Mosquito lines and maintenance
To introduce Wolbachia into Brazilian Ae. aegypti, an Australian line infected with the wMel strain21 was backcrossed for 8 generations to a natural mosquito population of Rio de Janeiro, Brazil24. Following the genetic background introgression, additional crosses and knockdown resistance (kdr) screening were undertaken to replicate natural insecticide resistance profiling and generate the line wMelRio. To assure a minimal variation in this profiling overtime, and sustain a homogeneous genetic background, wMelRio colony was refreshed with 10% wild males once in every five generations40.
To maintain wMelRio, immatures (i.e. larval stages L1 to L4) were reared in dechlorinated water, at 28 °C, and fed Tetramin Flakes (Tetra GmbH, Herrenteich, Germany) until pupal formation. Following adult emersion, groups of 1000 females and 800 males were sorted and kept in BugDorm cages (MegaView Science Co Ltd., Taiwan) at 25 °C, with 10% sucrose solution ad libitum. Every three days, females were fed human blood (from blood donation centers; see details under ‘ethical considerations’), through Hemotek artificial feeders (Hemotek Ltd, UK). Note that, to avoid arboviral contamination of our colony, all blood samples were formerly tested negative for DENV, ZIKV, CHIKV, MAYV and YFV by multiplex qPCR assays36,68. Egg-laying was induced by placing dampened strips of filter paper (i.e. partially immersed in water-containing cups) inside the cages for 2–3 days, after which they were gradually dried at room temperature. Strips loaded with eggs (i.e. ovistrips) were kept at room temperature until further use, either for colony maintenance or field release. Eggs older than 40 days were discarded due to a decay in overall quality60.
Egg releases
Mass-reared wMel-infected Brazilian Ae. aegypti, wMelRio, were released as eggs in Jurujuba (22°56′ 00″ S, 43°07′ 00″ W), a lower-middle-class community in the city of Niterói (state of Rio de Janeiro, Brazil). Located by the shores of Guanabara bay, this community has grown from a typical fisherman settlement, with informal occupancy, to a total population of 2797 residents in 890 houses. Jurujuba encompasses a total area of 2.53 km2, divided into seven smaller sectors (i.e. sub-areas or localities within the neighborhood): Ponto Final, Várzea, Brasília, Cascarejo, Praia de Adão e Eva, Peixe-Galo and Salinas.
wMelRio eggs were released in the field through special deployment devices, referred to as mosquito release containers (MRCs), which consisted of small white plastic buckets (19 cm height × 18 cm top diameter × 15.5 cm base diameter) with four small holes on the side wall, only a few centimeters away from the top lid. Each MRC was loaded with 1 L of water, 0.45 g of Tetramin Tropical Tablets (i.e. one and a half tablet) (Tetra GmbH, Herrenteich, Germany) and an ovistrip containing approximately 150–300 eggs. After six to seven days, about 80% of the immatures were pupae, and after 11 to 12 days, most of the adults had already emerged and left the device from the wall holes. Every 15 days, MRCs were checked and reloaded so that another rearing and release cycle could take place. Release sites were spatially distributed as evenly as possible (Supplementary Fig. S1), so as to maximize the spread of Wolbachia-harboring individuals and promote mating with their wild peers. The release strategy was optimized by splitting the sites into two groups, A and B, with alternate MRC loading schedules. Thus, while MRCs from group A were releasing adults, those from group B were being loaded with new ovistrips, water and food. In the following week, an opposite situation occurred, with MRCs from group B releasing adults. The release schedules, as well as the number of allocated MRCs, varied according to each Jurujuba’s sector (Supplementary Table S1).
Ethical considerations
All methods were carried out in accordance with relevant guidelines and regulations. Study protocol for Wolbachia field release was approved by the National Research Ethics Committee (CONEP, CAAE 02524513.0.1001.0008) and three government agencies: IBAMA (Ministry of Environment), Anvisa (Ministry of Health) and MAPA (Ministry of Agriculture, Livestock and Supply) to obtain the RET (Special Temporary Registry, 25351.392108/2013-96). Prior to mosquito releases, an informed consent was obtained from 70% of Jujuruba households. Also, a written informed consent was obtained from households that hosted BG-sentinel mosquito traps.
For the maintenance and mass-rearing of Wolbachia-infected Ae. aegypti, adult females were fed human blood from a donation center (Hospital Antonio Pedro, Rio de Janeiro State University), with supporting regulatory approval (CONEP, CAAE 59175616.2.0000.0008) We only used blood bags which would have been discarded by the donation center, mainly due to insufficient volume to meet their quality assurance policy. Samples had no information on donor’s identity, sex, age and any clinical condition, but were tested negative for several diseases, including Hepatitis B, Hepatitis C, Chagas disease, syphilis, HIV and HTLV, as part of the Brazilian Government routine screening.
