Sensitive Detection of Helicobacter pylori by Using ... - Europe PMC

JOURNAL OF CLINICAL MICROBIOLOGY, Jan. 1992, p. 192-200 0095-1137/92/010192-09$02.00/0 Copyright © 1992, American Society for Microbiology

Vol. 30, No. 1

Sensitive Detection of Helicobacter pylori by Using Polymerase Chain Reaction C. L. CLAYTON,'* H. KLEANTHOUS,l P. J. COATES, D. D. MORGAN,' AND S. TABAQCHALI' Department of Medical Microbiology' and Department of Histopathology,2 St. Bartholomew's Hospital Medical College, West Smithfield, London EC1A 7BE, United Kingdom Received 21 June 1991/Accepted 17 October 1991

A polymerase chain reaction (PCR) for the specific detection of Helicobacter pyloni was developed with a single primer pair derived from the nucleotide sequence of the urease A gene of H. pylori. We achieved specific amplification of a 411-bp DNA fragment in H. pyloni. After 35 cycles of amplification, the product could be detected by agarose gel electrophoresis and contained conserved single Hinfl and AluI restriction sites. This fragment was amplified in all 50 strains of H. pyloni tested, but it was not detected in other bacterial species, showing the PCR assay to be 100% specific. PCR DNA amplification was able to detect as few as 10 H. pyloni cells. PCR detected H. pylori in 15 of 23 clinical human gastric biopsy samples, whereas culturing and microscopy detected H. pylori in only 7 of the samples found to be positive by PCR. Additional primer pairs based on the urease genes enabled the detection of H. pylori in paraffin-embedded human gastric biopsy samples. The detection of H. pylori by PCR will enable both retrospective and prospective analyses of clinical samples, elucidating the role of this organism in gastroduodenal disease.

ods, however, require the use of 32P radioactivity and can take up to 3 days to provide autoradiographic results. To facilitate an understanding of H. pylori infection and the role of this organism in pathogenesis, we sought to improve current methods of detection. The polymerase chain reaction (PCR) can selectively amplify the copy number of a target gene more than 106-fold (37). PCR, therefore, has great potential for improving the ability to diagnose infectious diseases caused by fastidious or slow growing microorganisms, and it has already been used to detect many such bacterial pathogens, including Mycobacterium tuberculosis (12, 18), Mycobacterium leprae (32), Legionella pneumophila (24, 44), Borrelia burgdorferi (36), and Treponema pallidum (5). We previously cloned the urease structural genes of H. pylori and found them to be both highly conserved and useful as specific probes for the detection of H. pylori (8). In this study, we report the development of an extremely sensitive assay for H. pylori that is based upon the amplification of specific internal regions of the urease genes of H. pylori. The specificity and sensitivity of this assay are evaluated, and the direct detection of H. pylori in gastric biopsy samples is described. (Part of this work was presented at the Third International Meeting of the European Helicobacter pylori Study Group on Gastroduodenal Pathology and Helicobacter pylori, 8 to 10 November 1990, Toledo, Spain.) (Some of the information in this work is included in international patent PCT/GB90/01979.)

Helicobacter pylori is a gram-negative spirally shaped bacterium that has been implicated in the pathogenesis of active chronic gastritis and peptic ulcer disease of humans (4, 17). Since the eradication of H. pylori from the gastric mucosa alleviates symptoms (26, 33), the detection of and treatment for H. pylori would be beneficial for certain patient groups (46). A wide array of diagnostic tests have been developed to detect infection with this organism (10, 14, 16, 25, 29, 42), as it represents one of the most chronic diseases of humans. The 14C-urea breath test is a reliable noninvasive assay for the detection of an active H. pylori infection (25, 34), but it is costly and time-consuming and involves exposure to radioactivity. Serodiagnostic tests have been developed (14, 31) but do not allow the delineation of a current active H. pylori infection from a past H. pylori infection. Current diagnostic tests for H. pylori infection, however, still involve invasive gastric endoscopy and detection of the organism in gastric biopsy specimens. H. pylori can be identified in gastric tissue specimens by a urea hydrolysis test, staining techniques, and culturing (10, 16, 40, 42). The urease test is a rapid method for detecting H. pylori in biopsy specimens, but false-positive and false-negative results have been reported (3, 11). Microscopic analysis of stained biopsy smears is also a nonspecific test and has a low sensitivity for detecting this organism (39, 42). Culturing of H. pylori, which is fastidious and slow growing, is both difficult and time-consuming, requiring 3 to 7 days of incubation (15, 16). A reliable rapid means of detecting H. pylori in tissue samples would therefore be useful for the clinical management of patients with severe symptoms, enabling the prompt introduction of antibiotic therapy. Molecular biological techniques have been applied to the diagnosis of H. pylori infection, and both a genomic H. pylori DNA probe (50) and an oligonucleotide probe developed for an H. pylori-specific 16S rRNA sequence (28) have been found to detect as few as 104 H. pylori cells. These nucleic acid hybridization meth-

*

MATERIALS AND METHODS Bacterial strains and culture media. Most of the H. pylori strains used in this study were isolated routinely from human gastric biopsy samples collected from patients attending the Gastroenterology Clinic at St. Bartholomew's Hospital. The other H. pylori strains used were H. pylori NCTC 11637T isolated in Australia and strains isolated in Birmingham (C. McNulty) (27), Dublin (M. Crowe, St. James Hospital), and Cardiff (S. Gray, Public Health Laboratory Service [PHLS]) (8). The biopsy samples collected at St. Bartholomew's

