Multi-Omics Analysis Of Antiviral Interactions Of Elizabethkingia anophelis And Zika Virus

Background The microbial communities residing in the mosquito midgut play a key role in determining the outcome of mosquito pathogen infection. Elizabethkingia anophelis, originally isolated from the midgut of Anopheles gambiae, has drawn much attention due to its close association with Aedes and Anopheles mosquitoes, primary vectors of dengue virus and malaria parasites, respectively. E. anophelis possesses a broad-spectrum antiviral phenotype, yet a gap in knowledge regarding the mechanistic basis of its interaction with viruses exists. Methodology/Principal findings To further understand the antiviral interactions between E. anophelis and Zika virus (ZIKV), we utilized a non-targeted multi-omics approach, analyzing lipids, proteins, and metabolites of cell monolayers co-infected with ZIKV and E. anophelis. We further assessed the gene expression of ZIKV when cultured in the presence of E. anophelis. ZIKV cultured in the presence of E. anophelis resulted in an attenuated replicative fitness and unproductive virus infection. Further, in this treatment, we observed lower levels of the nonstructural protein 5 (NS5) and RNA-directed RNA polymerase (RdRp) protein. Lastly, a significant decrease in arginine levels, an essential requirement for viral replication and progression of viral infection was observed. Conclusions/Significance This study provides insights into the molecular basis of E. anophelis antiviral phenotype. These findings improve our knowledge of how microbes and viruses interact to impact viral replication. In the future, our findings can be utilized to unravel the mechanism behind the antiviral phenotype of E. anophelis, and this can help develop novel paradigms for viral therapeutics. AUTHOR SUMMARY Zika is a re-emerging disease and is endemic in many regions of sub-Saharan Africa, Asia and Latin America. It remains a major public health threat and lacks FDA-approved therapeutics or vaccines, hence the urgent need for the identification of alternative approaches that limit the transmission of the pathogens by its primary vector, Aedes spp. The microbial communities residing in the mosquito midgut play a key role in determining the outcome of mosquito pathogen infection. Flavobacteria dominates the mosquito midgut including Elizabethkingia, which is a gram-negative bacillus prevalent in Aedes and Anopheles species of mosquitoes. E. anophelis, a poorly studied midgut microbe, has a broad-spectrum antiviral phenotype, yet the mechanism of its antiviral action is unknown. In this study, we have identified several pathways as well as Zika virus proteins perturbed when the Zika virus is cultivated in the presence of E. anophelis. Our findings do not only provide insights into microbial, virus interaction but could be harnessed to develop novel antiviral tools.


AUTHOR SUMMARY
Zika is a re-emerging disease and is endemic in many regions of sub-Saharan Africa, Asia and Latin America.It remains a major public health threat and lacks FDA-approved therapeutics or vaccines, hence the urgent need for the identification of alternative approaches that limit the transmission of the pathogens by its primary vector, Aedes spp.The microbial communities residing in the mosquito midgut play a key role in determining the outcome of mosquito pathogen infection.Flavobacteria dominates the mosquito midgut including Elizabethkingia, which is a gram-negative bacillus prevalent in Aedes and Anopheles species of mosquitoes.E. anophelis, a poorly studied midgut microbe, has a broad-spectrum antiviral phenotype, yet the mechanism of its antiviral action is unknown.In this study, we have identified several pathways as well as Zika virus proteins perturbed when the Zika virus is cultivated in the presence of E. anophelis.Our findings do not only provide insights into microbial, virus interaction but could be harnessed to develop novel antiviral tools.

