- Email: [email protected]
Development of a PCR ELISA assay for the identification of Campylobacter jejuni and Campylobacter coli
Molecular and Cellular Probes (2001) 15, 291–300 doi:10.1006/mcpr.2001.0374, available online at http://www.idealibrary.com on
Development of a PCR ELISA assay for the identification of Campylobacter jejuni and Campylobacter coli A. D. Sails,∗1 A. J. Fox,2 F. J. Bolton,1 D. R. A. Wareing,1 D. L. A. Greenway3 and R. Borrow2 1
Preston Public Health Laboratory, Royal Preston Hospital, PO Box 202, Sharoe Green Lane, Fulwood, Preston, Lancs., PR2 9HG, UK; 2Manchester Public Health Laboratory, Withington Hospital, Nell Lane, West Didsbury, Manchester, M20 2LR, UK and 3 The Department of Biological Sciences, The University of Central Lancashire, Preston, PR1 2HE, UK; ∗Current address: National Center for Infectious Diseases, Centers For Disease Control and Prevention, 1600 Clifton Road, Atlanta, Georgia, 30333, USA. (Received 20 February 2001, Accepted 20 June 2001) A polymerase chain reaction (PCR) assay was developed based on a solution-hybridization colorimetric end-point detection format (PCR ELISA) for the identification of Campylobacter jejuni and Campylobacter coli. PCR primers were designed to target a gene sequence with speciesspecific motifs. Five biotin-labelled probes targeted to the species-specific motifs were investigated for the detection of digoxygenin-labelled PCR products from C. jejuni and C. coli using the PCR ELISA format. Two probes were identified, one which reacts with both the C. jejuni and C. coli target sequences (probe CC2) and one probe which reacts with the C. jejuni target sequence only (probe CJ2). The specificity of the assay with the CJ2 and CC2 probes was investigated with a range of Campylobacter spp., Arcobacter spp., Helicobacter spp. and a range of unrelated organisms. The PCR ELISA assay and probes were demonstrated to be specific for C. jejuni and C. coli. The sensitivity of the PCR ELISA assay was demonstrated to be 10–100-fold more sensitive than a gel-based PCR method using the same primers. This PCR ELISA assay is sensitive, specific and significantly reduces the time needed for the identification of C. jejuni and C. coli and has the potential to facilitate early detection of these important gastro-intestinal pathogens. 2001 Academic Press KEYWORDS: Campylobacter, PCR, ELISA, identification, detection.
INTRODUCTION Campylobacter is the most frequent bacterial cause of gastro-enteritis in the UK and the rest of the developed world.1 Campylobacter jejuni is the most common species associated with gastro-enteritis but Campylobacter coli accounts for approximately 5% of infections.1 Campylobacters are slow-growing fastidious organisms, which are asaccharolytic and
generally biochemically unreactive, unlike other enteric pathogens such as Salmonella and Shigella.2 Many of the phenotypic tests used for the routine identification of members of the Enterobacteriaceae do not identify campylobacters. Many diagnostic microbiology laboratories consequently fail to identify campylobacters to the level of species and report isolates as Campylobacter spp.3 Identification to the level of genus is based on colony morphology, Gram
Author to whom all correspondence should be addressed: Prof. Andrew Fox, Manchester Public Health Laboratory, Withington Hospital, Nell Lane, West Didsbury, Manchester, M20 2LR, UK. Tel: 44 161 291 4631; Fax: 44 161 446 2180; E-mail: [email protected]
0890–8508/01/050291+10 $35.00/0
2001 Academic Press
292
A. D. Sails et al.
stain, growth only in a micro-aerobic atmosphere and oxidase tests4 and more extensive identification or biotyping schemes have been described.4–6 The polymerase chain reaction (PCR) is a rapid method for the detection and identification of bacterial pathogens.7 Several assays have been described for the identification of campylobacters based on the amplification of 16 Svedberg units (S)8,9 and 23S rRNA gene sequences.10 Other gene targets reported for the identification of campylobacters by PCR include the ceuE gene,11 the mapA gene,12 the GTPase gene13 and the flagellin A (flaA) gene.14 Jackson and colleagues15 described a novel PCR assay for the detection of Campylobacter species, which targeted a 256 bp region of an open reading frame (ORF) adjacent to and downstream from a novel two-component regulatory gene. The assay produced an amplicon from C. jejuni, C. coli and Campylobacter upsaliensis which when digested with the restriction enzymes Alu I, Dde I and Dra I yielded species-specific restriction fragment length polymorphism (RFLP) profiles. Differentiation was achieved between C. jejuni and C. coli because of the presence of a Dra I restriction site in the C. jejuni amplification product. Many of the PCR assays described for the identification of campylobacters use agarose gel electrophoresis for end-point detection, but this method lacks sensitivity, specificity, and can be subjective in its interpretation.16 The sensitivity and specificity can be improved by transferring the PCR products to a nylon membrane and detection by hybridization with a labeled probe.17 However the extra steps are labour intensive and the standardization of the procedure between laboratories is difficult. Adaptation of PCR assays into a solution hybridization colorimetric end-point detection format (PCR ELISA) facilitates specific and sensitive detection of PCR amplification products with the potential for increased throughput.18 Non-radioactive digoxygenin (DIG) labels are incorporated into the PCR product during the PCR amplification and the detection of labelled PCR products is based on biotinylated capture probes complementary to specific sequences in the target gene.18 The aim of this study was to develop a probe-based PCR ELISA assay to detect speciesspecific motifs within a previously described gene sequence (ORF-C)15 for the detection and identification of C. jejuni and C. coli.
Table 1. Non-Campylobacter organisms used in the specificity studies. Organism
Aeromonas hydrophila Bacillus cereus Bacillus subtilis Clostridium perfringens Escherichia coli Escherichia coli 0157 non toxigenic Escherichia coli 0157 VTEC Lactobacillus casei Lactococcus lactis Listeria monocytogenes Pseudomonas aeruginosa Salmonella enteritidis Salmonella typhimurium Shigella sonnei Enterococcus faecalis Staphylococcus aureus Coagulase negative staphylococcus Serratia marcescens Vibrio parahaemolyticus Vibrio alginolyticus Vibrio cholerae non 0:1 Yersinia enterocolitica Proteus mirabilis Proteus rettgeri Citrobacter freundii Legionella pneumophila Acinetobacter lwoffii Enterobacter aerogenes Enterobacter cloacae Proteus rettgeri Klebsiella pneumoniae
NCTCa reference number 8049 7464 10400 8237 10418 12900 12079 10302 662 11994 10662 400452/97b 12023 8574 775 6571 170454/97b 11935 10885 12160 11348 10460 10975 7475 9750 12821 5866 10006 11936 7475 350324/97
a
NCTC: National Collection of Type Cultures, PHLS, UK Clinical isolates obtained from Preston Public Health Laboratory. b
containing 15% (v/v) glycerol (BDH Ltd, Poole, UK). Campylobacter spp. and Arcobacter spp. recovered from −70°C storage were cultured on Columbia blood agar (BA) (CM 331, Oxoid Ltd, UK) containing 5% (v/v) whole horse blood. Helicobacter spp. recovered from −70°C storage were cultured on chocolate agar (heated Columbia blood agar, CA). All Campylobacter, Helicobacter and Arcobacter isolates were incubated micro-aerobically at 37°C for 48–72 h. All other organisms were cultured on BA either aerobically or anaerobically at 37°C for 48 h prior to testing. The organisms used in the study are listed in Tables 1, 3 and 4.
MATERIALS AND METHODS Bacterial isolates and culture conditions
Extraction of DNA from bacterial cultures
Bacterial isolates were stored at −70°C in Brain Heart Infusion broth (CM 225, Oxoid Ltd, Basingstoke, UK)
DNA was extracted from bacterial cultures on BA or CA using the Isoquick DNA extraction method
Campylobacter PCR ELISA
293
Fig. 1. C. jejuni, C. coli and C. upsaliensis ORF-C gene sequences and PCR ELISA probes.
according to the manufacturer’s instructions (Orca Research Ltd, USA) and the purified DNA was quantified spectroscopically using a GeneQuant II DNA calculator (Amersham-Pharmacia Biotech Ltd, Amersham, UK).
PCR primers and probes PCR primers were designed using Primerselect software (DNASTAR Inc., Madison, USA) to amplify a 184 bp target sequence in the ORF-C gene sequence identified by Jackson and colleagues15 containing species-specific motifs. Probes were designed to hybridize to the C. jejuni and C. coli amplification products (Fig. 1) in the species-specific regions of the gene sequence. Probes CC1, CC2 and CC3 were designed to target the C. coli sequence and the CJ1 and CJ2 probes were designed to target the C. jejuni sequence. The forward and reverse (ORF-C) primer sequences were as follows: ORF-CF: 5′GTTTGCGACCATATTGTTTGGCATTGTAG3′ ORF-CR: 5′TAATACCGTTGCCCAAATCCCTGAAAGC3′. Primers and probes were synthesized by Oswell DNA Service, UK, and were purified by high performance liquid chromatography (HPLC) with the probes being 5′ biotinylated.
