Biallelic mutations in SUPV3L1 cause an inherited neurodevelopmental disorder with variable leukodystrophy due to aberrant mitochondrial double stranded RNA processing

doi:10.21203/rs.3.rs-4356120/v1

Download PDF

Research Article

Biallelic mutations in SUPV3L1 cause an inherited neurodevelopmental disorder with variable leukodystrophy due to aberrant mitochondrial double stranded RNA processing

https://doi.org/10.21203/rs.3.rs-4356120/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

We describe eighteen individuals from twelve families with an autosomal recessive neurodevelopmental disorder and variable leukodystrophy harbouring biallelic variants in SUPV3L1. SUPV3L1 encodes the RNA helicase SUV3 (also known as SUPV3L1), with previous studies demonstrating a role for the protein as part of the mitochondrial degradosome. Patient mutations result in an accumulation of mitochondrial double stranded RNAs in human cells. An assessment of supv3l1 knock-out zebrafish confirmed the role of supv3l1 in neurodevelopment, with gross defects identified in mitochondrial biogenesis and microglial function. Zebrafish displayed a significant activation of the type 1 interferon pathway, which was supported by qPCR of blood RNA from four patients with biallelic SUV3 mutations. Altogether, we describe a clinico-radiological spectrum associated with biallelic SUPV3L1 mutations, demonstrating that loss of SUV3 function results in altered mitochondrial biogenesis, increased mitochondrial double stranded RNA, dysplastic microglia and activation of the type 1 interferon innate immune pathway.

Medical Genetics

SUPV3L1

SUV3

dsRNA

interferon

neurodevelopment

As the understanding of inherited white matter disorders (IWMDs) grows, so does the recognition that these diseases may result from disruption of any component of white matter structure or function [1, 2]. This recognition, driven in large part by advances in magnetic resonance imaging (MRI) and genomic medicine, has led to the identification of multiple leukodystrophy (LD) and LD-associated genes. These include not only classical primary white matter disorders, where disruption to the myelin network is the primary pathology underpinning disease, but also an increasing number of secondary LDs where the impact on myelin is consequent upon another, potentially systemic, disease process. Now, in addition to patients presenting with clinico-radiological phenotypes consistent with the “classical leukodystrophies” [3], there are growing numbers with heterogenous neurodevelopmental phenotypes with evidence of abnormal white matter on MRI.

The recent drive in novel gene discovery has identified large numbers of proteins involved in myelin development, maintenance and function, which when mutated cause white matter disorders [1]. Increasingly represented are genes involved in mitochondrial homeostasis, encoded by either the nuclear or mitochondrial genome. In a recent study of over 4000 patients with LD, mutations in mitochondria-associated genes accounted for approximately 5–10% [4]. While many of these mutations were in genes encoding Complexes I-IV, it is important to note that mitochondria are not only essential for energy production, but also play a key role in the regulation of cellular metabolism, proliferation and innate immune signalling, including the antiviral immune response [5–7].

The mitochondrial degradosome is a functionally conserved complex composed of the ATP-dependent RNA and DNA helicase SUV3 (encoded by SUPV3L1) and the PNPase PNPT1 (encoded by PNPT1). The degradosome plays a key role in maintaining mitochondrial integrity through the surveillance and degradation of double stranded (ds) RNA molecules. In addition to an accumulation of R loops in the mitochondrial genome, with consequent disruption of replication and mitochondrial genome integrity, loss of degradosome activity through PNPT1 mutation has been shown to result in leakage of mitochondrial dsRNA into the cytosol and activation of the antiviral immune response [8, 9]. While biallelic PNPT1 variants have been extensively reported [10, 11], much less is known about the clinical and cellular phenotypes associated with pathogenic variants in SUPV3L1.

Recently, van Esveld et al. reported a family with 2 patients presenting with a white matter neurodegenerative disorder where the major findings included global developmental delay, microcephaly, progressive loss of motor function, retinal atrophy, atrophy of the optic chiasm, midbrain, and pons, as well as hypomyelination [12]. Exome sequencing analysis revealed that both siblings were homozygous for a stop mutation in the SUPV3L1 gene (c.2215C > T, p.Gln739Ter).

Here we describe the largest cohort so far reported with biallelic variants in SUPV3L1, comprising eighteen individuals from twelve families. Through in-depth phenotyping, dsRNA complementation assays and in vivo zebrafish modelling, we define a clinical spectrum associated with biallelic SUPV3L1 mutations, and demonste that loss of SUV3 function results in altered mitochondrial biogenesis, increased dsRNA, dysplastic microglia and activation of the type 1 interferon innate immune pathway both in vivo and in patients. This, we define the helicase SUV3 as a critical player in the activation of an antiviral immune response, which when dysfunctional results in a novel white matter disorder.

Patient ascertainment

We ascertained eighteen patients from twelve unrelated families across twelve institutions. Biallelic SUPV3L1 variants were identified locally by next generation sequencing techniques following local procedures. GeneMatcher was used to identify patients with biallelic variants SUPV3L1 patients through international collaboration [13]. Clinical and radiological data were collected retrospectively in Leeds, and evaluated by a paediatric neurologist (LG) and neuroradiologist (DW). The study was approved by the Yorkshire & Humber – Leeds East Research Ethics Committee (REC ref. 18/YH/0070). All patients and/or parents gave appropriate informed consent to be part of this study, and to be included in any publication, from their local institution in accordance with the principles outlined in the Declaration of Helsinki.

Molecular and cellular assays

Plasmid preparation and validation

Gene blocks of attB flanked SUPV3L1 containing each missense variant tested were obtained from Integrated DNA Technologies (Coralville, Iowa). Gene blocks were cloned into pDONR221 donor vector using the Gateway™ BP Clonase™ II Enzyme mix (ThermoFisher Scientific) and further cloned into the required destination vector using the Gateway™ LR Clonase™ II Enzyme mix (ThermoFisher Scientific) according to the manufacturers protocol. Sanger sequencing of the vectors confirmed the presence of each variant.

