The Mediterranean fruit fly and its bacteria – potential for improving sterile insect technique operations
Boaz Yuval (corresponding author), The R.H. Smith Faculty of Agriculture, Food and Environment, Hebrew University of Jerusalem, POB 12, Rehovot 76100, Israel. E-mail: firstname.lastname@example.org
Mediterranean fruit flies (Ceratitis capitata Wiedemann, Diptera: Tephritidae), harbour a diverse community of bacteria in their digestive system. Molecular and culture-based techniques show that members of the Enterobacteriaceae form the dominant populations in the gut of the Mediterranean fruit fly. Among them, many are diazotrophs and actively fix nitrogen in vivo. Most prominent are Klebsiella spp., Enterobacter spp., Pectobacterium spp. Citrobacter freundii and Providencia stuartii. A marked shift in community composition was observed between different developmental stages: in larvae, the pectinolytic Pectobacterium were most abundant, suggesting that pectinolysis plays a role early during the fly’s life. Additionally, pseudomonads, some of which are known entomopathogens, constitute a minor, yet common and stable community in the C. capitata gut. Microbial communities in VIENNA 8 flies, a mass rearing genetic sexing strain, differ from wild flies, and irradiation further affects the microbial community. We found increased levels of the pathogenic species Pseudomonas in the industrially used strain. Furthermore, although members of the Enterobacteriaceae family remain the dominant bacteria group present in the fly’s gut, the levels of Klebsiella species decrease significantly in the days after irradiation. Eliminating the bacterial population in normal flies by using antibiotics affects measurable physiological and behavioural parameters related to fitness. Finally, we tested the hypothesis that inoculation of sterile flies with members of the original bacterial community results in enhanced competitiveness. We found that addition of the bacteria Klebsiella oxytoca to the post-irradiation diet significantly improves sterile male performance.
The interaction between insects and micro-organisms is multifaceted, ranging from pathogenicity to complex mutualistic relations. The increasing number of studies addressing the bacteria-insect interface is gradually revealing the intricacy of these interactions (Bourtzis and Miller 2003, 2009).
Until recently, the association between the tephritid fruit flies and bacteria received relatively little scrutiny (see Drew and Lloyd 1987, 1991; Lauzon 2003; Behar et al. 2009; for reviews of tephritid-bacteria interactions). Classical microbiological methods revealed that Mediterranean fruit flies (Ceratitis capitata, Diptera: Tephritidae), harbour a community of diazotrophic bacteria in their digestive system. Marchini et al. (2002), using culture-dependent methods, studied the bacteria associated with the oesophageal bulb (a foregut diverticulum) of C. capitata. They found that this organ harbours a discrete community of bacteria, comprised mainly, as in other fruit flies, of members of the family Enterobacteriaceae. Specifically, Klebsiella oxytoca and Enterobacter agglomerans were identified as the most representative species in the oesophageal bulb of wild and laboratory medfly populations respectively (Marchini et al. 2002). Similarly, a systematic study of the structure and diversity of microbial communities in eggs, larvae, host fruit, pupae and adult Mediterranean fruit flies, based on 16S rDNA sequences obtained from PCR-Denaturating Gradient Gel Electrophoresis (DGGE) and from isolated colonies revealed that members of the Enterobacteriaceae constitute the dominant populations in the gut of C. capitata. Most prominent were species of Klebsiella, which were found in different combinations with Citrobacter freundii, Enterobacter spp, Pantoea spp., Pectobacterium spp. and Providencia stuartii (Behar et al. 2005, 2008a).
These analyses also found that, as in other fruit flies, Klebsiella oxytoca is the most common species in the gut of C. capitata. Furthermore, a species of Pectobacterium was found to be commonly associated with the Mediterranean fruit fly, particularly in the larval stages. PCR-DGGE analyses also supplied evidence that the enterobacterial community, composed mainly of K. oxytoca, Pectobacterium cypripedii, Pantoea spp. and C. freundii, is present during all of the fly’s developmental stages and some elements are vertically transmitted from the female parent to its offspring during oviposition (Behar et al. 2008a). Strikingly, all these bacterial species are potential diazotrophs. In addition, K. oxytoca, Pantoea spp. and P. cypripedii are also known pectinolytic bacteria. Further to these ubiquitous bacteria, we also found small but persistent populations of potentially pathogenic bacteria, Pseudomonas spp., in wild and laboratory-reared flies (Behar et al. 2008b).
