Antiviral signalling in human IPSC-derived neurons recapitulates neurodevelopmental disorder phenotypes

Maternal immune activation increases the risk of neurodevelopmental disorders. Elevated cytokines, such as interferon-gamma (IFNγ), in offspring’s brains play a central role. IFNγ activates an antiviral cellular state, limiting viral entry and replication. In addition, IFNγ has been implicated in brain development. Here, we hypothesise that IFNγ-induced antiviral signalling contributes to molecular and cellular phenotypes associated with neurodevelopmental disorders. We find that transient IFNγ treatment of neural progenitors derived from human induced pluripotent stem cells (hIPSCs) persistently increases neurite outgrowth, phenocopying hIPSC-neurons from autistic individuals. IFNγ upregulates antiviral PML bodies and MHC class I (MHCI) genes, which persists through neuronal differentiation. Critically, IFNγ-induced neurite outgrowth requires both PML and MHCI. We also find that IFNγ disproportionately alters expression of autism and schizophrenia risk genes, suggesting convergence between these genetic and environmental risk factors. Together, these data indicate that IFNγ-induced antiviral signalling may contribute to neurodevelopmental disorder aetiology.


Introduction
Multiple lines of evidence point to immune activation during foetal development as an important risk factor for neurodevelopmental disorders (1). Epidemiological studies indicate that maternal infection during pregnancy increases incidence of autism spectrum disorder (ASD) and schizophrenia (SZ) (2)(3)(4)(5). Animal studies have shown that induction of an antiviral immune response during pregnancy using the dsRNA mimetic poly(I:C) leads to behavioural abnormalities in offspring that are thought to be relevant to neuropsychiatric disorders, including repetitive behaviour, altered social behaviour, deficits in prepulse inhibition and working memory (6). Also, transcriptomic studies of both ASD and SZ post mortem brains consistently show enrichment for inflammatory and innate immune genes (7)(8)(9)(10). However, despite this association, the pathological mechanisms through which transient inflammatory activation increases susceptibility to neurodevelopmental disorders remain unclear.
Among the inflammatory cytokines upregulated during maternal immune activation (11), IFNγ is of particular interest. It is an activator of innate cellular antiviral signalling and transcription programmes whose primary function is to defend the cell against infection by viral particles (12). Mid-pregnancy maternal serum IFNγ is increased during gestation of offspring with ASD (13) and circulatory IFNγ levels are elevated in neonates subsequently diagnosed with ASD relative to developmental delay controls (14). Intriguingly, within the brain, many antiviral IFNγ signalling targets also play important roles in neuronal development and synaptic activity, independent of microbial infection (15)(16)(17). More recently, IFNγ has also been described to play a role in social behaviour in rodents, through modulation of inhibitory neuronal GABAergic tone (18). Thus, it is now emerging that IFNγ has a physiological role beyond its antiviral and immune actions.
Previous studies have shown that antiviral activation establishes enduring cellular changes that persist beyond the acute inflammatory response. Exposure to IFNγ primes cells to induce an enhanced transcriptional response upon re-stimulation, allowing cells to mount a faster and more effective antiviral response (19,20).
Examination of transcriptional priming at the MHC locus revealed a critical role for antiviral promyelocytic leukemia protein (PML) nuclear bodies (20). MHC Class I (MHCI) gene expression has also been shown to be persistently upregulated in developing neurons following gestational poly(I:C) exposure (21). In the brain, seemingly independently of their antiviral functions, both PML and MHCI proteins play important roles in many aspects of neuronal development and function, including neurite outgrowth and axon specification (15), synaptic specificity (16), synaptic plasticity (22) and cortical lamination (23). Importantly, all of these processes have been shown to be altered following gestational poly(I:C) exposure in rodents (6,11,21,24). Thus, PML nuclear bodies and MHCI proteins may link antiviral inflammatory activation to neuronal abnormalities. However, the impact of this pathway on neurodevelopment following inflammatory activation has never been examined.
In this study, we examine the hypothesis that IFNγ-induced antiviral signalling perturbs neurodevelopmental processes associated with neurodevelopmental disorders. Here, we used hIPSCs to investigate how transient IFNγ exposure affects developing neurons. We hypothesised that transient developmental inflammatory activation could recapitulate molecular and cellular phenotypes associated with neurodevelopmental disorders. Indeed, we demonstrate that exposing hIPSCderived neural progenitor cells (NPCs) to IFNγ led to persistently increased neurite outgrowth in hIPSC-neurons, mimicking a phenotype observed in hIPSC-neurons from individuals with ASD (25)(26)(27). RNA sequencing was used to characterise the acute and persistent transcriptomic responses to IFNγ. Importantly, we observed that genes of the MHCI protein complex were among the most significantly upregulated.
This was accompanied by a persistent increased expression of MHCI proteins and number of PML bodies. Critically, both PML and MHCI proteins were required for IFNγ-dependent effects on neuronal morphology. Furthermore, we observed higher numbers of PML bodies in NPCs derived from individuals diagnosed with ASD than controls, consistent with transcriptomic perturbations in post-mortem human brain (10). Interestingly, IFNγ responding genes were enriched for those with genetic association to ASD and SZ. Moreover, these genes overlapped significantly with those differentially expressed in the brains of individuals with these disorders.
Together, these findings highlight a potential mechanism through which antiviral signalling could contribute to intrinsic neuronal phenotypes in neurodevelopmental disorders.

