Epidemiology of bovine hemoprotozoa parasites in cattle and water buffalo in Vietnam.

A PCR-based survey of hemoprotozoa parasites detected Babesia bigemina, Theileria orientalis and Trypanosoma theileri among cattle and water buffalo in Vietnam, and a new Babesia sp. closely related to Babesia ovata was detected in cattle only. In addition, Theileria annulata and Trypanosoma evansi were not detected in both cattle and water buffalo. Phylogenetic analysis detected T. orientalis MPSP genotypes 3, 5, 7 and N3 in cattle and 5, 7, N1 and N2 in water buffalo. Additionally, water buffalo-derived T. theileri CATL sequences clustered together with a previously reported cattle-derived sequence from Vietnam. This is the first report of a new Babesia sp. in cattle, and T. orientalis MPSP genotype 7 and T. theileri in water buffalo in Vietnam.

Bovine hemoprotozoa parasites, including species of Babesia, Theileria and Trypanosoma, infect cattle populations worldwide, causing significant economic damage to the livestock industry. Among the Babesia parasites infecting cattle, Babesia bovis and Babesia bigemina are virulent species reportedly causing infections in tropical and sub-tropical regions of the world [6]. Whereas, Babesia ovata, a less virulent species of Babesia, is known to be associated with clinical anemia in immunocompromised or Theileria orientalis-infected cattle [9, 31]. Theileria parva and Theileria annulata, lymphoproliferative Theileria parasites, severely compromise the health status of infected cattle [5]. In addition, T. orientalis, a non-lymphoproliferative Theileria species that has a worldwide distribution, occasionally causes severe anemia in infected cattle [28]. Although Trypanosoma congolense, Trypanosoma vivax and Trypanosoma brucei, which are endemic in Africa, are highly pathogenic, Trypanosoma evansi and Trypanosoma theileri are also sometimes reported to be involved in clinical diseases [7, 19, 37]. In general, although most of the bovine hemoprotozoa parasites are known to cause asymptomatic infections in buffalo, control strategies should focus on the elimination of these parasites among buffalo as well, as these animals can act as potential reservoirs [1, 15, 20, 23].

Recent studies conducted in Vietnam demonstrated that cattle populations were infected with B. bovis, B. bigemina, T. orientalis and T. theileri, and that water buffalo were infected with B. bovis, T. orientalis and T. evansi [12,13,14, 16, 30, 36, 39]. Previously, 96 cattle and 43 water buffalo reared in Thua Thien Hue province of Vietnam were analyzed for T. orientalis major piroplasm surface protein (MPSP) genotypes [16]. The MPSP genotypes 1, 3, 5, 7 and N3 were detected in cattle, whereas in water buffalo, genotypes 5, N1 and N2 were detected. By contrast, in Thailand, a country neighboring Vietnam, several other genotypes were detected in water buffalo [1]. Therefore, the possible presence of other MPSP genotypes in Vietnamese water buffalo cannot be ruled out. Extensive studies analyzing the genetic diversity of T. theileri revealed pronounced host specificity of the parasite genotypes that infect different host species, including cattle, water buffalo and deer [8, 10, 11, 24, 25]. By contrast, in a recent investigation in Sri Lanka, some of the T. theileri cathepsin-L like protein gene (CATL) fragments derived from cattle and water buffalo clustered together phylogenetically [38]. However, the host specificity of T. theileri genotypes is still unclear in Vietnam, as the parasite has not yet been detected in Vietnamese water buffalo [30]. In the case of B. ovata, despite being detected in cattle populations in a number of Asian countries, including Japan [21], China [4], Korea [32], Thailand [40] and Mongolia [40], this parasite has not been surveyed in Vietnam. Therefore, in the present study, several species of Babesia, Theileria and Trypanosoma were surveyed in Vietnamese cattle and water buffalo.

