FHL1 promotes chikungunya and o’nyong-nyong virus infection and ...

Abstract

Arthritogenic alphaviruses are positive-strand RNA viruses that cause debilitating musculoskeletal diseases affecting millions worldwide. A recent discovery identified the four-and-a-half-LIM domain protein 1 splice variant A (FHL1A) as a crucial host factor interacting with the hypervariable domain (HVD) of chikungunya virus (CHIKV) nonstructural protein 3 (nsP3). Here, we show that acute and chronic chikungunya disease in humans correlates with elevated levels of FHL1. We generated FHL1^−/− mice, which when infected with CHIKV or o'nyong-nyong virus (ONNV) displayed reduced arthritis and myositis, fewer immune infiltrates, and reduced proinflammatory cytokine/chemokine outputs, compared to infected wild-type (WT) mice. Interestingly, disease signs were comparable in FHL1^−/− and WT mice infected with arthritogenic alphaviruses Ross River virus (RRV) or Mayaro virus (MAYV). This aligns with pull-down assay data, which showed the ability of CHIKV and ONNV nsP3 to interact with FHL1, while RRV and MAYV nsP3s did not. We engineered a CHIKV mutant unable to bind FHL1 (CHIKV-ΔFHL1), which was avirulent in vivo. Following inoculation with CHIKV-ΔFHL1, mice were protected from disease upon challenge with CHIKV and ONNV, and viraemia was significantly reduced in RRV- and MAYV-challenged mice. Targeting FHL1-binding as an approach to vaccine design could lead to breakthroughs in mitigating alphaviral disease.

Introduction

Arthritogenic alphaviruses are a large group of positive-strand RNA viruses known to cause acute and chronic musculoskeletal disease^1,2. Notable examples, including chikungunya virus (CHIKV), o'nyong-nyong virus (ONNV), Ross River virus (RRV) and Mayaro virus (MAYV), cause recurrent episodic outbreaks and have shown the capacity to cause widespread epidemics^3,4,5. Despite their debilitating impact on human health and potential for global emergence, no licensed therapies or vaccines currently exist^6,7.

Alphaviruses encode four replicase proteins named nonstructural proteins 1-4 (nsP1-nsP4). NsP1, nsP2 and nsP4 form the membrane-bound core of viral RNA replicase. NsP3 is indispensable for viral replication and is known to interact with numerous host proteins⁸. Host factors such as Ras-GAP SH3 domain binding proteins (G3BPs), amphiphysin⁹, Y-box-binding protein 1 (YBX1)¹⁰, heat shock proteins¹⁰ and NAP1L1/4¹¹ interact with alphavirus nsP3^12,13,14,15. The C-terminal hypervariable domain (HVD) of nsP3 has been shown to be important for host protein interaction, however, the role of nsP3 HVD in alphavirus pathogenesis remains unclear^11,16,17,18.

Four-and-a-half LIM domain protein 1 (FHL1, also known as SLIM1) is a unique member of the LIM protein family that was recently discovered to play a key role in CHIKV and ONNV RNA replication¹⁹. FHL1 is involved in multiple cellular processes. Mutations in the FHL1 gene cause many distinct heart and musculoskeletal diseases, which have been categorized into FHL1-related myopathies or FHL1opathies²⁰. Notable examples include reducing body myopathy, Emery-Dreifuss muscular disorder (EDMD), myofibrillar myopathy (MFM), X-linked recessive hypoparathyroidism, and hypertrophic cardiomyopathy. Many of these diseases are progressive, and left untreated, have various implications for the quality of life of the affected individual, eventually resulting in death^{21,22,23,24,25}. Three major FHL1 splice variants – FHL1A, FHL1B and FHL1C – are found in humans. FHL1A is most commonly found in skeletal and cardiac muscle and fibroblasts, whereas FHL1B and FHL1C are found mostly in muscle, brain and testes^26,27,28,29. FHL1A was identified as the splice variant that plays a role in CHIKV and ONNV infections. It was reported that the deletion of FHL1 led to inhibition of CHIKV and ONNV infection, but did not affect other alphaviruses, such as RRV and MAYV, or flaviviruses. Notably, FHL1 was found to interact directly with CHIKV nsP3 HVD. The study further showed the cells from a patient with EDMD, deficient in FHL1, were resistant to CHIKV infection. Furthermore, based on limited in vivo studies using 9-day-old genetically engineered mice that lacked FHL1 (FHL1^−/y), Meertens et al. could not detect CHIKV replication in these mice. Histologically, there were no signs of necrosis and reduced cellular infiltrates compared to WT mice¹⁹. Interestingly, Lukash et al. recently reported that the FHL1-nsP3 HVD interaction may not be a prerequisite of viral replication, as CHIKV was still able to replicate in cells deficient for FHL1³⁰. Our results agree with Lukash et al. where we were able to detect replication of CHIKV, albeit attenuated in CHIKV-infected FHL1^−/− mice. The role of FHL1 in the development of alphavirus diseases has not been fully characterized.

