MLKL‒OPTN axis regulates herpesvirus-induced neurological sequelae

2.1 Cell lines, viruses and tissues

HeLa (OPTN +/+) and HeLa OPTN knockout (KO) (OPTN ‒/‒) cells were provided by Dr. Richard Youle (National Institutes of Health) and cultured in were grown in Dulbecco's Modified Eagle Media (DMEM) (Gibco) supplemented with 10% (v/v) Fetal Bovine Serum (FBS) (Gibco) and 1% (v/v) penicillin‒streptomycin (Gibco). Vero cells were obtained from ATCC. SV40-immortalised HCE cell line was obtained from Dr. Kozaburo Hayashi (National Eye Institute). Luhmes cells were provided by Dr. David Bloom (University of Florida) and cultured as previously described.¹⁹

The HSV-1 McKrae strain (HSV-1) was provided by Dr. Homayon Ghiasi (Cedars Sinai). HSV-1 KOS and 17 strains were provided by Dr. Patricia Spear (Northwestern University). HSV-1 K26RFP strain was provided by Dr. Prashant Desai (Johns Hopkins University). Human brain (normal) tissue slides (cat no. GTX2203) and human brain (multiple sclerosis) tissue slides (cat no. GTX2204) were obtained from GeneTex.

2.2 Generation of CRISPR/Cas9 KO model

To create OPTN ‒/‒ in the HCE cell line, we used three Alt-R CRISPR-Cas9 crRNAs: 5′-GTTCAGACACGATGCCCAAC-3′, 5′-CCTGGACACGTTTACCCCGG-3′ and 5′-CCAGTGGAGACTGTTCTCGT-3′. The three guides were pooled together to target exons 3 and 5, and exon 6 to ensure complete abrogation of the protein expression. The cells were electroporated using Amaxa Nucleofector II using program T-20. The crRNAs were incubated with TracrRNA at a 1:1 ratio (300 µM) and incubated at 95°C for 5 min. Cas9 enzyme was added to the complex of crRNA + TracrRNA at a ratio of 1:3 to form the ribonucleoprotein (RNP) complex and incubated for 10 min at room temperature (RT). After 72–96 h of nucleofection, cells were sorted for single-cell cloning and screened for the desired clone.

2.3 Plasmids, antibodies and chemical reagents

The full-length OPTN and OPTN 1–424 plasmid were provided by Dr. Beatrice Yue (University of Illinois Chicago).²⁰

The following antibodies and stains were used in this study: DAPI (4',6-diamidino-2-phenylindole) (D9542, Sigma), mouse monoclonal to LAMP1 (Abcam, #ab25630, [H4A3]), RIPK1 antibody (Abclonal #A7414), RIPK1 rabbit polyclonal antibody (Abclonal #A7414), RIPK3 rabbit polyclonal antibody (Abclonal #A5431), MLKL rabbit monoclonal antibody (Abclonal #A21894), phospho-MLKL-T357/S358/S360 rabbit polyclonal antibody (Abclonal #AP0949), epidermal growth factor receptor (EGFR) rabbit polyclonal antibody (Abclonal #A11351), mouse anti-EEA1 monoclonal antibody, clone 14 (BD Biosciences #610456), rabbit polyclonal anti-GAPDH (Proteintech 10494-1-AP), rabbit polyclonal anti-OPTN (C-terminal) (cayman chemical, no. 100000), goat anti-rabbit immunoglobulin G (IgG) (H + L) highly cross-adsorbed secondary antibody, Alexa Fluor 647 (Thermo Fisher, #A-21245), goat anti-mouse IgG (H + L) highly cross-adsorbed secondary antibody horseradish peroxidase (HRP) (Thermo Fisher, 31432), goat anti-rabbit IgG (H + L), cross-adsorbed secondary antibody HRP (Thermo Fisher, G-21234) and anti-HSV1 + HSV2 gB antibody (Abcam #ab6506),

The following chemicals were used in this study: bafilomycin A1 (Sigma) to inhibit autophagic flux, cycloheximide (Sigma) to block de novo protein synthesis and recombinant human EGF (Peprotech). Lipofectamine 2000 (Thermo Fisher) and RNAiMAX (Thermo Fisher) for transfections. Necrostatin 1s (Nec1s; Selleckchem) for inhibition of necroptosis. NSA (Selleckchem) for inhibition of MLKL.

