1.1. Avian and rodent samples.

Avian zoo samples were taken between 2016 and 2024, with exception of 2017 where birds were not sampled, from a total of 155 birds of 10 different species (Rhea americana; Eudocimus ruber; Bucorvus abyssinicus; Dromaius novaehollandiae; Pavo cristatus; Amazonetta brasiliensis; Gallus gallus; Chauna torquata; Leptoptilos crumenifer; Ara ambiguus). Samples from the Ecolotheque include sera from 33 birds of 7 different species (Anas platyrhynchos; Cairina moschata; Anas platyrhynchos domesticus; Meleagris gallopavo; Anser anser domesticus; Gallus gallus; Numida meleagris), collected only in October 2024, and blood and organs from a dead helmeted guineafowl (Numida meleagris) from March 2024. For these animals, 0.3 to 1.5 mL of heparinized plasma or serum were taken from the medial metatarsal vein or the brachial vein. The blood was centrifuged at 2000 G for 10 min, the serum was then separated from the clot and stored at -20°C until analysis.

Cloacal swab was taken by inserting the swab into the cloaca of the bird and then returning it to the liquid containing an RNA preservative (stored at -20°C until analysis). Complete brain and part of the liver and spleen were also taken after necropsy for a few animals of interest.

Great tits (Parus major) were blood-sampled throughout 2024 at the CEFE (a peri-urban area close to Montpellier zoological park), reaching 36 Dried Blood Spots (DBS, corresponding to 10–15 µL of blood deposited on Whatman Saver Cards). A total of 39 cloacal swabs were collected.

We collected two pools of pigeon droppings (Columba livia) 21 days after the death of three pigeons whose corpses were found in the same loft in Montpellier Center (urban area). The samples were placed in two 50 mL tubes and stored at -20°C until analysis.

Moreover, we analysed 31 DBS of black rats (Rattus rattus) captured on the Montpellier zoological park site in 2023 and 2024. Blood spots were collected from the heart on filter paper (Whatman), then dried for 24 hours and stored at room temperature in an individual resealable plastic bag with a desiccant pack (silicagel).

1.2. Molecular xenomonitoring (MX) samples.

Mosquito trapping at Montpellier zoological park consisted in two traps: one BG-Pro trap (Biogents, Germany) connected to the electrical grid – allowing a continuous functioning between two successive pick-ups – placed next to the enclosure of the greater rheas in Montpellier zoological park, and a second BG-Pro trap on 64h-autonomy battery (Powerbank 27 000 mAh) placed inside the enclosure to directly monitor mosquito presence and potential virus circulation in close proximity to the birds. The mosquito sampling at the Ecolotheque educational farm consisted of a BG-Sentinel trap connected to the mains, placed inside the duck enclosure and their pool. All three traps were associated with a 3D-printed plastic MX adapter to collect the faeces of mosquitoes kept alive in the adapter [33]. Two types of baits were used to attract Culicidae species: CO2 generated by BG-CO2-Generator (Biogents) and BG-Lure-Mozzibait [34,35]. Inside the adapter, a large feeder (which is twice the volume of the regular feeder) composed of a cotton ball soaked in a 10% honey water solution was placed to maintain humidity and sustain the trapped mosquitoes alive during the whole week. At the bottom of the MX adapter, a ring of regular kitchen aluminium foil (10µm thickness) was used to collect mosquito faeces. The traps were collected every week between the 15th of April and the 23rd of October 2024 (i.e., 28 sessions) for the first trap at Montpellier zoological park, between the 23rd of July and the 8th of October (i.e., 12 sessions) for the second trap, and from the 22nd of July to the 7th of October (i.e., 12 sessions) at the Ecolotheque. In the laboratory, the mosquitos trapped in the MX adapter were firstly collected with an entomological vacuum Prockopack, frozen-killed at -20°C, sexed and identified to the genus level. The aluminium ring was collected, then provided to IAGE (Ingénierie et Analyse en Génétique Environnementale) and stored at -80°C before RT-dPCR analysis.

In addition to these routine sampling sites, two additional MX collections were conducted opportunistically in response to confirmed autochthonous arboviral cases. In Perols, 3 MX traps were deployed between July 5 and 8, 2024. In Vendargues, 3 MX traps were deployed in three periods: September 20–23, September 23–30, and September 27–30 2024. These opportunistic trappings were also supplemented with a BG-CO2 Generator (Biogents) and a BG-Lure Mozzibait.