PCR amplification and DIG labelling The primer and magnesium chloride concentrations were optimized and PCR labelling reactions were performed using the PCR ELISA DIG Labellingplus kit (Roche Diagnostics Ltd, Lewes, UK) according to manufacturer’s instructions. PCR amplification was
performed in a 50 l reaction volume containing: molecular biology grade water, PCR reaction buffer (final concentration 2 mM Tris–HCl (pH 8·4), 5 mM KCl), 3 mM final concentration magnesium chloride, PCR ELISA DIG labelling Mixplus containing final concentrations of 200 M of each of dATP, dCTP, dGTP, 570 M dUTP, 30 M DIG-UTP, primers ORF-CF and ORF-CR (final concentration 0·5 M), Platinum Taq DNA polymerase (final concentration 2.5 U) (Life Technologies Ltd, Paisley, UK), uracil DNA glycosylase (final concentration 2 U), and 5 l of template DNA. Thermal cycling parameters were as follows: 95°C for 2 min; followed by 30 cycles of 94°C for 25 s, 55°C for 40 s, 72°C for 60 s; and a terminal extension step of 72°C for 5 min. Negative controls (molecular biology grade water) were included in each set of reactions.
Detection of DIG labelled PCR products by ELISA The PCR ELISA (DIG Detection kit, Roche Diagnostics Ltd, Lewes, UK) was used according to the manufacturer’s instructions. A negative detection control (molecular biology grade water) was included with each set of reactions. DIG labelled PCR products were denatured in 100 mM NaOH and 300 mM NaCl and hybridization buffer containing probe at a concentration of 50 ng/ml was added. Aliquots (200 l) were then transferred to the wells of a streptavidincoated microtitre plate strip, and the plate was incubated at 39°C in a shaking ELISA incubator (IEMS Thermomix, Labsystems Ltd, UK) for 90 min. The microtitre plate wells were then washed and antidigoxigenin peroxidase conjugate (2 mU diluted in
294
A. D. Sails et al.
conjugate buffer 200 l) was added to each well and the microtitre plate was incubated at 37°C with shaking for 30 min. Following washing ABTS (2, 2azino-di-3-ethylbenzithiazoline sulfonate) was added (200 l of a 1 mg/ml solution in substrate buffer) and the plate was incubated at 37°C with shaking for 30 min to allow colour development. The absorbance of the contents of each well was read at 405 nm (492 nm reference) and the extinction of the negative control was subtracted from each sample. Positive reactions were determined by a calculation of cut-off values as follows: cut-off=2×absorbance at 405 nm of the mean of ten replicates of the negative detection control (water).19
Detection of DIG labelled PCR products from C. jejuni, C. coli and C. upsaliensis using the novel PCR ELISA probes DNA was extracted from fresh overnight cultures of C. jejuni (NCTC11168), C. coli (NCTC11366) and C. upsaliensis (NCTC11540) and the DNA concentration was determined spectrophotometrically as described above. DNA template dilutions were prepared in molecular biology grade water to give 15 pg, 1·5 pg, 150 fg, 15 fg and 1·5 fg per 5 l aliquot. This range of DNA template concentrations were used in the PCR ELISA reaction and the labelled products detected using the five probes as described above.
Investigation of the specificity of the PCR ELISA assay with the CJ2 and CC2 probes
with the labelled products being detected using the CJ2 probe and PCR ELISA method as described above.
RESULTS Detection of DIG labelled PCR products from C. jejuni, C. coli and C. upsaliensis using the novel PCR ELISA probes Five oligonucleotide probes were investigated for the detection of labelled PCR products from C. jejuni, C. coli and C. upsaliensis using the PCR ELISA method. The comparative specificity of the different probes for the detection of C. jejuni, C. coli and C. upsaliensis DIG labelled PCR products is summarized in Table 2. Positive results were determined by calculating the cut off value as described above and were visible by a green coloration. The mean absorbance at 405 nm of 10 replicate negative detection controls was 0.125, to give a negative cut off value of 0.25. The CJ1 probe hybridized with the PCR products from C. jejuni and C. coli but not C. upsaliensis. The CJ2 probe hybridized with the C. jejuni PCR product only, demonstrating its specificity for this species. The CC1 probe hybridized to the PCR products from all three species however the CC2 and CC3 probes hybridized to the PCR products from C. jejuni and C. coli only.
Investigation of the specificity of the PCR ELISA assay with the CJ2 and CC2 probes
DNA was extracted from overnight cultures of a range of Campylobacter, Arcobacter, Helicobacter species and unrelated genera (Tables 1, 3 and 4) and the DNA was quantified spectrophotometrically as described above. The purified DNA (10 ng per reaction) was used as a template in the PCR ELISA and the labelled PCR products were detected using the CJ2 and CC2 probes as described above.
The PCR ELISA results for isolates of Campylobacter species and related organisms are summarized in Tables 3 and 4. The PCR ELISA assay with the CJ2 probe produced a positive reaction with all of the isolates of C. jejuni but not with any other organisms. The CC2 probe produced a positive signal with all isolates of C. coli and the closely related species Clostridium hyoilei. The CC2 probe also produced a positive signal with 15 of the 21 C. jejuni isolates tested. The CC2 probe did not produce a positive signal with any of the other organisms tested.
Comparison of the sensitivity of the PCR ELISA assay and gel-based detection of PCR products using varying amounts of template DNA
Comparison of the sensitivity of the PCR ELISA assay and gel-based detection of PCR products using varying amounts of template DNA
DNA extracted from C. jejuni (NCTC11168) was diluted in molecular biology grade water to give a range of DNA concentrations from 1 ng to 0.1 fg DNA. The DNA dilutions were then used as template in the PCR ELISA assay and a gel-based PCR assay
The smallest amount of C. jejuni template DNA giving a positive signal in the PCR ELISA assay with the CJ2 probe was 1·5 fg (Fig. 2). The smallest amount of C. jejuni template DNA which produced a clearly defined band in the conventional PCR assay using gel
Campylobacter PCR ELISA
295
Table 2. Polymerase chain reaction (PCR) ELISAa detection of digoxygenin (DIG) labelled PCR products from Campylobacter jejuni, Campylobacter coli and Campylobacter upsaliensis with the five probes. Campylobacter species
C. jejuni (NCTC11168)
C. coli (NCTC11366)
C. upsaliensis (NCTC11540)
a
DNA template level
15 pg 1·5 pg 150 fg 15 fg 1·5 fg 15 pg 1·5 pg 150 fg 15 fg 1·5 fg 15 pg 1·5 pg 150 fg 15 fg 1·5 fg
PCR ELISA end-point detection (Absorbance at 405 nm for each probe) CJ1
CJ2
CC1
CC2
CC3
2·528 2·794 2·381 −0·001 −0·001 0·413 0·423 0·566 −0·003 −0·003 0·000 −0·007 −0·002 0·004 0·000
>3·0 >3·0 >3·0 >3·0 0·604 −0·010 −0·008 −0·007 −0·006 −0·009 −0·009 −0·010 −0·008 −0·007 −0·001
2·395 2·481 3·071 −0·008 −0·008 2·812 2·942 2·291 −0·008 −0·007 2·278 2·874 2·555 2·687 −0·012
>3·0 >3·0 1·562 −0·016 −0·018 2·762 2·510 2·665 −0·030 −0·012 0·059 0·045 0·051 0·010 −0·013
>3·0 2·905 2·979 1·620 0·013 2·526 2·836 2·631 0·009 0·007 0·055 0·055 0·019 0·019 0·012
ELISA: enzyme-linked immunoabsorbent assay.
electrophoresis and ethidium bromide staining to detect the PCR products was 150 fg DNA (Fig. 2). A faint band was visible following gel electrophoresis of the PCR product from 15 fg of template DNA, but it was poorly defined. The PCR ELISA was therefore 10–100-fold more sensitive when compared with gelbased PCR detection.
DISCUSSION There are relatively few biochemical reactions that are useful for the identification of Campylobacter spp. Useful laboratory biochemical tests for the identification of Campylobacter spp. include catalase production, hippurate hydrolysis, hydrogen sulphide production, indoxyl acetate hydrolysis, nitrate reduction and urease production and growth temperature.4 The hippurate hydrolysis test is critical to distinguish between C. jejuni and C. coli isolates4 with C. jejuni being positive in the test. However, isolates of C. jejuni have been reported that do not express this phenotypic trait2 making them more difficult to identify unambiguously using phenotypic methods. This has led to the development of genetic methods for identification. PCR assays for the identification and speciation of C. jejuni and C. coli have been described.11,14,15,20,21 Many of these assays utilized gel electrophoresis as the end-point detection method, which can be subjective in its interpretation.16 Adapting PCR assays into solution hybridization format with colorimetric
end-point detection increases both specificity and sensitivity.20,21 In this study a PCR ELISA was investigated for the detection and identification of C. jejuni and C. coli. The primers described here were demonstrated to produce amplicons of the correct size from C. jejuni, C. coli and C. upsaliensis (data not presented). To produce an assay specific for C. jejuni and C. coli, novel probes were designed to hybridize to speciesspecific sequences within the DIG labelled PCR products from these species, which could then be detected using a commercial ELISA kit. The PCR ELISA method allows the differentiation between specific PCR products via the hybridization of the biotin-labelled probes.18 The specificity of hybridization is dependent on a number of factors including the length and GC content of the capture probe, and the temperature of hybridization. Increasing the temperature of hybridization increases specificity, although sensitivity may be reduced. Short capture probes of 18–20 nucleotides and elevated hybridization temperatures (up to 55°C) can identify a single base mismatch in the probe target sequence. In order to maximize the melting temperature (Tm) of the hybridized probe and target oligonucleotide duplex, probes were designed to be 33–36 bp long to give a Tm of approximately 60°C. Two probes CJ1 and CJ2 were designed to target the C. jejuni sequence, CJ1 was identical to the target sequence, but the second probe (CJ2) included two nucleotide mismatches in an attempt to increase the specificity of the hybridization to the C. jejuni PCR product. The
296
A. D. Sails et al. Table 3. Identification of Clostridium jejuni and Clostridium coli by polymerase chain reaction (PCR) ELISA.e Organism
C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C.
jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni coli coli coli coli coli coli coli coli
Sourcea
NCTC NCTC PH PH PP PB PB PB PB PB PC PC PC PC PC PC PL PL PM PS PS NCTC PC PC PL PL PP PP PP
Reference number
11168 11392 3212/91 3748/91 2380/90 451042/98 87881/95 87882/95 87880/95 87878/95 450359/96 450821/98 450286/96 450254/96 450361/96 88086/95 450888/98 450749/98 450098/98 77124/94 87024/95 11366 450788/98 450733/98 450761/98 450862/98 87209/95 87207/95 87208/95
Serotypeb
2 6 2 4 23 50 1 1 10 13 1 2 5 6 11 38,57 1 2 NDd ND ND 4 28 56 56 56 49 53 53
PCR ELISA resultc CJ2
CC2
+ + + + + + + + + + + + + + + + + + + + + − − − − − − − −
+ + + + + − − − − − + − + + + + + + + + + + + + + + + + +
a
Source: NCTC: National Collection of Type Cultures; all other isolates were obtained from the Campylobacter Collaborating Unit, Preston Public Health Laboratory, UK; PH: Human isolate; PP: Porcine isolate; PB: Bovine isolate; PC: Chicken isolate; PL: lamb isolate; PM: Milk isolate; PS: Sand isolate; b Penner Heat Stable serotype; c PCR ELISA negative cut off value was 0·25; d ND: Not determined. e ELISA: enzyme-linked immunoabsorbent assay.
stability of oligonucleotide duplexes is dependent on the oligonucleotides involved22 with one mismatch destabilizing a duplex slightly (stable mismatch: G-T, G-A) or significantly (unstable mismatch: A-A, T-T, C-T, and C-A). Additionally, the position of the mismatch can influence the melting point of the duplex. The CJ1 probe hybridized to both C. jejuni and C. coli amplicons in the PCR ELISA assay. The incorporation of a T-T and a C-A unstable mismatch in the CJ2 probe did not prevent hybridization with the C. jejuni DIG-labelled amplicon. The destabilizing effect of the two mismatches in the duplex may have been offset by the length of the oligonucleotide probe and/or the low G-C content. However, the mismatches incorporated in the CJ2 probe did destabilize the duplex formed between the probe and the C. coli
target sequence making the CJ2 probe specific for the identification of C. jejuni. Three probes were designed to hybridize with the C. coli amplicons with the CC1 probe being identical to the target sequence in the C. coli PCR product. The CC2 and CC3 probes incorporated two and three nucleotide mismatches respectively. The CC1 probe was six nucleotides different from the C. jejuni target sequence and the CC2 and CC3 probes were 8 nucleotides different from the C. jejuni target sequence. The CC1 probe hybridized to the amplicons from C. jejuni, C. coli and C. upsaliensis to form a stable duplex and therefore produced a detectable absorbance in the assay. The CC2 and CC3 probes hybridized to both the C. jejuni and C. coli DIG-labelled amplicons to form a stable duplex
Campylobacter PCR ELISA
297
Table 4. Identification of Campylobacter, Helicobacter and Arcobacter isolates by polymerase chain reaction (PCR) ELISA.d Organism
Campylobacter lari C. lari C. lari Campylobacter lari (UPTC)c C. lari (UPTC) C. lari (UPTC) Campylobacter upsaliensis Campylobacter helveticus Campylobacter fetus subsp. fetus C. fetus subsp. fetus C. fetus subsp. fetus Campylobacter fetus subsp. venerealis Campylobacter hyointestinalis C. hyointestinalis C. hyointestinalis Campylobacter mucosalis Campylobacter sputorum subsp. bubulus Campylobacter sputorum biovar faecalis Campylobacter hyoilei Helicobacter cinaedi Arcobacter butzleri A. butzleri A. butzleri A. butzleri Arcobacter skirowii
Sourcea
NCTC PSF PSF NCTC PSF PSF NCTC NCTC NCTC PB PL NCTC NCTC PB PB NCTC NCTC NCTC NCTC NCTC NCTC PC PC PC NCTC
Reference number
11352 77779/94 77715/94 11928 77625/94 77716/94 11540 12479 5850 450971/98 451145/98 10354 11609 DC54C DC89C 11000 11367 11416 12884 12423 12481 450427/98 450555/98 450556/98 12713
PCR ELISAb CJ2
CC2
− − − − − − − − − − − − − − − − − − − − − − − − −
− − − − − − − − − − − − − − − − − − + − − − − − −
a Source: NCTC: National Collection of Type Cultures; all other isolates were obtained from the Campylobacter Collaborating Unit, Preston Public Health Laboratory, UK; PSF: Shellfish isolate; PB: Bovine isolate; PL: lamb isolate; PC: Chicken isolate; b PCR ELISA negative cut off value was 0·25. c UPTC: Urease positive thermophilic Campylobacter; d ELISA: enzyme-linked immunoabsorbent assay.
but both probes did not hybridize to the C. upsaliensis amplicon. Therefore the CC2 and CC3 probes reacted with only C. jejuni and C. coli target sequences. Ritzler and Altwegg18 investigated the effect of different mismatches in 25 nucleotide probes on the specificity of a PCR ELISA assay for the detection of Bartonella henselae and Bartonella quintana. They concluded that the incorporation of one mismatch could be tolerated and only slightly reduced sensitivity, whereas two mismatches could not be tolerated and almost no hybridization was detectable. This is in contrast to the results obtained in this study with the Campylobacter species-specific probes, which could tolerate up to three mismatches in the probe sequence. This difference may be related to the longer length of the probes (33–35 nucleotide versus 25 nucleotide probes) or the lower G-C ratio in the target sequence. A PCR ELISA assay for the detection and identification of C. jejuni and C. coli was developed using the CC2 and CJ2 probes. The specificity of the PCR ELISA using the CJ2 and CC2 probes was investigated against a collection of
C. jejuni and C. coli isolates from raw food products, related species and non-related organisms. The CJ2 probe was specific for C. jejuni and the CC2 probe was specific for C. jejuni, C. coli and C. hyoilei. C. hyoilei is a recently described species associated with porcine proliferative enteritis23 and is closely related to C. coli. The role of C. hyoilei in human gastrointestinal disease has yet to be determined, but a report by Vandamme and colleagues24 concluded that C. hyoilei and C. coli are indistinguishable and should be classified as the same species, which may account for C. hyoilei reacting positively with the CC2 probe. Six of the 21 C. jejuni isolates used in the specificity study were not positive in the PCR ELISA assay with the CC2 probe. These C. jejuni isolates were of diverse serotype. The sensitivity of PCR ELISA end-point detection was compared with gel electrophoresis end-point detection using the same primers. The smallest amount of C. jejuni DNA which could be detected by the PCR ELISA assay was 1.5 fg which is approximately one genome equivalent assuming a
298
A. D. Sails et al.
Fig. 2. Comparison of the sensitivity of the PCR ELISA and gel electrophoresis PCR end-point detection methods. a 100 bp: 100 base pair ladder; b Amount of DNA template; c A405: Absorbance of PCR ELISA end-point detection with the CJ2 probe.
genome size of 1.6 Mb.25 This level of sensitivity is about 100-fold higher than gel electrophoresis of the amplification products. In contrast, the PCR ELISA assay of Ritzler and Altwegg18 was 10–20 times more sensitive when compared with detection of amplified product by gel electrophoresis and ethidium bromide staining. Ritzler and Altwegg18 also used a higher annealing temperature (55°C) and a shorter hybridization time (60 min in contrast to 90 min reported in this study), which can reduce the sensitivity of the assay. Positive reactions in the PCR ELISA were visibly identified by the formation of a green colouration in the wells of the microtitre plate well allowing positive and negative results to be determined without the use of a spectrophotometer. PCR assays are inherently sensitive and therefore susceptible to false-positive results due to the contamination of reactions with previous PCR products or extraneous DNA. The prevention of cross-contamination from amplified product to subsequent PCR reactions can be facilitated by strict laboratory practices and separating the areas used for template preparation, PCR reaction mix preparation and analysis of amplified products.26 The use of a uracil DNA glycosylase enzyme and the inclusion of dUTP instead of dTTP in the reaction mixture for the PCR ELISA
labelling reaction reduces problems associated with product carryover.27 The presence of ‘spurious’ bands following agarose gel analysis of PCR products can occur through ‘mispriming’ of the reaction and non-specific amplification of non-target DNA.16 The PCR assays reported by Waage and colleagues28 and Winters and Slavik29 both described ‘non-specific products’ or ‘bands of unknown origin’ in their PCR assays that hampered the interpretation of the results. The standardization of gel electrophoresis methods is problematic and transfer of methods between laboratories is difficult. The use of a PCR ELISA assay, which incorporates a target specific solution hybridization event, removes subjectivity and provides increased specificity. Also the incorporation of positive and negative controls in each set of tests facilitates the standardization of methods between laboratories. The investigation of naturally contaminated samples such as raw poultry and environmental samples for campylobacters by enrichment culture can yield C. jejuni and C. coli. But such methods can also yield other Campylobacter organisms such as C. fetus, C. lari (including urease positive isolates), C. hyointestinalis and Arcobacter spp., which are difficult to identify using phenotypic methods. However
Campylobacter PCR ELISA
the PCR ELISA assay described here is specific for C. jejuni and C. coli and can be completed in approximately 7 h. The PCR ELISA assay has been demonstrated to be capable of detecting one genome-equivalent therefore this assay may have the potential to detect C. jejuni and C. coli in foods and environmental samples. Further studies are currently underway to validate the assay for the detection of C. jejuni and C. coli in foods and environmental waters.
