Antimicrobial susceptibility profiles of porcine mycoplasmas isolated from samples collected in southern Europe

Background Mycoplasma (M.) hyopneumoniae, M. hyorhinis and M. hyosynoviae are significant pathogens for the porcine industry worldwide. The aim of the present study was to determine the antimicrobial susceptibility of six key antimicrobials (tylosin, tilmicosin, tylvalosin, lincomycin, tiamulin and valnemulin) routinely used for treating infections caused by these pathogens. Twenty-seven M. hyopneumoniae, 48 M. hyorhinis and 40 M. hyosynoviae field strains isolated from clinical samples from different Southern European countries between 2013 and 2018 using broth microdilution method were evaluated. Results Tylvalosin exhibited the highest in vitro activity among the macrolides assayed, with MIC90 values 4 to 5 two-fold dilutions lower than those of tylosin and tilmicosin. The pleuromutilin valnemulin showed one of the highest in vitro activities against the three mycoplasma species. On the contrary, lincomycin exhibited the highest MIC values of the antimicrobials tested. Conclusions The data obtained in the present study supports the use of pleuromutilins and macrolides for the control of infections caused by porcine mycoplasmas. The use of lincomycin for the treatment of porcine mycoplasma infections should be carefully evaluated due to the presence of circulating field isolates with decreased susceptibility to this antimicrobial.


Background
Mycoplasma (M.) hyopneumoniae, M. hyorhinis and M. hyosynoviae are considered the most relevant Mollicutes to porcine health worldwide, and together with M. suis, a non-culturable haemotropic mycoplasma, represent the main pathogenic mycoplasmas of pigs [1,2]. M. hyopneumoniae is a major porcine pathogen, due to its role as the aetiological agent of enzootic pneumonia and also, by interacting with other microorganisms, as a primary pathogen of the porcine respiratory disease complex [3], a disease considered as the most relevant health concern for pig producers [4]. Commercial vaccines are routinely used for the control of this pathogen. However, the analysis of M. hyopneumoniae vaccines under field conditions has shown variable efficacy [5], leading in practice to the regular use of antimicrobials against its infections. The antibiotics most frequently used against M. hyopneumoniae infections in pigs are aminocyclitols, aminoglycosides, fluoroquinolones, florfenicol, lincosamides, macrolides, pleuromutilins and tetracyclines [6]. On the other hand, M. hyosynoviae is one of the main bacterial pathogen involved in pig lameness [7][8][9]. Infections caused by this pathogen are characterised by a variable progression, leading in most cases to clinical arthritis. Up to date, there are no commercial vaccines available for the control of this microorganism, so the control measures rely on farm management and antimicrobial treatment. Enrofloxacin, lincomycin, tetracyclines, tiamulin and tylosin are commonly used for the treatment of M. hyosynoviae infections [10]. Furthermore, M. hyorhinis is a ubiquitous porcine pathogen, primarily associated with cases of polyserositis, arthritis and otitis in pigs. In some cases it has been linked to cases of pneumonia, acting as a secondary or opportunistic pathogen. Pneumonia caused by M. hyorhinis is clinically indistinguishable from cases produced by M. hyopneumoniae [8][9][10]. Tetracycline, tiamulin, enrofloxacin, tylosin, tilmicosin and lincomycin have demonstrated in vivo efficacy against this microorganism [11,12]. Commercial vaccines are not available for M. hyorhinis, so additional control measures involving improved farm management strategies and the reduction of environmental stressors are required. As described above, the control of porcine mycoplasmosis still depends strongly on the use of antimicrobials. However, the historical overuse of antibiotics in human and animal medicine has led to a current situation of alert, due to the development of resistance that reduces the therapeutic options. In this scenario, the analysis and monitoring of antimicrobial susceptibility has become pivotal in animal health management [13]. The aim of this study was to evaluate the in vitro activity of some of the most relevant antimicrobials used in cases of porcine mycoplasmosis against mycoplasma field strains isolated from Italian, Portuguese and Spanish clinical samples collected between 2013 and 2018.

Mycoplasma strains
A total of 27 M. hyopneumoniae isolates were obtained from cases of porcine respiratory disease. M. hyorhinis isolates (48) were divided into 18 isolates obtained from arthritic joints and 32 from respiratory diseases cases. The 40 M. hyosynoviae isolates were obtained from arthritis cases.