1.3. Environmental water samples.

Environmental water samples were collected from the same four sites where mosquitoes were sampled, using standardized, sterilized 1 L polypropylene bottles. The water was first homogenised at the surface and at depth, then directly recovered with the bottles. The first site was a large concrete pool (14,000 m³) within the enclosure of the greater rheas at the Montpellier zoological park. The second site was the Montpellier Ecolotheque, where a single pooled environmental water sample was generated with various samples from different avian resting areas, including small concrete pools and bathing zones used by geese, chickens, ducks, and guinea fowl. To ensure sampling homogeneity, the water column was gently mixed, incorporating both surface and deeper water layers while avoiding sediment disturbance, before collecting a 1 L targeted sample. Environmental water sampling was conducted monthly between July and September 2024. After collection, the water samples were kept in 4°C coolers and refrigerators, before analyses within 48h at IAGE laboratory. During each sampling event, after homogenization, the water physico-chemical parameters — pH, conductivity, salinity, and temperature — were measured using a HI98195 multi-parameter probe (Hanna Instruments, USA), and the turbidity was assessed through the measure of absorbance (570 nanometer), realized in laboratory with microplate spectrophotometer Thermo Scientific Multiskan SkyHigh.

Like the MX trapping, environmental water sampling was also conducted opportunistically during mosquito control operations. Hence, in addition to the routine sampling sites, two additional water collections were performed in response to confirmed autochthonous dengue cases in Perols and Vendargues. These samples were obtained from multiple Culex pipiens and Aedes albopictus larval breeding sites. In Perols, water samples were collected between July 5 and 8, 2024. In Vendargues, water samples were collected on September 20, 23, and 26, 2024.

2.1. ELISA test.

The presence of anti-orthoflavivirus antibodies was detected using the ID Screen Flavivirus Competition ELISA kit (IDVet, France) with sera or DBS of birds (zoo, CEFE) and rats. The plates are pre-coated with a flavivirus envelope protein. Fifty microliters of serum were added to each designated well and incubated for 1.5 hours, allowing any antibodies against orthoflavivirus in the serum to bind to the coated antigen. Afterward, 100 µL of secondary antibody is added and incubated for 30 minutes, during which it attaches to the coated antigen if antibodies are present in the serum. The plates are then washed, and 100 µL of the substrate solution is added for 15 minutes. A colour change indicates the presence of the secondary antibody, signalling a negative test result (DO > 0.788 = negative result; DO < 0.0630 = positive result). For DBS 2 punches were resuspended in a 2 mL tube with 100 µL of PBS1x/BSA/Tween 0.1% buffer and incubated overnight at 4°C, following the IDVet kit protocol. Then, 50 µL of the resuspended solution was added to the plate and incubated overnight at room temperature. After rinsing, 100 µL of conjugate was added to the plate.

2.2. Microneutralization assays.

Vero E6 cell cultures were used for microneutralization tests (MNT). These cells were cultured with Dulbecco’s modified Eagle’s medium (DMEM; Sigma-Aldrich) supplemented with 10% heat-inactivated fetal bovine serum (Dutscher) and Penicillin/Streptomycin antibiotics (Gibco), kept in an incubator at 37°C and 5% CO2. The MNT was carried out in flat-bottomed 96 well microplates. 15 µL of avian serum (zoo, CEFE) was subsequently added to the first column to obtain a starting dilution of 1/5. From this well, three-fold dilutions were made by pipetting 50 µL of each well in the next, until the last column. For DBS samples, 1 punch in 50 µL of PBS1x/BSA/Tween 0.1% buffer, incubate overnight at 4°C, add the 15 µL in the plate. The sera were mixed with 50 µL of virus suspension (at 100 tissue culture infectious dose 50 (TCID50) of USUV (France2018, MT863562) incubated at 37°C for 90 min to allow neutralization of the virus by antibodies. For DBS samples, incubated overnight at 37°C. Wells were then supplemented with 100 µL of Vero E6 cell suspension (DMEM with 2% foetal bovine serum) and incubated 3–6 days at 37°C and under 5% CO2. After this time, microplates were read under the microscope to determine the presence or absence of cytopathogenic effect (ECP) in samples and in positive and negative sera controls. Each sample was assigned a titer that was the reciprocal of the dilution in that well. Sera that showed neutralization at dilutions ≥ 1:15 were considered positive. Duplicates were performed to confirm positive results. For the samples collected between 2016 and 2019, MNT assays were conducted under identical conditions to verify the absence of anti-WNV antibodies using WNV lineage 2 (strain MT863560).

