Pathogens, Vol. 11, Pages 1443: Serotyping and Antimicrobial Susceptibility Profiling of Glaesserella parasuis Isolated from Diseased Swine in Brazil

1. IntroductionPigs can be colonized by different microorganisms before weaning, but some of those agents are potentially pathogens, such as G. parasuis. Infection caused by this agent has a serious impact on pig production. Therefore, the correct diagnosis of this disease is essential to establish appropriate control measures. Fibrinous polyserositis and arthritis caused by G. parasuis (Glasser’s disease) is usually diagnosed based on herd history, clinical signs, necropsy findings, and bacterial isolation [1].

G. parasuis is a Gram-negative, NAD-dependent, and fastidious bacterium, and its isolation in pure culture from diseased animals is usually difficult, particularly if the animals received antimicrobial treatments before sample collection. The isolation of the microorganism is important, not only to genotyping and serotyping but also for antimicrobial sensitivity determination, which are very useful for disease control. The isolation of the microorganism is important, not only to genotyping, virulence studies, and serotyping but also for antimicrobial sensitivity determination, which are also very useful for disease control.

The impact of Glässer’s disease can be reduced in some herds by the use of vaccination with commercial bacterins based on serotype and autogenous vaccines; however, the great serovar diversity and the limited cross-serovar protection still represent an important challenge for disease control. The evaluation of virulence factors associated with infection by G. parasuis and the identification of epitopes with good antigenicity are extremely important for the development of subunit vaccines able to block the early steps of infection and reduce the dissemination of pathogens in internal organs. For this reason, antibiotic therapy continues to be necessary for the management of Glässer´s disease outbreaks [1,2]. For G. parasuis control, β-lactams, phenicols, macrolides, sulphonamides, and tetracyclines are often recommended [3]. However, resistance to sulphonamides, tetracycline, quinolones, and florfenicol has already been described for G. parasuis among Chinese, Australian, Brazilian, and Czech herds [4,5,6,7], and in the Spanish swine herds, multidrug-resistant G. parasuis strains were described causing infection [8]. Here, we present the isolation, identification, serotyping, and antibiotic susceptibility profiling of a collection of G. parasuis isolated from diseased swine in Brazil. 3. ResultsG. parasuis strains were isolated from brain (1%–1/105), lung (91.4%–96/105), thoracic cavity (1.9%–2/105), and pericardium (5.7%–6/105) samples. In total, 35.2% of the strains were isolated, between 2009 and 2010, while the remaining 64.8% were obtained between 2011 and 2014, from nine Brazilian states (Figure 1). All strains were identified as G. parasuis by MALDI-TOF, and from PCR to the 16S rRNA gene, all were positive for group 1 vtaA gene detection. Among the 105 tested strains, serotypes 4 and 5 were the most prevalent (27.6% and 24.8%, respectively), followed by serotypes 13 (18.1%), 14 (13.3%), and 1 (11.4%) (Table 1). All 105 strains were susceptible to gentamycin (Table 2). The highest resistance rates were observed against tylosin, sulfonamides, fluoroquinolones, and clindamycin. The aminoglycosides presented a lower resistant rate against Brazilian G. parasuis, while the florfenicol, tiamulin, and β-lactams tested inhibited most of the strains studied. The distribution of MIC values is presented in Figure 2 and Table 3.A total of 98.1% (103/105) of isolates were resistant to at least one of the tested antimicrobial classes; only two isolates were characterized by susceptibility to all tested drugs. Multidrug resistance was detected in 89.5% (94/105) of tested strains, with 23.8% (25/105) being resistant to more than five antibiotic classes (Table 4). No significant difference was observed between the susceptibility profiles and the isolation year (p value = 0.43).No correlation between clusters according to resistance profiles and isolates’ origin and serotype was detected (Figure S1). The MSTs demonstrate the dispersion of G. parasuis serotypes among the resistance profiles obtained (Figure 3B); the dispersion of strains according to isolation date and resistance is observed in Figure 3A. 4. DiscussionThe observed predominance of serotypes 4, 5, 13, and 