Rapid and sensitive isolation of Campylobacter jejuni using immunomagnetic separation from patient specimens exposed to oxygen

Abstract

This study describes a method for detecting Campylobacter jejuni in patient stools with subsequent isolation using antibody-magnetic beads in conjunction with selective culture and PCR. Monoclonal antibodies specific for the flagellin A and major outer membrane protein of C. jejuni were generated; two clones (1C7 and 4B2) were used to coat magnetic beads for immunomagnetic separation (IMS). C. jejuni strain NCTC11168 was recovered from human stool samples spiked with varying concentrations (10¹-10⁵ CFU/mL) by Campylobacter (Campy)-IMS or a conventional culture-based method and plated on modified charcoal-cefoperazone-deoxycholate agar; the number of colonies was enumerated. The detection limits of Campy-IMS and conventional culture-based method with spiked stool samples were 10² and 10⁴ CFU/mL, respectively. The sensitivity of IMS-PCR was 10-10,000-fold higher than that of direct PCR. The recovery rate of C. jejuni from spiked stools stored for 12 to 72 h decreased from 72.3 to 5.9% with Campy-IMS and from 48.5 to 0.1% with the conventional culture-based method. Importantly, of 20 PCR (+)/bacterial culture (-) samples that were diagnosed as probable cases according to general criteria, 95% (19/20) were confirmed positive by Campy-IMS. Thus, this study suggests a solution to overcome the problems caused by the inconsistency between probable and confirmed cases of Campylobacter infection.

Importance: The isolation, cultivation, and maintenance of Campylobacter spp. are difficult because of the microaerophilic conditions and specific medium needed. Although selective media are useful for the initial isolation of Campylobacter, subsequent exposure of the sample to oxygen has a detrimental effect on the positive culture rate of Campylobacter, significantly lowering the isolation rate from patient samples. In this study, the detection limit was improved by combining immunomagnetic separation and PCR methods to quickly detect Campylobacter jejuni in clinical patient stool samples using antibody-magnetic beads. Therefore, this study is expected to improve confirmation of C. jejuni infection where diagnosis would previously fail with patient samples because of oxygen exposure, inappropriate diagnostic methods, and interference from other bacteria in the sample.

Keywords: Campylobacter jejuni; confirmed case; detection sensitivity; magnetic bead–antibody complex; monoclonal antibody; probable case; stool specimen.

Fig 1

Schematic of the study workflow.

Fig 2

Performance of different Abs for the capture of NCTC11168. Clones 1C7, 7D8, 4B2, 1C7 + 7D8, 1C7 + 4B2, 7D8 + 4B2, and 1C7 + 7CB + 4B2 were conjugated to MBs. Recovery from samples spiked with high (A, 10⁵ CFU) and low (B, 10³ CFU) concentrations of NCTC11168 was evaluated. NCTC11168 and MAb-MBs were incubated at room temperature for 10 min, and unbound cells were removed by washing. NCTC11168 bound to MBs was resuspended in PBS, and serial dilutions were plated on mCCDA, followed by incubation at 42°C for 48 h. Average measurements with standard deviation from three independent experiments are shown.

Fig 3

Evaluation of Campy-IMS specificity for NCTC11168 in samples containing contaminating pathogenic bacteria [E. coli DH5α, S. Typhimurium, V. parahaemolyticus, S. sonnei, STEC, enterotoxigenic E. coli (ETEC), and enteropathogenic E. coli (EPEC)]. NCTC11168 at 10¹–10⁵ CFU was included as a control. (A–E) NCTC11168 was mixed with other bacteria at the following ratios: 10¹:10⁹ CFU (A), 10²:10⁹ CFU (B), 10³:10⁹ CFU (C), 10⁴:10⁹ CFU (D), and 10⁵:10⁹ CFU (E).

Fig 4

Comparison of recovery rate and detection sensitivity between IMS-PCR and conventional methods. (A) Comparison of recovery rates between Campy-IMS and bacterial culture from five stool samples spiked with NCTC11168 from 10¹ to 10⁵ CFU. Campy-IMS was performed without pre-enrichment, while bacterial culture was performed according to the Korea Centers for Disease Control and Prevention Manual after 18 h of preincubation. Black, Campy-IMS; gray, bacteria culture. **P < 0.01 (unpaired t test). (B) Comparison of the sensitivity of direct and Campy-PCR. Lanes M: 100 bp ladder (Thermo Fisher Scientific); lanes 1–5, direct PCR at 10⁵ to 10¹ CFU/mL; and lanes 6–10, pelleted beads of Campy-IMS from 10⁵ to 10¹ CFU/mL.

Fig 5

Detection of wild-type C. jejuni in spiked stool samples. The recovery of the target strain by Campy-IMS from stool samples spiked with 10⁵ CFU/mL wild-type C. jejuni isolated from patients between 2012 and 2016 was evaluated.

Fig 6

Detection of NCTC11168 in spiked stool samples stored at 4°C. (A) Comparison of Campy-IMS and bacterial culture. Spiked specimens were prepared by mixing NCTC11168 (10⁵ CFU) with stool samples. (A) Stored samples were collected at 0, 12, 24, 48, and 72 h and isolated by the two methods (black, bacterial culture; white, Campy-IMS). *P < 0.05, **P < 0.01 (unpaired t-test). (B) Sensitivity of direct PCR vs. IMS-PCR. Samples for PCR were collected in the same manner as in panel (A). Lane 1, 100 bp ladder (Thermo Fisher Scientific); lane 2, sample stored for 0 h; lane 3, sample stored for 12 h; lane 4, sample stored for 24 h; lane 5, sample stored for 48 h; and lane 6, sample stored for 72 h.

Fig 7

Isolation of Campylobacter jejuni in PCR (+)/bacterial culture (−) human samples by Campy-IMS. Stool specimens were analyzed using conventional culture and PCR. Specimens in which the Campylobacter spp. gene was detected but yielded no cultured bacteria were classified into groups of 48–120 h from the time of specimen collection from the patient. Among these specimens, five were used in the experiment. Samples were designated as 48 (A), 72 (B), 96 (C), and 120 (D) h according to the time elapsed after collection.