2.4 Mice studies

2.5 Behavioural studies

Eight-week-old Optn +/+ and Optn ‒/‒ mice went through all the behavioural studies listed below. These mice were then infected with HSV-1 by corneal scarification. After 8 days of infection, the mice went through all the behavioural studies again to compare the effects of infection. Most of these behavioural studies used in this study have been adapted from previously standardised behavioural methods.

2.6 Western blot

Proteins were extracted using RIPA buffer and analysed by SDS‒PAGE followed by immunoblotting with standard blocking, antibody incubation and chemiluminescent detection protocols. Detailed methodology is described in our previous publication.⁵

ImageJ software was used for quantifying all the Western blot images. First, the scanned or digital image of the Western blot was imported into ImageJ software. Using the ‘Rectangular Selection’ tool, bands of interest corresponding to target proteins and loading controls (GAPDH) were outlined and selected. The ‘Gel Analysis’ tool was then utilised to plot lane profiles and quantify band intensities. Background subtraction was performed to minimise non-specific signals, followed by normalisation of target protein bands to loading controls. Data analysis included calculating the ratio of band intensities for each protein of interest to the corresponding loading control (relative to GAPDH). GAPDH and β-actin were used as loading controls throughout the study. As our work was initiated several years ago during the pandemic, the availability of certain reagents including antibodies at specific times influenced our choice between the two.

2.7 Immunoprecipitation

Cells were scraped from culture dishes and centrifuged to pellet cells. Cell pellets were lysed on ice for 1 h in immunoprecipitation (IP) buffer (250 mM NaCl, 50 mM Tris, .5 mM EDTA and .5% NP-40) with protease/phosphatase inhibitor added. Lysates were centrifuged to remove cell debris and the soluble portion was precleared with isotype control antibody (Santa Cruz) and rotein A/G conjugated beads (Santa Cruz) for 1 h with agitation at 4°C. Beads were pelleted by centrifugation and lysates were moved to new tubes. The lysates were then incubated with isotype or specific antibody (1:100) on ice for 1 h. Twenty microlitres of protein A/G beads was added and samples were agitated at 4°C overnight. The beads containing the immunoprecipitated proteins were pelleted by centrifugation, and the unbound portion was decanted. The beads were washed three times in IP buffer prior to processing for immunoblot analysis.

2.8 Immunohistochemistry and immunofluorescence

Ten-micrometre Formalin-Fixed, Paraffin Embedded (FFPE)Exper tissue sections were used for immunohistochemistry and immunofluorescence experiments. For dewaxing, slides were placed in xylenol for 10 min, then 100% ethanol for 10 min, then 90% ethanol for 10 min, and last 70% ethanol for 10 min. The slides were then washed in dH₂O for 5 min. For antigen retrieval, slides were placed in antigen retrieval solution (Vector Laboratories), and the container was microwaved at high for 5 min. For permeabilisation, slides were washed with PBS/Triton (.25%, v/v; PBST) solution twice for 10 min. Slides were then blocked in 5% bovine serum albumin in PBS for 60 min at RT. Slides were then stained with the primary antibody (1:100 in PBST) and kept in the humid box overnight at RT, followed by three washes in PBST. For immunofluorescence, slides were stained with the secondary antibody (1:100 in PBST) and DAPI for 60 min at RT. Slides were washed three more times in PBST then mounted with ProLong Diamond Antifade Mountant (Invitrogen), sealed with clear nail polish, then visualised by fluorescence microscopy. For immunohistochemistry, the anti-goat HRP-DAB Cell and Tissue Staining Kit (R&D systems) was used.

2.9 Quantitative polymerase chain reaction

Trizol (Thermo Scientific, 15596018) was used to extract total RNA from cultured cells, according to the manufacturer's instructions. Complementary DNA was then produced using High-Capacity cDNA Reverse Transcription kit (Thermo Scientific, 4368814). Real-time quantitative polymerase chain reaction (qPCR) was performed with Fast SYBR Green Master Mix (Life Technologies) on QuantStudio 7 Flex system (Life Technologies).

