The microbial RNA metagenome of Aedes albopictus (Diptera: Culicidae) from Germany

Aedes albopictus is a highly invasive mosquito species that has become widespread across the globe. In addition, it is an efficient vector of numerous pathogens of medical and veterinary importance, including dengue, chikungunya and Zika viruses. Among others, the vector potential of mosquitoes is influenced by their microbiome. However, this influence is very dynamic and can vary between individuals and life stages. To obtain a rough overview on the microbiome of Ae. albopictus populations in Germany, pooled female and pooled male individuals from seven German locations were investigated by total RNA sequencing. The mosquito specimens had been collected as larvae in the field and processed immediately after adult emergence, i.e. without females having fed on blood. RNA fragments with high degrees of identity to a large number of viruses and microorganisms were identified, including, for example, Wolbachia pipientis and Acinetobacter baumannii, with differences between male and female mosquitoes. Knowledge about the natural occurrence of microorganisms in mosquitoes may be translated into new approaches to vector control, for example W. pipientis can be exploited to manipulate mosquito reproduction and vector competence. The study results show how diverse the microbiome of Ae. albopictus can be, and the more so needs to be adequately analysed and interpreted. Supplementary Information The online version contains supplementary material available at 10.1007/s00436-022-07576-7.


Introduction
Aedes albopictus is a thermophilic mosquito species native to the Asian-Pacific region. Due to globalisation and its high ecological and physiological plasticity, it has become widespread in many regions in the world. Presently, Ae. albopictus is considered the most invasive mosquito species of the world (Benedict et al. 2007;Bonizzoni et al. 2013). Climate warming and the resulting mild winters favour the establishment, reproduction and spread of Ae. albopictus in temperate climes, such as Central Europe (e.g., Walther et al. 2017;Fălcuţă et al. 2020).
Aedes albopictus is highly vector-competent for numerous arboviruses, including dengue, chikungunya, yellow fever, Zika, West Nile and various encephalitis viruses (Paupy et al. 2009;Martinet et al. 2019). It thus has a major impact on human and veterinary health. The vector competence, i.e. the ability of a haematophagous arthropod to transmit a pathogen, can be influenced by the arthropod's microbiome (Engel and Moran 2013;Jupatanakul et al. 2014) which is defined by Berg et al. (2020) as 'a characteristic microbial community occupying a reasonable well-defined habitat which has distinct physio-chemical properties'.
It has been shown that the microbiome may have a general impact on the development, reproduction and physiology of an invertebrate (Minard et al. 2013;Coon et al. 2014Coon et al. , 2016Coon et al. , 2017. For example, the endosymbiont Wolbachia pipientis is known to be widely distributed in invertebrates (Yang 2000;Hilgenboecker et al. 2008). Bourtzis and O'Neill (1998) and Ahmad et al. (2017) have demonstrated that W. pipientis can affect both the reproduction of insects and the replication and dissemination of pathogenic viruses in an insect vector. These effects are major reasons why the study of the microbiome of mosquitoes has become so popular in recent years. However, the microbiome is not static, but may change during development and can be influenced by many factors such as sex, age and life stage of the host, geographic location, breeding habitat characteristics and food supply (Wang et al. 2011;Zouache et al. 2011;Boissière et al. 2012;Terenius et al. 2012;Jupatanakul et al. 2014;Chen et al. 2020). Water temperature and nutrient content of the breeding habitat, for example, can strongly influence its bacterial community and thus have an impact on the microbiome of developing mosquito larvae (Hörtnagl et al. 2010;Onchuru et al. 2016). In turn, the microbiome ingested from the breeding habitat may considerably influence larval growth and development (Coon et al. 2014(Coon et al. , 2016(Coon et al. , 2017. Insects take up a variety of microorganisms from their environment (Strand 2018). In the case of mosquitoes, this mainly occurs in the larval stage, when individuals are confronted with large numbers of microorganisms in their aquatic habitats during feeding. Larval nutrition can therefore have a major impact on the composition of the microbiome (Wang et al. 2011;Boissiere et al. 2012;Coon et al. 2016). By contrast, occasions to take up microorganisms in the adult stage are limited: both sexes feed on sugary plant juices and only females feed on blood, with the latter occasionally facilitating the uptake of disease agents. There is evidence that the insect host can exert some control over its microbiome via the innate immune response (Douglas 2015;Smith et al. 2015).
In recent years, the microbiome of several mosquito species has been studied, among them Ae. albopictus, Aedes japonicus, Anopheles gambiae and Culex pipiens, with the focus of most studies being on the midgut microbiota of adult mosquitoes (Wang et al. 2011;Gimonneau et al. 2014). It turned out that the microbiome of some species is extremely diverse, and a variety of bacterial phyla such as Actinobacteria, Proteobacteria, Bacteroidetes or Firmicutes could be detected (Moro et al 2013;Zotzmann et al. 2017;Wang et al. 2018).
The extent to which the microbiome of a species differs between populations and individuals is largely unexplored. Furthermore, nothing is known about microorganisms naturally occurring in Ae. albopictus in Germany, the influence they have on their host and whether they pose a threat to humans or may be exploited to their benefit. This study presents the first preliminary insights into the microbial RNA metagenome of Ae. albopictus from Germany which can be considered to represent the mosquito's microbiome. When interpreting the results, however, it is essential to keep in mind that RNA reads similar to a certain microbial species vary considerably in number, and contigs generated from them are of various lengths and have various degrees of probability to be identical to a certain microbial species.

