All DEL mouse models are affected by early post-natal mortality and their body weight is significantly lower than wild-type mice (Horev et al., 2011; Portmann et al., 2014; Arbogast et al., 2016). In Dolmetsch model, this effect could be corrected by improved nutrition and separation from wild-type littermates (Portmann et al., 2014). As adults, the body weight of DEL mice is similar to wild-type littermates (Horev et al., 2011) or reduced (Portmann et al., 2014; Pucilowska et al., 2015; Arbogast et al., 2016), showing decreased adiposity (Arbogast et al., 2016). This is in contrast with human findings where the 16p11.2 deletion has been associated with a highly penetrant type of obesity (Jacquemont et al., 2011). Herault model also displays a higher energy expenditure during the dark phase, a faster glucose clearance and lower blood levels of leptin and adiponectin (Arbogast et al., 2016).

Craniofacial dysmorphisms have been reported by (Arbogast et al., 2016), with decreased skull size and altered skull shape in the Herault DEL females. In contrast with humans (Qureshi et al., 2014), the brain size is modestly reduced in Mills DEL mice during early post-natal development (Pucilowska et al., 2015). This phenotype was observed also in Mills DEL adults (Horev et al., 2011; Pucilowska et al., 2015), with no significant changes in the grey matter (Kumar et al., 2018). In addition, cortical thickness is decreased in Mills DEL mice (Pucilowska et al., 2015), similarly to human carriers (Maillard et al., 2015). In particular, this model shows an aberrant cortical cytoarchitecture, with a reduction of upper cortical layer neurons at embryonic day 14.5, probably due to aberrant progenitor proliferation and premature cell cycle exit, leading to depletion of progenitor pools (Pucilowska et al., 2015; Pucilowska et al., 2018). Despite the general reduction in brain size, Mills and Dolmetsch DEL mice show increases in the relative volumes of several brain areas, including the midbrain, hypothalamus, striatum, nucleus accumbens, globus pallidus and cerebellar cortex (Horev et al., 2011; Portmann et al., 2014; Pucilowska et al., 2018). Mills DEL mice also exhibit decreased volumes in the ventral hippocampus, lateral septum, amygdala and entorhinal cortex (Pucilowska et al., 2018).

Interestingly, data obtained in both Dolmetsch and Mills models suggest major alterations in dopaminergic signalling. For instance, Portmann et al. observed a significant increase in medium spiny neurons expressing dopamine receptor 2 (Drd2+ MSNs) in the striatum of 16p11.2 DEL neonates, with no changes in medium spiny neurons expressing dopamine receptor 1 (Drd1+ MSNs) (Portmann et al., 2014). The observed increase in striatopallidal MSNs in the striatum resulted in a reduced sensitivity to sedation induced by risperidone, a D2 receptor antagonist (Portmann et al., 2014). Interestingly, an increased number of cells co-expressing Drd1 and Drd2 was observed, suggesting that the 16p11.2 deletion may also affect the process of MSNs specification. In the deep layers of the cortex, Drd1+ MSNs were significantly decreased as well as DARPP-32 expressing neurons. Moreover, tyrosine-hydroxylase (TH), a rate limiting enzyme in the dopamine (DA) synthesis pathway, was decreased in mesodiencephalic DA cells (Portmann et al., 2014). Imbalances in the ratio of D1 and D2-expressing MSNs were also observed in adult brains by Grissom et al. In particular, Mills DEL males, but not females, overexpressed the mRNA for D2 receptor and adenosine 2a receptor in the striatum (Grissom et al., 2018). Interestingly, transcriptomic analysis in Herault DEL model revealed dysregulations in gene expression in DEL mice in different brain regions, with the striatum being more severely impacted (Arbogast et al., 2016). This finding may explain some of the motor and cognitive deficits displayed by these animals, dependent on basal ganglia circuitry.

