ERA-NET NEURON: Investigation of the neuroinflammatory basis of the human type I

Lead Research Organisation: University of Manchester
Department Name: School of Biological Sciences

Abstract

Aicardi-Goutières syndrome (AGS) is a cause of very severe brain disease in children. A lot of
evidence exists to indicate that this brain damage is due to an inflammatory process involving the
production of a chemical called interferon. Interferon is normally produced by cells in response to
infection by a virus, and some children with AGS are initially misdiagnosed as having a viral-related disease because of the close clinical overlap of these states. In contrast to the production of interferon
secondary to virus, in AGS an excess of interferon occurs due to a primary genetic defect. Because of
the severity of the condition there is an urgent need to develop new treatments for AGS. We think this
should be possible if we better understand how the responsible genetic changes drive interferon
production.
We intend to use the very latest genetic and cell technologies to understand how brain cells are
damaged by inflammation in AGS. AGS is rare, and few parents feel able to agree to post-mortem if
their affected child dies. However, a family with affected twin girls has generously donated the brain of
one of their very recently deceased daughters. This almost unique resource will allow us to study
human neurological tissue in more detail than ever before, and using new techniques which have not
been available previously. Additionally, we are going to make use of state-of-the-art methods for
producing 'brain cells' in a test-tube - derived from cells of patients with AGS, so that we can study
brain tissue in the context of several different genetic sub-types of AGS. We also intend to analyse a
strain of mouse which has changes in one of the AGS-related genes as another approach to trying to
understand the human disease.
Although AGS is rare, the study of the genes and the proteins related to the disease has become of
very high scientific importance - partly because of the involvement of these genes / proteins in how
the body reponds to infection by HIV-1 (the virus that causes AIDS), and also because of an overlap
with diseases where the body attacks itself - so-called autoimmunity. Thus, not only will our work be relevant to the children and families affected by this dreadful condition, it may also have relevance for a much wider set of human medical disorders.

Technical Summary

Studies of Aicardi-Goutières syndrome (AGS) have proven of major scientific importance by directly informing the biology of infection - particularly HIV-1 (e.g. Nat Immunol 2010; 11: 1005-13; Nature 2011; 474: 654-7), interferon metabolism (e.g. Cell 2008; 134: 587-98), DNA maintenance (e.g. Cell
2012; 149: 1008-22; Nature 10.1038/nature13292), and retroelements (e.g. Nat Immunol 2014; 15:415-422). AGS is undoubtedly a neuroinflammatory disease, as evidenced by its remarkable clinical mimicry of in utero viral infection (e.g. Dev Med Child Neurol 2008; 50: 410-6), a robust association with elevated cerebrospinal fluid white cell counts, intrathecal overproduction of the
primary antiviral cytokine interferon (e.g. J Neurol Sci 1988;84:201-8) and the inflammatory marker neopterin (Dev Med Child Neurol 2009; 51: 317-23), the recent observation of brain-reactive
autoantibodies (Ann Rheum Dis annrheumdis-2014-205396), and the notable observation of a
progressive inflammatory encephalopathy in a transgenic mouse model chronically producing
interferon alpha from astrocytes - which recapitulates the neuroimaging and pathological features of AGS, and some viral encephalopathies, to a remarkable degree (J Immunol 1998; 161: 5016-26; Brain Res 1999; 835: 46-61). Although much is now understood about the genetic basis of AGS, very little is known of the
mechanics of innate and adaptive immune system engagement directly relevant to the, primary,
neurological phenotype - not least because, considering the sensitivities around infant post-mortem examination, particularly where a diagnosis has already been reached, neurological tissue from
patients affected by AGS is extremely scarce. Here we propose a multidisciplinary study of AGSassociated neuroinflammation. Our project, combining expertise of high order (spanning mouse and
human biology), involves a portfolio of complementary but discrete approaches - thus mitigating risks
related to any one particular aspect of the project. In pursuing these studies, including the use of stateof-the-art iPSC and deep sequencing technologies, we will derive data informing the future treatment of the devastating AGS phenotype (across all the recognised genotypes), and which may also be relevant to our understanding of congenital infection, the biology of HIV-1, and neuro-lupus.
