Members of Venezuelan Equine Encephalitis complex entry into host cells by clathrin-mediated endocytosis in a pH-dependent manner

Pixuna virus (PIXV) and Río Negro virus (RNV) are mosquito-borne alphaviruses belonging to the Venezuelan Equine Encephalitis (VEE) complex, which includes pathogenic epizootic and enzootic subtypes responsible for life-threatening diseases in equines. Considering that the first steps in viral infection are crucial for the efficient production of new progeny, the aim of this study was to elucidate the early events of the replication cycle of these two viruses. To this end, we used chemical inhibitors and the expression of dominant-negative constructs to study the dependence of clathrin and endosomal pH on PIXV and RNV internalization mechanisms. We demonstrated that both viruses are internalized primarily via clathrin-mediated endocytosis, where the low pH in endosomes is crucial for viral replication. Contributing knowledge regarding the entry route of VEE complex members is important to understand the pathogenesis of these viruses and also to develop new antiviral strategies.


Introduction
The Venezuelan equine encephalitis (VEE) antigenic complex belongs to the Togaviridae family, which include the Alphavirus genus. They constitute a group of arboviruses responsible for human and veterinary disease outbreaks around the world 1 , they share similar genetic characteristics and can be defined by their broad antigenic cross-reactivity 2 . The VEE complex comprises New World viruses that cause encephalitis in humans and horses and are classified in six antigenic subtypes divided into epizootic (IAB and IC) and enzootic subtypes mediated uptake 15 . In this sense, and considering that the entry route of different enzootic variants of the VEE complex remain poorly explored, it is important to describe the mechanisms of entry and uncoating of these viruses. Therefore, the aim of the present research was to elucidate the early events of the replication cycle of PIXV and RNV, two enzootic members of the VEE complex. To this end, we used chemical inhibitors and the expression of dominant-negative constructs to study the dependence of clathrin and endosomal pH on PIXV and RNV internalization mechanisms. Agents that disrupt clathrindependent receptor-mediated endocytosis such as Sucrose (SCR) and Chlorpromazine (CPZ) were used. SCR was necessary to generate a hypertonic medium to disturb the formation of clathrin vesicles on the cell membrane [16][17][18][19][20][21] . CPZ was used as it prevents clathrin-coated pit formation on the plasma membrane by promoting clathrin lattice assembly on endosomal membrane [22][23][24][25] . The lysosomotropic agents, Glucosamine (GlcN) and ammonium chloride (NH4Cl) were used to elevate acidic organelles pH owing to the presence of their amino group 10,26-30 . The above mentioned compounds have been widely used to decrease viral replication in several enveloped viruses 31- 35 . We demonstrate that both viruses are internalized primarily via clathrin-mediated endocytosis, where the low pH in endosomes is crucial for viral replication. The results reported here show new insights regarding VEE complex members entry route and infection. In addition, these findings provide a better understanding of the molecular mechanisms underlying their replication cycle pointing to attractive targets for the development of new strategies to prevent infections.

