Tiamulin resistance in porcine Brachyspira pilosicoli isolates

Research in Veterinary Science 80 (2006) 1–4 www.elsevier.com/locate/rvsc

Tiamulin resistance in porcine Brachyspira pilosicoli isolates M. Pringle *, A. Lande´n, A. Franklin Department of Antibiotics, National Veterinary Institute, SE-751 89 Uppsala, Sweden Accepted 16 February 2005

Abstract There are few studies on antimicrobial susceptibility of Brachyspira pilosicoli, therefore this study was performed to investigate the situation among isolates from pigs. The tiamulin and tylosin susceptibility was determined by broth dilution for 93 and 86 porcine B. pilosicoli isolates, respectively. The isolates came from clinical samples taken in Swedish pig herds during the years 2002 and 2003. The tylosin minimal inhibitory concentration (MIC) was >16 lg/ml for 50% (n = 43) of the isolates tested. A tiamulin MIC >2 lg/ml was obtained for 14% (n = 13) of the isolates and these were also tested against doxycycline, salinomycin, valnemulin, lincomycin and aivlosin. For these isolates the susceptibility to salinomycin and doxycycline was high but the MICs for aivlosin varied. The relationship between the 13 tiamulin resistant isolates was analyzed by pulsed-field gel electrophoresis (PFGE). Among the 13 isolates 10 different PFGE patterns were identified. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Brachyspira pilosicoli; Tiamulin; Antibiotic resistance; Spirochaete; PFGE

1. Introduction Treatment failure of porcine intestinal spirochaetosis (PIS) with tiamulin in a Swedish pig herd was reported in 2002 (Karlsson et al., 2002). The etiologic agent of PIS is an anaerobic spirochaete Brachyspira pilosicoli (Trott et al., 1996) and the disease is characterized by nonfatal diarrhoea in growing pigs causing reduction in growth rate. A recent study shows that B. pilosicoli is commonly isolated from pigs in herds with diarrhoeal problems and poor performance (Jacobson et al., 2003). Due to withdrawal of drugs authorized for use in pigs the antibiotic arsenal against PIS is diminishing. In many countries tiamulin, a pleuromutilin, is the drug of choice for treatment of PIS. Tiamulin resistance in B. pilosicoli is a threat to the pig industry as long as there are no other real control alternatives. Only sparse information is available regarding antimicrobial susceptibility in porcine B. pilosicoli. The *

Corresponding author. Tel.: +46 18 67 40 00; fax: +46 18 30 91 62. E-mail address: [email protected] (M. Pringle).

0034-5288/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.rvsc.2005.02.004

aim of this study was to investigate the situation in diarrhoeic pigs from Swedish herds through tests for tiamulin and tylosin susceptibility of field isolates of B. pilosicoli. Isolates with a tiamulin minimal inhibitory concentration (MIC) >2 lg/ml were also tested against doxycycline, salinomycin, valnemulin, lincomycin and aivlosin. Additionally the relationship between 13 tiamulin resistant isolates was analyzed by pulsed-field gel electrophoresis (PFGE).

2. Materials and methods 2.1. Bacterial isolates and growth conditions Swedish isolates of B. pilosicoli (n = 103) from clinical submissions of faecal samples to the National Veterinary Institute, Uppsala, Sweden, during 2002 and 2003 were studied. The bacteria were isolated as previously described (Fellstro¨m and Gunnarsson, 1995). One isolate of B. pilosicoli from each positive submission is routinely stored in liquid nitrogen. The majority of the

