Molecular detection of Sodalis glossinidius, Spiroplasma and Wolbachia endosymbionts in wild population of tsetse flies collected in Cameroon, Chad and Nigeria

Background Tsetse flies are cyclical vectors of African trypanosomiasis. They have established symbiotic associations with different bacteria, which influence certain aspects of their physiology. The vector competence of tsetse flies for different trypanosome species is highly variable and is suggested to be affected by various factors, amongst which are bacterial endosymbionts. Symbiotic interactions may provide an avenue for the disease control. The current study provided the prevalence of 3 tsetse symbionts in Glossina species from Cameroon, Chad and Nigeria. Results Tsetse flies were collected from five different locations and dissected. DNA was extracted and polymerase chain reaction PCR was used to detect the presence of Sodalis glossinidius, Spiroplasma sp and Wolbachia using specific primers. A total of 848 tsetse samples were analysed: Glossina morsitans submorsitans (47.52%), Glossina palpalis palpalis (37.26%), Glossina fuscipes fuscipes (9.08%) and Glossina tachinoides (6.13%). Only 95 (11.20%) were infected with at least one of the 3 symbionts. Among the infected, 6 (6.31%) were carrying mixed infection (Wolbachia and Spiroplasma). The overall symbiont prevalence was 0.88%, 3.66% and 11.00% respectively, for Sodalis, Spiroplasma and Wolbachia. Prevalence varied between countries and tsetse species. No Spiroplasma was detected in samples from Cameroon and no Sodalis was found in samples from Nigeria. Conclusion The present study revealed for the first time, the presence of infection by Spiroplasma in tsetse in Chad and Nigeria. These findings provide useful information to the repertoire of bacterial flora of tsetse flies and incite to more investigations to understand their implication in the vector competence of tsetse flies.


Conclusion
The present study revealed for the rst time, the presence of infection by Spiroplasma in tsetse in Chad and Nigeria. These ndings provide useful information to the repertoire of bacterial ora of tsetse ies and incite to more investigations to understand their implication in the vector competence of tsetse ies.