In our study, we demonstrate for the first time that FHL1 levels are increased in patients with acute and chronic chikungunya disease. We also generated FHL1 knockout (FHL1^−/−) mice on the C57BL/6J background and used them to examine the role of FHL1 in arthritogenic alphavirus disease. Knockout of FHL1A without affecting FHL1B/C is technically difficult, as the three isoforms have heavily overlapping exons. In mice infected with either CHIKV or ONNV, joint swelling was diminished in FHL1^−/− compared to wild-type (WT) mice. Additionally, data obtained from histopathological analysis, immunoassay and mass cytometry (CyTOF) demonstrated reduced lesions, lower expression of proinflammatory cytokines and reduced cellular infiltrates in infected FHL1^−/− mice compared to WT mice. In contrast, FHL1^−/− mice infected with either RRV or MAYV did not show reduced disease symptoms. We carried out a pull-down assay, which demonstrated a distinct interaction between FHL1 and the nsP3s of CHIKV and ONNV, but no such interaction was observed with the nsP3s of RRV and MAYV. Additionally, we created a CHIKV mutant (CHIKV-ΔFHL1) unable to interact with FHL1 (confirmed by pull-down assay) to test its pathogenicity and efficacy as a universal vaccine candidate against arthritogenic alphaviruses. CHIKV-ΔFHL1 infection was highly attenuated in Vero cells, showing significantly reduced replication kinetics compared to CHIKV-WT. Disease signs were significantly reduced in CHIKV-ΔFHL1 inoculated mice compared to CHIKV-WT infected, with CHIKV-ΔFHL1 inoculation offering protection from disease development upon subsequent challenge with CHIKV, ONNV and, to a degree, RRV and MAYV.

Results

FHL1 was upregulated during CHIKV infection

Viral infection leads to a range of changes in host factors to facilitate viral replication¹⁶. To determine whether levels of FHL1 are increased during CHIKV infection, FHL1 protein levels in serum samples from chikungunya patients were quantified. Serum samples collected from chikungunya patients with either acute or chronic disease, thirty-nine and seventeen respectively, were analyzed by ELISA; samples from healthy individuals were used as controls (Supplementary Table 1 and 2). FHL1 protein levels were significantly higher in the serum of both acute and chronic chikungunya patients compared to the healthy controls (Fig. 1a). Similar observations were made in a mouse model of acute CHIKV disease. Mice were infected with CHIKV-WT or mock-infected subcutaneously in the right footpad. At 1-, 3- and 5-days post-infection (dpi), blood was collected via tail bleeding, and the ipsilateral quadriceps were harvested at 7 dpi. The quadriceps were processed for wholemount immunofluorescence staining of FHL1 (the antibody was produced by GenScript Biotech based on Q13642-1, UniProt). We observed an increased FHL1 antibody signal in the quadriceps of infected mice compared to mock-infected mice (Fig. 1b), recapitulating our observations in CHIKV patients. To investigate if the elevated levels of FHL1 at the sites of inflammation is detectable on a transcription level, total RNA was extracted from quadriceps and FHL1 mRNA expression analyzed by qRT-PCR. A significant increase in FHL1 mRNA was observed in the ipsilateral quadriceps of CHIKV-infected mice at 7 dpi (Fig. 1c); a time which coincides with the second peak of foot swelling (Fig. 2a). Moreover, CHIKV-infected mice exhibited increased serum levels of FHL1 across all time points (1, 3 and 5 dpi) compared to mock-infected mice (Fig. 1d). These results indicate that expression of FHL1 is elevated as a result of CHIKV infection.

