Molecular detection of Theileria sergentii/orientalis/buffeli and Ehrlichia canis from aborted ovine and caprine products in Sardinia, Italy

Abstract The economic impact and losses caused by abortion of small ruminants represent an important threat to livestock industry worldwide. Infectious agents are the most commonly causes of small ruminant abortion and many of which pose a serious threat to human health. The management of abortion outbreaks is essential to understand the transmission, prevention and control of the zoonotic diseases. This study aimed to increase the knowledge about the common known zoonotic pathogens causing abortion (Coxiella burnetii, Chlamydia abortus and Toxoplasma gondii) circulating in Sardinia. In addition, the occurrence of other infectious agents that, until now, had never been identified in abortion samples and which might be cocirculating during the abortion outbreaks were also considered. In this study, 125 abortion samples collected from 91 small ruminant farms were screened for the presence of Babesia/Theileria spp., Ehrlichia canis, Anaplasma spp., Chlamydia spp., C. burnetii and T. gondii by PCR analyses and sequencing. This is the first evidence on the presence of Theileria sergenti/orientalis/buffeli group and Eh. canis in 22 (22/125; 18%) and 26 (26/125; 21%) abortion products from small ruminants, respectively. Chlamydia abortus, C. burnetii and T. gondii were also detected in brain, liver, spleen and placentae at 46% (58/125), 34% (42/125) and 2% (2/125), respectively. This study highlights that pathogens with epizootic and zoonotic potential are circulating in the island and could be involved directly or in association with other pathogens as possible cause of ruminant abortion. Further studies are needed to fully assess the impact of Theileria sergenti/orientalis/buffeli group and Eh. canis on ruminant abortion and their real zoonotic risk in the island.


| INTRODUC TI ON
Abortion in small ruminants is caused by a wide range of infectious agents, and it represents an important cause of economic losses of livestock industries over the world (Benkirane et al., 2015;Kardjadj et al., 2016).
The economic losses are evaluated from production losses, treatment and control costs associated with abortion events. Monitoring the causes of abortion is important for both protecting the healthiness of livestock production and for avoiding the onset of diseases that can be transmissible to humans, especially when the causative agents are zoonotic (van Engelen et al., 2014;Gebretensay et al., 2019). The infectious causes of abortion are not always easily diagnosed, and in most cases, the etiologic agents causing abortion remain unknown (Anderson, 2007). The improper or inadequate specimen collection, handling process and some factors linked to the autolysis process of the organs or environmental contamination, could also prejudge the final diagnosis of abortion in small ruminants. In Sardinia, sheep and goat farms currently account for about one third of value in national ruminant farms, registering more than 3 million of sheep and 0.3 million of goats (National Italian Database 2020 (BDN) June 2020, established by Ministry of Health at the National Surveillance Centre of the IZS in the Abruzzo and Molise Region). Annual losses due to ruminant abortion are estimated around 10 million euros per year. In a previous study, Toxoplasma gondii, Salmonella enterica serotype Abortusovis, Chlamydia abortus and Coxiella burnetii represented the major protozoan and bacterial infectious agents causing abortion in the island (Masala et al., 2007). In the last decade, some studies conducted in Sardinia highlighted the high diffusion of tick-borne zoonotic pathogens in arthropods collected from wild and domestic hosts, including sheep and goats Chisu, Foxi, Mannu, et al., 2018. Many tick-borne diseases cause considerable economic damage to livestock farms as the infection in ruminants can be accompanied by abortion and reducing livestock production (Hurtado & Giraldo-Ríos, 2018;Kivaria, 2006). In Sardinia, the climatic conditions exposed sheep to the risks of tick infestations throughout whole year (Satta et al., 2011). Since some of the pathogens vectored by ticks have never been investigated in abortion products, we performed a molecular survey with the following aims: (1) to identify infectious causes of abortion by testing all samples with the routinary panel of already known microorganisms-caused ovine and caprine abortion in Sardinia, (2) to include in this panel the molecular identity of other infectious agents which might co-circulate during the abortion outbreak and (3) to detect possible co-infections in the same aborted samples.