3.1. RNA extraction, RT-qPCR and sequencing.

RNA extractions were carried out of organs, cloacal swabs, sera, droppings and DBS using the QIAamp viral RNA mini kit (Qiagen), following the manufacturer’s protocol. For DBS sample, one punch per card was individually rehydrated in 175 µL of PBS1x/BSA/Tween 0.1% buffer and incubated overnight under agitation at 4 °C. A volume of 140 µL of the resulting suspension was then used for RNA extraction according to the kit protocol. For organs and droppings, samples were directly ground in 140 µL of PBS1x, and the entire lysate was used for extraction. Subsequent extraction steps were performed according to the QIAamp Viral RNA Mini Kit protocol. All RNA samples were eluted in 50 µL of elution buffer, as recommended. RNA quantity and purity were assessed by measuring absorbance using a Nanodrop spectrophotometer. Reverse transcription (RT) was then performed on 500 ng of total RNA using the LunaScript RT SuperMix Kit (New England BioLabs).

Quantitative PCR (qPCR) was performed using the SYBR Green I Master mix (Roche) on a LightCycler 480 instrument (Roche), with the following primers targeting the Usutu virus genome: forward primer 5′-AACAGACGGTGATGCGAACT-3′ and reverse primer 5′-TACAGCTTCGGAAACGGCTT-3′ [36]. Thermal cycling conditions included an initial activation at 95 °C for 10 minutes, followed by 45 cycles of amplification at 95 °C for 15 seconds, 60 °C for 15 seconds, and 72 °C for 15 seconds. A melting curve analysis was subsequently performed with the following steps: 95 °C for 5 seconds, 65 °C for 1 minute, and a continuous increase to 97 °C.

To validate selected results, a one-step TaqMan RT-qPCR assay was also conducted using RNA extracted from the same samples. As this was a one-step assay, reverse transcription and amplification were conducted in a single reaction. A volume of 5 μL of each RNA extract was directly used as input. Reactions were carried out using the SuperScript IV One-Step RT-PCR System (Invitrogen) with primers and probe targeting the NS5 region of the Usutu virus genome: forward primer 5′-AAAAATGTACGCGGATGACACA-3′, reverse primer 5′-TTTGGCCTCGTTGTCAAGATC-3′, and probe 5′FAM-CGGCTGGGACACCCGGATAACC-3′TAMRA [37]. Thermal cycling was conducted on a LightCycler 480 instrument (Roche) under the following conditions: reverse transcription at 50°C for 15 minutes; enzyme activation at 95°C for 2 minutes; followed by 50 amplification cycles consisting of denaturation at 95°C for 15 seconds and combined annealing/extension at 60°C for 1 minute. Fluorescence data were collected at the end of each extension step. A virus-derived RNA extract from a viral culture was included in each run as a positive control.

To confirm some samples, amplicons obtained at the end of the RT-qPCR were purified using the Gel and PCR Clean-up kit (NucleoSpin) and then sent for sequencing. Sanger sequencing was carried out by GENEWIZ from Azenta Life Sciences starting from 20 µl of purified amplicons, using one of the primers employed for amplification.