Background
Trypanosomiasis is one of the major endemic diseases in sub-Saharan Africa. It is caused by trypanosomes, an extracellular agellated protozoan parasite of the genus Trypanosoma. The disease exists in two forms: The human African trypanosomiasis or sleeping sickness and the animal African trypanosomiasis. In humans, the disease is caused by two subspecies of Trypanosoma brucei: Trypanosoma brucei rhodesiense responsible for the acute form of the disease in eastern and southern Africa, while Trypanosoma brucei gambiense is responsible for the chronic form of the disease in western and central Africa [1]. Approximately 56 million people are estimated to be at different levels of risk of contracting HAT and more than a surface of 1.18 million Km 2 are still at risk of T.b. gambiense infection [2]. In 2019 there were 992 cases recorded in Africa [1]. The animal form of the disease, animal African trypanosomiasis (AAT) is caused by several species and subspecies of trypanosomes including Trypanosoma congolense, Trypanosoma vivax, Trypanosoma simiae, Trypanosoma uniforme, Trypanosoma godfreyi, Trypanosoma brucei brucei and Trypanosoma grayi. They cause pathogenic infections in cattle, sheep, goats, pigs, dogs, camels and horses [3][4][5]. The disease is one of the major constraints to agricultural development on the continent.
Trypanosomes are cyclically transmitted between different vertebrate hosts by tsetse ies (Diptera: Glossinidae). Thirty-one species and subspecies of tsetse ies have been described. They can be grouped into three groups or subgenera based on common characteristics and morphology due to bio-ecological and genetic similarities: the riverine Palpalis, the savannah Morsitans and the forest Fusca [6,7]. They acquire trypanosomes when feeding on an infected mammalian host. The trypanosomes undergo a series of transformations and multiplication in their gut and give rise to infective forms which will be inoculated into a new host during the feeding [8].
Tsetse gut harbours a diversity of bacteria acquired from the environment or maternally transmitted [9].
Previous studies have shown that these bacterial populations vary considerably depending both on the tsetse species or sub-species and the geographic origin of the ies [10]. This microbial community in uences several aspects of tsetse's physiology, including nutrition, fecundity development and maturation of the innate immune system and vector competence [11,12].
Tsetse ies have established long-term associations with four vertically transmitted endosymbiotic bacteria including Wigglesworthia glossinidia, Sodalis glossinidius, Wolbachia sp, and Spiroplasma sp that was recently established as the fourth tsetse symbiont in G. fuscipes fuscipes, G. tachinoides, and G. palpalis palpalis [13]. They show different types of relation with their host.
All tsetse ies house Wiggleworthia glossinidia, the primary and obligate endosymbiont. It resides intracellularly within the bacteriome in the anterior midgut and also found extracellularly in milk gland of the y [11]. Wiggleworthia glossinidia provides dietary supplements absent from the y's vertebrate blood-restricted diet, supports larval development and contributes to maturation of the adult immune system [12,14].
Wolbachia endosymbionts are obligatory intracellular bacteria belonging to the Order Rickettsiales. Wolbachia infects a broad range of arthropod and larial nematode species and is probably the most prevalent endosymbiont found in insect germlines [11]. Within the genus Wolbachia, 17 supergroups (A to Q) are currently recognized based on sequences of the ve conserved genes fbpA, coxA, ftsZ, gatB, coxA, and hcpA and the amino acid sequences of the four hypervariable regions of the Wsp protein [11,15]. The majority of insect infections fall into supergroups A and B [16].
In tsetse, Wolbachia resides mainly in reproductive tissues and is maternally transmitted from generation to generation by trans-ovarian transmission. They are also transferred horizontally among arthropods. Wolbachia infection in the tsetse host results in a variety of reproductive abnormalities such as parthenogenesis, male killing, feminization and cytoplasmic incompatibility [16][17][18].
Cytoplasmic incompatibility results in embryonic mortality in the progeny derived from matings between insects with different Wolbachia infection status: when an infected male mates with an uninfected female or a female infected with a different strain of the bacterium [16]. The presence of this symbiont in the tsetse y Glossina morsitans has been associated with the induction of cytoplasmic incompatibility [17]. This effect confers indirect reproductive advantages to the infected females and is considered as a potential vector control alternative.
The tsetse's secondary and facultative symbiont is the commensal Sodalis glossinidius. It is a gramnegative organism belonging to the Enterobacteriaceae family. It is widely spread in numerous tissues of the y (midgut, fat body, milk gland, salivary glands and reproductive system) and can be found both intracellularly and extracellularly [23]. The Sodalis genome consists of one circular chromosome of 4.17 Mbp, three extrachromosomal plasmids designated pSG1, pSG2, and pSG3, as well as a phage, ФSG1.
However, its genome sequence shows reduced coding capacity with a large number of pseudogenes [24]. Sodalis glossinidius can be transmitted maternally via haemolymph, milk gland secretions, and horizontally during mating [11,25]. In tsetse ies, the speci c role of this symbiont is still not clear. However, it has been shown to affect host longevity and has been suspected to play a role in potentiating susceptibility to trypanosome infection in tsetse by in uencing the e cacy of the tsetse immune system possibly through lectin-inhibitory activity [26].
The genus Spiroplasma belongs to the Mollicutes class, and the Tenericutes phylum. Spiroplasma are found abundantly in insects either in the gut or haemolymph where they have developed a large variety of interactions with the host that can be classi ed as commensal, pathogenic or mutualist [13].
It has been shown that Spiroplasma confers protection against pathogens e.g Drosophila neotestacea from a nematode [27], the pea aphid against fungi [28] and Drosophila hydei against a parasitoid wasp Leptopilina heterotoma [29]. However, reproductive alterations such as cytoplasmic incompatibility, malekilling and sex determination have been related to numerous species of Spiroplasma [13].
Recently, Spiroplasma has been established as a new class of tsetse symbiont in G. fuscipes fuscipes, G. tachinoides, and G. palpalis palpalis. The interactions between Spiroplasma and Glossina seem to be bene cial because of its ability to extend lifespan and reduce the vector competence for Trypanosoma [11,30].
Current control measures against trypanosomiasis are mainly based on chemotherapy. In the absence of effective vaccine and to address the limitations associated with chemotherapy, disruption of trypanosomes transmission through vector control is crucial. Transmission of pathogens by vector depends on its vector competence, which can be affected by several factors, including vector endosymbionts [31]. Due to their importance, interactions between the symbionts and their hosts are being harnessed toward the development of novel approaches for vector and disease control [32][33][34].
The present study aims to investigate the presence of Sodalis, Spiroplasma and Wolbachia in wild population of tsetse ies from Cameroon, Nigeria and Chad.