2

M. Pringle et al. / Research in Veterinary Science 80 (2006) 1–4

samples came from different pig herds but nine herds were represented by more than one sample each. When there was no difference in tiamulin MIC between the isolates from one herd they were excluded from the study (n = 10). When there was a difference of three twofold dilutions or more, two isolates were included from each herd. In the tiamulin MIC distribution diagram 93 isolates from 87 herds are represented. Because of uncertainties in the reading of tylosin MICs (skipped wells) another seven isolates were removed and the tylosin MIC distribution diagram includes 86 isolates from 80 herds. Thawed isolates were grown on fastidious anaerobe agar (National Veterinary Institute, Uppsala, Sweden) and re-identified as B. pilosicoli by positive hippurate hydrolysis test and weak haemolysis on tryptone soya agar (Oxoid, Hampshire, UK) with 5% ox blood. For 13 tiamulin resistant isolates species identification was confirmed by PCR based on the 16S rRNA gene (Fellstro¨m et al., 1997). The purity of all isolates was assessed by phase contrast microscopy. All cultures were grown in 2.5 l jars in an anaerobic environment provided by gas generator envelopes (BBL GasPak Plus, Becton Dickinson) in 37 °C. 2.2. MIC determinations The antimicrobial susceptibility testing was performed by broth dilution as described previously (Karlsson et al., 2003). Briefly, antimicrobial agents were dried in serial twofold dilutions in tissue culture trays, in which a suspension of the bacteria was dispensed (0.5 ml/well) and incubated at 37 °C for 4 days. The medium for the susceptibility tests was brain heart infusion broth supplemented with 10% foetal calf serum. The MIC was read as the lowest concentration of the antimicrobial agent that prevented visible growth. Tiamulin and tylosin were tested for all isolates, and isolates with an MIC of tiamulin >2 lg/ml were considered to be resistant and also tested against doxycycline, salinomycin, valnemulin, lincomycin and aivlosin. The antimicrobial aivlosin (3-acetyl-400 -isovaleryltylosin) is a modification of tylosin. 2.3. Pulsed-field gel electrophoresis The PFGE was performed as described previously, except that only one restriction enzyme, MluI, was used (Fellstro¨m et al., 1999). The band patterns were analyzed visually.

3. Results The distribution of the tiamulin MICs for 93 B. pilosicoli isolates is shown in Fig. 1. The susceptible popula-

50% 40% 30% 20% 10% 0%

≤0.016 0.031 0.063 0.125 0.25

0.5

1

2

4

8

16

32

64

MIC (µg/ml)

Fig. 1. Distribution of tiamulin MICs for 93 Brachyspira pilosicoli isolates from 87 different Swedish pig herds during the years 2002 and 2003.

tion with tiamulin MICs of 0.03–0.125 lg/ml dominates. There is a tendency towards a trimodal distribution having two populations with decreased susceptibility, the first with a peak at 0.5 lg/ml and the second at 32–64 lg/ml. From one herd five isolates from different sampling occasions were tested. Two of these are represented in the MIC distribution diagram (Fig. 1) and all five in Table 1. The MICs of aivlosin, doxycycline and salinomycin for the 13 isolates with a tiamulin MICs >2 lg/ml are presented in Table 2. These isolates had ten different PFGE patterns (Fig. 2). The bimodal distribution of tylosin MICs for 86 B. pilosicoli isolates is shown in Fig. 3.

4. Discussion There is no accepted clinical breakpoint for tiamulin resistance in Brachyspira spp. Using the proposed clinical breakpoint >4 lg/ml 11% of the 94 B. pilosicoli isolates in this study were resistant to tiamulin (Rønne and Szancer, 1990). In SVARM, the Swedish Veterinary Antimicrobial Resistance Monitoring program, a microbiological cut off value of >2 lg/ml is used (SVARM, 2003). With this cut off, 14% of the isolates in this study would be designated as resistant. The mechanisms of tiamulin resistance in B. pilosicoli are not known. In vitro tiamulin resistance develops in a stepwise manner in B. pilosicoli (Karlsson et al., 2001). The situation in vivo could be the same, which would explain the tendency to the trimodal MIC distribution (Fig. 1). The MIC distribution makes it difficult to set cut off values and more studies to define a clinical breakpoint are needed. However, for monitoring of resistance it is more important to detect the low-level resistance (or decreased susceptibility) than to find the isolates with the highest MICs. The low-level resistance could be the first step towards higher MICs and hence all the more important to control.

M. Pringle et al. / Research in Veterinary Science 80 (2006) 1–4

3

Table 1 Antimicrobial susceptibility for five tiamulin resistant Swedish field isolates of B. pilosicoli from different sampling occasions at the same farm Isolate

Month of isolation

1085/02 4653/02 125/03 984/03 3218/03

Apr 2002 Dec 2002 Jan 2003 Mar 2003 Sept 2003

MIC (lg/ml) Tiamulin

Tylosin

Aivlosin

Doxycycline

Salinomycin

2 16 16 32 16

>128 16 32 16 8

16 2 4 8 2

60.25 60.25 60.25 60.25 60.25

0.5 0.5 0.5 0.5 0.5

Table 2 Antimicrobial susceptibility for 13 tiamulin resistant Swedish field isolates of B. pilosicoli Isolate