| Sample collection and molecular detection of pathogens
During 2018, 43 foetuses from 78 sheep flocks and 11 from 13 goat herds were collected by veterinary practitioners and then submitted to the molecular laboratory of the Istituto Zooprofilattico Sperimentale della Sardegna to be tested for detection of infectious agents known as potential causes of abortion and circulating in Mediterranean area . Thirty-two and four placentae from sheep and goats, respectively, were also collected in this study.
Brain, liver and spleen were collected from each foetus wherever possible. However, the condition of some of the foetuses was such that not all tissues could be collected. In fact, dogs or wild animals can frequently eat the placenta and occasionally consume dead foetuses or parts of them. Moreover, factors such as the inadequate transport conditions, autolysis and other causes may adversely affect the collection. Placental samples were homogenated by using the cotyledon and intercotyledon regions (Masala et al., 2007). Foetal organs and placental samples were washed with phosphate buffered saline (PBS) containing 1,000 U/ml of penicillin and 1,000 U/ml of streptomycin sulfate and then digested with 2% trypsin for 3 hr at 37°C. The brain was digested by using trypsin at 0.6%. Digested tissues were then filtered through sterile gauze and centrifuged at 3,000 × g for 10 min. After three washes, pellets were resuspended in Foetal Bovine Serum (FBS) containing 10% dimethylsulphoxide, aliquoted, and stored at −20°C until use (Masala et al., 2007). As part of the routinary diagnostic methods available for the abortion diagnosis of the Istituto Zooprofilattico Sperimentale della Sardegna, all collected samples were tested for the presence of bacterial (Anaplasma spp., Chlamydia spp., Coxiella burnetii, Ehrlichia canis,) and protozoal (Toxoplasma gondii, and Babesia/Theileria spp.) agents by PCR analyses as previously described Masala et al., 2007).

| Purification and sequencing of positive samples
To identify detected pathogens, positive PCR products were purified and sequencing was performed as described in previous studies Masala et al., 2007). All sequences obtained were assembled and edited with Chromas pro software version 2.6. and were then analysed by BLAST (www.ncbi.nlm.nih.gov/blast/ Blast. cgi) sequencing analysis in the GenBank database.

| Statistical analysis
Based on positive/negative samples, contingency 2 × 2 tables were performed for each pathogen and Chi-square test was used to assess significance. To quantify the degree of departure of the number of mixed infections from independence, index of coinfection (Ic) was used (Ginsberg, 2008

| RE SULTS
A total of 125 abortion products were collected from 91 different small ruminant farms located in several sites from Sardinia island.
Overall, 36 placentae (32 from sheep and 4 from goats), 51 brains (43 from sheep and 8 from goats), 26 liver samples (15 from sheep and 11 from goats) and 12 spleen samples from sheep were collected as illustrated in Table 1. Over 125 total samples, 51 (56%) were infected with at least one of the following pathogens: Babesia/Theileria spp., T. gondii, E. canis, C. burnetii, and Chlamydia spp., whilst 40/125 (44%) samples were co-infected with two or more pathogens as illustrate in Table 1.

Sequencing of the 18S rRNA performed on positive amplicons
showed that all sequences detected from both sheep and goats abortion samples were 100% identical to that of Theileria sergenti/ orientalis/buffeli strains, isolated worldwide. Theileria group DNA was mainly found from sheep brain (5/43; 12%) and liver (3/15; 20%) samples and from goat liver tissues (9/11; 82%) as shown in Table 2.
The presence of DNA from T. gondii was detected in two placentae (2/125; 1.6%) from sheep by nested PCR assays targeting 227 bp of the multicopy 18S-5.8S rRNA internal transcribed spacer (ITS1) region.
The remaining 29 samples resulted unreadable probably suggesting multiple Chlamydiales agents are present in aborted samples.
Following BLAST analyses used to determine the likely sequence identity, all the sequences detected were found to share 100% identity to the partial 16S rRNA gene sequences from Chlamydia abortus strains.
Specifically, seven, three and one combination of co-infections with two, three and four different pathogens were detected in sheep abortion samples, respectively. Instead, three, two and one combination of co-infections with two, three and four different pathogens, were detected in goat abortion samples, respectively.
In sheep, the most frequent but not statistically significant coinfection was those of C. abortus and C. burnetii, detected in 13 samples: 7 brains, 1 liver and 5 placentas. In goats the co-infection of Theileria complex and C. abortus has been detected in 1 brain, 9 livers and 2 placentas. Positive and statistically significant index of coinfection (Ic) was detected in double infection by E. canis/C. burnetii (5.581; p = 0.007) in sheep brain samples, Theileria complex/C. abortus (9.713; p = 0.005) in brain, liver and spleen samples and Theileria complex/C. burnetii (9.391; p = 0.012) in brain, liver and spleen from sheep. No one statistically significant Ic has been detected in samples from goats (Table 3).

