Molecular identification of trypanosome species in trypanotolerant cattle from the south of Gabon

The aim of this study was to provide information on trypanosome species infecting trypanotolerant cattle from southern Gabon. The study was conducted on 224 trypanotolerant cattle from three regions located in southern Gabon, using ITS1 primer-based PCR. Seventy-two (32%) N’dama cattle were found polymerase chain reaction (PCR) positive with trypanosomes. The overall prevalence of trypanosomosis was 57% (63/110), 4% (4/100), and 36% (5/14) in the Gala section of the Nyanga ranch, the Miyama ranch, and Ossiele, respectively. Trypanosoma congolense and Trypanosoma vivax were identified. In Gala section and Ossiele, T. congolense and T. vivax were found. In the Miyama ranch, only T. vivax was identified. Mixed infections were also found. The forest (9%) and savannah (63%) subgroups of T. congolense were identified. The presence of the two subgroups was detected in 16 out of 56 cattle (29%). T. congolense and T. vivax would appear to be the main agents responsible for bovine trypanosomosis in southern Gabon. Although trypanotolerant, N’dama cattle may serve as a reservoir, and this should be further studied. On the other hand, these trypanotolerant cattle can be reared in such tsetse infested areas, which gives them an advantage compared to other trypanosensitive breeds, and this shows that they represent a key factor in biodiversity which has to be promoted.


Introduction
Trypanosomes are extracellular protozoan parasites of vertebrates, including domestic cattle, from tropical regions. Trypanosomes cause diseases known as trypanosomosis. In sub-Saharan Africa, cattle trypanosomosis is caused mainly by Trypanosoma congolense, Trypanosoma brucei, and Trypanosoma vivax (responsible for the disease known as nagana in cattle). African mammals harbor other pathogenic trypanosomes such as Trypanosoma simiae, Trypanosoma suis, Trypanosoma uniforme, Trypanosoma evansi (responsible for surra, trypanosomosis of Camelidae and Equidae), and Trypanosoma equiperdum (responsible for dourine, equine trypanosomosis) [30]. Trypanosomes are mainly transmitted between mammalian hosts by the bite of an infected tsetse fly (biological vectors); however, some trypanosomes can be transmitted by hematophagous Diptera (mechanical vectors) belonging to the Tabanidae, Stomoxyinae, and Hippoboscidae [10,19,26]. This mechanical transmission is especially known for T. vivax and T. evansi, which are mainly found in areas beyond the range of tsetse flies [6,10].
In Gabon, the determination of breeding areas and the introduction of animal herds are problematic in terms of animal trypanosomosis. The few savannah areas of southern Gabon are crossed by many forest galleries and dotted with dense groves. These woodlands are the preferred shelters of tsetse that thrive in this type of environment. In addition, only little information is available on the status of bovine trypanosomosis in Gabon, especially regarding trypanosomes and tsetse species as well as the affected geographical areas. However, the presence of Glossina haningtoni and Glossina tabaniformis has been confirmed in forests and forest galleries of the Nyanga valley, southern Gabon [29]. A survey conducted in 1986 in the Okouma ranch, in the south-east, revealed the presence of G. nashi and G. tabaniformis [14]. G. palpalis palpalis was identified in different regions of the country: in the region of the Nyanga valley [29], as the main species, and during active outbreaks of human trypanosomosis, in Komo Mondah and Bendjé, in the north [11,12], and in the north-east (Ogooué-Ivindo province) [32]. In this last region, G. frezili, G. fuscipes fuscipes, G. nashi, and G. fusca congolense have also been identified [32]. In the Nyanga valley, Trypanosoma congolense and Trypanosoma vivax have been identified in G. p. palpalis. During the Komo Mondah and Bendjé outbreaks, entomological research revealed infection of tsetse by T. brucei, T. vivax, and T. congolense. The prevalence and incidence of bovine trypanosomosis in Gabon are very poorly known. The few studies that have been conducted concern the Okouma ranch, south-east of Gabon. Trypanosoma vivax was identified in cattle at a rate of 26% using parasitological tools [14].
The aim of this study was to provide information on the circulation of trypanosome species in trypanotolerant cattle from southern Gabon and to determine the molecular prevalence of infection of these species in order to evaluate the risk of bovine trypanosomosis in the main cattle breeding regions of Gabon.

