Plasmid Constructs

Reporter RNA constructs were engineered with the baculovirus immediate early promoter (IE3) [15] for transcription in C6/36 cell culture and the Ae. aegypti poly-ubiquitin (PUb) promoter [16] to transcribe the reporter RNA in transformed Ae. aegypti mosquitoes. The reporter constructs were developed from SINV sequences using the TE3’2J/TR339 strain of Sindbis and from SINV replicon pBG254 and pBG60 previously described [17,18]. The 5’ and 3’ UTR sequences were included from SINV (5’ end to the start of the nonstructural protein and the 3’ end from the c-terminus of E1 through the poly A tail). The first 143 residues of nsP1 were inserted in frame with an enhanced Green fluorescent protein (eGFP) gene followed by stop codons used to identify transfected cells and stop translation initiated from the 5’ end of the reporter RNA. We then inserted the subgenomic promoter sequence followed by the mCherry gene, which will produce a subgenomic RNA that mCherry can be translated from in the presence of transcomplemented nsP1-4 proteins. 50 adenosine residues were added downstream of the 3’ UTR to produce a polyadenylated end (pBG426) as previously described [19].

eGFP was removed from pBG426 by BglII restriction enzyme digest and the plasmid was re-ligated to produce plasmid pBG446. Alternatively, eGFP was removed from pBG426 with BglII and replaced with the antibiotic resistance gene puromycin acetyltransferase (PAC) [20], which had been amplified with primers containing BglII sites (BG661- ATGCAGATCTTTCGTGAAGACCC and BG662-CCTGAGATCTGGCACCGGGCTTGC). The PAC gene was ligated into the BglII site, producing plasmid pBG461.

pBG460 was designed using the West African strain of Chikungunya virus (CHIKV) 37997(pCHIK-37997–5GFP, GenBank accession number EU224271) [21]. The CHIKV sequence included identical regions (5’ UTR, nsP1, SGP) as the SINV sequence (pBG426) and was synthesized by GenScript. The CHIKV reporter sequence was amplified from the synthesized plasmid using a reverse primer with a 5’ end XhoI site and a forward primer with the 5’ end having 20bp overlap with the 3’ end of the IE promoter (BG657-GTTCATGTTGGATATTGTTTCATGGCTGCGTGAGACACACG and BG658-CGGGCCCTCAAGACTCGAG ). The IE promoter was amplified with a forward primer containing a 5’ NheI site and reverse primer with 5’ overlap of the 5’ CHIKV UTR (BG655-GTCGGGTCCATTGTCCGTGTG and BG656- CGTGTGTCTCACGCAGCCATGAAACAATATCCAACATGAAC). The two PCR products were the templates for fusion PCR for 10 rounds followed by addition of primers BG655 and BG658 for 30 more cycles. The resulting full-length fragment was ligated into the NheI and XhoI sites in pBG426.

Additional subgenomic promoters from Western equine encephalitis virus (WEEV)) (pBG447) or CHIKV (pBG448) were inserted into pBG446 by ligating a virus specific SGP/mCherry PCR product into the XbaI site. The WEEV and CHIKV SGPs were amplified with forward primers containing 5’ XbaI sites (BG582- TACATCTAGACTCTACGGCTGACCTAAATAGG and BG581- TACATCTAGACTGTACGGTGGTCCTAAATAGG). The reverse primer annealed to the 3’ end of mCherry and had a 5’ XbaI site (BG576- ATATTCTAGACTACTTGTACAGCTCGTCCATGC).

