Z-DNA binding protein 1 mediates necroptotic cell death in primary murine microglia following herpes simplex virus-1 infection
Alexander J. Suptela, Ian Marriott
TL;DR
This study shows that ZBP1 helps microglia fight HSV-1 infection by causing cell death, but the effect depends on the virus strain and may worsen brain damage.
Contribution
The study reveals ZBP1's role in HSV-1-induced necroptosis in microglia and highlights strain-specific differences in cell death pathways.
Findings
ZBP1 acts as a restriction factor for HSV-1 in primary murine microglia.
HSV-1 isolates induce necroptosis in microglia via ZBP1-dependent and independent mechanisms.
Apoptosis in microglia occurs only with a lab-adapted HSV-1 strain and is ZBP1-independent.
Abstract
The mechanisms by which microglia respond to viral central nervous system (CNS) pathogens are now becoming apparent with the demonstration that they express an array of pattern recognition receptors that include cytosolic sensors for exogenous nucleic acids. We have previously shown that microglia express Z-DNA binding protein 1 (ZBP1) and found that this sensor contributes to their inflammatory responses to the clinically relevant DNA virus, herpes simplex virus-1 (HSV-1). More recently, we showed that ZBP1 serves as a restriction factor for HSV-1 in murine astrocytes and is associated with the induction of both necroptotic and apoptotic cell death pathways in these cells. Here, we demonstrate that this cytosolic DNA sensor similarly functions as a HSV-1 restriction factor in primary murine microglia. However, unlike astrocytes, we have determined that a neuroinvasive…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6- —University of North Carolina at Charlotte
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsNeuroinflammation and Neurodegeneration Mechanisms · interferon and immune responses · Nuclear Receptors and Signaling
Introduction
It is now appreciated that microglia can detect and respond to invading pathogens in the central nervous system (CNS) via the expression of a variety of pattern recognition receptors (PRRs) (Chauhan et al. 2008, 2009; Furr et al. 2010, 2011; Furr et al., 2012; Jiang et al. 2014; Crill et al. 2015). Upon the detection of pathogen and/or damage associated molecular patterns (PAMPs and DAMPs, respectively), PRRs initiate signaling cascades that result in the production of proinflammatory and/or antiviral mediators by microglia (Town et al. 2006; Huang et al. 2018) in a similar manner to that seen in peripheral host cells (Gewirtz et al. 2001; Diebold et al. 2004; Upton et al. 2012; Orzalli et al. 2015; Lahaye et al. 2018). However, whether such responses act in a beneficial or detrimental manner in the CNS during infection is less well understood and may be context dependent (Hofer et al., 2013; Goldmann et al. 2016; Blank and Prinz 2017; McDonough et al. 2017). Of these PRRs, cytosolic DNA sensors, such as Z-DNA binding protein 1 (ZBP1; also known as DNA-dependent activator of interferon regulatory factors (DAI), are of particular interest in CNS infection due to their ability to detect DNA from CNS-specific pathogens, such as herpes simplex virus type-1 (HSV-1) (Jeffries et al. 2022; Wang et al. 2024).
The importance of ZBP1 in the mediation of host responses to viral infection has been extensively described (Jeffries and Marriott 2020; Jeffries et al. 2022; Karki et al. 2022; Suptela and Marriott 2023; Oh and Lee 2023). ZBP1 was initially determined to be a critical component in eliciting antiviral type-1 interferon responses to DNA viruses (Takaoka et al. 2007; Upton et al. 2012), but subsequent studies revealed that it is also capable of inducing pro-inflammatory mediator production via the activation of nuclear factor kappa B (NF-kB) following interaction with dsDNA (Rebsamen et al. 2009). More recently, ZBP1 has been implicated in the initiation of cell death pathways, such as necroptosis, in peripheral cell types and astrocytes (Upton et al. 2012; Daniels et al. 2019; Rothan et al. 2019; Jeffries et al. 2022). Upon detection of dsDNA, ZBP1 interacts with receptor-interacting protein kinase 3 (RIPK3) via their shared RIP homotypic interaction motif (RHIM) domains (Rebsamen et al. 2009). RIPK3 subsequently phosphorylates mixed lineage kinase domain-like (MLKL), driving its oligomerization and pore formation in cell membranes that precipitates necroptotic cell death (Upton et al. 2012; Upton et al., 2017). Importantly, we have recently described the ability of ZBP-1 to serve as a restriction factor for HSV-1 infection in primary murine astrocytes, and we have shown that this is associated with the induction of both necroptotic and apoptotic death pathways in these cells (Jeffries et al. 2022).