3.2. Environmental water analyses and xenomonitoring analyses.

3.2.1. RNA/DNA extraction for water analyses. Extractions of total DNA/RNA for dPCR analyses were performed by IAGE Company following the patented method [38] with an optimized pre-treatment. Briefly, the water samples were kept at 4°C after sampling until nucleic acid extraction with volumes comprised between 100 mL and 1 L. The Environmental water sample were then thoroughly mixed, 15 mL were taken from each homogenized sample and centrifuged for 15 min at 3234 g at 4°C. For each sample, pellet was conserved on ice until further treatment. Supernatant was recovered and applied to an Amicon Ultra-15 Centrifugal Filter Unit (cut-off: 10 kDa, Cat. No. UFC901096, Merck Millipore, Germany) previously hydrated with 2,5 mL of MQ water and spun for 10 min at 3234 g, at 4°C. The tube was then centrifuged for 35 minutes at 3234 g, at 4°C. If the resulting volume was inferior to 500 µL, the tube was spin again for 10 minutes at 3234 g, at 4°C. The resulting volume was used to resuspend the corresponding pellet for each sample. Then, 400 µL of Nucleo Protect VET Buffer (Cat. No. 740750, Macherey Nagel, Germany) was added before grinding with two 3 mm stainless steel beads on a Geno/Grinder instrument (Spex Sample Prep LLC, USA) for 15 s at 1500 cpm, thrice. Extractions were then performed using the Indimag Pathogen kit (Cat. No. SP947457, Indical Bioscience, Germany) following manufacturer’s instructions.

3.2.2. RNA/DNA extraction for xenomonitoring analyses. The aluminium foil from the MX adapter was coiled and placed individually in 50 mL plastic tubes (Greiner Bio-one, Cat. No.227261) before being soaked in a 1.2 mL VXL lysis buffer (Cat. No.1069974, Qiagen, Germany) and vortexed for 5 min. The supernatant was recovered in 1.5 mL tubes and 200 µL were used for DNA/RNA extraction using the Indimag Pathogen Kit (Cat. No. SP947457, Indical bioscience, Germany) following manufacturer’s instructions. Then, the sample pre-treatment for total nucleic acid extraction followed the same protocol as the water sample analyses.

3.2.3. RT-dPCR. Digital-PCR reactions were conducted using a simplex assay developed by IAGE on a QIAcuity Eight Plateform System (Qiagen, Germany), using the QIAcuity One-Step Viral RT-PCR Kit (Cat. No. 210212) and QIAcuity Nanoplate 26 K 24-wells (Cat. No. 250001). The dPCR detection system was adapted from [39] with primers USUTU F (5′-CGTTCTCGACTTTGACTA-3′) and USUTU R (5′-GCTAGTAGTAGTTCTTATGGA-3′) optimized for enhanced specificity to USUV by the IAGE company (https://www.iage-france.com). The dPCR reaction mixtures were prepared in a preplate as follows. For each reaction 10 µL of 4 × One-Step Viral RT-PCR Master Mix, 0.4 μL of 100 × Multiplex Reverse Transcription Mix, 1 μL of the primers/probes mix (final concentration 0.45 µM of both primers and 0.125 µM of probe), and 2 µL of DNA extract (for both xenomonitoring assays and water samples) were combined with H2O up to 40 µL final reaction volume. Reaction mixtures were transferred into a QIAcuity Nanoplate and the plate was loaded onto the QIAcuity Eight instrument, a fully automated system. The workflow included (i) priming and rolling step to generate and isolate the chamber partitions (26,000 partitions), (ii) amplification step under the following cycling protocol: 50 °C for 40 min for reverse transcription, 95 °C for 2 min for enzyme activation, 95 °C for 5 s for denaturation, and 58 °C for 60 s for annealing/extension for 40 cycles; and (iii) imaging step by reading fluorescence emission after excitation of the probe at the appropriate wavelength. Data were analysed using the QIAcuity Software suite V3.1.0.0.

1.1. Captive exotic Avifauna.

Analyses were carried out on 155 specimens of 10 different avian species across a period running from 2016 to 2024 (Fig 1). No samples were available in 2017.

Using ELISA serological tests, we identified antibodies against Orthoflaviviruses in 41 out of 155 birds (26.5%, CI95% [19.5-33.4]), and 15 of the 41 were greater rheas (Rhea americana), (36.5%, CI95% [21.8-51.3]; Fig 2, S1 Table). The Orthoflavivirus incidence rate, used here to calculate the rate of seroconverted birds per species over a given monitoring period (per 10 bird-years), was determined based on 127 individuals from 9 different species. The highest incidence rate was observed in Rhea americana 3.44 [1.72;6.15] (other birds: Bucorvus abyssinicus 1.43 [0.29;4.17]; Pavo cristatus 0.71 [0.09;2.58]; Ara ambiguus 0.62 [0.02;3.48]; Gallus gallus 0.59 [0.01;3.28]; Amazonetta brasiliensis 0.57 [0.07;2.06]; Eudocimus ruber 0.43 [0.18;0.84]). We therefore observed the seroconversion of many birds throughout our study, with incidence data indicating that, for most of them, this seroconversion occurred between 2016 and 2019 but has also continued up to 2024. Notably, new animals seroconvert each year.