Sodalis glossinidius infection prevalence
The presence of Sodalis glossinidius was investigated in 678 eld-collected tsetse ies, originating from Cameroon (149), Chad (259) and Nigeria (270) using a Sodalis Hemolysin gene-based PCR. The Sodalis infection rates were 2.01% in Cameroon, 1.16% in Chad and 0.0% in Nigeria (Table 1). The prevalence of S. glossinidius infection did not differ signi cantly between Cameroon and Chad, while it was null in Nigeria.
The sequence of our amplicons showed a high similarity (> 99%) with other Sodalis's Hemolysin partial gene sequences found in NCBI database. The Maximum Likelihood phylogenetic tree ( Fig. 1) showed that Cameroonian amplicons formed a different clade with amplicons from Chad.

Spiroplasma infection prevalence
A total of 847 y samples originated from Cameroon (154), Chad (423) and Nigeria (270) were used for the molecular detection of Spiroplasma using a 16S rRNA-based PCR approach with wspecF/wspecR primers. Only 31 samples were positive (3.66%). The prevalence varied signi cantly (p = 0.00) between countries (Table 1). Spiroplasma was detected in 0.0%, 0.95% and 10.00% of samples respectively from Cameroon, Chad and Nigeria.
The sequence of our amplicons showed a high identity (> 99%) with other Spiroplasma 16S rRNA partial gene sequences found in NCBI database. The Maximum Likelihood phylogenetic tree (Fig. 2) showed that Chadian and Nigerian amplicons belong to 2 different clades.

Wolbachia infection prevalence
A total of 582 tsetse samples (gut or abdomen) were screened for the presence of Wolbachia with a 16S rRNA-based PCR approach using the wspecF/wspecR set of primers. The samples were collected in 3 countries (Cameroon: 155, Chad: 157 and Nigeria: 270). The results of the screening (Table 1) showed that the prevalence of Wolbachia infections varied, but not signi cantly (p = 0.06) between the three countries. The Wolbachia prevalence in Cameroon, Chad and Nigeria was 9.68%, 7.00% and 14.07% respectively.
The sequence of our amplicons showed a high identity (> 99%) with other Wolbachia 16S rRNA partial gene sequences found in NCBI database. The Maximum Likelihood phylogenetic tree (Fig. 3) showed that Cameroonian and Nigerian amplicons clustered together while they are far away from that from Chad.
All 16S rRNA and hemolysin gene sequences generated in this study have been deposited into GenBank under accession numbers OQ448931 to OQ448937 and OQ458709 to OQ458712.
Prevalence according to the y species All the 4 species were found to be infected at different rates by the three symbiotic bacteria except G. tachinoides ies that were not infected by S. glossinidius (

Discussion
Owing to their restricted source of meal, the vertebrate blood, tsetse ies are highly dependent on their microbial ora which provide them with complementary nutrients [23]. Symbiotic associations have been established with some bacteria. In recent years, endosymbionts have received increased attention because of their large distribution in several insects and their effects on host physiology. Symbiotic interactions in tsetse ies have been discussed for their implication in vector competence of the ies [11].
In this study, tsetse symbiotic bacteria (Sodalis, Spiroplasma and Wolbachia) were detected in 11.20% of the tsetse samples examined with varying prevalence within countries, collection sites and tsetse species.
This overall symbiont infection rate is lower compared to that of other previous studies [35][36][37]. Varying levels of infection rates may be attributed to environmental factors (vegetation, humidity, temperature) encountered in different ecological settings and to the intrinsic characters of each tsetse species.