Study areas
The study was carried out on cattle from three areas located in two provinces of the south-east of Gabon: the Miyama ranch and the rural town of Ossiele, in the Haut-Ogooué province; the Gala section of the Nyanga ranch in Nyanga province (Fig. 1). In addition to breeding cattle, these areas were selected because animal trypanosomosis has frequently been suspected on the basis of clinical signs.
The Miyama ranch is located about 30 km from Franceville, the main city. The livestock is composed of about 200 animals. N'dama trypanotolerant cattle are reared. The ranch is in a vegetation area composed of savannah and forest transitions.
The rural town of Ossiele is located in the department of Djoué, south of the town of Onga, the capital city of the department. In this region, the vegetation is also made of forest and savannah transitions. In these two regions, the surrounding watercourses are located several kilometers away, making them difficult to access by animals.
The Nyanga ranch is located in the largest savannah zone of the country in the department of Mongo-Moulengui Binza, 65 km from the town Tchibanga (the capital city of Nyanga province) and extends over 100,000 hectares. The Nyanga ranch includes three divisions (Nyanga, Bibora, and Voungou) divided into sections. The study was conducted within the Gala section of Nyanga division 1.

Animal sampling and blood collection
The samples were collected in cattle, at random (without any distinction in age and sex). Animals that received trypanocide treatment (diminazene aceturate, and two weeks later isometamidium chloride) less than three months before the start of our study were excluded from sampling because the chemoprevention provides a three-month protection period. Approximately 5 mL of whole blood was collected from each animal in ethylenediaminetetraacetic acid (EDTA) vacuum tubes VENOJECT Ò , by puncture of the caudal vein. EDTA tubes containing whole blood were centrifuged at 1300 rpm for 10 min [7] and the plasma was recovered with buffy coat and stored at À20°C until shipment to the laboratory of the Centre International de Recherches Médicales de Franceville (CIRMF), Gabon, for analysis.

DNA extraction and PCR
Total DNA extraction was conducted from 200 lL of a mixture plasma-buffy coat, using the Qiamp DNA Mini Kit (Qiagen) according to the manufacturer's guidelines. The DNA was stored at À80°C until analysis.
Trypanosome research was carried out using two PCRs. In the initial PCR, the intergenic sequence (ITS1) primers were used to amplify the partial intergenic sequence region of the ribosomal DNA (rDNA) of trypanosomes [22]. Amplification was performed in 25 lL reaction volumes, containing 2.5 lL of 10X PCR buffer (Platinum Ò Taq DNA Polymerase, Invitrogen), 2 lL of dNTP (10 lM each), 2 lL MgCl 2 (50 mM), 0.24 lL of each primer (10 lM), 0.3 lL enzyme Platinum Taq (Platinum Ò Taq DNA Polymerase, Invitrogen), 12.72 lL of RNAse-free water (Invitrogen), and 5 lL of DNA. Amplification generally involved 5 min at 95°C followed by 40 cycles of 1 min at 95°C, 1 min at 60°C, and 1 min at 72°C followed by a final elongation at 72°C for 10 min. The PCR products of approximately 700 base pairs (bp), characteristic of T. congolense, were subjected to T. congolense subtype-specific PCR to distinguish them between Forest, Savannah, and Kilifi. The primer sets used are shown in Table 1. The PCR conditions are the following for all three PCRs: 5 min at 95°C followed by 35 cycles including 1 min at 95°C, 1 min at 55°C, and 1 min at 72°C followed by a final elongation at 72°C for 10 min.

Sequence analysis and phylogenetic analysis
Sequence cleaning and assembling of contigs was conducted in ChromasPro 1.5. The nucleotide sequences were compared to those available in the public database using the algorithm ''Blastn'' of the NCBI BLAST [1]. Phylogenetic analyses were conducted to infer the relationship with other Trypanosoma spp. The sequence alignment was performed using the multiple sequence alignment program, ClustalW, contained in MEGA 5 [28]. A neighbor-joining tree was constructed using a Kimura 2-parameter substitution model, with bootstrap values calculated for 1000 replicates. The new sequences were deposited in GenBank with Accession Numbers KX452152-KX452176.

Statistical analysis
Analyses were performed using R software version 3.1.0 (Development Core Team 2014, USA). The chi-square (v 2 ) test of Pearson was used for the comparison of the proportions of animals infected by different trypanosome species between study areas.

