Introduction
Trypanosoma vivax causes trypanosomosis in cattle. This haemoprotozoan originating from Africa is believed to have been introduced in South America around the year 1830, with the transport of infected cattle from Senegal (Ventura et al., Reference Ventura, Paiva, Silva, Takeda, Buck and Teixeira2001; Osório et al., Reference Osório, Madruga, Desquesnes, Soares, Ribeiro and Costa2008). In sub-Saharan Africa, the transmission of this protozoa occurs biologically through the tsetse fly (Glossina spp.). In Central and South America, transmission occurs mechanically by biting flies (tabanids, Stomoxys calcitrans and Haematobia irritans) or by iatrogenic means (Cadioli et al., Reference Cadioli, Barnabé, Machado, Teixeira, André, Sampaio, Fidélis Junior, Teixeira and Marques2012; Dagnachew and Bezie, Reference Cadioli, Barnabé, Machado, Teixeira, André, Sampaio, Fidélis Junior, Teixeira and Marques2015; Bastos et al., Reference Bastos, Faria, Madrid, Bessa, Linhares, Fidelis Junior, Sampaio, Cruz, Cruvinel, Nicaretta, Machado, Costa and Lopes2017).
The first outbreak of trypanosomosis in Brazil occurred in the state of Pará in 1972 (Shaw and Lainson, Reference Shaw and Lainson1972). From the year 2000 to 2017, occcurences have been reported in nearly 60% of the states of the country: Mato Grosso do Sul, Mato Grosso, Paraíba, Maranhão, Tocantins, Minas Gerais, São Paulo, Rio Grande do Sul, Pernambuco, Alagoas, Goiás, Sergipe, Piauí, Rio de Janeiro and Rio Grande do Norte (Silva et al., Reference Silva, Silva, Schneider, Freitas, Mesquita, Mesquita, Laura, Dávila and Pereira1996; Paiva et al., Reference Paiva, Lemos, Nakasato, Mori, Brum and Bernardo2000; Linhares et al., Reference Linhares, Dias Filho, Fernandes and Duarte2006; Batista et al., Reference Batista, Riet-Correa, Teixeira, Madruga, Simões and Maia2007; Carvalho et al., Reference Carvalho, Abrão, Facury Filho, Paes and Ribeiro2008; Guerra et al., Reference Guerra, Júnior, Batista, Santos, Abreu-Silva and Santos2008; Silva et al., Reference Silva, Costa, Polenz, Polenz, Teixeira, Lopes and Monteiro2009; Cadioli et al., Reference Cadioli, Barnabé, Machado, Teixeira, André, Sampaio, Fidélis Junior, Teixeira and Marques2012; Pimentel et al., Reference Pimentel, Ramos, Ramos, de Araújo, Borba, Faustino and Alves2012; Andrade Neto et al. Reference Andrade Neto, Afonso, Mendonça, Souto, André and Machado2015; Costa et al., Reference Costa, Abreu, Machado, Thomé, Massard, Santos and Brito2016; Bastos et al., Reference Bastos, Faria, Madrid, Bessa, Linhares, Fidelis Junior, Sampaio, Cruz, Cruvinel, Nicaretta, Machado, Costa and Lopes2017; Vieira et al., Reference Vieira, Macedo, Santos, Silva, Mendonça, Faustino, Ramos, Alves, Ramos and Carvalho2017; Lopes et al., Reference Lopes, Prado, Martins, Esmeraldo, Beserra, de Sousa Filho, Evangelista and Souza2018).
In the present study, we evaluated the spatial distribution, prevalence and risk factors of the acute occurrence of trypanosomosis in cattle in the state of Goiás, Brazil, and performed molecular analyses for the identification of T. vivax during an outbreak, with comparisons to previous findings based on the 18S rRNA gene.
Materials and methods
Ethics statement
This study received approval from the Animal Use Ethics Committee of the Federal University of Goiás, Brazil (certificate number: 032/15) and was conducted in compliance with the ethical principles governing animal experimentation of the Brazilian National Animal Experimentation Control Council (CONCEA).
