Background

Batai virus (BATV) [1, 2] belongs to the genus Orthobunyavirus within the family Peribunyaviridae [3]. Initially detected in Culex gelidus trapped in Malaysia in 1955, it has since been identified in southern Slovakia (referred to as Calovo virus, CVOV) [4, 5] as well as in various European countries (for a review, see [6]). Another variant, Chittoor virus (CHITV), has been found in Anopheles barbirostris in India [7].

This zoonotic and especially veterinary important virus is transmitted by mosquitoes and biting midges, with mosquitoes considered as the most important vector group [8]. It affects a variety of vertebrate hosts, including pigs, horses, ruminants and various bird species. BATV has been detected in Africa, Europe and Asia. Human infections are rare and associated with mild flu-like symptoms. Infections of pigs, wild birds and harbour seals have been detected, and in ruminants severe outcomes such as abortions, premature births and genetic defects have been noted [1, 9].

The genomic structure of orthobunyaviruses is tripartite consisting of single-stranded RNA genomes [10]. This tripartite genome organisation leads to the appearance of reassortants, most frequently amongst co-circulating, genetically closely related strains [11].

Reassortments within the genus Orthobunyavirus may lead to viruses capable of inducing severe symptoms in humans [12]. Ngari virus, which carries the L- and S-segment of Bunyamwera orthobunyavirus and the M-segment of BATV, is associated with increased viral titres in infected mammalian cells as well as increased pathogenicity compared with the parental viruses [13, 14]. Ngari virus has been responsible for at least two outbreaks of haemorrhagic fever in humans in Central Africa between 1998 and 1999 [15, 16].

Surveillance studies conducted in Germany and Italy have confirmed the presence of antibodies against BATV in cattle, sheep and goats. Overall, these studies have demonstrated a seroprevalence up to 44% [17, 18]. However, in Europe, BATV-associated disease has not yet been reported in ruminants or humans. Notably, a BATV infection has been detected in a German captive harbour seal that exhibited encephalitis symptoms [9].

Furthermore, BATV has been repeatedly detected in Germany Anopheles maculipennis s.l., in Germany and twice in Italy [19,20,21]. Additionally, BATV has been identified in various other taxa, including Culex pipiens [22].

Recent laboratory studies with the Asian lineage of BATV showed that the mosquito species Culex quinquefasciatus as well as Culex tritaeniorhynchus are able to transmit the virus, whereas Aedes aegypti could only be infected [23]. British Cx. pipiens could also only be infected with BATV, while Aedes detritus was a competent vector under laboratory conditions [24].

Taken together, several mosquito species in Central Europe could potentially act as vectors for BATV. We recently showed that especially Culex torrentium, one of the three most frequent Culex species in Central Europe [Cx. p. biotype pipiens (Cx. p. pipiens), Cx. p. biotype molestus (Culex p. molestus) and Cx. torrentium] is a potent vector for arboviruses, e.g. West Nile virus (WNV) and Sindbis virus [25, 26]. In addition, the exotic species Aedes albopictus has infested more than 20 countries in Europe and is established along the Upper Rhine Valley in Germany and France and is known as a competent vector for chikungunya virus (CHIKV) and dengue virus (DENV) [27,28,29,30].

We assessed the vector competence of field-caught Culex species Cx. p. pipiens and Cx. torrentium as well as the invasive species Ae. albopictus, Aedes japonicus japonicus along with the laboratory colony of Ae. aegypti as a reference. Vector competence, in this context, refers the inherent ability of a mosquito to be infected and subsequently transmit the virus [31], confirmed by the presence of infectious viral particles in the mosquito’s saliva. Additionally, we investigated the impact of varying temperatures on the risk of BATV transmission by these different mosquito species.

Methods

Culex egg rafts were collected in Hamburg, Neugraben-Fischbek, Germany (longitude 53.467821/latitude 9.831346) in 2018 and 2019. Egg rafts were individually reared at room temperature with a 12:12 light:dark photoperiod. Molecular identification as Cx. p. pipiens and Cx. torrentium was performed using DNA extraction of a pool of five L1/L2 larvae per egg raft (DNeasy blood & tissue kit, Qiagen, Hilden, Germany) in a multiplex quantitative real-time PCR (HotStarTaq master mix kit, Qiagen, Hilden, Germany) as described [32].

Aedes albopictus were reared from a laboratory colony originally collected from Heidelberg, Germany (F26-29) and Ae. aegypti were reared from a historic laboratory colony from the Bayer company (Leverkusen, Germany). Ae. japonicus were reared from eggs collected with ovitraps in southwestern Germany (longitude 8.671355/lattitude 49.523888) in summer 2019. All adult mosquitoes were reared at 26 °C, with a relative humidity of 70% and a 12:12 light:dark photoperiod with 30 min twilight.