Corresponding author. 192

VOL. 30, 1992

SENSITIVE DETECTION OF H. PYLORI BY USING PCR

Hospital were inoculated onto selective agar plates containing horse blood (8% [vol/vol]), vancomycin (10 ,ug/ml), colomycin (10 U/ml), trimethroprim (5 jig/ml), and amphotericin B (2 ,ug/ml) and incubated at 37°C under microaerophilic conditions for up to 7 days. The organisms were identified as H. pylori on the basis of morphology in Gram stains and by the oxidase, catalase, and rapid urease tests (23). All of the strains were maintained frozen at -70°C in Oxoid no. 2 broth containing 15% (vol/vol) horse serum and 15% (vol/vol) glycerol. Helicobacter mustelae W0831 and F6 were obtained, respectively, from D. Buckley (Beechams Pharmaceuticals, Epsom, United Kingdom) and D. Jones (PHLS, Manchester, United Kingdom). Campylobacter jejuni NCTC 50452, Campylobacter coli NCTC 50453, and Wolinella succinogenes NCTC 11488 were obtained from PHLS, Colindale, London, United Kingdom, and three urease-positive thermophilic campylobacter strains (8) were provided by F. Bolton (PHLS, Preston, United Kingdom). A urease-producing Clostridium sordellii strain (NCTC 6927) and a routine clinical isolate of Bacteroides ureolyticus (BU14; St. Bartholomew's Hospital) were also studied. The urease-positive members of the family Enterobacteriaceae studied were routine clinical isolates (St. Bartholomew's Hospital) identified by API 20E tests (API-bio Merieux Ltd., Basingstoke, United Kingdom). Working stocks of all strains were obtained by culturing on Columbia blood agar (GIBCO, Uxbridge, United Kingdom) at 37°C under appropriate atmospheric conditions. Synthetic oligonucleotides. All oligonucleotides were synthesized on an Applied Biosystems synthesizer (391-EP) by the automated phosphoramidite coupling method and purified as described by the manufacturer. Concentrations of oligonucleotides were determined spectrophotometrically. Oligonucleotides used as primers for the detection of H. pylori were derived from the sequenced urease genes (7) (EMBL accession number X17079). Primers HPU1 (5'GCC AATGGTAAATTAGTT3') and HPU2 (5'CTCCTTAAT TGTTTTTAC3') amplified a 411-bp product from urease gene A (nucleotides [nt] 304 to 714). Primers HPU54 (5'TG GGATTAGCGAGTATGT3') and HPU18 (5'CCCATT TGACTCAATG3') amplified a 132-bp product from urease gene B (nt 1971 to 2102). Primers HPU55 (5'AATTGCA GAAATATCAC3') and HPU17 (5'ACTTTATTGGCTG GTTT3') amplified a 115-bp product also from urease gene B (nt 2284 to 2398). Primer HPUI1 (5'ATTGACATTGGCGG TAAC3') (nt 559 to 576) internal to the 411-bp amplified product obtained with primers HPU1 and HPU2 was used as a probe for Southern hybridization. Primers HPUI54 (5'AA CATGATCATCAAAGGC3') (nt 2059 to 2076) and HPUI55 (5'CACATTGAAGTCAATTCT3') (nt 2325 to 2340) internal to the 132- and 115-bp products obtained with HPU54 and HPU18 and with HPU55 and HPU17, respectively, were also used as probes for Southern hybridization. The human 3-globulin gene primers BGLO1 (5'ACACAACTGTGTT CACTAGC3') and BGLO2 (5'CAACTTCATCCACGTTC ACC3') amplified a 110-bp fragment from DNA released from human tissue (9, 38). Southern hybridization. Oligonucleotide primer HPUI1 was labelled at the 5' end with 32p ([32P]ATP; Amersham Ltd., Amersham, United Kingdom) and 20 U of T4 polynucleotide kinase in a buffer containing 50 mM Tris-HCl (pH 7.4), 10 mM MgCl2, and 1 mM dithiothreitol. The solution was incubated at 37°C for 1 h, and the labelled probe was separated from the unincorporated isotope by ethanol precipitation and washing in 70% (vol/vol) ethanol. Amplified DNA was transferred from an agarose gel to a

193

nylon membrane (Hybond N; Amersham Ltd.) by the method of Southern (43). The membrane was prehybridized for 4 h at 42°C in a hybridization buffer containing 6x SSC (lx SSC is 0.15 M NaCl plus 0.015 M sodium citrate), 1Ox Denhardt's solution (0.2% [wt/vol] bovine serum albumin, 0.2% [wt/vol] polyvinylpyrrolidone, 0.2% [wt/vol] Ficoll), and 50 ,ug of denatured salmon sperm DNA per ml. The labelled oligonucleotide probe was added to the hybridization buffer at a concentration of 1 ng/ml, and hybridization was carried out for 14 h at 5°C below the melting temperature of the probe. Following hybridization, the membrane was washed twice for 5 min in 6x SSC-0.1% (wt/vol) sodium dodecyl sulfate at room temperature and once for 5 min at the melting temperature of the probe. It was exposed to X-ray film at -70°C for 14 h, and the film was developed. Extraction of target DNA. Bacterial genomic DNA was extracted as described previously (8) from H. pylori 630, H. mustelae W0831, W. succinogenes NCTC 11488, C. jejuni NCTC 50452, C. coli NCTC 50453, urease-positive campylobacter strain 83/12830/1, Proteus mirabilis S, Morganella morganii J, Providencia rettgeri C, and Klebsiella pneumoniae W. Suspensions of each bacterial strain were prepared in sterile distilled water (200 ,ul) by inoculation from agar plates with a standard loop. The samples were boiled for 10 min and centrifuged for 5 min at 14,000 x g. One microliter of the supernatant or 50 ng of purified DNA was used as the target DNA in the amplification assay. Supernatants of boiled cells of Ureaplasma urealyticum and Ureaplasma diversum were provided by J. Willoughby (St. Andrew's University, Fife, United Kingdom). H. pylori 630 was used to prepare serial 10-fold dilutions for viable count estimations on Columbia blood agar plates, and the dilutions were also processed as described above for PCR analysis. Clinical samples. Human antral gastric biopsy specimens were dissected and added to 50 ,ul of sterile distilled water. The biopsy specimens were vortexed, boiled for 10 min, and centrifuged for 5 min at 14,000 x g. A negative biopsy specimen was processed and used for amplification with each batch of biopsy specimens prepared and tested by PCR. Paraffin-embedded human antral gastric biopsy specimens were prepared by immersing the tissue in 10% (vol/vol) formalin saline and embedding the tissue in paraffin wax on an automatic tissue processor by standard histological techniques (1). The embedded biopsy specimens were analyzed by cutting sections (4 by 10 ,um) on a microtome. Sterile distilled water (300 ,ul) was added to the sections, and the sections were boiled for 15 min. The samples were cooled on ice for 5 min and centrifuged for 5 min at 14,000 x g, and the supernatants were collected and stored at -20°C (9). A suspension of H. pylori 630 (105 cells per ml) was solidified in agar and embedded in paraffin wax for use as a PCR positive control. A negative paraffin wax section was processed in between the clinical samples to control for cross-contamination on cut sections. The microtome blade was wiped with 1 M HCI to control for contamination. Forty microliters of the biopsy supernatants were used in the PCR assay. Routine diagnostic tests. Biopsy specimens tested by PCR were also inoculated onto selective agar for culturing as described above. Smears were prepared from the biopsy specimens on slides, stained with a modified Gram stain (10), and examined by light microscopy. For the urease test, a crushed portion of each biopsy specimen was inoculated into 500 p.l of buffered 2% (wt/vol) Christensen's urea broth and incubated at 37°C, as were positive and negative controls, and any color change was noted after 4 and 24 h. Biopsy specimens separate from those analyzed above

194

CLAYTON ET AL.