In the present study, we have examined the importance of ZBP1 in the mediation of microglial responses to a clinically relevant DNA virus, HSV-1. We show that this cytosolic DNA sensor functions as a viral restriction factor for HSV-1 in a similar manner to that seen previously in murine astrocytes (Jeffries et al. 2022). However, unlike astrocytes, we have determined that a neuroinvasive clinically-derived HSV-1 isolate induces necroptosis, but not apoptosis, in microglia in a ZBP1-dependent, as well as a ZBP1-independent but RIPK1-mediated, manner. Interestingly, we have found that a laboratory adapted HSV-1 strain can elicit apoptosis in these CNS cells in a ZBP-1-independent manner, in addition to both ZBP1-dependent and independent necroptosis, and such strain-specific effects lend support to the use of clinically relevant viral isolates in the study of host cell responses.
Materials and methods
Isolation and culture of primary murine microglia
Microglia were isolated from ZBP1-expressing C57BL/6J (ZBP1+/+; Jackson Laboratories) or ZBP1-deficient C57BL/6J (ZBP1-/-) neonatal (day 2–3) mice according to method previously employed by our laboratory (Chauhan et al. 2008; Crill et al. 2015). ZBP1-/- mice were a kind gift from Dr. Laura Knoll (University of Wisconsin-Madison, Madison, WI) and were generated using CRISPR-Cas9 technology (Guo et al. 2018). Six to eight brains were used to prepare each flask of cells and were minced and forced through a wire screen in the presence of 0.25% trypsin prior to washing with RPMI 1640 (Gibco, Cat# 31800089) containing 10% fetal bovine serum (FBS; Gibco, Cat# A5256701) and subsequent resuspension in fresh RPMI 1640 containing 10% FBS and 1% penicillin/streptomycin. Mixed glial culture was maintained for two weeks prior to microglial isolation. Microglial cultures were isolated by trypsinization (0.25% trypsin containing 1mM EDTA) as we have previously described (Chauhan et al. 2008; Crill et al. 2015; Saura et al. 2003) and cultured in RPMI 1640 containing 10% FBS and 20% conditioned medium isolated from a murine monocyte-like cell line that secretes colony stimulating factor-1 (CSF-1) (LADMAC; ATCC CRL-2420). No significant differences in phenotype, cell proliferation, or spontaneous cell death were observed between microglia derived from WT and ZBP1-deficient mice. These studies were performed in accordance with relevant institutional polices and federal guidelines regarding the use of animals for research purposes.
Viral stock Preparation and infection of cultured microglia
The HSV-1 MacIntyre strain (HSV-1(MacIntyre), which was first isolated from the brain of a patient with lethal HSV encephalitis, was obtained from ATCC (VR-539) and the laboratory strain (HSV-1(F) was a kind gift from Dr. Edward Mocarski (Emory University, Atlanta, GA). HSV-1 viral stocks were prepared by infecting Vero cells (ATCC CCL-81) at a multiplicity of infection (MOI) of 0.01 until 100% of cells displayed cytopathic effects (48–72 h). Following this incubation period, the flasks of infected Vero cells were frozen at −80 °C for approximately 15 min before being warmed to room temperature inside a biosafety cabinet. The cell suspension was then removed, and pulse sonicated to release intracellular virions. Following sonication, the cell slurry was centrifuged at approximately 4000 RCF to pellet the cell debris and the cell-free virus-containing supernatants were collected. Virus-containing supernatants were aliquoted and mixed 1:1 with sterile milk for increased stability during storage and freeze-thaw cycles. Viral titers were obtained via standard plaque assay with Vero cells as performed previously (Jeffries et al. 2020, 2022).
Murine microglia (3 × 10^4^ cells per well in 96-well plates (cell viability assessments) and 1 × 10^5^ cells per well in 12-well plates (Western blot analysis and ELISAs)) were infected with either HSV-1(MacIntyre) or HSV-1(F) at an MOI of 0.2 or 2 and the virus was allowed to adsorb for approximately 1 h in Medium 199 containing Earles salts (Gibco, Cat# 11150059). After 1 h, cells were washed with 1X PBS and maintained in RPMI 1640 containing 10% FBS, 1% penicillin/streptomycin, and 20% LADMAC conditioned media for the indicated times prior to cell viability measurements or collection of supernatants or whole-cell lysates. Featured pharmacological inhibitors (fludarabine 10 µM, Selleckchem Cat# S1491; GSK547 50 nM, Cat# S8787; Selleckchem, GSK843 2 µM, Cat# SML2001; Sigma Aldrich, and Z-IETD-FMK 20 µM; InvivoGen Cat# inh-ietd) were reconstituted in DMSO and added to cultures following infection.