Clathrin is required for PIXV and RNV entry into Vero cells
Considering that the ability of alphaviruses to enter into the host cell via alternative pathways has already been reported [13][14][15]36 , we evaluated the dependence of PIXV and RNV entry into Vero cells by clathrin-mediated endocytosis. To this end, we first employed chemical inhibitors, widely used for clathrin-mediated endocytosis, such as SCR 16-21 and CPZ 22-25 .
According to the literature and to the possible drug-induced cytotoxic effects, cell culture viability in the presence of MTT was assayed at different concentrations of these inhibitors ( Figure 1A). Thus, we defined 0.2 M for SCR and 25 µM for CPZ as the experimental conditions (cell viability ≥ 90%). Besides, the inhibitory action of both drugs on clathrinmediated endocytosis was effectively assessed determining TRITC-transferrin internalization reduction. The fluorescence inside the cell cytoplasm observed in cultures treated with SCR or CPZ was significantly lower (64.6% and 55.9%, respectively; p < 0.001) compared to control cells (100%, Figure 1B). After PIXV or RNV internalization, in the presence of drugs, we collected the supernatants at 4, 8 and 24 hours postinfection (hpi) and quantified extracellular infectious particles by plaque assay (plaque-forming unit, PFU/ml). Infected cultures with PIXV (multiplicity of infection: MOI 0.1) and treated with 0.2 M SCR or 25 uM CPZ showed a significant decrease in PUF/ml of about 2 log10 and 1-2 log10 at 8 and 24 hpi, respectively. In the same way, infected cultures with RNV (MOI 0.1) and treated with both drugs displayed a significant reduction of virus particles in the extracellular medium of about 1-2 log10 and 2 log10 at 8 and 24 hpi, respectively ( Figure 1C). As we expected, no differences in extracellular virus yield was observed at the end of the eclipse phase (4 hpi) 37 for either viruses. Additionally, cell cultures were treated and infected with PIXV or RNV (MOI 10). At 8 hpi, cultures were fixed and processed for immunofluorescence in order to calculate the percentage of infected cells. In the representative image, the culture shows a visible reduction of green fluorescence labeling after treatment with both inhibitors in comparison to mocktreated cells ( Figure 1D). Cultures treated with SCR or CPZ showed a significant reduction in 7 unit, PFU/ml). D. Representative images of cultures treated with 0.2 M SCR or 25 µM CPZ and infected with PIXV or RNV (green) at MOI 10 and fixed at 8 hpi; nuclei (blue). E.
Percentage of infected cells with PIXV or RNV against control (100% value). Data represent the mean ± SD from 3 independent experiments. Significance was calculated by one-way ANOVA test, followed by Tukey post-hoc analysis to enable specific group comparisons, with p value ≤ 0.05 considered as statistically significant. **p < 0.01; ***p < 0.001. Scale bar: 20 μm.
To support the participation of the clathrin pathway in PIXV and RNV entry, we employed wild type (WT) and dominant negative (DN) mutant constructs of the clathrin-coat associated cellular protein Eps15, a required adaptor for clathrin-mediated endocytosis 38

PIXV and RNV infection requires active endosomal acidification
Endosomal vesicles may play a crucial role in virus infection, providing the appropriate condition for conformational changes in envelope viral proteins, resulting in the formation of a fusion pore and release of the viral nucleocapsid into the cytosol 39 . Considering that the endocytic pathway is highly regulated and involves several intermediate organelles, it is important to examine the fate of viral proteins after their internalization in early and late endosomes. The Rab family of small GTPases is essential for membrane trafficking and the classification of endocytic loads into specific subcellular compartments 40 . In this sense, we performed a quantitative colocalization analysis of PIXV and RNV with Rab5-GFP and Rab7-GFP, as early and late endosome markers, respectively 41 ( Figure 3A). Using Pearson's colocalization coefficient 42 , we observed significant colocalization PIXV and RNV proteins with Rab5-GFP and Rab7-GFP ( Figure 3A). After 8 hpi, a great coefficient of colocalization was found for both endosomal compartments and the viral proteins (PIXV: r= 0.68 ± 0.11 for Rab5 and r= 0.78 ± 0.06 for Rab7; RNV: r=0.69 ± 0.1 for Rab5 and r= 0.75± 0.07 for Rab7) ( Figure 3B). These results show that PIXV and RNV associate to early and late endosomes at the early stage of the infectious cycle. To study the low pH-dependence of PIXV and RNV replication, we first evaluated the effect of two lysosomotropic agents, GlcN and NH4Cl, inhibitors of endosomal vesicles acidification 10,26-33 , in Vero cells. Cell culture viability by MTT was assayed at different concentrations of either of these two inhibitors ( Figure 4A). Thus, we defined 15 mM GlcN and 25 mM NH4Cl as the experimental conditions (cell viability ≥ 90%). To assure the effect of both compounds in intracellular acidic vesicles, we pretreated Vero cells with the drugs plus acridine orange. Cultures treated with NH4Cl showed significant differences in the fluorescence intensity of the cells cytoplasm (56.3 %; p < 0.001) compared to controls (100%). However, this effect was not observed in culture cells treated with GlcN, where acridine orange fluorescence inside the cell cytoplasm did not display significant differences (87.1%) compared to the control condition ( Figure 4B). We then performed the cytoplasmic accumulation assay with neutral red, a very sensitive probe for lysosomal microenvironments and a biological pH indicator 43 . In figure 4C, a 40% and 80% inhibition of the neutral red signal can be observed in cultures treated with GlcN and NH4Cl, respectively, compared to untreated cells. Therefore, the participation of endosomal pH in PIXV and RNV replication in  Percentage of infected cells with PIXV or RNV against controls (100% value). Data represent the mean ± SD from 3 independent experiments. Significance was calculated by one-way ANOVA test, followed by Tukey post-hoc analysis to enable specific group comparisons, with p value ≤ 0.05 considered as statistically significant. *p < 0.05; **p < 0.01; ***p < 0.001. Scale bar: 20 μm.

