Antimicrobial susceptibility of Clostridium perfringens isolated from piglets with or without diarrhea in Brazil

Abstract

The minimum inhibitory concentration (MIC) was determined for 13 antibiotics against Clostridium perfringens isolated from Brazilian piglets. The collection of isolates was performed in June to October 2010. All isolates were susceptible to amoxicillin and ceftiofur, whereas most were resistant to tetracycline and lincomycin. Avilamycin and narasin were more effective against isolates from non-diarrheic than from diarrheic piglets. The other antimicrobials were less active in need of high concentrations to inhibit the growth of the C. perfringens type A. These results suggest the need for further studies evaluating molecular factors related to the antimicrobial resistance of C. perfringens.

Clostridium perfringens type A is an anaerobe commonly isolated from the guts of piglets and is considered a normal inhabitant of piglet intestines. Current evidence suggests that this microorganism is involved in diarrhea in nursing piglets, especially those less than four days of age (17). The main source of immediate post-farrowing infection in piglets is spores in the environment and in the feces of sows (8). Data concerning the prevalence of this pathogen in piglets are rare in the scientific literature. A retrospective study conducted at the Veterinary Diagnostic Laboratory at Iowa State University, USA identified C. perfringens type A as the main causative agent of neonatal diarrhea in the years 2005 and 2006, composing 48% of the 273 strains isolated from diarrheic piglets (22).

Previously, only alpha toxin and enterotoxin were thought to play a role in the pathogenicity of C. perfringens type A-related diarrhea in piglets. However, Songer and Glock (17) suggested that these toxins are not the ones responsible for lesions and liquid accumulation in the intestinal tract of diseased animals. Rather, a novel toxin encoded by the cpb2 gene, called beta-2 toxin, is speculated to be the main factor driving the development of clinical diarrhea by C. perfringens type A in neonatal piglets (19). The adoption of a high pig density in intensive systems and other recent changes in Brazilian pig farming have increased the infection pressure and the stress levels to which animals are subjected. Therefore, conditions facilitate the emergence of pathogenic agents that had once coexisted in equilibrium with the host and its microbiota. Antibiotics have become an essential tool for pig farmers as a means of the therapy, prophylaxis, and growth promotion (1).

In contrast, the World Health Organization has questioned the use of antibiotics as growth promoters. Therefore, a number of countries, such as members of the European Union, have reduced the use of antibiotics. The use of medicated feed in swine production was restricted due to the risk of antibiotic residues in the meat and in the selection of resistant strains, which could lead to human infections with resistant bacteria (2,20). In Brazil, antimicrobials are still widely used in swine production, and little is known about C. perfringens antibiotic susceptibility. The objective of this work was to evaluate the minimum inhibitory concentration (MIC) of 13 antibiotics against 50 C. perfringens type A isolates isolated from neonatal piglets with or without diarrhea.

The C. perfringens type A isolates studied were isolated from one- to seven-day-old piglets with or without diarrhea. Samples were taken from 15 different pig farms located at Triângulo Mineiro and Alto Paranaíba, Minas Gerais, an important swine-producing region of Brazil. The collection of isolates was performed in June to October 2010. All pig farms sampled had a history of diarrhea outbreaks in the farrowing house, especially in the pigs’ first week of life. The pig farms not using antibiotic growth promoters in the diet and fecal samples were collected from piglets which had not been treated at birth with any antibiotics and coccidiostats. Fecal samples were collected directly from the rectal ampulla into sterile containers and subjected to a bacteriologic culture on C. perfringens selective agar (DIFCO^TM SPS agar, Difco Laboratories, Detroit, USA). Plates were incubated at 37ºC in an anaerobic atmosphere with 10% H₂, 10% CO₂, 80% N₂ and were examined after 24 hours (15). Presumptive identification of C. perfringens was determined by colonial and microscopical morphology and confirmed by biochemical tests and multiplex-PCR (21). All isolated were classified as C. perfringens type A, and 31 (62%) were positive for the cpb2 gene.

The MIC was determined by the agar dilution method, as recommended by the CLSI (5). The following 13 antibiotics used in the Brazilian swine industry were evaluated: amoxicillin, avilamycin, ceftiofur, florfenicol, josamycin, leucomycin, lincomycin, narasin, neomycin, norfloxacin, streptomycin, tetracycline, and tylosin (16). For each antimicrobial tested, the MIC₅₀ and MIC₉₀, i.e., the minimum concentration that inhibited growth of 50% and 90% of the isolates, respectively, were calculated. Bacteroides fragillis (ATCC 25285) was used as a control strain.

The MIC of the 13 antimicrobials tested against 50 C. perfringens type A isolates are summarized in Table 1. The antibiotics that showed the best MIC₅₀ and MIC₉₀ values were amoxicillin, ceftiofur and narasin. These results are in agreement with those reported by Tansuphasiri et al. (18) and Rood (13). Both described a high sensitivity of C. perfringens isolated from pigs to beta-lactams. In broiler chickens and dogs, the results for beta-lactams were similar to those described in this study (10,15). However, one report demonstrated that isolates of C. perfringens from cattle exhibited decreased sensitivity to beta-lactams antibiotics (14).