For vector competence assays, human blood was obtained from Fundação Hemominas as part of a research agreement with Instituto René Rachou (Fiocruz Minas) (OF.GPO/CCO-Nr224/16).
Wolbachia field monitoring and density level assessment
Ae. aegypti field population was monitored with BG-Sentinel traps (Biogents AG, Regensburg, Germany), spread across Jurujuba in a semi-homogeneous fashion (Supplementary Fig. S2, Supplementary Table S2, Supplementary Datasheet S1). These monitoring sites were chosen among suitable households who formally agreed with hosting of a trap, and had to be reallocated according to necessity (i.e. household quits hosting the trap). Working traps were checked weekly by removing the catch bags (e.g. small meshed envelopes placed inside the BG-Sentinels to collect trapped insects) and bringing them to the laboratory for species identification and Wolbachia screening. Catch bags were barcoded according to the trap ID and site, so as to create a pipeline for field samples.
Screening for Wolbachia in Ae. aegypti samples was undertaken by qPCR. Briefly, individual DNA was extracted by homogenizing head/thorax body parts in Squash Buffer (10 mM Tris–Cl, 1 mM EDTA, 25 mM NaCl, pH 8.2) supplemented with Proteinase K (200 ug/ml) and incubating at 56 °C for 5 min. Extraction ended by enzyme inactivation at 98 °C for 15 min. DNA amplifications were carried out with FastStart Essential DNA Probes Master (Roche), using specific primers and probes to Wolbachia pipientis WD0513 and Ae. aegypti rps17 markers (Supplementary Table S3). Thermocycling conditions were set on a LightCycler 96 Instrument (Roche), as follows: 95 °C for 10 min (initial denaturation), and 40 cycles of 95 °C for 15 s and 60 °C for 30 s. Samples were analyzed using absolute quantification, by comparison to serial dilutions of either gene product, cloned and amplified in the pGEMT-Easy plasmid (Promega). Negative control samples were normalized between plates, and were used as reference to determine a minimum threshold for positive samples.
DENV and ZIKV isolation and replication in mosquito cells
ZIKV was kindly provided by Instituto Aggeu Magalhães (IAM, Fiocruz) through viral isolation of a symptomatic patient sample from Recife (PE, Brazil) in 2015 (ZikV/H.sapiens/Brazil/BRPE243/2015). DENV was sourced following a viral isolate from a patient sample diagnosed with Dengue type 1 in Contagem (MG, Brazil), also in 2015 (Den1/H.sapiens/Brazil/BRMV09/2015). Both ZIKV and DENV samples were accompanied by patients’ written consent (CONEP, reference number 862.912), being further catalogued into the national database of genetic patrimony and associated knowledge (SISGEN, access number AA1D462).
In vitro culture of viral particles were done as previously described36. Briefly, ZIKV and DENV were replicated in Aedes albopictus C6/36 cells, grown at 28 °C in Leibovitz L-15 medium (ThermoFisher) supplemented with 10% fetal bovine serum (FBS) (ThermoFisher). After seven days, supernatants were harvested and virus titers were assessed, first by Reverse Transcription (RT)-qPCR, and later by plaque assay with VERO cells grown under 37 °C, 5% carbon dioxide, in Dulbecco’s Modified Eagle Medium (DMEM) (ThermoFisher) supplemented with 3% Carboxymethylcellulose (Synth) and 2% FBS.
Vector competence assays
To perform vector competence assays with field samples of Ae. aegypti, ovitraps were mounted in both Ponto Final (Jurujuba) and Urca, a Wolbachia-free area in Rio de Janeiro. Ovitraps were collected from the field over 13 weeks, from April to June 2017, which corresponds to the time-frame between 14 and 16 months along the post-release phase in Ponto Final. Once in the insectary, eggs samples were reared to the adult stage in a controlled insectary environment (refer to ‘mosquito lines and maintenance’ for details).
