, S., et al. (2022). Absence of familiarity triggers hallmarks of autism in mouse model through aberrant tail-of-striatum and prelimbic cortex signaling. Neuron 110, 1468-1482.e5. doi:10.1016/j.neuron.2022.02.001. Kulinich, A. O., Reinhard, S. M., Rais, M., Lovelace, J. W., Scott, V., Binder, D. K., et al. (2020). Beneficial effects of sound exposure on auditory cortex development in a mouse model of Fragile X Syndrome. Neurobiol. Dis. 134, 104622. doi:10.1016/j.nbd.2019.104622. Lee, J., Chung, C., Ha, S., Lee, D., Kim, D.-Y., Kim, H., et al. (2015). Shank3-mutant mice lacking exon 9 show altered excitation/inhibition balance, enhanced rearing, and spatial memory deficit. Front. Cell. Neurosci. 9. doi:10.3389/fncel.2015.00094. Lee, S.-Y., Vuong, T. A., So, H.-K., Kim, H.-J., Kim, Y. Bin, Kang, J.-S., et al. (2020). PRMT7 deficiency causes dysregulation of the HCN channels in the CA1 pyramidal cells and impairment of social behaviors. Exp. Mol. Med. 52, 604–614. doi:10.1038/s12276-020-0417-x. Leekam, S. R., Nieto, C., Libby, S. J., Wing, L., and Gould, J. (2007). Describing the sensory abnormalities of children and adults with autism. J. Autism Dev. Disord. 37, 894–910. doi:10.1007/s10803-006- 67 0218-7. Liu, C., Li, C., Hu, C., Wang, Y., Lin, J., Jiang, Y., et al. (2018). CRISPR/Cas9-induced shank3b mutant zebrafish display autism-like behaviors. Mol. Autism 9, 23. doi:10.1186/s13229-018-0204-x. Liu, H., Huang, X., Xu, J., Mao, H., Li, Y., Ren, K., et al. (2021). Dissection of the relationship between anxiety and stereotyped self-grooming using the Shank3B mutant autistic model, acute stress model and chronic pain model. Neurobiol. Stress 15, 100417. doi:10.1016/j.ynstr.2021.100417. Lörincz, A., Notomi, T., Tamás, G., Shigemoto, R., and Nusser, Z. (2002). Polarized and compartmentdependent distribution of HCN1 in pyramidal cell dendrites. Nat. Neurosci. 5, 1185–1193. doi:10.1038/nn962. Lovelace, J. W., Ethell, I. M., Binder, D. K., and Razak, K. A. (2018). Translation-relevant EEG phenotypes in a mouse model of Fragile X Syndrome. Neurobiol. Dis. 115, 39–48. doi:10.1016/j.nbd.2018.03.012. Lovelace, J. W., Ethell, I. M., Binder, D. K., and Razak, K. A. (2020a). Minocycline Treatment Reverses Sound Evoked EEG Abnormalities in a Mouse Model of Fragile X Syndrome. Front. Neurosci. 14, 1– 16. doi:10.3389/fnins.2020.00771. Lovelace, J. W., Rais, M., Palacios, A. R., Shuai, X. S., Bishay, S., Popa, O., et al. (2020b). Deletion of Fmr1 from Forebrain Excitatory Neurons Triggers Abnormal Cellular, EEG, and Behavioral Phenotypes in the Auditory Cortex of a Mouse Model of Fragile X Syndrome. Cereb. Cortex 30, 969– 988. doi:10.1093/cercor/bhz141. Lovelace, J. W., Wen, T. H., Reinhard, S., Hsu, M. S., Sidhu, H., Ethell, I. M., et al. (2016). Matrix metalloproteinase-9 deletion rescues auditory evoked potential habituation deficit in a mouse model of Fragile X Syndrome. Neurobiol. Dis. 89, 126–135. doi:10.1016/j.nbd.2016.02.002. Lukose, R., Schmidt, E., Wolski, T. P., Murawski, N. J., and Kulesza, R. J. (2011). Malformation of the superior olivary complex in an animal model of autism. Brain Res. 1398, 102–112. doi:10.1016/j.brainres.2011.05.013. Maciag, D., Simpson, K. L., Coppinger, D., Lu, Y., Wang, Y., Lin, R. C. S., et