IFNγ increases neurite outgrowth in a hIPSC model of neurodevelopment
We and others have previously observed alterations in the morphology of hIPSC neurons derived from autistic individuals (25)(26)(27). Since inflammatory mechanisms have been implicated in both neural development and neurodevelopmental pathology, we hypothesised that activation of antiviral signalling pathways may influence neuronal architecture. To address this, we used hIPSC-NPCs from three control male individuals with no psychiatric diagnoses (M1, M2 and M3; Supplementary Table 1; Fig. S1; 21,23) and treated these cells with IFNγ (25 ng/ml) daily on days (D) 17-21 of differentiation. Subsequently, IFNγ was excluded from cell culture media and neuronal differentiation was continued, resulting in postmitotic hIPSC-neurons (Fig.1A). Cells were fixed on D26, D30, D35 and D40, stained for β III-tubulin (Tuj1) and high-content automated neurite tracing was carried out (Fig.1, B and C). Both IFNγ treatment and days in culture were associated with a significant increase in total neurite length per cell across the time-course examined ( Fig. 1D; IFNγ treatment P = 0.0004; days in culture P < 0.0001). We used Sidak's multiple comparison test to compare IFNγ-treated and untreated neurons at individual timepoints and observed a significant increase in total neurite length in IFNγ-treated lines at D30, (P = 0.010) after which increased variability was observed. This is similar to the increased neurite length observed in ASD-derived neurons in previous studies (25)(26)(27).
In previous studies of cells from individuals with ASD, increased neurite branching, number, and length has also been reported (25)(26)(27) with IFNγ (Fig. 2, D and E). We also found that the genes downregulated by acutely treated neurons were enriched 2-fold for the GO term 'plasma membrane' (FDR = 0.002). Full differential expression and GO enrichment results are reported in supplementary files S1 and S2 respectively.
The persistent transcriptional response to IFNγ was of interest, given the enduring impact of IFNγ on neuronal morphology described above. To this end, we compared untreated neurons (30UU) and pre-treated neurons (30TU; Fig. 2A). Remarkably, neurons exposed to IFNγ at the NPC stage showed enduring transcriptional changes, with 26 genes significantly upregulated and 2 downregulated in postmitotic neurons 9 days after treatment (Fig. 2F). Notably, the upregulated gene set was enriched for MHCI genes, with the GO term 'MHC class I protein complex' enriched 357-fold (FDR = 3.33 x 10 -8 , Fig. 2H). Strikingly, the top five ranked DEGs in pretreated neurons were all involved in MHCI antigen presentation. Also of interest, the metabotropic glutamate receptor gene GRM3 was upregulated while the GABAergic transcription factor gene LHX6 was downregulated in the pre-treated neurons.
Interestingly, GRM3 has been implicated as a SZ risk gene (29), whereas LHX6 is required for specification and correct spatial positioning of parvalbumin interneurons, neurite outgrowth, and has been reported to be deficient in prefrontal cortices of individuals with SZ (30) (Fig 2F).
We also found evidence of IFNγ-induced cellular priming, where repeated exposure at the NPC and neuronal stages (30TT) induced considerably more DEGs than a single neuronal treatment (30UT) when compared to untreated neurons (30UU). This double hit induced 1,091 DEGs, 45% more than were detected after a single neuronal treatment (Fig. 2G). This is consistent with previous reports of IFNγinduced transcriptional priming (19,20). Notably, the genes downregulated by neurons that received a double hit were enriched 4-fold for the GO term 'synapse' (FDR = 0.049). Taken together, these results demonstrate that IFNγ exposure induces widespread and persistent transcriptional changes during human neuronal differentiation. Considering their previously proposed role in neurite outgrowth in mouse (15,31), these data highlight MHCI proteins as candidates to explain the IFNγ-induced morphological phenotype described above.

