Survey of vector-borne and nematode parasites involved in the etiology of anemic syndrome in sheep from Southern Brazil

2020. *Corresponding author: Rafael Felipe da Costa Vieira. E-mail: rvieira@ufpr.br Abstract Although anemia has been historically linked to Haemonchus contortus infection, other infectious agents, such as hemotropic mycoplasmas and tick-borne disease pathogens, may also lead to anemic crisis in sheep. This study has aimed to investigate infections related to anemia in a sheep herd from Bandeirantes City, Paraná State, southern Brazil. Seven out of forty-two (16.6%; 95% CI: 8.32–30.6%) sheep were positive for hemoplasmas by a PCR targeting the 16S rRNA gene and all tested negative for A. marginale/A. ovis and Babesia / Theileria spp. by PCR based on msp 4 and 18S rRNA genes, respectively. Two (4.7%; 95% CI: 1.32–15.79%) animals were infested with Rhipicephalus microplus ticks. Fecal egg counting was performed in 38 sheep and 24 (63.15%; 95% CI: 47.2–76.6%) presented > 500 eggs per gram. Phylogenetic analysis of partial sequences of the detected hemotropic Mycoplasma sp. 16S and 23S rRNA genes confirmed that the animals were infected with Mycoplasma ovis . Polymorphism analysis of partial 16S rRNA sequences showed three different genotypes of M. ovis infecting sheep assessed in the present study. Mycoplasma ovis and gastrointestinal nematodes M. ovis In the present study, two M. ovis 16S rRNA gene (MN173878 MN173879) M. ovis isolated from Japanese serows ( Capricornis crispus ) from (AB571119), MN173880 sequence clustered a M. ovis sequence from Turkish sheep (MF377458). analysis for polymorphisms haplotypes of M. ovis


Introduction
Brazilian sheep herd has been estimated to comprise more than 18 million animals. The southern region of Brazil is home to 30% of the herd, which are mainly reared for meat production (IBGE, 2016). Anemia is considered a concerning condition for animal health leading to production losses. Gastrointestinal (GI) parasites and vector-borne pathogens (VBP) have been considered an important cause of anemia in sheep worldwide.
The GI parasite Haemonchus contortus, which parasitizes the sheep's abomasum and causes acute hemorrhagic anemia, may lead to sudden death and important losses in sheep industry (Lane et al., 2015;Taylor et al., 2007). Although anemia is commonly linked to H. contortus infection in sheep, it is noteworthy that other infectious agents, such as hemotropic mycoplasmas (hemoplasmas) and VBP may also lead to anemic crisis (Yeruham et al., 1998;Neimark et al., 2004;Hornok et al., 2009;Alessandra & Santo, 2012). Nevertheless, VBP have not been historically included in the differential diagnosis of anemia by veterinary practitioners in Brazil.
Anaplasma marginale, the causative agent of bovine anaplasmosis, has been detected in goats from the northeastern region of Brazil (Da Silva et al., 2018) and sheep from Iran (Yousefi et al., 2017). However, the epidemiological and clinical effects of A. marginale infection in small ruminants remain to be fully established. On the other hand, Anaplasma ovis has been reported infecting sheep and goats from different countries, such as Italy (De La Fuente et al., 2005), Hungary (Hornok et al., 2007), Iran (Jalali et al., 2013), China (Zhang et al., 2016;Yang et al., 2018), and Turkey (Aktas & Ozubek, 2018). Although A. ovis infection often causes mild pathogenicity in sheep (Hornok et al., 2007;Renneker et al., 2013), its occurrence is highly involved in concurrent infections with different hemoparasite species in sheep (Jalali et al., 2013;Yang et al., 2015;Sevinc et al., 2018;Ringo et al., 2018), which may contribute to aggravating animal clinical conditions (Aktas & Ozubek, 2018). To date A. ovis has never been reported infecting small ruminants in Brazil. Despite the high host specificity of Rhipicephalus microplus to cattle (Dantas-Torres et al., 2009;Ma et al., 2016), this tick species may be found parasitizing small ruminants (Brito et al., 2005;Da Silva et al., 2018).
A wide range of Piroplasmorida species, mainly from the genera Theileria and Babesia, may infect sheep. Although piroplasmid infections in small ruminants may be considered negligible when compared to cattle infections, the number of studies regarding these tick-borne agents has raised with the increasing economic interest in these animals (Schnittger et al., 2003;. These infections may cause fever, anorexia, mucosal pallor, hemoglobinuria, and anemia in sheep (Yeruham et al., 1998;Hassan et al., 2015). Rhipicephalus, Hyalomma, and Haemaphysalis ticks have been incriminated as putative vectors for these agents (Morzaria, 1998;Tian et al., 2004;Uilenberg, 2006). In Brazil, studies on piroplasmid infection in sheep have not been reported so far.
Co-infection with M. ovis and other pathogens, such as GI nematodes and tick-borne pathogens, may lead to hemolytic anemia and reflect on production decay and mortality (Hornok et al., 2009;Abdullah et al., 2013). Although ovine production has a notable importance, only one study on M. ovis has been reported in sheep from Brazil (Souza et al., 2019), whereas the occurrence of A. marginale and piroplasmids in sheep from this country remains unknown. Accordingly, the present study aimed to investigate the role of VBP and GI nematodes in the occurrence of anemia in a sheep herd from southern Brazil.