al. (2006). Neonatal antidepressant exposure has lasting effects on behavior and serotonin circuitry. Neuropsychopharmacology 31, 47–57. doi:10.1038/sj.npp.1300823. Malmierca, M. S. (2003). “The Structure And Physiology Of The Rat Auditory System: An Overview,” in International Review of Neurobiology , 147–211. doi:10.1016/S0074-7742(03)56005-6. Mansour, Y., Ahmed, S. N., and Kulesza, R. (2021). Abnormal morphology and subcortical projections to the medial geniculate in an animal model of autism. Exp. Brain Res. 239, 381–400. 68 doi:10.1007/s00221-020-05982-w. Mansour, Y., Mangold, S., Chosky, D., and Kulesza, R. J. (2019). Auditory Midbrain Hypoplasia and Dysmorphology after Prenatal Valproic Acid Exposure. Neuroscience 396, 79–93. doi:10.1016/j.neuroscience.2018.11.016. Marco, E. J., Hinkley, L. B. N., Hill, S. S., and Nagarajan, S. S. (2011). Sensory Processing in Autism: A Review of Neurophysiologic Findings. Pediatr. Res. 69, 48R-54R. doi:10.1203/PDR.0b013e3182130c54. Masini, E., Loi, E., Vega-Benedetti, A. F., Carta, M., Doneddu, G., Fadda, R., et al. (2020). An Overview of the Main Genetic, Epigenetic and Environmental Factors Involved in Autism Spectrum Disorder Focusing on Synaptic Activity. Int. J. Mol. Sci. 21, 8290. doi:10.3390/ijms21218290. Matsuzaki, H., Iwata, K., Manabe, T., and Mori, N. (2012). Triggers for Autism: Genetic and Environmental Factors. J. Cent. Nerv. Syst. Dis. 4, JCNSD.S9058. doi:10.4137/JCNSD.S9058. Mei, Y., Monteiro, P., Zhou, Y., Kim, J.-A., Gao, X., Fu, Z., et al. (2016). Adult restoration of Shank3 expression rescues selective autistic-like phenotypes. Nature 530, 481–484. doi:10.1038/nature16971. Monteggia, L. M., Eisch, A. J., Tang, M. D., Kaczmarek, L. K., and Nestler, E. J. (2000). Cloning and localization of the hyperpolarization-activated cyclic nucleotide-gated channel family in rat brain. Mol. Brain Res. 81, 129–139. doi:10.1016/S0169-328X(00)00155-8. Monteiro, P., and Feng, G. (2017). SHANK proteins: roles at the synapse and in autism spectrum disorder. Nat. Rev. Neurosci. 18, 147–157. doi:10.1038/nrn.2016.183. Moosmang, S., Biel, M., Hofmann, F., and Ludwig, A. (1999). Differential Distribution of Four Hyperpolarization-Activated Cation Channels in Mouse Brain. Biol. Chem. 380, 975–980. doi:10.1515/BC.1999.121. Muhle, R., Trentacoste, S. V., and Rapin, I. (2004). The Genetics of Autism. Pediatrics 113, e472–e486. doi:10.1542/peds.113.5.e472. Murphey, D. K., Herman, A. M., and Arenkiel, B. R. (2014). Dissecting inhibitory brain circuits with genetically-targeted technologies. Front. Neural Circuits 8. doi:10.3389/fncir.2014.00124. Nagode, D. A., Meng, X., Winkowski, D. E., Smith, E., Khan-Tareen, H., Kareddy, V., et al. (2017). Abnormal Development of the Earliest Cortical Circuits in a Mouse Model of Autism Spectrum Disorder. Cell Rep. 18, 1100–1108. doi:10.1016/j.celrep.2017.01.006. Nakamura, M., Valerio, P., Bhumika, S., and Barkat, T. R. (2020). Sequential Organization of Critical Periods in the Mouse Auditory System. Cell Rep. 32, 108070. doi:10.1016/j.celrep.2020.108070. 69 Nguyen, A. O., Binder, D. K., Ethell, I. M., and Razak, K. A. (2020). Abnormal development of auditory responses in the inferior colliculus of a mouse model of Fragile X Syndrome. J. Neurophysiol. 