Cellular receptors: Hunting for the culprit

Orthoflaviviruses utilize receptor-mediated endocytosis to gain access to host cells. It is proposed that attachment molecules recruit and/or concentrate virus particles at the surface of the cell, thereby facilitating engagement with a primary receptor and eventual endocytosis. Numerous putative receptors have been identified in mammalian systems including C-type lectins (DC-SIGN, L-SIGN, mosGCTL-1/3/7, mannose receptor), phosphatidylserine receptors (TIM/TAM families, CD300a), integrins receptors (αvβ3, αvβ5), glycosaminoglycans (heparan sulfate, heparan sulfate proteoglycan), laminin receptors (heat shock proteins, 37/67 kDa receptor), and claudin-1/prohibitin-1 [5,6]. These findings suggest that orthoflaviviruses have evolved mechanisms to leverage a broad spectrum of molecules to interact with mammalian cells (Fig 1). It should be noted that many of these putative receptors have been shown to mediate virus attachment to host cells, but not necessarily internalization which is a prerequisite for bona fide receptors. Notably, mosquitoes lack orthologues for many of these molecules.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Orthoflaviviruses exploit multiple putative receptors to initiate infection.

Sankey plot highlighting the diversity of putative receptors (left) associated with orthoflavivirus infection (right) and the experimental systems from whence they were identified across diverse cellular systems (center). The highlighted flows represent the putative receptors identified in mosquito vector studies (magenta) and their corresponding orthologs (light blue) detected in insect cells and/or human cells [5–7]. The figure was generated Microsoft Excel using the SankeyArt GmbH plug-in (https://www.sankeyart.com/excel/).

https://doi.org/10.1371/journal.ppat.1013617.g001

Similarly, several mosquito receptors have been shown to facilitate viral entry in both mosquito cell lines and Aedes aegypti. Key examples include prohibitin which interacts with DENV, CLEC5A, and mosGCTL to mediate JEV, DENV, and ZIKV attachment, and a C-Type lectin in collaboration with a CD45 phosphatase homolog which facilitate WNV infection in mosquitoes [6,7]. Moreover, additional attachment factors (50, 67, and 80 kDa proteins) present in C6/36 cells and the midgut of Ae. aegypti influence DENV attachment in a serotype-specific manner [5]. It has recently been reported that a 31 kDa protein serves as a receptor for DENV2 in Ae. aegypti mosquitoes [8]. Interestingly, this protein was found to localize intracellularly, and partial knockout or dsRNA suppression of this gene had no effect on infection prevalence. Prevalence was only reduced upon dsRNA suppression in the heterozygous knockout line [8]. These data suggest a role for this 31 kDa protein during infection, but fail to definitively validate it as a bona fide receptor. Moving forward, it will be important that putative receptors, in both vertebrate and invertebrate systems (Fig 1), be validated with overexpression experiments in nonpermissive cells/hosts, knockout experiments, binding assays, and inhibition assays.

Orthoflaviviruses, like all arthropod-borne (arbo)viruses, are maintained in a transmission cycle between two highly divergent hosts, namely the invertebrate vector and vertebrate host. Consequently, the utilization of a receptor conserved between insect and mammalian hosts would be evolutionarily advantageous. Emerging evidence suggests that alphaviruses such as Western equine encephalitis virus, Semliki Forest virus, Sindbis virus, and Eastern equine encephalitis virus utilize low-density lipoprotein receptor (LDLR), very low-density lipoprotein receptor, and apolipoprotein E receptor 2 and their invertebrate orthologues as receptors for entry into mammalian and mosquito cells [9,10]. Similarly, members of the Flaviviridae, including hepatitis C virus (HCV; Hepacivirus), DENV, and JEV, have been shown to utilize lipoprotein receptors, specifically LDLR and scavenger receptor class B type 1, to gain entry into mammalian cells [11,12]. The conservation of lipoprotein receptors from mammals to mosquitoes, coupled with the affinity of Flaviviridae and Togaviridae for these proteins, raises the possibility that lipoprotein receptors may be important in mediating orthoflavivirus infection of mosquitoes. While viruses can utilize multiple receptors to enter different cell types, it would seem evolutionarily advantageous for viruses infecting disparate hosts like vertebrates and invertebrates to utilize a shared or highly similar molecule as a receptor in both hosts (Fig 2A). Future studies characterizing lipoprotein receptor-orthoflavivirus interactions could reveal new targets for the development of novel vector control interventions.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Multifactorial influences on mosquito midgut infection.