INTRODUCTION
Zika virus (ZIKV) is an emerging virus infection [1][2][3][4][5][6], endemic in many regions of sub-Saharan Africa [7][8][9][10][11][12][13][14][15] as well as in countries in South Asia [16][17][18][19].ZIKV is a focus of intense research due to its rapid geographic spread in the Americas and its association with birth defects (e.g.microcephaly) and neurological syndromes [20][21][22].It is mainly transmitted by the bite of infected Aedes mosquitoes, in addition, sexual, transplacental, as well as transmission through blood transfusion have been documented [23].There is no FDA-approved medication or vaccine to treat or prevent ZIKV infection despite WHO declaring Zika a public health emergency of international concern [24].Medications for symptomatic relief are the only source of relief for infected individuals [25].Control of the mosquito vector is the primary method to limit transmission of ZIKV [23].However, an upsurge in insecticide resistance as well as changing weather patterns, which can complicate vector control efforts has led to significant geographical range expansion of Aedes spp.[26], underscoring the urgent need for novel, efficacious, and cost-effective alternatives to traditional mosquito control, and methods to limit pathogen transmission.
At the field application level, Wolbachia has demonstrated significant success in containing the spread of dengue and Zika in natural populations [47,48].An expansion of the biological control toolbox by exploration for a wide array of microbes that bear antiviral properties is urgent.
Elizabethkingia is a rod-like, gram-negative, aerobic, non-fermenting, non-motile, and nonspore-forming bacteria widely distributed in natural environments, including soil, freshwater, and hospitals [49].Originally isolated from the midgut of Anopheles gambiae [50], it has drawn much attention due to its association with Anopheles and Aedes spp., primary vectors of Plasmodium parasites and dengue virus (DENV), respectively [51].Due to its initial isolation from the midgut of Anophelis gambiae, there is a potential for E. anophelis to interact with other co-habiting microbial species in the mosquito midgut and with the host mosquito.
E. anophelis was noted as the dominant microbial genus in a lab colony of adult An. funestus and An.arabiensis collected from Mozambique and South Africa [52].Further, sequencing of gut bacterial communities of An. gambiae from Cameroon in West Africa resulted in 95% of sequence tags associated with Elizabethkingia spp.[53] .In addition, Elizabethkingia was identified in a Brazilian Ae. aegypti colony among the non-fed, blood-fed fed as well as the ZIKV-infected mosquitoes [54].In our previous findings, we identified E. anophelis as the most abundant gut microbiome species in both noninfectious and infectious blood-fed lab-reared Ae. albopictus collected from Long Island, New York [55].A past study reported that 10 7 E. anophelis per midgut, resulted in a significant reduction of Plasmodium falciparum oocyst load to approximately zero.Further, the study identified the capacity of E. anophelis to inhibit P. berghei ookinete development in vitro.The study postulated that E. anophelis inhibits parasite development either through direct interaction with the parasites or through the production of inhibitory factors [56].A different study reported that extracts obtained from cultured E. meningoseptica showed activity against gram-positive Staphylococcus aureus and gramnegative E. coli as well as Candida albicans.Further, the extracts were active against blood and gametocyte transmission stages of Plasmodium falciparum [57].Consistent with previous studies, we previously showed that E. anophelis has a broad-spectrum antiviral activity significantly reducing viral loads of ZIKV, dengue virus (DENV), and chikungunya virus (CHIKV) in vitro while negatively impacting infection rates of ZIKV in Ae. albopictus mosquitoes [58].
This study aimed to identify the pathways that E. anophelis targets to effect, the anti-viral phenotype, as well as delineate the impact of E. anophelis on ZIKV gene expression.To address this aim, we cultured ZIKV in the presence of E. anophelis and utilized a multi-omics approach, leveraging non-targeted metabolomics, proteomics, and lipidomics assay.The findings of our study will provide insights into the key points targeted by E. anophelis to impact ZIKV replication.Further, we contribute to a body of knowledge that can be harnessed to develop new avenues for pathogen and vector control tools.