123, 2101–2121. doi:10.1152/jn.00706.2019. Noma, A., and Irisawa, H. (1976). Membrane currents in the rabbit sinoatrial node cell as studied by the double microelectrode method. Pflugers Arch. Eur. J. Physiol. 364, 45–52. doi:10.1007/BF01062910. Notomi, T., and Shigemoto, R. (2004). Immunohistochemical localization of Ih channel subunits, HCN14, in the rat brain. J. Comp. Neurol. 471, 241–276. doi:10.1002/cne.11039. Peça, J., Feliciano, C., Ting, J. T., Wang, W., Wells, M. F., Venkatraman, T. N., et al. (2011). Shank3 mutant mice display autistic-like behaviours and striatal dysfunction. Nature 472, 437–442. doi:10.1038/nature09965. Peñagarikano, O., Abrahams, B. S., Herman, E. I., Winden, K. D., Gdalyahu, A., Dong, H., et al. (2011). Absence of CNTNAP2 leads to epilepsy, neuronal migration abnormalities, and core autism-related deficits. Cell 147, 235–246. doi:10.1016/j.cell.2011.08.040. Pirbhoy, P. S., Jonak, C. R., Syed, R., Argueta, D. A., Perez, P. A., Wiley, M. B., et al. (2021). Increased 2-arachidonoyl-sn-glycerol levels normalize cortical responses to sound and improve behaviors in Fmr1 KO mice. J. Neurodev. Disord. 13, 47. doi:10.1186/s11689-021-09394-x. Pirbhoy, P. S., Rais, M., Lovelace, J. W., Woodard, W., Razak, K. A., Binder, D. K., et al. (2020). Acute pharmacological inhibition of matrix metalloproteinase-9 activity during development restores perineuronal net formation and normalizes auditory processing in Fmr1 KO mice. J. Neurochem. 155, 538–558. doi:10.1111/jnc.15037. Pitzer, C., Kurpiers, B., and Eltokhi, A. (2022). Sex Differences in Depression-Like Behaviors in Adult Mice Depend on Endophenotype and Strain. Front. Behav. Neurosci. 16. doi:10.3389/fnbeh.2022.838122. Reinhard, S. M., Rais, M., Afroz, S., Hanania, Y., Pendi, K., Espinoza, K., et al. (2019). Reduced perineuronal net expression in Fmr1 KO mice auditory cortex and amygdala is linked to impaired fear-associated memory. Neurobiol. Learn. Mem. 164, 107042. doi:10.1016/j.nlm.2019.107042. Rendall, A. R., Perrino, P. A., Buscarello, A. N., and Fitch, R. H. (2019). Shank3B mutant mice display pitch discrimination enhancements and learning deficits. Int. J. Dev. Neurosci. 72, 13–21. doi:10.1016/j.ijdevneu.2018.10.003. Robertson, C. E., and Baron-Cohen, S. (2017). Sensory perception in autism. Nat. Rev. Neurosci. 18, 671–684. doi:10.1038/nrn.2017.112. 70 Robinson, R. B., and Siegelbaum, S. A. (2003). Hyperpolarization-Activated Cation Currents: From Molecules to Physiological Function. Annu. Rev. Physiol. 65, 453–480. doi:10.1146/annurev.physiol.65.092101.142734. Rodenas-Cuadrado, P., Ho, J., and Vernes, S. C. (2014). Shining a light on CNTNAP2: Complex functions to complex disorders. Eur. J. Hum. Genet. 