738/02 953/02 991/02 1854/02 1890/02 1921/02 2015/02 2448/02 2507/02 4492/02 4653/02 1142/03 3065/03

MIC (lg/ml) Tiamulin

Valnemulin

Aivlosin

Tylosin

Lincomycin

Doxycycline

Salinomycin

32 4 64 4 64 8 64 64 4 32 16 64 16

2 1 2 4 4 1 1 4 1 2 0.5 4 2

8 8 32 >32 >32 4 2 >32 4 4 2 8 >32

4 32 >128 >128 >128 8 4 >128 64 16 16 >128 >128

4 62 32 16 32 4 4 32 16 16 62 64 64

60.25 60.25 60.25 60.25 60.25 60.25 0.5 60.25 60.25 0.5 60.25 60.25 0.5

0.5 0.5 0.25 0.5 0.25 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.25

50% 40% 30% 20% 10% 0%

≤2

4

8

16

32

64

128

>128

MIC (µg/ml)

Fig. 2. Different PFGE patterns from Swedish isolates of B. pilosicoli with decreased tiamulin susceptibility digested with MluI. Lane 1 and 16, Lambda Ladder PFG Marker, New England, BioLabs; lanes 2–14, isolates from different farms; lane 15 duplicate of the isolate in lane 14.

From six herds isolates with a tiamulin MIC difference of 3–10 twofold dilutions were found within the herd. This difference indicates that there was more than one clone present in a herd. Fossi et al. (2003) have shown by PFGE that it is not uncommon with more than one clone of B. pilosicoli in one pig herd. For this reason, it is important to test several isolates from one herd for antimicrobial susceptibility, especially before an eradication program. From one herd five isolates from different sampling occasions during 2002 and 2003 were tested (Table 1).

Fig. 3. Distribution of tylosin MICs for 86 Brachyspira pilosicoli isolates from 80 different Swedish pig herds during the years 2002 and 2003.

Four of these isolates have tiamulin MICs (16–32 lg/ml) that most likely make them untreatable with the drug. The resistance pattern for these isolates indicates that this might be a single clone that persists and dominates in the herd for a prolonged time. Tiamulin resistant isolates of B. pilosicoli were found at 13 different farms. The majority of these farms are situated in the same region of Sweden and a possible relationship between the isolates was investigated. Among the resistant isolates ten different PFGE patterns (different clones) were identified (Fig. 2). This finding identifies a development of tiamulin resistance in B. pilosicoli at

4

M. Pringle et al. / Research in Veterinary Science 80 (2006) 1–4

several different locations instead of spreading of one resistant clone through trade with pigs. Hence, instead of measures to prevent spread of the disease, the pattern of tiamulin usage at farm level is probably of greater importance to limit the occurrence of tiamulin resistance in B. pilosicoli. All the 13 tiamulin resistant isolates were susceptible to doxycycline and salinomycin. The valnemulin MICs were higher than the normal susceptible population MICs that are in between 0.016 and 0.063 lg/ml (Karlsson et al., 2001). The MICs of tylosin and aivlosin varied and the isolates with the highest aivlosin MICs were also tylosin resistant. In a previous study high tylosin MICs (>128 lg/ml) obtained for Swedish Brachyspira hyodysenteriae isolates were followed by a slight rise of the aivlosin MICs (Karlsson et al., 2004a) indicating that the 23S rRNA mutation causing macrolide resistance in B. hyodysenteriae (Karlsson et al., 1999) also affects the binding of aivlosin. The molecular structure of tylosin and aivlosin is similar but the longer disaccharide side chain on the aivlosin molecule is probably enhancing the binding to the ribosome. However tylosin resistant B. hyodysenteriae isolates with higher aivlosin MICs (>32 lg/ml) have also been found indicating an additional change in these isolates (Karlsson et al., 2004a). Of the 13 B. pilosicoli isolates, the tylosin susceptible had an aivlosin MIC of 2–8 lg/ml and all but one of the tylosin resistant, P32 lg/ml. The isolates with high tylosin MICs also had higher lincomycin MICs, which is due to the fact that macrolides and lincosamides have overlapping binding sites on the ribosome. Interestingly two of the tiamulin resistant isolates seem to have increased their susceptibility to lincomycin (MIC 62 lg/ml). In B. pilosicoli two 23S rRNA mutations causing macrolide resistance have been described (Karlsson et al., 2004b). The proportion of tylosin susceptible B. pilosicoli isolates was higher than among Swedish B. hyodysenteriae (causing swine dysentery) isolates (SVARM, 2003). The distribution of the MICs was bimodal and with a cut-off at 16 lg/ml 50% of the isolates were resistant. Considering the limited number of antibiotics available, tylosin is still a treatment alternative for PIS in Sweden despite the rather high percentage of resistant isolates. In conclusion, tiamulin resistance is present among Swedish B. pilosicoli isolates from pigs. The PFGE results indicate that this resistance has developed independently at separate farms.