14 corroborates previous Brazilian reports [19,20,21,22]. These serotypes are also predominant in European and Chinese swine herds, as previously described [8,23,24]. Despite not having a proven correlation, these serotypes have been associated with moderately and highly virulent G. parasuis strains [1], which agrees with all of the studied isolates being positive for group 1 vtaA gene amplification, which has also been associated with virulent G. parasuis [10].The main issue for the evaluation and comparison of G. parasuis antimicrobial susceptibility results is the lack of specific standardized breakpoints. Most published studies apply respiratory disease breakpoints, when available, or make an approximation from specific Actinobacillus pleuropneumoniae and/or Histophilus somni breakpoints [5,7,8,25]. In the present study, breakpoints were selected with the following criteria: those described for swine respiratory diseases were preferably chosen in CLSI documents, and when there was no description in the CLSI or EUCAST, a literature reference was used.Considering the resistance profiles identified, it was possible observe that the highest resistance rates were found against tylosin (97.1%), sulfadimethoxine (89.5%), danofloxacin (80%), trimethoprim/sulfamethoxazole (62.5%), enrofloxacin (54.3%), and clindamycin (50.5%). Resistance against tylosin and clindamycin are becoming very frequent in swine pathogens since these drugs were extensively used in the last 40 years as growth promoters (this method was banned in Brazil in 2020). G. parasuis presenting high MIC values against tylosin and clindamycin are also described by De la Fuente et al. [8] in Spain and Miani et al. [6] in Brazil. Resistance to sulphonamides observed in this study has been described in the literature, to varying degrees in different countries. In Denmark, Aarestrup et al. [26] reported that 3.8% of G. parasuis strains were resistant to trimethoprim/sulphamethoxazole. In Spain and the UK, de la Fuente et al. [8] reported that 53.3% of Spanish and 10% of British strains were resistant to trimethoprim/sulphamethoxazole. In the present study, a higher level of resistance to trimethoprim/sulphamethoxazole and sulfadimethoxine was observed. Zhao et al. [26] described the identification of both sul1 (6.29%) and sul2 (1.4%), in 143 G. parasuis strains from China and ranked these genes as most associated to trimethoprim/sulfamethoxazole resistance. Zhou et al. [7] reported a resistance rate to enrofloxacin of 70% (78/110), and Zhao et al. [26] described 55.9% (80/143), with both studies being conducted in China. These rates are similar to the results observed in our strains (54.3%). Strains evaluated by Zhao et al. [26] presented different gyrA and parC mutations, affecting 125 of 143 strains. The association between these mutations and fluoroquinolone resistance is largely described in Gram-negative bacteria and Gram-positive bacteria from pigs. Contrary to these observations, Miani et al. [6] described different results in Brazil, indicating that the fluoroquinolones class is a good choice in G. parasuis control. These authors reported a very low MIC90 against quinolones (MIC 90 = 0.25 µg/mL to enrofloxacin and 0.12 µg/mL to danofloxacin) when evaluated in 50 clinical isolates. Our findings suggest that this antimicrobial class must be avoided because of growing resistance rates (MIC 90 >2 µg/mL to enrofloxacin and >1 µg/mL to danofloxacin) and the importance to fluoroquinolones of controlling human pathogens. Resistance to the tetracycline drugs tested (oxytetracycline and chlortetracycline) was present in 40 and 26.7% of strains, respectively. The spread of the tetracycline class resistance in swine production is expected because it is one of the most used in-feed antimicrobials in preventive and metaphylactic programs in Brazil [27]. Strains of G. parasuis carrying tet B (23.78%) and tet C (3.5%) genes were described in China and are probably circulating in studied strains [26]. The MIC90 of both tetracyclines evaluated here was 4 µg/mL (considering 105 strains). Recent studies describe MIC90 of 8 µg/mL [6] in Brazil and MIC90 of 16 µg/mL in Germany [28].Aminoglycosides resistance rates were zero against gentamicin, and MIC90 values were low against neomycin and spectinomycin in evaluated strains, but these drugs are not the most indicated to respiratory or systemic disease treatment in swine. The global results indicate variable MIC90 values of this class; Brogden et