**Fig. 1: FHL1 levels are increased in CHIKV infected humans and mice.**

**Fig. 2: FHL1 facilitates CHIKV disease development and viral replication in mice.**

The severity of CHIKV disease was less pronounced in FHL1^−/− mice

To define the role of FHL1 in CHIKV disease pathogenesis, WT and FHL1^−/− mice were inoculated subcutaneously in the footpad with 10⁴ PFU of CHIKV or mock-infected with PBS. CHIKV-infected FHL1^−/− mice showed significantly reduced foot swelling compared to infected WT mice (Fig. 2a). Reduced CHIKV titers were observed in the serum of FHL1^−/− mice compared to that of WT mice at 1-, 3-, and 5 dpi (Fig. 2b). Furthermore, in both ipsilateral and contralateral quadriceps and ankle joints, CHIKV-infected FHL1^−/− mice had significantly reduced virus titers compared to WT mice at 3 and 7 dpi (Fig. 2c–f).

Hematoxylin and eosin (H&E) staining of the quadriceps and ankles was used to enumerate cell infiltrates in these tissues. In mock-infected WT and FHL1^−/− mice, comparable cellular infiltrates were observed in the quadriceps (Fig. 3a, b) and ankle joints (Fig. 3g, h). Compared to CHIKV-infected WT mice, infected FHL1^−/− mice showed reduced cellular infiltrates in the ipsilateral quadriceps at 3- and 7 dpi (Fig. 3c–f). Interestingly, no significant differences were found at 3 dpi in the ipsilateral ankles of CHIKV-infected FHL1^−/− and WT mice (Fig. 3i, j). However, at 7 dpi, infected WT mice showed significantly higher cellular infiltrates in the ipsilateral ankles than infected FHL1^−/− mice (Fig. 3k, l).

**Fig. 3: FHL1 facilitates cell infiltration in the quadriceps and ankles of mice with CHIKV disease.**

Additionally, the ankles of CHIKV-infected mice collected at 3 and 7 dpi were stained with safranin-O to assess differences in articular cartilage thickness. For mock-infected mice, the articular cartilage thickness was comparable between WT and FHL1^−/− mice (Supplementary Fig. 2a, b). Ipsilateral ankles of CHIKV-infected FHL1^−/− and WT mice also exhibited similar articular cartilage thickness at 3 dpi (Supplementary Fig. 2c, d). In contrast, a notable reduction in articular cartilage thickness was observed in the ipsilateral ankle of infected WT mice compared to infected FHL1^−/− mice at 7 dpi (Supplementary Fig. 2e, f). Taken together, these results demonstrate FHL1 is necessary for the development of acute CHIKV disease in mice.

CHIKV-infected FHL1^−/− mice exhibit reduced levels of proinflammatory cytokines and limited immune cell infiltration

Next, we examined the effects of FHL1 knock-out on the cytokine profile and immune response following CHIKV infection in mice. The levels of proinflammatory cytokines and chemokines, including G-CSF, GM-CSF, IFN-γ, IL-1β, IL-6, IL-12p70, IL-17A, CCL-2, CCL-3, CCL-4, CCL-5, CXCL1 and TNF-α in the ipsilateral quadriceps were significantly lower in FHL1^−/− mice compared to WT mice both at 3 and 7 dpi (Fig. 4). This was also observed for Eotaxin, IL-1α, IL-2, IL-3, and IL-12p40 levels at 7 dpi, but not at 3 dpi. Interestingly, the anti-inflammatory IL-10 and the allergy-associated cytokines, IL-9 and IL-13, were also significantly lower in FHL1^−/− than WT mice at both time points (Fig. 4a).