Archived blood DNA samples sourced from cattle (n=258) and water buffalo (n=49) reared in Thua Thien Hue province of Vietnam, which had been previously used to detect and genetically characterize B. bovis [39], were screened for B. bigemina, B. ovata, T. annulata, T. orientalis, T. evansi and T. theileri using previously described parasite-specific PCR assays [3, 18, 22, 24, 27, 31]. The PCR reaction mixtures and cycling conditions were the same as those previously reported [29, 30, 40]. Briefly, 10 µl PCR reactions for B. bigemina, B. ovata, T. annulata, T. evansi and T. theileri contained 1 µl of 10× PCR reaction buffer, 200 µM dNTPs (Applied Biosystems, Branchburg, NJ, U.S.A.), 0.5 µM of forward and reverse primers (Table 1), 0.5 units of Taq polymerase (Applied Biosystems), 5.9 µl of double distilled water (DDW) and 1 µl of DNA sample. For T. orientalis, 10 µl reaction mixture contained 5 µl of 2× Ampdirect plus (Shimadzu Biotech., Kyoto, Japan), 0.1 µM of forward and reverse primers (Table 1), 0.1 µl of Extaq DNA polymerase (Takara, Tokyo, Japan), 3.7 µl of DDW and 1 µl of DNA sample. After an enzyme activation step at 95°C for 5 min, PCR reaction mixtures were subjected to 35 (T. orientalis) or 45 (B. bigemina, B. ovata, T. annulata, T. evansi and T. theileri) cycles, each consisting of a denaturing step at 95°C for 30 sec, an annealing step at the appropriate temperature (Table 1) for 1 min and an extension step at 72°C for 1 min. After a final elongation step at 72°C for 7 min, PCR products were analyzed by agarose gel electrophoresis and then visualized under UV light. The findings demonstrated that both cattle and water buffalo were infected with B. bigemina, T. orientalis and T. theileri, as summarized in Table 2. In addition, three DNA samples from cattle tested positive in the PCR assay targeting B. ovata. However, none of the surveyed samples tested positive for T. annulata or T. evansi. Positive rates of the parasite species were analyzed by OpenEpi software (http://www.openepi.com/Proportion/Proportion.htm) and a Chi-squared test (https://www.medcalc.org/calc/comparison_of_proportions.php) to determine the 95% confidence intervals and to calculate the P values, respectively. P values <0.05 were considered to indicate statistically significance. The findings demonstrated that the positive rate of T. orientalis in cattle was significantly higher than that of other parasite species detected in the present study. Similarly, T. theileri-positive rate in cattle was significantly higher than the positive rate of B. bovis and B. bigemina. B. bovis and B. bigemina are usually transmitted by one-host ticks [6], whereas T. orientalis is transmitted by three-host tick species [5]. Therefore, in theory, a tick infected with T. orientalis may transmit the parasite to more number of host animals as compared to B. bovis- or B. bigemina-infected tick. This could be a reason for the higher positive rate of T. orientalis as compared with that of other parasite species. Additionally, the differences in the densities of specific vectors that can transmit different species Babesia, Theileria and Trypanosoma could also explain the difference between the positive rates of these parasite species. On the other hand, although B. bigemina-positive rate was significantly lower than that of B. bovis, T. orientalis and T. theileri in water buffalo, the small sample size may not allow us to make fair comparisons.

Table 1.

Diagnostic PCR primers used in the present study

Parasite	Target gene	Primer sequence (5ʹ–3ʹ)	Amplicon size (bp)	Annealing temp. (°C)	Reference
B. bigemina	Apical membrane antigen - 1	F: TACTGTGACGAGGACGGATC	211	62	[27]
		R: CCTCAAAAGCAGATTCGAGT
B. ovata	Apical membrane antigen - 1	F: GATACGAGGCTGTCGGTAGC	504	56	[31]
		R: AGTATAGGTGAGCATCAGTG
T. annulata	Merozoite-piroplasm surface antigen	F: ATGCTGCAAATGAGGAT	768	52	[18]
		R: GGACTGATGAGAAGACGATGAG
T. orientalis	Major piroplasm surface protein	F: CTTTGCCTAGGATACTTCCT	776	58	[22]
		R: ACGGCAAGTGGTGAGAACT
T. evansi	Minicircle DNA	F: CAACGACAAAGAGTCAGT	373	55	[3]
		R: ACGTGTTTTGTGTATGGT
T. theileri	Cathepsin L-like protein	F: CGTCTCTGGCTCCGGTCAAAC	289	55	[24]
		R: TTAAAGCTTCCACGAGTTCTTGATGATCCAGTA

F, forward primer; R, reverse primer.

Table 2.

The findings of the PCR assays targeting Babesia, Theileria and Trypanosoma species among Vietnamese cattle and water buffalo

Animal typea)	Sample No.	B. bovisb)		B. bigemina		B. ovatac)		T. orientalis		T. theileri
		Positive No. (%)	CId)	Positive No. (%)	CI	Positive No. (%)	CI	Positive No.(%)	CI	Positive No. (%)	CI
Cattle	258	23 (8.9)	6.0–13.0	28 (10.9)	7.6–15.2	3 (1.2)	0.4–3.4	182 (70.5)	64.7–75.8	88 (34.1)	28.6–40.1
Buffalo	49	16 (32.7)	21.2–46.6	2 (4.1)	1.1–13.7	0		22 (44.9)	31.8–58.7	16 (32.7)	21.2–46.6

a) T. annulata and T. evansi were not detected in both cattle and water buffalo. b) B. bovis infection data were obtained from a previous study that analyzed the same DNA samples [39]. c) Although these samples were positive for B. ovata by PCR, sequencing and phylogenetic analyses demonstrated that they were infected with a Babesia sp. closely related to B. ovata. d) 95% confidence interval.

