Abstract
Infections caused by Campylobacter species pose a severe threat to public health worldwide. However, in Grenada, the occurrence and characteristics of Campylobacter in food animals, including pigs, remain mostly unknown. In this study, we identified the sequence types (STs) of Campylobacter from young healthy pigs in Grenada and compared the results with previous studies in Grenada and other countries. Antimicrobial resistance patterns and diversity of the Campylobacter clones were evaluated. Ninety-nine Campylobacter isolates (97 Campylobacter coli and 2 Campylobacter jejuni) were analyzed by multilocus sequence typing. Eighteen previously reported STs and 13 novel STs were identified. Of the 18 previously reported STs, eight STs (ST-854, −887, −1068, −1096, −1445, −1446, 1556, and −1579) have been associated with human gastroenteritis in different geographical regions. Among these 18 previously reported STs, ST-1428, −1096, −1450, and −1058 predominated and accounted for 18.2%, 14.1%, 11.1%, and 8.1% of all isolates, respectively. Of the 13 novel STs, ST-7675 predominated and accounted for 20% (4 of 20 isolates), followed by ST-7678, −7682, and −7691, each accounting for 10% (2 of 20 isolates). Antimicrobial resistance testing using Epsilometer test revealed a low resistance rate (1–3%) of all C. coli/jejuni STs to all antimicrobials except for tetracycline (1–10.1%). Some of the C. coli STs (13 STs, 24/99 isolates, 24.2%) were resistant to multiple antimicrobials. This is the first report on antimicrobial resistance and multidrug resistance patterns associated with Campylobacter STs recovered from swine in Grenada. This study showed that pigs in Grenada are not major reservoirs for STs of C. coli and C. jejuni that are associated with human gastroenteritis worldwide.
Introduction
C
Molecular epidemiology is required to determine the likely sources of human Campylobacter infections (Wilson et al., 2008). The multilocus sequence typing (MLST) method uses the relative conservation in sequence of seven housekeeping genes in which variations are likely to be selectively neutral (Maiden et al., 1998). MLST analysis has been used to identify Campylobacter clonal complexes (CCs) and STs in mongooses (Miller et al., 2014), poultry (Miller et al., 2010; Stone et al., 2013), and sheep and goats (Stone et al., 2014) in Grenada. The objective of this study was to identify the genetic clones of Campylobacter from young healthy pigs in Grenada and compare the results with previous studies in other animals in Grenada as well as studies from other countries. The antimicrobial resistance patterns and diversity of the Campylobacter clones that infect pigs in Grenada were also evaluated.
Materials and Methods
Sample collection
In a previous study, 172 Campylobacter isolates were obtained from 180 pigs of 6 to 12 weeks of age from 14 pig farms within six parishes of Grenada between May and July 2014 (Matthew-Belmar et al., 2015). The 172 Campylobacter isolates in the previous study were identified using a biochemical method, which resulted in 92 Campylobacter jejuni and 80 Campylobacter coli (Matthew-Belmar et al., 2015). Of the 172 Campylobacter isolates, 99 isolates were viable for culture and DNA extraction and were identified using polymerase chain reaction (PCR), which is a more reliable method of identifying Campylobacter species.
Culture, DNA extraction, and PCR-based identification of viable Campylobacter isolates
Ninety-nine viable isolates were inoculated into campylobacter blood-free selective agar (mCCDA) medium (Oxoid Ltd., Basingstoke, England) and incubated microaerobically using Campy GasPak (BBL Becton Dickson and Co., Cockeysville, MD) at 42°C for 48 h to resuscitate the isolates for DNA extraction. DNA was extracted using the Qiagen Dneasy Kit following the manufacturer's instructions (Qiagen Sciences, MD). The extracted DNA from the isolates was shipped to the Veterinary Preventive Medicine Department, Ohio State University, OH, for MLST analysis. For identification of Campylobacter species, multiplex polymerase chain reaction was used as described by Denis et al. (1999).