J. CLIN. MICROBIOL.

1 2 3 4 5 6 7 891012314 r

_1018

6%

P"

10

.;...,iA:. _f

1

2

3

4

5

6

7

A 8

9

10 11 12 13 14

B FIG. 1. Detection of H. pylori by DNA amplification and hybridization. Shown are agarose gel electrophoresis (A) and corresponding Southern hybridization (B) of amplified PCR products from pTCP3 plasmid DNA (1 ng) (lanes 1), serial 10-fold dilutions of H. pylori 630 (106 cells to 1 cell) (lanes 2 to 8), and human gastric biopsy samples (lanes 9 to 14). Arrows indicate the 411-bp product obtained with primers HPU1 and HPU2 and by hybridization to oligoprobe HPUI1. The limit of detection was 10 to 100 whole bacterial cells. The marker used was a 1-kb ladder of molecular weight standards (GIBCO).

were collected for an endoscopy rapid urease test (RUT) (45) and for histological examination. The endoscopy RUT was performed in the endoscopy theater by adding the biopsy specimens to 0.5 ml of 10% (wt/vol) urea in deionized water containing phenol red indicator. A positive result was recorded when there was a color change from yellow to pink within 1 min. The biopsy specimens used for histological examination were processed in the Department of Histopathology. They were Formalin fixed and paraffin embedded, and sections were cut and stained by standard techniques (1). The sections were examined for evidence of gastritis and the presence of H. pylori. H. pylori was detected by use of fast cresyl violet stain. PCR. The template DNA was added to 100 ,u of a reaction mixture containing 10 mM Tris-HCl (pH 8.3), 50 mM KCI, 1.5 mM MgCl2, 0.01% (wt/vol) gelatin, 0.2 mM each deoxynucleotide (Pharmacia Ltd., Milton Keynes, United Kingdom), and 0.5 p.M each oligonucleotide primer. Taq polymerase (2.5 U; Perkin-Elmer Corp., Norwalk, Conn.) was added, and the reaction mixture was overlaid with 100 ,ul of mineral oil to prevent evaporation. PCR was performed with an automatic thermal cycler (Hybaid Ltd., Twickenham, Middlesex, United Kingdom). The amplification cycle consisted of an initial denaturation of target DNA at 95°C for 5 min followed by denaturation at 94°C for 1 min, primer annealing at 45°C for 1 min, and extension at 72°C for 1 min. The final cycle included extension for 5 min at 72°C to ensure full extension of the product. Samples were amplified through 35 consecutive cycles. Paraffin-embedded samples were subjected to an initial 25 cycles of amplification. Aliquots of 5 plI were removed, added to 95 of a fresh reaction mixture, and subjected to a further 25 cycles of amplification. The completed reactions were analyzed by electrophoresis of a 20-pd aliquot through 0.7 to 1.5% (wt/vol) agarose gels containing 0.5 pug of ethidium bromide per ml, and the bands were visualized by excitation under UV light. Reagent negative control reactions were performed with each batch of amplifications and consisted of tubes containing distilled water in place of the DNA samples. For reducing the

occurrence of false-positive results, separate designated pipettes were used for setting up reactions and for post-PCR work, reagents were divided into aliquots, disposable gloves were worn, premixtures were prepared, and the DNA was added last, as outlined by Kwok and Higuchi (22). With each set of clinical PCR amplifications a corresponding amplification was performed with H. pylori template DNA obtained from cells together with a volume of DNA supernatant extracted from the clinical samples to control for falsenegative reactions due to inhibitory substances. Restriction endonuclease digestions. Samples (5 Pdl) of amplification products obtained by PCR were subjected to restriction endonuclease digestion for 2 h at 37°C in 20-pul volumes as recommended by the manufacturer. The digested samples were analyzed by agarose (1.5 to 4.0% [wt/vol]) gel electrophoresis.

RESULTS Development of the PCR assay. The positions of primers HPU1 and HPU2 within the nucleotide sequence of the urease A gene predicted that a 411-bp fragment would be generated following PCR of either H. pylori chromosomal DNA or a recombinant plasmid containing the cloned gene. A fragment of the appropriate size was visualized following PCR of released DNA obtained from H. pylori 630 and from the urease clone pTCP3 (8) (Fig. 1A). Sensitivity of the PCR assay. The sensitivity of the PCR assay was investigated with serial dilutions of bacterial cells of H. pylori 630. The lowest numbers of bacterial cells detected, as determined by serial dilution and plating of samples on agar, ranged from 10 to 100 CFU (Fig. 1A, lanes 2 to 7). The same level of sensitivity was also found for mixed cultures containing H. pylori cells (results not shown). Southern hybridization with the internal oligonucleotide probe HPUI1 reproducibly increased the sensitivity of detection 10-fold, allowing the detection of 10 H. pylori cells (Fig. 1B, lane 7). Specificity of the PCR assay. Supernatants of 50 boiled H. pylori strains examined all yielded the 411-bp amplified


VOL. 30, 1992

195

TABLE 1. Specificity of PCR with H. pylori urease gene oligonucleotide primersa Presence (+) or absence (-) of the following PCR 132 bp (HPU1-HPU2) (HPU54-HPU18)

Bacterial species (no. of strains tested)

product':

411 bp

H. pylori NCTC 11637T (1) H. pylori clinical isolates (49) H. mustelae W0831 and F6 (2) C. jejuni NCTC 50452 (1) C. jejuni clinical isolates (10) C. coli NCTC 50453 (1) C. coli clinical isolates (10) W. succinogenes NCTC 11488 (1) Urease-positive thermophilic campylobacters (3) Other urease-positive organisms C. sordellii NCTC 6927 (1) B. ureolyticus BU14 (1) P. mirabilis clinical isolates (4) M. morganii clinical isolates (4) P. rettgeri clinical isolates (4) K. pneumoniae clinical isolates (4) U. urealyticum clinical isolates (7) U. diverus animal isolates (7)