Mosquito origin
The Ae. albopictus specimens investigated in this study had been collected as larvae in the field at seven sites in Germany in 2020 ( Fig. 1): Mengen, Freiburg-Waldsee, Freiburg-Zähringen, Kernen, Munich, Fürth and Jena. These were the locations successfully checked for the presence of Ae. albopictus aquatic stages from all German cities known to possess established populations at the time of the study. Individuals were obtained by sieving potential breeding containers in cemeteries (Mengen, Freiburg-Waldsee, Freiburg-Zähringen, Munich, Jena), in gardens of a settlement (Kernen) and an allotment garden complex (Fürth).
Collected larvae were kept in water from their breeding habitat until adult emergence while being fed ad libitum with ground TabiMin fish food pellets (Tetra, Melle, Germany). Shortly after hatching, adults were killed by freezing at − 20 °C, without being offered blood or a sugar solution. They were morphologically identified on a chilling table using a stereomicroscope according to the determination key by Becker et al. (2010) and stored in 75% ethanol until further processing. Except for the Freiburg-Zähringen site, from where only one female and no male were available, two females and two males per site were analysed.

Nucleic acid extraction and sequencing
For nucleic acid extraction, the mosquitoes were removed from the ethanol and dried for about 1 min at room temperature for the alcohol to evaporate. Subsequently, 13 female and 12 male mosquitoes were pooled by sex and then completely homogenised in 500 µl serum-free ZB5d medium (FLI-intern cell culture medium = Eagle's minimal essential medium with Earle's and Hank's salts plus non-essential amino acids) containing 5 µl of a ready-to-use mixture of penicillin-streptomycin and 1 µl of a ready-to-use mixture of gentamicin-amphotericin (Thermo Fisher Scientific, Dreieich, Germany). Three steel beads with a diameter of 3 mm (TIS GmbH, Gauting, Germany) were added and the samples agitated for 2 min at 30 Hz in a TissueLyser II (Qiagen, Hilden, Germany). Nucleic acid was then extracted from 200 µl of the supernatant using the NucleoMag VET kit (Macherey-Nagel, Düren, Germany) according to the manufacturer's instructions, but without the addition of carrier RNA. The concentration of extracted RNA (12.2 ng/µl for the female sample, 11.3 ng/µl for the male sample) was measured using a NanoDrop Lite (Thermo Fisher Scientific).
Further processing of the sample for total RNA sequencing with Ion Torrent technology, including manual library preparation, was performed following the protocol described by Wylezich et al. (2018). Briefly, the extracted RNA was transcribed into double-stranded cDNA using the cDNA Synthesis System Kit (Roche, Mannheim, Germany), then fragmented by an M220 Focused-ultrasonicator (Covaris Ltd., Brighton, UK) and prepared for Ion Torrent-compatible library generation by means of the GeneRead L Core Kit (Qiagen) and Ion Xpress barcode adapters (Thermo Fisher Scientific). The resulting library was subjected to quality control in a 2100 Bioanalyzer (Agilent Technologies, Santa Clara, USA), using the High Sensitivity DNA Kit (Agilent Technologies) and quantification with a KAPA Library Quantification Kit (Roche). The library was then sequenced on an Ion Torrent S5 XL System (Thermo Fisher Scientific) according to the manufacturer's instructions.

Data analysis
Sequencing results were edited and analysed using Geneious Prime version 2021.0.1 (Biomatters, Auckland, New Zealand). For this, sequences trimmed to a minimum length of 25 bp by the Geneious BBDuk tool were merged by the BBMerge tool, using a merge rate set to 'high'. Both merged data and data that could not be merged were assembled de novo to cluster all closely related sequences into contigs, based on a 'custom sensitivity' setting. The obtained consensus sequences were sorted according to their length, resulting in groups of contigs with similar base pair length. Each contig of these groups was subsequently aligned with GenBank entries (www. ncbi. nlm. nih. gov), with equal results being summarised.

