Ethics statementAll 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-5035541. The virus was isolated on January 6, 2016, from the placenta of a human who had traveled to Honduras in 201541. The viral isolate was procured under Texas Tech University Institutional Review Board-approved protocols. NIH material transfer agreement document that establishes the purpose, usage and limitations of the transferred virus isolate was signed by Texas Tech University.Cells, virus, bacteriaTo culture Vero E6 and CCL-81 cell lines, six well plates were seeded with 3 ml of 2.0 × 105 cells/ml. The maintenance media utilized was Dulbecco’s Modified Eagle Medium (DMEM, ThermoFisher Scientific, 11,965,092), supplemented with 10% fetal bovine serum (FBS, Gibco, 16,140,071) and 1% penicillin/streptomycin (Gibco, 15,140,148). The cells were incubated at 37 °C with 5% CO2 until they attained 90% confluency. The confluent cells were examined under a microscope (EVOS XL Core Imaging system).To preclude any potential contamination, both Vero E6 and CCL-81 cell lines used in this study were freshly procured from ATCC. Once in the lab, during their growth process and amplification, we continuously ensured their morphology and characteristics were according to what is expected in the respective cell lines. Strict aseptic conditions were maintained by adding 1% penicillin/streptomycin (Gibco, 15,140,148) to the DMEM maintenance media and filtering the DMEM media and FBS through Nalgene 0.2 μm filter unit (ThermoFisher Scientific, 567-0020).To generate ZIKV stocks for experiments, the virus was passaged three times on Vero E6 cell lines. On the day of virus inoculation of the confluent Vero E6 cell monolayers, maintenance media was aspirated from 90% confluent cell monolayers. ZIKV at a multiplicity of infection (MOI) of 0.1 was used to inoculate each well, DMEM maintenance media was used to mock infect the negative control wells. To allow the virus to adsorb to cells, the plates were incubated at 37 °C for 1 h while rocking every 15 min to distribute viral inoculum and moisten the cell monolayer. After 1 h of incubation, 3 ml of DMEM maintenance media supplemented with 10% FBS was added to each well, and the plates were incubated at 37 °C with 5% CO2 for 96 h36. At 96 h post infection (hpi), cytopathic effect was visible, the cell monolayer was significantly disrupted and shrunk as well as detached from cell bottom. ZIKV was harvested from the culture supernatant as previously reported. Briefly, the supernatant was aspirated using a Pasteur pipette and centrifuged at 300X g for 10 min at 4 °C 42,43. To eliminate the cell debris, the supernatant was carefully aspirated without disturbing the pelleted cell debris and transferred to cryovials. The virus was held at − 80 °C until downstream procedures.We assessed the integrity of the ZIKV isolate used in this study by aligning the assembled whole genome RNAseq reads to the existing ZIKV genome strain in the database HND 2016 —19,563 (GenBank accession no. KX906952).To prepare bacteria glycerol stocks of E. anophelis and E. coli, bacteria were cultured in Luria–Bertani (LB) media (Sigma Aldrich, L3397) in an orbital shaker (Eppendorf New Brunswick Innova 40/40R) while shaking at 225 rpm at 30 °C (E. coli) and 28 °C (E. anophelis) until the bacteria culture attained an optical density (OD) of 0.5. The negative control was LB media without any bacteria cultured. After attaining log phase growth, 1:1 ratio of bacteria and sterile glycerol solution (ThermoFisher, 15,514,011) was prepared and mixed by inversion and immediately stored at − 80 °C.Elizabethkingia anophelis strain Ag1, NR-50124 (GenBank: CP023402.1) was obtained through BEI Resources, NIAID, NIH. The E. anophelis strain Ag1 was isolated in 2010 from the midgut of a mosquito (Anopheles gambiae, strain G3) in Las Cruces, New Mexico USA44. Escherichia coli was kindly provided by Peter Keyel of Texas Tech University, Biology department.Determination of virus titer by plaque assayVirus titer was determined by plaque assay using the protocol described in45 with a few modifications. In summary, as described above, six well plates were seeded with 3 ml of 2.0 × 105 cells/ml Vero CCL-81, maintained in DMEM media supplemented with 10% FBS and incubated at 37 °C with 5% CO2 until they attained 90% confluency. Upon attaining confluency, maintenance media was aspirated from the confluent cell monolayer, the cell monolayers were inoculated with 100 μl of eight-fold serially diluted ZIKV. Undiluted frozen ZIKV was included as a positive control and the negative control wells were mock inoculated with DMEM maintenance media. The virus was plated in duplicates.To adsorb the virus inoculum to the confluent cells, the cell monolayers were incubated at 37 °C with 5% CO2 for 1 h. After the 1 h incubation