Results
Twenty-nine rDNA sequences were generated from cattle samples in this study. Blast analysis confirmed that close matches were found with T. congolense and T. vivax, respectively, for the sequences of 710-and 250-bp products.
Twenty-five of 29 sequences generated in this study were included in the phylogenetic analysis. The T. vivax sequence obtained from cattle from the Miyama ranch shared 77% identity with the T. vivax reference sequence from a cow from Burkina Faso (JX910370), and 69% identity with four other sequences from cows from Burkina Faso (JX910372, JX910375, JX910376, and JX910379). In the phylogenetic tree reconstructions, based on the nucleotide sequences of the ITS1 region of the ribosomal DNA, the Gabonese T. congolense strains from the Gala section grouped into two major clusters. Cluster I comprised 14 T. congolense strains belonging to the Savannah subtype; whereas 10 T. congolense strains from Gabon and three other strains, including one from Kenya (FJ712718), were grouped within cluster II, bringing together strains belonging to the Forest subtype (Fig. 2). T. congolense sequences from cluster I shared 34-76% identity. The T. congolense variants grouped within cluster II displayed 43-77% identity with each other at the nucleotide level. These 10 T. congolense sequences shared 39-65% identity with the Kenyan reference sequence (FJ712718) and 37-64% with the reference sequence JX910374 from a cow from Burkina Faso.
The T. congolense-positive samples (n = 56) were then screened using a specific PCR for the three subgroups of T. congolense: T. congolense savannah, T. congolense forest, and T. congolense kilifi. T. congolense savannah was the dominant subgroup with 35 animals infected (62.5%) followed by T. congolense forest with 5 infected animals (8.9%). The T. congolense savannah subgroup is significantly more predominant (v 2 = 35, ddl = 1, p < 0.001) than the T. congolense forest subgroup in Gala and Ossiele. Furthermore, 16 of 56 (28.6%) animals had mixed infections with both subgroups. No infection with the T. congolense kilifi subgroup was discovered.