The presence of major alterations in dopamine-mediated circuits is further supported by electrophysiological recordings on striatal MSNs from Dolmetsch DEL mice. These studies revealed an increased ratio of AMPA to NMDA receptor-mediated excitatory post synaptic currents (EPSC) and an increased miniature EPSC (mEPSC) frequency. Conversely, the paired-pulse ratios (PPRs) across multiple interstimulus intervals (ISIs) were significantly decreased (Portmann et al., 2014). These results suggest that the release probability of excitatory synapses on MSNs may be augmented.

In the hippocampus, no alterations were found in the excitability profile in Herault DEL model (Arbogast et al., 2016). However, in the medial prefrontal cortex (mPFC) of Mills DEL mice, pyramidal neurons displayed deficient NMDA-receptor-mediated glutamatergic transmission and reduced frequency of action potential (AP) firing (Wang et al., 2018). Compromised functional connectivity on the orbitofrontal, insular and auditory axis, and between the septum and the hippocampal regions, has been also reported in Mills DEL mice (Openshaw et al., 2023).

At the behavioural level, mild motor impairments have been observed in 16p11.2 DEL mice, such as deficits in righting from upside-down position (Mills DEL model) (Brunner et al., 2015) and lack of gait fluidity and tremor (Dolmetsch DEL model) (Portmann et al., 2014). However, normal motor coordination in the rotarod test was reported in Herault DEL model (Arbogast et al., 2016). Higher locomotor activity in familiar environments and stereotyped behaviours have also been broadly reported in all models (Horev et al., 2011; Portmann et al., 2014; Yang et al., 2015c; Arbogast et al., 2016; Angelakos et al., 2017). However, when tested in novel environments, such as in the open field test, both Dolmetsch and Mills DEL mice showed initial hypoactivity that gradually disappeared over the course of the first 10 min (Portmann et al., 2014; Pucilowska et al., 2015) which might reflect deficits in motor initiation (Portmann et al., 2014) or increased anxiety, as shown by Pucilowska et al., in the elevated plus maze test (Pucilowska et al., 2015).

DEL mice also display a wide range of cognitive deficits, including impairments in spatial memory (Mills DEL model) (Wang et al., 2018), novel object recognition (Mills, Dolmetsch and Herault DEL models) (Portmann et al., 2014; Yang et al., 2015b; Pucilowska et al., 2015; Arbogast et al., 2016; Pucilowska et al., 2018), novel object location (Dolmetsch DEL model) (Yang et al., 2015b) and passive avoidance (Mills DEL model) (Tian et al., 2015). Dolmetsch DEL model also exhibited prominent cognitive impairments in the touchscreen pairwise visual discrimination acquisition and reversal (Yang et al., 2015b), while deficits in contextual fear conditioning have been observed in Mills DEL model (Tian et al., 2015; Pucilowska et al., 2018), but not in Dolmetsch DEL model (Yang et al., 2015b). Consistently with the impaired fronto-temporal connectivity and GABAergic dysfunction observed by Openshaw et al., Mills DEL mice showed deficits in pre-pulse inhibition (PPI), a measure of sensorimotor gating, but enhanced performance in attentional tasks (Openshaw et al., 2023).

In terms of social behaviour, ultrasonic vocalizations during male-female interactions are significantly reduced in DEL mice (Mills and Dolmetsch DEL models) (Yang et al., 2015c; Stoppel et al., 2018), that could be rescued upon chronic activation of GABA_B receptors (Stoppel et al., 2018). Unexpectedly, all DEL models display normal behaviour in the three-chamber social preference test, in contrast with the social deficits affecting human carriers (Yang et al., 2015a; Brunner et al., 2015; Arbogast et al., 2016). This could be partially due to the aberrant increase of oxytocin levels exhibited by the Mills DEL mice (Pucilowska et al., 2018) that may mask potential social deficits. In addition, the genetic background can profoundly influence the manifestation of social impairments. As demonstrated by Arbogast et al., a significant decrease in social preference for the second stranger in the three-chamber test can be observed in Herault DEL model with a hybrid C57/Bl6N X C3B background, but not in mice with a pure C57/Bl6N background (Arbogast et al., 2016). Housing conditions also seem to affect the emergence of social and cognitive deficits. For instance, impairments in recognition memory and ultrasonic vocalisation are displayed by Dolmetsch DEL mice only in standard mixed-genotypes housing conditions, but not by animals housed with individuals of the same genotype (Yang et al., 2015a).

