Investigation of antimicrobial susceptibility and virulence factor genes in Trueperella pyogenes isolated from clinical mastitis cases of dairy cows

Abstract Trueperella pyogenes is an opportunistic pathogen causing important diseases including mastitis and metritis in domestic animals such as dairy cows leading to prominent economic losses in food production industry. The aim of this study was to investigate bacterial species, antimicrobial susceptibility, and presence of virulence factor genes and genotyping of T. pyogenes isolates associated with summer mastitis cases from 22 different farms around Tehran, Iran. Fifty‐five percent of dairy cows with clinical mastitis symptoms was infected by T. pyogenesis indicated that this pathogen is the most important contributor to clinical mastitis in dairy cows in the present study. A significant correlation was illustrated between presence of virulence factor genes of isolated pathogen, biochemical patterns, and the utter infected types. Multidrug resistance susceptibility observed between isolates indicated the important need for prudent use of antimicrobials in treatment of mastitis caused by T. pyogenes and increased concerning of consumer health associated with recent problems of antimicrobial resistance. The categorization of isolates was implemented into seven different clonal related types by COX‐PCR at 80% of similarity cutoff with significance relationship to clonal types, CAMP test result and sampling time and biochemical profile. Regarding to the results obtained at the present study, T. pyogenes can be considered as an important typically cause of purulent and acute form of clinical bovine mastitis and loss of dairy productivity. Further studies with more sample size and high‐throughput omic methods in various sampling time and areas are suggested for study of this pathogen precisely.


| INTRODUC TI ON
Trueperella pyogenes is a Gram-positive, nonsporeforming, nonmotile, opportunistic pathogen, short and rod-shaped (coryneform) bacterium belonging to the Actinemycetales order. T. pyogenes was known previously as Arcanobacterium pyogenes and reclassified because of phospholipid composition, phylogenetic properties, and the presence of Vitamin K 2 . Growing condition of T. pyogenes is anaerobic; however, in a 7% CO 2 atmosphere, it grows optimally (Pohl et al., 2018). It is widely considered ubiquitous, the inhabitant of skin, oropharynx, gastrointestinal, upper respiratory, and urogenital tracts of different domestic animal species. There are many suppurative infections caused by T. pyogenes including pneumonia in pigs; metritis, mastitis, liver abscesses, otitis, peritonitis, pyodermitis, endocarditis, osteomyelitis, lymphadenitis, and uterine diseases in cows and sheep, and abscesses in various wildlife species. As T. pyogenes is not a normal microflora in human and companion animals, infections caused by this bacterium are not common and occasionally associated with occupational exposures (Ribeiro et al., 2015). Several virulence factors, needed for pathogenesis of T. pyogenesis, existed in this pathogen and were reported by . T. pyogenes infection leading to tissue damage is associated with cholesterol-dependent cytolysin, pyolisin encoded by plo gene, which is the primary virulence factor.
Other important virulence factors are colonization and mucosal adherence involved in pathogenicity and considered for this bacterium which were described as fimbriae, collagen binding protein, and neuraminidases encoded with fim (A, C, E and G), cpbA, and nan(H and P) genes, respectively .
Mastitis, one of the most important and costly diseases of dairy industry, is the inflammation of mammary glands. The major effects of this disease caused by mastitis pathogens include decrease in lactation and milk quality by enhancement of somatic cell count (SCC). Mastitis and the related diseases lead loss to producers approximately 2 billion dollars annually in the united states (Srednik et al., 2018). The pathogens causing mastitis include Staphylococcus aureus, Streptococcus dysgalactiae, Trueprella pyogenes, Escherichia coli, Enterococcus spp., Bacillus spp., and Pasteurella haemolytica.
However; the major bacterial agents of mastitis are S. aureus, E. coli, and T. Pyogenes. No apparent change is visible in the udder in subclinical infections; however, decrease in milk production is occurred.
On the other hand, clinical mastitis is known by swelling and pain in the udder with lactation decrease (Bogni et al., 2017). The most important factors which increase the risk of infection by T. pyogenes leading to metritis disease as a metabolic illness through the teat ends and vaginal canals is change in energy demand such as onset of lactation period resulted in increasing the risk of metabolic illness.
Another factor has been determined for this disorder recently is change in feeding behavior (Pomeroy et al., 2017). Administration of antibiotics is the main choice for treatment of mastitis cases.
However, overuse of antibiotics as treatment, preventive, and/or growth promoter agents in animal husbandry has led to the development of antimicrobial resistance in different bacterial species including T. pyogenes. Preventive activities and early treatment with correct antibiotic decrease the mentioned cost leading to higher economic proficiency in animal and food production at the cattle fields (Zhang, Zhao, et al., 2017).
Given the widespread use of antimicrobial agents, antimicrobial susceptibility tests along with molecular detection of antibiotic resistance genes are necessary after the diagnosis of the causative agent, to make the appropriate choice of therapy. Also, resistance to chloramphenicol, macrolides, and beta-lactam antibiotics have been reported previously for T. pyogenes isolates .
However, other treatments such as intravaginal immunization recently have been reported useful for prevention of urinary tract infection with whole-cell vaccine or inactivated T. pyogenes pyolysin as an alternative treatment by antibiotics and preventive strategy for controlling the disease . The objectives of the present study were to investigate bacterial species associated with mastitis cases and subsequently determine the prevalence of T. pyogenes isolates involved in those infections. Furthermore, after genotyping of T. pyogenes isolates, the associations between detected clonal types, antimicrobial resistances, presence of virulence factors encoding genes, and persistence of infection were evaluated.