Sodalis prevalence
The overall S. glossinidius infection rate of 0.88% obtained in the present study is lower than the 12.69% global prevalence of Sodalis in tsetse samples from 15 African countries [35]. However, in the same study, similar prevalence was obtained in some West African countries: 0.48% in Burkina Faso, 0.00% in Mali, Ghana and Senegal. The Sodalis prevalence in Chad (1.16%) was lower than the 9.0% previously reported in the same area [38]. The prevalence in Cameroon was also lower than the 37.2% previously reported in the neighbouring area [36]. The variation may be due to PCR screening approach. The primers used in the previous works (pSG2 primers) were targeting a portion of the extrachromosomal plasmid 2 while in this present work, the primers (Hem primers) were targeting the nuclear hemolysin gene. Comparing different sets of primers, it has been reported that the use of the hemolysin gene provided a more reliable assessment of prevalence than the pSG2 that was giving higher but non-consistent prevalence [39]. Moreover, tsetse ies screened in Cameroon for the present study were all G.p. palpalis while the previous study screened G. tachinoides and G. m. submorsitans. Up to date in Nigeria, there is no published data on S. glossinidius prevalence in tsetse ies.
The presence of S. glossinidius has been suspected to be involved in vector competence of tsetse y. But the con rmation is still under debate. Numerous reports showed a positive association between Sodalis and trypanosome establishment in the midgut possibly through lectin-inhibitory activity involving the production of N-acetyl glucosamine [26,38,40]. However, other studies reported the absence of association between the presence of Sodalis and that of Trypanosome [35][36][37].

Spiroplasma prevalence
The Spiroplasma general infection rate of 3.66% obtained in the present study is much lower than 17.17% and 44.5% reported respectively in Burkina Faso for G. tachinoides and in Uganda for G. f. fuscipes [30,41]. The study in Uganda showed a high variation of prevalence across the 26 sampling sites ranging from 0-62% and was correlated with the geographic origin and the season of collection of G. f. fuscipes [30]. The prevalence in Nigeria (10.0%) was signi cantly higher (p = 0.00) compared to that of Cameroon (0.0%) and Chad (= 0.95%). Few investigations have been done on the prevalence of Spiroplasma in tsetse ies in Africa. Up to date, there is no existing data on tsetse Spiroplasma prevalence in the 3 countries where our samples were collected.
Spiroplasma is known to induce reproductive abnormalities and pathogen protective phenotypes in various arthropods hosts [11]. A negative correlation of Spiroplasma with trypanosome infection was found in G. f. fuscipes, indicating that Spiroplasma infections may have an important effect in y's resistance to infection with trypanosomes [30,42]. In an experimental study conducted by [42] on G. f. fuscipes, it was discovered that Spiroplasma infection induced changes in sex-biased gene expression in the reproductive tissues, adepletion in the availability of nutrients in pregnant females resulting in delayed larval development, and compromised sperm tness. These ndings indicate that Spiroplasma could be exploited for reducing tsetse population size and therefore, the disease transmission [42].

Wolbachia prevalence
The Wolbachia global prevalence of 11.00% and countries prevalence (Table 1) obtained in the present study are in the range of prevalence reported by [18]. All these values were lower than 80.5% and 78.9% reported in Zambia for G. m. morsitans and G. pallidipes respectively [37]. In Cameroon and Chad, the prevalence previously obtained in the same and neighbouring study sites (67.6% and 14.5% respectively) were much higher than those of the present study [36,38]. The primers used in this study were targeting the Wolbachia speci c 16S rRNA while the previous studies used primers targeting the Wolbachia surface protein (wsp) gene. The comparison of results generated by 16S rRNA and wsp primers by [43] did not nd a signi cant difference between the 2 primers, even though the prevalence with 16S rRNA primers (65%) was higher than the prevalence with wsp primers (60.5%). The tsetse species were globally different amongst the two studies. The prevalence in Nigeria (14.07%) was relatively higher than those of the two other countries (Cameroon: 9.68% and Chad: 7.00%). However, up to date there is no published data on Wolbachia infection in tsetse ies in Nigeria.
Wolbachia has received increased interest in recent years because of its high rates of distribution in a wide range of arthropods and nematodes, its unique effects on host physiology and its potential in disease control. Concerning trypanosomiasis, there is no con rmed implication of Wolbachia in tsetse vector competence. Investigations on the tripartite association between tsetse y, Wolbachia and trypanosomes reported contrasting results. When in G. f. fuscipes, it was found a negative association between Wolbachia and trypanosome [44] suggesting a prevention of trypanosome infections by the presence of Wolbachia; some experimental studies did not nd an association between the presence of Wolbachia and the ability of tsetse ies to harbor trypanosomes [36, 37,43].

