Many pathogen populations consist of genetically diverse strains. Genetic diversity of pathogens can have important implications for public health such as vaccine development. The study of this genetic diversity can help with the development of control strategies. Jonas Durand (PhD) found 21 different strains of B. afzelii and B. garinii in ticks on a 1 ha plot near Neuchâtel [1]. Using data from a long-term study (11 years), we have shown that some strains are 10 times more common than others [1, 2] and we want to understand why. Working together with the research group of Lise Gern, we have recently used experimental infections in laboratory mice to show that there is substantial variation in transmission efficiency between strains of B. afzelii [3]. Interestingly, the strains that had higher fitness in the lab were also those strains that were most common in our wild population of I. ricinus ticks [2, 3]. The strain-specific spirochete load in the tick is another important phenotype (see below).

Arthropod vectors often carry multiple strains of a vector-borne pathogen. These strains have the opportunity to interact inside the arthropod vector. The outcome of these interactions may influence what strains are transmitted from the vector to the population of vertebrate hosts. Jonas Durand (PhD) showed that multiple-strain infections are common in wild I. ricinus ticks [1, 4]. In its natural life cycle, Borrelia spirochetes can spend 8–10 months in the tick in between blood meals. Thus the different strains in a tick have a lot of time to interact with each other. Dolores Genné (PhD candidate) is testing whether Borrelia strains compete inside the tick vector and whether this competition influences the transmission success to the vertebrate host.

Immune responses that protect the host from infection can be highly specific for one particular strain. In vertebrate hosts, strain-specific antibody responses often target surface proteins on the pathogen that are highly variable between strains. Identification of immunodominant antigens that mediate strain-specific protective immunity is therefore important for understanding the epidemiology of multiple-strain pathogens and vaccine development. In LB pathogens, outer surface protein C (OspC) is necessary for establishing infection in the vertebrate host. OspC induces a strong antibody response that induces strain-specific protective immunity. Maxime Jacquet (PhD) conducted a vaccination trial with recombinant OspC proteins to confirm this phenomenon in B. afzelii [5]. Local population of B. afzelii and B. garinii each contain 10–11 different ospC alleles [1]. We want to understand what factors maintain this ospC polymorphism. We believe that the ospC gene facilitates super-infection of previously infected reservoir host and would like to test this hypothesis in the future.

Genetic differences between host individuals can explain variation in resistance/susceptibility to infectious diseases. There is much interest in identifying the genes that are responsible for variation in host resistance. The toll-like receptors (TLR) are a family of innate immune receptors in vertebrates that were recently discovered. Work by Tschirren and colleagues found that genetic variation in toll-like receptor 2 (TLR2) in a wild population of bank voles was associated with variation in B. afzelii infection [6, 7]. Andrea Gomez (PhD candidate) established a breeding population of bank voles and is currently testing whether the TLR2 polymorphism causes variation in B. afzelii infection.

The host specificity of a parasite is arguably its most important feature. Generalist parasites can jump easily between host species whereas specialist parasites are restricted in their host range. The two most common LB pathogens in Europe, B. afzelii and B. garinii are specialized on rodent and bird reservoir hosts, respectively. This host specificity appears to be mediated by the vertebrate complement system [8, 9]. Experimental infections with B. garinii are rare because there is no reliable avian model. Cindy Bregnard (PhD candidate) is currently trying to develop an in vivo model for testing the complement-based host specificity hypothesis of B. afzelii and B. garinii.

The risk of Lyme disease depends on the density of infected nymphs, which is the second blood-seeking stage in the tick life cycle. The spirochete population (or load) in the nymph can range from 1000 to 10000 bacteria [5]. Maxime Jacquet (PhD) recently showed that the population of B. afzelii spirochetes in the tick can decline by 90% over a period of 6 months [5]. Dolores Genné (PhD candidate) is testing how temperature influences this decline in spirochete load over time. Jonas Durand (PhD) showed that strains with the highest spirochete load in wild ticks are the most common in nature [4]. The tick spirochete load appears to be a critical life history trait for the Borrelia pathogen.

Vector-borne pathogens can manipulate their vertebrate host or arthropod vector to increase their transmission success. Vector-borne pathogens could benefit from manipulating the host choice behaviour of their arthropod vectors to direct the latter to choose competent rather than incompetent hosts. Jérémy Berret (MSc) tested whether Borrelia influenced the attraction of ticks to host odours but found no evidence for adaptive manipulation [10]. Océane Courbat (MSc candidate) is continuing this work with experimentally infected ticks. Research in the USA found that B. burgdorferi can suppress the immunity of laboratory mice [11]. Inès Ben Messaoud (MSc candidate) is testing this phenomenon in B. afzelii. Yating Li (MSc candidate) is currently testing whether B. afzelii can manipulate the antibody response in bank voles to benefit the tick vector.

LB is a serious disease in humans but what is the effect of Borrelia pathogens on the fitness of the reservoir host and tick vector? An experimental infection study found no effect of infection with B. burgdorferi on the running behaviour of the white-footed mouse, Peromyscus leucopus [12]. I analysed a long-term mark-recapture data set but found no effect of Borrelia infection on the survival of P. leucopus [13]. I am very interested to test whether Borrelia pathogens influence the reproductive success of their rodent reservoir hosts. Anouk Sarr (technician) is testing whether B. afzelii influences the fitness of its tick vector.

Co-feeding transmission of vector-borne pathogens occurs when the vector is transferred directly between arthropod vectors that feed in close proximity to each other on the same host. This mode of transmission has been demonstrated in a number of vector-borne pathogens such as West Nile virus, tick-borne encephalitis virus, and LB pathogens [14]. Our research group has made a number of contributions to co-feeding transmission of LB pathogens [3, 14, 15]. We have shown that co-feeding transmission can vary among strains of B. afzelii [3, 5] and that this phenotype is associated with strains that have high fitness and that are common in nature [2, 3]. An important remaining question is whether nymphs infected via co-feeding transmission as larval ticks are actually infectious to vertebrate reservoir hosts [14]. Alessandro Belli (MSc candidate) has recently shown that such nymphs are infectious for rodent reservoir hosts.