Regulation of MHCI genes by PML nuclear bodies
In non-neuronal cells, IFNγ has been shown to induce long-lasting changes in signaldependent transcription of MHC proteins through the formation of PML bodies (20).
PML bodies are dynamic, DNA-binding protein complexes which mediate myriad transcriptional functions from viral gene silencing (32) to neurogenesis (23) and neuronal homeostatic plasticity (22). Because the MHCI pathway was highly enriched in the pre-treated neurons and PML was similarly upregulated in acutely P < 0.0001). Conversely, treatment of postmitotic neurons had no effect on PML bodies, with or without pre-treatment at D17-21 (Fig. 3D). These results indicate a time-window of sensitivity in neuronal differentiation, during which IFNγ exposure leads to a persistent increase in PML bodies.
PML nuclear bodies are known to be specifically disrupted following binding of arsenic trioxide (As 2 O 3 ) (33). Therefore, we pre-treated NPCs with As 2 O 3 from D16-17, followed by co-treatment with As 2 O 3 and IFNγ from D17-18, then counted PML bodies per nucleus (Fig. 3, E i, F). As previously observed (Fig. 3A), IFNγ treatment led to a notable increase in PML bodies per nucleus (P < 0.0001); this was blocked by co-treatment with As 2 O 3 . Alone, As 2 O 3 had no detectable impact on number of PML bodies. To test whether As 2 O 3 treatment also prevented the persistent increase in PML bodies the experiment was repeated, this time allowing NPCs to differentiate into neurons (Fig. 3, E ii, G). Again, we observed a significant increase in PML bodies with IFNγ treatment alone (P < 0.0001) which was entirely prevented by cotreatment with As 2 O 3 . We could thus conclude that As 2 O 3 prevented both acute and persistent PML body induction.
To investigate whether PML bodies were required for IFNγ-dependent transcriptional activation of MHCI genes, we performed quantitative (q)PCR on NPCs exposed to  (Fig. 3, E i, I). We used a probe specific to the pre-spliced HLA-B RNA transcript. HLA-B was selected as it showed the greatest upregulation in pretreated neurons (Fig. 2F). IFNγ treatment led to a dramatic increase in HLA-B pre-mRNA ( Fig. 3, I, J and K; P < 0.0001). This induction was largely prevented by co-

IFNγ induces HLA-B transcription near PML nuclear bodies
If PML bodies directly regulate MHCI gene expression, we would predict them to be in close proximity to the site of transcription. To investigate this spatial relationship, we carried out RNA FISH using the probe described above, specific to pre-spliced HLA-B RNA transcript. Because splicing is highly localised (34,35), we reasoned that the site of the pre-spliced transcript could be used as a proxy for the location of transcription. Co-staining for PML and HLA-B pre-mRNA revealed that, following IFNγ treatment, HLA-B spots were frequently located immediately adjacent to, or overlapping with, PML bodies ( Fig. 3N). To investigate this further, we measured the density of PML bodies (spots per micron) within HLA-B spots or HLA-B spot perimeters (see Methods for full definitions) and compared this to the density of PML bodies across the nucleus as a whole in IFNγ-treated NPCs and neurons (Fig. 3,O and P;S3C and S3D). In IFNγ-treated NPCs, the increased density of PML bodies in HLA-B pre-mRNA spot perimeters did not reach statistical significance (P = 0.09).
However, in IFNγ-treated neurons, a significantly higher density of PML bodies was observed in the HLA-B pre-mRNA spots (P < 0.0001) and spot perimeters (P = 0.0007) than the nucleus as a whole, indicating a positive spatial association. By contrast, in untreated NPCs and neurons, PML spots were never observed to overlap with HLA-B pre-mRNA spots or spot perimeters ( Fig. S3C and S3D). To confirm the existence of a non-random spatial relationship, we carried out random shuffle nearest neighbour analysis on IFNγ-treated NPCs (36). Actual distances between HLA-B spots and PML bodies were observed to be significantly shorter than simulated distances following randomisation (P < 0.0001), confirming a positive spatial relationship (Fig. 3, Q and R). These results confirm that PML bodies are in closer proximity to the site of HLA-B transcription than would be expected by chance, supporting the hypothesis that PML is required for IFNγ-induced MHCI gene transcription.