Of the 41 ELISA-positive zoo birds, 37 could be tested for USUV-specific antibodies by virus-specific microneutralization test (MNT), as we no longer had material for the other 4 (Eudocimus ruber). Of these 37 birds tested by MNT USUV, 28 tested positive for the presence of antibodies against USUV (75.6%, CI95% [61.8-89.5]), including the 15 greater rheas that tested positive by ELISA (53.5%, CI95% [35.1-72]), representing the total number of individuals tested (Fig 3, S2 Table). Since 2021, only one rhea has not developed neutralizing antibodies against USUV. In greater rheas, we observed different patterns of immunity maintenance. Some individuals exhibited a peak in neutralizing antibody titers in one year, followed by a significant decline, while others experienced fluctuations with alternating increases and decreases in antibody levels over time. Additionally, some individuals showed an increase in antibody titers after a previous decline, suggesting regular reinfections and, consequently, continuous circulation of USUV within the Montpellier Zoo. Notably, microneutralization assays for WNV performed on samples collected during the early years (2016–2019) of the study did not reveal any WNV positive cases among the avian species monitored throughout the study.

In parallel with our serological analyses, we extracted and analysed organs, blood samples, and cloacal swabs by RT-qPCR mainly from 2024, along with some organs and cloacal swabs dating from 2020 to 2023. From bird sera and cloacal swabs collected in August and October 2024, we detected the presence of USUV RNA in 11 of the 26 birds tested (42.31%, CI95% [23.32-61.30]), including 5 positive greater rheas out of 15 (33.3%, CI95% [9.48-57.19]) (Table 1, S3 Table). We were unable to obtain a complete sequence of our samples due to the limited amount of material available. Nevertheless, several amplicons were sequenced, confirming the positivity of the samples with sequences likely related to the Africa 3 lineage strains (≈98% identity) (S1 Appendix).

Download:

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

Table 1. *+Molecular analyses of rheas by RT-qPCR.

https://doi.org/10.1371/journal.pntd.0013506.t001

1.2 USUV infections in mosquitoes’ faeces.

Since these greater rheas appear to be particularly susceptible to USUV infection, we set up a xenomonitoring mosquito trap near their enclosure from May to September 2024 to determine whether we could detect USUV viral RNA. Only mosquito faeces collected from the trap were analysed, not the mosquitoes themselves. We detected a positive signal in September, indicating the presence of the virus in mosquitoes near these animals. A total of 30 Aedes (24 females, 6 males), 208 Culex (148 females, 60 males) and 2 Culiseta (1 of each sex) were captured inside the enclosure during the 28-sessions course of the sampling, and 13 Aedes (12 females, 1 male), 360 Culex (239 females, 121 males) and 32 Culiseta (29 females, 3 males) outside of it, in 12 sessions (S2 Appendix). The faeces of these mosquitoes were analysed, and USUV was detected in samples collected during the week of September 3–10, 2024 for the traps inside the rheas’ enclosure. In this trap, a total of 32 female and 28 male Culex spp. mosquitoes were captured, along with eight female and one male Aedes albopictus.

1.3. USUV detection in rheas’ pool water.

The particularly high circulation of USUV among greater rheas led us to question the possibility of detecting this virus in the water of their concrete pool, which was emptied and cleaned monthly. Based on the hypothesis that these animals might shed the virus in their faeces, we collected water samples from the pool from July to September 2024, during the period when arboviral circulation is known to be most intense in the region. Interestingly, we detected a positive signal in the pool water in July, before detection in other matrices such as mosquitoes, highlighting the relevance of this matrix for the viral detection of the USUV (Table 2).

Download:

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

Table 2. Molecular table analyses of the zoo pool of the greater rheas by dPCR with physicochemical parameters. The limit of detection (LOD) for the environmental water sample is: 5.97 × 10⁴ copies/L.

https://doi.org/10.1371/journal.pntd.0013506.t002