These findings prompt the obvious question – are the bacteria that form the major microbial populations contributing to the fitness of the fly host? Below, we briefly review a number of recent studies aimed at answering this question. In the first study, we cleared bacteria from the guts of flies and monitored female oviposition and male copulatory success. In a second series of studies, we investigated how inoculating flies with bacteria affects their longevity and, in the case of sterile males, copulatory success.
Effect of Gut Bacteria on Nutritional Status and Reproductive Success
We recently studied the contribution of gut bacteria to fly fitness (Ben-Yosef et al. 2008a). After clearing bacteria from the guts of adults, we were able to test the hypothesis that bacteria contribute to reproductive success of Mediterranean fruit fly adults by enhancing copulatory success in males and egg production in females. To do so, the reproductive success of antibiotic-treated flies – maintained on either nutrient rich or nutrient poor diets – was compared to that of flies containing a full complement of enteric bacteria. Following eclosion, flies were fed a full diet containing peptides, sugar and minerals, or a sugar diet, lacking peptides. Sub-groups from each diet were fed a mixture of the antibiotics ciprofloxacin and piperacillin. We quantified the presence of bacteria, food consumption, weight gain, lipid and protein levels, oviposition in females and copulatory success of males in the four treatment groups.
The antibiotic treatment effectively cleared the gut of bacteria. Relative amounts of food consumed – with or without antibiotics – were similar in all treatment groups. The antibiotics did not inhibit feeding, and their ingestion did not affect dry weight or the amount of protein stored, yet females feeding on the full diet without antibiotics had higher lipid levels than females on antibiotics. Females fed the full diet produced significantly more eggs than females on the sugar diet, but the presence of bacteria did not affect numbers of eggs produced. However, in the absence of bacteria, oviposition rate of nutritionally stressed females was significantly accelerated.
The presence of bacteria conferred a marginally significant mating advantage to both sugar-fed (P = 0.09) and protein-fed males (P = 0.08), when competing with antibiotic-treated males that were fed on the same diet. In sugar-fed males, the addition of antibiotics did not affect the latency to copulate (P = 0.907). Conversely, in the males fed a full diet, the presence of bacteria was significantly associated with a shorter latency to mate (P = 0.033; see Ben-Yosef et al. 2008a for full details).
The diet-bacteria interaction had another manifestation, as it significantly affects longevity. Treating males and females with antibiotics affected their longevity in a diet-dependent fashion: prolonging life when flies were fed with sugar, yet having no effect when combined with the full diet (Ben-Yosef et al. 2008b). These results provide several important clues about the effect of antibiotics on the flies and on the nature of the interaction between the fly and its gut microbiota. The diet-dependent prolongation of longevity elicited by suppressing the gut microbiota indicates that the nature of fly-bacteria interaction was dictated by the diet. The fact that bacteria were associated with a decreased longevity only in sugar-fed flies suggests that antagonistic bacterial activity was induced in response to unfavourable environmental conditions present only in the gut of these flies. Alternatively, qualitative or quantitative changes in the species composition of the gut bacterial community, induced by the diet, could have been the cause of a different net effect on the longevity of the host fly. Thus, the gut microbiota may be alternating between mutualism/commensalism and parasitism in response to the composition of their host’s diet.