The previous studies that analyzed 96 cattle DNA samples from Thua Thien Hue province in Vietnam found positive rates lower than those determined in the present investigation for B. bigemina, T. orientalis and T. theileri, whereas B. bovis-positive rate was lower in cattle surveyed in the present work [16, 30]. The variations in the distribution of specific transmission vectors in different sampling localities within Hue province and the difference between the sample numbers might explain these discrepancies. Of 208 cattle and 38 water buffalo DNA samples that tested positive for at least one parasite species, 92 and 16, respectively, were infected with multiple parasite species (Table 3). However, the co-infection rates for any two parasite species were not significantly higher than the expected values. Although T. theileri infections are generally considered to be benign, the parasite may be associated with clinical disease when co-infected with other parasite species [19, 37]. As 73 of 88 T. theileri-positive cattle DNA samples were co-infected with Babesia and Theileria parasites, future studies in Vietnam should focus on the clinical significance of T. theileri in co-infected animals.

Table 3.

Multiple infections of Babesia, Theileria and Trypanosoma in the surveyed DNA samples

Combinationa)	Cattle		Water buffalo
	Positive no. (%)b)	CIc)	Positive no. (%)	CI
4 parasites
B. bovis + B. bigemina + T. orientalis + T. theileri	4 (4.3)	1.7–10.7	1 (6.3)	1.1–28.3
3 parasites
B. bovis + B. bigemina + T. orientalis	2 (2.2)	0.6–7.6	0
B. bovis + B. bigemina + T. theileri	2 (2.2)	0.6–7.6	0
B. bovis + T. orientalis + T. theileri	1 (1.1)	0.2–5.9	1 (6.3)	1.1–28.3
B. bigemina + T. orientalis + T. theileri	7 (7.6)	3.7–14.9	0
2 parasites
B. bovis + T. orientalis	7 (7.6)	3.7–14.9	5 (31.3)	14.2–55.6
B. bovis + T. theileri	1 (1.1)	0.2–5.9	3 (18.8)	6.6–43.0
B. bigemina + T. orientalis	8 (8.7)	4.5–16.2	0
B. bigemina + T. theileri	4 (4.3)	1.7–10.7	1 (6.3)	1.1–28.3
Babesia sp. + T. orientalis	2 (2.2)	0.6–7.6	0
Babesia sp. + T. theileri	1 (1.1)	0.2–5.9	0
T. orientalis + T. theileri	53 (57.6)	47.4–67.2	5 (31.3)	14.2–55.6
Total	92		16

a) B. bovis infection data were obtained from a previous study that analyzed the same DNA samples [39]. b) Expressed as a percentage of the total number of co-infected cattle (n=92) or water buffalo (n=16). c) 95% confidence interval.

Five (3 cattle and 2 water buffalo), three, eighteen (10 cattle and 8 water buffalo) and eight (5 cattle and 3 water buffalo) PCR amplicons from B. bigemina-, B. ovata-, T. orientalis- and T. theileri-specific PCR assays, respectively, were cloned, and 2 clones per PCR amplicon were sequenced, as previously described [27]. As the sequences were different between the two clones for 1 B. bigemina apical membrane antigen-1 (AMA-1) (cattle), 12 T. orientalis MPSP (6 cattle and 6 water buffalo) and 2 T. theileri CATL (cattle) gene fragments, a total of 6, 30 and 10 of these gene sequences, respectively, together with 3 AMA-1 gene sequences amplified by the PCR assay targeting B. ovata, were registered in GenBank. Six B. bigemina AMA-1 gene sequences, including four (LC125406–LC125409) from cattle and two (LC125410 and LC125411) from water buffalo, shared high identity scores (98.6–100%) with a B. bigemina AMA-1 gene sequence (AB845438) isolated in Sri Lanka, confirming the findings of the PCR assay. The AMA-1 gene sequences (LC125412, LC125414 and LC125415) amplified by the B. ovata-specific PCR assay shared only 93.5–93.7% identity scores with known B. ovata sequences (AB634843, AB703297, AB703298 and AB733631). In a maximum likelihood phylogenetic tree constructed based on the Kimura 2-parameter model [17] using the MEGA software version 6.06 [35], these AMA-1 gene sequences clustered and formed a sister clade to the B. ovata clade (Fig. 1A). These findings suggested that the AMA-1 gene sequences amplified by the B. ovata-specific PCR could have been derived from a Babesia species that is closely related to B. ovata. To test this hypothesis, a fragment of 18S rRNA was amplified from the DNA samples that tested positive in the B. ovata-specific PCR, using a pair of forward (SSBab18SF1, 5ʹ-CATTACAACAGTTATAGTTTCTTTGG-3ʹ) and reverse (SSBab18SR1, 5ʹ-GTTAAATACGAATGCCCCCAACC-3ʹ) primers. The sequences of the 18S rRNA gene fragments (694 bp) isolated from these DNA samples were identical to each other. These newly determined sequences shared identity scores of 97.3% and 97.1% with known B. ovata (AY603400 and LC125457) and B. bigemina (X59604 and DQ785311) sequences, respectively. On phylogenetic analysis based on the maximum likelihood method and Tamura-Nei model [34], the Vietnamese 18S rRNA sequence formed a sister clade to B. ovata (Fig. 1B), confirming the hypothesis that the AMA-1 gene fragments amplified by the B. ovata-specific PCR were derived from a Babesia sp. that is closely related to B. ovata but has not been described previously.