MIC values for M. hyopneumoniae
The MIC range, MIC 50 and MIC 90 for each antimicrobial tested against M. hyopneumoniae are presented in Table 1. For macrolides, isolates tested against tylvalosin showed an MIC range of 0.016-0.06 μg/ml and MIC 50/90 values of 0.03/0.06 μg/ml. Results obtained for the other two macrolides tested, tylosin and tilmicosin, were higher than those described for tylvalosin, with an MIC range of 0.06-1 and 0. 25

Discussion
This investigation was aimed at evaluating the antimicrobial susceptibility patterns of selected members of the macrolide, lincosamide and pleuromutilin antimicrobial families against isolates of the most clinically relevant porcine mycoplasmas collected from Southern European countries. Macrolides, lincosamides and pleuromutilins are extremely relevant for porcine health. For instance, the World Organisation for Animal Health (OIE) classifies these antimicrobial families as either critically (macrolides) or highly important antimicrobials (lincosamides and pleuromutilins) in veterinary medicine [14]. Furthermore, these antimicrobials have remained pivotal for the treatment of infections caused by porcine mycoplasmas in the last decades [3,12,15,16]. Due to the limited number of isolates analysed per country, a valid comparison of the MIC values based on their geographical origin could not be achieved.
In our study tylvalosin exhibited the highest in vitro activity among the macrolides assayed against the three mycoplasma species, with MIC 90 values 16 to 32-times lower than those of tylosin and tilmicosin. Similar results were obtained by Tavío et al. [17] after evaluating field strains of M. hyopneumoniae, as well as for the avian mycoplasma M. synoviae and Mycoplasma sp. 1220 [18,19]. Freely accessible tylvalosin MIC values for mycoplasmas are limited up to date to M. hyopneumoniae, M. hyorhinis, M. synoviae, Mycoplasma sp. 1220 and M. gallisepticum [17][18][19][20]. However, the MIC values observed in our work for this antimicrobial against M. hyorhinis and M. hyosynoviae, and the in vitro [17] and in vivo efficacy [21] demonstrated for the treatment of M. hyopneumoniae infections, gives prominence to tylvalosin as one of the most effective macrolides against porcine mycoplasmosis, together with tulathromycin, a 15-membered macrolide. Up to date, there are limited comparative data between tylvalosin and tulathromycin MIC values for porcine mycoplasmas. Felde et al. [22] described tulathromycin MIC 90 values of 1 and 4 μg/ml, 1-4 two-fold dilutions higher to those observed for tylvalosin after analysing Central European M. hyopneumoniae isolates. Conversely, Klein et al. [23] described MIC 90 values for tulathromycin of 0.004 μg/ml, after analysing 50 European M. hyopneumoniae isolates, however, the authors did not include tylvalosin in their analysis. In addition, decreased susceptibility to tylosin and tilmicosin has been previously described in European M. hyosynoviae and M. hyopneumoniae isolates [10,24]. Although tylvalosin was not evaluated in these studies decreased susceptibility for this antimicrobial has yet to be described. The relatively recent application of tylvalosin for therapy purposes in pigs may explain the absence of decreased susceptibility for this antimicrobial. Also, Andersen and others [25], described the presence of alternative bacterial resistance mechanisms for tylosin in comparison to tylvalosin, which can explain the differences in MIC values between members of the 16membered ring class of macrolides. The tylosin and tilmicosin MIC values were in agreement with those found in M. hyosynoviae [10], M. hyorhinis [12] and M. hyopneumoniae [17,23]. Based on our data, the macrolides could still be utilised as first-line antimicrobial treatment for the control of mycoplasma disease in pigs. However, evidence about the emergence of macrolideresistant M. hyopneumoniae strains [17,22,24,26],   [24,26]. Current strategies for the control of M. hyopneumoniae include the respiratory exposure of gilts to lung homogenates containing viable strains of this pathogen [27]. However, the presence of circulating strains with decreased susceptibility to certain antimicrobials, as found in our work, or even harbouring multidrug resistance phenotypes [26], requires a careful evaluation of the antimicrobial susceptibility patterns of the strains included in the homogenate prior administration to gilts, in order to reduce the dissemination of resistant isolates of M. hyopneumoniae in the farm. Lincomycin has been previously described as effective in vitro against M. hyorhinis [12,28]. Conversely, our data suggest the presence of resistant strains of M. hyorhinis against this antimicrobial in the population studied, in agreement with the data published by Bekö et al. [20] in Hungarian isolates of this pathogen. M. hyorhinis is a common commensal of the upper respiratory tract of pigs, acting as an opportunist pathogen of immunocompromised animal in a variety of clinical presentation [10]. In our study, the population M. hyorhinis analysed showed a marked decrease in susceptibility for lincomycin. This antimicrobial has been extensively used as part of infeed medication in porcine farming, administered orally in premix, oral powder and oral solution for years [29]. Therefore, it can be hypothesised that the intensive selective pressure due to in-feed medication for long periods has facilitated the development of a marked decreased susceptibility in our M. hyorhinis strain population. This selective pressure may have enabled in the same way the lincomycin susceptibility and MIC value distribution observed in the M. hyosynoviae and M.   [10]. Differences in effectivity between pleuromutilins were observed in the three mycoplasma species analysed as previously described [17], with valnemulin exhibiting 8 to 16 times higher activity than tiamulin. MIC values for tiamulin and valnemulin against M. hyopneumoniae were in agreement with those previously reported by Tavío et al. [17], and similar to the MIC values observed in isolates from Central Europe, where all isolates were susceptible to a concentration equal or lower than 0.039 μg/ml [22]. However, Klein et al. [23], reported lower MIC values for 50 European isolates of M. hyopneumoniae, with a difference in MIC 90 found between studies of 8 to 16 times. Also, Hannan et al. [16] described lower MIC values for M. hyopneumoniae and M. hyosynoviae against both pleuromutilins. These data suggest a potential decreased activity against pleuromutilins for these two porcine pathogens. M. hyorhinis MIC data presented the broadest range of values for both pleuromutilins tested, in conjunction with the highest MIC values (8 μg/ml for tiamulin and 1 μg/ml for valnemulin).
Other authors have described lower MIC values for both pleuromutilins in M. hyorhinis for Hungarian isolates [18], with similar results to those described for M. hyopneumoniae in the same region after testing the same antimicrobials [22]. Reduced susceptibility to pleuromutilins in mycoplasmas has been linked to mutations in the 23S rRNA gene [30], mutations that also conferred cross-resistance to lincomycin, tilmicosin and tylosin. Even though there are no described mechanisms of pleuromutilin resistance in M. hyorhinis, macrolide and lincomycin decreased susceptibility has been associated to multiple mutations in the same gene [31], suggesting a possible cross-resistance phenomenon between pleuromutilins, lincomycin, tilmicosin and tylosin based on the high MIC values for these antimicrobials observed in the M. hyorhinis population studied.