 
Description MRC PhD studentship (through UoM MRC Doctoral Training Partnership)
Amount £60,000 (GBP)
Organisation Medical Research Council (MRC) 
Sector Public
Country United Kingdom
Start 09/2018 
End 03/2022
 
Description Stroke Association, non-clinical lectureship award (Ref: TSA LECT 2017/02).
Amount £199,986 (GBP)
Funding ID TSA LECT 2017/02 
Organisation Stroke Association 
Sector Charity/Non Profit
Country United Kingdom
Start 08/2017 
End 07/2022
 
Title SAMHD1 animal model 
Description Project title: Investigation of the neuroinflammatory basis of the human type I interferonopathy Aicardi-Goutières syndrome (AGS) Summary of the work undertaken at the University of Manchester through the MRC-funded contribution to the NEURO-IFN five centre project (Grant Ref: MR/M501803/1. Starting 1/2/15). SAMHD1 animal model Background In 2009 we identified mutations in SAMHD1 to cause one of the seven subtypes of Aicardi-Goutières syndrome (AGS), a genetically determined encephalopathy associated with an upregulation of type I interferon (IFN) [1]. Clinically, SAMHD1-related AGS differs from other AGS subtypes in that many patients demonstrate a cerebral vasculopathy and are prone to intracerebral haemorrhage (ICH) and stroke [2-4]. The pathological basis of this phenotype is predicted to relate to a discrete role for SAMHD1 in neurovascular homeostasis - likely through an immune/inflammatory-mediated mechanism. The immune function of SAMHD1 has been the subject of intense study; however, an understanding of its role in the neurovasculature remains unclear. Mutant SAMHD1 mice have been generated, but these mice do not exhibit neurovascular disease [5, 6]. In contrast, using an antisense morpholino approach, we published a temporal samhd1 gene knockdown zebrafish embryo model, apparently recapitulating the cerebrovascular and innate immune phenotypes observed in AGS [7]. Morpholinos cannot integrate into the genome, so that permanent genetic modification is impossible using this strategy. Thus, for long-term investigation a genetically stable mutant samhd1 model is necessary, and the development of such has been the main focus of our ERA-NET funded work in Manchester. Although this theme was not originally envisaged as part of the ERA-NET application, we considered that the possibility to derive an animal model recapitulating the neuro-inflammatory features associated with the human disease due to SAMHD1 mutations could be fully justified within the remit of the NEURO-IFN grant. Considering the insights that a zebrafish animal model of inflammatory cerebrovascular disease might afford, we decided to develop a stable loss of function SAMHD1 system. Three mutant samhd1 lines are now available in Manchester (1 x ENU mutant, 2 x CRISPR). Given the nature of the human condition and the samhd1 morphant phenotype, severe developmental phenotypes were expected in homozygous mutants for each of these three lines. However, all homozygotes are viable and survive to adulthood. Further phenotypic exploration has focussed primarily on one of these mutant lines - the CRISPR/Cas9-induced 23bp deletion (?23bp) line - details of which are described below. Embryos injected (F0) with guide-RNA (gRNA) specific for exon 4 of the zebrafish samhd1 gene and Cas9 RNA were assessed for indels using restriction diagnostics and raised to adulthood before outcrossing onto a WT background (F1). Sanger sequencing of F1 DNA identified the presence of two heterozygous mutant alleles - a 2bp deletion and a 23bp deletion. Both alleles are predicted to cause truncation of the translated protein. Heterozygous F1 fish were grown to adulthood and subsequently in-crossed to test for homozygosity. Phenotypic assessments have focussed on the ?23bp line (Fig. 1). Gross developmental abnormalities were assessed in mixed genotype clutches; however, no genotype:phenotype correlations were observed. Due to the retention of maternal RNA/proteins in larval yolk, early developmental phenotypes are often inhibited in zebrafish until the yolk gradually recedes over the first few days of life. To test for this, we performed survival analyses between 5 and 14 days post-fertilisation. However, once again, no significant