Dynamics of experimental infection with Leptospira serogroup Pomona in different sheep breed

Most of the information about leptospirosis was obtained from experimental infections with rodents, which in spite of being relevant, do not provide all the answers about the disease in animals and human beings, given the variability of interactions which exist between the agent and the different hosts. Therefore, the objective of the present study was to verify the pathogenesis of the infection in native breeds of sheep of Brazil challenged with a strain of the serogroup Pomona. A total of 10 sheep were divided into two groups with five animals according to the breed. In each group four sheep were challenged by intraperitoneal route with a strain of the serogroup Pomona and one was used as control. Sheep were monitored for 60 days, with collection of blood for serologic diagnosis, as well as vaginal fluid and urine for microbiological and molecular analyses. After this period the animals were submitted to euthanasia and Research, Society and Development, v. 10, n. 2, e19510212373, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v10i2.12373 2 necropsy, with collection of tissues for microbiological, molecular and histopathologic diagnosis. All the challenged sheep presented anti-Leptospira antibodies. Crossbred sheep presented a lower concentration of titers and the antibodies were detected for a shorter period of time when compared to the Santa Inês sheep, with statistical difference in the concentration of the titers on the days 15 (p<0.05), 45 (p<0.05) and 60 (p<0.05) post-infection. There was no significant difference between the groups when comparing the positivity rates of the microscopic agglutination test (MAT) (p>0.05). Eight positive reactions in the urine and vaginal fluid PCR were detected in both groups, being four (50%) in urine and four (50%) in the vaginal fluid, however without statistical difference (p>0.05). In both groups there was a greater proportion of PCR positive samples in kidneys (71.4%) in relation to uterus (28.6%), however without significant difference (p>0.05). There was no significant difference between the groups when comparing the positivity rates of the PCR (p>0.05). The isolation of leptospires from the urine and kidney of a crossbred sheep was possible. Therefore, it is possible that the native sheep, especially the crossbred, may have a relation of resistance with strains of the serogroup Pomona. However the intensity and duration of this relation need to be elucidated and longer-lasting investigations of natural and experimental infections are necessary in order to determine the epidemiological nature of this relation.

necropsy, with collection of tissues for microbiological, molecular and histopathologic diagnosis. All the challenged sheep presented anti-Leptospira antibodies. Crossbred sheep presented a lower concentration of titers and the antibodies were detected for a shorter period of time when compared to the Santa Inês sheep, with statistical difference in the concentration of the titers on the days 15 (p<0.05), 45 (p<0.05) and 60 (p<0.05) post-infection. There was no significant difference between the groups when comparing the positivity rates of the microscopic agglutination test (MAT) (p>0.05). Eight positive reactions in the urine and vaginal fluid PCR were detected in both groups, being four (50%) in urine and four (50%) in the vaginal fluid, however without statistical difference (p>0.05). In both groups there was a greater proportion of PCR positive samples in kidneys (71.4%) in relation to uterus (28.6%), however without significant difference (p>0.05). There was no significant difference between the groups when comparing the positivity rates of the PCR (p>0.05). The isolation of leptospires from the urine and kidney of a crossbred sheep was possible. Therefore, it is possible that the native sheep, especially the crossbred, may have a relation of resistance with strains of the serogroup Pomona. However the intensity and duration of this relation need to be elucidated and longer-lasting investigations of natural and experimental infections are necessary in order to determine the epidemiological nature of this relation. Keywords: Resistance; Leptospirosis; Santa Inês; Crossbred sheep; Pomona.