22, 171–178. doi:10.1038/ejhg.2013.100. Rotschafer, S. E., and Cramer, K. S. (2017). Developmental emergence of phenotypes in the auditory brainstem nuclei of Fmr1 knockout mice. eNeuro 4. doi:10.1523/ENEURO.0264-17.2017. Rotschafer, S. E., Marshak, S., and Cramer, K. S. (2015). Deletion of Fmr1 alters function and synaptic inputs in the auditory brainstem. PLoS One 10, 1–15. doi:10.1371/journal.pone.0117266. Rotschafer, S., and Razak, K. (2013). Altered auditory processing in a mouse model of fragile X syndrome. Brain Res. 1506, 12–24. doi:10.1016/j.brainres.2013.02.038. Ruby, K., Falvey, K., and Kulesza, R. J. (2015). Abnormal neuronal morphology and neurochemistry in the auditory brainstem of Fmr1 knockout rats. Neuroscience 303, 285–298. doi:10.1016/j.neuroscience.2015.06.061. Rylaarsdam, L., and Guemez-Gamboa, A. (2019). Genetic Causes and Modifiers of Autism Spectrum Disorder. Front. Cell. Neurosci. 13. doi:10.3389/fncel.2019.00385. Sanes, D. H., and Kotak, V. C. (2011). Developmental plasticity of auditory cortical inhibitory synapses. Hear. Res. 279, 140–148. doi:10.1016/j.heares.2011.03.015. Santoro, B., Chen, S., Lüthi, A., Pavlidis, P., Shumyatsky, G. P., Tibbs, G. R., et al. (2000). Molecular and Functional Heterogeneity of Hyperpolarization-Activated Pacemaker Channels in the Mouse CNS. J. Neurosci. 20, 5264–5275. doi:10.1523/JNEUROSCI.20-14-05264.2000. Santoro, B., and Tibbs, G. (1999). The HCN Gene Family: Molecular Basis of the HyperpolarizationActivated Pacemaker Channels. Ann. N. Y. Acad. Sci. 868, 741–764. doi:10.1111/j.17496632.1999.tb11353.x. Sarn, N., Thacker, S., Lee, H., and Eng, C. (2021). Germline nuclear-predominant Pten murine model exhibits impaired social and perseverative behavior, microglial activation, and increased oxytocinergic activity. Mol. Autism 12, 1–19. doi:10.1186/s13229-021-00448-4. Satterstrom, F. K., Kosmicki, J. A., Wang, J., Breen, M. S., De Rubeis, S., An, J.-Y., et al. (2020). LargeScale Exome Sequencing Study Implicates Both Developmental and Functional Changes in the Neurobiology of Autism. Cell 180, 568-584.e23. doi:10.1016/j.cell.2019.12.036. Schaaf, C. P. (2014). Nicotinic acetylcholine receptors in human genetic disease. Genet. Med. 16, 649– 656. doi:10.1038/gim.2014.9. 71 Schmeisser, M. J., Ey, E., Wegener, S., Bockmann, J., Stempel, A. V., Kuebler, A., et al. (2012). Autisticlike behaviours and hyperactivity in mice lacking ProSAP1/Shank2. Nature 486, 256–260. doi:10.1038/nature11015. Schridde, U., Strauss, U., Bräuer, A. U., and van Luijtelaar, G. (2006). Environmental manipulations early in development alter seizure activity, I h and HCN1 protein expression later in life. Eur. J. Neurosci. 23, 3346–3358. doi:10.1111/j.1460-9568.2006.04865.x. Scott, K. E., Schormans, A. L., Pacoli, K. Y., De Oliveira, C., Allman, B. L., and Schmid, S. (2018). Altered auditory processing, filtering, and reactivity in the cntnap2 knock-out rat model for neurodevelopmental disorders. J. Neurosci. 38, 8588–8604. doi:10.1523/JNEUROSCI.075918.2018. Sethna, F., Moon, C., and Wang, H. (2014). From FMRP Function to Potential Therapies for Fragile X Syndrome. Neurochem. Res. 39, 1016–1031. doi:10.1007/s11064-013-1229-3. Shah, M. M. (2014). Cortical HCN channels: function, trafficking and plasticity. J. Physiol. 592, 2711– 2719. doi:10.1113/jphysiol.2013.270058. Shah, M. M. (2016). Recording Hyperpolarization-Activated Cyclic Nucleotide-Gated Channel Currents ( I h ) in Neurons. Cold Spring Harb. Protoc. 