Ethics statement
All experiments using viruses or bacteria were performed under biosafety level 2 (BSL-2) conditions at Texas Tech University with Institutional Biosafety Committee approval.
The ZIKV strain utilized in this study (GenBank:KX262887) was obtained through BEI Resources, NIAID, NIH: Zika Virus, R103451, NR-50355.The virus was isolated on January 6, 2016, from the placenta of a human who had traveled to Honduras in 2015.To generate stock for experiments, the virus was passaged three times on Vero E6 cell line to create stock virus and titrated by standard plaque forming assays.Stocks were stored in liquid Nitrogen prior to use.

Co-infection of Vero E6 monolayers with bacteria supernatants and Zika virus
Six well plates were seeded with Vero E6 cells three days prior.Upon attaining confluency, the monolayers were inoculated sequentially with 75 μl of 8.0 log10 E. anophelis CFU/ml and ZIKV at a multiplicity of infection (MOI) of 0.1 (ZIKV/E.anophelis); 75 μl of 8.0 log10 E. coli and ZIKV at an MOI of 0.1(ZIKV/E.coli) and mock infected with DMEM maintenance media and ZIKV at an MOI of 0.1(ZIKV/DMEM).Infected cells were cultured at 37°C with 5% CO2 for 48 hours [55].
Supernatants was harvested from each well, ZIKV RNA was extracted from the supernatants using QIAamp viral RNA kit (Qiagen, 52906) according to the manufacturer protocol.

Zika virus RNA quantification by RT-qPCR
Quantitative real-time qRT-PCR assay was performed using primers targeting the NS1 region of the ZIKV (ZIKV 1086 CCGCTGCCCAACACAAG; ZIKV 1162C CCACTAACGTTCTTTTGCAGACAT; ZIKV 1107-FAM AGCCTACCTTGACAAGCAGTCAGACACTCAA) [59].Standard curve was generated using an 8-fold dilution of previously plaque-tittered ZIKV for the quantification of copies of ZIKV genome present in the samples.Plaque assay of a select number of samples were used to confirm the qRT PCR results.The experiment was conducted as six biological and two technical replicates.

Growth curves
Supernatants harvested from wells inoculated with ZIKV/ E. anophelis or ZIKV/ E. coli above were filtered using a 0.02-micron filter (Whatman) to exclude the bacteria.Thereafter, the filtered supernatants (ZIKV/E.anophelis or ZIKV/E.coli) were added to confluent Vero E6 monolayers at equal ratios at 0.1 MOI, ZIKV/DMEM was added as a negative control.After incubating at 37°C with 5% CO2 for 1 hour, the viral supernatants inoculum was removed and 3 ml of DMEM containing 10% FBS was added.Infected cells were cultured at 37°C with 5% CO2 and viral culture supernatants were collected at 24, 48, 72 and 96 hpi time points.Virus titers were determined by RT-qPCR targeting the NS1 region as described above.Standard curves were generated using a 10-fold serial dilution of previously tittered ZIKV for the quantification of copies of the ZIKV genome present in samples.

Plaque assay
A total of 100 μl of the filtered supernatants of ZIKV/E.anophelis and ZIKV/DMEM (as control) were serially diluted eight-fold with DMEM containing 1% penicillin/streptomycin. Subsequently, 100 μl of each diluent were added to CCL-81 cell monolayers in six well plates in duplicates and incubated at 37°C with 5% CO2.The supernatants containing the virus was discarded after 1 hour of incubation and 3 ml methylcellulose overlay was then added to each well.The infected cells were cultured for another 4 days and then fixed overnight at room temperature by adding 10% formaldehyde solution and stained overnight with 0.5% crystal violet solution.The plaques were counted for the calculation of virus titers.