115 bp

(HPU55-HPU17)

+ +

+ +

+ +

-

-

-

-

-

-

H. pylori strains and strains of 12 other species and urease-positive thermoph c campylobacters were tested. Designations in parentheses are primer pairs used in PCR.

a Fifty b

product with HPU1 and HPU2 on agarose gel electrophoresis. W. succinogenes, C. jejuni, C. coli, and urease-positive bacteria, such as H. mustelae and other species, were found not to yield amplified products from supernatants (Table 1 and Fig. 2A) or from purified DNA (results not shown). Primer pair HPU54-HPU18 amplified a 132-bp product of the urease gene from H. pylori, and primer pair HPU55-HPU17 amplified a 115-bp product. These primer pairs were also found to be specific for H. pylori (Table 1 and Fig. 2B). Southern hybridization with internal probe HPUI1 was used to confirm that the 411-bp amplified sequence was part of the urease gene of H. pylori (Fig. 1B). The 411-bp amplified

fragment of H. pylori was found to contain conserved internal AluI and Hinfl restriction enzyme sites. The 411-bp amplified fragment obtained from nine clinical H. pylori isolates on digestion with AluI yielded 347- and 64-bp products (Fig. 3A), and digestion with Hinfl yielded 277- and 134-bp products (Fig. 3B). PCR assay of gastric biopsy samples. PCR with primers HPU1 and HPU2 enabled the detection of H. pylori directly from biopsy samples. In an initial study, culture-positive antral gastric biopsy samples were also detected by PCR and subsequent agarose electrophoresis and confirmed by Southern hybridization with oligonucleotide probe HPUI1 (Fig. 1, lanes 9 and 13). The biopsy sample found to be positive by

1 2 3 4 5 6 7 8 9 1011 1213 14 1 2 3 4 5 6 7 8 9 10 506

411X/:

A

394

394z-

220

-

B

132-0 FIG. 2. Specific detection of H. pylori by PCR. Shown is agarose gel electrophoresis of amplified PCR products obtained from cells of H. pylori NCTC 11637T, H. mustelae W0831 and F6, P. mirabilis S, M. morganii J, P. retgerii C, K. pneumoniae W, Yersinia enterocolitica B2, urease-positive campylobacter strain 83/12830/1, C. jejuni NCTC 50452, C. coli NCTC 50453, and W. succinogenes NCTC 11488 (lanes 3 to 14, respectively) with urease primers HPUl and HPU2 (A) and HPU54 and HPU18 (B). Arrows indicate 411- and 132-bp amplified fragments obtained only with H. pylori (lane 3). Lane 2, PCR negative control. Lane 1, 1-kb ladder of molecular weight standards.

277--

344

A

298

B

134-0 134

FIG. 3. Agarose gel electrophoresis of restriction enzyme-digested amplified PCR products obtained from cells of H. pylori. (A) AluI-digested 411-bp product (HPU1-HPU2 primer pair) of H. pylori NCTC 11637T (lane 1), H. pylori 630 (lane 2), and seven clinical H. pylori strains cultured from gastric biopsy samples (lanes 3 to 9) (Table 2). (B) HinfI-digested 411-bp product of the same strains. Arrows indicate 347- and 64-bp fragments obtained with AluI and 277- and 134-bp fragments obtained with Hinfl. The 64-bp AluI fragment is just visible. Lane 10, 1-kb ladder of molecular weight standards.

196

CLAYTON ET AL.

J. CLIN. MICROBIOL.

TABLE 2. Comparison of PCR and diagnostic tests for H. pylori detection in antral gastric biopsy specimens

Resultb determined by: Patient

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Age (yr)

Clinical diagnosisa

81 67 27 39 60 82 57 46 83 74 70 65 59 77 51 20 63 48 29 39 40 70 43

G Gu An Duo Dys An Dys Dysp Dys An Cancer An Dys G Dysp Dysp Dys G Dysp, DU Dysp Dysp G Dysp

Gram

Urease

staining

test

+ + -

+ + -

+ -

+

+

-

+

+ + + -

+ +

PCR

+ + + + -

+ + + +

+ + +

+

-

-

Bacteriological cultunng

RUT

Diagnosisc

+ + -

+ + -

CG CG N

+

+

+

-

+ + + +

+ +

CG N

+ + -

+ +

-

-

+

-

NT NT + NT NT NT NT

+

+ +

-

NT

+

-

NTd

+ a G, gastritis; Gu, gastric ulcer; An, anemia; Duo, duodenitis; Dys, dysphagia; Dysp, dyspepsia; DU, duodenal ulcer. b +, positive; -, negative. I CG, chronic gastritis; MG, mild gastritis; N, normal gastric mucosa. d NT, not tested.

Southern hybridization (Fig. 1B, lane 13) was also found to yield a visible amplified fragment by further amplification of the initial PCR sample (results not shown). The biopsy sample found to be positive by PCR and visualized in Fig. 1, lane 9, was found to have culturable H. pylori that was also detectable by the urease test. The biopsy samples in Fig. 1, lanes 10 to 12 and 14, were found to be negative by the urease test and bacteriological culturing. Antral gastric biopsy samples obtained from 23 other patients referred for upper gastrointestinal tract endoscopies were analyzed for H. pylori by PCR, Gram staining, urease testing, and bacteriological culturing (Table 2). All patients were symptomatic and mostly had upper abdominal discomfort or pain; clinical preendoscopic diagnoses included peptic ulcer disease, gastritis, and nonulcerative dyspepsia. PCR with primers HPU1 and HPU2 detected H. pylori in 15 of the biopsy samples. The PCR results obtained for biopsy samples 1 to 11 are shown in Fig. 4A, lanes 3 to 13. The positive control (boiled H. pylori 630 cells) and the negative control are shown in Fig. 4A, lanes 1 and 2, respectively. The amplified products were confirmed to originate from H. pylori by Southern hybridization with oligonucleotide probe HPUI1 (Fig. 4B). Southern analysis on this occasion did not detect H. pylori in any biopsy samples that did not yield visible amplified products. Further confirmation was obtained by digestion of the amplified products with AluI and Hinfl (results not shown). None of the clinical biopsy samples tested were found to contain inhibitors of Taq polymerase. The seven H. pylori strains cultured from these biopsy samples were used for the PCR assay and restriction enzyme digestion shown in Fig. 3. Bacteriological culturing and microscopy detected H. pylori in 7 of 23 (30%) of the biopsy samples, and these were all