The IE3 promoter was replaced with the Ae. aegypti poly ubiquitin promoter (PUb) [16] and inserted into the transposon backbone containing the 3xP3 promoter and an eGFP gene for mosquito transgenesis [22]. The IE3 promoter was replaced with the PUb promoter through overlapping PCR amplification. The SINV reporter construct was amplified with a forward primer containing 5’ 19bp overlap with the PUb and a reverse primer with a terminal AscI site (BG 671- GCAAAGGCAAAACCAGCTCATTGACGGCGTAGTACACAC and BG 675-CTGGCGCGCCGCCCTCAAGACTCGAG ). The PUb fragment was amplified with a forward primer containing a 5’ AscI site and a reverse primer with 5’ 20bp overlap with the beginning of the reporter construct. (BG669- CTGGCGCGCCTATCTTTACATGTAGC and BG670- GTGTGTACTACGCCGTCAATGAGCTGGTTTTGCCTTTGC). The two PCR fragments were fused for 10 rounds, then primers BG669 and BG675 were added to amplify the fusion PCR product. The PCR product and destination plasmid were digested with AscI and ligated into the transposon backbone (pMos[3xP3-eGFPaf]) resulting in pBG471 [22]. All clones were verified with sequencing.

Cell Culture, Transfection, and Viruses

C6/36 Aedes albopictus cells were grown in 6-well cell culture plates with L-15 media containing 10% FBS, 100U/ml penicillin/streptomycin, and 5% NaHCO3. Cells were maintained in a 28°C incubator. C6/36 cells were transfected with Mirus 293T transfection reagent following manufacturer’s protocols for 1µg DNA. 12 hours post transfection; the cells were infected with SINV at an MOI of 10. At 36-48 hours post infection, the cells were examined for mCherry expression. Stable reporter expressing cells were selected with 2µg/ml of Puromycin and drug was replaced every 3-4 days.

TE3’2J SINV virus stocks for in vitro infections were produced from plasmid transfection into Baby Hamster Kidney (BHK) cells [8] . MRE16 5’ds GFP and MRE16 5’ds were generated from infectious clones [23,24]. Chikungunya virus (La Reunion strain LR2006-OPY-1) and Western Equine Encephalitis virus (McMillian strain) infections were performed in biosafety level 3 [25,26]. The CHIKV LR strain was used because we had access to an infectious clone containing a 5’ duplicated subgenomic promoter and GFP gene for visual detection of infection. Sindbis and West Nile virus (Kunjin subtype) (KUNV) infections were performed at biosafety level 2 [27]. 24 hours after plating cells or 12 hours post transfection, cells were infected by replacing media and adding in sufficient amounts of virus for desired MOI.

Fluorescence quantification

Fluorescent images were acquired with a Nikon Diaphot 200 inverted fluorescent microscope. Images were analyzed and fluorescence was quantified using Image J software [28,29]. Specifically, images were separated to RGB color channels. The red or green stack was selected and the threshold was adjusted to detect only fluorescent cells that were brighter than negative controls. Following the threshold adjustment, fluorescent cells were analyzed for pixel counts, total area, average size, area fraction, and integrated density. Relative fluorescence corresponds to integrated density (intDen) value. Fold change in fluorescence was determined by dividing intDen values of infected images by the uninfected control.

RNA and Protein quantification

RNA was extracted using Trizol reagent as previously described [30] . Northern Blots to detect RNAs were performed with DIG-labeled RNA probes (Roche DIG Northern kit) specific for the mCherry gene in the reporter and subgenomic RNA [31]. qRT-PCR was performed using the Brilliant III Ultra-Fast SYBR Green QPCR kit (Stratagene, Agilent Technologies) with primers corresponding to the mCherry gene. Total protein was extracted from triturated mosquitoes in PBS. Primary rabbit anti-GFP antibody (Cat#ab290, Abcam, Cambridge, MA) or mouse anti-mCherry antibody (ab125096, Abcam, Cambridge, MA) were used to detect GFP and mCherry in western blot assays. Secondary goat polyclonal antibodies conjugated to horseradish peroxidase that are anti-rabbit (anti-rabbit, Cat#ab97051; anti-mouse; Cat#ab97023 Abcam, Cambridge, MA) were used to detect primary antibodies. Northern and Western blots were imaged using a Chemidoc XRS for chemiluminescent detection with HRP peroxide/luminol (Thermo-Scientific, Rockford IL) and CDP-Star (Sigma-Aldrich, St. Louis MO).