| Bacterial isolation and growth conditions
The study was conducted on 400 postpartum cows from 22 different farms around Tehran (range of number of cows was 8 to 14 cows per farm), which had clinical of summer mastitis. Clinical mastitis was diagnosed based on the presence of grossly altered secretion, systemic manifestations, and condition of the udder. Also, dairy cows with mastitis were treated by erythromycin, penicillin, kanamycin, neomycin, ceftriaxone, and tylosin antibiotics. Milk samples were collected over 9 months between April and December 2017 in individual sterile containers, stored in ice-packed coolers, and sent to the bacteriology laboratory for further culturing and identification of potentially infectious agents. Smears were provided from all samples followed by bacterial cultivation on brain heart infusion agar (BHI) supplemented with 5% sheep blood and MacConkey agar (Merck, Germany), and incubation for 48 hr at 37℃ under aerobic and anaerobic conditions. Gram-stained smears of all plated isolates were prepared. To identify T. pyogenes isolates, several biochemical tests were performed on the isolates (Malinowski et al., 2011).
Isolated bacterial isolates other than T. pyogenes were also identified using morphological comparison and biochemical tests.

| Antimicrobial susceptibility test
Resistance profiles of the isolated T. pyogenes isolates were determined on Mueller-Hinton agar (Merck, Germany) supplemented with 5% sheep blood using the Kirby-Bauer disk diffusion method according to the Clinical and Laboratory Standards Institute protocols (Weinstein et al., 2019). Twenty-three antibacterial disks (MAST) were used, including gentamicin (GM 120 µg), ampicillin (AP 25 µg),

| DNA extraction and polymerase chain reaction (PCR)
The isolated bacteria were cultured in TSB broth (Merck, Germany) supplemented with 5% bovine serum and incubated at 37 ℃ for 48 hr.
In the next step, 3 ml of cultured TSB broth was centrifuged at 8,000 g for 10 min at 4 ℃, followed by washing the pellet once with saline solution. At last, genomic DNAs were extracted using a commercial