| D ISCUSS I ON AND CON CLUS I ON
In Sardinia, ovine and caprine livestock industries represent the top in the agricultural sector. Although abortion in small ruminants has considerable economic impact on the productivity of herds in Sardinia, the implementation of analyses to ascertain the putative pathogens-causing abortion needs to be improved. T. sergenti/orientalis/buffeli group was also detected in asymptomatic and symptomatic calves and goats from Sardinia (Zobba et al., 2020).
In addition, the transplacental transmission of T. orientalis in cattle has been proposed (Baek et al., 2003;Swilks et al., 2017), and evidence of the complex in colostrum samples suggests also a transmammary transmission (Hammer et al., 2016). Further research is required to determine if T. sergenti/orientalis/buffeli group could be considered the causal agent of ovine/caprine abortion or a collateral cause in co-presence with other microorganisms.
As far as we know, this study reported the first molecular detection of E. canis from placenta and foetal organs from both sheep and goats flocks from Sardinia. Although Eh. canis is known as the pathogen that commonly infects dog worldwide, infections have also been described in sika deer (Li et al., 2016), goats (Zhang et al., 2017), humans (Maeda et al., 1987;Perez et al., 2006), cats (Braga et al., 2014) and arthropods (Sainz et al., 2015) and there is thus growing evidence that Eh. canis has a wider host range than previously thought.
Ehrlichia canis is mainly vectored by R. sanguineus s.l., which is largely present in the Mediterranean basin where E. canis is endemic (Sainz et al., 2015). In previous studies, the presence of Eh. canis was highlighted in Rh. sanguineus s.l., Hy. marginatum, Rh. bursa and Hy. marginatum ticks collected from sheep and goats (Chisu, Foxi, Mannu, et al., 2018). Although Eh. canis pathogenicity in domestic ruminants is still unknown, detection of this microorganism in abortion products warrants further investigations.
The results of this study indicate that Ch. abortus DNA was the most frequently detected bacterial pathogen in abortion products.
The intracellular bacteria Ch. abortus, the causative agent of enzootic abortion of ewes (EAE), affects the placenta of sheep and goats leading to abortion in late gestation (Essig & Longbottom, 2015). This pathogen is mainly of veterinary importance being involved in abortion of ruminants as well as other domestic and wild mammals, reptiles and amphibians (Opota et al., 2015). C. abortus represents also a zoonotic risk to humans since it can colonise the human placenta and lead to foetal death and miscarriage (Essig & Longbottom, 2015;Opota et al., 2015). The last report of the pathogen in the island dates back to 2007 when it was detected in 12.5% of placentae from caprine aborted samples and from 2.4% of foetuses and 14.4% of placentae of ovine aborted samples (Masala et al., 2007) and Dermacentor tick genera from Sardinia .
Ruminants are hosts to a wide range of tick species that vectors members of the order Chlamydiales, including C. abortus. Since C.
abortus strains detected in aborted samples are the same detected in ticks, their involvement as routes of transmission of C. abortus is yet to be fully defined. However, considering the significant economic losses in farms due to this pathogen, the early and accurate diagnosis of C. abortus is necessary, so that appropriate control measures can be adopted to limit or prevent the spread of infection.

Coxiella burnetii is another important infectious agent of
Sardinian livestock, causing significant economic losses to the livestock industry (Masala et al., 2004). Domestic mammals, mainly goats, sheep, and cattle, are the most important reservoir of infection of this bacterium. Transmission from animal reservoirs to humans occurs primarily through the inhalation of contaminated aerosols (Schneeberger et al., 2014). In this study, the presence of C. burnetii was also confirmed in abortion products. C. burnetii was mainly detected in placentae (18/32; 56%) and in brain samples (10/43; 23%) from sheep and in liver organs from goats (5/11; 45%) with a prevalence rate of infection that was higher than any previously reported in the island (Masala et al., 2007). Moreover, recent studies identified C. burnetii in R. sanguineus s.l., R. annulatus and H. marginatum ticks from Sardinia (Chisu, Foxi, Mannu, et al., 2018;Chisu et al., 2020) confirming that it is likely that arthropods could act as vectors of the pathogenic C. burnetii in Sardinian ruminants.
Toxoplasmosis, a disease caused by Toxoplasma gondii, is an important zoonosis and a major cause of abortion in sheep (Zedda et al., 2010). In this study, T. gondii has been detected in very low numbers and only two cases have been reported in ovine placen- Pathogens causing abortions can coexist and cause concurrent infections with more than one pathogen and can increase the risk of atypical forms of clinical disease. These results highlight that two or more pathogens were simultaneously present in placenta, brains, liver and spleen indicating that ovine and caprine farms are exposed to multiple pathogens. Since one or more pathogens can coexist in the same tick, we can suppose that the arthropods could vector and transmit more pathogens simultaneously to the host . Co-infecting pathogens might alter the efficiency of transmission, cause cooperative or competitive pathogen interactions and alter disease severity amongst hosts (Swanson et al., 2006). So the direct responsibility of the single pathogens in abortion could not be definitively established without the confirmation of the presence of the lesions caused by the different infectious agents by macroscopic, microscopic and histological examination of the abortion products. These information will be of interest to laboratory diagnosticians that should consider the possibility that other abortifaciens agents are circulating in the island and could be the cause of ovine and caprine abortion.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interests.

E TH I C S A PPROVA L
The authors declare that all applicable international, national and/or institutional guidelines for the care and use of animal were followed.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1002/vms3.510.

DATA AVA I L A B I L I T Y S TAT E M E N T
Not applicable.