Aedes aegypti transformation

Aedes aegypti mosquitoes from the Higgs White Eye Strain (HWE) [32] were hatched and allowed to mature to adulthood. Females were given a bloodmeal consisting of defibrinated sheep’s blood (Colorado Serum Company. Denver, CO) 4 days before oviposition. On the day of injection, females were presented with an oviposition paper inside a 50ml conical tube, and eggs were collected for injection [33]. 1,736 eggs were injected at the posterior pole with plasmid pBG471 (Mariner Mos1 and reporter construct) and a helper plasmid containing the mariner transposase gene using an Eppendorf FemptoJet injector [34]. Following injection, eggs were returned to oviposition paper for 4-5 days to mature. The 327-hatched larvae (18% survival) were grown and separated into individual containers as pupae. The emerged adults were pooled together into 65 families depending on gender (1 male with 15 HWE wild-type (wt) females, or 12 female with 3 HWE wt males). Each of the 65 families was bloodfed three times and eggs were collected. The eggs were hatched and screened for eGFP expression in the eyes of the larvae on a fluorescent dissecting microscope. Nine families contained positive larvae and were outcrossed with HWE for 3 generations then were intercrossed for subsequent generations. Two families did not produce viable offspring, but the remaining 7 lines were stable through at least 10 generations. Each generation is screened for positive GFP eye expression as a marker for the presence of the transposon.

Mosquito infections

7-14 day old mosquitoes were orally infected with a blood meal containing GFP-expressing SINV (MRE16 5’dsGFP) [8]. Virus for bloodfeeds was prepared by infecting Vero cells with the respective virus 36-48 hours prior to bloodfeeds [35,36]. The virus was collected prior to feeding by scraping cells, centrifuging supernatant, removing media to desired volume (5mls) and then mixed with 5 ml sheep’s blood and 1 ml ATP [37]. Mosquitoes were fed through an artificial glass feeder with hog’s gut as the membrane. Bloodfed mosquitoes were separated and maintained following the feeding and the bloodmeal was titered to verify virus titers. Mosquitoes engorged with a bloodmeal were assumed infected, and at 7dpi and 14dpi the mosquitoes were screened for GFP (infection) and mCherry (reporter of infection) using a dissecting fluorescence microscope. Midguts were dissected from whole mosquitoes to examine reporter protein expression within internal tissues. Alternatively, whole mosquitoes were triturated, supernatant was filtered through 0.2uM syringe filter, and samples were assayed for virus titers by plaque assay [30].

Rationale for design of Sindbis virus subgenomic reporter constructs

Sindbis virus infection was detected by utilizing the viral subgenomic promoter to induce synthesis of a subgenomic RNA expressing mCherry only during active virus replication. Plasmids were engineered to contain alphavirus RNA elements (5’ UTR, SGP, 3’ UTR) with reporter proteins replacing the viral proteins. The 5’ end of the reporter genome was aligned with the transcription start site of the DNA pol II baculovirus IE promoter to allow full-length reporter RNAs to be transcribed in mosquito cells [8] . The 5’ and 3’ UTRs were maintained to allow the reporter RNA to be replicated in the presence of viral non-structural proteins [38,39]. The nonstructural proteins were removed from the 5’ open reading frame, except for the first 143 amino acids of nsP1 which contain conserved sequence elements required for RNA replication [9] (Figure 1). Initial constructs contained an eGFP gene fused to the nsP1 fragment that allowed for visual detection of cells expressing the full-length reporter RNA (Figure 1 and 2). The structural proteins, which are encoded from a subgenomic RNA at the 3’ end of the genome, were replaced with a mCherry fluorescent protein under control of the subgenomic promoter. The IE promoter drives expression of the reporter RNA in transfected cells which was evident by visual detection of eGFP expression (Figure 2A). Open reading frames located at the 3’ end of the genome cannot be translated from the full length reporter RNA due to stop codons upstream of the subgenomic promoter. Transcription from the subgenomic promoter, which is initiated by the nonstructural protein replication complex, generates a short subgenomic RNA that is competent for mCherry translation. The full-length reporter RNAs do not contain the open reading frames for the nonstructural proteins, so the reporter RNA can only be replicated and produce a subgenomic RNA only when replication competent virus infects the cell and provides the nsP1-4 proteins in trans. The infecting virus translates its own nonstructural replication complexes, which bind to the full-length reporter RNAs already present in the cell and replicates the reporter RNA. The subgenomic RNA is transcribed from the negative strand copy of the replicating full-length reporter RNA, and the fluorescent mCherry protein is translated from the newly synthesized subgenomic RNA (Figure 1B).