period, the virus inoculum or DMEM maintenance media was aspirated, 3 ml of methylcellulose overlay was then added to each well. The infected cells were cultured until 4 dpi when plaques were visible when the cell monolayers were observed under the microscope (EVOS XL Core Imaging system). After the 4th day, the methylcellulose overlay was aspirated using a transfer pipette and then fixed overnight at room temperature by adding 3 ml 10% formaldehyde solution and stained overnight with 2 ml 0.5% crystal violet solution.To determine the virus titer, after the plaques had been stained and were visible, the plates were inverted on a light box and plaques were counted from the wells that contained 10–80 plaques. The plaque morphology was distinguished to ensure that they had not fused, any well with plaques that had merged were not selected to determine the virus titer. The viral titer was determined by multiplying the average number of plaques in the selected dilution factor of the two replicates and the inverse of the total dilution factor and volume of the inoculum used to inoculate the cell monolayer46.In this study we encountered significant challenges when we attempted to utilize Vero E6 cell monolayers to carry out plaque assay due to the constant detachment of the cell monolayer from the bottom of the wells upon addition of the methylcellulose overlay hence resorted to use of the Vero CCL 81.Bacteria culture preparation for cell monolayer inoculationTo prepare E. anophelis or E. coli for cell monolayer inoculation, a sterile pipette was used to scrape frozen glycerol stock and streaked onto a freshly prepared LB agar plate which was incubated overnight at 30 °C (E. coli) and 28 °C (E. anophelis) in a cell incubator. After the overnight growth, a sterile pipette tip was used to select a single colony from the LB agar plate and dipped into a 2 ml LB media in a culture tube. The inoculated LB media was incubated at 30 °C (E. coli) and 28 °C (E. anophelis) in an orbital shaker and grown to 8.0 log10 CFU/ml. The negative control was LB media without any bacteria cultured.The bacterial viable titer in each solution [expressed as Colony Forming Units (CFU)] was determined by an eight-fold serial dilution of the bacterial culture supernatant. A total of 100 μl of the serially diluted bacterial supernatant was pipetted and spread onto LB agar plate and incubated overnight under their respective growth conditions [30 °C (E. coli) and 28 °C (E. anophelis)]. The LB agar plates were plated in duplicates. LB plate with less than 100 colonies was selected to determine the bacteria titer. the colonies were counted and calculations perfomed by multiplying the total number of colonies by the dilution factor and dividing the product by the volume of the culture inoculum47.Co-infection of Vero E6 monolayers with bacteria supernatants and Zika virusIn our previous study findings, we observed that antiviral effect of E. anophelis was only measured when E. anophelis and ZIKV were inoculated sequentially but not co-inoculated40. To sequentially infect Vero E6 confluent cell monolayers with bacteria and ZIKV, maintenance media was aspirated from the confluent Vero E6 cell monolayers and initially inoculated with either 75 μl of 8.0 log10 CFU/ml E. anophelis or 75 μl of 8.0 log10 CFU/ml E. coli or mock infected with DMEM maintenance media (negative control).The infected cell monolayers were incubated at 37 °C for 1 h with 5% CO2. The plates were rocked by tilting at approximately 45° angles every 15 min to moisten the cell monolayer and to evenly distribute bacterial inoculum. After 1 h, 3 ml DMEM maintenance media supplemented with 10% FBS was added to the infected cell monolayer and incubated at 37 °C with 5% CO2 for 24 h40. After the overnight incubation with bacteria or mock infection with DMEM, maintenance media was aspirated from the cell monolayers, ZIKV at a multiplicity of infection (MOI) of 0.1 was used to inoculate each well to a total of six biological replicates and two technical replicates. To allow the virus to adsorb to cells, the plates were incubated at 37 °C for 1 h with 5% CO2 while rocking every 15 min to distribute viral inoculum and moisten the cell monolayer. After 1 h, 3 ml of DMEM maintenance media supplemented with 10% FBS was added to each well, and the plates were incubated at 37 °C with 5% CO2 for 48 h36.The virus was harvested from the supernatant after 48 h as described above by centrifuging at 300 X g for 10 min at 4 °C. The supernatant was carefully aspirated without disturbing the pelleted cell debris and transferred to cryovials at − 80 °C42. The virus was held at − 80 °C until RNA extraction.ZIKV RNA was extracted from the viral supernatants using QIAamp viral RNA kit (Qiagen, 52,906) according to the manufacturer protocol. Briefly, the virus