11.
12.
13.
ACKNOWLEDGEMENTS This work was supported by The Ministry of Agriculture, Food and Fisheries, UK (research programme FS1242). The authors would also like to thank Dr Ibrahim Attabay at Bristol University for the provision of the Campylobacter hyointestinalis isolates.
REFERENCES 1. Skirrow, M. B. (1994) Diseases due to Campylobacter; Helicobacter and related bacteria. Journal of Comparative Pathology 111, 113–49. 2. On, S. L. and Holmes, B. (1992) Assessment of enzyme detection tests useful in identification of Campylobacteria. Journal of Clinical Microbiology 30, 746–9. 3. Anon. (1993) Advisory Committee on the Microbiological Safety of Food Interim Report on Campylobacter. HMSO, London. 4. Bolton, F. J., Wareing, D. R. A., Skirrow, M. B. & Hutchinson, D. N. (1992) Identification and Biotyping of Campylobacters. In Identification Methods in Applied and Environmental Microbiology, (Board, R. G., Jones, D. & Skinner, F. A., eds). Pp. 151–62. Oxford, England: Blackwell Scientific Publishers. 5. Lior, H. (1984) New, extended biotyping scheme for Campylobacter jejuni, Campylobacter coli and “Campylobacter laridis”. Journal of Clinical Microbiology 20, 636–40. 6. Bolton, F. J., Holt, A. V. & Hutchinson, D. N. (1984) Campylobacter biotyping scheme of Epidemiological value. Journal of Clinical Pathology 37, 677–81. 7. Bej, A. K., Steffan, R. J., DiCesare, J., Haff, L. & Atlas, R. M. (1990) Detection of coliform bacteria in water by polymerase chain reaction and gene probes. Applied and Environmental Microbiology 56, 307–4. 8. Eyers, M., Chapelle, S., Van Camp, G., Goossens, H. & De Wachter, R. (1993) Discrimination among thermophilic Campylobacter species by polymerase chain reaction amplification of 23S rRNA gene fragments. Journal of Clinical Microbiology 31, 3340–3. 9. Cardarelli-Leite, P., Blom, K., Patton, C. M., Nicholson, M. A., Steigerwalt, A. G., Brenner, D. J., Barrett, T. J. & Swaminathan, B. (1996) Rapid identification of Campylobacter species by restriction fragment length polymorphism analysis of a PCR-amplified fragment of the gene coding for 16S rRNA. Journal of Clinical Microbiology 34, 62–7. 10. Hurtado, A. & Owen, R. J. (1997) A molecular scheme
14.
15.
16.
17.
18.
19.
20.
21.
22.
299
based on 23S rRNA gene polymorphism’s for rapid identification of Campylobacter and Arcobacter species. Journal of Clinical Microbiology 35, 2401–4. Gonzalez, I., Grant, K. A., Richardson, P. T., Park, S. F. & Collins, M. D. (1997) Specific identification of the enteropathogens Campylobacter jejuni and Campylobacter coli by using a PCR test based on the ceuE gene encoding a putative virulence determinant. Journal of Clinical Microbiology 35, 759–63. Stucki, U., Frey, J., Nicolet, J. & Burnens, A. P. (1995) Identification of Campylobacter jejuni on the basis of a species-specific gene that encodes a membrane protein. Journal Clinical Microbiology 33, 855–9. Van Doorn, L-J., Giesendorf, B. A. J., Bax, R., van der Zeijst, B. A. M., Vandamme, P. & Quint, W. G. V. (1997) Molecular discrimination between Campylobacter jejuni, Campylobacter coli, Campylobacter lari and Campylobacter upsaliensis by polymerase chain reaction based on a novel putative GTPase gene. Molecular and Cellular Probes 11, 177–85. Comi, G., Ferroni, P., Cocolin, L., Cantoni, C. & Manzano, M. (1996) Detection and identification of Campylobacter coli and Campylobacter jejuni by twostep polymerase chain reaction. Molecular Biotechnology 3, 266–8. Jackson, C. J., Fox, A. J. & Jones, D. (1996) A novel polymerase chain reaction assay for the detection and speciation of thermophilic Campylobacter spp. Journal of Applied Bacteriology 81, 467–73. Fredricks, D. N. & Relman, D. A. (1999) Application of the polymerase chain reaction to the diagnosis of infectious diseases. Clinical Infectious Diseases 29, 475–88. Giesendorf, B. A., Quint, W. G., Henkens, M. H., Stegeman, H., Huf, F. A. & Nesters, H. G. (1992) Rapid and sensitive detection of Campylobacter spp. in chicken products by using the polymerase chain reaction. Applied and Environmental Microbiology 58, 3804–8. Ritzler, M. & Altwegg, M. (1996) Sensitivity and specificity of a commercially available enzyme-linked immunoassay for the detection of polymerase chain reaction amplified DNA. Journal of Microbiological Methods 27, 233–8. Borrow, R., Claus, H., Guiver, M., Smart, L., Jones, D. M., Kaczmarski, E. B., Frosch, M. & Fox, A. J. (1997) Non-culture diagnosis and serogroup determination of meningococcal B and C infection by a sialyltransferase (siaD) PCR ELISA. Epidemiology and Infection 118, 111–7. Lawson, A. J., Logan, J. M., O’Neill, J., Desai, M. and Stanley, J. (1999) Largescale survey of Campylobacter species in human gastroenteritis by PCR and PCREnzyme-Linked Immunosorbent assay. Journal of Clinical Microbiology 37, 3860–4. O’Sullivan, N. A., Fallon, R., Carroll, C., Smith, T. and Maher, M. (2000) Detection and differentiation of Campylobacter jejuni and Campylobacter coli in broiler chicken samples using a PCR/DNA probe membrane based colorimetric detection assay. Molecular and Cellular Probes 14, 7–16. Ikuta, S., Takagi, K., Wallace, R. & Itakura, K. (1987) Dissociation kinetics of 19 base paired oligonucleotide-DNA duplexes containing single mismatched base pairs. Nucleic Acid Research 15, 797–811.
300
A. D. Sails et al.
23. Alderton, M. R., Korolik, P., Coloe, F. E., Dewhirst, F. E. & Paster, B. J. (1995) Campylobacter hyoilei sp. nov., associated with porcine proliferative enteritis. International Journal of Systematic Bacteriology 45, 61–6. 24. Vandamme, P., Van Doorn, L. J., al Rashid, S. T., Quint, W. G., van der Plas, J. and On, S. L. (1997) Campylobacter hyoilei Alderton et al. 1995 and Campylobacter coli Veron and Chatelain 1973 are subjective synonyms. International Journal of Systematic Bacteriology 47, 1055–60. 25. Parkhill, J., Wren, B. W., Mungall, K., Ketley, J. M., Churcher, C., Basham, D., Chillingworth, T., Davies, R. M., Feltwell, T., Holroyd, S., Jagels, K., Karlyshev, A. V., Moule, S., Pallen, M. J., Penn, C. W., Quail, M. A., Rajandream, M. A., Rutherford, K. M., van Vliet, A. H., Whitehead, S. & Barrel, B. G. (2000) The genome sequence of the foodborne pathogen
26. 27.
28.
29.
Campylobacter jejuni reveals hypervariable sequences. Nature 403, 665–8. Kwok, S. & Higuchi, R. (1989) Avoiding false positives with PCR. Nature 339, 237–8. Longo, M. C., Berninger, M. S. & Hartley, J. L. (1990) Use of uracil DNA glycosylase to control carry-over contamination in polymerase chain reactions. Gene 93, 125–8. Waage, A. S., Varund, T., Lund, V. and Kapperud, G. (1999) Detection of small numbers of Campylobacter jejuni and Campylobacter coli in environmental water, sewage, and food samples by a semi-nested PCR assay. Applied and Environmental Microbiology 65, 1636– 43. Winters, D. K. and Slavik, M. F. (1995) Evaluation of a PCR based assay for the specific detection of Campylobacter jejuni in chicken washes. Molecular and Cellular Probes 9, 307–10.