SUPV3L1 complementation assay

2.5x10⁵ U2OS cells were seeded on to coverslips and 24 hours later transfected with a custom siRNA targeting the 5’UTR of SUPV3L1 (Horizon technologies) using lipofectamine RNAiMAX (Thermo Fisher Scientific). 24 hours later, a second transfection was performed to introduce wild-type or variant-containing SUPV3L1 using TransIT-X2 (Mirrus Bio). After 48 hours, U2OS cells were fixed with 4% para-formaldehyde. Following fixation, cells were permeabilised with 0.02% Triton X-100 for 5 minutes before washing with 1x phosphate-buffered saline (PBS) for a further 5 minutes. Wells were then blocked with 5% milk:PBS solution for 60 minutes at room temperature. Primary anti-dsDNA antibody (K1; Cell Signalling Technology) was prepared at a 1:500 dilution in a 3% milk:PBS solution and added to the wells for 60 minutes at room temperature. After three washes with PBS, cells were incubated with an Alexa Fluor 488 anti-mouse IgG secondary antibody (Thermo Fisher Scientific, 1:500) and Hoescht (1:1,000). Secondary antibodies were prepared with 1% milk in PBS and incubated at room temperature, out of light, for 60 minutes. Once completed the cells were washed three times with PBS before mounting. Images were captured using a Nikon A1R confocal microscope using NIS Elements software. ImageJ software was used to count dsRNA, using the SpotCounter plugin.

Zebrafish Modelling

Zebrafish husbandry and ethics

All zebrafish were raised in the Bateson Centre at the University of Sheffield in UK Home Office approved aquaria and maintained following standard protocols [14]. Tanks were maintained at 28°C with a continuous re-circulating water supply and a daily light/dark cycle of 14/10 hours. All procedures were performed under an animal Project Licence to standards set by the UK Home Office. We used the Tg(mpeg1:mCherryCAAX)sh378 labelling the membrane of macrophages and microglia [15] and the Tg(AnnexinV:mVenus)sh632 [16] to label dying cells.

Generation of supv3l1 zebrafish crispants

Synthetic SygRNA® consisting of gene-specific CRISPR RNAs (crRNA) (Sigma) and transactivating RNAs (tracrRNA) (Merck) in combination with CAS9 nuclease protein (Merck) was used for gene editing. TracrRNA and crRNA were resuspended to a concentration of 100µM in nuclease free water containing 10mM Tris-HCl ph8. SygRNA® complexes were assembled on ice immediately before injection using a 1:1:1 ratio of crRNA:tracrRNA:Cas9 protein. We used the CHOPCHOP website [17] to design the following crRNA sequence specific to the zebrafish supv3l1 gene (ENSDARG00000077728) targeting exon1, where the PAM site is indicated in brackets: GAAGACGCGGAGGGATCAGT(CGG). A 0.5nl drop of SygRNA®:Cas9 protein complex was injected into one-cell stage embryos. The resulting supv3l1 crispants were used for the experiments, alongside a scrambled crRNA sequence for control group [18].

Construction of Tg(mpeg1.1:mts-mNeonGreen)sh631

The Tol2kit multisite Gateway method was used with p5E-mpeg1.1 [19], pME-mts-mNeonGreen, p3E-polyA and pDestTol2pA2 [20] to create p(mpeg1.1:mts-mNeonGreen) transgene plasmid. The pME-mts-mNeonGreen middle-entry plasmid incorporated mts-mNeonGreen PCR amplified from a custom gBlock from IDT. The mts-mNeonGreen had the mitochondrial targeting sequence from zmLOC100282174 [21] as an N-terminal fusion to mNeonGreen [22] via a (GGGGS)3 flexible linker and was codon optimised using CodonZ3 [23]. DNA sequencing confirmed the full transgene plasmid sequence (Core Genomics Facility University of Sheffield). The transgene plasmid was co-injected with tol2 transposase mRNA into Zebrafish embryos, the fish raised to adulthood and 3-day post fertilisation (dpf) larvae from outcrosses were screened for green fluorescent macrophages using a Zeiss Axio ZoomV16 microscope. Such F1 larvae were raised to adulthood and screened by outcrossing. This identified the line sh631 that transmitted the transgene to 50% of progeny, indicating a single transgenic insertion.

Quantitative PCR

RNA was extracted from dissected heads of 5-day post fertilisation (dpf) zebrafish larvae using the Trizol/chloroform method. cDNA was synthetised using the SuperScript II kit (Invitrogen) with 2mg of RNA following manufacturer instructions and diluted in 1:20 for qPCR. qPCR primers were tested for efficiency (85%-105%) using a cDNA serial dilution. The qPCR reaction was run in a CFX96 Bio-Rad machine using validated primers as previously described [18]. Expression was normalised to rpl13 as reference gene and relative to scrambled control set at 1.

Zebrafish confocal imaging and analysis

To assess microglial morphology, 5dpf scrambled and supv3l1 crispant larvae injected in double reporter background Tg(mpeg1:mCherryCAAX)sh378 and Tg(AnnexinV:mVenus)sh632 were anaesthetised and embedded in low melting point agarose containing tricaine (0.168 mg/ml; Sigma-Aldrich) and imaged using a 40x objective on a UltraVIEW VoX spinning disk confocal microscope (PerkinElmer Life and Analytical Sciences). Z-stacks were 100µm thick using 0.5µm slices. For microglial morphology analysis, sub-stacks of 50µm were used to create a maximum projection and contours of each microglia (avoiding pigment cells) were drawn using a pen tablet (Intuos from Wacom). Using Fiji, the circularity index (0–1) of each microglia was automatically recorded to assess the circularity of each cell with the value of 0 being not circular and 1 as being perfectly circular.

To count the number of apoptotic cells, the same animals were imaged using the 10x lens on a UltraVIEW VoX spinning disk confocal microscope (PerkinElmer Life and Analytical Sciences) using the brightfield alongside GFP and DsRed channels. Z- Stacks of 100µm with 2µm per slice were acquired and the brightfield image was used to draw a region of interest around the optic tectum as previously described [18].