MLST of Campylobacter isolates
To ascertain the genotypic relationship of Campylobacter isolates and to assess similarity to strains associated with human infections, 99 isolates (97 C. coli and 2 C. jejuni) were analyzed by MLST as described previously (Dingle et al., 2001; Sanad et al., 2011; Kashoma et al., 2014).
Antimicrobial susceptibility testing of Campylobacter isolates
The viable 99 Campylobacter isolates were tested for susceptibility to seven antimicrobials: ampicillin, tetracycline, erythromycin, ciprofloxacin, gentamicin, chloramphenicol, and metronidazole (AB–Biodisk, Solna, Sweden) by determining the minimum inhibitory concentration (MIC) using the Epsilometer test (E-test) strips (AB Biodisk, Solna, Sweden). The E-test was performed according to the manufacturer's instructions on Mueller–Hinton agar (Remel) with 5% sheep blood. C. jejuni (ATCC 33291) susceptible to all the tested antimicrobials and giving reproducible MICs was used as control (Udayamputhoor et al., 2003). The MIC of a drug was read directly from the scale printed on the E-test strip at the point of intersection between the bacterial growth zone and the strip. The breakpoints established by the Clinical and Laboratory Standards Institute (CLSI, 2016) for Campylobacter species were used in the present study, and the MIC values used to classify a strain as resistant were ciprofloxacin, ≥4 μg/mL; erythromycin, ≥32 μg/mL; and tetracycline, ≥16 μg/mL. The National Antimicrobial Resistance Monitoring System–Enteric Bacteria (NARMS) (NARMS, 2012) breakpoints were used when CLSI breakpoints were not available. For chloramphenicol and gentamicin, the breakpoint (≥32 and ≥8 μg/mL, respectively) was used (NARMS, 2012). For ampicillin, the breakpoint (≥32 μg/mL) suggested by Guevremont et al. (2006) was used. The breakpoint for resistance to metronidazole was set as ≥16 μg/mL, as recommended by Lorian (1991).
Statistical analysis
The differences in the number of Campylobacter isolates showing resistance to multiple antimicrobials and those showing resistance to single antimicrobial as well as the number of isolates showing resistance to the different antimicrobial in this study and previous studies in Grenada were compared using chi-squared (χ2) (contingency table) analysis created in data analysis plus available in Microsoft Excel 2010 program. A value of p < 0.05 was considered statistically significant.
Results
Of the 99 viable Campylobacter isolates, 98% (97 of 99) were identified as C. coli, while only 2% (2 of 99) were identified as C. jejuni as per PCR result obtained during MLST procedure.
Table 1 shows the allelic profiles and CCs from the 99 Campylobacter isolates analyzed in this study. A total of 18 previously reported STs were generated from the 99 Campylobacter isolates (97 C. coli and 2 C. jejuni) with eight STs occurring singly and ST-1428 being the most common (Table 1). Of the 18 previously reported STs identified in this study, ST-1428, −1096, −1450, and −1058 predominated and accounted for 16.2%, 14.1%, 11.1%, and 8.1% of all isolates, respectively (Table 1). The common sources and countries where the Campylobacter clones obtained in this study have been previously identified are presented in Table 1.
Information in the PubMLST database retrieved January 29, 2016 (Campylobacter PubMLST, 2010).
Place(s) where the Campylobacter strains have been recorded.
Clonal complex not assigned as yet.
CCs and novel sequence types: presented in Table 2.
CC, clonal complex.
Based on the allelic profiles of the housekeeping genes, 13 novel STs were identified and represent 20 of the 99 isolates (Table 2). Five of the 13 novel STs (STs-7678, −7680, −7686, −7687, and −7688) from this study were assigned to the ST-828 CC. The other eight STs (STs-7673, −7675, −7677, −7681, −7682, −7684, −7685, -and −7691) have not yet been assigned to a CC (Table 2). The only two C. jejuni isolates typed in this study belonged to ST-7673. Of the 13 novel STs, ST-7675 predominated and accounted for 20% of the novel isolates, followed by ST-7673, −7678, −7682, and −7691, which accounted for 2 of the 20 novel isolates each. The remaining eight novel STs (ST-7677, −7680, −7681, −7684, −7685, −7686, −7687, and −7688) represented one of the 20 novel isolates each (Table 2).