Download:

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

Figure 1. Diagram of Subgenomic reporter RNA constructs.

A) Transcription of each reporter RNA is initiated from the baculovirus IE promoter (IE), with 5’ UTR, 3’ UTR, and subgenomic promoter sequences derived from SINV (pBG426, 446, 447, 448, and 461) or CHIKV (pBG460). All constructs include the first 143 residues of nsP1 to provide conserved RNA sequence elements and 50 adenosines at the 3’ terminus. pBG426 encodes eGFP fused to nsP1 for detection of transfected cells expressing the reporter RNA. pBG446 does not contain GFP and is a simplified construct compared to pBG426. pBG447 and 448 have an additional virus specific (WEEV or CHIKV) subgenomic promoter inserted downstream of the SINV SGP. pBG461 contains the puromycin acetyltransferase gene fused to nsP1 for selection in stable cell lines. pBG460 is a CHIKV specific reporter construct engineered directly from CHIKV sequences. B) Diagram of reporter RNA replication and mCherry expression.

https://doi.org/10.1371/journal.pone.0084930.g001

Download:

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

Figure 2. Expression of Alphavirus Reporter RNAs in SINV Infected Cells results in mCherry Production.

C6/36 cells were transfected with pBG426 and subsequently infected with pBG167 TE3’2J SINV for 36 hrs. A) Brightfield, GFP, and mCherry fluorescence was determined for each sample. Uninfected (top) and infected (bottom). B) Fluorescence increase with different MOIs of SINV. C6/36 cells were transfected with pBG446 and infected for 36 hrs at the indicated MOI. Images were collected from each experimental well and the mCherry fluorescence was quantified. C) Increased fluorescence with increased infection times. C6/36 cells were transfected with pBG446 and infected at MOI = 10. Images were collected at the indicated times and fluorescence quantified.

https://doi.org/10.1371/journal.pone.0084930.g002

Reporter RNA constructs can detect infection in mosquito cells

Reporter RNA expressing plasmids transiently transfected into Aedes albopictus C6/36 cells show eGFP expression within 6 hours after transfection, indicating transcription of the reporter RNA is occurring within the transfected cells. The low number of eGFP positive cells reflects the low transfection efficiencies we commonly observe with C6/36 cell transfection. mCherry expression is not observed in uninfected cells, whereas mCherry fluorescence can be observed following infection with Sindbis virus (Figure 2A). Interestingly, we observed that cells that strongly expressed eGFP tended to have reduced mCherry expression upon infection, and conversely observed that cells expressing low levels of eGFP tended to display higher levels of mCherry following infection. mCherry fluorescence was quantified from images collected at given time points and multiplicity of infections. Significant expression of the reporter was detected with a virus MOI of as little as 0.02 at 36 hrs post infection and infection with an MOI of 10 provided significant mCherry fluorescence at 3 hours post infection (Figure 2B and 2C). These results indicate that infectious Sindbis virus can provide the replication complex in trans to activate the subgenomic promoter on reporter RNAs and express subgenomic RNA encoded proteins. We removed the eGFP to simplify the system (pBG446), and consistently observed mCherry expression during SINV infection but not in the absence of SINV (data not shown). Therefore, this system is able to visually detect SINV infection in cultured mosquito cells.