RNA-sequencing and data analysis
Total ZIKV RNA was extracted from supernatants harvested after 48 hours from VeroE6 monolayers inoculated with ZIKV/E.anophelis or ZIKV/DMEM using RNeasy kit (Qiagen) according to manufacturer's protocol.The RNA library construction and high-throughput sequencing were performed by Genewiz, Azenta Life Sciences.Multiplexed libraries were sequenced for 150bp at both ends using an Illumina HiSeq6000 platform.Three biological replicates of each condition were RNA sequenced.
The high throughput Illumina sequence data were processed using the Galaxy online tools (usegalaxy.org).Reads were first trimmed and filtered to remove the ambiguous nucleotides and low-quality sequence using trimmomatic (HEADCROP, 15; SLIDING WINDOW, 4, 30; MINLEN, 50).The filtered reads were mapped to the ZIKV reference genome NCBI ViralProj 411812 (RefSeq: GCF_002366285.1), using RNA STAR.ZIKV genome is a polyprotein, and hence, a homogenous distribution of reads was expected along the polyprotein sequence.This was not the case in this study and hence the annotation of the unique coding sequence of the genome was split into the unique proteins coded within the polypeptide.The new annotation generated was used to count reads aligning uniquely to the specific genes with htseq count.Subsequently, differential expression (DE) was calculated using the reads aligning uniquely to ZIKV transcripts using DESeq2 [60].Significance was determined using DESeq2 exact test for the negative binomial distribution, corrected with a False Discovery Rate (FDR) at P < 0.05.

Untargeted analysis of proteins, lipids and metabolites
Confluent monolayers of Vero E6 were infected with ZIKV and DMEM (ZIKV/DMEM); ZIKV and heat inactivated E. anophelis (ZIKV/Heat inactivated E. anophelis ) or ZIKV and live E. anophelis (ZIKV/live E. anophelis).The samples were cultured at 37°C with 5% CO2 for 48 hours.At 48 hours post infection (hpi), supernatants were harvested from wells containing ZIKV/DMEM, ZIKV/Heat inactivated and ZIKV/live E. anophelis.To assess whether ZIKV replication was successful, ZIKV RNA was extracted from the supernatants using QIAamp viral RNA kit (Qiagen, 52906) according to manufacturer protocol.Samples were stored at -80°C before submission to the Center for Biotechnology and Genomics at Texas Tech University for untargeted screening of proteins, lipids and metabolites.Microfuge tubes handled by the individual involved in the sample preparations were sent to the sequencing core for normalization of the variation introduced during sample collection and preparation.The maintenance media that was utilized to maintain the cells was collected as blank negative control.The samples were pooled into four pools with six biological replicates [ZIKV/DMEM (Mock infection); ZIKV/Heat inactivated E. anophelis and ZIKV/live E. anophelis].
To characterize the protein changes that occur when ZIKV is co-cultured with E. anophelis, we performed protein analysis on the four pools by BCA protein assay.Based on the protein concentrations, aliquots of the four samples containing 50 µg proteins were aliquoted into a separate tube and volume was adjusted to 50µl using 50mM ammonium bicarbonate buffer added to the solution after incubation and vortexed for 30 s followed by addition of 75 µl of cold water, vortex for 30 s and centrifuged at 5000 rpm for 15 min.The top layer (aqueous phase), which contained polar/hydrophilic metabolites, was collected for metabolomics analysis.The bottom layer (organic phase), which contained non-polar/hydrophobic lipids, was collected for lipidomics analysis.Both the aqueous layer and organic layers were air dried, then resuspended before LC-MS/MS analysis.
Metabolomics samples were resuspended to a volume of 50µl using a 1:1 solution of 50% methanol, 50% water while Lipidomics samples were resuspended in a volume of 50µl of a solution of 65% ACN, 30% Isopropanol, 5% water.Samples were analyzed using Thermo Vanquish UHPLC/ Thermo Q Exactive HF.After running pool samples for tests, the injection volume for metabolomics samples was decided as 5µl, whereas lipidomics samples as 10µl.
The identification and quantification of metabolites and lipids were performed by Thermo Compound Discoverer 3.1 and LipidSearch 4.2, respectively.The metabolites/lipids with significant changes (p<0.05) were identified after performing the Mann-Whitney U test.