trend for homozygous lethality was observed. Indeed, homozygous mutants (F1 and F2) are viable, breed, and survive to over 12 months of age. Whole brains from three age-matched (~13 months) WT, heterozygous and homozygous siblings have been harvested, fixed and sectioned. Histological analysis will be performed beyond the end of the grant, to assess for neurovascular/inflammatory pathology. Since SAMHD1-related AGS is associated with an increased risk of ICH, and a significant proportion of samhd1 morphant zebrafish larvae exhibit brain haemorrhage [7], we tested for brain bleeding phenotypes in the WT and homozygous samhd1 ?23bp siblings using the haemoglobin stain o-dianisidine during larval stages. In untreated larvae, no significant difference in the proportion of animals exhibiting brain haemorrhages was observed between WT and homozygous mutants (Fig.2). However, following treatment with atorvastatin, a chemical known to induce neurovascular rupture and brain-specific bleeding in zebrafish embryos [8], an increase in the proportion of larvae displaying brain haemorrhages was observed in homozygous mutants in comparison to WT. Within the 0.5-1µM dose range, these differences were statistically significant (Fig.2), suggesting that homozygosity for the samhd1 23bp deletion is associated with a degree of neurovascular instability. Future work will continue to investigate this observation. For example, we have now crossed the samhd1 ?23bp allele onto a kdrl:mCherry transgenic reporter background [9] to allow for observation of blood vessel formation in the brain. Increased expression of type I IFN stimulated genes (ISGs) is a biochemical hallmark of all AGS subtypes [10]. We previously showed that temporal gene knockdown of samhd1 is associated with significant upregulation of zebrafish ISG expression [7]. To determine whether germline mutation of samhd1 in zebrafish also induces a type I IFN signature, we performed quantitative PCR on WT and homozygous mutant siblings. Fish were harvested and RNA was extracted at 5, 14 and 28 days post-fertilisation (dpf), and qPCR performed using a previously established panel of Taqman probes [7]. At each time point, an increased expression of some of these genes was observed in homozygotes. However, inconsistencies between genes and ages exist (Fig.3). To further investigate this matter, we treated larvae with Poly (I:C) - a synthetic double-stranded RNA molecule which is commonly used to evoke type I IFN signalling [11]. RNA from untreated and treated larvae was harvested at 5dpf. As before, untreated homozygotes exhibited increased expression of some of the genes within our panel in comparison to WT, and a trend for augmented expression was observed in homozygotes treated with Poly (I:C) (Fig.4). However, the biological relevance of these data remains unclear, and further investigation of this point is ongoing using other stimulation techniques. Morpholino knockdown of samhd1 was also associated with increased expression of the zebrafish type I IFN, ifnphi1 [7]. Thus, the samhd1 ?23bp allele has been crossed onto the ifnphi1:mCherry transgenic reporter background in preparation for future experiments. To determine the effects of the 23bp deletion on protein translation, we performed western blotting on WT and homozygote siblings using established protocols [7]. However, we experienced significant issues with antibody specificity and/or recognition of the samhd1 protein. Numerous antibody batches have been tested from a range of suppliers. However, we have still not obtained a satisfactory result to date. Due to the severity of SAMHD1-related AGS and the samhd1 morphant zebrafish phenotype, we expected to observe a strong phenotype in our stable mutant samhd1 zebrafish line. One potential explanation for the absence of such might be the phenomenon of 'genetic compensation' that can occur in stable mutant zebrafish [12]. Of note, the samhd1-like gene is a recently annotated zebrafish transcript that is annotated as a paralogue of samhd1. Structurally, the translated samhd1-like protein shares significant homology with the C-terminal of samhd1. Mutant constructs, resembling the structure of this paralogue, retain HIV