Fig. 1.

Phylogenetic trees of the AMA-1 and 18S rRNA gene sequences. The AMA-1 (panel A) and 18S rRNA (panel B) gene sequences of Babesia and Theileria parasites, together with the sequences isolated from the Vietnamese DNA samples that tested positive in the B. ovata-specific PCR assay, were used to construct phylogenetic trees. The Vietnamese sequences are highlighted in boldface type letters. Bootstrap values are provided at the beginning of each branch. The scale bars in panels A and B represent 0.5 and 0.1 substitutions per site, respectively. Note that the Vietnamese sequences from Babesia sp. formed sister clades to B. ovata in both of the phylogenetic trees.

The genotypic diversity of T. orientalis was analyzed using 16 (LC125416–LC125431) cattle-derived and 14 (LC125432–LC125445) water buffalo-derived MPSP gene sequences, respectively. A maximum likelihood phylogeny constructed based on the Tamura 3-parameter model [33] placed the cattle-derived MPSP gene sequences within genotypes 3 (n=3), 5 (n=9), 7 (n=2) and N3 (n=2), and the previously reported genotype 1 was not detected [16] (Fig. 2). The MPSP gene sequences of water buffalo origin clustered with genotypes 5 (n=3), 7 (n=2), N1 (n=7) and N2 (n=2). In Vietnamese water buffalo, this is the first report of genotype 7, which has been implicated in several clinical cases of oriental theileriosis among cattle in India [2].

Fig. 2.

Phylogenetic tree of T. orientalis MPSP. The sequences isolated in the present investigation are indicated by boldface type letters. Bootstrap values are provided at the beginning of each branch. The scale bar represents 0.1 substitutions per site. Note that the newly generated cattle-derived sequences were found within MPSP genotypes 3, 5, 7 and N3, while those derived from water buffalo occurred within genotypes 5, 7, N1 and N2.

A neighbor-joining phylogenetic tree [26] was constructed using T. theileri CATL gene sequences, based on the Tamura 3-parameter model [33]. Seven T. theileri CATL gene sequences (LC125446–LC125452) isolated from cattle DNA samples in the present study belonged to five different clades (IB, IE, IK, IL and IIG), two of which (IB and IE) had been previously identified in Vietnam [30] (Fig. 3). Clades IL and IIG were formed by two individual CATL gene sequences determined in the present investigation. The three water buffalo-derived sequences (LC125453–LC125455) clustered with the previously determined cattle-derived sequences isolated in Vietnam, Brazil and Sri Lanka to form clade IIF. In previous investigations, the genotypes of T. theileri have been determined based on several marker genes, including small subunit rRNA (ssrRNA), internal transcribed spacer 1 (ITS1), cytochrome b (Cyt b), spliced leader (SL), glycosomal glyceraldehyde 3-phosphate dehydrogenase (gGAPDH) and CATL, which were found to be host specific in cattle, water buffalo and deer [8, 10, 11, 24, 25]. By contrast, similar to the findings of the present study, some of the CATL genotypes were shared between cattle and water buffalo in Sri Lanka, questioning the host specificity of T. theileri CATL genotypes [38]. Further studies, such as experimental infections and complete sequence analysis from cultured (cloned) parasites, are now needed to clarify the host specificity of these parasites.

Fig. 3.

Phylogenetic tree of T. theileri CATL gene sequences. The sequences determined in the present study are shown in boldface type letters. Bootstrap values are provided at the beginning of each branch. The scale bar represents 0.05 substitutions per site. Note that the cattle-derived sequences from Vietnam belonged to genotypes IB, IE, IK, IL and IIG and that the water buffalo-derived sequences clustered together with the previously reported cattle-derived sequences from Vietnam, Brazil and Sri Lanka to form genotype IIF.

In summary, our survey of bovine hemoprotozoa parasites among Vietnamese cattle and water buffalo is the first to report a new Babesia sp. in cattle, and T. theileri and T. orientalis MPSP genotype 7 in water buffalo in Vietnam. These findings provide valuable insight into the epidemiology of bovine hemoprotozoa parasites infecting livestock in Vietnam.