TUNEL staining and quantification

Fish larvae at 5dpf were fixed in 4% PFA and the TUNEL assay was performed according to standard protocol using the ApopTag Kit (Millipore) and as previously described [18]. Samples were imaged on the inverted W1 Nikon Spinning Disk confocal microscope using the brightfield and GFP channel by acquiring stacks of approximately 100µm with 2µm per slice. Using Fiji, the optic tectum region was outlined with the free hand drawing tool using the brightfield image and counting of number of apoptotic cells was performed using the automatic thresholding and particle count. All imaging analysis were blinded until after counting was performed.

Statistical analysis

All statistical analysis were performed in GraphPad Prism where data was entered using a column (2 samples, 1 variable only) spreadsheet. Sample distribution was assessed using frequency of distribution analysis and two-tailed Mann-Whitney U test was used for statistical analysis. All experiments were repeated using different batches of larvae born on different dates, with the number of biological replicates and n (experimental unit) number stated for each experiment in figure legends. p values are indicated and a star system is used instead for graph with multiple comparisons: *=p < 0.05, **=p < 0.01, ***=p < 0.001, ****=p < 0.0001. Following the recommendation of the American Statistical Association we do not associate a specific p value with significance [24].

We identified eighteen patients from twelve families in whom biallelic SUPV3L1 variants segregate with their disease phenotype. Fourteen different SUPV3L1 variants were identified that were rare (minor allele frequency < 0.001 in gnomAD v4.0) and considered to be likely pathogenic (CADD v1.7 score > 15, predicted damaging (PolyPhen2), deleterious (SIFT) or to impact splicing (SpliceAI > 0.4) (Table 1). Variants identified included missense, frameshift, nonsense and splice site alterations. The evolutionary conservation of all amino acid residues in which a missense variant was identified was investigated and can be found in Supplementary Fig. 1.

Table 1

Summary of variants identified in *SUPV3L1*.
DNA variant	Protein variant	dbSNP: rs number	gnomAD v4.0 Frequency	PolyPhen2	SIFT	CADD (v1.7)	SpliceAI	Family
c.196C > T	p.Pro66Leu	rs752705209	1.8x10^− 5 (14/781,038)	Prob. Dam. 1.00	DEL 0.00	27.3	0.00	9*
c.329G > A	p.Gly110Asp	Not Present	Not Present	Benign 0.082	DEL 0.00	22.5	0.00	5
c.371T > C	p.Phe124Ser	Not Present	Not Present	Poss. Dam. 0.566	DEL 0.02	27.1	0.00	11
c.458-2A > G	p.(splice)	Not Present	1.4x10^− 5 21/1,440,516	N/A	N/A	32.0	0.99 (Acc.L)	2*
c.854-3A > G	p.(splice)	Not Present	Not Present	N/A	N/A	13.03	0.50 (Acc.L)	4
c.1016A > T	p.Glu339Val	Not Present	Not Present	Benign 0.085	DEL 0.01	26.8	0.46 (Don.L)	10
c.1070C > A	p.Ala357Glu	Not Present	Not Present	Prob. Dam. 0.959	DEL 0.00	26.4	0.00	2*
c.1093C > T	p.Arg365Trp	rs527626577	8.7x10^− 6 (14/1,613,968)	Poss. Dam. 0.457	DEL 0.00	33.0	0.00	1
c.1177G > A	p.Ala393Thr	rs1025320000	3.7x10^− 6 (6/1,613,896)	Prob. Dam. 0.936	DEL 0.00	29.5	0.01 (Don.G)	8
c.1240C > T	p.Pro414Ser	Not Present	1.6x10^− 6 (1/626,536)	Prob. Dam 0.968	DEL 0.00	26.2	0.00	9*
c.1321dup	p.Tyr441Leufs*10	Not Present	Not Present	N/A	N/A		0.00	3*
c.1469A > G	p.Asp490Gly	Not Present	Not Present	Prob. Dam. 0.995	DEL. 0.00	25.1	0.13 (Don.L)	12
c.1924A > C	p.Ser642Arg	rs750670256	5.0x10^− 6 (8/1,612,906)	Prob. Dam. 0.999	DEL 0.00	32.0	0.02 (Don.L)	3*
c.2215C > T	p.Gln739*	Not Present	Not Present	N/A	N/A	39.0	0.00	6, 7
Variant nomenclature based on transcript ENST00000359655.4. * Signifies a compound heterozygous variant.
Abbreviations. PolyPhen2: Prob.Dam, Probably Damaging; Poss. Dam., Possibly Damaging. SIFT: DEL, Deleterious; TOL, Tolerated. SpliceAI; Acc.L, Acceptor Loss; Don.G, Donor Gain; Don.L, Donor Loss.

Biallelic SUPV3L1 mutation causes a variable neurodevelopmental phenotype

Detailed clinical characteristics are provided in Supplementary Table 1. All patients presented in childhood, with onset under 12 months in all except one case where data was available (unknown age of onset P15 and P18), the majority with developmental delay (13/18). Three patients presented prior to any noted delay, one in the antenatal period with abnormal antenatal scans, one at birth with neonatal thrombocytopenia and hypoglycaemia, and one postnatally with feeding issues and weight loss. No other patients had haematological or biochemical abnormalities reported. Other features at presentation included tonal abnormalities (hypotonia 7/18), nystagmus (3/18), acquired microcephaly (1/18) and ambiguous genitalia (1/18). P12 presented aged 4 years with acute encephalopathy, ataxia and dystonia following a viral illness but, prior to this, development was normal except for mild speech articulation difficulties.

Over time all surviving patients demonstrated some degree of developmental delay, being described as severe in 5 patients, and 14/17 had intellectual disability. At last review 16/18 had spasticity, 15/18 microcephaly, 8/18 areas of skin hypopigmentation, 5/18 seizures, 2/18 movement disorders and 2/18 retinal dystrophy or unspecified retinal changes. 11 patients achieved walking, although only 4 independently. The 7 others continue to require significant support aged between 4–20 years. At the time of writing 1 patient (P11) was deceased –antenatal scans demonstrated lissencephaly in this individual, who experienced respiratory distress syndrome, neonatal seizures and early spasticity.