Properties and animals for the study of T. vivax
The properties registered with the Agriculture and Livestock Defense Agency of the state of Goiás (AGRODEFESA-GO) that did not receive any medication with specific action against T. vivax, had cattle with acute problems suggestive of trypanosomosis and reported recent animal deaths were selected for visits between May 2015 and May 2017. The owners were contacted by phone for visit arrangements.
Forty-two ranches were visited in 26 cities in the state of Goiás: Alexânia, Anápolis, Bonfinópolis, Buriti Alegre, Caldas Novas, Campo Alegre de Goiás, Corumbaíba, Cromínia, Edealina, Gameleira de Goiás, Goianápolis, Goianésia, Goiatuba, Guapó, Ipameri, Itaberaí, Itauçu, Jataí, Mairipotaba, Mambaí, Morrinhos, Pontalina, Porteirão, Quirinópolis, Santa Bárbara de Goiás and Urutaí. A total of 4049 blood samples were collected from cattle. Approximately 73% of the bovines evaluated were the Girolando breed (1/2 Holstein + 1/2 Gyr and 3/4 Holstein + 1/4 Gyr). The other breeds were Holstein (7/8 Holstein + 1/8 Gyr and 15/16 Holstein + 1/16 Gyr), Gyr, Jersey or crossbreed/Nelore.
Animal histories and blood collections were taken from at least 90% of the animals on each range. Owners in all regions of the state of Goiás were contacted, regardless of the type of ranch (dairy, beef or mixed). On each dairy farm, the history of daily milk production before and after the occurrence of T. vivax was investigated.
Parasitological diagnosis of T. vivax on ranches during the outbreak
The Woo method (Woo, Reference Woo1970) was used to determine the presence of T. vivax in the blood samples. Approximately 4 mL of blood was collected from the caudal vein of each animal in a tube containing an anticoagulant (EDTA). Immediately after collection, the samples were homogenized and the blood was transferred to microhaematocrit tubes. Each tube was placed in a micro-centrifuge. After 5 min of centrifugation (13 000 RCF), the tube was examined under an optical microscope (magnification: 400×) for the study of T. vivax trypomastigotes.
Spatial distribution of T. vivax and risk factors
The state of Goiás is located in the midwestern region of Brazil and has an area of 340 106 km2. To facilitate the interpretation of the spatial distribution of the registered cases, the data were grouped by mesoregion. The division of mesoregions (central, eastern, northern, northwestern and southern) as well as the division of the municipalities in which T. vivax was detected in cattle followed the division established by the Brazilian Institute of Geography and Statistics (IBGE, 2019).
Possible risk factors for cattle to acquire T. vivax in the state of Goiás were determined using a questionnaire administered at each property visited. The questionnaire addressed the following: type of ranch (dairy, beef or mixed); animal category presenting problem (lactating cows, dry cows, heifers, calves, bulls, steers, calves); whether animals had been purchased in the previous 90 days and the place of purchase; whether abortions occurred in this period; whether sick animals presented difficulty moving; whether a reduction in milk production had recently occurred; whether oxytocin was administered to lactating cows and whether the same syringe and needle were shared among different animals; whether haematophagous flies (such as Tabanus, S. calcitrans or H. irritans) were present on the property; and whether artificial insemination was employed on the ranch.
Molecular identification of T. vivax and phylogenetic analysis
Blood samples were sequenced (18S rRNA gene) and the phylogenetic analysis was performed for 20 of the 24 ranches visited on which the parasitological diagnosis confirmed animals positive for T. vivax: Alexânia, Bonfinópolis 01, Bonfinópolis 02, Caldas Novas, Goianápolis, Inhumas, Ipameri 01, Ipameri 02, Ipameri 03, Ipameri 04, Ipameri 05, Ipameri 06, Itauçu 01, Itauçu 02, Morrinhos, Pontalina 01, Pontalina 02, Quirinópolis, Urutaí 01, Urutaí 02 and Urutaí 03. The positive control was Ipameri 01 registered in GenBank (accession code MK392089). The samples from the four remaining ranches (Campo Alegre de Goiás, Cromínia, Mairipotaba and Santa Bárbara de Goiás) that had animals infected with T. vivax were insufficient to perform the analyses.