Acknowledgements ¨ sterdahl and MargaThe authors thank Anthony O reta Horn af Rantzien for technical assistance. The

study was supported by the Swedish Farmers´ Foundation for Agricultural Research.

References Fellstro¨m, C., Gunnarsson, A., 1995. Phenotypical characterisation of intestinal spirochaetes isolated from pigs. Research in Veterinary Science 59, 1–4. Fellstro¨m, C., Pettersson, B., Thomson, J., Gunnarsson, A., Persson, M., Johansson, K.-E., 1997. Identification of Serpulina species associated with porcine colitis by biochemical analysis and PCR. Journal of Clinical Microbiology 35, 462–467. Fellstro¨m, C., Karlsson, M., Pettersson, B., Zimmerman, U., Gunnarsson, A., Aspan, A., 1999. Emended descriptions of indole negative and indole positive isolates of Brachyspira (Serpulina) hyodysenteriae. Veterinary Microbiology 70, 225–238. Fossi, M., Pohjanvirta, T., Pelkonen, S., 2003. Molecular epidemiological study of Brachyspira pilosicoli in Finnish sow herds. Epidemiology and Infection 131, 967–973. Jacobson, M., Ha˚rd af Segerstad, C., Gunnarsson, A., Fellstro¨m, C., de Verdier Klingenberg, K., Wallgren, P., Jensen-Waern, M., 2003. Diarrhoea in the growing pig – a comparison of clinical, morphological and microbial findings between animals from good and poor performance herds. Research in Veterinary Science 74, 163– 169. Karlsson, M., Fellstro¨m, C., Heldtander, M.U., Johansson, K.-E., Franklin, A., 1999. Genetic basis of macrolide and lincosamide resistance in Brachyspira (Serpulina) hyodysenteriae. FEMS Microbiology Letters 172, 255–260. Karlsson, M., Gunnarsson, A., Franklin, A., 2001. Susceptibility to pleuromutilins in Brachyspira (Serpulina) hyodysenteriae. Animal Health Research Reviews 2, 59–65. Karlsson, M., Franklin, A., Stampe, M., Fellstro¨m, C., 2002. Treatment failure in spirochaetal diarrhoea. Svensk Veterina¨rtidning 54, 245–247. Karlsson, M., Fellstro¨m, C., Gunnarsson, A., Lande´n, A., Franklin, A., 2003. Antimicrobial susceptibility testing of porcine Brachyspira (Serpulina) species isolates. Journal of Clinical Microbiology 41, 2596–2604. Karlsson, M., Aspan, A., Lande´n, A., Franklin, A., 2004a. Further characterization of porcine Brachyspira hyodysenteriae isolates with decreased susceptibility to tiamulin. Journal of Medical Microbiology 53, 281–285. Karlsson, M., Fellstro¨m, C., Johansson, K.-E., Franklin, A., 2004b. Antimicrobial resistance in Brachyspira pilosicoli with special reference to point mutations in the 23S rRNA gene associated with macrolide and lincosamide resistance. Microbial Drug Resistance 10, 204–208. Rønne, H., Szancer, J., 1990. In vitro susceptibility of Danish field isolates of Treponema hyodysenteriae to chemotherapeutics in swine dysentery (SD) therapy. Interpretation of MIC results based on the pharmacokinetic properties of the antibacterial agents. In: Proceedings, International Pig Veterinary Society, 11th Congress, Lausanne, Switzerland, p. 126. SVARM, 2003. Swedish Veterinary Antimicrobial Monitoring, 2004. The National Veterinary Institute (SVA), Uppsala, Sweden. ISSN 1650-6332. Available from: . Trott, D.J., Stanton, T.B., Jensen, N.S., Duhamel, G.E., Johnson, J.L., Hampson, D.J., 1996. Serpulina pilosicoli sp. nov., the agent of porcine intestinal spirochetosis. International Journal of Systematic Bacteriology 46, 206–215.