Development of a PCR ELISA assay for the identification of Campylobacter jejuni and Campylobacter coli A. D. Sails,∗1 A. J. Fox,2 F. J. Bolton,1 D. R. A. Wareing,1 D. L. A. Greenway3 and R. Borrow2 1
Preston Public Health Laboratory, Royal Preston Hospital, PO Box 202, Sharoe Green Lane, Fulwood, Preston, Lancs., PR2 9HG, UK; 2Manchester Public Health Laboratory, Withington Hospital, Nell Lane, West Didsbury, Manchester, M20 2LR, UK and 3 The Department of Biological Sciences, The University of Central Lancashire, Preston, PR1 2HE, UK; ∗Current address: National Center for Infectious Diseases, Centers For Disease Control and Prevention, 1600 Clifton Road, Atlanta, Georgia, 30333, USA. (Received 20 February 2001, Accepted 20 June 2001) A polymerase chain reaction (PCR) assay was developed based on a solution-hybridization colorimetric end-point detection format (PCR ELISA) for the identification of Campylobacter jejuni and Campylobacter coli. PCR primers were designed to target a gene sequence with speciesspecific motifs. Five biotin-labelled probes targeted to the species-specific motifs were investigated for the detection of digoxygenin-labelled PCR products from C. jejuni and C. coli using the PCR ELISA format. Two probes were identified, one which reacts with both the C. jejuni and C. coli target sequences (probe CC2) and one probe which reacts with the C. jejuni target sequence only (probe CJ2). The specificity of the assay with the CJ2 and CC2 probes was investigated with a range of Campylobacter spp., Arcobacter spp., Helicobacter spp. and a range of unrelated organisms. The PCR ELISA assay and probes were demonstrated to be specific for C. jejuni and C. coli. The sensitivity of the PCR ELISA assay was demonstrated to be 10–100-fold more sensitive than a gel-based PCR method using the same primers. This PCR ELISA assay is sensitive, specific and significantly reduces the time needed for the identification of C. jejuni and C. coli and has the potential to facilitate early detection of these important gastro-intestinal pathogens. 2001 Academic Press KEYWORDS: Campylobacter, PCR, ELISA, identification, detection.
INTRODUCTION Campylobacter is the most frequent bacterial cause of gastro-enteritis in the UK and the rest of the developed world.1 Campylobacter jejuni is the most common species associated with gastro-enteritis but Campylobacter coli accounts for approximately 5% of infections.1 Campylobacters are slow-growing fastidious organisms, which are asaccharolytic and
generally biochemically unreactive, unlike other enteric pathogens such as Salmonella and Shigella.2 Many of the phenotypic tests used for the routine identification of members of the Enterobacteriaceae do not identify campylobacters. Many diagnostic microbiology laboratories consequently fail to identify campylobacters to the level of species and report isolates as Campylobacter spp.3 Identification to the level of genus is based on colony morphology, Gram
Author to whom all correspondence should be addressed: Prof. Andrew Fox, Manchester Public Health Laboratory, Withington Hospital, Nell Lane, West Didsbury, Manchester, M20 2LR, UK. Tel: 44 161 291 4631; Fax: 44 161 446 2180; E-mail: [email protected]
0890–8508/01/050291+10 $35.00/0
2001 Academic Press
292
A. D. Sails et al.
stain, growth only in a micro-aerobic atmosphere and oxidase tests4 and more extensive identification or biotyping schemes have been described.4–6 The polymerase chain reaction (PCR) is a rapid method for the detection and identification of bacterial pathogens.7 Several assays have been described for the identification of campylobacters based on the amplification of 16 Svedberg units (S)8,9 and 23S rRNA gene sequences.10 Other gene targets reported for the identification of campylobacters by PCR include the ceuE gene,11 the mapA gene,12 the GTPase gene13 and the flagellin A (flaA) gene.14 Jackson and colleagues15 described a novel PCR assay for the detection of Campylobacter species, which targeted a 256 bp region of an open reading frame (ORF) adjacent to and downstream from a novel two-component regulatory gene. The assay produced an amplicon from C. jejuni, C. coli and Campylobacter upsaliensis which when digested with the restriction enzymes Alu I, Dde I and Dra I yielded species-specific restriction fragment length polymorphism (RFLP) profiles. Differentiation was achieved between C. jejuni and C. coli because of the presence of a Dra I restriction site in the C. jejuni amplification product. Many of the PCR assays described for the identification of campylobacters use agarose gel electrophoresis for end-point detection, but this method lacks sensitivity, specificity, and can be subjective in its interpretation.16 The sensitivity and specificity can be improved by transferring the PCR products to a nylon membrane and detection by hybridization with a labeled probe.17 However the extra steps are labour intensive and the standardization of the procedure between laboratories is difficult. Adaptation of PCR assays into a solution hybridization colorimetric end-point detection format (PCR ELISA) facilitates specific and sensitive detection of PCR amplification products with the potential for increased throughput.18 Non-radioactive digoxygenin (DIG) labels are incorporated into the PCR product during the PCR amplification and the detection of labelled PCR products is based on biotinylated capture probes complementary to specific sequences in the target gene.18 The aim of this study was to develop a probe-based PCR ELISA assay to detect speciesspecific motifs within a previously described gene sequence (ORF-C)15 for the detection and identification of C. jejuni and C. coli.
Table 1. Non-Campylobacter organisms used in the specificity studies. Organism
Aeromonas hydrophila Bacillus cereus Bacillus subtilis Clostridium perfringens Escherichia coli Escherichia coli 0157 non toxigenic Escherichia coli 0157 VTEC Lactobacillus casei Lactococcus lactis Listeria monocytogenes Pseudomonas aeruginosa Salmonella enteritidis Salmonella typhimurium Shigella sonnei Enterococcus faecalis Staphylococcus aureus Coagulase negative staphylococcus Serratia marcescens Vibrio parahaemolyticus Vibrio alginolyticus Vibrio cholerae non 0:1 Yersinia enterocolitica Proteus mirabilis Proteus rettgeri Citrobacter freundii Legionella pneumophila Acinetobacter lwoffii Enterobacter aerogenes Enterobacter cloacae Proteus rettgeri Klebsiella pneumoniae
NCTCa reference number 8049 7464 10400 8237 10418 12900 12079 10302 662 11994 10662 400452/97b 12023 8574 775 6571 170454/97b 11935 10885 12160 11348 10460 10975 7475 9750 12821 5866 10006 11936 7475 350324/97
a
NCTC: National Collection of Type Cultures, PHLS, UK Clinical isolates obtained from Preston Public Health Laboratory. b
containing 15% (v/v) glycerol (BDH Ltd, Poole, UK). Campylobacter spp. and Arcobacter spp. recovered from −70°C storage were cultured on Columbia blood agar (BA) (CM 331, Oxoid Ltd, UK) containing 5% (v/v) whole horse blood. Helicobacter spp. recovered from −70°C storage were cultured on chocolate agar (heated Columbia blood agar, CA). All Campylobacter, Helicobacter and Arcobacter isolates were incubated micro-aerobically at 37°C for 48–72 h. All other organisms were cultured on BA either aerobically or anaerobically at 37°C for 48 h prior to testing. The organisms used in the study are listed in Tables 1, 3 and 4.
MATERIALS AND METHODS Bacterial isolates and culture conditions
Extraction of DNA from bacterial cultures
Bacterial isolates were stored at −70°C in Brain Heart Infusion broth (CM 225, Oxoid Ltd, Basingstoke, UK)
DNA was extracted from bacterial cultures on BA or CA using the Isoquick DNA extraction method
Campylobacter PCR ELISA
293
Fig. 1. C. jejuni, C. coli and C. upsaliensis ORF-C gene sequences and PCR ELISA probes.
according to the manufacturer’s instructions (Orca Research Ltd, USA) and the purified DNA was quantified spectroscopically using a GeneQuant II DNA calculator (Amersham-Pharmacia Biotech Ltd, Amersham, UK).
PCR primers and probes PCR primers were designed using Primerselect software (DNASTAR Inc., Madison, USA) to amplify a 184 bp target sequence in the ORF-C gene sequence identified by Jackson and colleagues15 containing species-specific motifs. Probes were designed to hybridize to the C. jejuni and C. coli amplification products (Fig. 1) in the species-specific regions of the gene sequence. Probes CC1, CC2 and CC3 were designed to target the C. coli sequence and the CJ1 and CJ2 probes were designed to target the C. jejuni sequence. The forward and reverse (ORF-C) primer sequences were as follows: ORF-CF: 5′GTTTGCGACCATATTGTTTGGCATTGTAG3′ ORF-CR: 5′TAATACCGTTGCCCAAATCCCTGAAAGC3′. Primers and probes were synthesized by Oswell DNA Service, UK, and were purified by high performance liquid chromatography (HPLC) with the probes being 5′ biotinylated.