Tsetse species
Spiroplasma was found to infect the 4 Glossina species, but with a low infection on G.m. submorsitans and G.f. fuscipes. On the contrary, the investigations by [13] found Spiroplasma infection only in palpalis group and nothing in the morsitans and fusca groups attributing this result to their frequent infection with Wolbachia which may have led to the development of competitive exclusion with Spiroplasma. Wolbachia was detected in all the 4 species with varying prevalence within species. This variation agrees with ndings from similar studies on tsetse symbionts. In the present study, G. tachinoides was found to be more infected with Wolbachia than the 3 other tsetse species. Sodalis was not found in G. tachinoides. However, Sodalis is found to be infecting G. tachinoides in other studies [35,36].

Conclusion
The present study revealed for the rst time, the presence of infection by Spiroplasma in tsetse in Chad and Nigeria. The infection rates of Sodalis, Spiroplasma and Wolbachia varied between countries and collection sites probably because of environmental factors. Few tsetse ies harboured symbiont coinfections. These ndings provide useful information to the repertoire of bacterial ora of tsetse ies and incite to more investigations to understand the implication of these symbiotic bacteria in the vector competence of tsetse ies. More data could help the development of environmentally friendly methods for both tsetse and trypanosomiasis control.

Study areas
Tsetse ies used in this study were collected from three countries: Cameroon, Chad, Nigeria (Fig. 4).
In Cameroon ies were trapped in Dodeo (Latitude 7° 27.994' N and Longitude 12° 04.101′ E) located in the "Faro et Déo" division of the Adamawa region. The village is close to the Cameroon-Nigeria Border. The vegetation is predominantly characterized by gallery forest along rivers and the climate type is of Sudano-Sahelian with two seasons: A rainy season from May to October and a dry season from November to April. This area has a dense hydrographic network and offers pastures for livestock [5,36].
In Chad, ies were collected in Maro and Lake Iro areas, all situated in the Moyen-Chari province in southern Chad. The area has a climate of Sudano-Sahelian type with two seasons of equal duration. A rainy season from May to October and a dry season from November to April. Lac Iro is situated between the Latitude 9°59' N and Longitude 19°26′ E, along the Salamat River. The vegetation is dominated by oodplains and dense forests containing shrubs. The area is considered as a buffer zone of the Zakouma national park where domestic and wild animals can meet [38]. Maro (8°24.807' N and 18°46.139' E) is located at the Chad-Central African Republic border. The area is crossed by many rivers and their multiple tributaries; the most important is the Chari River and its con uent the Grande Sido [45].
Maro is close to the Bamingui-Bangouran National Parc. In

Sampling of tsetse
Biconical traps were used for this study. They were set up in various tsetse y-favourable biotopes at distances of at least 100 m intervals. Tsetse ies were collected once a day and transported in cool boxes to the base camp. The collected ies were identi ed using morphological identi cation keys [47,48], numbered and dissected on the same day.

Dissection
Flies were dissected in a drop of sterile saline solution as described by [5,46]. The wings were removed rst, followed by legs, proboscis, salivary glands and gut. After each y, dissection tools (forceps, pins and slide) were decontaminated by immersion in 5% sodium hypochlorite for approximately 10 minutes, followed by immersion in 70% ethanol and nal immersion in sterile normal saline.
Wings were stored dry for geometric morphometrics. Legs, proboscis and salivary glands were preserved in nucleic acid preservation agent (NAPA: 25 mM sodium citrate, 10 mM EDTA, 70 g ammonium sulphate/100 mL solution, pH 7.5) in 1.5 mL cryotubes. Gut tissue was homogenised in 200 µL of 50 mM Tris-Cl, pH 9.0, using four 2.38 mm metal beads (MoBio Laboratories, Carlsbad, CA, USA). Fifty microliters of the homogenate were added to 500 µL of NAPA.
Most of the ies from Chad were dead. For this reason, after the removal of wings, legs and proboscis, the entire abdomen was transferred in ethanol.
DNA extraction DNA was extracted from homogenised gut tissue in NAPA with the DNeasy Blood and Tissue Kit (Qiagen, Germany) according to the manufacturer's instructions with slight adjustments (100 µL of homogenate were used for puri cation and 100 µL elution buffer were used at elution step).
For the dead ies from Chad, DNA was extracted from the abdomen using 5% Chelex-100 Resin (BIO-RAD, Hercules, California, USA). Brie y, the abdomen was transferred from ethanol to a new 1.5 mL centrifuge tube and allowed to dry. The abdomen was then crushed using the micro-pipette tips. Thereafter, 100 µL of chelex 5% solution were added. The mixture was vortexed and incubated at 56° C for 30 min in a Thermomixer. After this incubation, the mixture was vortexed and incubated for an additional 5 min at 95° C. The mixture was then mixed thoroughly before brief centrifugation at 7000 rpm for 1 min and stored at -20°C.