Despite the higher prevalence of autism spectrum disorder in males, sex-specific phenotypes have not been systematically investigated in DEL mice. Kumar et al. found prominent male-specific structural changes in medial fibre tracts proximate to the striatum, overlapping with specific gene expression patterns associated with neurite outgrowth and MAPK pathway (Mills DEL model) (Kumar et al., 2018). At the behavioural level, male-specific deficits in perinatal communication have been observed (Mills DEL model) (Agarwalla et al., 2020). Consistently with ASD and ADHD patients’ phenotypes, male-specific sleep/wake decrements in total sleep time and longer bouts of continuous wakefulness have been reported (Mills DEL model) (Angelakos et al., 2017). Moreover, Mills DEL males display reduced motivation and impaired reward learning, which is consistent with the observed increase in the mRNA coding for dopamine D2 receptors associated with behavioural inhibition (Grissom et al., 2018).

In comparison with DEL mice, DUP mice show opposite phenotypes in terms of body weight and metabolism, with increased body size, lower energy expenditure during the light and dark phase, a lower glucose clearance and higher blood levels of leptin (Herault model) (Arbogast et al., 2016). Craniofacial dysmorphisms have also been observed in DUP mice, showing altered skull shape but no changes in the skull size (Herault model) (Arbogast et al., 2016). At the neurostructural level, DUP mice are not significantly different from wild types, although a trend toward reduced volumes in several brain regions can be observed (Mills model) (Horev et al., 2011).

In contrast with DEL mice, both Herault and Mills DUP mice display hypolocomotion (Horev et al., 2011; Arbogast et al., 2016; Bristow et al., 2020; Rein et al., 2020) with potential sex-differences. Hypo locomotion was observed in Mills DUP females only during home cage monitoring whereas males exhibited this behaviour in the novel environment of the open field arena. This observation may indicate that additional factors, such as the levels of stress, may have an impact on sex-specific phenotypes. These results are also consistent with the increased anxiety behaviour observed only in males, possibly linked to the reduced hippocampal-orbitofrontal-amygdala connectivity (Bristow et al., 2020). Importantly, this circuitry has been implicated in thought disorder, a hallmark of schizophrenia (Sumner et al., 2018). In Mills DUP mice, typical phenotypes linked to schizophrenia, such as MK-801-induced hyperlocomotion and deficits in pre-pulse inhibition, were not observed by Rein et al. (Rein et al., 2020), although there is a report of female-specific reduction of pre-pulse inhibition (Bristow et al., 2020). Mills DUP model also displays social impairments reminiscent of ASD, including deficits in social approach and in the three-chamber test (Rein et al., 2020) as well as reduced time spent in proximity with cage mates (Bristow et al., 2020). In contrast, in Herault DUP model, social deficits in the three-chamber test could be observed only in the hybrid C57/Bl6N X C3B background, similarly to the DEL mice (Arbogast et al., 2016).

Consistently with the observed deficits in hippocampal-orbitofrontal-amygdala connectivity, Mills DUP mice display impairments in spatial working memory (Bristow et al., 2020), that have been also reported in DUP human carriers (Schobel et al., 2009). Similarly to the deficits observed in patients in the continuous performance task (Fleck et al., 2001), Mills DUP mice show slower learning and more impulsive responding (Bristow et al., 2020), as well as prefrontal cortex-dependent cognitive impairments in the temporal order recognition memory (Rein et al., 2020). Novel object recognition memory was either found unaffected by the CNV (Rein et al., 2020) or significantly enhanced (Arbogast et al., 2016), possibly due to the different models and experimental protocols used in these studies.