Histological examination

H. pylori

+ +

NT +

+ +

MG CG CG CG CG N CG CG CG CG N N CG N CG CG

NT

found to be positive by PCR (Table 2). The culturing and microscopy diagnostic methods were correlated for five of these seven positive biopsy samples. The urease test detected H. pylori in only 3 of 23 (13%) of the biopsy samples (Table 2). This assay did not detect the culturable H. pylori found in five biopsy samples and did not detect the visible H. pylori seen in four biopsy samples by Gram staining and light microscopy. The endoscopy RUT was positive for 12 of 15 (80%) of the biopsy samples, and H. pylori was detected by histological staining in 10 of 21 (48%) of the biopsy samples. Histological examination revealed that 15 of 21 (71%) patients examined had mild to chronic'gastritis and that 6 patients exhibited normal gastric mucosa. PCR assay of paraffin-embedded gastric biopsy samples. Paraffin-embedded gastric biopsy samples taken from five patients in 1988 and stored for at least 2 years were analyzed for H. pylori by use of boiled paraffin section supernatants for PCR without previous dewaxing. Primer pair HPU1HPU2 did not yield amplified products with the paraffin sections despite double PCR amplification. However, four of the five samples yielded the expected 132-bp product with primer pair HPU54-HPU18 (Fig. 5A, lanes 3 to 6) and the expected 115-bp product with primer pair HPU55-HPU17 (Fig. SB, lanes 3 to 6). The paraffin-embedded H. pylori 630 positive control and a negative control are shown in Fig. 5, lanes 2 and 9. The simple boiling DNA extraction procedure yielded DNA of sufficient amount and quality to yield amplified products with human p-globulin primers. The 110-bp PCR product amplified from section 4593 with the p-globulin primers is shown in Fig. 5A and B, lane 1. The amplified products were confirmed to originate from H. pylori by both Southern hybridization and restriction endonuclease digestion. Internal oligonucleotide probe

|l-~ ~ ~506

VOL. 30, 1992


197

1 2 3 4 5 6 7 8 9 10 11 12 13 14

~

~

1

2 3 4 5 6 87

9 1011 12 13 14

A6e41! ~ ~s :