Phenylalanine and Arginine assay
We observed a reduction in levels of phenylalanine and arginine among the ZIKV/E.anophelis samples in the metabolomic assay.To confirm changes in phenylalanine and arginine concentration upon culturing ZIKV in the presence of E. anophelis, a secondary biochemical assay was performed to verify the metabolomic predictions.Both analyses were performed using commercial assay kits.
Phenylalanine assay (Sigma-Aldrich MAK005) utilizes a coupled enzyme assay that results in the deamination of phenylalanine and the production of NADH which reacts with the probe resulting in a fluorescent (lex = 535 nm/lem = 587 nm) product, proportional to the phenylalanine present.
Confluent VeroE6 monolayers were inoculated with ZIKV/DMEM or ZIKV/E.anophelis and cultured for 48 hours.A total of 50µl supernatant was harvested at 48 hpi and filtered in a 10kDa MWCO spin filter (Sigma-Aldrich) by centrifuging at 13 000 g for 10 minutes to remove insoluble material.Samples were diluted 1:100, phenylalanine assay buffer, enzyme mix, and developer were mixed together, standards and sample blank were run in duplicates, this was mixed, incubated for 20 minutes at 37°C.The fluorescence intensity was measured at (lex = 535 nm/lem = 587 nm.A standard curve was generated and the amount of phenylalanine present in the ZIKV/DMEM and ZIKV/E.anophelis supernatant was determined from the standard curve. L-Arginine assay (Sigma-Aldrich MAK370) is an enzyme-based assay whereby L-Arginine is converted into a series of intermediates which reacts with the probe resulting in a stable colorimetric signal at 450 nm (A450).A total of 2 µl of sample cleanup mix was added to 100 µl of ZIKV/DMEM and ZIKV/E.anophelis supernatants, incubated at 37°C for 1 hour and centrifuged in Corning Spin-X UF concentrator (Corning) at 13 000 g for 10 minutes at 4°C.Arginine enzyme mix, assay buffer was added to the samples and standards and incubated at 37°C for 30 minutes.The reaction mix containing probes was then added and incubated at 37°C for 60 minutes.The assay was run in duplicates.The absorbance was measured at 450nm.The amount of arginine present in the ZIKV/DMEM and ZIKV/E.anophelis supernatant was determined from the standard curve.

Statistical analysis
Data analysis was carried out using GraphPad Prism software.Statistical evaluation was performed using one-way ANOVA or student's unpaired t-test for single factor analysis.Unless stated otherwise the experiments were done in six biological replicates and two technical replicates.The data are presented as means.

Elizabethkingia anophelis attenuates Zika virus replicative fitness and yield of infective virus.
We have previously demonstrated the antiviral effect of E. anophelis on ZIKV, DENV, and CHIKV [58].In the current study, ZIKV cultured in the presence of E. anophelis resulted in a significant reduction of ZIKV RNA copy numbers (unpaired t-test; P< 0.0001) (Figure 1A) but this was not the case when cultured in the presence of E. coli (unpaired t-test; P = 0.4947) (Figure 1B);(Supplementary file no. 1).
Next, we queried whether culturing ZIKV in the presence of E. anophelis would impact virus replicative fitness.To preclude the effect of the bacteria, we filtered out the bacteria (E. anophelis or E. coli) from the viral supernatants before performing the growth curve assay.ZIKV exposure to E. anophelis resulted in subsequent suppressed replication throughout the growth curve assay (Figure 1C; Supplementary file 2).We observed a significant variation in the ZIKV RNA copy numbers between ZIKV/E.anophelis and ZIKV/DMEM (mock infection) samples at each time point.The ZIKV/DMEM (mock infection) demonstrated increased replication relative to ZIKV/E.anophelis at all time points and, at 5 dpi, there was an approximately a two-fold difference in RNA copy numbers between the two treatments.In contrast, we did not measure any difference in genome copy numbers in ZIKV/DMEM and pre-filtered ZIKV/E.coli samples at each time point throughout the growth curve assay (Figure 1D; Supplementary file 2).Next, we assessed the impact of E. anophelis on the formation of infectious virus.As above, we cultured ZIKV in the presence of E. anophelis, and harvested ZIKV/DMEM and ZIKV/E.
anophelis supernatants at 48 hpi, filtered the bacteria, and seeded confluent CCL-81 with serial dilutions up to 10 -8 .We did not observe any plaque formation in any well in replicate plates inoculated with filtered ZIKV/E.anophelis (Figure 2).The control wells inoculated with ZIKV/DMEM supernatants had visible plaques up to 4 th dilution in one plate and the 5 th dilution in the replicate plate (Figure 2).