supernatants were lysed with lysis buffer. Carrier RNA was added to the lysis buffer to enhance the binding of viral RNA to the columns. After incubation at room temperature to ensure complete lysis, equal volume absolute ethanol was added to the lysed virus and the solution loaded onto a spin column in order to bind the RNA to the column. To remove contaminants and stabilize the RNA, the column- bound RNA was washed twice with wash buffer and finally eluted into a clean microcentrifuge tube. The RNA was stored at − 80 °C until downstream procedures.In a separate experiment, the pH of Vero cells supernatant following E. anophelis/ZIKV, E. coli/ZIKV and DMEM/ZIKV (negative control) co-inoculation was assessed. Phenol red was added to each well to a final concentration of 2.5% and pH-induced color changes were assessed.Zika virus RNA quantification by quantitative real-time PCRTo quantify copies of ZIKV genome present in the RNA samples, 5 μl of total RNA was reverse transcribed using qScript 1-Step virus tough mix (Quantabio) in a thermocycler (Biorad CFX96 Real-Time System) by incubation at 55 °C for 10 min. The cDNA synthesis step was inactivated by incubating the solution at 95 °C for 1 min. The ZIKV NS1 gene was then amplified at 95 °C for 10 s for 45 cycles, 60 °C for 60 s. Primers and probe targeting the NS1 region of the ZIKV used in this study were: ZIKV 1086 CCGCTGCCCAACACAAG; ZIKV 1162C CCACTAACGTTCTTTTGCAGACAT; ZIKV 1107-FAM AGCCTACCTTGACAAGCAGTCAGACACTCAA)1. NS1 gene was selected due to the fact that it is highly conserved among the ZIKV strains, approaching 99.3% sequence identity48 and hence a suitable gene to ensure efficiency and reproducibility of results. Frozen ZIKV was used as a positive control and sterile water was used as a negative control. In addition, a non-template control was included in the reaction.To calculate copies of ZIKV genome present in the samples, a standard curve was generated using an eightfold serial dilution of previously plaque-tittered ZIKV. The experiment was conducted as six biological and two technical replicates. The differences in virus quantities were measured using one-way ANOVA in GraphPad prism. Plaque assay of 16 randomly picked samples were used to confirm the qRT PCR results using the plaque assay protocol described above. The randomly picked samples were representative of samples drawn from each biological and technical replicate.Growth curve assayTo assess the impact of E. anophelis on the growth rate of ZIKV, supernatants harvested from wells inoculated with E. anophelis/ZIKV or E. coli/ZIKV were filtered using a 0.2 μm UNIFLO syringe filter (Whatman, WHA9914-2502) to exclude the bacteria. Confluent Vero E6 cell monolayers were inoculated with 100 μl of filtered E. anophelis/ZIKV or E. coli/ZIKV or DMEM/ZIKV (negative control) as described above. 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 genomes copy numbers were determined by qRT-PCR targeting the NS1 region as described above.Further, plaque assay of E. anophelis/ZIKV and DMEM/ZIKV was carried out as described above after filtering the virus supernatants as described above to exclude the bacteria.Untargeted analysis of proteins, lipids and metabolitesTo gain insights on excreted extracellular metabolites as a proxy analysis of ZIKV and E. anophelis interaction, we carried out an untargeted analysis of the proteins, lipids and metabolites excreted in the culture supernatants after sequential inoculation of confluent Vero E6 with live E. anophelis and ZIKV (live E. anophelis/ZIKV), heat inactivated E. anophelis and ZIKV (Heat inactivated E. anophelis/ZIKV) and sequential mock infection with DMEM and ZIKV (DMEM/ZIKV) as described above. The culture supernatants were harvested as described above and held at + 80 °C until analyzed. To ascertain that ZIKV replicated to the extent previously observed, ZIKV RNA was extracted from 140 μl of the supernatants using the QIAamp viral RNA kit (Qiagen, 52,906) as described above and qRT PCR to measure the genome copy numbers was perfomed on the viral RNA as described above. We stored the culture supernatants that corresponded to the viral RNA analyzed separately until submission to the Center for Biotechnology and Genomics at Texas Tech University for untargeted extracellular metabolic analysis.To obtain heat inactivated bacteria, E. anophelis was cultured at 28 °C to OD 0.5 (8.0 log10 CFU/ml) shaking at 225 rpm in an orbital shaker (Eppendorf New Brunswick Innova 40/40R) as described above. The grown bacteria were centrifuged at 10,000 rpm for 15 min and the supernatants was heat inactivated on a dry bath at 100 °C overnight.Microfuge tubes handled by the individual involved in the sample preparations were sent to the Center for Biotechnology and Genomics at Texas Tech University sequencing core for