* -394

_~~~~~~~~~~9

~~~~~~~~~B

A

FIG. 4. PCR detection of H. pylori in human gastric biopsy samples. Shown are agarose gel electrophoresis (A) and corresponding Southern hybridization (B) of amplified PCR products obtained from gastric biopsy samples collected from patients 1 to 11 (Table 2) (lanes 3 to 13, respectively). Arrows indicate the 411-bp product obtained with primers HPU1 and HPU2 and by hybridization to oligoprobe HPUI1 (lanes 3, 4, 6, and 9 to 13). H. pylori 630 cells were used as a positive control (lanes 2), and the PCR negative control is shown in lanes 1. Lanes 14, 1-kb ladder of molecular weight standards.

HPUI54 or HPUI55 was used for Southern analysis; Fig. 5C shows hybridization of the HPUI55 probe to the amplified products obtained with HPU55 and HPU17. Digestion of the 132-bp product of HPU54 and HPU18 with Hinfl yielded two 1 W..

2 3 4

5

6 7 8 9

.?

220 132 --- b

A

-154

60-bp fragments (Fig. 6A), and digestion of the 115-bp product of HPU55 and HPU17 with FokI yielded 77- and 38-bp fragments (Fig. 6B). Two of the patients found to be positive for H. pylori by PCR had gastric ulcers (Fig. 5, lanes 3 and 5), and the two other patients had gastritis (Fig. 5, lanes 4 and 6). The patient found to be negative for H. pylori by PCR had a gastric ulcer. H. pylori was detected by histological staining in three of the four patients found to be positive by PCR, and the biopsy sample from the patient with the gastric ulcer and found to be negative for H. pylori by PCR was found not to have detectable H. pylori. The processed sections were all found not to contain any inhibitors of PCR.

DISCUSSION

B

134

115---

--75 1

2

3

4

5

6

7

8

.4F

we:::>

FIG. 5. PCR detection of H. pylori in stored paraffin-embedded human gastric biopsy samples. Shown is agarose gel electrophoresis of amplified PCR products obtained from paraffin-embedded sections 4593, 10412, 7213, 4591, and 1841 (lanes 3 to 7, respectively) with primers HPU54 and HPU18 (A) and primers HPU55 and HPU17 (B). Paraffin-embedded H. pylori 630 cells were used as a positive control (lanes 2), and the PCR negative control is shown in lane 9. A PCR product of 110 bp amplified from section 4593 by human 3-globulin primers BGLO1 and BGLO2 is shown in lanes 1. Arrows indicate the 132- and 115-bp fragments obtained with the primer pairs. Lanes 8, 1-kb ladder of molecular weight standards. (C) Southern hybridization of panel B with oligoprobe HPUI55.

A rapid, sensitive, and specific test for H. pylori would be of great value because of the clinical importance of this pathogen and the large number of laboratory identifications being routinely undertaken around the world on a daily basis. The aims of this study were to develop a simple method for the extraction of H. pylori DNA present in clinical biopsy samples and to develop a specific and sensitive PCR by use of primers based on the sequences of the H. pylori urease genes. As the H. pylori urease enzyme seems to be required for the survival of this organism in the acidic gastric environment, there is probably a strong selective pressure to maintain the amino acid sequence of this enzyme, resulting in the observed conservations of the DNA sequence among strains (8). PCR primers HPU1 and HPU2 were designed to avoid regions of homology with the sequenced structural urease genes of P. mirabilis (20), K. pneumoniae (29), and U. urealyticum (2). The other primer pairs used in this study were random primers previously used for sequencing the urease genes of H. pylori (7). Studies with H. pylori cells revealed a reproducible sensitivity of about 10 to 100 cells with a simple boiling DNA extraction procedure for PCR. There was a slight increase in sensitivity in Southern blot analysis with 32P-labelled internal probe HPUI1. Dot blot analysis may increase the sensitivity, as has been described for L. pneumophila (44). The

198

CLAYTON ET AL.

J. CLIN. MICROBIOL. 1 2

3 4 5 6

7

1

2

3

4

5

6

7

::

--

134

115-0-

77-...-

--134

B A FIG. 6. Agarose gel electrophoresis of restriction enzyme-digested amplified PCR products obtained from paraffin-embedded gastric biopsy samples 4593, 10412, 7213, and 4591 (lanes 2 to 5, respectively). (A) Products obtained with primers HPU54 and HPU18 and digestion with Hinfl. (B) Products obtained with primers HPU55 and HPU17 and digestion with FokI. Arrows indicate the 60-bp doublet fragment obtained by Hinfl digestion of the 132-bp product (HPU54-HPU18; lane 6) (A) and the 77-bp fragment obtained by FokI digestion of the 115-bp product (HPU55-HPU17; lane 6). The digested amplified products of the positive control H. pylori 630 cells are shown in lanes 2. Lane 7, 1-kb ladder of molecular weight standards. The 115-bp product was not digested to completion with FokI, and the 38-bp fragment is too small to be seen.

primer pairs used in PCR specifically detected H. pylori; other urease-producing species did not yield any amplified products, even with the sets of random primers. There is a need, however, to test the primer pairs HPU1-HPU2, HPU54-HPU18, and HPU55-HPU17 for specificity against a number of other bacterial species, especially those found in the upper respiratory flora, as these are often the major contaminants of biopsy samples. Blanchard (2) has similarly used sets of primers based on the urease genes of U. urealyticum for both PCR detection and biotyping of this organism. The reference "gold standard" used for the comparative detection of H. pylori in clinical biopsy samples was bacteriological culturing. PCR detected H. pylori in all seven biopsy samples found to contain culturable H. pylori. Seven further samples found to be negative by culturing, Gram staining, and the urease test were found to be positive by PCR. These samples found to be positive by PCR may reflect the high sensitivity of this method and may have contained low numbers of nonviable or viable cells which were below the levels of detection of the routine diagnostic tests. However, a major drawback of PCR is its extreme sensitivity, which can lead to false-positive reactions due mainly to sample contamination by PCR product "carryover" (22). The urease test appeared to have a low sensitivity in detecting H. pylori when compared with the other diagnostic techniques. This low sensitivity has been reported by others (3, 11, 13, 45, 47), and the false-negative results in our study are probably due to low numbers of bacteria within the specimens. The endoscopy RUT appeared to be much more sensitive in detecting H. pylori than did the routinely used urease test, and three biopsy specimens that were found to be H. pylori negative by the other assays were found to be positive by the endoscopy RUT. Histological staining of biopsy specimens with cresyl violet detected H. pylori in five specimens that did not show H. pylori in Gram stains. However, the endoscopy RUT and histological examinations were performed on biopsy specimens separate from those used for the other diagnostic tests and may have merely reflected the reported patchy colonization of the gastric mucosa by H. pylori (16, 51). Three of the six patients with a normal gastric mucosa were diagnosed not to have H. pylori by PCR and the other diagnostic assays. PCR, however, detected H. pylori in two of the patients found not to have H. pylori by the other tests and to exhibit normal gastric mucosa. These results may represent true falsepositive PCR results, although the appropriate PCR controls