Microglial protein expression assessments
Whole-cell lysates were collected from microglia using Triton lysis buffer (10 mM Tris HCl pH 10.5, 5 mM MgCl_2_, and 1% (v/v) Triton X-100) and analyzed via Western blotting. Samples were electrophoresed on a 4–12% Bolt™ Bis-Tris Plus Mini Protein gel (Invitrogen Cat# NW04122BOX) and transferred to Immobilon-P transfer membranes (Millipore Cat# IPVH00010). Membranes were blocked with either 5% milk (for ZBP1) or 5% BSA (for pMLKL) for 1 h and then incubated overnight at 4 °C with primary antibodies directed against ZBP1 (AdipoGen Cat# AG-20B-0010-C100) or pMLKL (Abcam Cat# ab187091) or β-actin (Abcam Cat# ab8227). Membranes were then washed with 1X TBS containing 0.1% Tween-20 then incubated for one hour at room temperature in the presence of a horseradish peroxidase (HRP)-linked anti-rabbit (Cell Signaling, Cat# 7074) or anti-mouse (Cell Signaling, Cat# 7076) IgG secondary antibody. Chemiluminescence of bound enzyme was detected with SuperSignal™ West Pico PLUS (ThermoFisher Scientific Cat# 34580). Precision Plus dual color (Bio-Rad) and PageRuler™ pre-stained (ThermoFisher) molecular weight markers were employed to verify the identity of pMLKL and β-actin bands. Immunoblots shown are representative of at least three separate experiments using an Azure 300 imager (Azure Biosystems), and pMLKL expression was quantified using AzureSpot Pro software (Azure Biosystems) and normalized to immunoblot b-actin expression to correct for changes in total protein loading attributable to cell death.
Inflammatory cytokine release assessments
Since mRNA levels may not faithfully reflect biologically relevant protein production and release, sandwich enzyme-linked immunosorbent assays (ELISAs) were performed to quantify TNF, IFN-β, IL-6, IL-1b, and IL-18 secretion by murine microglia. Murine TNF, IL-1b, and IL-18 ELISAs were performed using commercially available DuoSet kits from R&D Systems (Cat# DY410, DY401, and DY7625-05, respectively). The murine IFN-β ELISAs were performed using a polyclonal goat anti-mouse IFN-β capture antibody (Biolegend, Cat# 508102) and a biotinylated hamster anti-mouse IFN-β detection antibody (Biolegend, Cat# 508105). The murine IL-6 ELISAs were performed with rat anti-mouse IL-6 capture antibody (BD Pharmingen, Cat# 554400) and a biotinylated rat anti-mouse IL-6 detection antibody (BD Pharmingen, Cat# 554402). Bound detection antibody was detected using HRP-linked streptavidin (R&D systems, Cat# DY998) followed by the addition of tetramethylbenzidine (TMB) substrate (Surmodics, Cat# TMBW-1000-01). HCl was used as stop solution to quench the reaction and absorbance was measured at 450 nm (SpectraMax iD5). The concentration of each sample was determined by extrapolation to a standard curve generated using recombinant mouse IL-6 (BD Biosciences, Cat# 554582), IFN-β (Biolegend, Cat# 581309), TNF, IL-1b, or IL-18 (contained in R&D Systems kits above).
Cell viability assessments and calculation of total percentage cell death and kinetics
Cell viability was measured every 2 h for 24 h post-infection using the RealTime-Glo™ MT cell viability assay (Promega, Cat# G9713) according to the manufacturer’s instructions. In brief, NanoLuc^®^ enzyme and MT cell viability substrate was combined with RPMI 1640 containing 10% FBS, 1% penicillin-streptomycin, and 20% LADMAC conditioned medium with or without pharmacological inhibitors for RIPK3, RIPK1, or caspase 8, and added to microglia cultures one-hour post-infection. Luciferase activity was measured every two hours using a SpectraMax^®^ iD5 plate reader for 24 h beginning at two hours following infection. Luciferase readings were normalized to uninfected controls and the resulting values were subtracted from a value of one and multiplied by 100% to calculate percentage cell death at 24 h. Data were further normalized within experimental groups by subtracting the percentage of dead cells at 2 h from values at all subsequent time points and any resulting negative values were considered as zero. Slopes of the rate of cell death and standard deviation for each treatment were calculated via linear regression in Microsoft Excel.
Statistical analysis
All data are presented as mean ± SEM. Two-way analysis of variance (ANOVA) with Bonferroni’s or Tukey’s post hoc tests or Student’s t tests were performed when appropriate using GraphPad Prism (GraphPad Software, La Jolla, CA). Results were considered statistically significant if recorded p-values were less than 0.05.