PCR amplification and DIG labelling The primer and magnesium chloride concentrations were optimized and PCR labelling reactions were performed using the PCR ELISA DIG Labellingplus kit (Roche Diagnostics Ltd, Lewes, UK) according to manufacturer’s instructions. PCR amplification was
performed in a 50 l reaction volume containing: molecular biology grade water, PCR reaction buffer (final concentration 2 mM Tris–HCl (pH 8·4), 5 mM KCl), 3 mM final concentration magnesium chloride, PCR ELISA DIG labelling Mixplus containing final concentrations of 200 M of each of dATP, dCTP, dGTP, 570 M dUTP, 30 M DIG-UTP, primers ORF-CF and ORF-CR (final concentration 0·5 M), Platinum Taq DNA polymerase (final concentration 2.5 U) (Life Technologies Ltd, Paisley, UK), uracil DNA glycosylase (final concentration 2 U), and 5 l of template DNA. Thermal cycling parameters were as follows: 95°C for 2 min; followed by 30 cycles of 94°C for 25 s, 55°C for 40 s, 72°C for 60 s; and a terminal extension step of 72°C for 5 min. Negative controls (molecular biology grade water) were included in each set of reactions.
Detection of DIG labelled PCR products by ELISA The PCR ELISA (DIG Detection kit, Roche Diagnostics Ltd, Lewes, UK) was used according to the manufacturer’s instructions. A negative detection control (molecular biology grade water) was included with each set of reactions. DIG labelled PCR products were denatured in 100 mM NaOH and 300 mM NaCl and hybridization buffer containing probe at a concentration of 50 ng/ml was added. Aliquots (200 l) were then transferred to the wells of a streptavidincoated microtitre plate strip, and the plate was incubated at 39°C in a shaking ELISA incubator (IEMS Thermomix, Labsystems Ltd, UK) for 90 min. The microtitre plate wells were then washed and antidigoxigenin peroxidase conjugate (2 mU diluted in
294
A. D. Sails et al.
conjugate buffer 200 l) was added to each well and the microtitre plate was incubated at 37°C with shaking for 30 min. Following washing ABTS (2, 2azino-di-3-ethylbenzithiazoline sulfonate) was added (200 l of a 1 mg/ml solution in substrate buffer) and the plate was incubated at 37°C with shaking for 30 min to allow colour development. The absorbance of the contents of each well was read at 405 nm (492 nm reference) and the extinction of the negative control was subtracted from each sample. Positive reactions were determined by a calculation of cut-off values as follows: cut-off=2×absorbance at 405 nm of the mean of ten replicates of the negative detection control (water).19
Detection of DIG labelled PCR products from C. jejuni, C. coli and C. upsaliensis using the novel PCR ELISA probes DNA was extracted from fresh overnight cultures of C. jejuni (NCTC11168), C. coli (NCTC11366) and C. upsaliensis (NCTC11540) and the DNA concentration was determined spectrophotometrically as described above. DNA template dilutions were prepared in molecular biology grade water to give 15 pg, 1·5 pg, 150 fg, 15 fg and 1·5 fg per 5 l aliquot. This range of DNA template concentrations were used in the PCR ELISA reaction and the labelled products detected using the five probes as described above.
Investigation of the specificity of the PCR ELISA assay with the CJ2 and CC2 probes
with the labelled products being detected using the CJ2 probe and PCR ELISA method as described above.
RESULTS Detection of DIG labelled PCR products from C. jejuni, C. coli and C. upsaliensis using the novel PCR ELISA probes Five oligonucleotide probes were investigated for the detection of labelled PCR products from C. jejuni, C. coli and C. upsaliensis using the PCR ELISA method. The comparative specificity of the different probes for the detection of C. jejuni, C. coli and C. upsaliensis DIG labelled PCR products is summarized in Table 2. Positive results were determined by calculating the cut off value as described above and were visible by a green coloration. The mean absorbance at 405 nm of 10 replicate negative detection controls was 0.125, to give a negative cut off value of 0.25. The CJ1 probe hybridized with the PCR products from C. jejuni and C. coli but not C. upsaliensis. The CJ2 probe hybridized with the C. jejuni PCR product only, demonstrating its specificity for this species. The CC1 probe hybridized to the PCR products from all three species however the CC2 and CC3 probes hybridized to the PCR products from C. jejuni and C. coli only.
Investigation of the specificity of the PCR ELISA assay with the CJ2 and CC2 probes
DNA was extracted from overnight cultures of a range of Campylobacter, Arcobacter, Helicobacter species and unrelated genera (Tables 1, 3 and 4) and the DNA was quantified spectrophotometrically as described above. The purified DNA (10 ng per reaction) was used as a template in the PCR ELISA and the labelled PCR products were detected using the CJ2 and CC2 probes as described above.
The PCR ELISA results for isolates of Campylobacter species and related organisms are summarized in Tables 3 and 4. The PCR ELISA assay with the CJ2 probe produced a positive reaction with all of the isolates of C. jejuni but not with any other organisms. The CC2 probe produced a positive signal with all isolates of C. coli and the closely related species Clostridium hyoilei. The CC2 probe also produced a positive signal with 15 of the 21 C. jejuni isolates tested. The CC2 probe did not produce a positive signal with any of the other organisms tested.
Comparison of the sensitivity of the PCR ELISA assay and gel-based detection of PCR products using varying amounts of template DNA
Comparison of the sensitivity of the PCR ELISA assay and gel-based detection of PCR products using varying amounts of template DNA
DNA extracted from C. jejuni (NCTC11168) was diluted in molecular biology grade water to give a range of DNA concentrations from 1 ng to 0.1 fg DNA. The DNA dilutions were then used as template in the PCR ELISA assay and a gel-based PCR assay
The smallest amount of C. jejuni template DNA giving a positive signal in the PCR ELISA assay with the CJ2 probe was 1·5 fg (Fig. 2). The smallest amount of C. jejuni template DNA which produced a clearly defined band in the conventional PCR assay using gel
Campylobacter PCR ELISA
295
Table 2. Polymerase chain reaction (PCR) ELISAa detection of digoxygenin (DIG) labelled PCR products from Campylobacter jejuni, Campylobacter coli and Campylobacter upsaliensis with the five probes. Campylobacter species
C. jejuni (NCTC11168)
C. coli (NCTC11366)
C. upsaliensis (NCTC11540)
a
DNA template level
15 pg 1·5 pg 150 fg 15 fg 1·5 fg 15 pg 1·5 pg 150 fg 15 fg 1·5 fg 15 pg 1·5 pg 150 fg 15 fg 1·5 fg
PCR ELISA end-point detection (Absorbance at 405 nm for each probe) CJ1
CJ2
CC1
CC2
CC3
2·528 2·794 2·381 −0·001 −0·001 0·413 0·423 0·566 −0·003 −0·003 0·000 −0·007 −0·002 0·004 0·000
>3·0 >3·0 >3·0 >3·0 0·604 −0·010 −0·008 −0·007 −0·006 −0·009 −0·009 −0·010 −0·008 −0·007 −0·001
2·395 2·481 3·071 −0·008 −0·008 2·812 2·942 2·291 −0·008 −0·007 2·278 2·874 2·555 2·687 −0·012
>3·0 >3·0 1·562 −0·016 −0·018 2·762 2·510 2·665 −0·030 −0·012 0·059 0·045 0·051 0·010 −0·013
>3·0 2·905 2·979 1·620 0·013 2·526 2·836 2·631 0·009 0·007 0·055 0·055 0·019 0·019 0·012
ELISA: enzyme-linked immunoabsorbent assay.
electrophoresis and ethidium bromide staining to detect the PCR products was 150 fg DNA (Fig. 2). A faint band was visible following gel electrophoresis of the PCR product from 15 fg of template DNA, but it was poorly defined. The PCR ELISA was therefore 10–100-fold more sensitive when compared with gelbased PCR detection.
DISCUSSION There are relatively few biochemical reactions that are useful for the identification of Campylobacter spp. Useful laboratory biochemical tests for the identification of Campylobacter spp. include catalase production, hippurate hydrolysis, hydrogen sulphide production, indoxyl acetate hydrolysis, nitrate reduction and urease production and growth temperature.4 The hippurate hydrolysis test is critical to distinguish between C. jejuni and C. coli isolates4 with C. jejuni being positive in the test. However, isolates of C. jejuni have been reported that do not express this phenotypic trait2 making them more difficult to identify unambiguously using phenotypic methods. This has led to the development of genetic methods for identification. PCR assays for the identification and speciation of C. jejuni and C. coli have been described.11,14,15,20,21 Many of these assays utilized gel electrophoresis as the end-point detection method, which can be subjective in its interpretation.16 Adapting PCR assays into solution hybridization format with colorimetric
end-point detection increases both specificity and sensitivity.20,21 In this study a PCR ELISA was investigated for the detection and identification of C. jejuni and C. coli. The primers described here were demonstrated to produce amplicons of the correct size from C. jejuni, C. coli and C. upsaliensis (data not presented). To produce an assay specific for C. jejuni and C. coli, novel probes were designed to hybridize to speciesspecific sequences within the DIG labelled PCR products from these species, which could then be detected using a commercial ELISA kit. The PCR ELISA method allows the differentiation between specific PCR products via the hybridization of the biotin-labelled probes.18 The specificity of hybridization is dependent on a number of factors including the length and GC content of the capture probe, and the temperature of hybridization. Increasing the temperature of hybridization increases specificity, although sensitivity may be reduced. Short capture probes of 18–20 nucleotides and elevated hybridization temperatures (up to 55°C) can identify a single base mismatch in the probe target sequence. In order to maximize the melting temperature (Tm) of the hybridized probe and target oligonucleotide duplex, probes were designed to be 33–36 bp long to give a Tm of approximately 60°C. Two probes CJ1 and CJ2 were designed to target the C. jejuni sequence, CJ1 was identical to the target sequence, but the second probe (CJ2) included two nucleotide mismatches in an attempt to increase the specificity of the hybridization to the C. jejuni PCR product. The
296
A. D. Sails et al. Table 3. Identification of Clostridium jejuni and Clostridium coli by polymerase chain reaction (PCR) ELISA.e Organism
C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C. C.
jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni jejuni coli coli coli coli coli coli coli coli
Sourcea
NCTC NCTC PH PH PP PB PB PB PB PB PC PC PC PC PC PC PL PL PM PS PS NCTC PC PC PL PL PP PP PP
Reference number
11168 11392 3212/91 3748/91 2380/90 451042/98 87881/95 87882/95 87880/95 87878/95 450359/96 450821/98 450286/96 450254/96 450361/96 88086/95 450888/98 450749/98 450098/98 77124/94 87024/95 11366 450788/98 450733/98 450761/98 450862/98 87209/95 87207/95 87208/95
Serotypeb
2 6 2 4 23 50 1 1 10 13 1 2 5 6 11 38,57 1 2 NDd ND ND 4 28 56 56 56 49 53 53
PCR ELISA resultc CJ2
CC2
+ + + + + + + + + + + + + + + + + + + + + − − − − − − − −
+ + + + + − − − − − + − + + + + + + + + + + + + + + + + +
a
Source: NCTC: National Collection of Type Cultures; all other isolates were obtained from the Campylobacter Collaborating Unit, Preston Public Health Laboratory, UK; PH: Human isolate; PP: Porcine isolate; PB: Bovine isolate; PC: Chicken isolate; PL: lamb isolate; PM: Milk isolate; PS: Sand isolate; b Penner Heat Stable serotype; c PCR ELISA negative cut off value was 0·25; d ND: Not determined. e ELISA: enzyme-linked immunoabsorbent assay.
stability of oligonucleotide duplexes is dependent on the oligonucleotides involved22 with one mismatch destabilizing a duplex slightly (stable mismatch: G-T, G-A) or significantly (unstable mismatch: A-A, T-T, C-T, and C-A). Additionally, the position of the mismatch can influence the melting point of the duplex. The CJ1 probe hybridized to both C. jejuni and C. coli amplicons in the PCR ELISA assay. The incorporation of a T-T and a C-A unstable mismatch in the CJ2 probe did not prevent hybridization with the C. jejuni DIG-labelled amplicon. The destabilizing effect of the two mismatches in the duplex may have been offset by the length of the oligonucleotide probe and/or the low G-C content. However, the mismatches incorporated in the CJ2 probe did destabilize the duplex formed between the probe and the C. coli
target sequence making the CJ2 probe specific for the identification of C. jejuni. Three probes were designed to hybridize with the C. coli amplicons with the CC1 probe being identical to the target sequence in the C. coli PCR product. The CC2 and CC3 probes incorporated two and three nucleotide mismatches respectively. The CC1 probe was six nucleotides different from the C. jejuni target sequence and the CC2 and CC3 probes were 8 nucleotides different from the C. jejuni target sequence. The CC1 probe hybridized to the amplicons from C. jejuni, C. coli and C. upsaliensis to form a stable duplex and therefore produced a detectable absorbance in the assay. The CC2 and CC3 probes hybridized to both the C. jejuni and C. coli DIG-labelled amplicons to form a stable duplex
Campylobacter PCR ELISA
297
Table 4. Identification of Campylobacter, Helicobacter and Arcobacter isolates by polymerase chain reaction (PCR) ELISA.d Organism
Campylobacter lari C. lari C. lari Campylobacter lari (UPTC)c C. lari (UPTC) C. lari (UPTC) Campylobacter upsaliensis Campylobacter helveticus Campylobacter fetus subsp. fetus C. fetus subsp. fetus C. fetus subsp. fetus Campylobacter fetus subsp. venerealis Campylobacter hyointestinalis C. hyointestinalis C. hyointestinalis Campylobacter mucosalis Campylobacter sputorum subsp. bubulus Campylobacter sputorum biovar faecalis Campylobacter hyoilei Helicobacter cinaedi Arcobacter butzleri A. butzleri A. butzleri A. butzleri Arcobacter skirowii
Sourcea
NCTC PSF PSF NCTC PSF PSF NCTC NCTC NCTC PB PL NCTC NCTC PB PB NCTC NCTC NCTC NCTC NCTC NCTC PC PC PC NCTC
Reference number
11352 77779/94 77715/94 11928 77625/94 77716/94 11540 12479 5850 450971/98 451145/98 10354 11609 DC54C DC89C 11000 11367 11416 12884 12423 12481 450427/98 450555/98 450556/98 12713
PCR ELISAb CJ2
CC2
− − − − − − − − − − − − − − − − − − − − − − − − −
− − − − − − − − − − − − − − − − − − + − − − − − −
a Source: NCTC: National Collection of Type Cultures; all other isolates were obtained from the Campylobacter Collaborating Unit, Preston Public Health Laboratory, UK; PSF: Shellfish isolate; PB: Bovine isolate; PL: lamb isolate; PC: Chicken isolate; b PCR ELISA negative cut off value was 0·25. c UPTC: Urease positive thermophilic Campylobacter; d ELISA: enzyme-linked immunoabsorbent assay.
but both probes did not hybridize to the C. upsaliensis amplicon. Therefore the CC2 and CC3 probes reacted with only C. jejuni and C. coli target sequences. Ritzler and Altwegg18 investigated the effect of different mismatches in 25 nucleotide probes on the specificity of a PCR ELISA assay for the detection of Bartonella henselae and Bartonella quintana. They concluded that the incorporation of one mismatch could be tolerated and only slightly reduced sensitivity, whereas two mismatches could not be tolerated and almost no hybridization was detectable. This is in contrast to the results obtained in this study with the Campylobacter species-specific probes, which could tolerate up to three mismatches in the probe sequence. This difference may be related to the longer length of the probes (33–35 nucleotide versus 25 nucleotide probes) or the lower G-C ratio in the target sequence. A PCR ELISA assay for the detection and identification of C. jejuni and C. coli was developed using the CC2 and CJ2 probes. The specificity of the PCR ELISA using the CJ2 and CC2 probes was investigated against a collection of
C. jejuni and C. coli isolates from raw food products, related species and non-related organisms. The CJ2 probe was specific for C. jejuni and the CC2 probe was specific for C. jejuni, C. coli and C. hyoilei. C. hyoilei is a recently described species associated with porcine proliferative enteritis23 and is closely related to C. coli. The role of C. hyoilei in human gastrointestinal disease has yet to be determined, but a report by Vandamme and colleagues24 concluded that C. hyoilei and C. coli are indistinguishable and should be classified as the same species, which may account for C. hyoilei reacting positively with the CC2 probe. Six of the 21 C. jejuni isolates used in the specificity study were not positive in the PCR ELISA assay with the CC2 probe. These C. jejuni isolates were of diverse serotype. The sensitivity of PCR ELISA end-point detection was compared with gel electrophoresis end-point detection using the same primers. The smallest amount of C. jejuni DNA which could be detected by the PCR ELISA assay was 1.5 fg which is approximately one genome equivalent assuming a
298
A. D. Sails et al.
Fig. 2. Comparison of the sensitivity of the PCR ELISA and gel electrophoresis PCR end-point detection methods. a 100 bp: 100 base pair ladder; b Amount of DNA template; c A405: Absorbance of PCR ELISA end-point detection with the CJ2 probe.
genome size of 1.6 Mb.25 This level of sensitivity is about 100-fold higher than gel electrophoresis of the amplification products. In contrast, the PCR ELISA assay of Ritzler and Altwegg18 was 10–20 times more sensitive when compared with detection of amplified product by gel electrophoresis and ethidium bromide staining. Ritzler and Altwegg18 also used a higher annealing temperature (55°C) and a shorter hybridization time (60 min in contrast to 90 min reported in this study), which can reduce the sensitivity of the assay. Positive reactions in the PCR ELISA were visibly identified by the formation of a green colouration in the wells of the microtitre plate well allowing positive and negative results to be determined without the use of a spectrophotometer. PCR assays are inherently sensitive and therefore susceptible to false-positive results due to the contamination of reactions with previous PCR products or extraneous DNA. The prevention of cross-contamination from amplified product to subsequent PCR reactions can be facilitated by strict laboratory practices and separating the areas used for template preparation, PCR reaction mix preparation and analysis of amplified products.26 The use of a uracil DNA glycosylase enzyme and the inclusion of dUTP instead of dTTP in the reaction mixture for the PCR ELISA
labelling reaction reduces problems associated with product carryover.27 The presence of ‘spurious’ bands following agarose gel analysis of PCR products can occur through ‘mispriming’ of the reaction and non-specific amplification of non-target DNA.16 The PCR assays reported by Waage and colleagues28 and Winters and Slavik29 both described ‘non-specific products’ or ‘bands of unknown origin’ in their PCR assays that hampered the interpretation of the results. The standardization of gel electrophoresis methods is problematic and transfer of methods between laboratories is difficult. The use of a PCR ELISA assay, which incorporates a target specific solution hybridization event, removes subjectivity and provides increased specificity. Also the incorporation of positive and negative controls in each set of tests facilitates the standardization of methods between laboratories. The investigation of naturally contaminated samples such as raw poultry and environmental samples for campylobacters by enrichment culture can yield C. jejuni and C. coli. But such methods can also yield other Campylobacter organisms such as C. fetus, C. lari (including urease positive isolates), C. hyointestinalis and Arcobacter spp., which are difficult to identify using phenotypic methods. However
Campylobacter PCR ELISA
the PCR ELISA assay described here is specific for C. jejuni and C. coli and can be completed in approximately 7 h. The PCR ELISA assay has been demonstrated to be capable of detecting one genome-equivalent therefore this assay may have the potential to detect C. jejuni and C. coli in foods and environmental samples. Further studies are currently underway to validate the assay for the detection of C. jejuni and C. coli in foods and environmental waters.