| Screening of genes encoding putative virulence factors
Single PCR method was used to evaluate the presence of putative T. pyogenes virulence genes including plo, nanH, nanP, cbpA, fimA, fimC, fimE, and fimG, macrolide resistance genes erm (X) and erm (B), and tetracycline resistance gene tet (W) (Bicalho et al., 2012;Silva et al., 2008). PCR assays were performed in a reaction mixture with the final volume of 25 µl containing 10 pmol of each primer (BIONEER, Korea), 12.5 µl Taq DNA Polymerase 2x Master Mix Red (Cat No. A180306, AMPLIQON), and 2 µl of DNA template. Primer sequences and PCR conditions are listed in (Table 1). PCR reactions were performed in a TECHNE thermal cycler (TC-512 England).
Finally, PCR products were analyzed as previously mentioned.
Positive controls including plo (MF458305), nanP (MF688763), nanH , and tetW (MF996853) genes and correspond to code of the reference strain, T. pyogenes ATCC 19411, and negative controls consisting of the PCR mixture without addition of DNA templates were included in all PCR runs (Figures 1 and 2).
Genomic DNA of a reference strain of T. pyogenes (ATCC 19,411) and distilled water were used in BOX-PCR as positive and negative controls, respectively. Reaction mixtures were prepared in a total volume of 25 µl containing a primer concentration of 2 µM (BIONEER, Korea), 12.5 µl Taq DNA Polymerase 2x Master Mix Red (Cat No. A180306, AMPLIQON), and 100 ng of template DNA. The reactions were carried out as follows: initial denaturation at 95℃ for 2 min, then 34 cycles of 95℃ for 1 min, 53℃ for 1 min, and 72℃ for 5 min, and a final extension step at 72℃ for 10 min. The amplification products (5 µl) were resolved by electrophoresis on 1.5% agarose gel for 3 hr at 70 V and visualized as above. The 1/0 (presence/absence) binary matrices were processed by NTsys program (NTSYSpc version 2.10e) to calculate the similarity coefficients and construct the clustering tree with an unweighted pair-group method using arithmetic averages (UPGMA). The dendrogram was confirmed to be stable based on our repeated verifications ( Figure 3). The similarity cutoff level to identify clonally related isolates was set up at 75% (Hadimli & Kav 2011;Silva et al., 2008).

| Statistical analysis
The data were described by absolute and relative frequencies.
Statistical analyses were done by SPSS software version 23.0 using chi-square and Fisher's exact tests. The p value ≤.05 was considered to be statistically significant.

| Isolation and identification of T. pyogenes
Fifty-five T. pyogenes isolates (13.75%) were isolated from 400 collected samples, 12 (21.82%) of which were isolated as pure cultures, while the other 43 isolates (78.18%) were obtained from mixed cultures. Other bacterial species which were isolated along with T. pyogenes are listed in (Table 2). Overall, T. pyogenes was the predominant bacterium isolated from the clinical samples. From 55 T. pyogenes isolates, 18, 28, and 9 isolates were from cows with mastitis symptoms during the spring, summer, and fall respectively. Small, irregular Gram-positive coccobacilli were observed in smears obtained from the samples. After 48 hr, small white colonies with β hemolysis were detected on the blood agar plates, but no growth was observed on the MacConkey agar plates. The isolates differed in the intensity of hemolytic zone surrounding the colonies. However, most of them showed a strong β hemolysis (78.18%). The remaining isolates had a moderate zone of β hemolysis on the blood agar. Fifty isolates (72.7%) exhibited strong hemolysis enhanced by Staphylococcus aureus (ATCC 29,213) in CAMP test. The rest of isolates had a weak CAMP reaction. Smears from cultured colonies also contained irregular Gram-positive coccobacilli. Catalase, oxidase, motility, urease, nitrate reduction, and esculin tests were negative for all isolates.
On the contrary, the results of gelatin and Loeffler serum hydrolysis were positive. In litmus milk, production of acid, clot, reduction, and
Complete antimicrobial susceptibility testing results are shown in Table 3.
Recently, the multidrug resistance properties of pathogens have been introduced as a critical problem in the world (Lim et al., 2014).
Overusing of antibiotherapy, uncompleted treatment period, inappropriate drug prescription, horizontal antimicrobial resistance gene