PML and MHCI mediate IFNγ-induced neurite outgrowth
Having demonstrated a role for PML bodies in IFNγ-induced MHCI gene transcription, we next asked whether PML and MHCI were required for the IFNγdependent increased neurite outgrowth. MHCI proteins have previously been reported to be expressed during, and required for, neurite outgrowth in primary cultured rodent neurons (15). To examine the role of MHCI proteins in IFNγ-induced neurite outgrowth, we first carried out instant (i)SIM super-resolution microscopy to determine the subcellular localisation of MHCI proteins HLA-A, HLA-B and HLA-C ( Fig. 4A). We found MHCI proteins to be present in the neurites, growth cones and cell bodies of untreated neurons. We then examined the effects of IFNγ and As 2 O 3induced PML disruption on MHCI abundance in neurites and growth cones, following the experimental outline schematised in Fig. 3E ii. IFNγ treatment induced a significant increase in MHCI intensity within growth cones and neurite compartments Significantly increased actin intensity was also observed in both neurites and growth cones following IFNγ treatment ( Fig. 4E; neurites: P = 0.0020; growth cones: P = 0.0021). Co-treatment with As 2 O 3 prevented this increase in both compartments (P = 1 for both). Actin staining intensity was also increased in growth cones relative to adjoining neurites following IFNγ treatment (P < 0.0001), and this enrichment was prevented by co-treatment with As 2 O 3 ( Fig. 4F; As: P = 0.57; As + IFNγ: P = 0.13).
Together, these results indicate that PML-dependent IFNγ-activated signalling pathways have a functional impact on growth cone composition and actin dynamics.
To determine whether intact PML bodies are required for IFNγ-dependent increased neurite outgrowth, we treated NPCs with IFNγ and As 2 O 3 then continued differentiation of these cells into post-mitotic neurons and assessed neurite outgrowth. As previously described, we observed an increase in total neurite length per cell in the IFNγ-treated relative to the untreated neurons ( Fig. 4, G, H and I; P = 0.04). Importantly, this increase was prevented by co-treatment with As 2 O 3 (P = 0.98), supporting a requirement for PML in IFNγ-induced neurite outgrowth. These results support a model whereby IFNγ activates PML body formation, leading to MHCI gene transcription, expression of MHCI in growth cones and increased neurite outgrowth (Fig. 4J).

IFNγ-mediated increase in MHCI and neurite outgrowth requires expression of B2M
To rule out non-specific effects of As 2 O 3 and confirm the requirement for MHCI in IFNγ-dependent neurite outgrowth, we investigated the effect of blocking MHCI cell surface expression on IFNγ-induced morphological changes. To achieve this, we took advantage of the requirement for the B2M protein for cell surface expression of MHCI (37). Firstly, we sought to demonstrate that IFNγ-induced expression of B2M protein was required for MHCI expression in our NPCs. Indeed, we observed a significant increase in both B2M and MHCI expression following exposure to IFNγ