Results
A total of 5,000,504 RNA reads were generated for the female Ae. albopictus pool and 4,653,856 reads for the male pool. In the microbial RNA metagenome of female mosquitoes, RNA fragments with high identities to 42 different microorganismal species from 37 different families were detected, whereas in males RNAs with high identities to a total of 38 different species from 36 families were found (Table 1, Supplementary Tables 1 and 2). In the pool of female mosquitoes, there were a total of 213 contigs (in the range of 69.91-100% percent identity (p.i.), 91-100% query cover and 32-1674 bp), 20 (9.35%) of which matched with eukaryotic, 136 (64.02%) with bacterial and 57 (26.63%) with viral species. In the male mosquito pool, a total of 1380 contigs (82.99-100% p.i., 96-100% query cover, 28-1702 bp) was analysed, 19 (1.87%) of which could be assigned to eukaryotes, 338 (24.29%) to bacteria and 1023 (73.83%) to viruses (Fig. 2). Often, an identification at the species level was not possible and only the genus could be determined.
Contigs with high identities to 13 species and 12 genera of viruses/microorganisms were identified in mosquitoes of both sexes (Fig. 3). Contigs arguing for seven microbial species formerly described in Ae. albopictus were found in both female and male mosquitoes. Four additional microbial species suggested by the contigs had been found in other mosquito species previously. Another two species of bacterial contigs female Ae. albopictus were suggestive of had also been found in other mosquito species, but were not present in our pool of male mosquitoes. Instead, a bacterial species indicated by contigs found in the male pool, but not in the female one of this study, had previously been detected in other mosquito species. Table 1 shows all viruses and microorganisms whose RNA was identified in Ae. albopictus with a p.i. of at least 97% and a query cover of 100%, representing a minimum contig length of 56 bp. RNAs with less p.i. and query cover, shorter lengths or RNAs that could not be assigned to a species, but only to a genus or higher systematic level, are listed in Supplementary Tables 1 and 2. Only the first group of RNAs is discussed in the following.
RNAs with identity to Aedes albopictus anphevirus (89.23-100% p.i., 42-370 bp, 281 contigs) and Aedes Table 1 Species of viruses/microorganisms whose RNA matched that isolated from Ae. albopictus to at least 97% (see Supplementary Tables 1  and 2 (Seifert et al. 1993;Minard et al. 2013), A. baumannii also from lice and fleas (Kempf et al. 2012). The ubiquitous E. coli is a potentially pathogenic (Kaper et al. 2004), widely spread intestinal bacterium of humans and other vertebrates, which had been recognised in adult Anopheles funestus and An. gambiae before (Straif et al. 1989). Elizabethkingia anophelis has emerged as a human pathogen in Africa and Asia (Lau et al. 2016) and had previously been detected in adult An. gambiae (Kämpfer et al. 2011). Wolbachia pipientis is a widely distributed essential bacterial symbiont of mosquitoes, which had frequently been described from Ae. albopictus (e.g. Wiwatanaratanabutr 2013; Park et al. 2016) and other mosquito species (e.g. Kittayapong et al. 2000).
Finally, RNA corresponding to that of Candida sake (100% p.i., 104-189 bp, 3 contigs) was detected in both Ae. albopictus females and males in this study. The fungus belongs to a genus widely distributed in arthropods but has also been extracted from the oral cavity of HIV-positive humans (Hoegl et al. 1998).
In addition to RNAs with high identities to the above bacteria found in only one sex of Ae. albopictus, RNA with a high identity to the fungus Conidiobolus coronatus, which has a human-pathogenic potential (Fischer et al. 2008), could be detected in the female mosquitoes.

