Challenges limiting the appraisal of tsetse microbiome include the difficulties to identify bacterial species in these flies. To fill these gaps in understudied ecological settings, PCR-based method was used to identify S. glossinidius and Wolbachia in wild population of G. m. submorsitans caught in the area of lake Iro in the south of Chad. The identification of S. glossinidius and Wolbachia in wild population of G. m. submorsitans of Lake Iro is in line with previous studies reporting these two symbiotic micro-organisms in wild populations of G. m. morsitans, G. tachinoides, G. p. palpalis, G. pallidipes, G. f. quanzensis and G. brevipalpis [2010, 30, 31, 25, 26, 22, 23].
The S. glossinidius infetion rate of 9.0% obtained in the present study is similar to 9.3% reported in Liberia for G. p. palpalis . This rate in higher than 1.4% reported in Zambia for wild populations of G. pallidipes , but lower than 15.65%, 17.5%, 54.9% and 93.7% previously reported respectively in the Democratic Republic of Congo for G. f. quanzensis, in Zambia for G. m. morsitans, in Cameroon for G. p. palpalis and in Zambia for G. brevipalpis [7, 31, 23]. These results confirm the high heterogeneity of S. glossinidius infection rates according to tsetse species [31, 7]. Nevertheless, reliable comparisons between data from by different studies require to understand the study designs. In the present study, S. glossinidius was identified in tsetse body while in other studies, whole tsetse or parts of the insect such as the abdomen, the thorax and the legs were used. The heterogeneity observed in the S. glossinidius infection rates could be explained by some variations in methodological approaches, the intrinsic characters of each tsetse species and environmental factors (vegetation, humidity, temperature) encountered in different ecological settings. In natural conditions where environmental factors vary and have impacts on the biology of tsetse flies, the relationship between tsetse and its symbiotic micro-organisms is affected. As the survival and the transmission of these symbiotic micro-organisms are linked to tsetse biology because of their limited metabolic capacities, each environmental factor affecting the biology of tsetse could change its interactions with symbiotic micro-organisms. In such a scenario, S. glossinidius could not undergo horizontal transmission with the same efficiency and in consequence, its infection rates could vary with environmental factors. When environmental variations are removed like in experimental studies or in insectarium , the symbiotic association between tsetse and its symbionts is not affected. As already reported in G. p. gambiense and G. m. morsitans, high vertical transmission and high infection rates of symbiotic micro-organisms are observed in tsetse flies .
The significant association (r = 4.992; P = 0.025; 95% CI= [0.178–5.012]) observed between S. glossinidius and trypanosome infections indicates that the presence of S. glossinidius seems to favor trypanosome infections in G. m. submorsitans of the area of lake Iro in the south of Chad. Although these results are not in agreement with those reporting no significant association between the presence of S. glossinidius and trypanosome infections in G. austeni , G. brevipalpis, G. m. morsitans and G. pallidipes , our findings are in line with those reported in G. p. palpalis of some sleeping sickness foci of Cameroon  and other tsetse species [20, 8, 11]. The discrepancies observed in the tripartite association between tsetse, S. glossinidius and trypanosomes may result from differences in the biology of different tsetse species as well as the bioclimatic conditions impacting the relationship between tsetse and its symbiotic micro-organisms. Moreover, our results showing significant association (r = 3.147; P = 0.043; 95% CI= [0.178–5.012]) between S. glossinidius and T. simiae, but no association for other trypanosome species identified in this study suggest that the tripartite association between tsetse, S. glossinidius and trypanosomes could vary according to trypanosome species. Better understanding these tripartite associations requires more in-depth investigations on wild populations of different tsetse species of various tsetse infested areas.
The identification of Wolbachia in wild populations of G. m. submorsitans may have some implications in the development of new vector control strategies. On the basis of its capacity of inducing cytoplasmic incompatibility and to be transmitted from mother to offspring, Wolbachia can be genetically modified with the objective of producing biomolecules able to interfer with the establishment and/or the development of trypanosomes in tsetse flies. If that occurs, the vectorial competence of tsetse will be affected and disease transmission could be blocked through genetically modified Wolbachia strains that conferred resistance to tsetse fly .