2.10 Confocal immunofluorescence microscopy

Cells were cultured in 35 mm glass bottom dishes (Cellvis #D35-10-1.5-N). Cells were fixed in 4% paraformaldehyde for 10 min and permeabilised with .1% Triton-X for 10 min at RT for intracellular labelling, followed by incubation with primary antibody for 1 h at RT. When a secondary antibody was needed, cells were incubated with respective FITC- or Alexa Fluor 647-conjugated secondary antibody (Sigma–Aldrich F9137 or Thermo Scientific A21244) at a dilution of 1:100 for 1 h at RT. NucBlue Live ReadyProbes Hoechst stain (Thermo Scientific R37605) was included with secondary antibody stains when applicable, according to the manufacturer's specifications. Samples were examined under LSM 710 confocal microscope (Zeiss) using a 63× oil immersion objective. The fluorescence intensity of images was calculated using ZEN software.

2.11 Plaque assay

Viral egress was measured using a plaque assay. Monolayers of Optn +/+ and Optn ‒/‒ cells were plated in six-well plates and infected with the McKrae-WT virus at MOI .1. Media were collected at different time points post-infection and titred on Vero cells. Briefly, primary incubation of collected media was performed with Opti-MEM (Life Technologies) for 2 h. Vero cells were then incubated with growth media containing 1% methylcellulose for 72 h followed by fixing with 100% methanol and staining with crystal violet solution. Formed plaques were counted and analysed.

2.12 siRNA transfection

A Dicer-substrate short interfering RNAs (DsiRNAs) TriFECTa Kit (IDT) with predesigned siRNA molecules were used for transfections in this study. Cells were plated and grown to 50% confluency. Cells were then transfected as per the manufacturer's protocol using RNAiMAX at 1 µL/mL in OptiMEM (Thermo Fisher). Multiple concentrations for each premade siRNA molecule were tested and it was determined that OPTN siRNA 1 at 10 nM and MLKL siRNA 1 at 30 nM produced effective knockdown with minimal cell death after 48 h of transfection.

2.13 Statistical analysis

The data shown in the figures are means ± standard error of means (SEM). The statistical tests were performed on GraphPad Prism. Asterisks indicate significant difference: ^*p < .05, ^**p < .01, ^***p < .001 and ^****p < .0001.

3.1 MLKL is a cellular factor that promotes HSV-1 infection in OPTN-deficient conditions

Our previous work showed that OPTN-deficient cells were likely to undergo necroptosis-induced cell death.⁵ In consideration of the critical role MLKL plays during necroptosis, we examined the expression of MLKL and p-MLKL (indicates activation of MLKL by RIPK3) in Optn ‒/‒ HCE cells and tissues at different stages of HSV-1 infection. Optn +/+ and Optn ‒/‒ HCE cells were infected with HSV-1 McKrae strain (HSV-1) for 6 and 24 h of infection. We did not observe changes in MLKL expression in the presence or absence of OPTN at 6 h post-infection, but there was upregulation of MLKL 24 h post-infection in Optn ‒/‒ HCE cells (Figure 1A). Similarly, a higher expression of MLKL was also observed in HeLa cells for different HSV-1 strains (Figure S1A‒C). We additionally noted a higher expression of p-MLKL in Optn ‒/‒ cells after 24 h of HSV-1 infection (Figures 1A and S7A). We also conducted a time course experiment to evaluate the expression of MLKL during the first 6 h of infection. Interestingly, we did observe a higher expression of MLKL in OPTN ‒/‒ cells at 2 and 4 h post-infection (Figures 1B and S7B). We did not observe phosphorylation of MLKL during these early stages of HSV-1 infection. This indicated a role of MLKL that is not dependent on its phosphorylation by RIPK3. To further confirm if the upregulation of MLKL during the early stages of infection was dependent on RIPK3, we conducted the time course experiment in a RIPK3-deficient cell line HeLa. We observed similar upregulation of MLKL in Optn ‒/‒ cells at 2 h and 4 h post-infection (Figures 1C and S7C). Next, we evaluated if MLKL deficiency impacts HSV-1 infection. To analyse this, we conducted a transient knockdown of MLKL in HCE and HeLa cells. The growth of HCE or HeLa cells were not affected by knockdown of MLKL (data not shown). The accumulation of gB, an HSV-1 envelope protein in siCTRL HCE cells, was much higher than that in siMLKL cells (Figures 1D and S7D,E). We also used 1 MOI 17 GFP‒HSV-1 strain, a GFP-tagged HSV-1, to infect the cells and observed that the GFP intensity of siCTRL Optn ‒/‒ HeLa cells was much higher than that of siMLKL Optn ‒/‒ HeLa cells (Figure 1G,H). Consistently, the titre of HSV-1 in siCTRL Optn ‒/‒ cells was also markedly higher than siMLKL Optn ‒/‒ HeLa cells infected with HSV-1 for 24 h (Figure 1I). Collectively, these data demonstrate that upregulated MLKL aids HSV-1 in the absence of OPTN and that the negative regulation of MLKL by OPTN constituting an MLKL‒OPTN axis is necessary to control HSV-1 infection.