Disease relevance of IFNγ-induced transcriptional alterations
Previous studies have demonstrated that exposure to viral insults during development increases the risk of neurodevelopmental and psychiatric disorders (1).
With this in mind, we examined whether genes with genetic association to ASD and SZ were preferentially dysregulated in NPCs and neurons in response to IFNγ (Fig.   6A). Intriguingly, we found ASD-associated genes from the SFARI database (38) to be significantly overrepresented among those downregulated by NPCs in response to IFNγ (OR = 1.7, FDR = 0.02). These genes included the candidates NLGN3, SHANK2, UPF3B and NRXN3 (Fig. S4A). Interestingly, we also observed significantly fewer ASD risk genes than were expected by chance among those upregulated by NPCs in response to IFNγ (OR = 0.6, FDR = 0.02). These findings are consistent with a model in which reduced function of ASD risk genes contributes to neurodevelopmental dysregulation following antiviral immune activation. We then tested SZ risk genes curated by the PsychENCODE Consortium (PEC) (39). Again, we found an enrichment among the genes downregulated by NPCs in response to IFNγ (OR = 1.7, FDR = 0.004). Genes in this group included GRIN2A, PRKD1 and TSNARE1 (Fig. S5A). The SZ risk genes were also enriched among those upregulated by neurons in response to IFNγ (OR = 1.7, FDR = 0.009). Notable genes in this group included TCF4, ATXN7 and ZNF804A (Fig. S5B). Taken together, these data demonstrate that IFNγ exposure during human neuronal differentiation disproportionately alters genes with genetic association to ASD and SZ. Therefore, we see convergence between genetic and environmental risk factors for these disorders.
We were also interested to examine the overlap between the IFNγ-responding genes detected in this study with those found to be differentially expressed in the postmortem brains of patients with ASD and SZ. We utilised data from the PEC crossdisorder study, the largest transcriptome-wide analysis of ASD and SZ brains carried out to date (10). Remarkably, we found a highly significant enrichment of genes from both disorders (Fig. 6B). There was a significant overlap between genes upregulated to IFNγ was observed. Lastly, we also found a significant overlap between genes upregulated by neurons in response to IFNγ with those upregulated in the brains of SZ patients (OR = 2.7, FDR = 9.9 x 10 -12 ). The overlapping upregulated genes at both time points were enriched for the gene ontology terms "IFNγ-mediated signalling pathway" and "defence response to virus" in both disorders (File S3).
These data indicate that activation of IFNγ-dependent antiviral signalling modifies gene expression in a manner consistent with the dysregulation observed in the brains of patients with ASD and SZ.

NPCs from idiopathic ASD individuals display increased basal PML nuclear bodies.
Finally, to investigate the relevance of PML bodies to neurodevelopmental disorders, we examined hIPSC-derived NPCs originating from individuals diagnosed with ASD.
Cells from three individuals diagnosed with ASD alongside the three control cell lines were differentiated into NPCs as described previously ( Fig. 1A; S1), fixed and stained for PML on D18 (Fig. 6, C and D). The NPCs from individuals with ASD had more PML bodies per nucleus than controls, indicating over-activation in basal conditions ( Fig. 6, C, D; P = 0.036). This finding is consistent with published transcriptomic data from post mortem brain tissue (9,10,40) which showed increased expression of PML and HLA-B in the cerebral cortices of individuals with ASD ( Fig. 6, E and F). This, along with the enrichment of ASD-associated genes among the IFNγ-responding genes detected in this study provides compelling evidence for the relevance of antiviral signalling to ASD pathogenesis.