Females (7–10 days old) were starved for 24 hours (Aedes) or 48 hours (Culex). The artificial blood feeding was conducted at 24 °C for 2 hours. The blood meal consisted of 50% human blood (expired blood preservation), 30% of an 8% fructose solution, 10% filtrated bovine serum (FBS) and 10% virus stock, and was fed using a cotton stick (Culex) or two 50 µl drops (Aedes) on the bottom of the vial as previously described [33]. The virus stock contained BATV of the European lineage [strain 53.2, Genbank numbers HQ455790 (S-segment), HQ455791 (M-segment) and HQ457992 (L-segment)] isolated from An. maculipennis s.l. collected in Southern Germany [19] at a final concentration of 107 plaque forming units per millilitre (PFU/mL). BATV stock was produced and quantified via TCID50 on Vero cells (Chlorocebus sabaeus; CVCL_0059, obtained from ATCC, cat. no. CCL-81), results were converted in PFU/mL and the stock was diluted to reach a final concentration of 107 PFU/mL.

Only fully engorged females were used in the following experiments (ten females per vial). An 8% fructose solution was available via soaked cotton pads over the timeframe of the experiment. In general, mosquitoes were incubated for 14 days at 70% humidity and oscillating temperature profiles with mean temperatures of 18, 21, 24 and 27 °C and variations of ± 5 °C within 24 hours to mimic day–night temperature variations as previously described [29]. A diurnal temperature range of approximately 10 °C is commonly observed in the summer months in Germany [34]. The temperature maximum was reached in the middle of the light period, the temperature minimum in the middle of the dark period. Temperature profiles will be referred to by their mean temperature in the following text.

Culex mosquitoes were tested for all four mean temperatures in parallel. Aedes mosquitoes were tested at the highest mean temperature and at one lower temperature in parallel (21 °C for Ae. aegypti/Ae. japonicus and 24 °C for Ae. albopictus).

The salivation assay was performed at 14 days post infection (dpi) in alignment with previous studies [28, 29]. In summary, mosquitoes were anaesthetised using CO2 to facilitate the removal of legs and wings. The proboscis was then placed into a 10 µL tip containing phosphate-buffered saline (PBS) and incubated for 30 min. To test for viable virus particles, each saliva/PBS mix was incubated on Vero cells seeded in a 96-well plate for 7 days. To confirm the presence of BATV RNA, supernatant of Vero cells showing cytopathic effect were prepared for additional RNA testing as recently described by Jansen et al. [29]. RNA was isolated using the QIAamp Viral RNA Mini kit (Qiagen, Hilden, Germany). BATV RNA was detected using the quantitative real-time RT–PCR (qRT–PCR) as previously described [19] using the primers BATAI-Fwd (5′-GCTGGAAGGTTACTGTATTTAATAC-3′) and BATAI-Rev (5′-CAAGGAATCCACTGAGTCTGTG-3′) and the probe BATAI-P (5′-FAM-AACAGTCCAGTTCCAGACGATGGTC-BHQ). A series of a synthetic BATV (1.15 × 103, 1.15 × 104 and 1.15 × 105 copies) standards spanning the qRT–PCR product with an additional 5′ GTA and 3′ ACG overhang (5′-GTAGCTGGAAGGTTACTGTATTTAATACCGTAACAGTCCAGTTCCAGACGATGGTCAGTCACAGACTCAGTGGATTCCTTGACG-3′) was used as a positive control and for quantification of RNA copies within the sample, the threshold for positive PCR results was 100 copies per mosquito.

Every mosquito excluding legs and wings was homogenised using a micro homogeniser (Thermo Fisher Scientific, Waltham, Massachusetts, USA) in 500 µL Dulbecco’s modified Eagle medium (DMEM) and RNA was isolated using the 5× MagMax Pathogen RNA/DNA kit (Thermo Fisher Scientific, Waltham, Massachusetts, USA) as indicated in the manual. BATV RNA was detected via qRT–PCR as mentioned above. The mean number of RNA copies per mosquito was determined per temperature and species (log10 BATV RNA copies/mosquito).

We determined the feeding rate (FR, the number of engorged mosquitoes per number of mosquitoes that were offered an infectious blood meal) infection rate (IR, number of viral RNA positive mosquitoes per number of engorged mosquitoes), transmission rate (TR, the number of mosquitoes with BATV positive saliva per number of viral RNA positive mosquito bodies), transmission efficiency (TE, the number of mosquitoes with BATV positive saliva/number of engorged mosquitoes) and survival rate (SR, number of surviving mosquitoes on day 14 per number of engorged mosquitoes).