2016, pdb.prot091462. doi:10.1101/pdb.prot091462. Sohal, V. S., and Rubenstein, J. L. R. (2019). Excitation-inhibition balance as a framework for investigating mechanisms in neuropsychiatric disorders. Mol. Psychiatry 24, 1248–1257. doi:10.1038/s41380019-0426-0. Song, T.-J., Lan, X.-Y., Wei, M.-P., Zhai, F.-J., Boeckers, T. M., Wang, J.-N., et al. (2019). Altered Behaviors and Impaired Synaptic Function in a Novel Rat Model With a Complete Shank3 Deletion. Front. Cell. Neurosci. 13. doi:10.3389/fncel.2019.00111. Stanfield, C. L. (2013). Principles of Human Physiology . 5th editio. Pearson. Strumbos, J. G., Brown, M. R., Kronengold, J., Polley, D. B., and Kaczmarek, L. K. (2010). Fragile X mental retardation protein is required for rapid experience-dependent regulation of the potassium channel Kv3.1b. J. Neurosci. 30, 10263–10271. doi:10.1523/JNEUROSCI.1125-10.2010. Sturman, O., Germain, P.-L., and Bohacek, J. (2018). Exploratory rearing: a contextand stress-sensitive behavior recorded in the open-field test. Stress 21, 443–452. doi:10.1080/10253890.2018.1438405. Sztainberg, Y., and Zoghbi, H. Y. (2016). Lessons learned from studying syndromic autism spectrum disorders. Nat. Neurosci. 19, 1408–1417. doi:10.1038/nn.4420. The Dutch-Belgian Fragile X Consorthium, Bakker, C. E., Verheij, C., Willemsen, R., van der Helm, R., 72 Oerlemans, F., et al. (1994). Fmr1 knockout mice: A model to study fragile X mental retardation. Cell 78, 23–33. doi:10.1016/0092-8674(94)90569-X. Tomchek, S. D., and Dunn, W. (2007). Sensory Processing in Children With and Without Autism: A Comparative Study Using the Short Sensory Profile. Am. J. Occup. Ther. 61, 190–200. doi:10.5014/ajot.61.2.190. Tremblay, R., Lee, S., and Rudy, B. (2016). GABAergic Interneurons in the Neocortex: From Cellular Properties to Circuits. Neuron 91, 260–292. doi:10.1016/j.neuron.2016.06.033. Truong, D. T., Rendall, A. R., Castelluccio, B. C., Eigsti, I. M., and Holly, R. (2015). Auditory processing and morphological anomalies in medial geniculate nucleus of Cntnap2 mutant mice. Behav. Neurosci. 129, 731–741. doi:10.1037/bne0000096. Tsugiyama, L. E., Ida‐Eto, M., Ohkawara, T., Noro, Y., and Narita, M. (2020). Altered neuronal activity in the auditory brainstem following sound stimulation in thalidomide‐induced autism model rats. Congenit. Anom. (Kyoto). 60, 82–86. doi:10.1111/cga.12353. Vicidomini, C., Ponzoni, L., Lim, D., Schmeisser, M. J., Reim, D., Morello, N., et al. (2017). Pharmacological enhancement of mGlu5 receptors rescues behavioral deficits in SHANK3 knockout mice. Mol. Psychiatry 22, 689–702. doi:10.1038/mp.2016.30. Visel, A., Alvarez-Bolado, G., Thaller, C., and Eichele, G. (2006). Comprehensive analysis of the expression patterns of the adenylate cyclase gene family in the developing and adult mouse brain. J. Comp. Neurol. 496, 684–697. doi:10.1002/cne.20953. Wahl-Schott, C., and Biel, M. (2009). HCN channels: Structure, cellular regulation and physiological function. Cell. Mol. Life Sci. 66, 470–494. doi:10.1007/s00018-008-8525-0. Wang, X., Bey, A. L., Katz, B. M., Badea, A., Kim, N., David, L. K., et al. (2016). Altered mGluR5-Homer scaffolds and corticostriatal connectivity in a Shank3 complete knockout model of autism. Nat. Commun. 