Results
ZBP1 functions as an HSV-1 restriction factor in primary microglia
ZBP1 has previously been demonstrated to act as an HSV-1 restriction factor in peripheral myeloid cells and primary murine astrocytes (Guo et al. 2018; Jeffries et al. 2022). Here, we have investigated the ability of this sensor to limit infection in microglia derived from ZBP1+/+ and ZBP1-/- mice. Lack of ZBP1 protein expression was confirmed in microglia derived from ZBP1 knockout mice by immunoblot analysis (Supplemental Figure S1). ZBP1+/+ and ZBP1-/- derived microglia were infected with a clinically-derived neuroinvasive strain of HSV-1, HSV-1(MacIntyre), and the number of PFU released from infected cells was determined by conventional plaque assays in Vero cells as performed previously (Jeffries et al. 2020, 2022). As shown in Fig. 1A, infectious viral particle release by ZBP1-deficient microglia was significantly greater than that seen with wild-type cells.Fig. 1. Interferon-independent restriction of HSV-1 replication in primary murine microglia by ZBP1. Panel (A) Microglia derived from ZBP1-expressing (ZBP1+/+) or ZBP1-deficient (ZBP1-/-) mice were infected with HSV-1(MacIntyre) at an MOI of 2 for 1 h. Following infection, cells were either untreated or exposed to 10 µM fludarabine. Supernatants were collected at 24h post-infection and standard plaque assays were performed in Vero cells to determine the number of plaque-forming units (PFU) released from HSV-1(MacIntyre) infected microglia. Panel (B) Murine microglia derived from ZBP1-expressing or ZBP1-deficient mice were uninfected (Mock) or infected with HSV-1(MacIntyre) at an MOI of 0.2 or 2. At 24 h post-infection, the concentration of IL-6, IL-18, TNF, and IFN-β, were quantified by ELISA. Data are shown as the mean of 3–6 independent experiments ± SEM. An asterisk indicates a significant difference from similarly treated ZBP1+/+ cells and dagger symbols indicate a significant difference from uninfected cells (p < 0.05; n = 3)
To determine whether the higher levels of viral release were due to a reduction in the release of antiviral mediators, we measured IFN-β secretion by microglia following HSV-1 infection. As shown in Fig. 1B, both ZBP1+/+ and ZBP1-/- derived microglia produced minimal levels of IFN-β production, with statistically significant amounts only being seen in ZBP1-/- derived microglia with HSV-1 infection at an MOI of 0.2. Furthermore, we determined that treatment of either ZBP1+/+ or ZBP1-/- derived microglia with the STAT1 inhibitor, fludarabine, had no effect on infectious particle release (Fig. 1A). In addition, we assessed the effect of genetic ZBP1 deficiency on HSV-1-induced release of the inflammatory cytokines IL-6, TNF, IL-1b, and IL-18, by these cells. We report that ZBP1+/+ derived microglia released significant, albeit modest, levels of TNF at an MOI of 2.0, and ZBP1 deficient microglia released low but significant levels of IL-6 at MOIs of 0.2 and 2, and TNF at an MOI of 2.0 following HSV-1 infection (Fig. 1B). Furthermore, this production was significantly different from that seen with ZBP-1 expressing microglia following infection at an MOI of 2.0 (Fig. 1B). In contrast, HSV-1 infection failed to elicit detectable release of IL-1b by either ZBP1+/+ or ZBP1-/- microglia (data not shown) and significance production of IL-18 was only observed by ZBP1 deficient cells at an MOI of 2.0 (Fig. 1B). As such, these data are inconsistent with ZBP1-mediated viral restriction being due to differences in cytokine production.
HSV-1 infection induces necroptosis in microglia
Several studies have shown that ZBP1 can mediate necroptosis in non-CNS cell types (Upton et al. 2012; Thapa et al. 2016; Kuriakose et al. 2016; Maelfait et al. 2017; Koehler et al. 2017; Guo et al. 2018; Jiao et al. 2020; Yang et al. 2020) and, more recently, in astrocytes (Jeffries et al. 2022). To determine if this also occurs in microglia, we measured the final percentage of cell death at 24 h and the rate of cell death in microglia derived from ZBP1+/+ and ZBP1-/- mice following HSV-1 infection with clinically-derived HSV-1 (HSV-1(MacIntyre) and a laboratory strain (HSV-1(F). We found that there was significant difference between ZBP1+/+ and ZBP1-/- derived microglia in the rate and final percentage of virus-induced cell death at 24 h following challenge with both HSV-1(MacIntyre) (Figs. 2 and 3A) and HSV-1(F) (Fig. 2). Importantly, we found that levels of pMLKL, the executor of necroptosis, were significantly increased in ZBP1+/+ derived microglia following infection with HSV-1(MacIntyre) (MOI of 0.2) and that pMLKL levels were significantly lower in similarly infected ZBP1 deficient cells (Fig. 3B).Fig. 2HSV-1-induced cell death in primary murine microglia is mediated by ZBP-1. ZBP1-expressing and ZBP1-deficient microglia were infected with a clinical strain of HSV-1 (HSV-1(MacIntyre)) or a laboratory strain (HSV-1(F)). One-hour post-infection, cells were treated with a vehicle control (DMSO), the RIPK1 inhibitor GSK547 (50 nM), the RIPK3 inhibitor GSK843 (2 µM), and/or the caspase 8 inhibitor Z-IETD-FMK (20 µM). Cell viability was measured with a RealTime-Glo™ MT assay every 2 h post-infection starting at hour 2. Data are reported as the final percentage of cell death at 24 h relative to non-infected cells and as the rate of cell death. Data are shown as the mean of 3–5 independent experiments ± SEM. Asterisks indicate a significant difference from similarly treated ZBP1-expressing cells, while dagger symbols indicate significant difference from similarly challenged cells treated with DMSO vehicle only (p < 