**Fig. 4: Knock-out of FHL1 reduces levels of proinflammatory cytokines and chemokines in CHIKV-infected mice.**

The ipsilateral quadriceps, popliteal draining lymph nodes (DLNs) and spleens of CHIKV-infected FHL1^−/− and WT mice were collected and examined with Mass Cytometry and fluorescent flow cytometry. There were no significant differences in major immune cell populations observed between WT and FHL1^−/− mice at 0 or 3 dpi in the ipsilateral quadriceps (Fig. 4b) or DLNs (Supplementary Fig. 3a), while a reduction in some populations was detected in spleens of FHL1^−/− mice at 3 dpi (Supplementary Fig. 3b). In contrast, at 7 dpi, macrophages, monocytes, natural killer (NK) cells, CD4⁺ and CD8⁺ T cells were all significantly reduced in the ipsilateral quadriceps (Fig. 4b), DLNs (Supplementary Fig. 3a) and spleen (Supplementary Fig. 3b) of FHL1^−/− mice, compared to WT mice.

Mass cytometry data of DLNs and spleens were processed by Uniform Manifold Approximation and Projection (UMAP) for dimension reduction and visualization. Clusters of major cell types were identified on the UMAP plots based on the heatmaps of the expression levels of fundamental cell markers, such as CD3, CD4, CD8, CD11b, CD19, etc. Position shifts on the UMAP plots represent phenotype changes of the cells (Supplementary Fig. 4a, f). Cell markers that principally contributed to CHIKV-induced shifts included CCR7, Ly6A/E (Sca-1) and Ly6C in various cell types. In the DLNs, these surface antigens were found to be increased on T and B lymphocytes, as well as CD11b⁺ myeloid cells in both CHIKV-infected FHL1^−/− and WT mice (Supplementary Fig. 4b–e). However, while cells in the infected WT mice tended to maintain relatively high levels of CCR7, Ly6A/E and Ly6C expression over time, levels were significantly lower in the infected FHL1^−/− mice, with this difference greater at 7 dpi (Supplementary Fig. 4b–e). The differential expression between CHIKV-infected FHL1^−/− and WT mice was also observed in splenocytes at 7 dpi, although interestingly, in contrast to the DLNs, Ly6C expression was significantly higher in B cells and CD11b⁺ myeloid cells at 3dpi (Supplementary Fig. 4g–j). The profile of these cell surface antigens is in line with the lower disease symptoms observed in CHIKV infected FHL1^−/− mice. Furthermore, the higher levels of Ly6C in the infected FHL1^−/− mice suggests a potential novel role for Ly6C and monocytes in the early stage of CHIKV disease pathogenesis.

We further investigated whether recruited leukocytes in infected mouse muscle contributed to elevated FHL1 expression in this tissue during disease using immunofluorescence staining. Several CD4, Ly6C/G and CD11b cell clusters in the myofiber interstitial space expressing high levels of FHL1 were observed, indicating that myeloid cells such as monocytes and macrophages were important, albeit not majorly, for secreted FHL1 during peak CHIKV disease (Supplementary Fig. 5).

Impact of FHL1 on other arthritogenic alphavirus infections in vivo

From the observations above, we hypothesized that FHL1 might also play a significant role in infection and disease of other clinically important arthritogenic alphaviruses such as RRV, MAYV and ONNV.