Within the cohort, most patients did not purse a degenerative course – 4/5 with severe developmental delay remained stable, one patient deteriorating following onset of seizures. 13/18 with mild-moderate delay are reported as making on-going developmental progress aged 3–32 years. P9 was making good developmental progress until an episode of regression aged 12 years, shortly after which she was lost to follow up. A total of six patients experienced episodes of regression during their disease course.

SUPV3L1 -associated disorder has a variable radiological phenotype including abnormal white matter, temporal pole cysts and calcification

Imaging was available for 16 of the 18 patients (14 MRI, 1 foetal MRI, 4 CT head). White matter was abnormal in 11/15 patients with MRI imaging (Fig. 1), ranging from mild delay through to confluent, bilateral T2 hyperintensity. In the one patient for whom follow up imaging was available, there was an improvement in T2 signal over time (Fig. 2). Subcortical cysts in the temporal pole were seen in 5 patients. This feature could not be assessed in P5-7 due to incomplete imaging, but was absent in P9, P13, P14, P16, P18. Additional features included atrophy (cerebellar 10/15, cerebral 5/15) and abnormality of the corpus callosum (5/15). In 3 of 4 patients where CT imaging was available, small foci of calcification were seen within the basal ganglia.

P11 presented antenatally, with an MRI at 28 weeks of gestation demonstrating lissencephaly, ventriculomegaly, partial agenesis of the corpus callosum, agenesis of the cerebellar vermis, pons hypoplasia, posterior fossa cyst and multiple subcortical white matter cystic lesions. A further scan at age 3 days (not available for review) was said to have confirmed the antenatal findings with no additional features recorded.

Three patients (P9, P12, P18) exhibited no evidence of abnormal white matter, though mild cerebral atrophy was seen in P9, and cerebral atrophy plus a thin, dysplastic corpus callosum were observed in P18. P12 had normal MRI imaging, both at presentation and following a second episode of regression at 6 years. Additional brain MRI images from all available cases can be found in Supplementary Fig. 2.

SUPV3L1 variants do not cluster or map into any functional domain

SUV3 is a mitochondrial double-stranded RNA (mtdsRNA) helicase. To determine if the variants identified clustered into any particular region or domain, we mapped our observed substitutions onto SUV3 in 2D and 3D structures (Fig. 2A, B). Only 2/10 missense variants mapped onto any known functional domain, including the mitochondrial localisation signal (amino acids (aa):1–21), the nucleotide binding domain (aa: 207–214) or the helicase domain (aa: 378–475)). Both of these variants, p.Alal393Thr and p.Pro414Ser, are located in the helicase domain and may therefore directly impact helicase activity. The remaining 8 variants are located outside of any functional domains, suggesting a potential impact on the overall structure of SUV3. For example, analysis of the 3D crystal structure of SUV3 shows Arg-365 forms an interaction with Glu-482, with the homozygous p.Arg365Trp substitution identified in P1 likely to result in the loss of this charged interaction and an impact on overall protein structure. These data suggest that pathogenicity is not due to impairment of a specific function of SUV3 (in particular, DNA/RNA binding or helicase activity).

SUPV3L1 mutations fail to rescue dsRNA levels following siRNA knockdown of SUPV3L1

As mapping of the variants provided little insight into the potential impact of the missense variants observed in our study, we sought to confirm pathogenicity with a cell-based functional assay using dsRNA levels as a functional readout. We first created a custom siRNA, mapping to the 3’-UTR of endogenous SUPV3L1, and assessed the presence of mtdsRNA following transfection of the siRNA in U2OS cells. 48hrs after transfection of the siRNA, cells were fixed and immunofluorescence performed using an anti-dsRNA antibody (K1). This revealed a significant increase in dsRNA compared to the non-targeting control (Supplementary Fig. 3). To determine if this increase in dsRNA could be rescued by re-introduction of SUPV3L1, we co-transfected wild-type SUPV3L1 with the custom siRNA, and assessed levels of dsRNA. In this way, we observed a significant decrease in dsRNA in cells co-transfected with siRNA and SUPV3L1^WT compared to siRNA alone, consistent with a rescue of the phenotype. We subsequently introduced all missense variants into SUPV3L1^WT and assessed their ability to rescue the increase in dsRNA. For all missense variants tested, in contrast to transfection with SUPV3L1^WT, we observed no significant rescue of increased dsRNA levels, indicating that all variants have an impact on protein function and are likely pathogenic (Fig. 2C).

SUPV3L1 mutations result in variable interferon signature

mtdsRNA are known immunogenic substrates and have been shown to activate the antiviral immune response when accumulated in patients with pathogenic PNPT1 variants. A previous study, however, did not observe a similar response in SUV3 depletion, presumed due to restriction of mtdsRNA to the mitochondria [9]. We therefore wanted to investigate the impact of impaired SUV3 in humans with biallelic SUPV3L1 mutations. The interferon signature was assessed in four patients (P4, P6, P7, P12) for whom blood samples were available (Supplementary Fig. 4). Although differing testing methods were used, a marked up-regulation of interferon signalling genes (ISGs) suggestive of a type 1 interferon response was observed in three out of four patients on single testing. P6 was tested on two occasions, showing a positive response initially, though less marked than her sibling, followed by a subsequent negative response. The role of aberrant SUV3 in innate immune activation is therefore variable and may change over time.

Loss of supv3l1 in a zebrafish model results in immune dysfunction and activation of antiviral immune pathway

To investigate the effect of SUPV3L1 loss of function on brain development, we generated a supv3l1 crispant using CRISPR/Cas9, and used the previously published survival and mitochondrial phenotype [25] to assess efficiency of the CRISPR/Cas9 complex targeting the supv3l1 gene. Injected embryos with supv3l1 crRNA showed reduced survival compared to scrambled-injected controls, necessitating culling by 8dpf due to signs of necrosis in the liver, lack of swimming and deflated swim bladder (Fig. 3A,B). Mitochondrial dysfunction was confirmed using a novel reporter line Tg(mpeg1:mls-neon)sh631 labelling mitochondria in the macrophage/microglia lineage, with significantly more microglia displaying mitochondria fusion and fission in supv3l1 crispants (Fig. 3C-E).