Discussion
The mosquitoes in this study were pooled from seven sites within Germany known to be populated by Ae. albopictus. Since the tiger mosquito is controlled in Germany by Bti (Bacillus thuringiensis israelensis) larvicide as soon as local reproduction is detected (Becker et al. 2017(Becker et al. , 2022, the finding of larvae is difficult and was limited in the framework of this study. Due to the pooling of the collected samples, no statement can be made about the geographical origin of the microorganisms or the individual colonisation of Ae. albopictus specimens. Moreover, all viruses and microorganisms referred to in this study were identified exclusively by their RNAs and the alignment of those with sequences in the used databases. Therefore, it is not known whether exactly these species were present or other (unknown) species with closely related RNA sequences, whether they were viable viruses, living symbionts or similar and able to replicate/multiply in Ae. albopictus, and whether they were arbitrarily taken up from the environment and had no additional correlation to the mosquitoes.
As adult mosquitoes emerged from collected larvae were tested here, and no food sources whatsoever had been offered to the investigated adults, the detected RNAs, or the microbes characterised by them, must be supposed to have been transmitted transstadially from mosquito larva to pupa and through metamorphosis to adult. During metamorphosis, the midgut of mosquitoes is transformed and the digestive cells are histolised (Fernandes et al. 2014). The survival of microorganisms or the persistence of RNA, respectively, must therefore be supposed to be possible only intracellularly or with certain mechanisms of adapted symbionts. To clarify this and check for the viability of microorganisms, instead of mere RNA, cultivation attempts are necessary, but were not carried out in this study.
As we studied the microbial RNA metagenome of complete mosquitoes, it cannot be determined in which organ or tissue the found RNAs had been localised. Principally, the composition of the bacterial fauna in different organs of a mosquito can be very variable, and some bacteria colonise several organs in the mosquito at the same time (Gao et al. 2020). The tropism of microbes might give information about their migration paths in the mosquito, or about transmissibility from mosquito parent to offspring or mosquito female to blood host. For the latter, the emergence of the microorganism in the mosquito salivary glands would be requisite (Anderson et al. 2010). Whether this was the case for the potential human pathogens RNAs found were suggestive of, such as A. baumannii, E. coli or A. hydrophila, cannot be determined retrospectively. In addition to the localisation of the microorganisms in the mosquito, the pathogen load which was also not determined in this study might be decisive for a mosquito to become a vector. Thus, the mere presence of a pathogen in the mosquito might not be sufficient for transmission (Beerntsen et al. 2000).
The presence in the microbiome of certain bacteria is beneficial to the mosquito. For example, A. baumannii and A. johnsonii improve blood digestion and nectar assimilation in Ae. albopictus (Minard et al. 2013). However, the influence on mosquito development, reproduction and physiology of most microorganisms found in the microbiome is largely unknown.
The most common phyla ever found in the microbiome of adult Ae. albopictus include Proteobacteria, Bacteroidetes, Firmicutes and Actinobacteria (Mancini et al. 2018). The most common bacterial genera found in Aedes, Anopheles and Culex species are Enterobacter, Escherichia, Klebsiella, Pseudomonas and Serratia as detected in mosquitoes from the USA, England and India (Demaio et al. 1996;Touré et al. 2000;Pidiyar 2002). RNA with identities to all four bacterial phyla as well as to all five genera were found in the Ae. albopictus samples from Germany. In addition, RNA fragments indicating viruses and fungi, such as Riboviria and Ascomycota, were identified. In summary, RNAs with identities to a high number of microorganisms were detected in the German Ae. albopictus some of which represent microorganisms already described from this mosquito species previously, such as W. pipientis, A. baumannii or Usinis virus (Minard et al. 2013;Wiwatanaratanabutr 2013;Batson et al. 2021). Some other viruses or microbial species suggested by the RNA analysis in this study, such as High Island virus, Guapiaçu virus and E. anophelis, have not been detected in Ae. albopictus before, but in other mosquito species and other invertebrates (Kämpfer et al. 2011;Sadeghi et al. 2017;Batson et al. 2021;Oliveira Ribeiro et al. 2021). Furthermore, RNA fragments suggesting microorganisms previously not described from mosquitoes at all, such as L. humi, Z. resiniphila and C. aureum, were found.
It has also been shown that the midgut microbiome of adult mosquitoes may reduce a mosquito's susceptibility to pathogens (Dong et al. 2009;Bahia et al. 2014) and have a general influence on its vector competence (Dodson et al. 2014;Jupatanakul et al. 2014). It can thus be harnessed by manipulating its microorganisms to artificially reduce vector competence. In culture, for example, E. coli was genetically modified in order to express two surface molecules that suppress the development of Plasmodium berghei. Unfortunately, E. coli had difficulties in colonising the mosquitoes and disappeared from their midgut shortly after infection (Riehle et al. 2007). In addition, there are studies that show that an infection of mosquitoes with Wolbachia leads to a strong inhibition of the development of potential pathogens. The infection with Wolbachia of the wMel strain, for example, leads to a highly reduced replication of dengue virus in Ae. aegypti. The reason for this seems to be the Wolbachialinked upregulation of the immune system of the mosquito (Blagrove et al. 2012). Such a way of influencing vector competence is certainly also possible with the help of other organisms from the microbiome.
Since insecticides and physical measures are often inefficient tools for mosquito control (Bourtzis et al. 2016;Pang et al. 2017;Flores and O'Neill 2018), W. pipientis is also exploited for innovative biological control by manipulating mosquito reproduction. Wolbachia infection can lead to the feminisation of genetically male insects, to the killing of male siblings by females or cytoplasmic incompatibility, which ensures that females can successfully mate only with males harbouring the same Wolbachia strain .
Efficient control tools are particularly important in areas where mosquitoes serve as vectors of human disease agents such as Zika virus, yellow fever virus and malaria parasites. Knowledge about the natural occurrence of microorganisms in mosquitoes can therefore contribute to developing and designing new forms of vector control.
It is difficult to explain the differences in the RNA presence between female and male Ae. albopictus in this study. Individuals of both genders were collected from the same breeding sites of the seven locations, so differences between the microbial RNA cannot be attributed to developmental conditions. However, RNA with high identities to 25 species of microorganisms were found in the males but not in the females, and RNA with high identities to 29 species of microorganisms were found in the females but not in the males, with clear differences in the distribution of microorganism to kingdoms (in females, most contigs were assigned to bacteria, whereas in males most contigs were assigned to viruses). For example, RNA fragments arguing for plants were found in the pool of males, but not in the pool of females. That sex has an influence on the microbiome has been shown in previous studies . Also, Rani et al. (2009) detected Chryseobacterium, Pseudomonas and Serratia species only in females of Anopheles stephensi. One possible explanation for the differences might be owing to the limited number of individuals examined, with such differences becoming smaller the more individuals are studied per site. Another explanation might be that female and male larvae in fact have different food preferences and therefore take up different microorganisms with their food, but this is mere speculation and cannot be substantiated. Although it is known that larvae consume a wide range of food (Gimnig et al. 2002;Ye-Ebiyo et al. 2003) and have different feeding preferences depending on species (Merritt et al. 1992), no data exist about different feeding preferences of female and male larvae of the same species.
Especially in cases of a low percent identity and a low query coverage of the found contig sequences with sequences in GenBank, results have to be considered carefully with regard to the occurrence of the respective microorganisms in German Ae. albopictus. This applies, for example, to Kocuria rhizophila. RNA found in female Ae. albopictus matched this soil bacterium with a percent identity of 85.43% only and was registered with only one read. Due to the low percent identity, it is unlikely that the detected RNA belonged to exactly this species. Also, in the case of very short sequence lengths, the linked microorganisms must be viewed critically. This was the case for Wuchereria bancrofti-RNA where the sequence length was 45 bp only. Although percent identity and query coverage were both close to 100%, the probability of the RNA belonging to another organism, possibly a filarial species not described so far, is very high. Hitherto, W. bancrofti has not been reported from Ae. albopictus, and its geographic distribution range is restricted to subtropical and tropical regions (Service 2001).
Possible contamination must also be considered and, in fact, has already been described in similar studies. Genera such as Flavobacterium, Micrococcus, Microbacterium, Chryseobacterium, Neskia and Acidovorax have been found contaminating laboratory reagents like DNA extraction kits (Salter et al. 2014). In this study, Flavobacterium, Micrococcus, Chryseobacterium and Acidovorax could be detected in both sexes of German Ae. albopictus, Neskia only in females.