restriction activity when overexpressed in cells [13], indicating that the function of this protein is comparable to samhd1 itself. Therefore, it is feasible that expression of samhd1-like could compensate for loss of samhd1 in homozygous mutants. Precedent for this possibility exists in the literature, in a tdp43/tdp43-like double mutant model of amyotrophic lateral sclerosis [14]. Although genomic sequences are highly homologous, differences exist, and 5 mismatches in the target sequence used for the samhd1 gRNA and the same region in samhd1-like are present. Sequence analysis using samhd1-like-specific primers indicates that the WT transcript exists in homozygous mutant animals. Given these data, new samhd1-like-specific gRNA have been generated and we are now raising a samhd1/samhd1-like double mutant line which will then be assessed for neurovascular and inflammatory defects. Work in the last year In an attempt to identify novel molecular dysregulation consequent upon the samhd1 ?23bp allele, we performed RNA-Seq on WT and homozygous mutant animals. RNA was extracted from pooled groups (n=20) of whole larvae at 5dpf. WT and homozygous mutant pooled groups were harvested as four biological replicates. Differential gene expression analyses were performed by the Bioinformatics Core Facility at the University of Manchester. Ingenuity Pathway Analysis (IPA) software was used for further assessment of transcript expression. Most strikingly, IPA revealed that the expression of nine genes (sqle, nsdhl, ebp, dhcr24, hsd17b7, msmo1, lss, sc5d and cyp51a1) belonging to the 'Cholesterol Biosynthesis' pathway was significantly reduced in the homozygous mutant group compared to WT (Fig. 5). qPCR verification of these data revealed a reduction in the expression of all six genes tested in homozygous samhd1 ?23bp larvae in comparison to WT larvae at 5dpf (Fig. 6A). A comparable, albeit smaller, reduction in expression of these genes was also observed in RNA derived from adult brain tissue obtained from WT, heterozygous and homozygous samhd1 ?23bp fish at ~ 1.5 years of age (Fig. 6B). We are currently performing experiments to measure cholesterol levels per se in this model. However, at the transcriptional level, these data indicate that the samhd1 ?23bp allele is associated with a reduction in cholesterol. Statins are HMG-CoA reductase inhibitors commonly prescribed for the lowering of cholesterol levels and to reduce the risk of cardiovascular disease. Interestingly, treatment of zebrafish embryos with atorvastatin induces ICH. This is predicted to be due to a reduction in cholesterol and essential lipids necessary for stabilisation of endothelial cell-cell contacts within the neurovasculature, leading to weakness of blood vessel walls and rupture [15]. Although the developing zebrafish brain is likely to represent an 'extreme' model of such a mechanism, recent epidemiological studies have also identified low cholesterol as a risk factor for ICH in humans [16, 17], suggesting that this phenomenon has clinical and translational relevance. Although transcriptomic analyses reveal that homozygous samhd1 ?23bp larvae demonstrate reduced cholesterol levels, these fish do not exhibit any physical phenotype in the absence of treatment. However, upon stimulation with 'sub-threshold' doses of atorvastatin, they are vulnerable to ICH (Fig.2). These data indicate that the putative genetically-induced deficiency in cholesterol synthesis alone is not sufficient to drive an ICH phenotype. However, treatment with low dose atorvastatin in some way 'breaches' a threshold of neurovascular in / stability, manifest by ICH. Although a modest increase in the expression of type I IFN induced gene transcripts has been observed in immune cells obtained from Samhd1 knockout (KO) mice, these animals do not display any physical phenotype [6]. We hypothesised that a lack of phenotype might be explained by a 'cholesterol deficiency' threshold, where an environmental challenge is required to sufficiently reduce cholesterol levels for induction of a neurovascular phenotype akin to that seen in our zebrafish model. To test this possibility, we harvested RNA from brain and liver of WT and Samhd1 KO