To analyse the effect of dysfunctional mitochondria in microglia and the wider brain, 5dpf zebrafish were live imaged at high resolution to measure microglial morphology and cell death in the optic tectum region, using Tg(mpeg1:mCherryCAAX)sh378 and Tg(AnnexinV:mVenus)sh632 reporter lines respectively. Microglia in brains of 5dpf supv3l1 crispants displayed an activated phenotype, with rounder cell body measured using the FiJi circularity index (Fig. 4A-B,D), and showed an unusual membrane ruffled phenotype (Fig. 4B- high magnification panel). Using the Tg(AnnexinV:mVenus)sh632 reporter line to label apoptotic cells, we counted a higher number of apoptotic cells in supv3l1 crispant brains, which was confirmed using TUNEL staining (Fig. 4E,F).

To assess the neuroinflammatory state of the brain, we extracted RNA from 5dpf zebrafish heads of supv3l1 and control fish, and performed qPCR for a panel of inflammatory genes (Fig. 4G,H). We identified a more than 400-fold upregulation of expression of genes specific to the antiviral immune pathway, including the human orthologue of IFN-1 called ifnF1 [26] and interferon-induced genes such as isg15 and mxa (Fig. 4H). Activation of the antiviral immune response was confirmed using a reporter line for the mxa gene Tg(cryaa:Dsred;mxa:mCherryF)ump7tg, highlighting a systemic activation of the antiviral immune pathway (Fig. 4I).

Through an international collaboration, this study demonstrates the impact of biallelic SUPV3L1 variants on neurodevelopment and adds to the list of known LD-associated genes. From these clinical findings we see that SUPV3L1-related disease encompasses a wide clinical spectrum from neonatal haematological disturbance through to infant-onset motor disorder with as-of-yet preserved cognition and ambulation, and acute encephalopathy in an otherwise normally developed child. 16/18 individuals in our cohort demonstrated a significant neurodevelopmental disorder, manifesting as a moderate to severe motor delay with intellectual impairment. Other common clinical features include microcephaly and spasticity. The disease course was static in the more severely affected cases, while slow developmental progress, up to and including our oldest patient aged 32 years, was seen in the less severely affected. A third of the cohort (6/18) experienced one or more episodes of developmental regression, often without subsequent full return of function. Episodes of regression, often on a background of preceding developmental delay, are a recognised feature of disorders of mitochondrial function [27].

Eight patients (44%) had areas of skin hypopigmentation, at least two being consistent with vitiligo and progressive over time. Hypopigmentation occurs secondary to defects in melanin production, with causes including post-infectious/inflammatory, post-traumatic and autoimmune responses. Implicated in this pathway is RNA pattern-recognition receptor MDA5 which, when activated, induces a type 1 interferon response [28].

Eleven of fourteen patients for whom cranial MRI was available demonstrated abnormal white matter, though this was variable, ranging from mild, patchy T2 hyperintensity through to confluent bilateral abnormalities. In the remaining patients, myelination was normal aged 2, 5 and 32 years. Just under a third of patients (4/14) had anterior temporal subcortical cysts as reported in both congenital infections (cytomegalovirus and rubella) but also other leukodystrophies and RMND1-related mitochondrial leukoencephalopathy [29, 30]. Cranial CT demonstrated small foci of calcification in the basal ganglia in three of four patients. Intracranial calcification is seen in a number of other leukodystrophies, most notably Aicardi-Goutières syndrome (AGS). This observation, along with neonatal thrombocytopenia, meant that AGS was the initial differential in P6, but extended genetic screening of AGS-associated genes was negative.

Dawidziuk et al. reported the first probable cases of SUPV3L1 mutations causing human disease in a pair of siblings with microcephaly, one of whom had white matter changes [31]. These individuals were identified from a cohort of patients with congenital microcephaly who underwent exome sequencing, and in whom, no other candidate variants were identified. Van Esveld et al. reported a further two cases where SUPV3L1 variants were considered responsible for neurodegeneration in a sibling pair, with a family history of two other affected siblings and an affected cousin from whom samples were not available for testing [12]. The siblings were homozygous for the same c.2215C > T, p.Gln739* variant identified in two families (F6 and F7) in our cohort, both of whom were ascertained through a larger screen of patients with intellectual disability from the Middle East [32]. Given that all cases with the p.Gln739* variant are from the Middle East, this indicates it to be a likely founder variant in this population, inherited from a common ancestor. The phenotype described by van Esveld and colleagues was of a progressive spastic paraparesis with growth restriction and hypopigmentation [12]. One of the siblings also demonstrated subtle white matter changes in the form of focal areas of T2 hyperintensity as well as some delayed myelination.

Helicases are highly conserved motor proteins with a role in almost all processes involving nucleic acid replication, repair and degradation [33]. Maintaining effective turnover of unwanted nucleic acid material is critical for mitochondrial health and function. First discovered in yeast in 1992, SUPV3L1 encodes an ATP-dependent RNA/DNA helicase that, when bound to polyribonucleotide nucleotidyltransferase 1 (PNPT1 or PNPase), forms an essential part of the human mitochondrial degradosome [34–36]. If one of the two components of the heteropentamer is disrupted, the activity of the degradosome is abolished. Of note, biallelic variants in PNPT1 are also associated with abnormal myelination in patients, neurodevelopmental delay and mitochondrial dysfunction [9, 37]. Our study provides additional evidence that pathogenic variants in genes encoding components of the mitochondrial degradosome can result in neurodevelopmental disturbances, characterised by defects in mtdsRNA processing and function. Similar to variants in PNPT1, mtdsRNA accumulation due to loss of SUV3 leads to the activation of the antiviral immune response, as observed in vivo in our zebrafish supv3l1 model and consistent with increased expression of interferon stimulated gene transcripts in blood of 4 patients from our cohort, 1 of whom had skin hypopigmentation.