were negative, and as others have reported that there is not a total correlation between chronic gastritis and H. pylori (30), the results obtained may be correct. The lack of total agreement between chronic gastritis and PCR in our patients

is similar to the results obtained by Nedenskov-Sorensen et al. (30), who found nine patients without signs of chronic gastritis to be colonized with H. pylori and four patients with chronic gastritis not to be infected with H. pylori. It is likely that the PCR DNA amplification assay shows a superior sensitivity to the current diagnostic assays for the detection of H. pylori in gastric biopsy samples, rather than that the results are false positives, and further clinical studies with multiple biopsy samples should help in evaluation. In our future studies, a lack of correlation between PCR and the other diagnostic methods will be reevaluated by repeating the PCR with other primer pairs. Our results are similar to those recently reported by Valentine et al. (48), who used PCR to detect H. pylori in both gastric biopsy and gastric aspirate samples. PCR detection of H. pylori in their biopsy samples showed a good correlation with detection by routine diagnostic methods. They used PCR oligonucleotides that were based on a randomly cloned genomic DNA fragment, and the sensitivity of their method in detecting H. pylori cells agreed with the results of our study. However, these workers used a time-consuming DNA extraction procedure which is not really applicable to routine clinical laboratory use, and we have found the simple boiling DNA extraction procedure to be as sensitive. We recommend the use of primer pair HPU1-HPU2 to amplify H. pylori DNA in boiled clinical gastric biopsy samples, visualization of the specific 411-bp amplified sequence on ethidium bromide-stained agarose gels and, if required, verification by restriction enzyme digestion with Hinfl or AluI. This procedure avoids the use of radioactivity for confirmation of the amplified product and provides a same-day result. Primers HPU1 and HPU2 did not yield the 411-bp amplified product with the boiled paraffin-embedded biopsy supernatants. However, for four of the five embedded biopsy supernatants, the correctly sized, smaller fragments of 132 and 115 bp were yielded by the two other primer pairs. A similar result has been reported by Coates et al. (9), who found that a small, 110-bp fragment of the human ,-globulin gene was consistently amplified in boiled paraffin-embedded nasopharyngeal sections by PCR but that a larger, 355-bp fragment of this gene was not amplified. The DNA obtained

VOL. 30, 1992


from paraffin-embedded samples is usually of poor quality because of fixation effects (19, 35, 41), and the simple boiling DNA extraction procedure probably also fragments the DNA so that only products of small sizes can be amplified. Treatment of biopsy samples with proteinase K and extraction of DNA should yield more intact DNA (19) and also result in an increase in the sensitivity of the PCR for the detection of H. pylori. Such treatment of paraffin-embedded gastric biopsy samples is currently being evaluated along with the simple boiling DNA extraction method. Valentine et al. (48) also reported on the amplification of H. pylori sequences from a single paraffin-embedded biopsy sample, although the level of detection was poor despite extensive sample processing prior to PCR. T. pallidum sequences from paraffin-embedded tissue were similarly amplified recently (5). The detection of H. pylori in paraffin-embedded biopsy samples by PCR will enable a retrospective analysis of clinical samples and elucidate the role of this organism in gastroduodenal disease. PCR amplification of H. pylori DNA sequences has the potential to be a rapid and highly sensitive and specific method for the laboratory diagnosis of H. pylori infection. The technique could be used to quickly predict a relapse of infection after antibacterial therapy of H. pylori-infected patients, as has been reported for the monitoring of chlamydial infection (6). PCR analysis of clinical biopsy samples with the boiling DNA extraction method is a convenient and sensitive method for routine use. We are currently evaluating a nested primer amplification approach that has been reported recently for M. leprae detection by PCR (32); this approach uses primers internal to HPU1 and HPU2 to increase the specificity and sensitivity of PCR for the diagnosis of H. pylori-associated disease. The PCR technique would be more applicable to routine use in clinical laboratories if products could be detected by fluorometric or colorimetric assays, providing a focus for future PCR research. PCR should provide a rapid means of detecting viable but nonculturable H. pylori that has been suggested to exist in saliva, fecal, and environmental samples (21, 49) and should allow the determination of the source and route of transmission of this important pathogen. ACKNOWLEDGMENTS This work was supported by the Medical Research Council. We thank Ann Carey for typing the manuscript. REFERENCES 1. Bancroft, J. D., and A. Stevens. 1990. Theory and practice of histological techniques. Churchill Livingstone, London. 2. Blanchard, A. 1990. Ureaplasma urealyticum urease genes; use of a UGA tryptophan codon. Mol. Microbiol. 4:669-676. 3. Borromeo, M., J. R. Lambert, and K. J. Pinkard. 1987. Evaluation of "CLO-Test" to detect Campylobacter pyloridis in gastric mucosa. J. Clin. Pathol. 40:462-463. 4. Buck, G. E. 1990. Campylobacter pylori and gastroduodenal disease. Clin. Microbiol. Rev. 3:1-12. 5. Burstain, J. M., E. Grimprel, S. A. Lukehart, M. V. Norgard, and J. D. Radolf. 1991. Sensitive detection of Treponema pallidum by using the polymerase chain reaction. J. Clin. Microbiol. 29:62-69. 6. Claas, H. C. J., J. H. T. Wagenvoort, H. G. M. Niesters, T. T. Tio, J. H. Van Riisoort-Vos, and W. G. V. Quint. 1991. Diagnostic value of the polymerase chain reaction for chlamydia detection as determined in a follow-up study. J. Clin. Microbiol.

29:42-45.

7. Clayton, C. L., M. J. Pallen, H. Kleanthous, B. W. Wren, and S. Tabaqchali. 1990. Nucleotide sequence of two genes from

199

Helicobacter pylori encoding for urease subunits. Nucleic Acids Res. 18:362. 8. Clayton, C. L., B. W. Wren, P. Mullany, A. Topping, and S. Tabaqchali. 1989. Molecular cloning and expression of Campylobacter pylori species-specific antigens in Escherichia coli K-12. Infect. Immun. 57:623-629. 9. Coates, P. J., A. J. d'Ardenne, G. Kahn, H. 0. Kangro, and G. Slavin. 1991. Simplified procedures for applying the polymerase chain reaction to routinely fixed paraffin wax sections. J. Clin. Pathol. 44:115-118. 10. Coudron, P. E., and D. F. Kirby. 1989. Comparison of rapid urease tests, staining techniques, and growth on different solid media for detection of Campylobacter pylori. J. Clin. Microbiol. 27:1527-1530. 11. Das, S. S., L. A. Bain, Q. N. Karim, L. G. Coelho, and J. H. Baron. 1987. Rapid diagnosis of Campylobacter pyloridis infection. J. Clin. Pathol. 40:701-702. 12. Eisenach, K. D., M. D. Cave, J. H. Bates, and J. T. Crawford. 1990. Polymerase chain reaction amplification of a repetitive DNA sequence for Mycobacterium tuberculosis. J. Infect. Dis. 161:977-981. 13. Engstrand, L., C. Pahlson, S. Gustavsson, and A. Schwan. 