Mosquitoes, midges and microbiota: European vector diversity and the spread of pathogens

Tim W.R. Möhlmann

Research output: Thesisinternal PhD, WU


With a constantly growing human population, increased travel and trade around the globe, and global climate change, we face new challenges such as food security and increased risks of infectious diseases. Insect vectors, such as mosquitoes and biting midges, can transmit pathogens. These pathogens cause diseases in humans and animals, thereby negatively affecting health, welfare, and the economy. To tackle these challenges, solutions are needed that include human, veterinary, wildlife, environmental and ecological aspects. Therefore, the aim of this thesis was to answer the question whether environmental and geographical aspects influence vector abundance and community composition, of both the vectors and their bacterial community. With this knowledge about vector communities, we then aimed to elucidate if these field collections of biting midges can be used to predict risks of vector-borne veterinary diseases. Differences among the bacterial communities of field-collected vector species might explain differences in their vector competence. Therefore, we aimed to know whether gut bacteria can influence virus infection and transmission in mosquitoes and biting midges. Since not all vector species can transmit each virus, we finally aimed to know whether endemic and exotic mosquito and biting midge species are competent vectors of the potentially zoonotic Shuni virus.

Studies on species diversity of mosquitoes and biting midges often focus on a specific habitat, region or country, making it difficult to compare and validate results across Europe. To facilitate wider comparisons, this thesis focused on monitoring species diversity of mosquitoes and biting midges in three habitat types located in three countries across Europe (chapters 2 and 4). A total of 27 locations in Sweden, The Netherlands and Italy, comprising farm, peri-urban and wetland habitats were sampled monthly from July 2014 to June 2015, except for the winter months. Indices of species richness, evenness, and diversity were calculated. In addition, vector community compositions were analysed using non-metric multidimensional scaling.

A total of 11,745 female mosquitoes and 50,085 female biting midges were trapped during 887 and 442 collection nights respectively. For both vectors, differences in species communities were more distinct across the three countries than the three habitat types. The highest diversity for mosquitoes and biting midges was found in Sweden. Comparing the habitats, species diversity was the highest at farms for mosquitoes, whereas it was lowest at farms for biting midges. Most individuals were trapped in Italy for both insect groups. A core mosquito and biting midge community could be identified for the three countries, with Culex pipiens and Obsoletus group species as the most abundant mosquito and biting midge species respectively. Biotypes of the Cx. pipiens complex and species of the Obsoletus group show differences in ecological and behavioural characteristics, that are relevant for pathogen transmission. Differences in vector communities across countries imply different patterns of disease emergence and spread throughout Europe. How specific species and their associated communities affect disease risk remains unclear. However, Cx. pipiens and species in the Obsoletus group are expected to have a substantial contribution to the spread of vector-borne pathogens in Europe.

Culex pipiens is the main vector of West Nile virus in Europe. This mosquito species consists of two morphologically identical biotypes, pipiens and molestus, which can form hybrids. Until now, population studies of Cx. pipiens had not differentiated between biotypes and hybrids at the European scale, nor have they used comparative surveillance approaches. I therefore aimed to elucidate the relative abundance of Cx. pipiens biotypes and hybrids in the three habitat types at different countries across Europe (chapter 3). Collected Cx. pipiens mosquitoes were identified to biotype with qPCR. From northern to southern latitudes there was a significant decrease in biotype pipiens and an increase in biotype molestus. Our results emphasize the need to differentiate Cx. pipiens to the biotype level, especially for proper future WNV risk assessments for Europe.

Culicoides species from the Obsoletus group are important vectors of bluetongue and Schmallenberg virus in Europe. This group consists of several species that cannot easily be identified using morphological characteristics. I aimed to elucidate the species composition of the Obsoletus group in three habitat types in different countries across Europe (chapter 5). Biting midges were identified using PCR and gel electrophoresis. Species composition was unique for most country-habitat combinations. Culicoides chiopterus and C. dewulfi were only found in substantial numbers in sample locations from The Netherlands, whereas the majority of the identified biting midges were either C. obsoletus s.s. or C. scoticus. The wide distribution of these two species across all habitat types and countries, in addition to their high abundance at livestock farms, make C. obsoletus s.s. and C. scoticus the most likely candidates for rapid farm-to-farm pathogen transmission throughout Europe. To gain more insight in the potential role of these vectors in the spread of pathogens, field data should be incorporated into mathematical models, to better assess the risk of vector-borne disease outbreaks.