In contrast with DEL mice showing hypoactivity in the mPFC neurons, Mills DUP mice display hyperexcitability due to a significant impairment in GABAergic synaptic transmission, while glutamatergic transmission was found unchanged (Rein et al., 2020). This is consistent to the excitatory/inhibitory imbalance observed in ASD patients (Nelson and Valakh, 2015).

In order to determine the effect of the 16p11.2 duplication on gene expression, Rein et al. performed RNA-sequencing on mPFC and identified 388 differentially expressed genes, most of which were downregulated, including epigenetic markers, ASD/ID risk genes and the sodium ion channel SCN9a. A significant downregulation was detected for Npas4, a transcription factor promoting the formation of GABAergic synapses, that was found reduced also in post-mortem PFCs from idiopathic ASD patients (Rein et al., 2020). Restoration of Npas4 levels in Mills 16p11.2 DUP mice was sufficient to rescue the synaptic and behavioural deficits, thus suggesting the pathogenic role of Npas4 in the GABAergic dysfunction underlying the 16p11.2 DUP phenotype (Rein et al., 2020).

Rat models for 16p11.2 CNVs have also been generated recapitulating craniofacial phenotypes, with mirror effects between the DEL and the DUP conditions (Qiu et al., 2019). Converging and male-specific deficits in social behaviour and novel object recognition have been also observed in 16p11.2 DEL and DUP rats in two different genetic backgrounds (Martin Lorenzo et al., 2023). RNA sequencing analysis in the hippocampus found 267 differentially expressed genes dysregulated in DEL and DUP rat models. Among these genes, 100 were downregulated and 120 upregulated in both models, which could explain some overlapping phenotypes, independent from gene dosage. Differential functional analysis revealed 23 upregulated pathways in both DEL and DUP rats, associated with morphogenesis of the primary cilium. However, pathways related to synaptic function and metabolism were mostly deregulated in DEL rats, while pathways associated with transcription, epigenomic regulation and hormone regulation were mostly affected in DUP animals (Martin Lorenzo et al., 2023).

Recently, Yang et al. carried out anatomical and electrophysiological analysis of developing interneurons in 16p11.2 DEL rats, after the identification of a subset of interneurons in human foetal cerebral cortex potentially vulnerable to genetic autism risk factors. In 16p11.2 DEL rats at P21, the number or position of INs was unchanged in either CA1 or somatosensory cortex. However, somatostatin-expressing INs in CA1 display hyperexcitability, with an enlarged axon initial segment. This finding, although limited to a single developmental stage and one type of INs, supports the idea that the 16p11.2 deletion may perturb the electrophysiological properties of developing INs, thereby affecting the excitation/inhibition (E/I) balance (Yang et al., 2023).

The sialophorin, or CD43, is a glycoprotein expressed on the surface of T lymphocytes promoting adhesion and activation during immune responses. This gene, together with CORO1A and KIF22 have been suggested to play a role in reduced lymphocytes count in 16p11.2 DEL patient presenting a concomitant low dosage of BOLA2 duplicone, located on 16p11.2 BP4-BP5 flanking region (Giannuzzi et al., 2019; Giannuzzi et al., 2022). In male 16p11.2 DEL mice, SPN protein is overexpressed mainly in telencephalic and cerebellar regions presenting decreased fiber density (Kumar et al., 2018).