Reduction of Zika virus NS5 levels
We utilized transcriptomic analysis to investigate ZIKV gene expression responses to E.
anophelis.We carried out an RNA-seq analysis of supernatants originating from Vero E6 monolayers inoculated with ZIKV/E.anophelis and ZIKV/DMEM (mock infection).We observed a statistically significant reduction of the ZIKV nonstructural protein 5 (NS5) levels in ZIKV/E.
anophelis supernatants (Supplementary file 3) (P=0.0004).Further, both the membrane and pre-membrane genes showed a minimal increase in levels.Additionally, though not statistically significant, we observed a decrease in the amount of the envelope, NS1, NS2B, NS3, and NS4A genes.In addition, when we queried the proteins identified in the multi-omics analysis against ZIKV polyprotein we observed a statistically significant increase in the relative ZIKV/E.anophelis

Significant reduction of proteins responsible for replication of RNA-based viruses.
Analysis of the unsupervised principal component analysis (PCA) revealed a unique cluster of proteins associated with ZIKV/live E. anophelis samples, a few proteins shared among ZIKV/live E. anophelis and ZIKV/heat-inactivated E. anophelis conditions, while ZIKV/DMEM and ZIKV/Heat inactivated E. anophelis samples shared common proteins (Figure 3A).

Figure 3A
We observed a significant reduction in the levels of RNA-directed RNA polymerase (RdRp) protein among the ZIKV/live E. anophelis samples.RdRp is important in the replication of RNAbased viruses [61] (Figure 3B).In addition, a reduction in polymerase cofactor VP35, a viral structural protein which apart from inhibiting host innate immune response, acts as a cofactor in ZIKV/DMEM ZIKV/live E. anophelis ZIKV/Heat inactivated E. anophelis We also observed the upregulation of lipids such as sphingomyelin and ceramide, that are known to be associated with the viral membrane entry.Point coloration represents the fold change observed in ZIKV/E.anophelis samples.
Analysis of the lipid profile demonstrated multiple deviations, in particular, lysophosphatidylcholine, a lysolipid, known to promote positive curvature and enhance membrane permeability was reduced among ZIKV/live E. anophelis samples.All ZIKV/live E. anophelis samples except one, showed a reduction of Acyl Carnitine, an ester of I-carnitine and fatty acids.The rest of the lipid profiles identified in this study increased levels: diglycerides, ceramides, phosphatidylethanolamines, sphingomyelins, phosphatidylcholines, and triglycerides (Figure 4B).

Reduction of metabolites necessary for viral replication.
Among the ZIKV/live E. anophelis samples, we observed a reduction of arginine and phenylalanine levels known to be important in several viral replication processes.We further validated the changes in phenylalanine and arginine concentration through a secondary biochemical assay which demonstrated a reduction of both arginine (Supplementary file 8) and phenylalanine levels(Supplementary file 9) in supernatants associated with ZIKV/live E. anophelis relative to ZIKV/DMEM (mock infection).Furthermore, among the ZIKV/live E. anophelis, there was a decrease of arachidonic acid, a precursor of eicosapentaenoic acid, present in the cell membranes of most body cell samples.
Additionally, ZIKV cultured in the presence of E. anophelis resulted in a hypoglycemic condition with low sucrose levels observed.Reduced levels of cytidine were detected among ZIKV/live E. anophelis samples.Cytidine is a component of RNA and a nucleoside formed when cytosine is attached to a ribose ring.Lastly, Pyridoxal, the natural form of vitamin B6, a co-enzyme involved in the metabolism of carbohydrates, lipids, proteins, and nucleic acids was reduced in the ZIKV/live E. anophelis samples (Figure 5).