11.
12.
13.
ACKNOWLEDGEMENTS This work was supported by The Ministry of Agriculture, Food and Fisheries, UK (research programme FS1242). The authors would also like to thank Dr Ibrahim Attabay at Bristol University for the provision of the Campylobacter hyointestinalis isolates.
REFERENCES 1. Skirrow, M. B. (1994) Diseases due to Campylobacter; Helicobacter and related bacteria. Journal of Comparative Pathology 111, 113–49. 2. On, S. L. and Holmes, B. (1992) Assessment of enzyme detection tests useful in identification of Campylobacteria. Journal of Clinical Microbiology 30, 746–9. 3. Anon. (1993) Advisory Committee on the Microbiological Safety of Food Interim Report on Campylobacter. HMSO, London. 4. Bolton, F. J., Wareing, D. R. A., Skirrow, M. B. & Hutchinson, D. N. (1992) Identification and Biotyping of Campylobacters. In Identification Methods in Applied and Environmental Microbiology, (Board, R. G., Jones, D. & Skinner, F. A., eds). Pp. 151–62. Oxford, England: Blackwell Scientific Publishers. 5. Lior, H. (1984) New, extended biotyping scheme for Campylobacter jejuni, Campylobacter coli and “Campylobacter laridis”. Journal of Clinical Microbiology 20, 636–40. 6. Bolton, F. J., Holt, A. V. & Hutchinson, D. N. (1984) Campylobacter biotyping scheme of Epidemiological value. Journal of Clinical Pathology 37, 677–81. 7. Bej, A. K., Steffan, R. J., DiCesare, J., Haff, L. & Atlas, R. M. (1990) Detection of coliform bacteria in water by polymerase chain reaction and gene probes. Applied and Environmental Microbiology 56, 307–4. 8. Eyers, M., Chapelle, S., Van Camp, G., Goossens, H. & De Wachter, R. (1993) Discrimination among thermophilic Campylobacter species by polymerase chain reaction amplification of 23S rRNA gene fragments. Journal of Clinical Microbiology 31, 3340–3. 9. Cardarelli-Leite, P., Blom, K., Patton, C. M., Nicholson, M. A., Steigerwalt, A. G., Brenner, D. J., Barrett, T. J. & Swaminathan, B. (1996) Rapid identification of Campylobacter species by restriction fragment length polymorphism analysis of a PCR-amplified fragment of the gene coding for 16S rRNA. Journal of Clinical Microbiology 34, 62–7. 10. Hurtado, A. & Owen, R. J. (1997) A molecular scheme
14.
15.
16.
17.
18.
19.
20.
21.
22.
299
based on 23S rRNA gene polymorphism’s for rapid identification of Campylobacter and Arcobacter species. Journal of Clinical Microbiology 35, 2401–4. Gonzalez, I., Grant, K. A., Richardson, P. T., Park, S. F. & Collins, M. D. (1997) Specific identification of the enteropathogens Campylobacter jejuni and Campylobacter coli by using a PCR test based on the ceuE gene encoding a putative virulence determinant. Journal of Clinical Microbiology 35, 759–63. Stucki, U., Frey, J., Nicolet, J. & Burnens, A. P. (1995) Identification of Campylobacter jejuni on the basis of a species-specific gene that encodes a membrane protein. Journal Clinical Microbiology 33, 855–9. Van Doorn, L-J., Giesendorf, B. A. J., Bax, R., van der Zeijst, B. A. M., Vandamme, P. & Quint, W. G. V. (1997) Molecular discrimination between Campylobacter jejuni, Campylobacter coli, Campylobacter lari and Campylobacter upsaliensis by polymerase chain reaction based on a novel putative GTPase gene. Molecular and Cellular Probes 11, 177–85. Comi, G., Ferroni, P., Cocolin, L., Cantoni, C. & Manzano, M. (1996) Detection and identification of Campylobacter coli and Campylobacter jejuni by twostep polymerase chain reaction. Molecular Biotechnology 3, 266–8. Jackson, C. J., Fox, A. J. & Jones, D. (1996) A novel polymerase chain reaction assay for the detection and speciation of thermophilic Campylobacter spp. Journal of Applied Bacteriology 81, 467–73. Fredricks, D. N. & Relman, D. A. (1999) Application of the polymerase chain reaction to the diagnosis of infectious diseases. Clinical Infectious Diseases 29, 475–88. Giesendorf, B. A., Quint, W. G., Henkens, M. H., Stegeman, H., Huf, F. A. & Nesters, H. G. (1992) Rapid and sensitive detection of Campylobacter spp. in chicken products by using the polymerase chain reaction. Applied and Environmental Microbiology 58, 3804–8. Ritzler, M. & Altwegg, M. (1996) Sensitivity and specificity of a commercially available enzyme-linked immunoassay for the detection of polymerase chain reaction amplified DNA. Journal of Microbiological Methods 27, 233–8. Borrow, R., Claus, H., Guiver, M., Smart, L., Jones, D. M., Kaczmarski, E. B., Frosch, M. & Fox, A. J. (1997) Non-culture diagnosis and serogroup determination of meningococcal B and C infection by a sialyltransferase (siaD) PCR ELISA. Epidemiology and Infection 118, 111–7. Lawson, A. J., Logan, J. M., O’Neill, J., Desai, M. and Stanley, J. (1999) Largescale survey of Campylobacter species in human gastroenteritis by PCR and PCREnzyme-Linked Immunosorbent assay. Journal of Clinical Microbiology 37, 3860–4. O’Sullivan, N. A., Fallon, R., Carroll, C., Smith, T. and Maher, M. (2000) Detection and differentiation of Campylobacter jejuni and Campylobacter coli in broiler chicken samples using a PCR/DNA probe membrane based colorimetric detection assay. Molecular and Cellular Probes 14, 7–16. Ikuta, S., Takagi, K., Wallace, R. & Itakura, K. (1987) Dissociation kinetics of 19 base paired oligonucleotide-DNA duplexes containing single mismatched base pairs. Nucleic Acid Research 15, 797–811.
300
A. D. Sails et al.
23. Alderton, M. R., Korolik, P., Coloe, F. E., Dewhirst, F. E. & Paster, B. J. (1995) Campylobacter hyoilei sp. nov., associated with porcine proliferative enteritis. International Journal of Systematic Bacteriology 45, 61–6. 24. Vandamme, P., Van Doorn, L. J., al Rashid, S. T., Quint, W. G., van der Plas, J. and On, S. L. (1997) Campylobacter hyoilei Alderton et al. 1995 and Campylobacter coli Veron and Chatelain 1973 are subjective synonyms. International Journal of Systematic Bacteriology 47, 1055–60. 25. Parkhill, J., Wren, B. W., Mungall, K., Ketley, J. M., Churcher, C., Basham, D., Chillingworth, T., Davies, R. M., Feltwell, T., Holroyd, S., Jagels, K., Karlyshev, A. V., Moule, S., Pallen, M. J., Penn, C. W., Quail, M. A., Rajandream, M. A., Rutherford, K. M., van Vliet, A. H., Whitehead, S. & Barrel, B. G. (2000) The genome sequence of the foodborne pathogen
26. 27.
28.
29.
Campylobacter jejuni reveals hypervariable sequences. Nature 403, 665–8. Kwok, S. & Higuchi, R. (1989) Avoiding false positives with PCR. Nature 339, 237–8. Longo, M. C., Berninger, M. S. & Hartley, J. L. (1990) Use of uracil DNA glycosylase to control carry-over contamination in polymerase chain reactions. Gene 93, 125–8. Waage, A. S., Varund, T., Lund, V. and Kapperud, G. (1999) Detection of small numbers of Campylobacter jejuni and Campylobacter coli in environmental water, sewage, and food samples by a semi-nested PCR assay. Applied and Environmental Microbiology 65, 1636– 43. Winters, D. K. and Slavik, M. F. (1995) Evaluation of a PCR based assay for the specific detection of Campylobacter jejuni in chicken washes. Molecular and Cellular Probes 9, 307–10.