1986. Monoclonal antibodies for rapid identification of Campylobacter pyloridis. Lancet ii:1402-1403. 14. Evans, D. J., Jr., D. G. Evans, D. Y. Graham, and P. D. Klein. 1989. A sensitive and specific serologic test for detection of Campylobacter pylori infections. Gastroenterology 96:10041008. 15. Goodwin, C. S., and J. A. Armstrong. 1990. Microbiological aspects of Helicobacter pylori (Campylobacter pylori). Eur. J. Clin. Microbiol. Infect. Dis. 9:1-13. 16. Goodwin, C. S., E. D. Blincow, J. R. Warren, T. E. Waters, C. R. Sanderson, and L. Easton. 1985. Evaluation of cultural techniques for isolating Campylobacter pyloridis from endoscopic biopsies of gastric mucosa. J. Clin. Pathol. 38:1127-1131. 17. Graham, D. Y. 1989. Campylobacter pylori and peptic ulcer disease. Gastroenterology 96:615-625. 18. Hance, A. J., B. Grandchamp, V. Levy-Frebault, D. Lecossier, J. Rouzier, D. Bocart, and B. Gicquel. 1989. Detection and identification of mycobacteria by amplification of mycobacterial DNA. Mol. Microbiol. 3:843-849. 19. Jackson, D. P., F. A. Lewis, G. R. Taylor, A. W. Boylston, and P. Quirke. 1990. Tissue extraction of DNA and RNA and analysis by the polymerase chain reaction. J. Clin. Pathol. 43:499-504. 20. Jones, B. D., and H. L. T. Mobley. 1989. Proteus mirabilis urease: nucleotide sequence determination and comparison with jack bean urease. J. Bacteriol. 171:6414 6422. 21. Krajden, S., M. Fuksa, J. Anderson, J. Kempston, A. Boccia, C. Petrea, C. Babida, M. Karmali, and J. L. Penner. 1989. Examination of human stomach biopsies, saliva, and dental plaque for Campylobacter pylori. J. Clin. Microbiol. 27:1397-1398. 22. Kwok, S., and R. Higuchi. 1989. Avoiding false positives with PCR. Nature (London) 339:237-238. 23. Langenberg, M. L., G. N. Tytgat, M. E. I. Schipper, P. J. G. M. Reitra, and H. C. Zanen. 1984. Campylobacter-like organisms in the stomach of patients and healthy individuals. Lancet i:1348. 24. Mahbubani, M. H., A. K. Bej, R. Miller, L. Haff, J. Dicesare, and R. M. Atlas. 1990. Detection of Legionella with polymerase chain reaction and gene probe methods. Mol. Cell. Probes 4:175-187. 25. Marshall, B. J., and I. Surveyor. 1988. Carbon-14 urea breath test for the diagnosis of Campylobacter pylori associated with gastritis. J. Nucl. Med. 29:11-16. 26. Marshall, B. J., J. R. Warren, E. D. Blincow, M. Phillips, C. S. Goodwin, R. Murray, S. J. Blackbourn, T. E. Waters, and C. R. Sanderson. 1988. Prospective double-blind trial of duodenal ulcer relapse after eradication of Campylobacter pylori. Lancet ii:1437-1442. 27. McNulty, C. A., J. C. Gearty, B. Crump, M. Davis, I. A. Donovan, V. Melikian, D. M. Lister, and R. Wise. 1986. Campylobacter pyloridis and associated gastritis: investigator blind, placebo controlled trials of bismuth salicylate and erythromycin

200

J. CLIN. MICROBIOL.

CLAYTON ET AL.

ethylsuccinate. Br. Med. J. 293:645-649. 28. Morotomi, M., S. Hoshina, P. Green, H. C. Neu, P. Logerfo, I. Watanabe, M. Mutai, and I. B. Weinstein. 1989. Oligonucleotide probe for detection and identification of Campylobacter pylori. J. Clin. Microbiol. 27:2652-2655. 29. Mulrooney, S. B., and R. P. Hausinger. 1990. Sequence of the Klebsiella aerogenes urease genes and evidence for accessory proteins facilitating nickel incorporation. J. Bacteriol. 172:58375843. 30. Nedenskov-Sorensen, P., S. Aase, A. Bjorneklett, 0. Fausa, and G. Bukholm. 1991. Sampling efficiency in the diagnosis of Helicobacter pylori infection and chronic active gastritis. J. Clin. Microbiol. 29:672-675. 31. Newell, D. G., and B. J. Rathbone. 1989. The serodiagnosis of Campylobacter infection. Serodiagn. Immunother. Infect. Dis. 3:1-5. 32. Plikaytis, B. B., R. H. Gelber, and T. M. Shinnick. 1990. Rapid and sensitive detection of Mycobacterium leprae using a nested primer gene amplification assay. J. Clin. Microbiol. 28:19131917. 33. Rauws, E. A. J., W. Langenberg, H. J. Houthoff, H. C. Zanen, and G. N. J. Tytgat. 1988. Campylobacter pyloridis associated chronic active antral gastritis. A prospective study of its prevalence and the effects of antibacterial and antiulcer treatment. Gastroenterology 94:33-40. 34. Rauws, E. A. J., E. A. V. Royen, W. Langenberg, J. V. Woensel, A. A. VrU, and G. N. Tytgat. 1989. C-14-urea breath test in Campylobacter pylori gastritis. Gut 30:798-803. 35. Rogers, B. B., L. C. Alpert, E. A. S. Hine, and G. J. Buffone. 1990. Analysis of DNA in fresh and fixed tissue by the polymerase chain reaction. Am. J. Pathol. 136:541-548. 36. Rosa, P. A., and T. G. Schwan. 1989. A specific and sensitive assay for the Lyme disease spirochete Borrelia burgdorferi using the polymerase chain reaction. J. Infect. Dis. 160:10181029. 37. Saiki, R. K., D. H. Gelfand, S. Stoffel, S. J. Scharf, R. Higuchi, G. T. Horn, K. B. Mullis, and H. A. Erlich. 1988. Primer directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239:487-491. 38. Saiki, R. K., S. Scharf, F. Faloona, K. Mullis, G. Horn, H. A. Erlich, and N. Arnhein. 1985. Enzymatic amplification of p-globulin genomic sequences and restriction site analysis for diagnosis of sickle cell anaemia. Science 230:1350-1354. 39. Schaber, E., F. Umlauft, G. Stoffler, F. Aigner, B. Paulweber,

and F. Sandhofer. 1989. Indirect immunofluorescence test and enzyme-linked immunosorbent assay for detection of Campylobacter pylori. J. Clin. Microbiol. 27:327-330. 40. Schnell, G. A., and T. T. Schubert. 1989. Usefulness of culture, histology and urease testing in the detection of Campylobacter pylori. Am. J. Gastroenterol. 84:133-137. 41. Shibata, D., W. J. Martin, and N. Arnheim. 1988. Analysis of DNA sequences in forty-year old paraffin embedded thin tissue sections. A bridge between molecular biology and clinical histology. Cancer Res. 48:4564-4566. 42. Simor, A. E., N. B. Cooter, and D. E. Low. 1990. Comparison of four strains and a urease test for rapid detection of Helicobacter pylori in gastric biopsies. Eur. J. Clin. Microbiol. Infect. Dis.

9:350-352. 43. Southern, E. M. 1975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98:503-517. 44. Starnbach, M. N., S. Falkow, and L. S. Tompkins. 1990. Species-specific detection of Legionella pneumophila in water by DNA amplification and hybridization. J. Clin. Microbiol. 27:1257-1261. 45. Thillainayagam, A. V., A. S. Arvind, R. S. Cook, I. G. Harrison, S. Tabaqchali, and M. J. G. Farthing. 1991. The diagnostic efficiency of an ultrarapid endoscopy room test for Campylobacter pylori. Gut 32:467-469. 46. Vaira, D., J. Holton, and L. Barbara. 1991. Helicobacter pylori and gastroduodenal disease. Gastroenterol. Int. 4:70-76. 47. Vaira, D., J. Holton, S. Cairns, M. Falzon, and P. Salmon. 1988. Four hour rapid urease test (RUT) for detecting Campylobacter pylori; is it reliable enough to start treatment? J. Clin. Pathol. 41:355-356. 48. Valentine, J. L., R. R. Arthur, H. L. T. Mobley, and J. D. Dick. 1991. Detection of Helicobacter pylori by using the polymerase chain reaction. J. Clin. Microbiol. 29:689-695. 49. West, A. P., M. R. Millar, and D. S. Tompkins. 1990. Survival of Helicobacter pylori in water and saline. J. Clin. Pathol. 43:609. 50. Wetherall, B. L., P. J. McDonald, and A. M. Johnson. 1988. Detection of Campylobacter pylori DNA by hybridisation with non-radioactive probes in comparison with a 32P-labelled probe. J. Med. Microbiol. 26:257-263. 51. Wyatt, J. I., J. Primrose, and M. F. Dixon. 1988. Distribution of Campylobacter pylori in gastric biopsies, abstr. 350A. J. Clin. Pathol. 15:155.