Results

For Ae. aegypti, infection rates of 40% at 21 °C and 29% at 27 °C were detected, the mean number of copies ranged between 4.1 and 4.2 log10 RNA copies/mosquito. Transmission could not be detected (Table 1).

Table 1 Results of vector competence studies with BATV for tested Aedes species
Full size table

Aedes albopictus females were only infected at the two higher temperatures of 27 °C and 24 °C. A rather low infection rate of 3.3% was detected at 24 °C (Table 1). At 27 °C the infection rate was slightly higher with 11.7%. Mean numbers of RNA copies per mosquito ranged between 4.8 and 7.6 log10 RNA copies/mosquito. Transmission could not be detected at either of the investigated temperatures.

For Ae. japonicus, infection but no transmission could be shown at the tested temperature of 27 °C (IR of 50%) and 21 °C (IR of 86%) (Table 1). Mean number of RNA copies per mosquito ranged between 5.03 and 5.99 log10 RNA copies/mosquito.

Culex p. pipiens females could be infected with BATV after incubation at 21, 24 and 27 °C, with infection rates between 8.1% and 50% (Table 2).

Table 2 Results of vector competence studies with BATV tested Culex species
Full size table

For Culex, no specific temperature effect was detected for the three higher temperatures, while no infection could be detected at the lowest temperature of 18 ± 5 °C. Mean number of RNA copies per mosquito ranged between 5.1 and 6.0 log10 RNA copies/mosquito. Transmission could not be detected for Cx. p. pipiens. Culex torrentium showed infection at all temperatures, there was no hint towards a temperature dependency concerning the infection. Overall infection rates were higher compared with Cx. p. pipiens, with values between 22.6% and 93.3% (Table 2). C. torrentium was also able to transmit the virus at the highest of the tested temperatures with a low transmission efficiency of 3%. At this temperature, mean number of RNA copies per mosquito reached the highest values of 6.0 log10 RNA copies/mosquito, in comparison with 4.6–5.4 log10 RNA copies/mosquito at the other temperatures (Table 2).

In addition, we measured the survival of Cx. p. pipiens, Cx. torrentium, Ae. aegypti and Ae. albopictus (only at 24 °C) at 14 days after infection. Independent of the incubation conditions or the tested mosquito species, survival rates never fell below 79%. For Cx. torrentium even survival rates of 100% were detected at the highest temperatures.

Discussion

The presence of BATV antibodies has been studied in Eastern Germany across various livestock species including sheep, goat and cattle [17, 18, 35]. These studies have revealed seroprevalences as high as 44.7%. Antibodies have also been detected in bovine serum samples from the Novarra region in Northern Italy in 2011 [36]. Furthermore, BATV RNA has been detected in aedine and culicine mosquitoes in Germany [19, 22] and in a pool of An. maculipennis s.l. mosquitoes in Italy [21]. These findings collectively suggest that the virus is circulating in Central Europe, particularly in regions such as Eastern Germany. Despite the absence of documented BATV infections in humans, a BATV infection has been detected in Germany in a captured harbour seal showing symptoms of encephalitis [9].

However, no further documented BATV infections have been reported in humans or livestock in Central Europe. Despite this, it remains crucial to continue investigating BATV, as the overall risk of arbovirus transmission is on the rise.

In recent years, the risk of introduction and establishment of arboviral transmission cycles within Central Europe has grown. Notable examples are CHIKV epidemics in Italy and dengue virus (DENV) case clusters in Spain, France and Italy have been described [37, 38]. These outbreaks are attributed to factors such as the expanding distribution of the known CHIKV and DENV vector Ae. albopictus, as well as rising temperatures. Furthermore, there is also circulation of endemic viruses such as WNV. It emerged in Germany in 2018, and caused epidemics in Greece and Italy since 2010 [39, 40] with higher temperatures being one of the driving factors [25].

Specifically, BATV could pose a threat, parallels can be drawn from Cache Valley virus (CVV) another member of the Bunyamwera serogroup. In small ruminants, CVV infection may lead to foetal death or severe malformation of the foetus [41]. Cache Valley virus circulates in North, Central and South America and has been isolated from over 40 mosquito species [41]. Although human cases are rare, symptoms can range from mild illness with fever to severe cases of encephalitis. Notably, recent studies in the USA have revealed an increase in CVV infections. They showed that the invasive species Ae. albopictus transmits this virus and that Ae. albopictus is widespread in the area where CVV cases have been detected [42]. Based on these findings, we conducted tests on Aedes species, particularly the invasive ones, to assess their potential impact on the transmission of BATV in Central Europe.