al. [28], in Germany, describe MIC90 of gentamicin of 4 µg/mL and neomycin of 16 µg/mL. The MIC90 of spectinomycin (≤8.0 µg/mL) was lower than described for Spanish and British strains [8], Chinese strains [7], and in Brazil [6].Resistance rates and MIC90 values against b-lactams were low but looking at the different representative drugs tested it is possible to observe crescent resistance rates. These rates were lower than ceftiofur, followed by ampicillin, and higher than penicillin. These findings agree with values observed in British isolates and differ from Spanish strains, as described by De la Fuente et al. [8]. As described by Brogden et al. [28], the observation of bimodal or broad MIC distributions for several antimicrobial agents (Figure 2) indicates the presence of non-wild type isolates, with acquired mechanisms of resistance. The presence of β-lactamases, for example, is possible in isolates with elevated MICs of penicillin, ampicillin, and ceftiofur. Genes related to β-lactam resistance described in G. parasuis strains to date are blaTEM-1 and blaROB-1 [26].Some macrolide compounds tested (tilmicosin and tulathromycin), pleuromutilin (tiamulin), and phenicol (florfenicol) showed low rates of resistance and low MIC90 values. These findings are quite interesting because these drugs are largely used for treating respiratory and enteric infectious diseases in Brazilian swine, especially tiamulin [27]. Florfenicol resistance was observed in 1.3% of tested strains; other studies described no resistance to this antimicrobial [7,8,28], unlike Miani et al. [6], who described 40% of florfenicol resistance. Zhao et al. [26] described 9% (13/143) of G. parasuis strains studied carrying gene floR, which was previously described in this bacterial species [29]. The emergence of florfenicol resistance in G. parasuis strains was related to a novel small plasmid pHPSF1 bearing floR. This plasmid is similar to other Pasteurellaceae plasmids, suggesting that these species prefer to exchange genetic elements with each other [26]. Chromosomal genomic islands, carrying 13 antimicrobial resistance genes, were described in G. parasuis strains from China. These islands harbored resistance genes for tetracyclines [tet(B)], β-lactams (blaROB-1), sulphonamides (sul2), chloramphenicol (catIII), and aminoglycosides [aph(300)-Ib, aph(6)-Id, and aph(30)-Ia]. These results highlight the important roles that mobile elements play in capturing and diffusing antibiotic resistance genes between plasmids and chromosomes in Pasteurellaceae [30]. Antimicrobial resistance is currently one of the world’s biggest concerns in terms of animal and public health. The world organization for animal health (OIE) standards provide global recommendations for controlling antimicrobial resistance, including lists of antimicrobial agents of veterinary importance to treat animal diseases. In parallel, the world health organization (WHO) has also developed a list of critically important antimicrobial agents in human medicine [31]. Our data indicate that multidrug resistance is present in 89.5% of Brazilian G. parasuis strains tested. Antimicrobial use as a preventive, metaphylactic, and therapeutic treatment is certainly contributing to the selection of these strains. It is particularly concerning that, for example, three out of six antimicrobials, for which resistance was described here in more than 50% of strains, are categorized by the WHO as critically important antimicrobials with maximum priority. Excessive antimicrobial use around the time of colonization of the upper respiratory tract can interfere with the colonization by G. parasuis, and this interference is not limited to this species but also influences the rest of the microbiota. The reduction in bacterial diversity in microbiota in piglets can cause the poorer performance of the immune system, as previously described [20,31,32]. The use of strategic antimicrobial treatments may only be advised in a few limited situations, mainly to treat piglets during a disease outbreak, which is important not only for health but also for welfare issues. Alternative control measures should be taken to minimize the potential increase in Glässer’s disease cases caused by resistant G. parasuis [31]. Therefore, providing veterinarians with data on the prevalent serotypes and on the resistance profiles of the members of this bacterial species is an effective way of contributing to the correct choice of antimicrobials.

留言 (0)

沒有登入
gif