sex-matched mice at ~3 weeks of age (in collaboration with Prof. Jan Rehwinkel, University of Oxford). Taqman qPCR was performed using probes targeting the mouse orthologues described in Fig.6. No differences between genotypes were observed in gene expression in either brain or liver (Fig. 7). These data suggest that loss of mouse Samhd1 is not associated with a cholesterol biosynthesis deficiency at the transcriptional level. The role of SAMHD1 in viral restriction and innate immunity is extremely well documented [18]. However, these functions do not easily explain why, specifically, AGS5 patients are susceptible to ICH. Indeed, we have hypothesised that SAMHD1 has an unrecognised function in the maintenance of neurovascular integrity. Given the link between low cholesterol and the risk of human ICH [16, 17], our zebrafish data indicate a novel function of SAMHD1 in cholesterol biosynthesis worthy of further study. In this regard, we note that it has been shown that limiting cholesterol synthesis in macrophages can engage a type I IFN response [19]. We are continuing to investigate the possible link between SAMHD1, AGS5 and cholesterol biosynthesis, and have recently been awarded an MRC PhD studentship to continue this work - with Dr Kasher as the PI. References 1. Rice, G.I., et al., Mutations involved in Aicardi-Goutieres syndrome implicate SAMHD1 as regulator of the innate immune response. Nat Genet, 2009. 41(7): p. 829-32. 2. du Moulin, M., et al., Cerebral vasculopathy is a common feature in Aicardi-Goutieres syndrome associated with SAMHD1 mutations. Proc Natl Acad Sci U S A, 2011. 108(26): p. E232; author reply E233. 3. Ramesh, V., et al., Intracerebral large artery disease in Aicardi-Goutieres syndrome implicates SAMHD1 in vascular homeostasis. Dev Med Child Neurol, 2010. 52(8): p. 725-32. 4. Xin, B., et al., Homozygous mutation in SAMHD1 gene causes cerebral vasculopathy and early onset stroke. Proc Natl Acad Sci U S A, 2011. 108(13): p. 5372-7. 5. Behrendt, R., et al., Mouse SAMHD1 has antiretroviral activity and suppresses a spontaneous cell-intrinsic antiviral response. Cell Rep, 2013. 4(4): p. 689-96. 6. Rehwinkel, J., et al., SAMHD1-dependent retroviral control and escape in mice. EMBO J, 2013. 7. Kasher, P.R., et al., Characterization of samhd1 Morphant Zebrafish Recapitulates Features of the Human Type I Interferonopathy Aicardi-Goutieres Syndrome. J Immunol, 2015. 8. Gjini, E., et al., Zebrafish Tie-2 shares a redundant role with Tie-1 in heart development and regulates vessel integrity. Dis Model Mech, 2011. 4(1): p. 57-66. 9. Ulrich, F., et al., Neurovascular development in the embryonic zebrafish hindbrain. Dev Biol, 2011. 357(1): p. 134-51. 10. Rice, G.I., et al., Assessment of interferon-related biomarkers in Aicardi-Goutieres syndrome associated with mutations in TREX1, RNASEH2A, RNASEH2B, RNASEH2C, SAMHD1, and ADAR: a case-control study. Lancet Neurol, 2013. 12(12): p. 1159-69. 11. Chen, S.N., P.F. Zou, and P. Nie, Retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs) in fish: current knowledge and future perspectives. Immunology, 2017. 12. Rossi, A., et al., Genetic compensation induced by deleterious mutations but not gene knockdowns. Nature, 2015. 524(7564): p. 230-3. 13. Arnold, L.H., et al., Phospho-dependent Regulation of SAMHD1 Oligomerisation Couples Catalysis and Restriction. PLoS Pathog, 2015. 11(10): p. e1005194. 14. Schmid, B., et al., Loss of ALS-associated TDP-43 in zebrafish causes muscle degeneration, vascular dysfunction, and reduced motor neuron axon outgrowth. Proc Natl Acad Sci U S A, 2013. 110(13): p. 4986-91. 15. Eisa-Beygi, S., et al., The 3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR) pathway regulates developmental cerebral-vascular stability via prenylation-dependent signalling pathway. Dev Biol, 2013. 373(2): p. 258-66. 16. Chang, J.J., et al., Higher low-density lipoprotein cholesterol levels are associated with decreased mortality in patients with intracerebral hemorrhage. Atherosclerosis, 2017. 269: p. 14-20. 17. Phuah, C.L., et al., Subacute decline in serum lipids prece 
Type Of Material Model of mechanisms or symptoms - human 
Year Produced 2018 
Provided To Others? Yes  
Impact None so far. 