Bluetongue virus (BTV) is transmitted by biting midges and has been circulating in Europe since a major outbreak occurred in 2006 and 2007. The unpredictability of the biting activity of biting midges leads to difficulty in computing accurate transmission models. In chapter 6, this thesis uniquely integrates field collections of biting midges with a multi-scale modelling approach. We inferred the environmental factors that influence the dynamics of biting midge catching, and then directly linked predicted biting midge catches to BTV transmission dynamics. Catch predictions were subsequently linked to the observed BTV prevalence amongst sentinel cattle during the 2007 outbreak in The Netherlands. With this, we were able to directly infer the bias between daily midge catch predictions and the true biting rate per cow per day. The expected biting rate per cow per day at a specific location was around 50% of the total biting midges collected with a trap in one day. Extending the model across Europe, for different seasons and years, indicated that whilst intensity of transmission is expected to vary widely from herd to herd, around 95% of naïve herds in western Europe have been at risk of sustained transmission over the last 15 years. Successful transmission however, is not only dependent on extrinsic factors that influence vectorial capacity (e.g. temperature, vector abundance, and host abundance), but also on intrinsic factors (e.g. host preference, vector physiology, infection, dissemination, and transmission success), that influence vector competence of specific mosquito or biting midge species or individuals. Recently, insect gut bacteria have been hypothesized to influence the interaction of a virus and its vector, thereby adding another factor that can influence vector competence.

Bacteria are part of the insect gut system and influence many physiological traits of their host, such as nutrient availability, development time, longevity, and reproduction. In addition, gut bacteria may even reduce or block the transmission of viruses in several species of arthropod vectors. Only a limited number of studies have investigated the bacterial communities in Culicoides biting midges. Understanding how bacterial communities vary among different species of biting midges, and their related life stages, will help to understand how bacterial communities can be manipulated and ultimately be used as novel tool to reduce pathogen transmission. In chapter 7, I investigated how bacterial communities are influenced by life stage, species identity, and geographic distance. To this end, the bacterial communities in multiple field-collected species from different habitats and countries, and different life stages of two lab-reared biting midges, were identified using Illumina sequencing of 16S rRNA. During the transition from the larval to the pupal stage the bacterial community drastically changed, and only Pseudomonas, Burkholderiaceae and Leucobacter bacteria were found throughout the entire biting midge life cycle. Bacterial communities among field-collected biting midges were unique for almost each species, meaning that bacterial communities of individuals within a species were highly similar, but communities among species were divergent. Besides this species identity effect, also geographic distance influenced the gut bacterial communities of farm-associated biting midges. These differences in bacterial communities among species and geography might contribute to the observed inter- and intra-species variability in vector competence, whereas stably associated bacteria such as Pseudomonas can be potential new candidates for paratransgenic strategies to control vector-borne pathogens.

To show how communities of gut bacteria may contribute to variability in vector competence within and among species, we investigated if gut bacteria can influence the ability of vectors to transmit arboviruses. In chapter 8, I therefore investigated the impact of gut bacteria on the susceptibility of C. nubeculosus and C. sonorensis biting midges for Schmallenberg virus, and of Ae. aegypti mosquitoes for Zika and chikungunya virus. Gut bacterial communities were modified by treating the adult insects with antibiotics. The gut bacterial communities were identified, and mosquito and biting midge susceptibility to arbovirus infection was tested by feeding insects with an infectious blood meal. Antibiotic treatment led to changes in gut bacterial communities, and this significantly increased infection rates of C. nubeculosus with Schmallenberg virus. Infection rates of Ae. aegypti mosquitoes with ZIKV or CHIKV did not change after antibiotic treatment. Asaia bacteria were abundant in untreated, and largely absent in C. nubeculosus biting midges and Ae. aegypti mosquitoes. Antibiotic treatment resulted in relatively more Delftia bacteria in both biting midge species, but not in mosquitoes. I conclude that the effect of gut bacteria on arbovirus infection is specific for each vector, virus, and bacterial species combination.