For virus challenging assays through oral-feeding, young females (4–6 days old) were starved for 20 to 24 h, and subsequently offered culture supernatant containing ZIKV or DENV mixed with human red blood cells (2:1 ratio), using an artificial membrane feeding system36. It is important to mention that, as for the colony maintenance protocol, blood samples used here were also submitted to quality control prior to its use in the assays, mainly due to putative arbovirus contaminations which could affect the experimental outputs. Likewise, all samples were tested negative for DENV, ZIKV, CHIKV, MAYV and YFV by multiplex qPCR assays36,68. Oral-infections were performed twice for each virus. ZIKV was offered first from fresh (initial virus titer of 4.8 × 108 PFU/mL) and second from frozen culture supernatant (initial virus titer of 7.6 × 106 PFU/mL). In contrast, DENV was offered from fresh supernatants only (virus titers of 2 × 106 PFU/mL and 6.5 × 107 PFU/mL), since frozen versions failed to infect. Specimens were allowed to feed for one hour, after which engorged females were selected and maintained with 10% sucrose solution ad libitum, during the whole extrinsic incubation period. At 14 days post-infection (dpi), viral loads were assessed in heads/thorax extracts by RT-qPCR (refer to ‘Viral diagnosis’ for more details).
For saliva-mediated virus challenging assays, ZIKV and DENV pre-exposed females (14 dpi) from Jurujuba (Wolbachia +) and Urca (Wolbachia −) were starved for about 16 h (overnight) before being knocked down and kept at 4 °C for wings and legs removal. Salivation was induced by introducing a 10 µL sterile filter tip, pre-loaded with 5 µl of a solution [30% sucrose (w/v) diluted in 50% fetal bovine serum (FBS) and 50% DMEM medium], into the mosquito proboscis for 30 min. Saliva samples were individually collected, and 276 nL was intrathoracically injected into young naive females (Urca) using a Nanoject II hand held injector (Drummond), as previously described36,68. Each saliva sample was used to inoculate 8–14 naïve Wolbachia-free Ae. aegypti specimens, of which 8 were screened for infective particles. ZIKV and DENV were quantified by RT-qPCR at 5 dpi and 7 dpi, respectively (refer to ‘Viral diagnosis’ for more details). Overall Intrathoracic Saliva Infection index (OISI) was obtained by averaging the percentages (± SD) of infected individuals in each group.
Viral diagnosis
To identify ZIKV and DENV particles in individual samples, whole specimens were processed into head/thorax homogenates for RNA/DNA extraction with the High Pure Viral Nucleic Acid Kit (Roche), according to manufacturer’s instructions30. Extracted samples were diluted in nuclease-free water to a concentration of 50 ng/μL. ZIKV, DENV and Wolbachia levels, in vector competence assays, were quantified by RT-qPCR using TaqMan Fast Virus 1-Step Master Mix (ThermoFisher) and specific primers and probes (Supplementary Table S3). Reactions were run on a LightCycler 96 Instrument (Roche), using the following thermocycling conditions: 50 °C for 5 min (initial RT step), 95 °C for 20 s (RT inactivation/DNA initial denaturation), and then 40 cycles of 95 °C for 3 s and 60 °C for 30 s. Each RNA/DNA sample was used in two reactions, one with ZIKV, DENV or Wolbachia primers, and another with Ae. aegypti rps17 endogenous control30. Absolute quantification was achieved by comparing amplification profiles with standard curves generated by serial dilutions of their respective gene products, amplified from a cloned sequence in pGEM-T Easy vector (Promega). Negative control samples (no virus RNA) served as reference to fix a minimum threshold for positive ones. ZIKV and DENV loads were defined as their copy number per sample (head/thorax or saliva), while Wolbachia loads were relative quantifications to the rps17 reference gene. Here, it is worth noting that, while Wolbachia titer is naturally variable and dependent on its whole-body density, the overall expression of rps17 is stable and particularly suitable for internal controls in assays with adult females69, as demonstrated previously by us and others30,62,68.
Map creation and source codes
The satellite image map of Jurujuba was created with ArcGIS Desktop 10.7 (Esri Inc., https://www.esri.com/en-us/arcgis/products/arcgis-desktop/overview) using Google Earth (Google LLC) source code, under the license and in accordance with the fair use described in ‘https://about.google/brand-resource-center/products-and-services/geo-guidelines/’. Maps with geotagged MRCs and BG-Sentinel traps were created with ArcGIS Desktop 10.7 and OpenStreetMap source code (OpenStreetMap contributors), under the license CC-BY-SA 2.0.
Statistical analyses
Graphs and statistical analyzes were performed in GraphPad Prism 8 (GraphPad Software Inc., https://www.graphpad.com). Kruskal–Wallis test followed by Dunn’s post-hoc multiple comparisons were used to analyze Wolbachia density data from field-collected and colony samples. ZIKV and DENV loads in head/thorax extracts, from both oral and saliva-challenging samples, were compared using the Mann–Whitney U test. For all statistical inferences, ⍺ was set to 0.05.
Source: Ecology - nature.com