Reporter RNA expression is virus specific

To determine if activation of the subgenomic promoter on the reporter constructs is virus specific, we tested if related alphaviruses or unrelated flaviviruses could activate the Sindbis virus subgenomic promoter. C6/36 cells were transfected with pBG446 and infected with different viruses. mCherry expression was visually detected by fluorescence microscopy at 36 hours post infection (MOI 10) and fluorescence intensity was calculated. Infection with two strains of SINV (TE and MRE) resulted in a significant increase in mCherry expression, indicating that the detection was not strain specific (Figure 3A). However, infection with a flavivirus (KUNV) or a New World alphavirus (WEEV) [40] did not result in detectible mCherry expression (Figure 3A). Old World Chikungunya virus (CHIKV) was able to induce expression of mCherry, although to a lesser extent than Sindbis virus infection. This confirms that the viral replication complex has specificity for binding to the reporter RNA sequences, likely the 5’ or 3’ UTR in conjunction with the SGP, and that some cross-reactivity can be observed between related alphaviruses [38,39].

Download:

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

Figure 3. Sindbis subgenomic reporter detects infection with different SINV strains and similar old world alphaviruses, but not new world alphavirus or flavivirus infections.

A) C6/36 cells were transfected with pBG446 and then infected with different viruses (SINV (TE3’2J strain), SINV (MRE16 strain), CHIKV (La Reunion strain), WEEV (McMillan strain), and West Nile (Kunjin subtype). B) Virus-specific subgenomic promoter constructs for WEEV and CHIKV (pBG447 and pBG448), pBG446 (SINV specific), and pBG460 (CHIKV specific) were transfected into C6/36 cells and subsequently infected with SINV, CHIKV, WEEV, or uninfected mock control. All Images were taken at 36hrs post infection and mCherry fluorescence was quantified.

https://doi.org/10.1371/journal.pone.0084930.g003

The subgenomic promoter is not sufficient to induce subgenomic RNA synthesis

To determine if this system could detect a broad range of alphaviruses, an additional subgenomic promoter was inserted behind the SINV SGP. Theoretically, a reporter system with multiple SGPs from different viruses would be able to detect infection of any of the viruses. The WEEV or CHIKV subgenomic promoter sequence was added 3’ to the SINV subgenomic promoter in the SINV construct (pBG447 and pBG448- Figure 1). C6/36 cells transfected with pBG447 or pBG448 and infected with WEEV or CHIKV respectively, did not result in significant expression of mCherry (Figure 3B). However, a construct that was engineered and developed entirely based on CHIKV sequences showed detectable amounts of mCherry fluorescence during CHIKV infection but not with other viruses (Figure 3B). These results indicate that virus specific 5’ UTR, 3’ UTR, and subgenomic promoters are all required for the production of the 26S RNA from the reporter RNA and subsequent protein expression.

Stable cell line expressing reporter RNA

To determine if mosquito cells could be stably transformed to express alphavirus reporter RNAs, C6/36 cells were transfected with reporter constructs containing a puromycin acetyltransferase (PAC) gene (pBG461- Figure 1) and transfected cells were selected with puromycin. A bulk stable cell line was established that expressed the reporter RNA and detectable amounts of mCherry reporter protein only when infected with SINV but not in untransfected C636 or uninfected control cells (data not shown). Reporter detection was significantly higher with the stable cell line (pBG461) than transient transfection (pBG446) during SINV infection (Figure 4A and 4B). Infections were performed with a recombinant double subgenomic sindbis virus that expresses eGFP (SINV-GFP) to visually track infection (green) and confirm the reporter (red). Interestingly, cells that are not highly infected (low amounts of SINV-GFP) tend to have higher reporter mCherry expression. Although the entire stable cell line is resistant to puromycin, indicating that all cells are transformed with our construct, only 5-6% of infected cells expressed detectable amounts of mCherry (Figure 4C). This implies that there is an intricate balance of reporter RNA and infection that needs to be achieved in order for the reporter to be detected. The cell line has been maintained for over 18 months, with consistent ability to detect SINV infection (data not shown). These results indicate that reporter RNAs can be stably expressed in mosquito cells and detect alphavirus RNA replication.