Conclusion
The microbiome of vector species such as Ae. albopictus holds much potential for the development of efficient control measures and the reduction of vector competence. However, the influence of many microorganisms on the mosquito is still largely unexplored.
Studying the composition of a mosquito's microbiome is difficult, as it is influenced by many factors and can vary considerably both within a species and between species. It is necessary to investigate the influences of the microbiome in more detail and to examine its diversity in a large number of mosquito microbiomes.
To clarify the question of which influence the microorganisms of the microbiome have on the mosquito, it would be helpful to cultivate the microorganisms. However, cultivation is only possible with a small number of species of the microorganisms detected so far. Only with detailed knowledge about the composition and influence of the microbiome on the mosquito can these microorganisms be used for innovative approaches to vector and disease management. The study demonstrates that the microbiome of German Ae. albopictus is comprehensive and might be worth further investigations.
Author contribution JR, DW, MB and HK designed the study; DW and HK collected and identified the mosquitoes; JR and DH carried out the molecular work; JR analysed and interpreted the data; JR and HK wrote the main manuscript text; JR prepared the figure; and all authors reviewed and edited the manuscript. All authors approved the final manuscript version.
Funding Open Access funding enabled and organized by Projekt DEAL.
Data availability Data supporting the conclusions of this article are included within the article and its supplementary tables.

Competing interests
The authors declare no competing interests.
Ethics approval Not applicable.

Consent to participate Not applicable.
Consent for publication Not applicable.

Conflict of interest
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.