7, 11459. doi:10.1038/ncomms11459. Wang, X., McCoy, P. A., Rodriguiz, R. M., Pan, Y., Je, H. S., Roberts, A. C., et al. (2011). Synaptic dysfunction and abnormal behaviors in mice lacking major isoforms of Shank3. Hum. Mol. Genet. 20, 3093–3108. doi:10.1093/hmg/ddr212. Wang, X., Xu, Q., Bey, A. L., Lee, Y., and Jiang, Y. (2014). Transcriptional and functional complexity of Shank3 provides a molecular framework to understand the phenotypic heterogeneity of SHANK3 causing autism and Shank3 mutant mice. Mol. Autism 5, 30. doi:10.1186/2040-2392-5-30. Wen, T. H., Afroz, S., Reinhard, S. M., Palacios, A. R., Tapia, K., Binder, D. K., et al. (2018). Genetic Reduction of Matrix Metalloproteinase-9 Promotes Formation of Perineuronal Nets Around 73 Parvalbumin-Expressing Interneurons and Normalizes Auditory Cortex Responses in Developing Fmr1 Knock-Out Mice. Cereb. Cortex 28, 3951–3964. doi:10.1093/cercor/bhx258. Werling, D. M., and Geschwind, D. H. (2013). Sex differences in autism spectrum disorders. Curr. Opin. Neurol. 26, 146–153. doi:10.1097/WCO.0b013e32835ee548. Wilson, H. L., Wong, A. C. C., Shaw, S. R., Tse, W.-Y., Stapleton, G. A., Phelan, M. C., et al. (2003). Molecular characterisation of the 22q13 deletion syndrome supports the role of haploinsufficiency of SHANK3/PROSAP2 in the major neurological symptoms. J. Med. Genet. 40, 575–84. doi:10.1136/jmg.40.8.575. Xiong, Q., Oviedo, H. V., Trotman, L. C., and Zador, A. M. (2012). PTEN regulation of local and long-range connections in mouse auditory cortex. J. Neurosci. 32, 1643–1652. doi:10.1523/JNEUROSCI.4480-11.2012. Yalcin, I., Aksu, F., and Belzung, C. (2005). Effects of desipramine and tramadol in a chronic mild stress model in mice are altered by yohimbine but not by pindolol. Eur. J. Pharmacol. 514, 165–174. doi:10.1016/j.ejphar.2005.03.029. Yang, M., Bozdagi, O., Scattoni, M. L., Wohr, M., Roullet, F. I., Katz, A. M., et al. (2012). Reduced Excitatory Neurotransmission and Mild Autism-Relevant Phenotypes in Adolescent Shank3 Null Mutant Mice. J. Neurosci. 32, 6525–6541. doi:10.1523/JNEUROSCI.6107-11.2012. Yang, S.-S., Li, Y.-C., Coley, A. A., Chamberlin, L. A., Yu, P., and Gao, W.-J. (2018). Cell-Type Specific Development of the Hyperpolarization-Activated Current, Ih, in Prefrontal Cortical Neurons. Front. Synaptic Neurosci. 10. doi:10.3389/fnsyn.2018.00007. Yi, F., Danko, T., Botelho, S. C., Patzke, C., Pak, C., Wernig, M., et al. (2016). Autism-associated SHANK3 haploinsufficiency causes Ih channelopathy in human neurons. Science (80-. ). 352, aaf2669– aaf2669. doi:10.1126/science.aaf2669. Yoo, Y.-E., Yoo, T., Lee, S., Lee, J., Kim, D., Han, H.-M., et al. (2019). Shank3 Mice Carrying the Human Q321R Mutation Display Enhanced Self-Grooming, Abnormal Electroencephalogram Patterns, and Suppressed Neuronal Excitability and Seizure Susceptibility. Front. Mol. Neurosci. 12. doi:10.3389/fnmol.2019.00155. Zhang, X.-C., Shu, L.-Q., Zhao, X.-S., and Li, X.-K. (2019). Autism spectrum disorders: autistic phenotypes and complicated mechanisms. World J. Pediatr. 15, 17–25. doi:10.1007/s12519-018-0210-2. Zhou, C., Yan, S., Qian, S., Wang, Z., Shi, Z., Xiong, Y., et al. (2019a). Atypical response properties of the auditory cortex of awake MECP2-overexpressing mice. Front. Neurosci. 13, 1–10. doi:10.3389/fnins.2019.00439.