0.05)Fig. 3HSV-1-induced necroptosis occurs in both a RIPK1-independent ZBP1-mediated and a RIPK-mediated ZBP1-independent manner in murine microglia. ZBP1-expressing and ZBP1-deficient microglia were infected with HSV-1(MacIntyre). One-hour post-infection, cells were treated with DMSO as a vehicle control (Panels **A **and B) or GSK547 (50 nM) (Panels **C **and D). Panels **A and C: Cell viability was measured with a RealTime-Glo™ MT assay every 2 h post-infection, starting at hour 2. Panels B **and D: 24 h post-infection, whole-cell lysates were collected and analyzed for pMLKL expression or β-actin by immunoblot analysis. Relative pMLKL expression was determined by densitometric analysis and normalized to β-actin expression to correct for changes in total protein loading attributable to cell death. Representative Western blots and the mean densitometric values ± SEM of 3 independent experiments are shown. An asterisk indicates a significant difference in final cell death or pMLKL expression from similarly treated ZBP1-expressing cells and dagger symbols indicate a significant difference from vehicle-treated cells (p < 0.05)
To determine if RIPK1 mediates necroptosis in ZBP1-deficient microglia following infection, we treated ZBP1+/+ and ZBP1-/- derived microglia with the RIPK1 inhibitor, GSK547, during infection with the HSV-1(MacIntyre) and HSV-1(F) strains. As shown in Figs. 2 and 3C, there remained a significant difference between ZBP1-expressing and ZBP1-deficient microglia in the rate and final percentage of virus-induced cell death at 24 h following challenge with HSV-1(MacIntyre) in the presence of GSK547. However, the rate of cell death was significantly lower for both ZBP1-expressing and ZBP1-deficient microglia, and the difference in the final percentage cell death between GSK547 treated versus untreated HSV-1(MacIntyre) infected ZBP1-deficient microglia approached significance (28.78 ± 6.42 for vehicle versus 11.16 ± 3.92 for GSK547 treated, p = 0.081, n ≥ 3).
A significant difference between ZBP1-expressing and ZBP1-deficient microglia also remained in rate and final cell death in cells infected with HSV-1(F) laboratory strain in the presence of GSK547, but both the rate and final percentage of cell death was lower than vehicle treated cells regardless of the expression of ZBP1 (Fig. 2). Interestingly, HSV-1(MacIntyre)-induced increases in pMLKL levels in ZBP1-expressing microglia were not significantly altered by GSK547-treatment while pMLKL expression was almost entirely abolished in GSK547-treated ZBP1 deficient cells (Fig. 3D). Together, this data suggests that HSV-1-induced necroptosis in murine microglia may occur in both a ZBP1-dependent manner that is RIPK1 independent, and a ZBP1-independent manner that is RIPK1 dependent.
ZBP1 mediates necroptotic, but not apoptotic, death pathways in HSV-1(MacIntyre) challenged microglia
Since both RIPK1- and ZBP1-mediated necroptosis have been shown to require RIPK3 activity to phosphorylate MLKL (Upton and Kaiser 2017; Jiao et al. 2020), we have inhibited RIPK3 with the inhibitor GSK843 to determine whether necroptosis is the primary mechanism underlying HSV-1-induced cell death. We have found that a significant difference remained between ZBP1-expressing and ZBP1-deficient microglia in the rate of death and the final percentage of virus-induced cell death at 24 h following challenge with the neuroinvasive HSV-1(MacIntyre) strain with RIPK3 inhibitor treatment (Figs. 2 and 4A), despite a lower level of pMLKL expression (Fig. 4B) that was statistically different between treated and untreated cells ZBP1-expressing cells at an MOI of 0.2 (0.82 -/+ 0.16 versus 0.24 -/+ 0.12 in untreated and GSK843 treated cells, respectively, p < 0.05, n = 3). Similarly, a significant difference remained in the rate of cell death between ZBP1+/+ and ZBP1-/- derived cells infected with the laboratory HSV-1(F) strain following treatment with GSK843 (Fig. 2).Fig. 4ZBP1-mediated necroptosis is the primary cell death pathway occurring in HSV-1-infected murine microglia. ZBP1-expressing and ZBP1-deficient microglia were infected with HSV-1(MacIntyre). One-hour post-infection, cells were treated with either GSK843 (2 µM; Panels A and B) or Z-IETD-FMK (20 µM; Panels **C **and D). Panels A and C; Cell viability was measured with a RealTime-Glo™ MT assay every 2 h post-infection, starting at hour 2. Panels **B **and D) 24 h post-infection, whole-cell lysates were collected and analyzed for phosphorylated MLKL expression or β-actin by immunoblot analysis. Relative phosphorylated MLKL expression was determined by densitometric analysis and normalized to β-actin expression to correct for changes in total protein loading attributable to cell death. Representative immunoblots and the mean densitometric values ± SEM of 3 independent experiments are shown. An asterisk indicates a significant difference in final cell death or P-MLKL expression from similarly treated ZBP1-expressing cells and dagger symbols indicate a significant difference from vehicle-treated cells (p < 0.05)
Importantly, GSK843 treatment significantly reduced HSV-1(MacIntyre)-induced cell death percentage and rate in both ZBP1-expressing and ZBP1-deficient microglia compared to vehicle treated cells (Fig. 2). Similarly, ZBP1 expressing microglia infected with HSV-1(F) and treated with GSK843 showed a significantly lower final percentage of cell death and rate of cell death versus infected untreated cells (Fig. 2). Taken together, these data indicate that HSV-1 induces necroptosis in murine microglia in both a ZBP1-dependent and a ZBP1-independent manner.