Introduction
Leptospirosis is an infectious and contagious disease caused by bacteria of the genus Leptospira which has a negative impact on the productive and reproductive rates of the herds, besides being an important zoonosis (Ellis, 2015). A recent systematic review found about one million cases of human leptospirosis per year worldwide, resulting in a mortality rate of approximately 6% (Costa et al., 2015). The transmission of leptospirosis is mainly influenced by environmental conditions, especially rainfall indices (Correia et al., 2017). However, in situations in which the environmental conditions are adverse the animal reservoirs have an expressive importance in the epidemiology of the disease (Picardeau, 2013). In this context, it is assumed that in the semiarid region the interaction between sheep and the etiologic agent happens differently from the other places, due to the uniqueness of the weather conditions and vegetation (Costa et al., 2016).
Some Leptospira serovars are commonly adapted to specific hosts (Pinto et al., 2016), however an in-depth analysis of this adaptation and the relation of adaptability of strains with certain animal hosts may be being neglected. For a long time it was believed that sheep participated only as accidental hosts of leptospirosis and that the infection depended directly of the action of other species, mainly cattle (Vermunt et al., 1994). Notwithstanding, experimental studies and field observations indicate an independence of the participation of other species in the infection , moreover sheep are already cited as an alternative source of maintenance of the serovar Hardjo (Lilenbaum et al., 2009;Arent et al., 2017a).
There are few reports on clinical manifestations in sheep , therefore, it is believed that this species may also has a possible resistance to other strains, not only to the Hardjo. The serogroup Pomona is one of the most recovered in the world and various strains have the swine as a maintenance host (Ellis, 2012). Although this serogroup causes an acute clinical infection and is economically significant in various animal species (Arent et al., 2017b), recent reports indicate an increase in the incidence and endemism of this serogroup in sheep in some areas (Vallée et al., 2017). Furthermore, these animals have been incriminated as being risk factors for other species in infections involving this serogroup (Subharat et al., 2012) Most of the information about leptopirosis was obtained from experimental infections with rodents, which in spite of being relevant do not supply all the answers about the pathogenesis of the disease in animals and human beings, due to the variability of interactions which exist between the agent and the different hosts (Gomes-Solecki, Santecchia, & Werts, 2017). Therefore, knowing that the breeds of sheep have showed to be expressive asymptomatic carriers of the agent , as well as the environmental cycle of the leptospires is influenced by regional factors and by the dynamic nature of the strains/animal species involved (Barragan et al., 2017), the objective of the present study was to verify the pathogenesis of the infection in sheep challenged with the strain L. interrogans serogroup Pomona. Research, Society and Development, v. 10, n. 2, e19510212373, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v10i2.12373

Selection of the animals
The experiment was conducted at the Research Center for the Development of the Semiarid Tropic (Nupeárido) of the UFCG, State of Paraíba, Brazil. For the experimental infection , 10 sheep were used aged between 12 and 18 months, average body weight of 26 Kg, not vaccinated for leptopirosis and which presented negative at serology (MAT titer ≤ 50), bacterial culture and PCR (urine and vaginal fluid) in three previous analysis with intervals of 30 days. The sheep were divided into groups according to the breed, being five crossbred (Group A) and five of the Santa Inês breed (Group B). All animals were accommodated in individual covered stalls (1.0x1.5m) without contact with other animals, distant from the soil at a height of 1m and with access to water and food ad libitum. The animals underwent an adaptation in the installations 20 days before the challenge.

Virulence test and experimental infection
In each of the experimental groups (A and B) four animals were challenged with 10 7 bacteria of the L. interrogans serogroup Pomona serovar Kennewicki (strain Fromm) isolated from pigs in Brazil (Miraglia et al., 2015). The virulence tests followed suggested protocols (Silva et al., 2008;Suepaul et al., 2010) with slight modifications (Barbosa;Martins, & Lilenbaum, 2016). The 3Rs policy for experimental science was applied in all steps (Barbosa;Martins, & Lilenbaum, 2016)

Collection of samples
The blood was collected by puncture of the jugular vein, using a disposable needle and a 8 mL vacuum tube (without anticoagulant) (Vacuum Tube, Vacuette ® , Porto, Portugal), subsequently to the local antisepsis with 2% iodinated alcohol solution. The serum was drained from the blood, transferred to microtubes, and frozen at −20 °C until further processing. With a vaginal speculum, urine samples were collected by means of a no. 8 sterile urethral probe, and were extracted using disposable sterile syringes. For the molecular analysis, aliquots of urine (2 mL) were distributed in microtubes containing 100 µL of PBS 10X. Samples of vaginal fluid were obtained in duplicate, collected directly from the vaginal fornix with the use of Research, Society and Development, v. 10, n. 2, e19510212373, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org /10.33448/rsd-v10i2.12373 5 disposable sterile swabs (Labor Import, Ref: 25507, Osasco, SP, Brazil) and the aid of a vaginal speculum. The second sample of vaginal fluid collected was immersed into a sterile tube with 2 mL de PBS 10X, homogenized and aliquoted in microtubes.
The aliquots were immediately refrigerated and transported within two hours to the laboratory, and were stored at -20°C until the DNA extraction was carried out.
During the necropsy, fragments (2cm 2 ) of kidney, ovary, uterine tube, uterus, bladder, lung, liver and spleen were aseptically collected in triplicate. One sample was processed and sown in proper culture media for the microbiological diagnosis, the second was stored and fixed in 10% formaldehyde buffer, being posteriorly submitted to staining with hematoxylin-eosin (HE) and microscopic evaluation, while the third sample was stored at -20°C until the DNA extraction was carried out. In the tissue samples, the DNA extraction was performed only of the kidney and uterus.