The quinolinate phosphoribosyltransferase is a key enzyme involved in the catabolism of quinolinate, an intermediate in the synthesis of nicotinamide adenine dinucleotide (NAD). QPRT-mediated NAD biosynthesis plays a fundamental role in neuronal differentiation and neurite growth during development (Haslinger et al., 2018; Neves et al., 2022). QPRT expression is reduced in cell lines derived from 16p11.2 DEL patients, and its knock-down in neuroblastoma cells during differentiation significantly altered neuritic growth and complexity, whereas QPRT knock-out (KO) increased cell death during differentiation (Haslinger et al., 2018). Furthermore, QPRT KO leads to the downregulation of several genes, including GABRB3, SNTG2, KCNQ, CNTNAP2. These genes are implicated in the formation of GABAergic synapses formation and are associated with ASD and epilepsy (Haslinger et al., 2018). QPRT also interacts with neuroligin 3 (NLGN3), a postsynaptic transmembrane protein involved in synapse formation and neuron-glia connections which is often found mutated in ASD patients (Shen et al., 2015).

Human zymogen granule protein 16 is highly expressed in mucus-secreting cells its overexpression significantly suppresses tumour growth through T cells-mediated immune response activation (Meng et al., 2022). Although its role in 16p11.2 CNVs is unclear, it has been correlated with increased size of several brain areas in mice selectively ablated for this single gene (Kretz et al., 2023).

Kinesin family member 22 (KIF22) has been shown to mediate cell proliferation by regulating mitotic spindle, microtubules stability (Tokai et al., 1996; Tokai-Nishizumi et al., 2005), as well as the expression of CDC25C, a cell cycle regulator (Nilsson and Hoffmann, 2000).

Among all genes of 16p11.2 region, KIF22 is specifically enriched in neural progenitors in G2/M phase, while its expression levels significantly decline as cells become post-mitotic (Morson et al., 2021). During CNS development, KIF22 transcripts are clearly segregated in the ventricular and subventricular zones, where cortical progenitors arise. KIF22, together with ALDOA, HIRIP3, PAGR1, and MAZ are also expressed in the ventricular zone of the ganglionic eminences where interneuron progenitors reside, supporting the hypothesis that 16p11.2 CNVs may disrupt the excitatory and inhibitory components of the developing brain. KIF22 dosage seems to regulate neurogenesis by affecting the length of the cell cycle, thereby determining the proliferative or neurogenic fate of the progenitors. Kif22 gene dosage variations linked to 16p11.2 CNVs would thus affect cell-cycle kinetics and perturb neuronal formation during development (Morson et al., 2021). In Zebrafish, KIF22 and ALDOA are the only deletion “dosage sensor” genes for 16p11.2 CNVs, meaning that their functional levels are sensitive to hemizygosity and thus lead to a pathological phenotype. In particular, reduction of KIF22 is associated with abnormal brain morphology and bent tail, as well as deficient axonal development (Blaker-Lee et al., 2012). The same effect on axonal development was observed in Drosophila, in which silencing of KIF22 homologs leads to the development of ectopic innervation and off-target recurrent axon branches in the neuromuscular junction (Park et al., 2016). Interestingly, these results are in agreement with nerve regeneration studies conducted in rats, where Kif22 has been identified as a potential therapeutic target for promoting peripheral nerve injury repair via Schwann cell proliferation and migration (Lu et al., 2022).

The MYC-associated zinc finger protein is a transcription factor involved in several, highly specific, molecular processes related to gene expression and cell cycle, leading to differentiation and developmental effects. During development, MAZ is a cofactor of CTCF during chromatin insulation of active and repressed genes within the Hox clusters (Ortabozkoyun et al., 2022), specifically involved in the differentiation of motor neurons in vertebrates. Among its targets, MAZ regulates the expression of several WNT morphogens, involved in the correct morphological development of different organs. For instance, MAZ has been related to birth defects in the genitourinary tract, that are commonly observed in the context of 16p11.2 CNVs (Haller et al., 2018) as well as in eye development, also presenting comorbidity in 16p11.2 DEL and DUP patients (Medina-Martinez et al., 2020). As KIF22 and ALDOA, MAZ is also enriched in the ventricular and subventricular zones compared to post-mitotic cells, suggesting a role in neurogenesis (Morson et al., 2021). As a transcription factor, MAZ takes part in the correct synchronization between neurogenesis and NOTCH1 mediated gliogenesis. It is well known that during development, NOTCH1 signalling is downregulated during neurogenesis and activated in gliogenesis, following a specific timing. The upregulation of ADAM10 activates NOTCH1-mediated gliogenesis. Liu and colleagues have demonstrated that MAZ enhances ADAM10 transcription in response to activation factors such as CT-1 in cultured neuronal progenitor cells (NPC) (Liu et al., 2016). In neuronal stem cells (NSC), MAZ regulates the expression of Rho-GDIγ, which in turn modulates the activity of Rho GTPases during neuritogenesis, axon formation and dendritic development in neuronal differentiation (Wang et al., 2013). Moreover, MAZ activity has been reported to enhance NMDA receptor subunit type 1 (NR1) promoter activity during neuronal differentiation of P19 cells, contributing to the assignment of the correct functional profile to differentiated excitatory neurons (Okamoto et al., 2002). Thus, changes in MAZ dosage related to 16p11.2 CNVs might be related to structural changes in neuronal/glial tissue balance.