3.2 MLKL facilitates endosomal trafficking of HSV-1 during the early stages of infection

Previously, we observed upregulation of MLKL during the early stages of infection (Figure 1B,C), which was independent of phosphorylation of MLKL required for canonical role of MLKL in executing necroptosis. Our imaging data also indicated an accumulation of MLKL near the nucleus at 4 h post-infection (Figure S1D). To investigate the non-canonical role of MLKL in the early stages of HSV-1 infection, we conducted our next set of experiments in the RIPK3-deficient HeLa cell lines. The initial stages of infection include viral entry followed by the transport of virions into the nucleus. Previous work already determined that OPTN is not involved in viral entry⁵; however, to our surprise, we observed a higher transport of HSV-1 virions into the nucleus in Optn ‒/‒ cells as compared to Optn +/+ cells (Figure S2A). MLKL has been previously suggested to enhance endosomal transport so we explored the idea that the discrepancy in increased virions reaching the nucleus in the absence of OPTN was due to upregulated MLKL enhancing endocytic pathways, which may inadvertently be aiding the transport of HSV-1 virions to the nucleus.²⁵ First, we analysed if there is faster transport of molecules in Optn ‒/‒ cells in general. For this, we monitored the intracellular transport of EGFR within cells following treatment with EGF. Immunofluorescence analysis indicated that the transport of EGFR was delayed in Optn-expressing cells with low MLKL expression in comparison to the Optn ‒/‒ cells. (Figure 2A). This faster translocation of EGFR observed in Optn ‒/‒ cells was delayed in MLKL depleted (siMLKL) Optn ‒/‒ cells (Figure S8C), further corroborating that higher expression of MLKL is instrumental in regulating the intracellular transport rate in Optn ‒/‒ cells.

A similar phenomenon was observed in HSV-1-infected cells where transport of HSV-1 was delayed in Optn +/+ cells supported by our confocal imaging (Figure 2B). Our electron microscopy data showed that more HSV-1 particles were transported to the plasma membrane in Optn ‒/‒ cells indicating higher endocytic pathway activation by MLKL co-opted by the virus (Figure 2C). In comparison, HSV-1 particles were still inside the cells in the Optn +/+ cells (Figure 2C). These observations corroborated our hypothesis that MLKL's enhancement of the endocytic pathway is hijacked by the virus when OPTN is deficient, and the presence of OPTN ensures curbing this MLKL-aided viral transport of HSV-1.

To further evaluate the importance of MLKL's modulation by OPTN in dictating endocytic pathways, we conducted experiments with a pharmacological MLKL inhibitor NSA. Immunofluorescence imaging indicated a similar rate of transport in Optn +/+ and Optn ‒/‒ cells treated with NSA (Figure 2D). Furthermore, we observed less transport of gB into the nucleus in Optn ‒/‒ cells treated with NSA in comparison to the mock-treated Optn ‒/‒ cells (Figures 2E and S7E,F). In summary, MLKL is the driving force for enhanced transport of HSV-1 particles in Optn ‒/‒ conditions; utilising NSA as an MLKL inhibitor specifically in OPTN-deficient conditions helps rescue the gB levels close to that in Optn +/+, indicating unregulated MLKL's role in driving HSV-1 transport to the nucleus. Thus, MLKL plays a critical role during early stages of HSV-1 infection, which does not require its activation by RIPK3.