Bacteriological culture
At the time of the collections, some drops of urine and a vaginal fluid swab were sewn in EMJH culture medium in the concentration of 10%, supplemented with a antimicrobial cocktail (Chakraborty et al., 2011). The seeded tubes were kept at room temperature until transport to the laboratory and then incubated at 28ºC in a bacteriological oven. After 24 hours in STAFF medium, the tubes were submitted to serial dilution (10 -1 , 10 -2 , 10 -3 ) in Fletcher semi-solid medium (Difco, BD, Franklin Lakes, NJ, EUA), with the addition of 5-Fluorouracil (1mg/mL-1) and incubated at 28°C, and were examined weekly regarding the presence of Leptospira sp. during the period of 12 weeks. The tissue samples obtained during the necropsy were macerated with the aid of sterile disposable needles and sown in proper culture medium, as previously described. A culture was deemed positive when the sample was suspected at microscopy and confirmed by PCR. Blood & Tissue kit. A PCR targeting the lipL32 gene (referred as specific for pathogenic leptospires) was performed as described (Hamond et al., 2014). Primers LipL32-45F (5'-AAG CAT TAC CGC TTG TGG TG-3 ') and Lip L32-286R (5'-GAA CTC CCA TTT CAG CGA TT-3') were used to amplify the gene LipL32, which is specific for pathogenic leptospiras (Stoddard et al., 2009). The strain L. interrogans serogrupo Pomona sorovar Kennewicki was used as a positive control and ultrapure water as negative control. Additionally, in order to minimize the effect of PCR inhibitors, the samples were neutralized to pH 7.6 with phosphate-buffered saline (PBS), immediately after collection (Lucchesi et al., 2004). Research, Society andDevelopment, v. 10, n. 2, e19510212373, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v10i2.12373 6

Histopathological diagnosis
A part of the fragments of the tissues collected were stored and fixed in 10% buffered formalin, cleaved and routinely processed in the histopathological, going through the stages of dehydration in increasing concentrations of alcohol, diaphanization in xylol, inclusion in paraffin and microtome (cuts in sections of 4-5μm). Subsequently, the slides were stained with hematoxylin-eosin (HE) and analyzed under optical microscopy.

Statistical analysis
The Mann-Whitney U test was used to compare the median of the antibody titers between the groups in each day post-infection. The comparison of the positivity rates of in the serology, PCR and cultivation among the groups was performed by Fischer's exact test. The level of significance used was 5%.

Results
According to Table 1 A1  NT  NT  NT  NT  NT  +  A2  NT  NT  NT  NT  NT  +   A3  NT  NT  NT  NT  NT  -A4  NT  NT  NT  NT  NT  +*   B1  NT  NT  NT  NT  NT  +   B2  NT  NT  NT  NT  NT  -B3  NT  NT  NT  NT  NT  +   B4  NT  NT  NT  NT