Discussion
Alphaviruses constitute a conserved group of arboviruses responsible for human and Semliki Forest virus fused with early endosomes, and VEEV was found to fuse with maturing and/or late endosomes 10-12 . In CHIKV, membrane fusion was observed predominantly with Rab5-positive endosomes and often occurred within 40 s after delivery to endosomes 53 . To further confirm our observations, we employed lysosomotropic agents to interfere with endosome acidification that are extensively used as an assay for alphaviruses entry 10,54 . As expected, PIXV and RNV required a progressive pH reduction in endosomes in order to release the viral nucleic acid into the cytosol.
In conclusion, this study demonstrates, for the first time, that PIXV and RNV partially enter via clathrin-dependent endocytosis into Vero cells. Then, both viruses require transport to early and late endosomes, where the acidic pH-dependent step results in the release of PIXV or RNV RNA into the cytoplasm for a successful infection (Supplementary Fig.1). This study contributes to our understanding of the entry process in enzootic subtypes from the VEE complex. Considering that arthropod borne viruses are significant sources of disease in man and domestic animals, this research extends the knowledge of virus entry into host cells and provide a better understanding of the molecular mechanisms underlying the viral internalization process. Ultimately, this knowledge would be useful to develop potent antivirals that could block the virus at the early stages of infection. Briefly, Vero cells were plated and incubated at 37 °C, 5% CO2 and 98% humidity. Confluent monolayers of Vero cells were infected at different multiplicity of infection (MOI); MOI 0.1 for standard plaque assays and MOI 10 and for immunofluorescence assays. Virus adsorption was performed at 4 °C per 1 h. After this first incubation, the inoculum was removed and monolayers were washed with PBS at 4 °C to eliminate any free virus particle, then DMEM with 1% FCS was added and the infected cultures were incubated at 37 °C, 5% CO2 and 98% humidity for 8, 12 or 24 h, depending on the experiment. The cytopathic effect was evaluated under direct visualization by phase contrast microscopy at low magnification using an Olympus IX81 microscope.

Virus titration by plaque assay
Infected Vero cell monolayers with serial dilutions of each culture supernatant collected from the different assays were overlaid with 1.5% ultra-pure agarose (Gibco BRL) in 4% DMEM 2x. Two days later, the cells were fixed with 10% formaldehyde in water and the overlaid medium was removed. The monolayers were stained with 1% crystal violet in water and the plaques were counted. Infectivity titers were expressed as PFU/ml. All the experiments with viruses were performed in a laminar flow hood under biosafety level 2 conditions.