Molecular detection of Wolbachia
The detection of Wolbachia was carried out by amplifying a 438 bp fragment of the 16S rRNA gene with the 16S W-Spec primers (Table 3) designed by [49].
PCR ampli cations were performed in 20 µL reaction mixture containing 1 X DreamTaq buffer (10X), 150 µM dNTPs, 0.2 µM of each primer, 0.5 U of DreamTaq polymerase and 2 µl of template DNA. The cycling condition was as described by. It started by 95°C for 5 min followed by 35 cycles of 95°C for 30 s, 30 s at 54°C, 1 min at 72°C and a nal extension step of 72°C for 10 min. After ampli cation, the PCR products were analyzed by electrophoresis on a 1.5% agarose gel containing Stain-G and visualized under UV light.

Molecular detection of S. glossinidius
The presence of S. glossinidius was determined by PCR with Hem primers ( Table 3) that target the gene encoding the haemolysin protein of the bacterium [50]. The primers targeted a 650 bp fragment of the hemolysin gene. The PCR reaction was performed in a nal volume of 20 µL containing 1X Dream taq buffer, 150 µM dNTPs, 0.2 µM of each primer, 0.5 U of Dream Taq polymerase and 2 µL of DNA template. PCR cycles were: 95°C for 5 min; 35 cycles of 95°C for 30 s, 54°C for 30 s and 72°C for 60 s; and a nal elongation at 72°C for 10 min.

Molecular detection of Spiroplasma
The screening for Spiroplasma was carried out by amplifying the 455 bp fragment of the 16S rRNA as described by [13] with speci c primers (Table 3).
PCR reactions were performed in 20 µl reaction mixture containing 1X DreamTaq buffer, 150 µM dNTPs, 0.65 µM of each primer, 0.5 U of DreamTaq polymerase and 2 µl of template DNA. The PCR temperature pro le was 95°C for 5 min followed by 35 cycles of 95°C for 30 s, 30 s at 59°C, 1 min at 72°C and a nal extension step of 72°C for 10 min.

Phylogenetic analysis
Obtained sequences were manually checked and edited using Geneious Pro version 5.5.9 software. The Basic Local Alignment Search Tool (BLASTn) from National Center for Biotechnology Information (NCBI) was used to con rmed the identity of the amplicons and to determine the closest related sequences in the GenBank.
Sample sequences and reference sequences obtained from NCBI were aligned using MUSCLE alignment tool with its default setting implemented in MEGAX [51] which was also used to infer phylogenetic relationships. Maximum likelihood method was performed with the Kimura-2 model [52] for Wolbachia and Spiroplasma, then Hasegawa-Kishino-Yano model [53] for Sodalis as determined by the MEGA model nder tool with 1000 bootstraps replicates.

Statistics Data analysis
Endosymbionts hosted by tsetse ies were expressed in percentage as symbiont prevalence. The Pearson's chi-square test (χ 2 ) was used to compare symbiont prevalences between countries and collection sites. The datasets generated and analysed during this study are included in this publication. The 16S rRNA and hemolysin partial gene sequences generated have been deposited into GenBank under accession numbers OQ448931 to OQ448937 and OQ458709 to OQ458712.
Competing interests: The authors declare that they have no competing interests.

Funding
Financial support for the project came from WANIDA (West African Network of Infectious Diseases ACEs) doctoral scholarship (WAN100629X), scholarship package funded by the Agence Française de    Study area. Tsetse ies were collected in Cameroon (Dodeo), in Chad (Maro and Lake Iro) and in Nigeria (Yankari Game reserve and in Ija-Gwari).