3.3 MLKL is selectively degraded by OPTN's ubiquitin binding domain through direct binding

Next, we examined the mechanism by which OPTN regulates MLKL expression. As an autophagy adapter protein, OPTN is known to directly interact with the ubiquitinated cargo and carry them for autophagic degradation. In this regard, we hypothesized that the direct binding of MLKL to OPTN promotes its autophagic degradation under normal and infected conditions. To test this hypothesis, we performed a co-IP assay by immunoprecipitating MLKL to study its potential direct interaction with OPTN. Our IP studies indicated a direct interaction between MLKL and OPTN (Figure 3A). The direct interaction between MLKL and OPTN suggested autophagic control of MLKL exerted by OPTN, and to test this, we used a cycloheximide chase assay to inhibit new protein synthesis. We found that the degradation of MLKL was OPTN-dependent as MLKL degradation ceased in Optn ‒/‒ conditions (Figure 3B). The presence or absence of Bafilomycin (autophagy inhibitor) had no effect on MLKL degradation in Optn ‒/‒ cells (Figure 3C). In comparison, degradation of MLKL was inhibited in the presence of Bafilomycin in Optn +/+ cells (Figure 3C).

Immunofluorescence imaging shows lysosomal uptake of MLKL en route to its autophagic degradation only in the presence of OPTN indicated by the colocalisation of MLKL and Lamp1 (lysosome marker), which is abrogated in the absence of OPTN (Figure 3E). Additionally, to determine the site of interaction between MLKL and OPTN, we explored the possibility that OPTN is incorporated into the MLKL necrosome complex. Wild-type cells were infected with 10 MOI K26-RFP HSV-1 infection and the intracellular location of OPTN was monitored. The results indicated that OPTN was indeed recruited to the necrosome complex (indicated by p-MLKL rings) in response to HSV-1 infection (Figure 3D). Furthermore, we also observed that MLKL oligomer formation was hindered in the presence of Optn (Figure S1A).

To further elucidate the novel direct interaction between MLKL and OPTN, we conducted domain mapping experiments to find the OPTN domain responsible for interaction with MLKL. Next, we pursued the idea that OPTN's ubiquitin binding domain (UBD), which is responsible for recognising cargo for autophagy, is the selective domain that interacts with and targets MLKL. To test this hypothesis, we transfected Optn ‒/‒ cells with expression vectors of full-length OPTN and OPTN fragments lacking the UBD domain OPTN1-424 (Figure 3F). Our results indicate that only the full-length OPTN plasmid reduced the MLKL expression, as shown by our Western blot and imaging data, and the UBD-deleted fragment was not successful in both binding to MLKL and sufficiently targeting it for degradation (Figure 3G,H). Collectively, our results show MLK's direct interaction with OPTN, which enables its recognition of autophagic degradation through OPTN's UBD.

3.4 HSV-1 infection causes aggregation of MLKL bodies in the brains of OPTN-deficient mice

To investigate whether MLKL upregulation occurs during HSV-1 infection in vivo, we infected Optn +/+ and Optn ‒/‒ mice with ocular HSV-1 and assessed MLKL levels in the eyes, trigeminal ganglia (TG) and brain. Consistent with our in vitro findings, Optn ‒/‒ mice exhibited higher infection rates 4 days post-infection (dpi) (Figure 4A), with the eye serving as the primary site of HSV-1 infection. In line with the data presented in Figure 1, elevated MLKL protein expression was observed in whole-eye lysates from Optn ‒/‒ mice 4 dpi (Figure S3A). Additionally, we noted increased MLKL, RIPK3 and RIPK1 transcripts in eye lysates and TG of HSV-1-infected Optn ‒/‒ mice compared to Optn +/+ mice (Figure S4A‒D). Given HSV-1's neurotropic nature, we assessed MLKL activation in the mouse brain, which revealed significantly higher RIPK3 and MLKL transcripts (Figure 4B). To understand the previously unexplored consequences of this significant activation of MLKL in brain tissue exacerbated by the loss of OPTN in response to HSV-1 infection, we used immunofluorescence imaging that revealed extensive MLKL and p-MLKL aggregates (referred to as MLKL bodies) in the brainstem of only Optn ‒/‒ mice 4 dpi (Figures 4C and S3B). This aggregation of MLKL bodies also occurred when we silenced OPTN in HSV-1-infected human mesencephalic neuronal cells (LUHMES), which suggested that MLKL was disposed to form aggregates (MLKL bodies) in neuronal cells and tissue in the absence of OPTN (Figure 4D).

