Ureaplasma diversum (U. diversum) is considered one of the most important members of the family Mycoplasmataceae causing reproductive failure in cattle. U. diversum has also been detected in pneumonic cattle14. In pigs, it was first detected in animals with and without pneumonia1,2 and recently in nasal swabs of live pigs11. However, there is a lack of information on this microorganism with regard to its distribution and participation in the multifactorial porcine respiratory disease complex (PRDC). While Mycoplasma hyopneumoniae (M. hyopneumoniae) is the primary agent of the PRDC, co-infections with other pathogens such as viruses and bacteria are frequent10. Although the concomi-tant presence of M. hyopneumoniae with other Mycoplasma species, including Mycoplasma hyorhinis (M. hyorhinis) and Mycoplasma flocculare has been reported5,7, there has been scant confirmatory evidence of the important role of these Mollicutes in M. hyopneumoniae infections. Furthermore, recent swine lung microbiome studies have demonstrated that members of the family Mycoplasmataceae are one of the most common microbial families in pig lungs with and without pneumonia13. Although other Ureaplasma species, such as Ureaplasma parvum and Ureaplasma urealyticum, have been detected in pig lungs13, there are only limited reports of the presence of U. diversum with other Mollicutes in pneumonic pig lungs1,2. The objectives of this study were: (i) to detect U. diversum in abattoir-collected pig lungs at the individual and herd level and (ii) to determine the concomitant presence of U. diversum and Mycoplasma hyopneumoniae.

A cross-sectional study was conducted in five abattoirs involving eight Argentinian herds (1-8) from three provinces (Table 1). The main characteristics of the herds, such as type of system, size (number of sows) and main aspects of health management are shown in Table 1. Lung samples were collected using systematic random selection procedures where the first unit is randomly selected followed by selection at equal intervals. The herd sample size was set in 20, 40 and 60 lungs for herds with less than 500 sows, between 500 and 1000 sows, and more than 1000 sows, respectively. Each lung was grossly inspected for enzootic pneumonia (EP)-like lesions. Briefly, looking for purple to grey areas of pulmonary consolidation, located in the apical, intermediate, acces-sory and cranial parts of the diaphragmatic lobes10. Then, the type of pneumonia was determined by histopathological examination. For each inspected lung two specimens were collected using the following criteria: if the lung showed any macroscopic lesion compatible with EP-like pneumonia, a BAL sample (for PCR) and a piece of lung (for histopatholog-ical examination, in 10% buffered formalin) were collected for the affected lobe. Each BAL specimen was collected by a cut made in the affected lobe and a sterile 3 ml plastic pipette with sterile saline was introduced into the secondary or tertiary bronchi adjacent to the lesion. In case no gross lesions were detected, the specimens were collected from the right middle lobe. For the histopathological analysis, lung tissues were formalin-fixed, paraffin wax embedded and haematoxylin and eosin (H-E) stained. To determine EP-compatible lesions, each sample was graded from 0 to 4, according to the criteria described by Calsamiglia et al.3. Lesions scoring 1-2 were considered nonspecific pneumo-nias and 3-4 were considered to be compatible with EP. DNA from 280 BAL samples was extracted using the commercial Puri-Prep S kit (Inbio-Highway, Argentina) according to the manufacturer’s instructions. For both, U. diversum and M. hyopneumoniae detection, species-specific nested PCRs were performed under the conditions reported by Cardoso et al.4 and Calsamiglia et al.3, respectively. Considering that Cardoso et al.4 reported a specificity of 73.1% using this M. hyopneumoniae and Ureaplasma sp.-U. diversum were detected in 29.3% (95% CI 24.0-34.6) and 17.1% (95% CI 12.7-21.6) of the analyzed BAL specimens, respectively. The concomitant presence of both microorganisms was detected in 12.5% (95% CI 8.6-16.4) of pig lungs. All herds were positive for at least one of the evaluated specimens; in all herds M. hyopneumoniae was detected alone or in combination with Ureaplasma sp. -U. diversum, while U. diversum was detected in 7 out of the 8 herds. The propor-tion of positive pigs in each herd ranged between 40% and 90%. Table 2 shows the proportions of PCR positives for M. hyopneumoniae alone, Ureaplasma sp.-U. diversum alone or mixed detection and the proportion of macroscopic and microscopic lung lesions. The 16S ribosomal RNA consensus sequence obtained - 173 pb - (GenBank accession number ON059750) showed 100% similarity with the same region of U. diversum strains ATCC 49782 (CP009770.1: GU227397.1), T95 (JN935894.1), ATCC 49783 (GU227398.1) and A417 (NR_025878.1), and Ureaplasma sp. strains USP45 (GU227392.1), USP4 (GU227390.1), USP3012 (GU227389.1) and USP47 (GU227388.1). Among the 280 lungs examined, 125 (44.6%) and 110 (39.3%) had no macroscopic and microscopic lesions of pneumonia respectively. Of the 170 lungs with microscopic lesions, 66 (38.8%; 95% CI 31.2-46.4) were diagnosed as EP-like lesions (scored as 3-4) and 104 (61.2%; 95% CI 53.6-68.8) as nonspecific pneumonias, scored as 1-2 (Table 2). Table 3 shows the proportion of each type of pneumonia diagnosed by histopathological examination for all PCR results from the BAL specimens; from the data, there is sufficient evidence to conclude that the histopathological diagnosis and PCR results are dependent on each other. Among the lungs diagnosed as EP-like lesions, an overlap of the confidence intervals was observed, indicating that there were no differences between the proportion of diagnoses by PCR, while among the lungs diagnosed as nonspecific lesions the proportion of PCR positives for M. hyopneumoniae alone was higher than for Ureaplasma sp.-U. diversum alone or mixed detection.