Information in the PubMLST database retrieved January 29, 2016 (Campylobacter PubMLST, 2010).
Clonal complex not assigned as yet.
Sequence type of the two C. jejuni isolates.
Table 3 shows the antimicrobial resistance patterns of the C. coli/jejuni STs recovered from pigs in Grenada. There was low resistance rate (1–3%) for all the C. coli STs to all the tested antimicrobials except for tetracycline (1–10.1%). The C. coli/jejuni STs isolated from this study were all susceptible to ciprofloxacin and only two STs, ST-1096 (1%) and ST-1428 (1%), showed resistance to chloramphenicol and erythromycin, respectively (Table 3). The number of isolates showing resistance to tetracycline was significantly higher compared with those showing resistance to the other antimicrobials (p < 0.001). Some of the C. coli STs (13 STs, 24/99 isolates, 24.2%) were resistant to multiple antimicrobial drugs (MDR to two or more antimicrobial drugs of different classes), while some (13 STs, 42/99 isolates, 42.4%) showed resistance to a single antimicrobial drug. However, four STs (ST-1177, −7680, −7684, and −7687) as well as the two C. jejuni (belonging to ST-7673) were susceptible to all the tested antimicrobials (Table 3). Isolates showing MDR were significantly fewer than those showing resistance to one antimicrobial drug (p = 0.007). Resistance to tetracycline (61.6%) was the most common resistance observed in this study, followed by metronidazole (16.2%) and ampicillin (13.1%), therefore MDR pattern observed was mainly to tetracycline and other antimicrobial(s). The most common MDR pattern observed was to ampicillin/tetracycline, followed by metronidazole/tetracycline. One isolate, ST-1428, showed MDR to four antimicrobials (ampicillin/erythromycin/metronidazole/tetracycline) and two other isolates belonging to ST-1096 and −7678 showed MDR to three antimicrobials (gentamicin/metronidazole/tetracycline and ampicillin/metronidazole/tetracycline, respectively) (Table 3).
Clonal complex not assigned as yet.
Sequence type of the two C. jejuni isolates.
AM, ampicillin; CL, chloramphenicol; EM, erythromycin, GM, gentamicin; MDR, multidrug resistance (resistance to two or more antimicrobials of different classes); MZH, metronidazole; TC, tetracycline.
Discussion
The MLST results highlight the Campylobacter genotypes found in pigs in Grenada and reveal that certain pig-associated isolates are unique to Grenada pigs. In this study, most of the isolates belonged to ST-828 CC, which is mainly associated with previously reported isolates from agricultural and environmental sources and human clinical cases (Sheppard et al., 2010b). In Grenada, both Stone et al. (2013) and Miller et al. (2010) identified Campylobacter clones that belonged to the ST-828 CC in poultry; however, none of the STs in this present study was detected in their studies, nor in other animals, including goats and sheep (Stone et al., 2014). Furthermore, researchers in other geographic areas have reported the presence of STs in the ST-828 CC in human, poultry, swine, and cattle (Sanad et al., 2011; Kashoma et al., 2014). There is no published information on STs of human isolates in Grenada. A U.K. study (Roux et al., 2013) indicated that sheep and chicken C. coli STs were most frequently found in humans, while those from pigs were rarer. In Denmark, only 10% of the isolates from pigs shared STs with isolates from humans, and these shared STs were found in poultry isolates as well (Litrup et al., 2007). Overall, the results of this study are in agreement with previous documentation that host association of Campylobacter genotypes transcends geographic variation (Sheppard et al., 2010a).
Unlike developed countries, Grenada is a small developing island where few individuals own private pig farms and rear pigs in open area for consumption and commercial purposes. The farms are highly disorganized and are in close proximity to humans and livestock. The possibility of intermingling between the pigs and other farm animals exist. Due to the unsystematic distribution of pig farms in Grenada, the relationship between genotype and farms was not determined.