To further understand the impact of these MLKL bodies in vivo, we evaluated the expression of different cell types in the brain in Optn +/+ and Optn ‒/‒ mice during HSV-1 infection (Figure S5). We observed a significant decrease in the expression of oligodendrocytes in Optn ‒/‒ mice brains during HSV-1 infection, as shown in our imaging data (Figures S5 and 5A). Myelin basic protein (MBP) is the second most abundant protein in the myelin accounting for 30% of the total CNS myelin protein. It is synthesised in the CNS oligodendrocytes and essential for myelin formation and its long-term maintenance. The pathogenic role of MBP is associated with MS where it could act as an autoantigen, inducing self-attack on myelin sheath, leading to demyelination. So MBP was used as a biomarker to evaluate expression of Oligodendrocytes. To assess the hypothesis that the decrease in oligodendrocytes driven by HSV-1 infection specifically occurring in Optn ‒/‒ mice is due to the aggregation of MLKL bodies that are driving necroptotic cell death in oligodendrocytes, we utilised immunohistochemistry. Our experiments revealed colocalisation of MBP and MLKL in HSV-1-infected brainstems of Optn ‒/‒ mice, corroborating the role of MLKL in the death of these oligodendrocytes during HSV-1 infection (Figure 4E). Taken together, we show evidence of extensive loss of oligodendrocytes as a consequence of HSV-1 infection in the brainstem of Optn ‒/‒ mice driven primarily by the aggregation of catastrophic MLKL bodies.

3.5 MLKL dysregulation shows MS-like motor defects in HSV-1-infected mice

Upregulated MLKL in the Optn ‒/‒ mice post-HSV-1 infection-induced oligodendrocyte death, specifically in the brainstem. Oligodendrocytes are cells responsible for producing the protective myelin sheath around neurons and play crucial roles in maintaining neuronal homeostasis. We proceeded to evaluate the effect of MLKL-driven oligodendrocyte death by assessing demyelination in these tissues. As hypothesized, we observed reduced myelin expression in the brains of Optn ‒/‒ mice (Figure 5B). Remarkably, oligodendrocyte death and myelin loss were specifically localised to the brain stem and cerebellum regions (Figure 5A,B). The cerebellum plays a critical role in coordinating voluntary movements, maintaining balance, and fine-tuning motor skills, contributing to the precision and smoothness of physical activities. To determine the effect of MLKL-induced oligodendrocyte death on the coordination and balance of mice after HSV-1 infection, we conducted a series of behavioural studies on Optn +/+ and Optn ‒/‒ mice. Initially, we performed the tape removal test to assess sensorimotor coordination. Before infection, both Optn +/+ and Optn ‒/‒ mice could sense and remove the tape at similar times. However, after HSV-1 infection, Optn ‒/‒ mice took significantly longer to remove the tape, indicating a substantial coordination defect (Figure 5D,E). Subsequently, to evaluate balance problems, we utilised the ledge test and observed that Optn ‒/‒ mice displayed shaking behaviour and reluctance to lower themselves into the cage, resembling cerebellar ataxia in humans (Figure 5F). The rotarod test, which examines motor skills, and the grip test, did not yield statistically significant results (Figure 5G,H).