Table 3: Proportion (percentage and confidence interval 95%) of BAL PCR-positive or negative specimens for Ureaplasma sp.-U. diversum and/or M. hyopneumoniae according to the type of lesion diagnosed by histopathological examination in 280 lungs.

The nested PCR used in this study for U. diversum detection was developed and originally tested in clinical specimens from bovines with a specificity of about 73%4. The obtained sequencing and alignment results showed that their specificity was not enough for the identification of U. diversum, since other close-related Ureaplasma species could be detected. Other specific molecular techniques should be used in future studies for specific U. diversum detection. Despite this fact, there are no antecedents in the literature about the detection of U. diversum and other close-related species in pig lungs at herd level. Given the low sensitivity of culture and the availability of PCR in our laboratory, we did not attempt a culture of Ureaplasma sp. -U. diversum in this study but relied on molecular diagnostics to confirm their presence in the BAL samples. Ureaplasma sp.-U. diversum were detected in BAL fluid from lungs of approximately 17% of finisher pigs of slaughter age, a proportion greater than that previously reported in pigs from Cuba and Canada, with 6.6% and 3.85%, respectively1,2. It is important to note here that the posi-tive proportion of both, Ureaplasma sp.-U. diversum and M. hyopneumoniae, might be overestimated, considering that the scalding tank has been highlighted as a critical point of cross contamination during the slaughtering process8. However, we recently estimated a prevalence of 16.5% of Ureaplasma sp. -U. diversum in nasal swabs specimens from 6-8 and 15-22 weeks-old pigs11. These results might indicate, that a similar proportion of pigs carry Ureaplasma sp.-U. diversum at different sites in the respiratory tract and at different ages. Further studies are needed to under-stand this phenomenon. On the one hand, the evaluation of lung lesions in abattoirs is limited because it only identifies chronic (or recent) lung lesions at the end of the production period and does not provide any information regarding the evolution of respiratory diseases in younger pigs12. On the other hand, a limitation of gross lung lesion evaluation in abattoirs is that other agents may also cause macroscopic lesions similar to those induced by mycoplasmas10,12. Thus, a histological examination is important in lesion characteri-zation. With regard to microscopic lesions, BALT hyperplasia has so far been associated with M. hyopneumoniae and

M. hyorhinis7,12 in spite of the fact that the role of M. hyorhinis in the PRDC is still controversial5,7. In this case, only Ureaplasma sp. -U. diversum were detected in 27.3% of EP-like lesions (scored as 3-4). Furthermore, negative PCR results were obtained in 25.8% of EP-like lesions. Not having searched for M. hyorhinis is a limitation of this study. However, a putative relationship between the presence of Ureaplasma sp. -U. diversum with lung lesions cannot be strictly ruled out, especially considering previous findings in which U. diversum has been detected in lungs with EP-like lesions². Experimental studies should be conducted to confirm if Ureaplasma sp. -U. diversum can produce chronic lung inflammation, resulting in BALT hyperplasia. Our second objective, to determine the concomitant presence of U. diversum and M. hyopneumoniae, resulted in evidence of mixed infection in almost 13% of BAL samples. This is an important finding since there are scarce antecedents about the presence of Ureaplasma sp. -U. diversum together with other Mollicutes2. However, this type of mixed infec-tion would not be rare, mainly due to the abundance of Ureaplasma and other Mollicute species within the pig lungs5,7,13. While pathogenic mechanisms of U. diversum in pigs remain unknown, there are already antecedents of the presence of the agent associated with lung lesions in other species. Human Ureaplasma sp. closely related to U. diversum are capable of causing pneumonia15, and U. diversum has been detected in pneumonic calves14. In pigs with pneumonia, not only U. diversum but also other species of the genus Ureaplasma have been detected1,13. The identification of surface molecules and capsular material involved in the interaction of U. diversum with the host, and virulence factors including the Mycoplasma immunoglobulin binding/Mycoplasma immunoglobulin protease (MIB-MIP) system9, most of them also present in M. hyopneumoniae, suggest that this organism has a pathogenic potential. This descriptive exploratory study provides information that can be used to design experimental and field-based studies to better define the pathogenic role of this organism within the PRDC.

This study represents a comprehensive investigation of Ureaplasma sp. -U. diversum presence in pig lungs at the individual and herd level. The concomitant presence of Ureaplasma sp. -U. diversum with M. hyopneumoniae was demonstrated in BAL, being the first study to investigate such an association on a large field-based scale.

Conflict of interest

None reported.

Acknowledgements

This study was financially supported in part by FONCyT-ANPCyT-MinCyT, República Argentina [PICT 02148/2018]; and SeCyT-UNRC, República Argentina [PPI 2020-2022]. J.S. was holder of an ''Ayudantía de Investigación’’ scholarship during the development of this study (UNRC).

Received 29 March 2022

accepted 17 February 2023