Discussion
The results show that there was an immunological response and invasion of the tissues by the agent in the challenged animals; however a discrepancy was identified on the level of infection between the groups. Even though statistical difference was not observed in the positivity rate of the MAT between the groups (p>0.05), a distinct standard in the duration and in the concentration of the antibodies was recognized. All the sheep of the Santa Inês breed were reactive in the MAT until the last collection carried out (D60), while only two of the four crossbred sheep presented positivity up to the D45 p.i., a similar result to other experimental study with sheep which used the same strain by a natural route of infection . Also in sheep, experimental challenges carried out with strains of the serogroup Sejroe revealed variations of detectable titers at 22 days (Rocha et al., 2018) up to 14 weeks p.i. (Sulivan, 1970, whilst that in situations of natural exposure registers of 14 and 20 months were noted for the serogroups Sejroe and Pomona, respectively (Vallée, 2015). The lack of correspondence between the studies with relation to persistence of titers may be due to the virulence of the inoculated strain, infecting dose and cut-off point used; in situations of natural infection (non-controlled) the risk of re-exposure to the agent exists and thus a greater time of detection of antibodies. However, the short period of seroreactivity in crossbred animals is surprising, thereby indicating that the infection in the species may be influenced according to the breed concerned.
Concerning the antibody concentration, the animals of the Santa Inês breed registered the highest titers, with statistical difference on D15, D45 and D60. Other reports have already described high seroconversion of antibodies in pure-bred sheep caused by strains of the serogroup Pomona, either experimentally  or naturally infected (Vermunt et al., 1994). However, the lower concentration of titers in crossbred sheep indicates a possible resistance of these animals against non-Sejroe strains. Apparently serovars of the same serogroup (Pomona, Monjakov and Kennewicki) have distinct interaction patterns which depend on the animal host and the environment (Ellis, 2012;Arent et al., 2017a). Therefore, it may be attributed to the fact that in some regions low frequency of antibodies for the serogroup Pomona are still reported in sheep , whereas in other regions this serogroup is considered to be endemic in the same species (Vallée et al., 2017). In Brazil there is no precise evidence of sheep bearing and transmitting strains of the serogroup Pomona, but due to the rusticity of the species raised in the semiarid region it cannot be ruled out that these animals may be capable, since this capacity was demonstrated in this.
Some authors have already registered problems in sheep implied to strains of the serogroup Pomona, such as miscarriages, stillborn and birth of weak lambs (Ellis et al., 1983), as well as the manifestation of acute disease in lambs presenting pale colored mucous membranes, jaundice, hemoglobinuria, hemoglobinemia and centrilobular necrosis (Vermunt et al., 1994). The sheep of this experiment did not present any clinical manifestation, not even anatomopathological lesions, similar to the reported in recent studies of experimental infections in Brazilian breeds of sheep Rocha et al., 2018), however the short period of the experiment may have been a limiting factor for the presentation of lesions implied to the Research, Society and Development, v. 10, n. 2, e19510212373, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org /10.33448/rsd-v10i2.12373 9 chronic phase of the disease. It is known that, despite being infectious, the serovars have a greater pathogenicity for those hosts which are not resistance (Mansell, & Benschop, 2014), therefore the absence of clinical signs and suggestive may indicate a possible adaptability. It is worth highlighting that even without clinical signs the animals have shown to be healthy carriers of the agent, since the elimination of viable leptospires in the urine of a crossbred sheep was detected (A4). It must be admitted that experimental situations may not accurately portray reality; however, there are reports of isolation of strains of L.
From the epidemiological point of view this may be a serious problem, as sheep may serve as a silent source of infection for other animals and human beings, and this may contribute to the negligence of the disease in the herds. Seropositive sheep for the Pomona strain have already been pointed out as a risk factor for deer which shared the same pasture to also react positively to this serogroup (Subharat et al., 2012).
The number of positive samples of vaginal fluid in the PCR was similar to the quantity of positive samples of urine in both groups; however it is probable that the intermittent elimination of the agent through the urine influenced negatively the evaluation of the elimination by this route Rocha et al., 2017;Rocha et al., 2018). With this in mind, this hypothesis is further reinforced by the fact that the majority of the sheep having been renal bearers of the agent. However, the detection of the agent in the vaginal fluid and in the uterus reinforces the importance of the extra-renal site of infection and the possibility of transmission in sheep, suggested previously (Lilenbaum et al., 2008;Arent et al., 2013;Costa et al., 2018;Rocha et al., 2018;Silva et al., 2018). With relation to the capability of detecting carrier animals, the results demonstrate that only one negative result in the PCR of urine and/or vaginal fluid is inadequate to trustingly consider an animal as negative, seen as the moment of colonization of the genital tract and the period of elimination by the urine are variable. Related to the agent-host interaction, it is believed that the maintenance hosts remain infected and so a balance is established, thus creating a constant reinfection cycle controlled by a peripheral immune response that can last from months to years (Monahan;Callanan, & Nally, 2009). The short period of the experiment did not allow the recording of these dynamics, however it is possible that if exposed the sheep may have a relation of resistance and can eliminate viable strains serogroup Pomona, even if this interaction is executed with less intensity and importance than what happens with porcine species.

Conclusion
It is possible that sheep, especially the crossbred ones, may have a relation of resistance with strains of the serogroup Pomona. However, the intensity and duration of this relation need to be elucidated. Longer-lasting investigations of natural and experimental infections are necessary to determine the epidemiological nature of this relation. Therefore, the monitoring of endemic herds and outbreaks of the disease is essential to better understand its dynamic, as well as an experimental challenge with a strain possibly adapted to sheep, as Autumnalis and Sejroe.

Conflict of interest statement
The authors have no conflicts of interest to disclose.