Ethical approval
This study was approved by the Ethics Committee for Animal Experimentation and Animal Welfare of Universidade Federal do Paraná (protocol 030/2019) and conducted according to the ethical principles of animal experimentation, adopted by the Brazilian College of Animal Experimentation.
A total of 42 female sheep from Bandeirantes municipality (23°06′28′′S 50°21′36′W), Paraná State, southern Brazil, were evaluated for the presence of hemoplasmas, tick-borne pathogens (Anaplasma spp., Babesia spp., and Theileria spp.), and GI parasite infection. The animals were co-grazed with cattle in a paddock, where tick infestation is common during the entire year.

Sampling
Blood samples (up to 5 mL) were collected from the sheep by venipuncture of the jugular vein in commercial tubes containing EDTA (BD Vacutainer  , Franklin Lakes, NJ, USA). Fecal samples were obtained directly by rectal collection, identified, and stored in isothermal recipients until analysis.
Ticks found on animals were directly removed using a commercial hook (O'TOM Tick Twister  , Lavancia, France), and kept in labeled absolute ethanol-containing tubes until identification according to morphological taxonomic keys (Barros-Battesti et al., 2006).

Evaluation of packed cell volume
The packed cell volume (PCV) was measured by centrifugation (10,000 rpm for five minutes). A PCV value of < 0.27 L/L was used as an indicator of anemia (Weiss et al., 2010). Thereafter, aliquots of blood were stored at -20 °C until molecular testing.

DNA extraction
Isolation of genomic DNA from sheep blood samples was performed using a commercial kit (llustra TM Blood GenomicPrep Mini Spin Kit, GE Healthcare, Little Chalfont, UK). Nuclease-free water was used as negative control to monitor cross-contamination.