To determine if FHL1 expression is increased during RRV infection, we collected serum and plasma from patients with confirmed RRV disease and from healthy individuals (Supplementary Table 3) and quantified FHL1 protein levels by ELISA. Similar to the results observed in samples from chikungunya patients, RRV disease patient serum and plasma samples exhibited significantly higher levels of FHL1 than samples from healthy controls (Fig. 5a, b). Interestingly, however, in a mouse model of acute RRV disease, we did not observe any significant differences in the disease score between RRV-infected WT and FHL1^−/− mice (Fig. 5c, d). Consistent with this observation, viral titers in serum samples collected at 1-, 3- and 5-dpi (Fig. 5e), and those in quadriceps and ankle samples at 3- and 7-dpi (Fig. 5f, g), were comparable between RRV-infected WT and FHL1^−/− mice. Similar observations were made using a mouse model of MAYV disease; disease severity and virus titers in the serum, quadriceps and ankles were all comparable in MAYV-infected WT and FHL1^−/− mice (Fig. 5h-k). In contrast, upon ONNV infection, we observed a significant reduction in foot swelling in FHL1^−/− mice compared to WT mice. The reduction in foot swelling in FHL1^−/− mice was noted as early as 3 dpi and significant differences were observed up to 10 dpi (Fig. 5l). Consistent with this observation, FHL1^−/− mice exhibited reduced ONNV titers in the serum, ipsilateral quadriceps, and ankles compared to WT mice at all time points analyzed (Fig. 5m–o).

**Fig. 5: Impact of FHL1 on RRV, MAYV and ONNV infection.**

Additionally, the replication kinetics of CHIKV, RRV, ONNV, and MAYV in primary murine fibroblasts derived from WT and FHL1^−/− mice was analyzed using multistep growth curves. The obtained data was consistent with our in vivo data. Replication of CHIKV and ONNV was reduced in FHL1^−/− fibroblasts compared to WT fibroblasts, whereas RRV and MAYV replication kinetics were similar in both cell types (Fig. 6a–d). Next, we fused HVD of nsP3 of RRV, MAYV, CHIKV and ONNV to EGFP, expressed these fusion proteins in U2OS cells and performed a pull-down assay experiment. It was found that while nsP3s of all four viruses interacted with G3BP1, an essential host co-factor for replication of arthritic alphaviruses³¹, only nsP3s of CHIKV and ONNV were able to interact with FHL1 (Fig. 6e). Combined, our results indicate that FHL1 is essential for CHIKV and ONNV infection and disease development, but not for RRV or MAYV.

**Fig. 6: FHL1 is essential for the replication of CHIKV and ONNV.**

CHIKV mutant defective for interaction with FHL1 shows attenuated replication in Vero cells and offers protection from subsequent alphavirus challenge in mice

In previous studies, deletions in nsP3 HVD (R₁₆₈₆-P₁₇₉₅ and A₁₇₉₆-D₁₈₃₉; here and other places residues numbered from the beginning of nonstructural polyprotein) led to attenuation of Semliki Forest virus (SFV) virulence in mice³². In 2014, Hallengard et al. showed CHIKV-3del5, a CHIKV mutant with a large deletion in nsP3 by substituting E¹⁶⁵⁶ to G¹⁷¹⁷ with a linker (amino acid sequence AYRAAAG), exhibited reduced replicative capacity and elicited high immunogenic protection in mice against CHIKV challenge, making it a promising CHIKV vaccine candidate (phase III clinical trial NCT04546724 completed)³³. The critical amino acids mediating the interaction between nsP3 of CHIKV and FHL1 were reported to be V₁₇₄₅(or P₁₇₅₆) to P₁₇₈₇^19,30, which are located within HVD of nsP3 (aa residues 1653–1863). We therefore hypothesized that mutations in nsP3 of CHIKV, preventing its interaction with FHL1, should lead to reduced replication and attenuation in disease and that such an attenuated variant could be explored as a potential live-attenuated vaccine candidate against CHIKV infection.