Proline-rich transmembrane protein type 2 (PPRT2) is a membrane protein, located at synaptic contacts, playing a role in synapse formation during development. It is also a crucial component of the neurotransmitters’ release machinery, by interacting with SNARE proteins and synaptotagmins 1 and 2 (Valtorta et al., 2016). PRTT2 has been found enriched in Drd2+ MSNs in 16p11.2 deletion mice, together with KCTD13, Fam57b, Sez6l2 and CORO1A (Portmann et al., 2014). Mutations in PPRT2 have been associated with paroxysmal movement disorders (Heron and Dibbens, 2013), benign infantile familial seizures and autistic developmental regression (Zhang et al., 2024), as well ataxia and seizures in 16p11.2 DEL patients (Vlaskamp et al., 2019; Padmanabha et al., 2024). Importantly, correction of PRTT2 copy number in 16p11.2 DUP mice could rescue hypersynchronous activity and enhanced glutamate release in cortical circuits, seizure susceptibility and social deficits (Forrest et al., 2023).

Pax-interacting protein 1 is a component of the histone methyltransferase MLL2/MLL3 complex, with possible role in DNA damage response (Cho et al., 2007; Gong et al., 2009). This protein, highly expressed in neural progenitors (Morson et al., 2021), appears to have a role in early embryonal development, as suggested by animal studies. For instance, loss of function of PAGR1a in Zebrafish is associated with reduced brain ventricle size and less defined midbrain-hindbrain boundaries (Blaker-Lee et al., 2012). In addition, null mice for PAGR1a are not viable, while the loss of one copy leads to abnormal development of extraembryonic tissues, such as the amnion, chorion, and visceral yolk sac (Kumar et al., 2014). A recent clinical report by Daum et al. identified three individuals, carrying a homozygous missense mutation in PAGR1a gene and showing microcephaly, severe developmental delay, dysmorphism, neurological deficits and death in infancy (Daum et al., 2022).

CDP-Diacylglycerol-Inositol 3-Phosphatidyltransferase (CDIPT) catalyses the biosynthesis of phosphatidylinositol and is highly expressed during foetal and neonatal brain development in rats (Nyquist and Helmkamp, 1989). CDIPT knock-down in Drosophila displays altered growth development of the neuromuscular junction, a model for studying synapse growth defects (Iyer et al., 2018). In addition, a missense mutation in CDIPT gene in Zebrafish causes photoreceptor cells death and cataract (Murphy et al., 2011).

Double C2-like domain-containing protein alpha (DOC2A) is a cytosolic protein interacting with SNARE complex and phospholipids, acting as a Ca2+ sensor and mainly expressed in glutamatergic neurons, triggering glutamate release (Courtney et al., 2018). In the context of neurodevelopmental disorder, it has been found specifically overexpressed in 16p11.2 DEL male mice in brain regions with decreased fiber density (Kumar et al., 2018). Interestingly, DOC2A-KO mice display abnormal morphology of hippocampal neurons in the dentate gyrus, defective hippocampal activity as well as social deficits and repetitive behaviours. DOC2A functions appear to be regulated by the Ca2+ binding protein Secretagogin (Wang et al., 2023).