The detection and molecular characterization of T. vivax were performed as described by Vieira et al. (Reference Vieira, Macedo, Santos, Silva, Mendonça, Faustino, Ramos, Alves, Ramos and Carvalho2017). Genomic DNA was extracted from 200 μL of bovine blood using a commercial kit (Kasvi, Brazil) following the manufacturer's recommendations. Polymerase chain reaction (PCR) was performed using the primers 18STnF2 (5-CAACGATGACACCCATGAATTGGGGA-3) and 18STnR3 (5-TGCGCGACCAATAATTGCAATAC-3), which amplified a fragment of 659 bp of the 18S rRNA gene of T. vivax. The identity of the DNA sequences was determined by comparisons with sequences available in GenBank using BLASTn. A phylogenetic tree was constructed using the unweighted pair group method with arithmetic mean (UPGMA).
Statistical analysis
The data on the total number of animals diagnosed with T. vivax in the mesoregions and municipalities of the state of Goiás, Brazil, were used to calculate the prevalence. Subsequently, prevalence data were arranged in ascending order both for mesoregions and municipalities, setting the odds ratio (OR) equal to 1 for the lowest observed prevalence, then calculating the other ORs in relation to this. The Z test was used to determine the significance, considering a 95% significance level (P ⩽ 0.05).
Regression analysis was employed to determine associations between the prevalence [dichotomized by the median (zero for values below and one for values above the median)] of T. vivax and all the epidemiological variables described above. With these data, a simple binary logistic regression analysis was applied for all epidemiological variables and only those with a P value ⩽ 0.20 in the univariate analysis were selected for the multivariate logistic regression analysis. The strength of the associations between the dependent and independent variables was estimated using ORs derived from the logistic regression estimates. Variables with a P ⩽ 0.05 in the multivariate analysis were considered to be significantly associated with the outcome. A correlation analysis was also performed on the prevalence of T. vivax and epidemiological data collected during the visits. For such, Spearman's correlation coefficients were calculated. All statistical procedures were conducted using Epi Info, version 7.1.5.2.
Results
Prevalence of T. vivax in the state of Goiás during outbreaks
Trypanosoma vivax was found on 24 ranches distributed among 14 municipalities only in Girolando dairy cattle and only in lactating and/or dry cows. Among a total of 4049 blood samples analysed using the Woo method, 358 (8.84%; CI 95% 7.97–9.72) were positive for the acute form of trypanosomosis, as shown in Table 1 and Figs. 1A–C.
Fig. 1. Mesoregions of Goiás State, Brazil (A). Spatial distribution of Trypanosoma vivax in Goiás State, Brazil between May 2015 and May 2017 (B). Distribution of cattle exploitation in Goiás State, Brazil (C). Adapted from Rocha et al. (Reference Rocha, Gonc¸alves, Coelho, Brito, Dias, Delphino, Ferreira, Amaku, Ferreira Neto, Figueiredo, Lo^bo and Brito2009).
Table 1. Association analysis between the mesoregions of the state of Goiás where Trypanosoma vivax was diagnosed in cattle
No positive animals were detected in North and Northwest mesoregions of the State.
a Mesoregions with odds ratio and relative risk >1 (95% CI > 1) are more likely to contain cattle infected with Trypanosoma vivax
Prevalence rates, ORs and RRs by mesoregion in the state of Goiás are displayed in Table 2 and Fig. 1B. No animals positive for T. vivax were found in 12 of the 26 municipalities of origin: Anápolis, Edealina, Buriti Alegre, Corumbaíba, Gameleira de Goiás, Goianésia, Goiatuba, Guapó, Itaberaí, Jataí, Mambaí and Porteirão. On four of the properties where T. vivax was not found (located in the municipalities of Mambaí, Buriti Alegre, Jataí and Quirinópolis), the blood analysis revealed high parasitism by Anaplasma marginale in cows, which was probably the cause of some deaths reported by the owners when answering the questionnaire.