Effects of Bacterial Inoculation on Fly Longevity
This experiment assessed the effect of the dominant Enterobacteriaceae community and the minor Pseudomonas community on C. capitata longevity (Behar et al. 2008b). Ingestion of bacteria by adult flies significantly affected mortality rate and average fly longevity in both sexes, regardless of previous exposure to antibiotics. In general, feeding on Pseudomonas had a negative effect on longevity, while feeding on the Enterobacteriaceae mixture had a beneficial effect. Compared to flies fed on sugar alone, flies fed on Pseudomonas died at a significantly faster rate and their average longevity was significantly reduced. Conversely, flies fed on Enterobacteriaceae died more gradually and lived significantly longer, on average, than flies fed on sugar. Thus we concluded that the dominant presence of the enterobacterial community in the gut of C. capitata contributes to its longevity. Previous work (reviewed above) has shown that this community is capable of contributing to the fly’s nitrogen and carbon metabolism, development and copulatory success. In addition, we suggest that the Enterobacteriaceae community within the gut of C. capitata also has an indirect contribution to its host fitness by acting (through an as yet unknown mechanism) as a barrier against deleterious bacteria (Behar et al. 2008b).
Bacteria in Sterile Males and their Manipulation to Improve Sexual Competitiveness
After characterizing the bacterial community in wild and laboratory-reared strains, we attempted to do the same for flies used for application of the Sterile Insect Technique (SIT). Accordingly the community inhabiting the gut of flies from the genetic sexing strain VIENNA 8, reared in the Bio-Fly Ltd. factory in Sde Eliyahu, Israel was characterized using PCR-DGGE and Amplified rDNA restriction analysis (ARDRA) (see Ben-Ami et al. 2010 for full details of methods). We found that an irradiation dose of 100 Gy significantly affects the structure of the microbial community. Phylogenetic analyses based on DGGE results revealed that the Enterobacteriaceae family is the dominant bacterial group in the gut, both before and after the irradiation. Gut samples obtained from irradiated males on the day of emergence contained the genera Salmonella, Citrobacter, Providencia, Morganella and Enterobacter, but no Klebsiella. The gut of non-irradiated flies at emergence and of 5-day-old irradiated flies also contained species of the genera Klebsiella and Pectobacterium. In addition, these analyses revealed that all the VIENNA 8 flies, both before and after irradiation, harbour a stable and dominant community of Pseudomonas spp. However, the frequency of the pseudomonads and of Salmonella populations decreased in flies 5 days after irradiation.
The absence of Klebsiella in irradiated flies assayed on the day of irradiation was striking. A complementary analysis, based on ARDRA, strengthened the observation that irradiation specifically affects the abundance of Klebsiella spp. Thus, while Klebsiella spp. are part of a dominant community among the total gut microbiota of the non-irradiated, 5-day-old irradiated flies and of wild flies (18.67%, 23.0% and 31.0% respectively), their prevalence in the gut of the irradiated flies on eclosion day is significantly lower (4.0%, P < 0.05; Ben-Ami et al. 2010). ARDRA also confirmed that the guts of the irradiated VIENNA 8 flies harbour dominant populations of Pseudomonas spp. at the time of emergence and release 5 days later (ibid.), yet failed to detect Salmonella.
In light of the absence of Klebsiella spp. in the irradiated gut and their abundance in wild flies, we attempted to improve the performance of sterile males by inoculating their adult diet with a suspension of Klebsiella oxytoca, one of the most commonly found bacterial species in the gut of Mediterranean fruit fly. Such an approach has been attempted previously, without confirming the establishment of the inoculate in the experimental flies (Niyazi et al. 2004). We used an antibiotic resistant strain that was recoverable following inoculation, and found that, significantly, sterile males inoculated in this manner copulated faster (i.e. exhibited a shorter latency to mate), than sugar-fed sterile males or sterile males fed a suspension of autoclaved Klebsiella oxytoca (Ben-Ami et al. 2010).
Taken together these results encourage us to continue to investigate the precise manner of the contribution of bacteria to the fitness of the Mediterranean fruit fly and other tephritids. In addition, we suggest that supplementing the post-teneral diet of sterile males with beneficial bacteria can significantly improve the effectiveness of the SIT. Following on our ‘proof of principle’, future work should identify the optimal combination of bacteria which, alone or in conjunction with other post-teneral treatments, can provide an optimal boost to the performance of sterile male fruit flies.
We thank Pablo Liedo and three anonymous reviewers for comments on the manuscript. Our research is supported by grants from BARD and IAEA.