 
Title SAMHD1 knock-out fish derived by CRISPR 
Description In 2009, we identified that mutations in SAMHD1 cause one of the seven subtypes of Aicardi-Goutiéres syndrome (AGS), a genetically determined encephalopathy associated with an upregulation of type I interferon (IFN) [1] . Clinically, SAMHD1-related AGS differs from other AGS subtypes in that many patients demonstrate a cerebral vasculopathy and are prone to intracerebral haemorrhage (ICH) and stroke [2-4]. The pathological basis of this phenotype is predicted to relate to a discrete role for SAMHD1 in neurovascular homeostasis - likely through an immune/inflammatory-mediated mechanism. We, and others, have started to define the biochemical function of SAMHD1 in relation to immunity; however, an understanding of a regulatory role for SAMHD1 in the neurovasculature remains unclear. Mutant SAMHD1 mice have been generated, but these mice do not exhibit neurovascular disease [5, 6]. In contrast, using an antisense morpholino approach, we recently published a temporal samhd1 gene knockdown zebrafish embryo model, apparently recapitulating the cerebrovascular and innate immune phenotypes observed in AGS [7]. However, morpholinos cannot integrate into the genome, so that permanent genetic modification is impossible using this strategy. Thus, for long-term investigation a genetically stable mutant samhd1 model is necessary and has been the initial focus of our ERA-NET related-work in Manchester. Although this work was not originally envisaged as part of the ERA-NET application, we consider that the possibility to derive an animal model recapitulating the neuro-inflammatory features associated with the human disease due to SAMHD1 mutations can be fully justified within this remit. Considering the insights that a zebrafish animal model of inflammatory cerebrovascular disease might afford, we decided to try to develop a stable loss of function SAMHD1 system. Three mutant samhd1 lines are now available in Manchester (1 x ENU mutant, 2 x CRISPR). Due to the nature of the human condition and the samhd1 morphant phenotype, severe developmental phenotypes were expected in homozygous mutants for each of these three lines. However, all homozygotes are viable and survive to adulthood. Exploration for subtle/mild phenotypes have been performed primarily in one mutant line - the CRISPR/Cas9-induced 23bp deletion line - details of which are described below. Embryos injected (F0) with gRNA specific for exon 4 of the zebrafish samhd1 gene and Cas9 RNA were assessed for indels using restriction diagnostics and raised to adulthood before outcrossing onto a WT background (F1). F1 genotyping identified the presence of two heterozygous mutant alleles - a 2bp deletion and 23bp deletion. Both alleles are predicted to cause truncation of the translated protein. Heterozygous F1 fish were grown to adulthood and subsequently incrossed to test for homozygosity. Phenotypic assessments have focussed on the 23bp deletion line. Gross developmental abnormalities were assessed in mixed genotype clutches however no genotype:phenotype correlations were observed. Due to the retention of maternal RNA/proteins in larval yolk, early developmental phenotypes are often inhibited until the yolk gradually recedes over the first few days of life. To test for this, we performed survival analyses between 5 and 14 days post-fertilisation; however no significant trend for homozygous lethality was observed. Indeed, homozygous mutants (F1 and F2) are viable, breed and survive to over 12 months of age. Whole brains from three age-matched (~13 months) WT, heterozygous and homozygous siblings have been harvested, fixed and sectioned and histological analysis will be performed to assess for neurovascular/inflammatory pathology. Defective swimming was observed in a single homozygous mutant. As SAMHD1-related AGS5 is associated with an increased risk for ICH and a significant proportion of samhd1 morphant zebrafish larvae exhibit brain haemorrhage [7], we tested for brain bleeding phenotypes in mutant samhd1 siblings using the haemoglobin stain, o-dianisidine, during larval stages. No significant haemorrhaging was observed in any genotype using this particular assay. In an attempt to identify structural defects within the neurovasculature, the samhd1 23bp deletion allele has recently been crossed onto a kdrl:mCherry transgenic reporter background [8] to allow for observations of blood vessel formation in the brain. This line will be available for assessment within the next 3 months. Increased expression of type I IFN signature genes (ISGs) is a biochemical hallmark of all AGS subtypes [9]. Additionally, we have shown temporal gene knockdown of samhd1 is associated with significant upregulation of putative zebrafish ISG expression [7]. Therefore, to determine whether germline mutation of samhd1 in zebrafish also induces a type I IFN signature, we performed quantitative PCR on WT and homozygous mutant siblings. Fish were harvested and RNA was extracted at 5, 14 and 28 days post-fertilisation (dpf) and qPCR was performed using a previously established panel of Taqman probes [7]. At each time point, increased expression of some of the genes was observed in homozygotes, however inconsistencies between genes and