An illustration depicting arbovirus midgut infection barrier in Aedes aegypti mosquitoes. It is known that midgut infection is initiated through only a handful of cells, but the reasons why are unclear. Here, we highlight five poorly defined factors potentially involved in mediating orthoflavivirus infection of mosquitoes. (A) Orthoflaviviruses utilize receptor-mediated endocytosis to infect cells; however, a bona fide receptor has yet to be identified, and it is unclear if such a receptor is homogeneously expressed across the midgut epithelium. (B) The mosquito midgut contains four primary cell types and numerous subpopulations, and it is unknown whether orthoflavivirus utilize a specific cell type to initiate midgut infection. (C) sNS1 is highly conserved and has been shown to enhance viral infection in mosquitoes, but the mechanism of enhancement is unresolved. Evidence suggests that sNS1 may act as a bridging molecule facilitating interactions between extracellular virus particles and the midgut epithelium. (D, E) Wolbachia-mediated inhibition of orthoflavivirus infection of mosquitoes is well described, but it is unclear if inhibition occurs at the point of infection, during replication, or during cell-to-cell spread. Further, it is unknown if Wolbachia infection of the midgut epithelium is homogenous (D) or variable across cell types (E) and the implications of this on orthoflavivirus infection and inhibition. The image was Created in BioRender. Brackney, D. (2025) https://biorender.com/dkr7waj.

https://doi.org/10.1371/journal.ppat.1013617.g002

Midgut cellular diversity and viral tropism

Single-cell RNA sequencing (scRNAseq) technology has revolutionized our understanding about the cellular composition of the mosquito midgut. The utilization of scRNAseq to profile cellular heterogeneity in mosquito tissues indicates that a noninfectious blood meal induces dramatic changes in midgut cellular composition [2,13]. The midgut is composed of roughly 5,000–10,000 cells which are organized into several epithelial cell types including the undifferentiated progenitors (intestinal stem cells, and the enteroblasts (EB)), absorptive enterocytes (EC), and the secretory enteroendocrine cells (EE) [2,13]. The ECs constitute nearly 70% of the total cell population. Based on scRNAseq data, EC and EB cell populations can be further delineated into distinct subtypes such as EC-like-1, EC-like-2, EC-like-3, dEB-1, and dEB-2, among others [2,13]. The origin of these cell subtypes is unknown. It is hypothesized that changes in the midgut transcriptional profile due to shift in diet from sugar to blood could result in the formation of transient differentiating cells that occur as distinct metabolic states. Investigations of the interaction between midgut epithelial cells and viral pathogens suggest that EC-like-2 cells are the most susceptible to West Nile virus infection, while Zika virus is more commonly associated with EC and EE cells [1,14]. These studies provide an early indication of a possible link between cell types and virus infection (Fig 2B); however, these investigations only focused on established infections (4 and 12 dpi) and not early time points during the initial stages of infection (i.e., 6–8 hpi) [1,14]. This is significant because during an infection, extracellular viruses infect a susceptible cell and then spread cell-to-cell forming foci. These two modes of transmission are fundamentally different and, therefore, insights gained from established infections may not be relevant to the initial infection events. Further, it is known that the acquisition of a blood meal can drastically shift the cellular composition and metabolic state of the midgut meaning cell populations present at the time of infection may not be the same as those present days later after the blood meal has been digested. Examining earlier time points will be critical to determining if orthoflaviviruses preferentially initiate infection of a specific cell type or metabolic state.