dsRNA is a known pathogen-associated molecular pattern that can trigger the assembly of MDA5 ATP-sensitive filaments, that in turn activate the mitochondrial antiviral-signalling protein via CARD domain-mediated interactions [38]. Additionally, macrophage cell models have shown that Pnpt1 deficiency-mediated MAVS-activation results in NLRP3 inflammasome activation and mitochondrial reprograming, which in turn results in impaired mitochondrial function and decreased levels of cellular ATP [39, 40]. Genetic defects that cause impaired nucleotide metabolism have been reported and are well established aetiologies of other IWMDs such as mutations in ADAR1, TREX1, SAMHD1, IFIH1 (which encodes MDA5), and RNASEH2B [41–45], resulting in a common pathway of genes involved in nucleotide metabolism that can trigger leukodystrophy-causing inflammation. The high incidence of skin hypopigmentation, also seen in two of the previously reported cases [12], referred to as vitiligo, could have an autoimmune origin linked to the enhanced antiviral immune response observed in our zebrafish model and in the individuals tested in this study. Therefore, mtdsRNA-mediated activation of the antiviral immune response emerges as a possible pathophysiological mechanism in SUPV3L1-associated leukodystrophy.

P11, P12 and P18 represent outliers in our cohort, in terms of clinical and radiological presentation. P11 presented antenatally with significant abnormalities on MRI imaging at 28 weeks of gestation, he died age 25 days due to respiratory distress and neonatal seizures. In addition to the homozygous p.Ala393Thr in SUPV3L1 confirmed to impact protein in our assays, exome sequencing identified a heterozygous TUBB3 (NM_006086.4:c.1243A > G, (p.Met415Val)) variant inherited from a healthy father, and therefore presumed non-pathogenic. Poirier et al. reported 12 patients with mutations in TUBB3 demonstrating cortical malformation, including polymicrogyria, gyral disorganization and simplification with or without pontocerebellar defects [46]. Importantly, within this study were a mother and daughter in whom the heterozygous mother demonstrated only mild cognitive impairment, but had frontoparietal predominant gyral disorganisation and cerebellar vermis dysplasia indicating that even monoallelic TUBB3 mutations can impact neuronal development. We cannot rule out that the TUBB3 variant may be contributing to the clinico-radiological phenotype in P11, accounting for the increased severity of the disease, abnormalities on MRI and early death.

Both P12 and P18 presented with regression following intercurrent illness aged 4 and 22 years respectively, on a background of mild developmental impairment. Neither had any other candidate variants identified on local sequencing. The SUPV3L1 variants were predicted probably damaging, and were not able to rescue the mtdsRNA phenotype in our complementation assay compared to SUPV3L1^WT. Although a milder phenotype, both presentations would be in keeping with an underlying mitochondrial disorder.

Mapping of the variants on the SUV3 protein revealed no clustering, with variants observed across the full length of the protein. The p.Gln739* nonsense variant is located in the final exon of SUPV3L1, and is therefore expected to escape nonsense mediated decay and form a truncated protein which may retain some function. Van Esveld investigated the same p.Gln39Ter variant in patient cells by western blot and showed the truncated protein was present, albeit at reduced levels [12]. This is consistent with previous studies which found complete knockout of SUPV3L1 to result in embryonic lethality [34], and may suggest that all variants identified in SUPV3L1 in humans are hypomorphic. A frameshift variant was also identified, which would be predicted to lead to loss of protein. However, this variant was identified in trans with the missense variant p.Ser642Arg, which may retain some function.

The development of animal models to study neurodevelopmental disorders, including leukodystrophies, in vivo has been invaluable for testing therapies [47–50]. However, many rodent models fail to recapitulate key aspects of the human disease or are embryonic lethal [51]. Supv3l1 knockout in mice, for example, results in embryonic lethality, with further investigation identifying a role for SUV3 in the cytoplasm to suppress mitotic homologous recombination [34]. A gene trap supv3l1 zebrafish mutant showed mitochondrial dysfunction, impaired liver development and reduced survival, thereby confirming a conserved role for SUV3 in vertebrate mitochondrial homeostasis [25]. Altogether, this indicates SUPV3L1 to be an essential gene with a fundamental role in degrading mtdsRNA. The crispant approach used to generate the supv3l1 zebrafish model in our study is aligned with the gene trap model as we also observed reduced survival, with most injected crispants requiring humane culling by 8dpf. This zebrafish model is therefore the first pre-clinical model of SUPV3L1-deficient white matter disorder. Further studies will include the generation of patient-specific zebrafish mutants to dissect the mechanisms leading to demyelination.

In summary, we have identified eighteen patients with a neurodevelopmental disorder in whom biallelic SUPV3L1 variants segregate with disease. We observed abnormal white matter and subcortical temporal lobe cysts to be common features, as well as skin hypopigmentation. Variant SUPV3L1 failed to rescue mtdsRNA levels in a complementation assay, and modelling of supv3l1 loss in fish demonstrated significant effects on mitochondrial biogenesis and microglial function with systemic activation of the antiviral immune pathway, which was supported in four patients in our cohort with available blood samples. The role of SUV3 in the antiviral immune response could potentially be used as a therapeutic target in the future, particularly given that immunomodulators are now being trialled for other similar neurodevelopmental disorders.

Acknowledgements

The authors would like to thank the family and the patients for participating in this study. The study was funded by a Sir Jules Thorn Award for Biomedical Research (JTA/09 to E.G.S and C.A.J.), a University of Leeds PhD studentship (to M.E) and the UK leukodystrophy research network LEUKOLABS.UK established with funds from the White Rose University consortium (to N.H. and J.A.P). J.A.P is supported by a UKRI Future Leaders Fellowship (MR/T02044X/1). S.A.R is supported by a Medical Research Council grant to SHIELD consortium MR/NO2995X/1. S.A.R and D.H.S are supported by an MRC Programme Grant (MR/W028506/1). Y.J.C. acknowledges the European Research Council (786142 E-T1IFNs), a UK Medical Research Council Human Genetics Unit core grant (MC_UU_00035/11), a state subsidy from the Agence Nationale de la Recherche (France) under the ‘Investissements d’avenir’ program bearing the reference ANR-10-IAHU-01. The research is supported by the National Institute for Health Research (NIHR) infrastructure at Leeds. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.