The overall Wolbachia infection rate of 14.5% obtained in the present study is lower than 25.32%, 44.3%, 88.8%, 98% and 100% reported respectively in G. p. palpalis , G. f. fuscipes , G. f. Quanzensis, G. austeni [33, 20] and G. m. morsitans . These results show a certain heterogeneity in the Wolbachia infection rates according to tsetse species. As already reported by Kante et al. , this heterogeneity could be related to specific biological characteristics of each tsetse subspecies. For identical stimulus, it has been reported that interactions between tsetse fly and its symbiotic micro-organisms vary according to specific biological response of each tsetse species or subspecies . Such variations affect not only the interactions between tsetse and its symbiotic micro-organism, but also the Wolbachia infection rates. Some discrepancies observed in the Wolbachia infection rates could be explained by some differences in the study design as well as the analytical methods. In the present study, Wolbachia was searched in tsetse body (whole tsetse without legs, wings and proboscis) while in other studies, investigations were undertaken on isolated tissues or whole tsetse fly. In addition, the fact that one molecular marker was used to detect Wolbachia infections has probabily underestimated its infections rates. Indeed, in tsetse flies from the same ecological setting, Kante et al.  reported significant differences in the Wolbachia infection rates when different molecular markers were used. In Camerooun for instance, the detection of wsp gene was two-fold more sensitive in tsetse from Campo while 16S rDNA showed higher sensitivity in flies from Fontem . In addition to differences in the sensitivity of molecular markers, the technical approaches could also have impacts on the Wolbachia infection rates. If the density of Wolbachia in some G. m. submorsitans is below the detection threshold of standard PCR-based method, some infections could pass undetected. Wamwiri et al.  highlight a high density of Wolbachia in G. austeni populations from Kenya and a low density in the same tsetse species of South Africa. In addition, a low density of Wolbachia has been reported in Rhagoletis cerasi  and Drosophila paulistorum . While searching for sensitive and reliable markers or tools for Wolbachia identification remains a goal to achieve, the use of one marker or standard PCR-based method may lead to an underestimation of the real Wolbachia infection rates.
The 9.8% of G. m. submorsitans harboring co-infections of Wolbachia and trypanosomes is low compared to 29.84% and 26% reported respectively in G. p. palpalis  and G. tachinoides in Cameroun . Although the technical approach and the study design could partially explain this low co-infection rate, such co-infections are probably not common in G. m. submorsitans of lake Iro. The absence of significant association (r = 1.754; P = 0.185; 95% CI = [0.360–1.219]) between Wolbachia and trypanosome infections suggests that the presence of this bacterium does not seem to be an obstacle for the establishment of trypanosomes. These results are in agreement with those of Kante et al.  reporting no association in G. p. palpalis from in sleeping sickness foci of Cameroon. They contrast data of Alam et al.  showing a negative correlation between Wolbachia and trypanosome infections and suggesting that the presence of this bacterium prevent trypanosome infections in G. f. fuscipes. The tripartite association between tsetse, Wolbachia and trypanosomes seems to vary according to tsetse species or subspecies.
Results of the present study showing that only 2.31% of tsetse flies were co-infected by Wolbachia and S. glossinidius are in agreement with the 5.43% previously reported in G. f. quanzensis . They indicate that co-infections between Wolbachia and S. glossinidius are rare in wild populations of G. m. submorsitans. The co-infection rate between S. glossinidius and Wolbachia is probably underestimated in the present study because the molecular markers used have been reported to be of low sensitiviety, especially when only one marker was used to detect symbiotic microorganisms. The low co-infection rate revealed between S. glossinidius and Wolbachia can be also explained by the biological effects of each of these bacteria. Indeed, association studies revealed that the presence of S. glossinidius seems to favor trypanosome infections while no association was reported between Wlobachia and trypanosome infections. In other studies, the negative correlation reported between trypanosomes and Wolbachia infections suggested that the presence of Wolbachia seems to prevent trypanosome infections [27, 30]. These observations suggest that some antagonistic actions, resulting from different biological actions of Wolbachia and S. glossinidius could occur in tsetse fly during trypanosome infections.
Our investigations on tripartite associations were based on presence/absence of trypanosome or S. glossinidius or Wolbachia. Instead of focusing on this presence/absence, the genetic characterization of S. glossinidius or Wolbachia strains could provide additional values to decript these associations. In previous investigations, it has been reported that the tripartite association could be affected by specific genotypes of S. glossinidius and some trypanosome species such as T. b. gambiense and T. b. brucei . For some trypanosome species, specific S. glossinidius genotypes have been reported to affect the vectorial competence of G. p. gambiensis and G. m. morsitans . Genetic characterization of bacteria populations could provide additionnal data to improve knowledge on this tripartite association, and also to better understand the real contribution of symbiotic microorganisms (S. glossinidius or Wolbachia) in the vectorial competence of tsetse flies. To obtain the real overview of the vector competence of tsetse flies, it is also important to take into consideration other factors such as the level of lectin in the tsetse gut at the time of parasite uptake, the fly species, the age, the teneral status of tsetse and its first blood meal on a non-infected host because these factors affect its ability to be infected and could mitigate the influence of symbiotic micro-organisms. Such factors could play a significant role in the success or failure of parasite establishment because the processes leading to this establishment involve complex interactions between these factors .