First, mutations substituting six amino acids (F1762G, F1763G, T1780R, A1781S, Q1785G and A1786T) were introduced into a construct expressing EGFP fused with HVD of nsP3 of CHIKV (CHIKV-ΔFHL1); a matching set of mutations was also introduced into a construct expressing EGFP fused with HVD of nsP3 of ONNV (ONNV-ΔFHL1). Pull-down assays performed with these constructs revealed that the introduced mutations completely abolished interaction of HVDs of CHIKV and ONNV with FHL1 while interactions with G3BP1 were not affected (Fig. 6e). The differences in the amino acid sequences of nsP3 HVD between CHIKV-WT, CHIKV-ΔFHL1 and CHIKV-3del5 are shown in Supplementary Fig. 6a. Similar to CHIKV-3del5, CHIKV-ΔFHL1 replicated to significantly lower titers in Vero cells compared to WT CHIKV, indicating a high degree of attenuated replication (Supplementary Fig. 6b). To assess the disease-causing capacity of CHIKV-ΔFHL1, WT mice were infected subcutaneously in the right footpad with 10⁴ PFU CHIKV-WT, CHIKV-ΔFHL1 or CHIKV-3del5, and monitored daily for CHIKV-induced footpad swelling. At 3 dpi, the first peak of foot swelling was observed in all the three groups of infected mice. However, beginning from 5 dpi until 11 dpi, a significant reduction in foot swelling was observed in CHIKV-3del5 and CHIKV-ΔFHL1-infected mice, compared to CHIKV-WT-infected mice (Fig. 7a). Furthermore, CHIKV-ΔFHL1 and CHIKV-3del5 replicated to a significantly lower titer in quadriceps muscle at day 3 post infection, compared to CHIKV-WT (Fig. 7b). At day 7 post infection, the titer of CHIKV-ΔFHL1 and CHIKV-3del5 were below the detection limit (Fig. 7c). These results suggest that, similar to CHIKV-3del5, CHIKV-ΔFHL1 is highly attenuated in vivo and may therefore have potential as a vaccine candidate.

**Fig. 7: CHIKV-ΔFHL1 is attenuated and protects vaccinated mice against infection of different alphaviruses.**

Next, we sought to determine the protective effects of vaccination with CHIKV-ΔFHL1. WT mice were inoculated subcutaneously with 10⁴ PFU of CHIKV-ΔFHL1 in the right footpad or were mock-vaccinated. At 16 days post vaccination, the mice were subsequently challenged with 10⁴ PFU of CHIKV, RRV, MAYV or ONNV. CHIKV-, MAYV- and ONNV-challenged mice were monitored for footpad swelling daily. Disease scores were not analyzed for RRV-challenged mice as this model requires mice to be 21 days of age upon infection to observe clinical disease signs^34,35,36. Our data demonstrates that CHIKV-ΔFHL1 inoculated mice did not exhibit any CHIKV- or ONNV- induced foot swelling upon challenge (Fig. 7d, g) and had significantly reduced disease signs during MAYV challenge (Fig. 7j). Additionally, there was no detectable viremia in CHIKV-ΔFHL1 inoculated mice challenged with CHIKV (Fig. 7e, f), or ONNV at 1- and 3-days post-challenge (Fig. 7h, i). In CHIKV-ΔFHL1 inoculated mice challenged with MAYV (Fig. 7k, l) or RRV (Fig. 7m, n), while viremia was observed, virus titers were significantly reduced (by 0.5 – 0.8 log₁₀), compared to control mice challenged with these viruses. These results indicate that CHIKV-ΔFHL1 vaccination can protect mice from arthritogenic alphavirus-induced disease and viremia.

The humoral immune response, particularly neutralizing antibodies, are an important component of the host defense mechanism providing protection against alphaviruses^37,38,39,40. Thus, the neutralizing capacity of antibodies induced by vaccination is indicative of the efficacy of the vaccine. Hence, we examined the neutralizing capability of serum from CHIKV-ΔFHL1-inoculated mice collected at 14 days post-vaccination using plaque reduction neutralization test (PRNT) against CHIKV, RRV, MAYV or ONNV. A 50% PRNT (PRNT₅₀) value for CHIKV at ~1:2500 dilution was detected, whereas the PRNT₅₀ was at 1:160 to 1:320 for ONNV, at 1:40 to 1:80 for RRV, and <1:20 for MAYV (Fig. 7o). Thus, PRNT₅₀ values correlated with observed protection against different alphaviruses. These results suggest that the neutralizing antibodies induced by CHIKV-ΔFHL1 are not only able to protect against CHIKV infection, but also offer cross-protection against other alphaviruses, including closely related ONNV as well as more distantly related RRV and MAYV.