Author Contributions

Project administration: J.A.P., N.H., C.A.J.; Conceptualisation: J.A.P., N.H., S.A.R., R.M., Y.J.C.; Funding Acquisition: J.A.P., N.H., S.A.R., D.H.D., H.H., C.A.J., E.G.S.; Resources: N.H., S.A.R., J.A.P., R.M., H.H.; Patient care: L.G., A.G.L.D., K.O., T.P., K.S., A.A., S.A., A.A.F., D.C., K.A., S.A-Z., A.A.F., E.G.S., J.L., Y.J.C., I.C., K.S.F., M.S., D.W., E.W., P.P., M.A.T., L.B.; Patient recruitment, genetic analysis and data collection: M.E., E.L.H., R.M., S.E., A.M.S.R., E.C.L., D.H.D., K.T., A.A-M., D.C., S.E., A.S., C.R-B., G.I.R., C.P., C.B., M.T., I.S., M.Z., S.S., R.A., H.G., G.S., T.S., M.S.Z., E.A., S.L.Z.; Methodology: N.H., S.E., J.A.P., G.W., S.A.H.; Supervision: J.A.P., N.H., S.A.R., C.A.J., E.G.S., Y.J.C.; Writing Original draft: L.G., N.H., J.A.P., Y.J.C.. All authors read an approved the final manuscript.

Data Availability

The datasets generated and analysed during the current study are available from the corresponding authors on reasonable request.

Conflict of interest

The authors declare that they do not identify any competing interests.