DISCUSSION
The results of our study demonstrate that ZIKV cultured in the presence of E. anophelis is associated with attenuated viral replicative fitness and lack of infectious virus.Indeed, the lack of robust viral replication in the ZIKV/ live E. anophelis samples is coupled with a reduction in levels of NS5 expression and low levels of RdRp protein.The NS5 protein is essential for the replication of the flaviviral RNA genome [62][63][64].NS5 nonstructural protein consists of the Nterminal region made up of methyltransferase (MT), this is followed by a short linker that joins the MT to the RdRp [65][66][67].Apart from adding the 5' RNA cap structure to assist in the translation of the viral polyprotein and counteracting antiviral signaling [68,69], the RdRp initiates RNA synthesis [70].The alteration of NS5 has been shown to affect the speed and fidelity of RNA replication, hence altering the viral fitness [71][72][73][74].We cannot rule out that the variation of NS5 levels observed in our study could also reflect low read counts, technical bias, and ribosome stalling at specific sites [75].Further investigations involving experimental validation are required as well as cultures on broad cell lines to assess the potential of host-specific antiviral responses.
Flavivirus membrane proteins play important roles in facilitating the entry of viruses into host cells and assembly [76].After the flavivirus maturation, both prM/E proteins undergo conformational changes resulting from a low pH environment, leading to cleavage of prM into pr and M resulting in a mature virus that can proceed to exit the cell through the secretory pathway [77][78][79].In this study, we observed increased levels of prM and E proteins in the ZIKV/ live E. anophelis samples relative to ZIKV/DMEM (mock infection).This may suggest the increased utility of prM and E proteins in generating mature virions in ZIKV/DMEM compared to the ZIKV/ live E. anophelis samples.Future studies aimed at dissecting what point of the ZIKV life cycle is impacted by E. anophelis are important to understand the basis of E. anophelis antiviral phenotype.
The significant reduction of both arginine and phenylalanine levels in ZIKV/live E. anophelis samples relative to ZIKV/DMEM (mock infection) is interesting.Arginine is an essential requirement for the replication of viruses and the progression of viral infection [80].The result of our study corroborates previous findings which demonstrated the impact of nutritional deficiencies on Herpes Simplex Virus (HSV) replication in human cells by eliminating arginine, resulting in no viral recovery from any tube containing arginine-deficient medium, with no cytopathic effect (CPE).When the arginine-deficient medium was replaced with a complete medium the CPE was recovered, and viral replication was restored [81].Further, arginine has been demonstrated to be essential for the replication of vaccinia virus in HeLa cells and the yield of infective virus was equivalent to the concentration of arginine in the medium.It was further shown to be important for RNA and DNA replication and was incorporated and utilized in the formation of complete virus particles.The study concluded that the requirement of arginine in the vaccinia replication cycle can be distinguished as required before the synthesis of virus DNA and associated with the formation of mature virus particles [80].
In this study, all ZIKV/live E. anophelis samples except one, showed reduced levels of Acyl Carnitine, an ester of I-carnitine and fatty acids.A previous study focused on identifying the specific cellular pathways and genes necessary for DENV replication demonstrated that fatty acid synthesis is critical for DENV replication and upon inhibition of fatty acid synthesis pathways, a significant inhibition of DENV replication was observed [82].Further, in the same study, the inhibition of the fatty acid biosynthesis inhibited the replication of the Yellow Fever virus and West Nile virus [82].