Polymerase Chain Reactions (PCR)
A conventional PCR for the mammalian endogenous gene glyceraldehyde-3-phosphate dehydrogenase (gapdh) was performed for all samples to monitor the DNA extraction, as previously described (Birkenheuer et al., 2003).
Thereafter, DNA samples were tested using a conventional pan-hemoplasma PCR assay targeting a fragment (900 bp) of the 16S rRNA gene of hemoplasmas (Hoelzle et al., 2011;Machado et al., 2017). This assay has been validated for the diagnostic of hemoplasmas and is able to detect 4.32 DNA copies/μL (Machado et al., 2017).
The primer set targeting a fragment of the 23S rRNA gene of hemoplasmas was designed using Primer3 software (Koressaar & Remm, 2007;Untergasser et al., 2012) and commercially synthesized (Integrated DNA Technologies, Coralville, IA, USA). Briefly, the 23S rRNA gene sequences of hemotropic Mycoplasma spp. (NR_076982, NR_103993, AB740012, HE613254, NR_121969, NR_103970, NR_076983) available in the GenBank  database were retrieved and aligned using Bioedit v. 7.0.5.3 software (Hall, 1999). Potential target sites for forward primers were manually identified; suitable reverse primers and PCR products were selected using Primer3 software (Koressaar & Remm, 2007;Untergasser et al., 2012). Based on the desired product size (800-1,000 bp), melting temperature, minimal pair complementarity, and minimal pair 3′-complementarity (to avoid primer-dimer or hairpin formation), one reverse primer was selected. To verify the proprieties of each primer, PCR suitability tests were performed using SMS software (Sequence Manipulation Suite, Edmonton, Alberta, Canada). The specificity of the prospective primers was checked in silico by BLASTn analysis (Altschul et al., 1999) in order to determine the identity with the sequences deposited in the GenBank  database. The primer set used to amplify an 800-bp fragment was 23S_HAEMO_F (5′-TGA GGG AAA GAG CCC AGA C-3′) and 23S_HAEMO_R (5′-GGA CAG AAT TTA CCT GAC AAG G-3′).
For the standardization of 23S rRNA PCR protocol, a PCR mixture containing 1X PCR buffer, 1.5 mM of MgCl 2 , 0.2 mM of each dNTP, 0.4 mM of each primer, 2.5 U of Taq DNA Polymerase (Taq  PCR Master Mix Kit, QIAGEN, Hilden, Germany), 5 μL of DNA template, made to a final volume of 25 μL with water. The cycling conditions consisted of denaturation at 94 °C for 3 min, followed by 35 cycles of denaturation at 94 °C for 30 s, annealing over a gradient of temperature (53, 54, 55, 56, and 57 °C) for 30 s, extension at 72 °C for 60 s, final extension at 72 °C for 10 min, and cooling at 4 °C. Additionally, a second gradient of annealing temperature was tested (52, 52.5, 53, 53.5, 54, and 54.5 ºC), and the annealing temperature was determined to be 54 ºC. The efficacy of a gradient of primer concentration was also determined (0.4 and 0.2 mM of each primer), and the concentration of 0.2 mM of each primer showed a band with higher intensity. Mycoplasma haemofelis DNA obtained from a naturally infected cat (Marcondes et al., 2018) and nuclease-free water were used as positive and negative controls, respectively. The amplified PCR products were subjected to gel electrophoresis on 1.5% agarose gels for 1 h at 100 V, and the gels were stained with SYBR (0.1 μL/mL gel) (SYBR™ Safe, Thermo Scientific, Waltham, MA, USA), and visualized under a 312 nm UV light transilluminator (LTB HE, Loccus do Brasil, São Paulo, BR).
Additionally, sheep DNA blood samples were also tested by PCR assays targeting a fragment (870 bp) of msp4 gene of A. ovis/A. marginale , and a fragment (551 bp) of the 18S rRNA gene of Theileria spp./Babesia spp. (Almeida et al., 2012). Anaplasma marginale and Babesia vogeli DNA obtained from naturally infected cattle (de Souza Ramos et al., 2019) and dogs (Mongruel et al., 2018), respectively, were used as positive controls. Nuclease-free water was used as negative control.

Sequencing
Amplicons obtained from three samples that were positive for hemoplasmas were extracted and purified from the gel by enzymatic purification (ExoSAP-IT™ PCR Product Cleanup Reagent, Thermo Scientific, Waltham, USA), evaluated by spectrophotometry for concentration and purity (NanoDrop™ One Spectrophotometer, Thermo Scientific, Waltham, USA), and sequenced from both directions by the Sanger's method (3500 Genetic Analyzer, Applied Biosystems, Foster City, CA, USA). Thereafter, the sequences were subjected to BLASTn analysis (Altschul et al., 1999) for determining the identity with sequences previously deposited in the GenBank  database. The nucleotide sequences of the 16S rRNA and 23S rRNA genes of M. ovis amplified in this study were submitted to the GenBank  database (GenBank  accession nos. MN173878-MN173880 and MN169108-MN169110, respectively).