ages exist. To further investigate the potential significance of the inflammatory response observed in homozygotes, we treated larvae with Poly (I:C) - a synthetic double-stranded RNA molecule which is commonly used to evoke type I IFN signalling [10]. RNA from untreated and treated larvae were harvested at 5dpf. As before, untreated homozygotes exhibited increased expression of some of the genes within our panel in comparison to WT. Poly (I:C) treatment stimulated increased expression of many of the genes within this panel, and a trend for augmented expression was observed in homozygotes. However, inconsistencies exist, and investigations are ongoing using other stimulation techniques. Morpholino knockdown of samhd1 was also associated with increased expression of the zebrafish type I IFN, ifnphi1 [7]. Therefore, the samhd1 23bp deletion has been crossed onto the ifnphi1:mCherry transgenic reporter background and will be ready for assessment in 2 months. To determine the effects of the 23bp deletion on protein translation, we have performed western blotting on WT and homozygote siblings using established protocols [7]. Unfortunately, we have experienced significant issues with antibody specificity and/or recognition of the samhd1 protein. Numerous antibody batches have been tested from a range of suppliers. However, we have still not obtained a satisfactory result to date. We are currently discussing the possibility of generating custom-made, zebrafish specific anti-samhd1 antibodies to address this problem. The potential type I IFN defect and lack of neurovascular phenotype observed in mutant samhd1 zebrafish resembles the situation in mice [5, 6]. However due to the known phenomenon of genetic compensation that is frequently experienced in stable mutant zebrafish lines [11], it is still possible that further genetic interrogation may lead to a clinically-relevant disease model. Of potential relevance, the samhd1-like gene is a recently annotated zebrafish transcript that might function as a paralogue of samhd1. Structurally, the translated samhd1-like protein shares significant homology to the C-terminal of samhd1. Of possible note, mutant constructs resembling this paralogue retain HIV restriction activity when overexpressed in cells [12], indicating that the function of this protein might be comparable to samhd1 itself - in which case it is possible that expression of samhd1-like could compensate for loss of samhd1 in homozygous mutants. Precedent for this theory exists in the literature, in a tdp43/tdp43-like double mutant model of amyotrophic lateral sclerosis [13]. Although genomic sequences are highly homologous, differences exist and 5 mismatches in the target sequence used for the samhd1 gRNA and the same region in samhd1-like are present. Sequence analysis using samhd1-like-specific primers indicates that the WT transcript exists in homozygous mutant animals. Given these data, new samhd1-like-specific gRNA has been generated and we are now developing a samhd1/samhd1-like double mutant line which will then be assessed for neurovascular defects. References 1. Rice, G.I., et al., Mutations involved in Aicardi-Goutieres syndrome implicate SAMHD1 as regulator of the innate immune response. Nat Genet, 2009. 41(7): p. 829-32. 2. du Moulin, M., et al., Cerebral vasculopathy is a common feature in Aicardi-Goutieres syndrome associated with SAMHD1 mutations. Proc Natl Acad Sci U S A, 2011. 108(26): p. E232; author reply E233. 3. Ramesh, V., et al., Intracerebral large artery disease in Aicardi-Goutieres syndrome implicates SAMHD1 in vascular homeostasis. Dev Med Child Neurol, 2010. 52(8): p. 725-32. 4. Xin, B., et al., Homozygous mutation in SAMHD1 gene causes cerebral vasculopathy and early onset stroke. Proc Natl Acad Sci U S A, 2011. 108(13): p. 5372-7. 5. Behrendt, R., et al., Mouse SAMHD1 has antiretroviral activity and suppresses a spontaneous cell-intrinsic antiviral response. Cell Rep, 2013. 4(4): p. 689-96. 6. Rehwinkel, J., et al., SAMHD1-dependent retroviral control and escape in mice. EMBO J, 2013. 7. Kasher, P.R., et al., Characterization of samhd1 Morphant Zebrafish Recapitulates Features of the Human Type I Interferonopathy Aicardi-Goutieres Syndrome. J Immunol, 2015. 8. Ulrich, F., et al., Neurovascular development in the embryonic zebrafish hindbrain. Dev Biol, 2011. 357(1): p. 134-51. 9. Rice, G.I., et al., Assessment of interferon-related biomarkers in Aicardi-Goutieres syndrome associated with mutations in TREX1, RNASEH2A, RNASEH2B, RNASEH2C, SAMHD1, and ADAR: a case-control study. Lancet Neurol, 2013. 12(12): p. 1159-69. 10. Chen, S.N., P.F. Zou, and P. Nie, Retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs) in fish: current knowledge and future perspectives. Immunology, 2017. 11. Rossi, A., et al., Genetic compensation induced by deleterious mutations but not gene knockdowns. Nature, 2015. 524(7564): p. 230-3. 12. Arnold, L.H., et al., Phospho-dependent Regulation of SAMHD1 Oligomerisation Couples Catalysis and Restriction. PLoS Pathog, 2015. 11(10): p. e1005194. 13. Schmid, B., et al., Loss of ALS-associated TDP-43 in zebrafish causes muscle degeneration, vascular dysfunction, and reduced motor neuron axon outgrowth. Proc Natl Acad Sci U S A, 2013. 110(13): p. 4986-91. 