Discussion
Epidemiological and animal studies support a role for antiviral inflammatory activation in the aetiology of neurodevelopmental disorders. While the molecular mechanisms that underlie this association are unclear, inflammatory cytokines are thought to play a central role (11). In this study, we used hIPSCs to examine the impact of IFNγ exposure during human neuronal differentiation. Intriguingly, we  (41,42) and a human cancer cell line (43). This study is the first report of this effect in human neurons. Neurite outgrowth is a fundamental stage of neuronal maturation, where neural progenitors extend processes which can later become axons or dendrites.
Alterations in neurite outgrowth in the developing brain would be predicted to have implications for neuronal connectivity and ultimately brain function. Increased neurite outgrowth is a consistent finding from studies of hIPSC-neurons derived from individuals with ASD (25)(26)(27). Abnormalities have also been observed in hIPSCneurons from SZ patients (44). Post-mortem studies of individuals diagnosed with 1 ASD and SZ have pointed to abnormal cortical neuron organisation, dendritic arborization and dendritic spine density (45). The macrocephaly present in some ASD cases has been attributed to increased dendrite number and size (46).
Interestingly, offspring of pregnant dams exposed to a single gestational dose of poly(I:C) also display altered cortical development, with perturbed dendritic and synaptic development (6,11,21,24). Further evidence comes from large-scale genetic studies of ASD and SZ, which consistently identify genes encoding synaptic proteins and those involved in neuronal maturation (39,47). Together, these reports highlight the relevance of disturbed neurite outgrowth to the pathophysiology of neurodevelopmental disorders.
A key question we sought to address was how transient immune activation could have a lasting impact on neuronal phenotype. PML nuclear bodies are chromatin associated organelles that play an important role in viral infection response and transcriptional regulation. Moreover, PML has been shown to be involved in neuronal development and function (22,23). We found that IFNγ established a long-lasting increase in PML body number during the neuronal differentiation of hIPSCs. PML bodies were spatially associated with the site of HLA-B transcription and their disruption prevented IFNγ-induced MHCI transcription. Furthermore, PML body disruption blocked IFNγ-induced neurite outgrowth. While this phenomenon has not been examined before in human neurons, interferon-induced transcriptional memory has previously been observed in human cancer cells (20), mouse embryonic fibroblasts and mouse bone-marrow derived macrophages (19). In the latter study, IFNγ pre-treated cells were shown to mount an enhanced antiviral transcriptional response, conferring resistance to viral infection (19). Importantly, PML nuclear bodies were shown to mediate IFNγ-dependent transcriptional memory for the MHC Class II DRA gene in human cancer cells (20). Building on these findings, our data indicate that IFNγ exposure in neural progenitor cells establishes PML-dependent transcriptional memory in MHCI genes that persists through human neuronal differentiation. Remarkably, we also detected a higher number of PML bodies per nucleus in unstimulated hIPSC-NPCs derived from individuals with ASD when compared with controls. This is consistent with the observation that the PML gene is upregulated in the post-mortem ASD brain (15). cones (16,31). They have been implicated in neurite outgrowth in mouse primary hippocampal neurons (15) and synaptic stability in the mouse visual system (48).
Intriguingly, gestational poly(I:C) also leads to an enduring increase in MHCI protein in mouse cortical neurites (21). It is noteworthy that the MHC locus has shown the strongest association with SZ in multiple major genome wide association studies (49,50), however, strong linkage disequilibrium in the region has made identifying the causal variants a challenge. Combined with the existing literature, our results indicate that MHCI proteins are involved in IFNγ-induced neurite outgrowth. Further research is required to determine the mechanisms through which MHCI proteins alter growth cone dynamics.
A growing body of evidence indicates that genetic variants associated with ASD and SZ manifest their risk during critical periods of early brain development (8,(51)(52)(53).
Alterations in the expression of many risk genes combine to disturb fundamental neurodevelopmental processes involved in neuronal differentiation and maturation (39,54). In this study, we found that IFNγ exposure during human neuronal differentiation disproportionately altered the expression of genes associated with ASD and SZ. More specifically, risk genes from both disorders were overrepresented among genes downregulated by NPCs in response to IFNγ. SZ risk genes were also over-represented among genes upregulated by neurons in response to IFNγ. In total, 66 ASD and 104 SZ risk genes responded to IFNγ in our study with several high-profile candidates among them. One such example is PTEN, a tumour suppressor gene considered a high confidence risk gene for ASD (38). The disease associated mutations promote cortical macrocephaly observed in some ASD cases (55). Interestingly, PTEN was downregulated in neurons after IFNγ exposure in our study, consistent with the reduced activity caused by ASD-associated variants.
Other outstanding examples include ZNF804A, FOXP1, TSNARE1 and TCF4. It is not clear whether these IFNγ-responding risk genes are involved in the neurite phenotype we describe above. However, similar dysregulation in response to developmental inflammatory stimuli in vivo would be expected to have implications for brain development. The convergence of genetic and environmental risk factors on the same genes is noteworthy and may speak to the association between maternal immune activation and neurodevelopmental disorders.
We also observed a significant overlap between genes that were differentially expressed in response to IFNγ in our model with those found to be dysregulated in post-mortem ASD and SZ brains. The overlapping genes were enriched for genes of the IFNγ signalling pathway and antiviral response genes. The upregulation of immune and inflammatory factors is a consistent finding of transcriptomic studies of ASD and SZ brains (10). While this signal has typically been assumed to be driven by microglia, our results suggest that it may also have a neuronal origin. This observation provides validity to our model as IFNγ alters gene expression in a manner consistent with the dysregulation observed in the brains of patients with these disorders.
In summary, we find that antiviral immune activation in human NPCs induces morphological and transcriptomic changes associated with neurodevelopmental disorders. NPCs are thought to be central to the aetiology of these conditions. Their disruption can have lasting implications for neuronal migration, maturation and function (56). Recently, Schafer and colleagues observed that hIPSC-neurons from individuals with ASD displayed increased neurite outgrowth (27). Importantly, they demonstrated that this phenotype was mediated by aberrant gene expression in NPCs and could be prevented by bypassing the NPC stage, through direct conversion to neurons. In our study, we find that pathological priming of NPCs with IFNγ has a remarkably similar effect. Thus, perturbation of normal gene expression in NPCs, whether genetic or environmental, leads to altered neuronal maturation.
The degree to which this phenotype contributes to neurodevelopmental disorders remains an open question. Further investigation is required to determine whether the IFNγ-responding ASD and SZ risk genes contribute to the neurite outgrowth phenotype and whether PML or MHCI mediate perturbed risk gene expression.
Nonetheless, our results highlight antiviral signalling as a plausible link between early immune activation and neurodevelopmental disorders. This work provides a framework for future study of immune activation and gene-environment interaction in human neural development.