However, it has recently been shown that ZBP1 additionally mediates apoptosis in murine astrocytes (Jeffries et al. 2022). To determine whether ZBP1-mediated microglial cell death also occurs via apoptosis, we performed parallel experiments using the caspase-8 inhibitor Z-IETD-FMK. As shown in Figs. 2 and 4C and Z-IETD-FMK had no significant effect on either the rate or final percentage of cell death in ZBP1-expressing microglia challenged with HSV-1(MacIntyre). Furthermore, a statistically significant difference remained in the rate and final percentage of cell death between ZBP1+/+ and ZBP1-/- derived cells following infection with this clinical strain the presence of Z-IETD-FMK (Figs. 2 and 4C). Interestingly, despite the lack of statistical significance in final cell death percentage, the kinetics of cell death in ZBP1-deficient microglia following infection with HSV-1(MacIntyre) were significantly lower in comparison to untreated infected cells (Figs. 2 and 4C). In contrast to the results obtained with the clinically-derived viral isolate, caspase-8 inhibition significantly reduced both the rate and final percentage of cell death in ZBP1-expressing microglia challenged with laboratory adapted HSV-1(F) and abolished the ZBP1-mediated differences in cell death seen with this strain (Fig. 2).
Recent studies have indicated that caspase-8 inhibition can promote RIPK1 activation, leading to necroptotic cell death (Degterev et al. 2005). To assess this possibility, we measured phosphorylated MLKL protein levels in HSV-1(MacIntyre) infected microglia following caspase-8 inhibition. As shown in Fig. 4D, both ZBP1+/+ and ZBP1-/- derived microglia showed statistically similar levels of phosphorylated MLKL following HSV-1 infection in the presence of Z-IETD-FMK. These results indicate that ZBP1-independent HSV-1(MacIntyre)-induced necroptosis in murine microglia can occur in the absence of caspase activity. Taken together, our data indicate that ZBP1-mediates necroptosis, but not apoptosis, in primary murine microglia infected with this clinically-derived HSV-1 isolate.
To determine if RIPK1 activation is responsible for the pMLKL detected in HSV-1(MacIntyre) infected microglia in the presence of a caspase-8 inhibitor, we concurrently treated cells with Z-IETD-FMK and the RIPK1 inhibitor GSK547. Interestingly, we found that co-inhibition had no significant effect on final cell death percentage in ZBP1-expressing microglia infected with HSV-1(MacIntyre) and that the induced rate and final percentage of cell death remained significantly different between ZBP1-deficient and ZBP1-expressing microglia (Figs. 2 and 5A). This suggests that, in the absence of caspase-8, RIPK1 can mediate ZBP1-independent HSV-1(MacIntyre)-induced necroptotic cell death. In contrast, co-inhibition of RIPK1 and caspase-8 significantly reduced both the rate and final percentage of cell death in ZBP1-expressing microglia challenged with laboratory adapted HSV-1(F) and abolished the ZBP1-mediated differences in cell death seen with this strain (Fig. 2). These data indicate that ZBP1-independent cell death occurs in HSV-1(F) infected microglia via both RIPK1 and caspase-8 dependent mechanisms.Fig. 5HSV-1-induced necroptosis is mediated by RIPK3 in murine microglia. ZBP1-expressing and ZBP1-deficient microglia were infected with HSV-1(MacIntyre). One-hour post-infection, cells were treated with a combination of GSK547 (50 nM) and Z-IETD-FMK (20 µM) (Panel A) or a combination of GSK843 (2 µM) and Z-IETD-FMK (20 µM) (Panel B). Cell viability was measured every two hours with a RealTime-Glo™ MT assay beginning at two hours following infection. Data are shown as the mean of 4–6 independent experiments ± SEM. Asterisk in Panel A indicates a significant difference in final cell death percentage from similarly treated ZBP1+/+ cells (p < 0.05)
Since RIPK3 inhibition significantly reduced the percentage of cell death of ZBP1-expressing microglia at 24 h following challenge with HSV-1(MacIntyre) but caspase-8 inhibition did not (Fig. 2), we hypothesized that RIPK3-mediated necroptosis is the principal mechanism underlying cell death in these cells following infection. To confirm this hypothesis, we simultaneously treated HSV-1(MacIntyre) infected microglia with GSK843 and Z-IETD-FMK. As shown in Figs. 2 and 5B, the HSV-1(MacIntyre)-induced rate and final cell death percentage of in ZBP1-expressing microglia was reduced to almost the same degree to that seen with the RIPK3 inhibitor alone, and the kinetics of cell death between ZBP1-expressing and ZBP1-deficient microglia remained significantly different when simultaneously treated with RIPK3 and caspase-8 inhibitors. Taken together, we propose that this neuroinvasive HSV-1 strain induces both ZBP1-dependent, and independent, RIPK3-mediated necroptosis in murine microglia that could serve to restrict DNA virus replication.