Type Of Material Model of mechanisms or symptoms - non-mammalian in vivo 
Year Produced 2016 
Provided To Others? Yes  
Impact None so far 
 
Title iPSC modelling 
Description iPSC modelling None of the available mouse models of AGS1-7 exhibit a neurological phenotype, a primary feature in the majority of patients mutated in these genes, and brain tissue from type I interferonopathy patients is scarce. These factors limit interrogation of the associated neuropathology and tissue specificity. As part of a long-standing collaboration with Alysson Muotri, San Diego, CRISPR/Cas9 was used to mutagenize H9 human embryonic stem cells and produce two TREX1 mutant (V63fs, and E83fs) cell lines, and a TREX1-deficient line generated by inducing pluripotency in fibroblasts from an AGS patient (homozygous V201D)(Figure 8)[20]. These lines were then differentiated into neuronal progenitor cells (NPCs), neurons and astrocytes (Figure 9). TREX1-deficient NPCs, neurons and astrocytes exhibit higher numbers of extranuclear ssDNA puncta per cell than controls, which are enriched for L1s and are reduced by treatment with the nucleoside analog reverse transcriptase inhibitors (RTIs) lamivudine and stavudine, anti-HIV-1 drugs that inhibit both HIV and L1 RT. In contrast, nevirapine, which selectively inhibits HIV-1 but not L1 RT, has a negligible effect on ssDNA content. Co-culture experiments indicate both cell-intrinsic and non-cell autonomous cytotoxicity (Figure 10). Furthermore, development of a human iPSC cerebral organoid culture system demonstrated self-assembled spheres, with internal cytoarchitecture reminiscent of a laminated neocortex and transcriptionally equivalent to a mid-foetal human brain (Figure 11). Initial results indicate that TREX1-deficient corticoids exhibit a reduction in size throughout differentiation, apparently recapitulating the microcephaly seen in many AGS patients. The above data are concordant with the initial results of our clinical trial involving the use of RTIs in AGS, using combined abacavir, zidovudine and lamivudine for 12 months, with changes in IFN signalling as our primary end-point (https://clinicaltrials.gov/ct2/show/NCT02363452), indicating that IFN-signalling might be induced through an RT activity (Figure 12)(see below). More generally, these data suggest that iPSCs represent a useful model system in which to study the biology of certain type I interferonopathies, a point further supported by a recent publication relating to ADAR1 [21]. Beyond the lifespan of this current grant, we will use such iPSC material to phenotype iPSC-driven cell subsets across a catalogue of genotypes, explore underlying disease mechanisms and test potential treatments (e.g. see [22]). For further details, see: Thomas CA., et al. Modelling of TREX1-dependent autoimmune disease using human stem cells highlights L1 accumulation as a source of neuroinflammation. Cell Stem Cell 2017;21:1-13. 20. Thomas, C.A., et al., Modelling of TREX1-dependent autoimmune disease using human stem cells highlights L1 accumulation as a source of neuroinflammation. Cell Stem Cell. 2017. 21: p. 319-331. 21. Chung, H., et al., Human ADAR1 prevents endogenous RNA from triggering translational shutdown. Cell, 2018. 172: p. 811-824. 22. Kothur, K., et al., An open-label trial of Jak 1/2 blockade in progressive IFIH1-associated. Neurology 2018. 90: p. 1-3. 
Type Of Material Model of mechanisms or symptoms - human 
Year Produced 2017 
Provided To Others? Yes  
Impact None so far. 
 
Description Collaboration as part of ERA-NET consortium 
Organisation Albert Ludwig University of Freiburg
Department Institute of Neuropathology
Country Germany 
Sector Academic/University 
PI Contribution Exchange of ideas and materials as part of the consortium
Collaborator Contribution Exchange of ideas and materials as part of the consortium
Impact Contribution to a common manuscript (Meuwissen et al. J Exp Med. 2016 Jun 27;213(7):1163-74)
Start Year 2015
 
Description ERANET 
Organisation Pfizer-University of Granada-Junta de Andalucía Centre for Genomics and Oncological Research
Country Spain 
Sector Academic/University 
PI Contribution Exchange of ideas and material (zebrafish)
Collaborator Contribution Exchange of ideas and material (zebrafish)
Impact None so far
Start Year 2016