Table 2. Association analysis between the municipalities of the state of Goiás, regarding the prevalence of Trypanosoma vivax diagnosed in cattle
a Municipals with odds ratio and relative risk >1 (95% CI > 1) are more likely to contain cattle infected with Trypanosoma vivax.
Risk factors associated with epidemiological variables
Among the epidemiological variables submitted to logistic regression analysis, the type of animal utilization (beef, dairy or mixed), animal category (lactating cows and others), purchase of animals in the previous 90 days, a reduction in milk production and the use of oxytocin in lactating cows sharing the same syringe and needle were significantly associated (P ⩾ 0.05) with the occurrence of T. vivax in cattle in the state of Goiás, Brazil (Table 3).
Table 3. Association between the prevalence of Trypanosoma vivax diagnosed in the state of Goiás with some epidemiological variables using logistic regression analysis
a Epidemiological variable with odds ratio >1 (95% CI > 1) is more likely to contain cattle infected with Trypanosoma vivax.
Specifically, significant positive correlations (P ⩽ 0.05) were found between the prevalence of T. vivax and animal category (ρ = 0.78), type of cattle (ρ = 0.68), purchase of animals in the previous 90 days (ρ = 0.38), place of purchase of these animals (ρ = 0.87) and use of oxytocin in cows during milking sharing the same syringe and needle (ρ = 0.69). The purchase of animals in the previous 90 days also had a significant positive correlation (P ⩽ 0.05) with the occurrence of abortions (ρ = 0.69), locomotion difficulty of the animals, a sudden reduction in daily milk production (ρ = 0.54) and the place of purchase of these animals (ρ = 0.41). Significant positive correlations (P ⩽ 0.05) were also found between the place of purchase of the animals and the type of cattle (ρ = 0.40) and the use of oxytocin in the cows during milking sharing the same syringe and needle (ρ = 0.38).
The mean daily production of milk before the outbreaks (1595 L ± 1201.9) on the 24 properties where T. vivax was diagnosed was higher (P ⩽ 0.05) than the mean number of litres produced daily (963.0 ± 686.2) on these same properties after the outbreaks caused by the infection of this protozoan. In other words, the occurrence of T. vivax (±90 days) led to a 39.62% reduction in average daily milk production.
Molecular identification of T. vivax and phylogenetic analysis
In the molecular diagnosis (PCR), 100% (20/20) of the samples tested positive for T. vivax. The sequences from the NCBI GenBank used to perform the phylogenetic analyses were Belém (AY363164), Poconé (AY363165), Aquidauana (AY362546), Ipameri (MK392089), Itambé (HM209400), São Miguel Aleixo (KX766453.1), Nigeria (U22316), Mozambique (EU477537), Trypanosoma evansi (AY904050.1), Trypanosoma theileri (KF924256.1), Trypanosoma cruzi (AB301942), Babesia bigemina (FJ426361) and Babesia bovis (AY150059). In the alignment of the DNA sequences, all T. vivax in this study were genetically homologous to T. vivax found in different states of Brazil (Mato Grosso, Mato Grosso do Sul, Pará, Pernambuco and Sergipe) and Nigeria (west Africa), as demonstrated by comparisons to sequences in the NCBI GenBank database. The homology search of the sequenced amplicons revealed 100% identity among the sequences of the 18S rRNA gene. However, we noticed a difference in comparison to T. vivax from Mozambique (east Africa) (Fig. 2).
Fig. 2. Phylogenetic tree based on 18SrRNA genes sequences of Trypanosoma vivax isolates (Brazil, Nigeria and Mozambique), Trypanosoma evansi, Trypanosoma theileri, Trypanosoma cruzi, Babesia bigemina and Babesia bovis. Sequences were compared using the unweighted pair group method with arithmetic mean (UPGMA). The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) is shown next to the branches. The scale bar represents the number of mutations per sequence position.