Study design
The initial objective was to determine whether IFNγ treatment of hIPSC-NPCs followed by continued differentiation into postmitotic neurons led to recapitulation of morphological characteristics of hIPSC-neurons derived from individuals with ASD.
Following confirmation of this, we carried out RNA-sequencing to determine the transcriptional changes brought about by acute treatment, pre-treatment and doubletreatment with IFNγ in NPCs and neurons. Further hypotheses were generated to test the requirement for PML nuclear bodies in IFNγ-dependent MHCI transcription and the requirement for PML and MHCI in IFNγ-dependent increased neurite outgrowth. We also carried out gene enrichment analysis to test the hypotheses that IFNγ would disproportionately alter the expression of ASD and SZ risk genes as well as those that are differentially expressed in post-mortem brains of individuals with these disorders. No data were excluded from any dataset. To account for variability between cultures, multiple biological replicates were generated where appropriate and as specified. Cells from a given hIPSC line were considered to be independent biological replicates when they were generated from hIPSC samples with a different passage number. Treatments were carried out on cells within the same biological replicate and paired or grouped statistical analysis was carried out, to limit the impact of this variability on measured outcomes. The numbers of hIPSC lines and biological replicates, sample sizes and statistical tests used are specified in figure legends.
Experiment specific parameters are outlined in the methodology section below.

HiPSC generation and neuralisation and treatment
Participants were recruited and methods carried out in accordance to the 'Patient iPSCs for Neurodevelopmental Disorders (PiNDs) study' (REC No 13/LO/1218).
Informed consent was obtained from all subjects for participation in the PiNDs study.

Ethical approval for the PiNDs study was provided by the NHS Research Ethics
Committee at the South London and Maudsley (SLaM) NHS R&D Office. HIPSCs were generated from human hair follicle keratinocytes, derived from three control males with no known psychiatric conditions and three males with diagnosed ASD (Supplementary Table 1). Three control and one ASD hIPSC lines were generated using a polycistronic lentiviral construct co-expressing the four reprogramming transcription factors, OCT4, SOX2, KLF4 and c-MYC, as previously reported (28).
On D21

Embryonic stem cell generation and neuralisation and treatment
The human embryonic stem cell ( (60) to quality trim the reads and remove adapters. The reads were then mapped to the human (GRCh38) reference genome using STAR (61). Count matrices were prepared using GenomicAlignments on Bioconductor (62) and differential gene expression analysis was carried out with DESeq2 (63). The threshold for statistical significance was Benjamini-Hochberg (BH) adjusted P value < 0.05. Genes with an adjusted P value of 0 were awarded padj = 1 x 10 -308 for the purpose of the volcano plots. Gene ontology analysis was carried out with DAVID (64) for biological process, molecular function and cellular component. Differentially expressed gene lists were tested for enrichment relative to a background of all genes awarded an adjusted P value by DESeq2 for that comparison. A BH-adjusted P value < 0.05 was the threshold for enrichment. The raw sequencing data can be accessed from www.synapse.org/IFNG.