In contrast, and similar to the results obtained with simultaneous RIPK1 and caspase-8 inhibition, co-inhibition of RIPK3 and caspase-8 significantly reduced both the rate and final percentage of cell death in ZBP1-expressing microglia challenged with laboratory adapted HSV-1(F) and abolished the ZBP1-mediated differences in cell death seen with this strain (Fig. 2). These data indicate that this laboratory HSV-1 strain can induce an additional ZBP1-independent but RIPK3-dependent cell death pathway occurring in murine microglia.
Discussion
HSV-1 is a highly successful neurotropic DNA virus that accounts for nearly 95% of all fatal cases of sporadic encephalitis (Bulakbasi et al., 2008). The damage associated with HSV encephalitis can be attributed to an overproduction of inflammatory mediators and/or out-of-control viral replication (Conrady et al. 2010). As such, it is vitally important to understand the mechanisms underlying the early immune responses to HSV-1 infection within the CNS and to determine whether they are beneficial or, alternatively, contribute to disease pathology. We have previously demonstrated that murine glia express ZBP1 and that it mediates, at least in part, their inflammatory responses to HSV-1 that can result in neuronal cell damage (Furr et al. 2011; Crill et al. 2015). More recently, we have shown that ZBP1 can serve to restrict viral replication in HSV-1 infected murine astrocytes via the induction of both necroptotic and apoptotic cell death (Jeffries et al. 2022). In the present study, we have extended our understanding of the role of ZBP1 in murine glia with the demonstration that this cytosolic sensor restricts DNA virus production/release in primary microglia. We show that ZBP1 deficient microglia release significantly more infectious viral particles following challenge with a clinical neuroinvasive HSV-1 isolate (Krinke and Dietrich 1990). Similar to our previous studies in astrocytes (Jeffries et al. 2022), this ability cannot be explained on the basis of reductions in IFN-b or inflammatory cytokine production as ZBP deficient microglia release more rather than less of these cytokines following infection, and pharmacological inhibition of STAT1 fails to significantly affect infectious viral particle release.
ZBP1 has been identified as an important mediator of necroptosis associated with DNA and RNA virus infections (Kuriakose et al. 2016; Guo et al. 2018; Zhang et al. 2020; Jiao et al. 2020; Yang et al. 2023; Wang et al. 2024). Furthermore, we have recently described the ability of HSV-1 to induce ZBP-1-dependent necroptosis in primary murine astrocytes (Jeffries et al. 2022). Here, we show that this cytosolic DNA sensor can similarly mediate necroptotic cell death in primary murine microglia following infection with either a clinically-derived neuroinvasive HSV-1 isolate or a laboratory adapted viral strain. This is based on the finding that virally-induced cell death and levels of phosphorylation of the necroptotic pathway component MLKL are significantly reduced in the absence of ZBP1 expression or following RIPK3 inhibition. Furthermore, we have found that there also appears to be an additional ZBP1-independent mechanism underlying HSV-1-induced necroptosis in murine microglia, perhaps via TNFR/RIPK1-mediated signaling. This ZBP-1 independent pathway is supported by the ability of both RIPK1 and RIPK3 inhibition to reduce cell death in ZBP1 deficient cells, and the demonstration that RIPK3 inhibition can also decrease pMLKL levels in these cells. Prior studies have indicated that caspase-8 can act as a negative regulator of RIPK1 mediated necroptosis via the interaction of the RHIM domains of RIPK1 and RIPK3 (Silke et al. 2015; Pasparakis and Vandenabeele 2015; Weinlich et al. 2017; Grootjans et al. 2017). However, in the present study pharmacological inhibition of caspase-8 failed to significantly alter pMLKL levels in ZBP1-deficient cells following infection with HSV-1(MacIntyre). Together, these data indicate that HSV-1 infection can induce necroptotic cell death in murine microglia via both ZBP1 dependent and independent pathways (as shown in Fig. 6) in a similar manner to that seen in other cell types (Wang et al. 2014, 2024; Guo et al. 2018; Lee et al. 2021; Jeffries et al. 2022; Rashidi et al. 2024). It should be noted, however, that future follow-up time course studies to assess the induction of necroptotic markers prior to 24 h post-infection are desirable since HSV-1 can complete its infection cycle within this timeframe.Fig. 6. Proposed mechanisms for the activation of cell death pathways in murine microglia following challenge with a clinical and a laboratory strain of HSV-1. Infection with HSV-1(MacIntyre) or HSV-1(F) leads to the release of viral DNA and recognition by ZBP1 and its subsequent interaction with RIPK3 mediated by their respective RHIM domains, which results in MLKL phosphorylation and necroptotic cell death. Additionally, HSV-1(MacIntyre) and HSV-1(F) infection causes the production and release of TNF, that can act in an autocrine or paracrine manner to initiate ZBP1-independent necroptosis via RIPK1 activation, in the absence of caspase-8, and its subsequent interaction with RIPK3. In contrast to the clinical strain, TNF activity induced by HSV-1(F) laboratory strain infection may induce apoptosis via caspase-8