Discussion

Previous studies have identified several host factors, including FHL1, G3BPs, glycosaminoglycans, T-cell immunoglobulin, mucin 1 and MXRA8 facilitate entry and replication of alphaviruses^{12,13,14,15,19}. However, despite extensive efforts, many gaps remain in our current understanding of the role(s) that host factors play in alphavirus infection and pathogenesis.

In the current study, we sought to characterize the role of FHL1 in the pathogenesis of alphavirus diseases. We found that FHL1 levels were significantly higher in the serum of acute/chronic CHIKV disease and RRV disease, which, to the best of our knowledge, represents the first report showing the elevation of FHL1 upon alphavirus infection. Presumably, the high levels of FHL1 are derived from the breakdown of infected muscle. The level of FHL1 in the patient serum was higher in the acute phase of the disease than in the chronic phase. These observations suggest FHL1 may have potential to be used as a marker CHIKV disease diagnosis.

Previously, our study identified that pentraxin 3 (PTX3), a host factor that activates complement pathways and facilitates pathogen recognition by macrophages and DCs, was highly induced in CHIKV and RRV disease patients^40,41. With this present study showing upregulation of FHL1, both of our studies suggest a novel strategy by which alphaviruses are likely to facilitate infection by upregulating certain host factors which are critical for virus replication. The underlying mechanisms and the repercussions for disease pathology warrant further investigation.

Our knowledge regarding the pathogenesis of chronic CHIKV disease is still limited. Persistence of viral replication in the joint tissues has been hypothesized to contribute to chronic CHIKV disease. It is therefore tempting to speculate that the ongoing high levels of FHL1 in chronic CHIKV infection may facilitate the persistence of infection which leads to chronic disease development.

Notably, FHL1 has been reported to increase muscle fiber size and oxidative slow fiber type expression, which are both associated with muscle development and strength⁴². Thus, the increased FHL1 levels in the serum of CHIKV and RRV disease patients, apart from facilitating virus replication, may also reflect tissue recovery. FHL1 therefore may have a dual role in alphavirus disease pathogenesis, which warrants further investigation. Whether increased levels of FHL1 are a host-driven response to muscle damage, a response to soluble factor(s) from infiltrating leukocytes, or are directed by the virus, as seen in flavivirus infection to immunopathological effect^43,44,45, remains to be investigated.

A previous study identified FHL1A as a key host factor required for optimum CHIKV and ONNV infections in vitro¹⁹ while another study demonstrated it to be important but not critical³⁰. While these findings are not necessarily mutually exclusive, they leave open the question of the importance of FHL1 for CHIKV infection and pathogenesis. Here, we comprehensively assessed the role of FHL1 in CHIKV infection and disease by employing FHL1-deficient mice in our well-established immunocompetent adult mouse model of infection that mimics CHIKV human disease (i.e. arthritis and myositis)³⁵. FHL1 deficiency did not eliminate the ability of CHIKV to replicate in vitro or in vivo. However, it was associated with significant reduction in CHIKV replication and foot swelling in mice. Inflammatory monocytes, macrophages and T cell infiltrates promote acute arthritogenic alphavirus disease^34,46. Indeed, infected mice without FHL1 presented with lower cellular infiltrates and milder inflammation in the quadriceps and ankles than infected WT mice. The reduction in disease signs and cellular infiltrates were likely due to the inability of CHIKV to replicate efficiently which, in turn, is likely a consequence ...

Search This Blog

Mucus