Aldolase A (ALDOA) is a glycolytic enzyme which catalyses the conversion of fructose 1,6-biphosphate into glyceraldehyde 3-phosphate and dihydroxyacetone phosphate. ALDOA is highly expressed in neural progenitors and its expression declines as the cells become post-mitotic (Morson et al., 2021). It is also part of ERK1/2 interactome during epidermal and neuronal differentiation (von Kriegsheim et al., 2009). Knock-down studies in Drosophila demonstrated the involvement of this gene in the locomotor function, specifically in climbing activity. In addition, ALDOA knock-down produces altered cell counts and patterning of different cell types, including increased photoreceptors neurons and misplaced bristle cells (Iyer et al., 2018). In Zebrafish embryos, ALDOA loss of function is associated with no response to touch (Blaker-Lee et al., 2012).

Protein phosphatase 4, catalytic subunit (PPP4C) plays a role in several cellular processes, such as microtubule growth and organization, DNA damage checkpoint recovery, apoptosis, and TNF-alpha signalling (Chen et al., 2008). However, its role in the context of 16p11.2 syndromes has not been completely elucidated. It has been found overexpressed in 16p11.2 DEL mice specifically in regions with decreased fiber density (Kumar et al., 2018), while studies in Zebrafish suggest the role of PPP4C in locomotor function and axon growth (Blaker-Lee et al., 2012).

T-box transcription factor 6 (TBX6) is a putative DNA-binding protein involved in the specification of paraxial mesoderm via SOX2 regulation (Takemoto et al., 2011). Heterozygous loss of function of TBX6 gene in human and mice has been associated with congenital anomalies of the kidney and urinary tract (CAKUT), thus relating this gene with the kidney defects often observed in 16p11.2 DEL carriers (Yang et al., 2020). In addition, the increased incidence of congenital vertebral malformations in 16p11.2 DUP patients could be ascribed to TBX6 dosage, as also suggested by mouse studies (Ren et al., 2020).

Yippee-like three is a member of putative Zinc-finger motif-coding genes and p53-regulated tumour suppressor (Kelley et al., 2010). Its role in 16p11.2 syndromes is unknown, although YPEL3 loss of function Zebrafish display abnormal brain morphology (Blaker-Lee et al., 2012).

Glycerophosphodiester phosphodiesterase domain-containing protein 3 (GDPD3) has lysophospholipase activity against several lisophospholipids (Ohshima et al., 2015). Loss of function Zebrafish for this gene is characterised by abnormalities in brain morphology (Blaker-Lee et al., 2012). To date, no other information is available regarding its specific role in neurodevelopmental disorders.

Coronin 1A (CORO1A) is an actin-binding protein previously associated with lymphocyte trafficking and homeostasis (Foger et al., 2006). Mutations in this protein have been found in patients with severe combined immunodeficiency (SCID) (Khoreva et al., 2024). Its function in the context of 16p11.2 CNVs has not been determined yet, although it is associated with low lymphocytes count in 16p11.2 DEL patients (Giannuzzi et al., 2022). Zebrafish carrying CORO1A loss of functions display abnormal body length, defective neural tubes, deficient axon tracts and movement defects (Blaker-Lee et al., 2012). In 16p11.2 DEL mice, CORO1A is enriched in Drd2+ MSNs (Portmann et al., 2014).

Chromosome 16 open reading frame 54 (C16orf54) is a protein-coding gene expressed in 11 normal tissues and mainly located in the cell membrane. Although its function is unknown, overall C16orf54 appears as a potential marker of the poor prognosis of various tumours, suggesting that C16orf54 may participate in the progression of tumours (Du et al., 2022). C16orf54 expression is significantly associated with various immune-related pathways, such as PI3K/Akt/mTOR and Wnt/Beta catenin signalling pathways, inflammatory response, and interferon-gamma response (Ding et al., 2023). However, its specific biological function is still obscure and its role in 16p11.2 CNVs has not been investigated yet.