We have previously reported that ZBP1 can also mediate apoptotic cell death in HSV-1 infected astrocytes (Jeffries et al. 2022). In contrast, we have found that the rate and final percent cell death of ZBP1-expressing microglia following challenge with a clinical HSV-1 isolate are insensitive to caspase-8 inhibition. Studies featuring co-treatment with both RIPK3 and caspase-8 inhibitors further supports the notion that ZBP1 mediates HSV-1(MacIntyre)-induced necroptosis in microglia via RIPK3, but not apoptosis, as the inhibition of both failed to decrease cell death below that seen following RIPK3 inhibition alone. As such, it appears that cell type specific differences exist in the mechanisms underlying HSV-1 associated cell death in microglia and astrocytes. Furthermore, we have found a virus strain-dependent difference in microglial cell death pathways elicited by HSV-1 as both the rate and final percent cell death of ZBP1-expressing microglia following exposure to a laboratory adapted HSV-1 strain was sensitive to caspase-8 inhibition. However, such apoptotic death appears to be ZBP1 independent as no further decrease in cell death was seen in the absence of ZBP1 expression or following inhibition of either RIPK1 or RIPK3. The activation of a ZBP1-independent caspase-8 mediated apoptotic pathway in microglia challenged with a laboratory HSV-1 strain, but not a clinical isolate (as shown in Fig. 6), further illustrates cell type specific differences between microglia and astrocytes in the mechanisms underlying HSV-1 induced glial cell death.
It is presently unclear whether the viral stain-specific differences in microglial cell death pathways reflect differences in viral entry and/or host cell recognition of each or represent an evasion adaptation of the neuroinvasive strain. For example, wild-type HSV-1 is known to enter cells by interactions between HSV glycoprotein D and membrane receptors such as herpesvirus entry mediator A and nectin-1, but laboratory strains of HSV-1 with defined mutations in glycoprotein D instead use nectin-2 (Krummenacher et al. 2004). It is therefore possible that such genetic mutations have occurred in HSV-1(McIntyre) or HSV-1(F) that result in differences in invasion of murine microglia and/or subsequent detection by their PRRs. Alternatively, the strain differences in the cell death pathways initiated in microglia may reflect an adaptation of the neuroinvasive strain that may serve to limit death of this resident CNS cell type, as HSV-1 is recognized to possess numerous evasion strategies to prevent detection and antiviral response activation by host cells upon infection (Su et al. 2016; Tognarelli et al. 2019; Zhu and Zheng, 2020; Zhang et al. 2021; Fukai et al., 2023). Regardless, of whether these strain-specific differences are beneficial or detrimental to the host, the present findings lend support to the use of clinically relevant viral isolates in the study of viral pathogenesis, and it will be interesting to see if such differences are similarly seen between other clinical and laboratory HSV-1 isolates.
Taken together, our data suggests that ZBP1 serves as a sensor for DNA viruses, such as HSV-1, in primary murine microglia and can mediate the induction of necroptotic cell death to limit infectious viral particle production. It should be noted that, while we have previously shown that cGAS and IFI16 do not elicit antiviral responses to HSV-1 in human microglia-like cells (Jeffries et al. 2020), we cannot rule out that the involvement of other nucleic acid sensors. Thus, it is possible that these sensors, such as AIM2, cGAS, IFI16, or RIG-I, may act independent of, or in cooperation with, ZBP1 in murine microglial responses to HSV-1. For example, a study has identified an ability of ZBP1 to interact with AIM2 to elicit a novel caspase-1-mediated cell death pathway (PANoptosis) in murine bone marrow-derived macrophages following HSV-1 infection (Lee et al. 2021). However, the observed HSV-1-mediated elevations in pMLKL levels, shown in Fig. 3B and D, and the failure of this virus to elicit demonstrable production of IL-1β or IL-18 production by ZBP1 expressing cells are not consistent with the activation of such a pathway in mouse microglia. In addition, the relative importance of ZBP1-mediated responses in host defense against DNA viruses or damaging neuroinflammation in vivo remains to be determined, and it is presently unclear whether such mechanisms also occur in primary human microglia in response to HSV-1 infection.
Supplementary Information
Below is the link to the electronic supplementary material.
Supplementary Material 1