Discussion
This work reports novel findings regarding epidemiological and genetic aspects of trypanosomosis in Brazil. Significant associations were found between the prevalence of T. vivax and the type of cattle, the purchase of animals and use of oxytocin with the same syringe and needle in cows during milking. These aspects explain the rapid dissemination of this protozoan parasite among dairy cattle in this study.
The prevalence of T. vivax in cattle is causally related to the sensitivity of the diagnostic method employed. In a comparative study of indirect immunofluorescence, conventional PCR (cPCR) and the Woo method, cPCR proved the most sensitive for detecting low parasitaemia (Alves et al., Reference Vieira, Macedo, Santos, Silva, Mendonça, Faustino, Ramos, Alves, Ramos and Carvalho2017; Rabelo et al., Reference Rabelo, Teodoro, Peixoto, Silva and Laviola2017; Bastos et al., Reference Bastos, Faria, Cavalcante, Madrid, Zapa, Nicaretta, Cruvinel, Heller, Couto, Rodrigues, Ferreira, Soares, Cadioli and Lopes2020). However, the Woo method can be used to identify animals during the acute phase of infection in outbreaks of trypanosomosis, which justifies the use of this method in the present investigation, the aim of which was to diagnose acute cases of trypanosomosis in cattle.
Over the past 16 years, T. vivax has spread rapidly across Brazil (Linhares et al., Reference Linhares, Dias Filho, Fernandes and Duarte2006; Batista et al., Reference Batista, Riet-Correa, Teixeira, Madruga, Simões and Maia2007; Carvalho et al., Reference Carvalho, Abrão, Facury Filho, Paes and Ribeiro2008; Guerra et al., Reference Guerra, Júnior, Batista, Santos, Abreu-Silva and Santos2008; Silva et al., Reference Silva, Costa, Polenz, Polenz, Teixeira, Lopes and Monteiro2009; Cadioli et al., Reference Cadioli, Barnabé, Machado, Teixeira, André, Sampaio, Fidélis Junior, Teixeira and Marques2012; Pimentel et al., Reference Pimentel, Ramos, Ramos, de Araújo, Borba, Faustino and Alves2012; Andrade Neto e André, Reference Andrade Neto, Afonso, Mendonça, Souto, André and Machado2015; Costa et al., Reference Costa, Abreu, Machado, Thomé, Massard, Santos and Brito2016; Bastos et al., Reference Bastos, Faria, Madrid, Bessa, Linhares, Fidelis Junior, Sampaio, Cruz, Cruvinel, Nicaretta, Machado, Costa and Lopes2017; Vieira et al., Reference Vieira, Macedo, Santos, Silva, Mendonça, Faustino, Ramos, Alves, Ramos and Carvalho2017; Lopes et al., Reference Lopes, Prado, Martins, Esmeraldo, Beserra, de Sousa Filho, Evangelista and Souza2018). In 11 of the publications cited, outbreaks occurred in Girolando dairy cattle; 10 reported the introduction of new animals to the herd and seven reported the use of oxytocin and/or vaccines as the predisposing factor for the occurrence of trypanosomosis in the respective herds. The sale of these animals occurs collectively at auctions, which directly contributes to the dissemination of this protozoan to other properties. The use of intravenous oxytocin in lactating cows is a common management practice to increase the milking speed and reduce the time spent milking on properties with Girolando dairy cows (Araújo et al., Reference Araújo, Carvalho, Marcondes, do Sacramento and Paulino2012). This practice is performed with the same needle and syringe on several animals, which contributes to the spread of diseases in the herd. Thus, in view of the results obtained in the present investigation and previous studies, we can state that the acquisition of new animals with T. vivax and the administration of exogenous oxytocin to cows during milking using the same syringe and needle are the main causes of the dissemination of trypanosomosis in the state of Goiás and other regions of Brazil.