Chromosome 16 open reading frame 92 (C16orf92) encodes for a testis-specific protein necessary for the fusion between the oocyte and the sperm (Fujihara et al., 2020). To date, no other information is available regarding additional biological functions or potential association of this protein with neurodevelopmental disorders.

Fam57b consists of three variants upregulated during adipogenesis although at different degrees/timing. Fam57b mediates the production of lactosylceramide upon binding with a circulating metabolite of vitamin D with possible role in bone fracture healing (Martineau et al., 2018). Ceramides are precursors of sphingolipids, and their dysregulation can affect vesicle fusions and endocytic recycling. Ceramide derivatives interact with SNARE docking complex and recruit Munc13, the binding partner of Doc2A (calcium sensitive exocytosis regulator) thus enhancing calcium intracellular flow (Camoletto et al., 2009). McCammon et al. identified a subset of interacting 16p11.2 pairs of genes affecting brain and ventricle morphology in Zebrafish. In particular, DOC2A and Fam57a interacts to affect propensity to seizures and body size (McCammon et al., 2017). Its loss of function causes movement defects in Zebrafish, no response to touch and axonal defects (Blaker-Lee et al., 2012). In addition, Fam57b knock-down in Drosophila causes altered growth development of the neuromuscular junction (Iyer et al., 2018). Fam57b has been found enriched in Drd2+ MSNs in 16p11.2 DEL mice (Portmann et al., 2014), however its specific function in the context of neurodevelopmental disorders is currently unknown.

TMEM219 encodes for a transmembrane protein also known as IGFBP-3R (Insulin-like growth factor binding protein-3 receptor). It interacts with IGFBP-3 to mediate caspase-8-mediated apoptosis and tumour suppression in prostate and breast cancer xenografts (Ingermann et al., 2010). TMEM219 interacts with chitinase 3-like-1 (Chi3l1) and the multimeric complex formed by TMEM219/Chi3l1/IL-13Rα2 activates MAPK, AKT and Wnt signalling, inhibits oxidant-induced apoptosis in the lung, induces TGF- β1 and promotes melanoma metastasis (Lee et al., 2016). However, its role in 16p11.2 CNVs is currently unknown.

HIRIP3 is highly expressed in human and mice embryos, in both cortex excitatory and striatal medium spiny neuron progenitors, as well as in the adult human brain (Morson et al., 2021). Early evidence suggested that HIRIP3 may be part of a histone H3.3 chaperone complex along with HIRA (Magnaghi et al., 1998; Ricketts et al., 2019), while HIRIP3 seems to partner with histone H2A/H2B (Assrir et al., 2007). HIRIP3 loss of function is known to produce movement defects in Zebrafish (Blaker-Lee et al., 2012).

Ino80e is a component of Ino80, an ATP-dependent chromatin remodelling complex. Although the specific role of Ino80e subunit is unknown, studies in yeast support the role of Ino80 complex in chromatin remodelling, DNA replication, stabilization of replication fork and resumption of replication after stress (Papamichos-Chronakis et al., 2006; van Attikum et al., 2007; Papamichos-Chronakis et al., 2011). Loss of function of Ino80e in Zebrafish is associated with defective neural tubes and abnormal body length (Blaker-Lee et al., 2012). This gene has been also found overexpressed in regions with decreased fiber density in 16p11.2 DEL male mice (Kumar et al., 2018).

The role of aspartate beta-hydroxylase domain-containing protein (Asphd1) is still unknown. In general, aspartate beta hydroxylases are rarely expressed in normal adult tissues, while they are overexpressed in several malignancies thus mediating cell proliferation and metastasis. They also appear to be a downstream target of ERK/MAPK and PI3K pathways (Hou et al., 2018).