The 18 previously reported STs detected in this study have been identified in swine and other animals as well as humans. ST-854 (331) represents the highest number reported to PubMLST and has been associated with multiple sources, including humans, poultry, ruminants, swine, environment, and other unspecified sources. ST-1068 (171) also represents a high number of reports and has been identified in many sources with ruminants being the highest animal reservoir source (Table 1). ST-1096 (76) has been recovered from numerous sources with swine being the most common animal source, but has not been seen in ruminants. Seven of the STs (ST-1124, −1177, −1428, −4606, −4939, −5746, and −7289) have only been associated with swine as a host species, while ST-1464 has been associated with swine and other unspecified sources (Table 1) (Campylobacter PubMLST, 2010).
Eight STs (ST-854, −887, −1068, −1096, −1445, −1446, 1556, and −1579) of the 18 previously reported C. coli/jejuni STs recovered in this study have been associated with human gastroenteritis in different geographical regions (Table 1). ST-854, −887, −1096, −1445, and −1579 accounted for 1.3–25% sporadic human gastroenteritis; ST-854, −887, −1068, −1096, −1446, and −1556 accounted for 1.2–12.5% unspecified human infection, while ST-854 accounted for 0.3% hospital inpatient infection (Table 1) (Campylobacter PubMLST, 2010).
Based on the results from our antimicrobial susceptibility analysis, resistance of the Campylobacter STs to tetracycline (1–10.1%) was predominant and the most common MDR pattern observed was to tetracycline/metronidazole or tetracycline/ampicillin. The high frequency of tetracycline resistance (61.6%) observed in this study differed significantly (p < 0.001) from the resistance to tetracycline (19.6%, 10/51) observed in poultry by Stone et al. (2013). Another study on sheep and goats in Grenada by Stone et al. (2014) revealed resistance of Campylobacter isolates to tetracycline (30.8%, 4/13), which is also significantly lower (p = 0.005) than the observation in this study. Other studies in different geographical areas have also shown a low resistance rate of Campylobacter isolates from different animals and humans to tetracycline in comparison with our observation (Moore et al., 2006; Kashoma et al., 2015). On the other hand, other studies have reported high resistance rates of Campylobacter isolates from animals and humans to tetracycline, which is in close agreement with the observation in this study (Hariharan et al., 1990; Rollo et al., 2010; Scott et al., 2012; Stone et al., 2014).
Several studies have shown that tetracycline resistance is common among a variety of bacteria from different sources in Grenada (Sylvester et al., 2014; Amadi et al., 2015a, c, b, d; Farmer et al., 2016). Tetracycline resistance is high in Grenada. It is noteworthy that chlortetracycline is routinely used as a feed additive for pigs in Grenada. Furthermore, oxytetracycline is used for treatment of pigs for bacterial infections (Sabarinath et al., 2011). There is no published information on tetracycline-resistant Campylobacter from pigs in other Caribbean islands. Tetracycline resistance was 100% in pig Campylobacter in Brazil (Biasi et al., 2011). On the other hand, of seven C. coli isolates from pigs, five were resistant to erythromycin, but sensitive to tetracycline (Harrow et al., 2004). The general use of tetracycline has contributed to emergence of resistant bacteria in the environment and clinical sources (Balsalobre et al., 2011).
In conclusion, the present study documents that pigs in Grenada are not major reservoirs for STs of C. coli and C. jejuni that are associated with human gastroenteritis worldwide. This study revealed that porcine of Grenada origin harbor novel STs that have not been reported in humans or animals worldwide. The resistance rate to drugs other than tetracycline was low. Resistance to erythromycin was very low, but requires continuous monitoring to determine the risk factor for the emergence of erythromycin-resistant Campylobacter strains. This is the first report that documents the molecular typing of Campylobacter species and resistance and MDR patterns associated with Campylobacter STs recovered from swine in Grenada.
Footnotes
Acknowledgments
The study had the approval of St. George's University Institutional Animal Care and Use Committee (IACUC-15002-R). The authors are thankful to St. George's University for providing funds for this research.
Disclosure Statement
No competing financial interests exist.