Phylogenetic analysis
The partial sequences of 16S and 23S rRNA genes were subjected to multiple alignment with sequences selected from GenBank  using MAFFT available on the GUIDANCE 2 server (Sela et al., 2015) for each gene. The best-fit model of nucleotide substitution was determined using jModeltest v.2.1.10 (Darriba et al., 2012) and was set as HKY+G based on the Akaike Information Criterion (AIC). Each Bayesian reconstruction was performed in Beast 1.8.0 (Li & Drummond, 2012) with three independent runs of 30 million MCMC steps sampled at every 5,000 trees, 10% of the burn-in. The phylogenetic tree was visualized with FigTree software version 1.4.3 (Rambaut, 2014) and the final layout was done with Inkscape version 0.92.3 (Albert et al., 2018). The 16S rRNA tree was rooted with Mycoplasma bovis and Bacillus subtilis whereas the 23S rRNA tree was rooted with Mycoplasma pneumoniae and Bacillus subtilis.

Genotype analysis of hemoplasmas
Mycoplasma ovis 16S rRNA sequences were analyzed to determine the number and diversity of found genotypes, using the DnaSP software version 5.10.1 (Librado & Rozas, 2009).

Statistical analysis
A non-parametric Mann-Whitney test or a parametric unpaired Student's t test was used to compare the mean PCV values between M. ovis-positive and negative sheep and EPG-positive and negative sheep. The Chi-square or Fisher's exact test was used to determine associations between anemia and presence of ticks with positivity to M. ovis and between anemia and EPG positive results. Odds ratio (OR), 95% confidence interval and p values were calculated for each variable. Results considered significantly different when p < 0.05. Data were compiled and analyzed in Epi Info ™ software (version 7.1.5, CDC).

Results
The mean PCV for sheep was 0.30 L/L. Nine out of 42 (21.42%; 95% CI: 11.71-35.94%) animals were anemic. A total of 38 sheep were evaluated by EPG. Four out of 42 animals did not present feces at the time of sampling and were not evaluated. Twenty-four (63.15%; 95% CI: 47.2-76.6%) animals presented EPG values > 500 and were considered positive for Strongylida-type eggs. A total of 24 adult tick specimens were collected from 2 out of 42 (4.76%; 95% CI: 1.32-1.57%) animals and all were identified as R. microplus.

