A Field-Suitable, Semisolid Aerobic Enrichment Medium for Isolation of Campylobacter jejuni in Small Numbers

Abstract

The objective of this study was to produce an economical, easy to prepare, field-suitable enrichment medium for detection of Campylobacter jejuni in small numbers. A semisolid aerobic enrichment medium was developed. Rates of recovery from inoculated medium, sterile swabs, and mixed cultures of C. jejuni and coliform bacteria were tested.

Campylobacter jejuni is a leading food-borne pathogen in humans (11). Isolation of C. jejuni is hampered by its fastidious nature. In particular, C. jejuni may be overgrown by coliform bacteria in mixed cultures. Even in the presence of coliform bacteria, if the samples are heavily contaminated with C. jejuni, direct plating onto selective media is probably adequate for isolation (7). In our laboratory, the sensitivity of direct plating with sterile cotton swabs saturated from the skin of pigeon carcasses, processing equipment, or cloacal swabs on CampyFDA agar medium (Remel, Inc., Lenexa, Kans.) was greater than or equal to 10³ CFU (five trials; data not shown). However, the infective dose of C. jejuni for humans is quite low, as few as 5 × 10² CFU (8), making a more sensitive assay desirable for studies aimed at identifying critical control points in hazard analysis critical control point-based food safety systems or for risk analysis. Numerous transport and enrichment media have been developed and tested for Campylobacter isolation from food, fecal, and environmental specimens (1, 2, 3). Most are labor intensive (4), may require microaerophilic or other specialized incubation conditions (12) or filtration (5), or are not optimal for field sampling because the media are difficult to handle or hard to transport. The objective of this study was to produce an economical, easy to prepare, field-suitable enrichment medium for detection of C. jejuni bacteria in small numbers. A semisolid aerobic enrichment medium was developed that fulfilled these criteria. First, the detection limit of the medium was quantitatively tested by direct inoculation of C. jejuni diluted in phosphate-buffered saline (PBS). Recovery was then determined for medium inoculated with a sterile cotton swab (such as that which would be used for sampling of skin or processing equipment) and from samples that were cocontaminated with Escherichia coli. Intervals of 48 and 72 h of enrichment at 37°C, to determine the effect of holding inoculated medium for up to 24 h before enrichment incubation, and the ease of use of the medium for transport and handling during field sampling were evaluated.

The formulation of the medium was modified from the published work of L. A. Sicinschi (10), who described a broth medium for 24-h enrichment of stool samples, followed by direct plating, and also from that of T. T. Tran (13), who described a blood-free, aerobic selective enrichment broth for food samples. Notably, Shimada and Tsuji (9) successfully developed a semisolid medium (GAM) that enriched and recovered Campylobacter at levels of less than 10 CFU/ml after 48 h of aerobic incubation at 37°C. Our medium was designed to contain a minimum of ingredients without sacrificing the recovery rate described by Shimada and Tsuji (9). The medium consisted of (grams per liter) thioglycolate medium without indicator (29.0), Proteose Peptone (10.0), beef heart for infusion (2.2), H₂KO₄P (2.5), and agar (0.8) (all from Difco, Sparks, Md.). The medium was autoclaved for 15 min at 121°C and allowed to cool in a 37°C water bath. Antimicrobials were filter sterilized with 0.22-μm syringe filters and added to the cooled (still liquid) medium. The antimicrobials tested included the antibiotics (micrograms per milliliter) rifampin (5.3), trimethoprim (5.0), cephalothin (15.0), and polymyxin B (0.3) and the antimycotic amphotericin B (2.0 μg/ml). The final pH of the medium was adjusted to 7.4 ± 0.2 by adding sterile NaOH. Three-milliliter aliquots were pipetted into 5-ml snap-top polypropylene tubes (Falcon; Becton Dickinson, Franklin Lakes, N.J.). Medium was stored at 4°C until use.

C. jejuni ATCC 29248 was used to test recovery from the medium. A field isolate of C. jejuni from chicken skin was utilized for the detection and mixed-culture detection assays. Isolates were frozen on beads (Microbank Beads; ProLab Diagnostics, Austin, Tex.) at −80°C. Prior to each assay, a bead was streaked onto CampyFDA plates and incubated under microaerophilic conditions at 42°C for 48 h. Microaerophilic incubation was accomplished by means of sealed jars and CampyPak (Fisher Scientific, Somerville, N.J.). A field isolate of E. coli originating from chicken litter was used to test recovery from mixed cultures. The E. coli isolate was retrieved from a frozen stock, stored at −80°C, streaked onto blood agar, and incubated for 24 h at 37°C before each assay. To quantify the bacteria used in these trials, standard optical density (OD) curves were established for both C. jejuni and E. coli by preparing known dilutions of bacteria in PBS and obtaining the OD of each dilution. The known dilutions were simultaneously plated (using a Spiral System plater) to determine the number of CFU per milliliter for each dilution at each OD.