Wolf, N.I., C. Ffrench-Constant, and M.S. van der Knaap, Hypomyelinating leukodystrophies - unravelling myelin biology. Nat Rev Neurol, 2021. 17(2): p. 88-103.
van der Knaap, M.S. and M. Bugiani, Leukodystrophies: a proposed classification system based on pathological changes and pathogenetic mechanisms. Acta Neuropathol, 2017. 134(3): p. 351-382.
Aicardi, J., The inherited leukodystrophies: a clinical overview. J Inherit Metab Dis, 1993. 16(4): p. 733-43.
Roosendaal, S.D., et al., Imaging Patterns Characterizing Mitochondrial Leukodystrophies. AJNR Am J Neuroradiol, 2021. 42(7): p. 1334-1340.
Mohanty, A., R. Tiwari-Pandey, and N.R. Pandey, Mitochondria: the indispensable players in innate immunity and guardians of the inflammatory response. J Cell Commun Signal, 2019. 13(3): p. 303-318.
Vazquez, C. and S.M. Horner, MAVS Coordination of Antiviral Innate Immunity. J Virol, 2015. 89(14): p. 6974-7.
Ren, Z., et al., Regulation of MAVS Expression and Signaling Function in the Antiviral Innate Immune Response. Front Immunol, 2020. 11: p. 1030.
Silva, S., L.P. Camino, and A. Aguilera, Human mitochondrial degradosome prevents harmful mitochondrial R loops and mitochondrial genome instability. Proc Natl Acad Sci U S A, 2018. 115(43): p. 11024-11029.
Dhir, A., et al., Mitochondrial double-stranded RNA triggers antiviral signalling in humans. Nature, 2018. 560(7717): p. 238-242.
Vedrenne, V., et al., Mutation in PNPT1, which encodes a polyribonucleotide nucleotidyltransferase, impairs RNA import into mitochondria and causes respiratory-chain deficiency. Am J Hum Genet, 2012. 91(5): p. 912-8.
Rius, R., et al., Clinical Spectrum and Functional Consequences Associated with Bi-Allelic Pathogenic PNPT1 Variants. J Clin Med, 2019. 8(11).
van Esveld, S.L., et al., Mitochondrial RNA processing defect caused by a SUPV3L1 mutation in two siblings with a novel neurodegenerative syndrome. J Inherit Metab Dis, 2022. 45(2): p. 292-307.
Sobreira, N., et al., GeneMatcher: a matching tool for connecting investigators with an interest in the same gene. Hum Mutat, 2015. 36(10): p. 928-30.
NÜSSLEIN-VOLHARD, C. and R. DAHM, eds. Zebrafish: A Practical Approach. 2003, Oxford University Press. 322.
Bojarczuk, A., et al., Cryptococcus neoformans Intracellular Proliferation and Capsule Size Determines Early Macrophage Control of Infection. Sci Rep, 2016. 6: p. 21489.
Kugler, E., et al., The effect of absent blood flow on the zebrafish cerebral and trunk vasculature. Vasc Biol, 2021. 3(1): p. 1-16.
Labun, K., et al., CHOPCHOP v3: expanding the CRISPR web toolbox beyond genome editing. Nucleic Acids Res, 2019. 47(W1): p. W171-W174.
Hamilton, N., et al., The failure of microglia to digest developmental apoptotic cells contributes to the pathology of RNASET2-deficient leukoencephalopathy. Glia, 2020. 68(7): p. 1531-1545.
Ellett, F., et al., mpeg1 promoter transgenes direct macrophage-lineage expression in zebrafish. Blood, 2011. 117(4): p. e49-56.
Kwan, K.M., et al., The Tol2kit: a multisite gateway-based construction kit for Tol2 transposon transgenesis constructs. Dev Dyn, 2007. 236(11): p. 3088-99.
Chin, R.M., et al., Optimized Mitochondrial Targeting of Proteins Encoded by Modified mRNAs Rescues Cells Harboring Mutations in mtATP6. Cell Rep, 2018. 22(11): p. 2818-2826.
Shaner, N.C., et al., A bright monomeric green fluorescent protein derived from Branchiostoma lanceolatum. Nat Methods, 2013. 10(5): p. 407-9.
Horstick, E.J., et al., Increased functional protein expression using nucleotide sequence features enriched in highly expressed genes in zebrafish. Nucleic Acids Res, 2015. 43(7): p. e48.
Wasserstein, R.L., A.L. Schirm, and N.A. Lazar, Moving to a World Beyond “p < 0.05”. The American Statistician, 2019. 73(sup1): p. 1-19.
Ni, T.T., et al., Conditional control of gene function by an invertible gene trap in zebrafish. Proc Natl Acad Sci U S A, 2012. 109(38): p. 15389-94.
Boudinot, P., et al., The Peculiar Characteristics of Fish Type I Interferons. Viruses, 2016. 8(11).
Muraresku, C.C., E.M. McCormick, and M.J. Falk, Mitochondrial Disease: Advances in clinical diagnosis, management, therapeutic development, and preventative strategies. Curr Genet Med Rep, 2018. 6(2): p. 62-72.
Dias Junior, A.G., N.G. Sampaio, and J. Rehwinkel, A Balancing Act: MDA5 in Antiviral Immunity and Autoinflammation. Trends Microbiol, 2019. 27(1): p. 75-85.
Ulrick, N., et al., RMND1-Related Leukoencephalopathy With Temporal Lobe Cysts and Hearing Loss-Another Mendelian Mimicker of Congenital Cytomegalovirus Infection. Pediatr Neurol, 2017. 66: p. 59-62.
La Piana, R., et al., Neuroradiologic patterns and novel imaging findings in Aicardi-Goutieres syndrome. Neurology, 2016. 86(1): p. 28-35.
Dawidziuk, M., et al., Exome Sequencing Reveals Novel Variants and Expands the Genetic Landscape for Congenital Microcephaly. Genes (Basel), 2021. 12(12).
Al-Kasbi, G., et al., The diagnostic yield, candidate genes, and pitfalls for a genetic study of intellectual disability in 118 middle eastern families. Sci Rep, 2022. 12(1): p. 18862.
Raney, K.D., A.K. Byrd, and S. Aarattuthodiyil, Structure and Mechanisms of SF1 DNA Helicases. Adv Exp Med Biol, 2013. 767: p. 17-46.
Pereira, M., et al., Interaction of human SUV3 RNA/DNA helicase with BLM helicase; loss of the SUV3 gene results in mouse embryonic lethality. Mech Ageing Dev, 2007. 128(11-12): p. 609-17.
Stepien, P.P., et al., The yeast nuclear gene suv3 affecting mitochondrial post-transcriptional processes encodes a putative ATP-dependent RNA helicase. Proc Natl Acad Sci U S A, 1992. 89(15): p. 6813-7.
Wang, D.D., et al., Human mitochondrial SUV3 and polynucleotide phosphorylase form a 330-kDa heteropentamer to cooperatively degrade double-stranded RNA with a 3'-to-5' directionality. J Biol Chem, 2009. 284(31): p. 20812-21.
Sato, R., et al., Novel biallelic mutations in the PNPT1 gene encoding a mitochondrial-RNA-import protein PNPase cause delayed myelination. Clin Genet, 2018. 93(2): p. 242-247.
Berke, I.C. and Y. Modis, MDA5 cooperatively forms dimers and ATP-sensitive filaments upon binding double-stranded RNA. EMBO J, 2012. 31(7): p. 1714-26.
Heid, M.E., et al., Mitochondrial reactive oxygen species induces NLRP3-dependent lysosomal damage and inflammasome activation. J Immunol, 2013. 191(10): p. 5230-8.
Hsu, C.G., et al., Pnpt1 mediates NLRP3 inflammasome activation by MAVS and metabolic reprogramming in macrophages. Cell Mol Immunol, 2023. 20(2): p. 131-142.
Giordano, A.M.S., et al., DNA damage contributes to neurotoxic inflammation in Aicardi-Goutieres syndrome astrocytes. J Exp Med, 2022. 219(4).
Maharana, S., et al., SAMHD1 controls innate immunity by regulating condensation of immunogenic self RNA. Mol Cell, 2022. 82(19): p. 3712-3728 e10.
Mannion, N.M., et al., The RNA-editing enzyme ADAR1 controls innate immune responses to RNA. Cell Rep, 2014. 9(4): p. 1482-94.
Oda, H., et al., Aicardi-Goutieres syndrome is caused by IFIH1 mutations. Am J Hum Genet, 2014. 95(1): p. 121-5.
Crow, Y.J. and D.B. Stetson, The type I interferonopathies: 10 years on. Nat Rev Immunol, 2022. 22(8): p. 471-483.
Poirier, K., et al., Mutations in the neuronal ss-tubulin subunit TUBB3 result in malformation of cortical development and neuronal migration defects. Hum Mol Genet, 2010. 19(22): p. 4462-73.
Adamson, K.I., E. Sheridan, and A.J. Grierson, Use of zebrafish models to investigate rare human disease. J Med Genet, 2018. 55(10): p. 641-649.
Biffi, A., et al., Correction of metachromatic leukodystrophy in the mouse model by transplantation of genetically modified hematopoietic stem cells. J Clin Invest, 2004. 113(8): p. 1118-29.
Marshall, M.S., et al., Long-Term Improvement of Neurological Signs and Metabolic Dysfunction in a Mouse Model of Krabbe's Disease after Global Gene Therapy. Mol Ther, 2018. 26(3): p. 874-889.
Priller, J., et al., Targeting gene-modified hematopoietic cells to the central nervous system: use of green fluorescent protein uncovers microglial engraftment. Nat Med, 2001. 7(12): p. 1356-61.
Rutherford, H.A. and N. Hamilton, Animal models of leukodystrophy: a new perspective for the development of therapies. FEBS J, 2019. 286(21): p. 4176-4191.

The authors declare no competing interests.

SupplementaryTablesandFigures.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Biallelic mutations in SUPV3L1 cause an inherited neurodevelopmental disorder with variable leukodystrophy due to aberrant mitochondrial double stranded RNA processing

Status:

Version 1

Abstract

Figures

Introduction

Methods

Patient ascertainment

Molecular and cellular assays

Plasmid preparation and validation

SUPV3L1 complementation assay

Zebrafish Modelling

Zebrafish husbandry and ethics

Generation of supv3l1 zebrafish crispants

Construction of Tg(mpeg1.1:mts-mNeonGreen)sh631

Quantitative PCR

Zebrafish confocal imaging and analysis

TUNEL staining and quantification

Statistical analysis

Results

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1