Table 1.

Distribution of the minimum inhibitory concentration (μg/mL) of 13 antimicrobials for 50 C. perfringens type A isolates isolated from one- to seven-day-old piglets, with and without diarrhea.

Antibiotic	Number of C. perfringens isolates with MIC values (μg/mL)
	0.25	0.5	1	2	4	8	16	32	64	128	256	>256	MIC50	MIC90
Amoxicillin	5	31	9	2	2	1	0	0	0	0	0	0	0.5	1
Avilamycin	0	0	15	2	16	9	0	0	2	1	0	5	4	128
Ceftiofur	21	9	2	4	6	5	0	0	0	0	3	0	0.5	8
Florfenicol	0	1	2	24	9	2	2	2	0	7	1	0	2	128
Josamycin	0	0	0	1	3	17	6	0	2	0	5	16	16	>256
Leucomycin	0	0	7	14	13	0	2	0	0	0	7	7	4	>256
Lincomycin	0	0	0	6	8	8	1	1	0	0	7	19	16	256
Narasin	0	2	1	41	0	2	0	0	0	0	0	4	2	8
Neomycin	0	0	1	0	0	0	0	0	0	1	4	44	>256	>256
Norfloxacin	0	0	3	0	0	2	1	3	9	23	9	0	128	256
Streptomycin	0	0	0	0	0	0	0	0	0	0	8	42	>256	>256
Tetracycline	0	0	0	4	0	4	21	2	0	5	14	0	16	256
Tylosin	13	23	5	2	1	0	3	3	0	0	0	0	0.5	16

In this study, tylosin was highly effective against piglet C. perfringens isolates. Our results contrast with those of Devriese et al. (6), who reported that a high percentage of piglet C. perfringens isolates were resistant to this antibiotic. It is interesting to note that while tylosin is an antibiotic commonly used for porcine proliferative enteropathy, our results suggest that it can also be effective for the treatment and control of C. perfringens in piglets.

Florfenicol and leucomycin showed a low MIC₅₀ but a high MIC₉₀. According to Post and Songer (12), these results suggest that in vivo resistance exists in a proportion of the isolates. Reports evaluating the efficacy of these antibiotics against isolates from pigs are rare; however, studies with other domestic species indicate a high degree of susceptibility of C. perfringens isolates to both antibiotics (6,11).

Neomycin, norfloxacin, streptomycin, josamycin, tetracycline, and lincomycin had high MICs, suggesting poor efficacy of these antibiotics against isolates of C. perfringens isolated from piglets. Genes responsible for resistance to tetracycline and lincomycin have previously been described in C. perfringens isolates isolated from humans and domestic animals (23,4,14,11,9,3,7,15). This finding may explain the high resistance ratio found for these two drugs in the present study. As for neomycin, norfloxacin, streptomycin, and josamycin, to our knowledge, no studies regarding the genetic mechanisms of resistance to bacteria of the genus Clostridium have been published. In this work, C. perfringens demonstrated a decreased inherent sensitivity to these drugs, instead of resistance mediated by acquired genes or mutations. It should be emphasized that this study is the first evaluating the MIC for josamycin against C. perfringens isolated from pigs.

When the inhibition profile of isolates from piglets with diarrhea was compared to isolates from animals without diarrhea, the most notable differences included sensitivity toward avilamycin and narasin. The MIC₉₀ of these two drugs was higher in the isolates obtained from diarrheic piglets. For avilamycin and narasin, concentrations of 2 and 8μg/mL, respectively, could inhibit the growth of isolates not associated with diarrhea, whereas a concentration greater than 256μg/mL was required to inhibit the growth of isolates from piglets with diarrhea, a 128- and 32-fold increase in concentration, respectively. These results of avilamycin and narasin might suggest the hypothesis of a possible gene-mediated resistance or mutations present in isolates from diarrheic animals. The inhibition profile of the other antibiotics was not influenced by the clinical status of the piglets. These results suggest that some antibiotics, particularly ionophores, cannot be used for the treatment of diarrhea caused by C. perfringens type A and that the utilization of these drugs should be restricted to growth promotion. When the inhibition profiles of cpb2-positive and cpb2-negative isolates were compared, no significant differences in the MICs were observed.

Antimicrobial sensitivity tests represent an in vitro estimate of the sensitivity or resistance of certain biological agents to a set of drugs. The results of these tests are generally well correlated with therapeutic results. However, there is no guarantee that in vivo treatment will be effective because many factors other than chemotherapy interact with the host organism. This study is the first that demonstrates a striking difference in antimicrobial susceptibility between C. perfringens isolates from piglets with and without diarrhea. These results suggest the need for further studies evaluating molecular factors related to the antimicrobial resistance of C. perfringens.