Conclusions
In conclusion, among the tested antimicrobials, valnemulin and tylvalosin displayed the greatest in vitro activity against the three porcine mycoplasmas species collected from Southern European countries. Although active for M. hyosynoviae, the high MIC values observed  for lincomycin against M. hyorhinis and M. hyopneumoniae isolates, suggest the need for a more rational approach to the use of this antimicrobial in cases of porcine mycoplasmosis. The molecular basis of the potential cross-resistance for lincomycin, tilmicosin, tylosin and both pleuromutilins in M. hyorhinis should be analysed. The differences observed between isolates from different European regions draw special attention to the need for standardised antimicrobial susceptibility testing. Besides, coordinated monitoring schemes for these pathogens in Europe are essentialto effectively tackle the potential emergence of resistant mycoplasma strains in order to maintain an optimal level of health and welfare in the porcine industry.

Mycoplasma strains
Strains were isolated from clinical samples submitted by field veterinarians for routine diagnosis to the mycoplasma diagnostic service of the Instituto Universitario de Sanidad Animal y Seguridad Alimentaria (Gran Canaria, Spain) between 2013 and 2018. Samples were sourced from farrow-to-finish farms and obtained from fattening pigs, gilts and sows with no previous antimicrobial treatment. Clinical samples analysed included nasal swabs from those animals with respiratory disease symptoms compatible with mycoplasmal pneumonia and synovial fluid from cases of lameness. The isolation of the strains was performed as previously described [32], using modified Friis broth supplemented with equal parts of horse and porcine serum [17]. Pure cultures were obtained as previously described [32,33] and the identity confirmed by biochemical characteristics and specific PCR testing [3,34]. Strain distribution by species and country of origin can be found in Tables 1, 3 and 5. The type strains of M. hyopneumoniae (J), M. hyorhinis (BTS-7) and M. hyosynoviae (S16) were used as a positive control. All control strains were sourced from the National Collection of Type Cultures (NCTC).

Antimicrobials
Tiamulin, tilmicosin, valnemulin and lincomycin were obtained from Fluka Analytical (St Louis, Missouri, USA). Tylosin was obtained from Serva (Heidelberg, Germany) and tylvalosin from Eco Animal Health (London, UK). All antibiotic stock solutions were sterilised by filtration through 0.2 μm pore size membrane filters (Millipore).

Minimum inhibitory concentration testing
Minimum inhibitory concentration (MIC) testing was performed using a microbroth dilution method as previously described [35], using 96-well round base polystyrene microtitre plates (Sarstedt, Nümbrecht, Germany). Antimicrobials were added into the wells following a doubling dilution pattern in order to obtain a final concentration of antibiotic per well that ranged from 0.002 to 64 μg/ml. 100 μl of each antimicrobial dilution was added into each well and inoculated with 100 μl of a 48-h mycoplasma culture. The bacterial load present in each inoculum was calculated as described before [17], and adjusted to a final concentration of 10 5 colour-changing units/ml of mycoplasma per well. Inoculated plates were then incubated at 37°C with constant shaking at 150 rpm in a humidified atmosphere until growth in the drug-free control wells was evident. Bacterial growth was examined daily until a colour change was observed for a maximum of 21 days. MIC testing was performed on three different days, and duplicates of each strain were performed on each of the testing days. MIC was defined as the lowest concentration that completely inhibited growth, shown by a lack of colour change at the time that the drug-free growth control exhibited a colour change, while the negative control remained unchanged. MIC 50 and MIC 90 are defined as the lowest concentration that completely inhibited growth in 50 and 90% of the population studied respectively, and MIC ranges were also calculated. Due to the lack of official breakpoints for porcine mycoplasmas, the percentage of resistant isolates was not calculated. MIC values of the type strains can be found in Tables 1, 3 and 5. A total of five independent observations per type strain was performed.