Discussion
Worldwide, anemia in small ruminants has been generally linked to GI nematode infections (Kaplan et al., 2004;Adogwa et al., 2005;Di Loria et al., 2009). Likewise, veterinary practitioners in Brazil have not historically included VBP in the differential diagnosis of anemia. Anemia may be more severe when the animal is coinfected with nematodes and M. ovis. Although a previous study reported an outbreak of M. ovis infection that caused death of sheep from Argentina (Aguirre et al., 2009), the anemic status related to M. ovis infection is generally less severe if sheep are kept under good nutritional conditions and do not have a severe worm burden (Messick, 2004). In the present study, one out of nine anemic sheep (PCV = 0.17 L/L) was infected with M. ovis and also presented a high EPG value (4,550). The Strongylida suborder includes important parasites for ovine livestock, such as H. contortus, that may   lead animals to severe anemic conditions (Amarante, 2014), and thus, may explain the anemic status of the animals studied herein. The remainder eight anemic sheep also presented high values of EPG, ranging from 800 to 5,200.
The present study aimed to investigate infections related to anemia in a sheep herd from southern Brazil. Herein, M. ovis was detected in 16.6% of the sampled sheep. Previous studies on the detection of M. ovis in Brazil have reported prevalence rates of 39.30% in goats (Machado et al., 2017), 40-87% in deer (Grazziotin et al., 2011a, b;André et al., 2020) and 78.8% in sheep (Souza et al., 2019). Although Souza et al. (2019) detected a higher percentage of positivity to M. ovis in sheep from the state of Rio Grande do Sul, divergences on climate conditions, herd management, and employed PCR assays used may explain differences on hemoplasma prevalence rates between studies. Additionally, association between anemia (p = 0.5278) or the presence of ticks (p = 0.3089) and M. ovis infection were not found, similar to that reported in previous studies involving sheep (Rjeibi et al., 2015;Souza et al., 2019) and goats (Machado et al., 2017). Moreover, associations between anemia and EPG values > 500 were not found as well.
It is important to state that there are some limitations on the present study. A convenience sampling was performed in order to investigate infections related to anemia in the studied sheep herd. Although all animals from the herd have been sampled, a low number of samples were provided. Furthermore, conventional PCR assays may present a low sensitivity when compared to quantitative PCR and may leads to false negative results (Willi et al., 2007). Thus, the prevalence found herein may be higher.
Phylogenetic analysis of the 16S and 23S rRNA gene sequences from sheep that were positive for hemoplasmas confirms that animals were infected with M. ovis. In the present study, while two M. ovis 16S rRNA gene sequences (MN173878 and MN173879) clustered together with M. ovis isolated from Japanese serows (Capricornis crispus) from Japan (AB571119), the MN173880 sequence clustered together with a M. ovis sequence from Turkish sheep (MF377458). The analysis for polymorphisms showed that three different haplotypes of M. ovis were infecting sheep in the studied herd. The presence of polymorphisms may cause incongruities in the post-test probability values, which may explain the low values obtained. Additionally, phylogenetic analysis of the M. ovis 16S rRNA gene showed marked differences from sequences isolated from the same herd. Genotype diversity of M. ovis has been previously reported in flocks from China (Wang et al., 2017) and Mexico (Martínez-Hernández et al., 2019) and, more recently, among deer from Brazil (André et al., 2020). Interestingly, the human-associated M. ovis genotypes were more related to genotypes detected in sheep and goats when compared to those found in deer (André et al., 2020). Therefore, future studies aiming to investigate the occurrence of M. ovis infection in sheep herd workers are needed.
Even though the 16S rRNA gene has been widely used for phylogenetic analysis, intra-genomic heterogeneities are considered a limiting factor (Rajendhran & Gunasekaran, 2011). In 2014, the genome sequencing of M. ovis strain Michigan was reported, showing the presence of two copies of the 16S rRNA gene (Deshuillers et al., 2014). The primers used in the present study cannot specifically target one or another copy from the M. ovis-16S rRNA, which may represent unreliable data in sequence analysis.
Regarding the partial 23S rRNA amplification from hemoplasmas, the three sequences amplified in the present study showed 100% identity with M. ovis 23S rDNA gene sequence isolated from the USA and deposited in GenBank  (accession no. NR121969). The sequences amplified in the present study clustered together with those of M. ovis isolates from sheep from the USA and also formed a large clade with M. ovis genotypes detected in deer from Brazil. There are few studies in which the phylogenetic relationship of M. ovis has been determined on the basis of 23S rRNA gene assembly. The analysis presented in this study shows similar results with that performed by Grazziotin et al. (2011b), wherein the M. ovis deer-related sequences clustered separately from those of the M. haemofelis group. Low post-test probability values were also obtained in the 23S rRNA phylogeny of our sequences, which may be related to the genotype diversity among them.
In the present study, all sheep tested negative for A. ovis/A. marginale and Babesia/Theileria spp. by PCR. Although there are no reports on the detection of A. ovis and A. marginale or piroplasm infection in sheep from Brazil, A. marginale has been detected in goats co-grazed with cattle in the northeastern region of the country (Da Silva et al., 2018). Anaplasma marginale is a common tick-borne pathogen that primarily infects cattle from tropical and subtropical areas (Kocan et al., 2010), being transmitted by R. microplus ticks in Brazil (Kessler, 2001).

Conclusion
Mycoplasma ovis and GI nematodes occurs in sheep from the northern region of Paraná State and were not related with anemia in the present study. Different genotypes of M. ovis occur in sheep from the northern region of Paraná State.