First, the recovery of C. jejuni from the medium was tested. C. jejuni was added to sterile PBS to yield a 10⁸-CFU/ml solution as determined by OD (λ = 620 nm). Serial 10-fold dilutions of 10⁸ to 10¹ CFU/ml were prepared. Tubes of medium were inoculated with 0.5 ml of the appropriate dilution by sterile pipette and incubated aerobically at 37°C with tube caps loosely applied. After 48 h of incubation, a 4-mm loop, (2.74 ± 0.5 [standard error of the mean] μg) of the medium was plated onto CampyFDA agar plates and incubated as previously described. Assays were done twice for dilutions of >10³ and four times for dilutions of <10³. Plates were recorded as positive or negative for growth of C. jejuni. There was 100% recovery for dilutions of >10² and 75% recovery for dilutions of 10⁰ and 10¹. C. jejuni was confirmed by typical morphology on Gram staining and oxidase and catalase reactions.

Next, detection limits of the medium for recovery of C. jejuni from sterile cotton swabs were determined. Serial 10-fold dilutions of C. jejuni in sterile PBS were prepared as previously described. A single sterile cotton swab was dipped into each dilution of PBS-C. jejuni and then placed into a tube of the medium. Tubes were incubated aerobically for 48 h at 37°C, and then a 4-mm loopful was plated onto CampyFDA agar and incubated under microaerophilic conditions for 48 h at 42°C. Assays were done four times for dilutions of >10³ and three times for dilutions of <10³. Plates were reported as positive or negative for growth. All assays were positive for the 10⁵ to 10⁷ dilutions. For the 10⁴ dilution, 67% of the plates were positive while there was 100% recovery for the 10³ dilution. There was 25% recovery from the 10² dilution, and no campylobacters were recovered from the 10¹ and 10⁰ dilutions.

Finally, the detection limits of the medium for recovery of C. jejuni from a mixed culture with E. coli were determined. Serial 10-fold dilutions of C. jejuni and E. coli in sterile PBS were prepared as previously described. Tubes of medium were inoculated with 0.25 ml of each bacterium by sterile pipette and incubated aerobically at 37°C with tube caps loosely applied. After 48 h of incubation, a 4-mm loop of the medium was plated onto CampyFDA agar plates and incubated as previously described. Assays were done four times. Plates were reported as positive or negative for growth of C. jejuni (Table 1).

TABLE 1.

Recovery of C. jejuni from enrichment medium in the presence of  E. coli

Dilution of C. jejuni or E. coli in PBS	No. of CFU inoculated/ ml of medium	Assay resulta				% Positive (all assays)
		1	2	3	4
107	2.5 × 106	+	+	+	−	75
106	2.5 × 105	+	+	−	−	50
105	2.5 × 104	+	+	+	+	100
104	2.5 × 103	+	−	+	−	50
103	2.5 × 102	+	+	+	−	75
102	2.5 × 101	+	+	+	−	75
101	2.5 × 100	+	+	−	−	50

^aAfter 48 h of enrichment, 2.74 ± 0.5 μg was plated in each assay. +, positive; −, negative. 

The first part of this trial demonstrated that as few as 1 CFU of C. jejuni per ml could be recovered after aerobic incubation in the medium in four separate assays. Detection rates were 100% for samples of ≥100 CFU/ml and 75% for samples of <10 CFU/ml using a 4-mm loop. Bacterial growth was observed as an opacity in the medium 1 to 2 mm below the surface. In the second part of this trial, the field utility of the medium was challenged by inoculating it with cotton swabs as would be performed with samples from tissues, equipment surfaces, or feces. Recovery rates of 50 to 100% were observed when the medium was inoculated with approximately 1 to 1 million CFU/ml. The variability in these results is probably a function of how the medium was inoculated versus how the medium performed. Variations in cotton swab size, length of time the swab contacted the inoculum, and how many C. jejuni bacteria were picked up on the swab could all have affected the inoculating dose of the medium. This points out the importance of using standardized sampling procedures for field studies to minimize variance between samples. Recovery was demonstrated at all of the dilutions tested. Enrichment periods of 48 and 72 h before plating onto CampyFDA were compared. Recovery rates were higher and more consistent following 48 h of enrichment. The final part of this trial tested the ability of the medium to recover Campylobacter in the presence of coliform bacteria, as would be expected in fecal or environmental samples. Recovery rates ranged from 50 to 100% for all dilutions. Twenty-five to 250 CFU were recovered in 75% of the assays, and ≤3 CFU were recovered in 50% of the assays. The effect of holding the inoculated medium at room temperature (25 to 29°C) for up to 24 h, compared to incubation of paired samples immediately after inoculation, was also tested, and no effect on the Campylobacter recovery rate was found (three trials; data not shown). These results are consistent with the evaluation of C. jejuni held in Cary-Blair medium with reduced agar at 25°C by Luechtefeld et al. (6).

This study demonstrated a novel enrichment medium for Campylobacter isolation that can also serve as a holding or transport medium. The medium is easy to prepare, handle, and transport and requires low labor input following inoculation, compared to other described media (1, 2, 3, 4). No microaerophilic or other specialized incubation conditions or special preparation of samples is required, as with the media described by Tran (12) and Korhonen and Martikainen (5), respectively. Ease of preparation, transport, and field handling is improved over that of broth media (10, 13). The medium was more economical to prepare (approximately $0.015 per tube), due to the reduced number of ingredients, than that of Shimada and Tsuji (9) and had equivalent detection limits.