The 39.62% reduction in average daily milk production found on properties with the presence of T. vivax may be explained by the persistent fever that the acute phase of this disease causes in cattle with high parasitaemia (⩾2 × 106). It is well known that sudden, persistent fever results in anorexia in animals (Cavalcante, Reference Cavalcante2000; Peixoto et al., Reference Peixoto, Cunha, França, Bezerra Junior, Brust, Terra and Armién2011). Moreover, other researchers describe a positive correlation with hyperthermia in cattle during the acute phase of trypanosomosis, which lends support to the inference offered here (Schenk et al., Reference Schenk, Mendonça, Madruga, Kohayagawa and Araújo2001; Desquesnes, Reference Desquesnes2004; Almeida et al., Reference Almeida, Freitas, Tebaldi, Alessi, Machado and Nascimento2010; Dagnachew and Bezie, Reference Dagnachew and Bezie2015).
The presence of tabanid flies, S. calcitrans or H. irritans, was not correlated (P > 0.05) with infection on the 24 properties in which T. vivax outbreaks were diagnosed. However, researchers have reported that animal–animal transmission in other regions of the country could be related to the presence of these mechanical vectors (Otte and Abuabara, Reference Otte and Abuabara1991; Silva et al., Reference Silva, Silva, Schneider, Freitas, Mesquita, Mesquita, Laura, Dávila and Pereira1996; Batista et al., Reference Batista, Rodrigues, Olinda, Silva, Vale, Câmara, Rebouças, Bezerra, García and Teixeira2012; Cadioli et al., Reference Cadioli, Barnabé, Machado, Teixeira, André, Sampaio, Fidélis Junior, Teixeira and Marques2012; Batista et al., Reference Batista, Moura, Lopes, Paiva, Araújo Júnior, Góis, Costa, Coelho and Freitas2018). The epidemiological data demonstrate the importance of restricting the practice of collective sales events (auctions) as well as eliminating the use of exogenous oxytocin in animals during the milking process to avoid serious problems for cattle breeders in Brazil. Therefore, further studies evaluating the dissemination capacity of T. vivax transmission by haematophagous flies and studies evaluating the propagation capacity of iatrogenic pathways of T. vivax are needed.
The phylogenetical analysis proved that the T. vivax found in Goiás, Brazil, was homologous to others from the same country and west Africa, but different from T. vivax from east Africa. These results are in agreement with data in the literature stating that T. vivax in Brazil was originally from Africa (probably west Africa) (Osório et al., 2008; Cortez et al., Reference Cortez, Ventura, Rodrigues, Batista, Paiva, Añez, Machado, Gibson and Teixeira2006; Rodrigues et al., Reference Rodrigues, Neves, Garcia, Viola, Marcili, Da Silva, Sigauque, Batista, Paiva and Teixeira2008; Pimentel et al., Reference Pimentel, Ramos, Ramos, de Araújo, Borba, Faustino and Alves2012; Vieira et al., Reference Vieira, Macedo, Santos, Silva, Mendonça, Faustino, Ramos, Alves, Ramos and Carvalho2017).
Conclusion
The prevalence of trypanosomosis in the state of Goiás, Brazil, during the acute phase of the disease was 8.84%. The main risk factors for the dissemination of T. vivax in the herds were the acquisition of new animals infected with this protozoan and the administration of oxytocin in cows using the same syringe and needle among several animals. In contrast, the presence of tabanid flies, S. calcitrans and H. irritans, was not a risk factor for the occurrence of this haemoparasite. The T. vivax identified in the present study were genetically homologous to each other as well as others found in Brazil and west Africa.
Financial support
This work was supported by Fundo para o Desenvolvimento da Agropecuária em Goiás (FUNDEPEC-GOIÁS) (agreement signed on 5 May 2016) and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – financial code 001.
Ethical standards
This study received approval from the Animal Use Ethics Committee of the Federal University of Goiás, Brazil (certificate number: 032/15) and was conducted in compliance with the ethical principles governing animal experimentation of the Brazilian National Animal Experimentation Control Council (CONCEA).