Gene enrichment analysis
The ASD risk gene list was from the SFARI Gene database (38) and included all 736 genes from category 1-4 of the January 15 th 2019 release. The SZ list included the 1,111 risk genes compiled by the PsychENCODE Consortium (PEC) (39). None of the risk genes included in this analysis fall in the MHC region. Differentially expressed genes from post-mortem brains of ASD and SZ patients were taken from the PEC cross-disorder study, which employed 51 ASD, 559 SZ and 936 control post-mortem frontal and temporal cerebral cortex samples (10). Enrichment of the above gene sets in our DEGs was tested using a two tailed Fisher's exact test in R.
BH correction for multiple testing was applied. Up and down-regulated genes were tested separately. Background was controlled for by testing gene frequency within the differentially expressed sets against the frequency in the equivalent tested but non-differentially expressed set (awarded an adjusted P value > 0.05 by DESeq2). ReadyProbes, ThermoFischer, R37110), applied in PBS for 30 minutes following secondary antibody incubation and three washes in PBS.

High Content Cellomic Screening
For the initial neuronal morphology experiments, high content screening was performed using CellInsight™ (Thermo Scientific)  replicates. Identical detection parameters were used between conditions to allow direct comparison and paired statistical analysis. Analogous analysis was carried out for the MS3 HLA null hESC experiment with the Opera Phenix high content screening system (Perkin Elmer). Plating and staining were carried out as described above. The Opera Phenix was also used to quantify B2M and MHCI protein levels for the B2M shRNAi-mediated knock down and MS3 HLA null hESC knock out studies. In both cases, cells were fixed and stained immediately after IFNγ treatment at D18. Protein expression was measured as average cellular intensity, normalised to no-primary controls.

Quantitative PCR
RNA samples were collected in Trizol Reagent (Thermo Fisher, 15596026) and RNA extraction was performed using an RNEasy kit (Qiagen, 74104). DNA digest was performed using a TURBO DNA-free kit (Thermo Fischer, AM1907) to remove residual genomic DNA from samples. Reverse transcription to generate cDNA, was performed using SuperScript III (Invitrogen, 18080-044). QPCR was performed using EvaGreen Hot FirePol qPCR Mix Plus (Solis Biodyne, 08-24-00001) in a Biorad Chromo4 PCR detection system. QPCR primer sequences can be found in Supplementary Table 3. Gene expression fold-change between conditions was calculated using the Pfaffl comparative Ct method (65).

RNA Fluorescence In Situ Hybridisation
RNA FISH was carried out using an RNAScope™ double Z probe targeting presplicing HLA-B RNA (ACD, probe named Hs-HLAB-intron). Hybridisation and signal amplification was carried out using RNAScope™ 2.5 HD Detection Reagent kit (ACD, 322350), following manufacturer instructions. Briefly, cells were grown in 8well chamber slides (Merck, C7057), fixed in 10% Neutral Buffered Formalin for 30 minutes at RT, washed in PBS and incubated with the probe for 2 hours at 40°C.
Slides were washed and signal was amplified through a 6-step amplification process, followed by signal detection with a Fast-red chromogen label.
Super-resolution images were taken using a Visitech-iSIM module coupled to a Nikon Ti-E microscope with a Nikon 100x 1.49 NA TIRF oil immersion lens (Nikon, Japan), using 405/488/561nm lasers. Super-resolution images were deconvolved to increase contrast and resolution, using a Richardson-Lucy algorithm specific to the iSIM mode of imaging using the supplied NIS-Elements Advanced Research software (Nikon,Japan,v.4.6). Identical laser gain and offset settings were used within each biological replicate to enable direct comparison between conditions.
Semi-automated image analysis was performed using ImageJ software. Gaussian filters were applied to PML and RNAScope images to reduce noise and binary images were generated. Nuclei were used as boundaries within which PML and RNAScope spots were counted. Identical processing parameters were applied across conditions to allow paired or grouped statistical analysis. Numbers of PML bodies per µM 2 were measured within RNAScope spots, within RNAScope spot perimeter regions and within whole nuclei. RNAScope spot perimeter regions were defined as detected spot perimeters +/-0.3 µm, generating a ring-shaped region of interest.
DiAna, an Image J Plugin (36), was used to assess the spatial distribution of objects.
Binary images were generated from confocal images of PML and RNAScope staining. Centre-to-centre distances were measured from PML spots to RNAScope spots. Binary images of nuclei were used as a bounding box (mask) and randomised shuffle of objects in both channels was carried out 100 times. Centre-to-centre distances in shuffled images were computed and cumulated frequency distributions were calculated for real and simulated distances. This was repeated for 9 images across 3 cell lines. Real and simulated distances within a given image were matched by percentile for paired analysis.

Figure Legends
sharing and permission to use PEC data. Finally, we would like thank Dr. Anthony