Arboviruses are notorious for causing unpredictable and large-scale epidemics and epizootics. Shuni virus (SHUV) has zoonotic potential and was recently associated with severe disease in livestock and wildlife. Although most viruses are transmitted by either mosquitoes or biting midges, isolations of SHUV from both vector species from the field suggested that SHUV may be transmitted by both. In chapter 9, I therefore tested whether laboratory-reared biting midge species (C. nubeculosus and C. sonorensis) and mosquito species (Cx. p. pipiens and Ae. aegypti), could transmit the virus. I found that SHUV was able to successfully disseminate in both biting midge species, whereas no evidence of infection or transmission in either mosquito species was found. Our results show that SHUV infects and disseminates in two different Culicoides species, suggesting that these insects could play an important role in the disease transmission cycle.

In the final chapter of this thesis, the contribution of this research to the preparedness for vector-borne infectious diseases is addressed. First, the importance of vector diversity and abundance to the contribution of disease spread is discussed. The discussion continues with the role of core species and the importance of biotypes and species groups. Furthermore, the discussion elaborates on the tripartite interaction of gut bacteria, viruses, and vectors, and how this system could be translated to other virus-insect interactions. In addition, I discuss how factors like vector community composition and their associated bacteria could influence disease risk dynamics.

I propose to use standardized protocols for monitoring of vectors across Europe, to facilitate comparisons of vector communities and to be able to better predict how vector-borne pathogens will spread at a European scale. In addition, I advise that identification of species biotypes and species groups becomes standard in monitoring and surveillance studies, especially for known vectors such as Cx. pipiens and biting midge species of the Obsoletus group. This will contribute to a better understanding of their ecology and to the development of new strategies to effectively control these insect vectors. I suggest that besides the control of vector populations, there may be alternative strategies to control pathogen transmission. First, knowledge about the structure and pathogenicity of the virus could aid in the development of vaccines. Second, new control methods can be developed based on bacteria that are associated with insect vectors. Either through enhancement of bacteria in the gut of vectors that reduce virus infection rates, or by manipulating stably associated bacteria, to use them as paratransgenic control tool against vector-borne pathogens.

I conclude that the results presented in this thesis improve the preparedness for potential disease outbreaks. First, by identifying mosquito and biting midge community compositions at different locations in Europe, that were used for model based predictions. Second, by providing a mathematical model that can predict areas at risk of disease transmission based on environmental factors, vector abundance, and seroprevalence data. Based on the model predictions, I conclude that the emergence of (new) vector-borne diseases is inevitable. Testing endemic and exotic vector species for their ability to transmit emerging pathogens like Shuni virus, increases the preparedness for such potential new disease outbreaks. In addition, we may use closely associated (core) bacteria that occur in many biting midge species, to incorporate into new control tools to reduce virus transmission.

Viruses will continue to cause outbreaks of diseases in animals and humans throughout the world. The integration of knowledge from different research fields will be crucial to prevent large disease outbreaks in the near future. Ultimately, with continued monitoring of vector populations in combination with predictive models, we will be able to predict when and where disease outbreaks are most likely to occur. This, in combination with tools to effectively control vector populations, and methods to reduce virus infection in hosts and vectors, will facilitate effective containment and control of future disease outbreaks.

Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • Wageningen University
  • Dicke, Marcel, Promotor
  • Koenraadt, Sander, Co-promotor
  • van Overbeek, Leo, Co-promotor
Award date22 Mar 2019
Place of PublicationWageningen
Print ISBNs9789463433952
Publication statusPublished - 22 Mar 2019


Dive into the research topics of 'Mosquitoes, midges and microbiota: European vector diversity and the spread of pathogens'. Together they form a unique fingerprint.

Cite this