Lipid profiles identified in this study were enriched: diglycerides, ceramides, phosphatidylethanolamines, sphingomyelins, phosphatidylcholines, and triglycerides.Our results coincide with the findings of a previous study aimed at profiling the lipidome of DENVinfected mosquito cells at various infection stages.At the binding and entry stage, the study utilized an inactivated virus that was capable of binding and entry but not replication, and they observed selective enrichment of lipids that can influence membrane structure and also have signaling functions such as sphingomyelin, ceramide, lysophospholipids and several intermediates such as mono-and diacylglycerol and phosphatic acid [83].Taken together, our findings suggest that the lipid profile perturbations in this study may point to virus interaction with the cellular membranes, but a lack of a robust viral replication as demonstrated by a downregulation of cellular components associated with the formation of viral replication factories.
In conclusion, this study has provided insights into the pathways as well as viral genes that are impacted by ZIKV and E. anophelis interaction.These results contribute to understanding the molecular determinants of E. anophelis antiviral phenotype.
One of the significant limitations of the study is the lack of inclusion of an in vivo analysis of the interaction between ZIKV and E. anophelis or a comparison of the findings of Vero cell lines with mosquito or human cell lines which could be different and alter our findings.In the future, to build on our study findings, we recommend an in vivo study and a broad comparison of different cell lines to gain a basic understanding of whether E. anophelis antiviral effects are host specific.
Despite these limitations, our findings provide evidence that a poorly studied midgut microbe, E. anophelis, has an antiviral phenotype, reducing the replicative fitness of ZIKV and formation of infectious virus, and is associated with perturbance of pathways associated with key pathways that are necessary for viral replication and formation of viral factories.Future studies to identify the mechanisms utilized by E. anophelis to negatively impact viral replication can build upon our current findings.Identification of such mechanisms can lead to novel anti-viral pathways.
solution.Proteins were denatured at 90˚C for 15 min and subsequently reduced by adding 200mM of DL-Dithiothreitol (DTT) by incubation at 60˚C for 45 min.Following protein reduction, 200mM of Iodoacetamide (IAA) and incubated at 37˚C for 45 min to allow protein alkylation.DTT was added a second time to quench the extra IAA and incubated at 37˚C for 30 mins.An aliquot of 2 μg of trypsin was added to each sample (trypsin: protein=1:25, m/m) and incubated at 37˚C for 18 h.After tryptic digestion, formic acid (FA) was added to the final concentration of 1% (v/v) to stop the digestion process.Samples were dried in a speed vacuum, then resuspended to 1 μg/μL with a solution of 20% acetonitrile (ACN) and 1% FA.After resuspension, 1 µl of each sample (1 μg of proteins) was injected and analyzed by LC-MS/MS (Thermo UltiMate 3000 nanoLC/ Thermo Fusion Lumos).The protein identification and quantification were performed by Thermo Proteome Discoverer 2.4 by querying the database of Chlorocebus Sabaeus (Green Monkey) acquired from Uniprot https://www.uniprot.org/proteomes/UP000029965,the proteins of ZIKV (GenBank:KX262887) and Elizabethkingia anophelis, strain Ag1, NR-50124 (GenBank: CP023402.1).The statistical significance (p<0.05) of protein expression changes was identified after performing the Mann-Whitney U test.To analyze the Metabolites and Lipids, a volume of 200 μl of extraction solution [dichloromethane (DCM): methanol=1:2 (v/v)] was added to 100µl of the sample aliquots above, vortexed for 30 seconds and incubated on ice for 45 min.A total volume of 75 µl of DCM was

Figure 1 .
Figure 1.Analysis of the impact of E. anophelis on viral replication and growth kinetics of

Figure 2 .
Figure 2. Determination of the impact of E. anophelis on ZIKV infectiousness.ZIKV/E.anophelissupernatants harvested at 48 hpi were filtered through a 0.2µm filter to preclude

Figure 3 .Figure 4 .
Figure 3. Assessment of E. anophelis on protein profile.A. Analysis of the unsupervised