Discussion
Prevalence of bovine trypanosomosis and molecular identification of trypanosome species were studied in three cattle farming areas in the south of Gabon, namely the Gala section of the Nyanga ranch, the Miyama ranch, and Ossiele. Phylogenetic analysis was conducted in order to confirm the BLAST results and therefore the identity of the nucleotide sequences obtained. The phylogenetic tree showed that the variants of T. congolense, all from the Gala section, displayed considerable genetic variation and were classified into two major clusters. T. congolense genetic variants have also been reported in East and West Africa [3,8]. The identification of this T. congolense genetic diversity in the same geographical area indicates the existence of various isolates. As suggested by Auty et al. [3], this genetic diversity cannot be explained only by geographical variation. In the Gala section, cattle are frequently in contact with many other wild animal species, suggesting the possibility of a transfer of strains specific to these wildlife host species [3] through specific species of tsetse. For example, host-specific strains of T. theileri were identified in cattle and water buffalo in the same geographical areas [23]. T. vivax obtained from cattle from the Miyama ranch shared 69-77% identity with T. vivax strains from cows   The molecular prevalence of trypanosomosis was significantly higher in the Gala section than in Ossiele and the Miyama ranch. Out of the 224 N'dama cattle studied, 72 were found PCR positive with trypanosomes, with an overall prevalence of infection of 32.14%. This result was higher than the one found in the cattle in Ghana, using parasitological methods (1.88%) and serology (22.08%) [9], and in the cattle in Northwest Ethiopia using parasitological methods (5.43%) [15]. Otherwise, Nimpaye et al. [21] found an infection rate of 27.08% in domestic animals in Cameroon, using PCR assays. Moreover, Takeet et al. [27] detected 15.1% positive infection with trypanosomes in cattle from Nigeria, using microscopy, while PCR detected 63.7% positive infections. The highest infection rate was found in Gala (57.3%) followed by Ossiele (35.7%) and the Miyama ranch (4%). Indeed, in these three areas, the vegetation is made of savannah and plains crossed by numerous forest galleries and dotted with dense groves. However, the number and proximity of watercourses differ. The Gala section contains several watercourses including rivers and streams. These watercourses, surrounded by moist and shaded forest galleries [29], are shelters for tsetse flies. However, in the Miyama ranch, the watercourses bordering the area are located in distant hilly areas making it difficult to access for animals. The low infection rate found in the cattle from the Miyama ranch could therefore be explained by less frequent contacts between cattle and tsetse flies for the above reasons. Furthermore, the absence of T. congolense can probably be due to the absence of a biologically competent vector.
Two species of pathogenic trypanosomes have been identified, namely T. congolense and T. vivax. These two trypanosome species are the main agents of nagana [25]. The presence of these two species in Gabon has been highlighted by an entomological study in the Nyanga valley [29], and in tsetse flies captured during an active outbreak of human trypanosomosis in Komo Mondah [13] and a historic outbreak in Bendjé, in the Ogooué-Maritime Province in western Gabon [11,12]. In Gala, T. congolense and T. vivax were identified. These findings support those of Taufflieb [29], who identified these two species in Glossina palpalis palpalis in the Nyanga valley with an infection rate of 3% and 2%, respectively. In addition, mixed infections with both trypanosome species have also been identified (4.5%). In the cattle from Ethiopia, T. congolense was found to be the dominant species (66.7%), compared to T. vivax (9.3%) [22], and 19.4% were mixed infections. According to Taufflieb [29], in the Nyanga area, G. p. palpalis is the main tsetse fly vector of trypanosomosis. The G. p. palpalis species lives in secondary forest, in vegetation along streams and swamps [31]. In the Miyama ranch, only T. vivax was found. This species was already identified by parasitological methods, with an infection rate of 26% in cattle from a ranch located in the south-east of Gabon [14]. The tsetse fly of the Fusca group, G. tabaniformis, was responsible for the transmission of T. vivax in this area [14]. Although a competent vector is present in the area, T. vivax can also be found outside of the area of distribution of tsetse because it can be transmitted mechanically by flies from the Tabanidae, Stomoxyinae, and Hippoboscidae families [10,18].
The overall prevalence of T. vivax (7.1%) was lower than that found in cattle and other domestic animals [20,21,24,27]. However, the overall prevalence of T. congolense (25%) was high in contrast to that previously reported in other domestic animals [20, 21,24]. The very low prevalence of T. vivax with respect to T. congolense would indicate a low transmission of T. vivax in the studied localities, which is better controlled by animals because their genetic diversity is more limited than those of T. congolense and T. brucei [2]. The prevalence of T. vivax found in this study using the primers ITS1 CF and ITS1 BR is much lower than that obtained by Leak et al. [14] from cattle reared in the same area, using parasitological methods. This difference could be explained, on the one hand, by the improvement over recent years of curative and preventive treatments against trypanosomes, and on the other by the lower sensitivity of ITS1 CF and BR ITS1 primers for T. vivax compared to species-specific primers [22].
The subtyping of T. congolense allowed us to determine the circulation of two subgroups out of three in the Gala section and Ossiele, including T. congolense savannah and T. congolense forest, respectively, with overall infection rates of 62.5% (57.14% and 5.36% in Gala and Ossiele) and 8.9% (7.14% and 1.79%). Multiple subgroups of T. congolense have already been found in multiple ecosystems, and in the same location [16]. Mixed infections with T. vivax and T. congolense were found. Mixed infections with multiple species of trypanosomes have previously been reported [17]. The infection of cattle with these two pathogenic species exacerbates disease severity even in trypanotolerant cattle. Mixed infections by the two subgroups were also detected (28.6%). In Gabon, the T. congolense forest subgroup was predominant in the outbreak of Bendjé, in western Gabon [11]. The infection rate with the forest subgroup found in this study is higher than that found in Cameroon, a neighboring country, by Nimpaye et al. [21] in domestic animals. The savannah subgroup exhibits a greater virulence than the other two subgroups, experimentally demonstrated in cattle and mice [4,5]. Further studies are needed to determine the real prevalence of T. vivax and the active potential of the infection. Indeed, these findings should be coupled with measurements of packed cell volume or a microscopic confirmation or with another tool, to determine (i) the risk of trypanosomosis in the main breeding areas of this country in Central Africa, because of the co-circulation of these two pathogenic species of trypanosomes, and (ii) the need to increase control measures for trypanosomosis in livestock breeding areas in Gabon, because although trypanotolerant, N'dama cattle could be a reservoir for these two parasites (T. vivax and T. congolense) and therefore a risk for programs targeting the elimination of animal trypanosomosis in the region and an obstacle to the introduction of other more trypanosusceptible but productive cattle breeds. On the other hand, these trypanotolerant cattle can be reared in such tsetse infested areas. This is an enormous advantage compared to other breeds, and it shows that they represent a key in biodiversity which has to be promoted.