Drug treatments
Sucrose (SCR) was used to generate a hypertonic medium to disturb the formation of clathrin vesicles on the cell membrane. Under this condition, clathrin-dependent endocytosis is inhibited 16-21 . Chlorpromazine (CPZ) was used in this study as an inhibitor of clathrinmediated endocytosis. CPZ specifically inhibits clathrin-coated pit formation on the plasma membrane 22-25 . The lysosomotropic agents, Glucosamine (GlcN) and ammonium chloride (NH4Cl) were used to elevate the pH of acidic organelles owing to the presence of its amino group 10,26-30 . Those compounds have been widely used to decrease viral replication in several enveloped viruses 31-33 . All compounds were purchased from Sigma-Aldrich and were dissolved in phosphate-buffered saline (PBS) to prepare master-stocks for storage.
To evaluate the efficiency of blockade of each endocytic pathway by the different inhibitors, uptake studies with TRITC-transferrin (Transferrin From Human Serum, Alexa Fluor™ 647 Conjugate #T23366, Thermo Fisher Scientific Inc., Waltham, Massachusetts, USA) for SCR and CPZ, neutral red and acridine orange for NH4Cl and GlcN were performed. We used those strategies because transferrin and its receptor are taken into cells through clathrin-mediated endocytosis, a mechanism that has been extensively characterized 58 . Neutral red and acridine orange are dyes commonly used as probes to measure the internal pH in acidic vesicles 59 . Untreated cells were used as control.
Cells either treated with inhibitors or not, were incubated with 15 μg/ml TRITC-transferrin or 100 μg/ml acridine orange (A6014, Sigma) in serum free medium during 15 min at 37°C.
Then, cells were processed for visualization in a fluorescence microscope as described below.
For the quantification of fluorescence intensity of TRICT-transferrin or orange acridine, images of 5 microscope fields, selected at random, were acquired per condition in each independent experiment. Using the FIJI software (https://fiji.sc/), fluorescence intensity was calculated and data were normalized against fluorescence of controls (100% value).
Background intensity was subtracted from images prior to quantification 60 .
To determine inhibition by lysosomotropic agents, we incubated cultured cells with drugs at a selected concentration plus 50 mg/ml neutral red (NR), as a biological lysosomal pH indicator. Cells were seeded in 100 μl per well of transparent 96-well microtiter plates.
Following drug treatments, the cell culture medium was replaced with medium containing 50 μg/ml NR dye and drugs. Following incubation for 3 h at 37 °C and 5% CO2 in a humidified environment and two washes with PBS, intracellular NR dye was extracted using a destaining solution (50% ethanol, 1% acetic acid and water) and absorbance was measured at 540 nm using a spectrophotometer (Biotek, Elx800). Percentage of inhibition of each drug was calculated normalizing data against the negative control absorbance (100% value).
To evaluate the effect of different drugs on PIXV and RNV infection, Vero cell monolayers were seeded into 6-well plates or 24-well plates and pretreated with drugs, as listed before, for 30 min at 37 °C. Synchronized infection of PIXV or RNV was achieved by infecting the cells at MOI of 0.1 or 10 at 4 °C for 1 h in the presence of the compounds. Then, the inoculum was removed and DMEM was added with different inhibitors throughout all designed experiments. In cultures infected at MOI 0.1, the supernatant was collected at 4, 8 and 24 hpi to determine extracellular viruses by a titration assay. Monolayers infected with MOI 10 were fixed at 8 hpi to proceed to an immunofluorescence assay to determine intracellular viruses.
Briefly, Vero cells were seeded at the density of 1 × 10 6 cells/well in a 96-well plate. After 24 h, cells were treated with drugs at different concentrations for 24 h, untreated cells served as control. After treatment, cells were washed with PBS and 100 μl MTT assay was mixed with 900 μl PBS into each well, followed by incubation at 37 °C for 2 h. The medium was then removed, washed, and 1 ml DMSO was added to each well. After gentle mixing, the absorbance was measured using an ELISA plate reader at 570 nm wavelength. Only DMSO was used as the blank (medium) control. For each drug, the concentration that caused only a 10% reduction in metabolic activity was used in subsequent experiments.

Cell transient transfection
The plasmids for EPS15 wild-type WT (Eps15-WT) and dominant negative DN Δ95/295 (Eps15-DN) constructs, both containing proteins fused to GFP 61,62 , were kindly provided by Dra. Nathalie Sauvonnet (Institut Pasteur, Paris, France). In colocalization experiments, early and late endosomes markers, Rab5-GFP and Rab7-GFP, respectively, were used. These In colocalization experiments of either PIXV or RNV with Rab5 and Rab7, images were acquired in a laser-scanning confocal microscope (Zeiss LSM 800 -Germany). For quantitative analysis, a single plane from z-stack cross sections images was deconvoluted using FIJI software 63 . For each cell, a focal plane with well-defined endomembrane structures was selected for analysis and colocalization was quantified using Pearson's correlation coefficient 42 . Coefficient values were then used to perform the statistical analysis.

Inmunofluorescence assay
We performed a previously described assay 37 . Briefly, cell monolayers were grown at 70% confluence on glass coverslips, washed three times with PBS, fixed with 4% Statistical analyses were carried out using unpaired student t-test and one-way ANOVA test, followed by a Tukey post-hoc analysis to enable specific group comparisons (Type I error set at 0.05), as appropriate.

Data Availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.   Percentage of infected cells with PIXV or RNV against control (100% value). Data represent the mean ± SD from 3 independent experiments. Significance was calculated by one-way ANOVA test, followed by Tukey post-hoc analysis to enable specific group comparisons, with p value ≤ 0.05 considered as statistically significant. **p < 0.01; ***p < 0.001. Scale bar: 20 μm.   Percentage of infected cells with PIXV or RNV against controls (100% value). Data represent the mean ± SD from 3 independent experiments. Significance was calculated by one-way ANOVA test, followed by Tukey post-hoc analysis to enable specific group comparisons, with p value ≤ 0.05 considered as statistically significant. *p < 0.05; **p < 0.01; ***p < 0.001. Scale bar: 20 μm.