Gut feeling: How NS1 supercharges midgut infections

The envelope glycoprotein of orthoflaviviruses (e.g., DENV, ZIKV, WNV, and YFV) has long been considered the viral factor responsible for host cell attachment and penetration; however, recent work suggests that the multifunctional and highly conserved NS1 may also play an important role in initiating infection [15,16]. NS1 is essential for virus replication and has both intracellular (iNS1) and secreted (sNS1) forms. The intracellular dimeric iNS1 plays a critical role during virus replication by remodeling ER membranes resulting in the formation of viral replication compartments [17]. The secreted form is not critical to viral replication but has been shown to directly interact with components of the innate immune system (complement factors and toll-like receptors) and modulate aspects of the host’s innate immune response [18]. Further, it can directly interact with endothelial cells resulting in increased expression/activation of cathepsin L, heparinase, and sialidase which can cause disruption of the endothelial glycocalyx layer inducing hyperpermeability [19]. It has also been shown that clinically relevant serum levels of sNS1 can enhance orthoflavivirus infection of the mosquito midgut [20,21]. Mechanistically, it was reported that enhancement occurs through sNS1 suppression of ROS production within the midgut epithelial cells. At 18 hours post sNS1 exposure (hpe), it was found that H₂O₂ levels were decreased, as well as the expression levels of dual oxidase and NADPH oxidase transcripts in the midguts that were exposed sNS1 [20]. Interestingly, similar results were not observed during the early events of midgut infection (i.e., 4 and 8 hpe). Because virus infection of the midgut occurs within hours of blood meal acquisition, the proposed model implies that virus particles enter cells and begin replicating, but the infection is ultimately cleared upon ROS induction which is subsequently quenched by sNS1 allowing for infection to occur more efficiently. While this study and others have demonstrated that elevated levels of ROS at the point of infection can correlate with decreased infection rates, it is unclear if increased ROS after viral entry has any effect on viral replication [20]. In fact, studies have found that viruses across multiple viral families including the Flaviviridae can induce and benefit from an induced ROS state [22]. While decreased levels of ROS were observed at 18 hours postNS1 exposure, it is possible that midgut infection could have occurred at much earlier time points prior to any reductions in ROS [20]. An alternative hypothesis is that sNS1 is acting as a soluble bridge between specific cellular attachment molecules and the virus (Fig 2C), thereby enhancing infection in both mammalian and insect cells [23]. In fact, it has been shown that sNS1 can directly interact with virus particles and that sNS1 can enhance orthoflavivirus attachment to both mammalian and mosquito cells [23]. Further investigation into the mechanisms involved in sNS1-mediated enhancement are warranted.

Silent gatekeeper: Wolbachia inhibition of Orthoflavivirus infection of the midgut

Wolbachia pipientis is an obligate intracellular bacterium that exists in different strains with varied ability to render transinfected mosquito populations resistant to orthoflavivirus infection. The mechanism(s) of inhibition are not fully resolved, but host immune priming, resource competition, and metabolic alterations are thought to contribute to this phenotype [24]. Further, there are many strains of Wolbachia each inducing varying degrees of inhibition. For example, wMel strongly inhibits DENV2 infection of the Ae. aegypti midgut whereas wAlbB has an intermediate inhibitory effect [25]. While numerous studies have demonstrated that wMel inhibits infection, this observation has been made at later time points post virus exposure, and it is therefore impossible to discern if inhibition occurs at the point of infection or during cell-to-cell spread [26]. It is known that Wolbachia bacteria can stably reside in midgut tissue, but it is unclear if all cells are equally infected at similar densities or if Wolbachia reside in specific cell types within the midgut epithelium, and how this affects the inhibitory phenotype. For instance, it has been shown that Wolbachia-mediated virus inhibition can occur in a density-dependent manner; however, these studies assessed Wolbachia densities at a tissue or organismal level and not the cellular level or were completed in cell culture [27–29]. If Wolbachia densities are heterogeneous across midgut epithelial cells this could affect the ability of orthoflaviviruses to establish midgut infections by limiting the number of susceptible cells or by preventing initiated infections from spreading cell to cell (Fig 2D). Alternatively, Wolbachia spp. or specific Wolbachia strains could preferentially infect specific cell types within the midgut epithelium. If the preferentially infected cell type is important for viral infection it could result in variable infection phenotypes (Fig 2E). Understanding these diverse cellular processes may further explain the complex interactions between Wolbachia and mosquito hosts, thereby elucidating how different strains influence host biology, physiology, and vector competence. To address these unknowns, future studies should focus on analyzing the midguts of orthoflavivirus-exposed Wolbachia-transinfected mosquitoes across a series of early time points (e.g., 6, 12, 18, and 24 hpi) through advanced signal amplifying immunofluorescence techniques (e.g., hybridization chain reaction) and scRNAseq.

Concluding remarks

The recent application of high-throughput screens and single-cell sequencing technologies has significantly advanced our understanding of virus-vector interactions; however, the critical viral and/or host factors mediating the initial infection events remain unresolved. Future work should focus on identifying and validating potential receptors, most likely orthologous genes present in both the vertebrate host and invertebrate vector, clarifying the role of specific cellular population/subpopulations during infection, determining the mechanism by which sNS1 enhances midgut infection, and how Wolbachia effects the early stages of midgut infection. The results from these studies will be crucial for developing novel control strategies that disrupt viral infection of the mosquito midgut.