Download:

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

Figure 4. Stably Transformed C6/36 cells express reporter mCherry during SINV infection A)

Brightfield, GFP, mCherry, and overlay images are shown from SINV-GFP infected C636 cells that are pBG446 transiently transfected (top) or pBG461 stably selected cells (bottom). Infected with an MOI of 10. Images collected at 48hours post infection. B) Average mCherry fluorescence is significantly higher in pBG461 stably transformed C6/36 cells than transiently transfected cells. C) Averages of total red fluorescence was calculated as a percent of total green fluorescence and is displayed for pBG446 transfected, pBG461stably selected, or C636 mock cells.

https://doi.org/10.1371/journal.pone.0084930.g004

Development of Reporter RNA Expressing Transgenic Aedes aegypti Mosquitoes.

Based on our in vitro results it seemed likely that this system could be used in mosquitoes to detect specific alphavirus infections. We developed transgenic mosquitoes that express SINV reporter RNAs and can express mCherry in response to infection in vivo. We utilized the Mariner transposon system (Mos1) to establish transgenic Ae. aegypti mosquitoes [41-44]. We replaced the IE3 promoter with an Ae. aegypti poly ubiquitin promoter (PUb) to provide more stable expression throughout the mosquito midgut than has been observed with the IE3 promoter [16]. The poly ubiquitin promoter construct (pBG471, Figure 5A) worked in vitro similar to the previous IE3 construct (pBG446) (data not shown). The transposon contained an eye-specific 3xp3 promoter that expressed eGFP in eyes to identify transgenic mosquitoes [22] . The reporter RNA sequence was inserted into the mariner transposon, which was co-injected with a helper plasmid expressing the mariner transposase [45] into 1,736 pre-dermoblast mosquito embryos. 327(18%) of the injected eggs were viable and hatched, were individually raised, screened for the eGFP eye marker, and outcrossed with wild type Higgs White Eye (HWE) Ae. aegypti mosquitoes. Each of the sindbis-induced mCherry (SIM) transgenic mosquito lines was separately hatched, screened, and females were bloodfed to maintain transgenic populations. Transgenic lines were verified by consistent eye specific eGFP expression and detection of the reporter DNA in the mosquito genomic DNA (Figure 5A).

Download:

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

Figure 5. Transgenic Aedes aegypti express mCherry following SINV infection

Verification of transgenic mosquitoes. A) Green eye marker for transgenesis in SIM mosquito line and Reporter DNA in transgenic mosquitoes. Genomic DNA was extracted from whole mosquitoes and used as template in PCR reaction with primers directed to the nsP1 and the SINV subgenomic promoter. The 600bp amplicon was present in SIM transgenic mosquitoes but not HWE mosquitoes. B) Transgenic Ae. aegypti expressing the mCherry reporter during SINV infection. SIM 177 transgenic family (left) and Higgs White Eye wt (right) infected with SINV-GFP are shown. 7 days post infection, images of dissected midguts with GFP (SINV) (top panels) and then the corresponding mCherry reporter expression (bottom panels).

https://doi.org/10.1371/journal.pone.0084930.g005

Sindbis virus can activate reporter RNAs in transgenic mosquitoes

To test if Sindbis infection could activate the subgenomic reporter RNA in the transgenic lines, we infected the transgenic mosquitoes with an eGFP expressing SINV (MRE16 5’dsGFP) by feeding female mosquitoes a bloodmeal containing 7 logs of infectious virus [36,46]. Visibly engorged mosquitoes were collected after the bloodmeal and kept for 4, 7, or 14 days post-infection. At the indicated times mosquitoes were cold anesthetized and eGFP and mCherry expression was assessed under a dissecting fluorescence microscope. Mosquitoes that were visibly infected with the SINV-GFP were sacrificed and dissected midguts were imaged. We were able to detect mCherry expression in a fraction of infected transgenic mosquitoes Approximately 2-4% of the SINV-infected transgenic mosquitoes displayed detectable mCherry expression. Similar low levels of reporter fluorescence was detected in the stable C636 cell line, indicating that this system needs to be further optimized to be effective at sensitive diagnostics and modeling of infection. However, despite the low level of reporter expression, mCherry fluorescence was positively detected in bodies and midguts of a subset of mosquitoes (Figure 5B, line SIM 177 shown). mCherry expression by fluorescence or Western blot was never observed in uninfected mosquitoes, indicating that reporter RNA activation and mCherry expression was specific to SINV infection.