normalization of the variation introduced during sample collection and preparation. The DMEM maintenance media that was utilized to maintain the cells was collected as a blank negative control. The samples were pooled to five samples: [DMEM/ZIKV (Mock infection) n = 6; Heat inactivated E. anophelis/ZIKV n = 6 and live E. anophelis/ZIKV n = 6] DMEM maintenance media was added as a negative control and an empty tube was also included [DMEM maintenance media n = 4; empty cryotube handled with gloves n = 1].To account for technical variation and ensure comparability between samples, we performed a bicinchoninic assay (BCA) to determine the total protein concentration. Based on the protein concentration, we injected equal amounts of protein into the instrument. In addition, all raw data were normalized using total protein abundance to account for instrumental bias.Bicinchoninic protein assay (BCA) was performed according to the manufacturer’s instructions to determine the total protein concentration, and based on the protein concentration, equal amounts of protein were injected into the instrument. Briefly, 5 μl of each biological sample was diluted with HPLC-grade water to a total volume of 160 μl. The working reagent was prepared according to the manufacturer’s instructions and added to samples in a 1:1 ratio. Standard solutions were prepared fresh and underwent the same procedure as biological samples. All samples were incubated for 30 min at 60 °C. After incubation, samples were spun down, and absorbance was read on the Thermo Scientific Multiskan FC (Cat. No. N07710).Based on the protein concentrations, aliquots of the five samples stated above [DMEM/ZIKV (Mock infection); Heat inactivated E. anophelis/ZIKV; live E. anophelis/ZIKV;DMEM maintenance media; empty cryotube handled with gloves] containing 50 μg proteins were aliquoted into a separate tube and volume was adjusted to 50 μl using 50 mM ammonium bicarbonate buffer solution. Proteins were denatured at 90 °C for 15 min and subsequently reduced by adding 200 mM of DL-Dithiothreitol (DTT) by incubation at 60 °C for 45 min. Following protein reduction, 200 mM 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 min. 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 100 μl of the sample aliquots above, vortexed for 30 s, and incubated on ice for 45 min. A total volume of 75 μl of DCM was 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 50 μl using a 1:1 solution of 50% methanol and 50% water while Lipidomics samples were resuspended in a volume of 50 μl of a solution of 65% acetonitrile (ACN), 30% Isopropanol and 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 5 μl, whereas lipidomics samples as 10 μl. 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 assayWe observed a reduction in levels of phenylalanine and arginine among the E. anophelis/ZIKV 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 the Sigma Aldrich assay kits (Sigma-Aldrich MAK005; Sigma-Aldrich MAK370).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 (λex = 535 nm/λem = 587 nm) product, proportional to the phenylalanine present.Confluent VeroE6 monolayers were inoculated with DMEM/ZIKV or E. anophelis/ZIKV and cultured for 48 h. A total of 50 μl supernatant was harvested at 48 hpi and filtered in a 10 kDa MWCO spin filter (Sigma-Aldrich, MAK005) by centrifuging at 13 000 g for 10 min to remove insoluble material. To prepare the standard curve for fluorometric detection, serial dilution of Phenylalanine standard solution was prepared and combined with phenylalanine assay buffer, enzyme mix and developer. The sample blank was generated by 1:25 dilution of phenylalanine assay buffer and the developer. The samples were deproteinized by spinning in a 10 kDa MWCO spin filter (Sigma, UFC9010), diluted with 1:100, phenylalanine assay buffer, enzyme mix, and developer. The standards, samples and sample blank were run in duplicates, incubated for 20 min at 37 °C. The fluorescence intensity was measured at (λex = 535 nm/λem = 587 nm (Agilent, BioTek). A standard curve was generated and the amount of phenylalanine present in the DMEM/ZIKV and E. anophelis/ZIKV 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 DMEM/ZIKV and E. anophelis/ZIKV supernatants, incubated at 37 °C for 1 h and centrifuged in Corning Spin-X UF concentrator (Corning) at 13 000 g for 10 min at 4 °C. Arginine enzyme mix, assay buffer was added to the samples and standards and incubated at 37 °C for 30 min. The reaction mix containing probes was then added and incubated at 37 °C for 60 min. The assay was run in duplicates. The absorbance was measured at 450 nm. The amount of arginine present in the DMEM/ZIKV and E. anophelis/ZIKV supernatant was determined from the standard curve.Statistical analysisData 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.