Furthermore, reassortments within the Bunyamwera serogroup occur naturally. The most notable example is a reassortment event between BATV and Bunyamwera virus, resulting in the emergence of Ngari virus. Reassortment has led to an increase in pathogenicity, contributing to two major haemorrhagic fever outbreaks in humans in Africa [15]. Given their close genetic relationship, knowledge about competent vectors for BATV could inform risk assessments related to Ngari virus outbreaks.

As observed before, the laboratory colony of Ae. aegypti could be infected with BATV at both of the tested temperatures (27 °C, 21 °C), which were chosen to reflect a tropical and a more moderate temperature. No transmission could be detected, which is in line with previous studies [22]. For the other studied invasive Aedes species (Ae. albopictus and Ae. japonicus) similar results were observed at the two tested temperature, being 27 °C and 24 °C for Ae. albopictus and 27 °C and 21 °C for Ae. japonicus (infection but no transmission). None of the tested Aedes species were competent vectors. However, the sample size of Ae. japonicus was smaller than that of the other tested species. With six tested mosquitoes, the minimal detection limit is a TE of 16%, but TEs below this might already be biological relevant, therefore Ae. japonicus could still be competent vector here defined as the presence of viable virus particles within the saliva. The number of at least 30 investigated specimens per condition is well established in the field of vector competence studies and allows to determine TEs as low as 3%. Biologically relevant vector competence can be determined (TE > 3%), but the effort of the experiments is still proportionate.

Vector competence studies with Ae. albopictus and CVV already revealed that different lineages of CVV have a remarkable effect on transmission [42]. Although no transmission of BATV by Ae. albopictus was detected in this study, Ae. albopictus still could possibly contribute to the transmission of other strains of BATV if they would be introduced. Therefore, it would be of interest to test Ae. albopictus and other Aedes species for different BATV strains.

No obvious effect of BATV infection on survival could be seen for any of the tested species. This includes Cx. torrentium the only species that tested positive for BATV in the saliva in this study. This is in line with recently published results, where negative effects on survival could only be shown for Ae. detritus, but no changes in mortality could be observed for Ae. aegypti or Cx. pipiens. [24].

As BATV is transmitted by over 40 [41] different species from different genera, we included additional information regarding two specific species: Cx. p. pipiens and Cx. torrentium. These species are most abundant Culex species in Europe and previous research has demonstrated that these two serve as potential bridge vectors [43]. Recently, it has been shown for the Asian lineage of BATV, that Cx. quinquefasciatus as well as Cx. tritaeniorhynchus are competent vectors [23]. In contrast, it has been shown that a Cx. pipiens laboratory colony (hybrids from Cx. p. pipiens and Cx. p. molestus) was not able to transmit the European variant of BATV [24]. Our results for the field-caught Cx. p. pipiens are in line with the results obtained for Cx. pipiens [24], which were also not able to transmit BATV. However, at the highest temperature of 27 °C, Cx. torrentium, the more prevalent species in Central Europe [44] is able to transmit the virus, but only with a low transmission efficiency of 3%. Our data for Cx. torrentium show that highest copy numbers in mosquito bodies are reached at the highest temperature.

It has been described that at 20 °C, the extrinsic incubation period of BATV is at least not longer than 7 days in Ae. detritus and moreover this study showed that the transmission rate is higher at 7 days compared with 14 days post infection [24]. Therefore, it would be very important to further analyse whether this is also the case for Cx. torrentium. To be an effective vector in nature, vector capacity – rather than just vector competence – plays an important role. Vector capacity encompasses physiological, ecological and environmental factors related to the vector, host and pathogen. Key factors include blood-feeding behaviour, temperature and abundances [31]. However, currently neither Cx. torrentium nor Ae. detritus seems to be the relevant vector responsible for the high seroprevalence detected in several surveillance studies in Eastern Germany [17, 18]. Ae. detritus is a halophilic species predominantly distributed in coastal areas [45] and not in the regions described in the studies [17, 18] and Cx. torrentium only transmits BATV at high temperatures with a TE of only 3%.

Conclusion

Within this study, Cx. torrentium was found to be a potential vector for BATV at high temperatures but with a low TE. To unravel the current infection cycle, more mosquito species need to be analysed for their vector competence if technically possible. BATV, for example, has been detected in Germany in a pool of An. maculipennis s.l. [19] and in pools of different mosquitoes also containing different Anopheles species as well as Ae. vexans [21]. Due to their host feeding patterns Ae. vexans are important vectors for the transmission from non-human mammals to humans [43]. Combined with the mass appearance of species upon flooding events, they could be an important vector and therefore would be an interesting species to test whether mosquitoes from the field are available. The same is true for An. maculipennis s.l.