Demyelination induced by unregulated MLKL in Optn ‒/‒ mice contributed extensively to significant defects that bore a resemblance to demyelinating disorders. Demyelination typically involves the shedding of myelin, which can be detected in the serum of patients. To evaluate myelin shedding in our model, we measured the levels of shedding MBP in mouse serum. HSV-1-infected Optn ‒/‒ mice displayed the highest levels of MBP in the serum, supporting our hypothesis of severe demyelination occurring due to dysregulated MLKL and MBP shedding compared to the Optn +/+ mice (Figure 5I). The series of behaviour tests we performed on mice highlighted impairments that were phenotypically similar to the extensively studied demyelinating disorder MS, often studied using the EAE mouse model. Brain lysates from EAE mice showed similar reductions in mRNA and protein levels of OPTN, which indicated that there may be a dysregulated MLKL pathway activation in these tissues as well (Figures 5J,K and S7G). Additionally, increasing MLKL and GFAP (a marker for CNS injury) protein levels alongside decreasing MBP, corresponding to the increasing disease score for EAE mice depicting the MS state, reinforced the dysregulation of the MLKL‒OPTN axis.

Strikingly, we also found higher levels of MLKL and RIPK3 transcripts in these EAE mice (Figure S4E). The spinal cord is frequently affected in MS, causing motor, sensory and autonomic dysfunction. Interestingly, similar to the brains of our HSV-1-infected Optn ‒/‒ mice, we detected aggregates of MLKL bodies in the spinal cords of these EAE mice (Figure 5L). Lastly, we also observed lower OPTN levels in brain tissues of MS lesions collected from humans compared to the normal counterparts (Figure 5M). Overall, the EAE mice with progressively worse MS disease exhibited similar increases in MLKL and loss of MBP and OPTN that was like changes we report in Optn-deficient HSV-1-infected mice. Collectively, these data suggested that the intricate link between MLKL and OPTN that serves to maintain myelin, with dysregulation of this balance resulting in severe impairments in motor functions and behaviours reported typically in MS and MS-like demyelinating disorders.

3.6 NSA treatment rescues the MLKL-induced demyelination during HSV-1

Since the upregulation of MLKL leads to demyelination in Optn ‒/‒ mice during HSV-1 infection, we investigated if inhibiting MLKL using NSA could mitigate demyelination in these mice. Our initial in vitro studies suggested that NSA treatment reduced MLKL expression in Optn ‒/‒ cells during HSV-1 infection (Figures 6A and S6B). Next, we conducted in vivo studies, and like the in vitro studies, we observed a reduction in MLKL and p-MLKL expression with NSA treatment in Optn ‒/‒ and Optn +/+ mice brains (Figures 6D,F and S6F). Consecutively, our in vivo study also revealed that NSA treatment rescued the loss of myelin in Optn ‒/‒ infected mice (Figure 6H). To elucidate if MLKL expression regulation in Optn ‒/‒ is dependent on higher HSV-1 titres, we conducted in vitro and in vivo experiments with Acyclovir (ACV), which is the standard of care therapy to treat herpes patients. Interestingly, the reduction of virus by ACV did not reduce the MLKL and p-MLKL expression levels (Figure S6A,D,F and S8A). ACV treatment was also unsuccessful in rescuing the damage phenotype of oligodendrocytes and demyelination defects associated with increased MLKL suggesting that standalone ACV therapy in response to virus-induced MLKL activation and upregulation is insufficient and while the initial trigger in the pathology may have been the virus, the tight regulation of the MLKL‒OPTN axis protects against this severe neuropathy (Figure 6H,J).

To further clarify if demyelination in these mice was a result of the cell death-dependent role of MLKL, we performed in vitro and in vivo studies with Nec1s, which is a known necroptosis inhibitor that inhibits phosphorylation of MLKL required to execute necroptosis. Nec1s treatment was able to reduce phosphorylation of MLKL but there was no change in basal levels of MLKL both in vitro and in vivo (Figure 6B,E,G, S8B). Also, there was minimal MBP expression rescue in Nec1s-treated mice but not as significant as compared to NSA treatment (Figure 6I,K). Together, these data indicated that reduction of p-MLKL is not sufficient to reduce demyelination, and there might be a cell death-independent role of MLKL acting in conjunction that contributes to the demyelination we see.