TA B L E 3
The results of antibiotic susceptibility test of T. pyogenes isolated from mastitis samples transferring, and using antibiotics as growth promoters are the key factors of multidrug susceptibility (Doyle, 2015 (Zhang, Zhao, et al., 2017). The presence of tetW gene and macrolide antibiotic resistance ones including ermB and ermX were demonstrated between 58% and 34.5% of isolates, respectively, which showed the importance of horizontal multidrug resistance gene transferring between bacteria. 94.7% of isolates in this study with tylosin resistance characteristic presented ermX gene corresponded with the result reported by Zastempowska and Lassa (2012). Also, they found tetW, ermX, and ermB genes in the T. pyogenesis isolates that identified between cows with clinical mastitis symptoms. T. pyogenesis isolates (Alkasir et al., 2016). With investigation of association between presence of virulence factor genes and some phenotypic properties of isolates, it is demonstrated that the isolates with complete hemolysis, positive CAMP test, and detected common virulence genes lead to disease with higher severity of clinical symptoms. Also; in this study, the isolates identified during summer season were presented the virulence factor genes with higher frequencies in comparison with other times as the most of these identified isolates were categorized in biotype 1 and 3 groups.

| Clonal relationship among the isolates
All 55 T. pyogenes isolates and the standard strain (ATCC 19,411) were discriminated in 7 main groups, including groups A, B, C, D, E, F, and G ( Figure 3). The majority of isolates belonged to group A with 7 clonal types and 31 isolates. Group B included with 2 clonal types and 9 isolates, group C one clonal type and 3 isolates, group D one clonal type and 2 isolates, group E one clonal type and one isolate, and group F included with 2 clonal types and 9 isolates, respectively.
The reference strain did not match with any isolate type and was categorized as type G. Among all 14 types, types IV, II, and XIII were the predominant ones with 8 and 9 isolates, respectively, while types III and XIII (each with one isolate) were the least common types ( Figure 3). Data regarding the relationship between clonal types, virulence factor genes, biochemical profile, CAMP test results, and the sampling time are summarized in (Table 5).
BOX-PCR is considered as a suitable universal fingerprinting technique for molecular typing of bacteria today (Masco et al., 2003).  tests, and strong clinical symptoms were observed in some isolates of T. pyogenesis in this study; however, they were indifferent sampling areas, same clonal related type, and included principally in biotypes 1 and 3. More antibiotic susceptibility genes were observed in some clonally related types; however, in some other ones with different biotype categorization, less virulence genes were detected. Zhao et al. in the year 2013 isolated T. pyogenesis from seeking forest musk deer. According to the obtained result from BOX-PCR typing, they found 3 related clonal types by the reference line at 0.73 (Zhao et al., 2013); however, isolates in this study were categorized and separated into 7 different groups significantly by the reference line at 0.8 generated automatically. Also, Silva et al., (2008) reported the same result as we obtained in this study. They isolated T. pyogenesis from dairy cows with clinical metritis and determined 10 clonal types by BOX-PCR technique with the similarity cutoff at 84%. They evaluated the relationship between virulence factor genes and clonally related types as it has been done in this study for T. pyogenesis isolated from dairy cows with clinical mastitis symptoms (Silva et al., 2008). More sample size and studies for investigation of different biotype and clonally related types are suggested for controlling this pathogen.

| CON CLUS ION
As concluded in this study, the highest percentage (55%) of dairy cows with clinical mastitis symptoms was caused by T. pyogenesis which indicated the importance of this pathogen contributing to clinical mastitis in dairy cows. Also, a significant correlation was observed between presence of virulence factor genes of isolated pathogen, the utter infected types, and biochemical patterns. Isolated T. pyogenesis showed multidrug resistance indicated the prudent use of antimicrobials is needed in treatment of diseases caused by this pathogen. Isolates were categorized into 7 different clonal related types by COX-PCR with significance relationship to clonal types, biochemical profile, CAMP test result, and sampling time. Further investigations with higher sample size in different sampling time and areas are suggested for more precise study of this pathogen.

ACK N OWLED G M ENT
This research was financially supported by Bu-Ali Sina University.
Also, we thank our colleagues in University of Tehran, Faculty of Veterinary Medicine who assisted us in this project.

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

DATA AVA I L A B I L I T Y S TAT E M E N T
All authors confirm that the data supporting the findings of this study are available within the article.