Detection of Helicobacter pylori DNA in fecal samples from infected individuals.

Abstract

Stool, gastric biopsy, and serum samples were collected from 22 subjects. DNA from stool was extracted, amplified, and hybridized with primers specific for the 16S rRNA gene of Helicobacter pylori. DNA from gastric biopsy specimens was analyzed similarly for comparison. Universal primers were used to confirm successful extraction of DNA from samples. Histologic, serologic, and DNA analyses were scored in a blinded fashion. Universal primer amplification verified successful DNA extraction from all stool and gastric tissue specimens. The gastric tissue DNA assay was positive for H. pylori in 11 of the 22 subjects, correlating completely with histologic and serologic results. Stool DNA was positive for H. pylori by our molecular assay in 8 of these 11 H. pylori-positive subjects. All subjects who were negative by histologic, serologic, and gastric tissue DNA analyses were also negative by stool DNA analysis. Compared to histology, serology, and gastric tissue DNA analyses, the sensitivity of our stool DNA assay was 73%, with a specificity of 100%.

Full text

Page 1

JOURNAL OF CLINICAL MICROBIOLOGY,
0095-1137/99/$04.00?0
Copyright © 1999, American Society for Microbiology. All Rights Reserved.
July 1999, p. 2236–2240 Vol. 37, No. 7
Detection of Helicobacter pylori DNA in Fecal Samples
from Infected Individuals
WILLIAM A. GRAMLEY,1ALI ASGHAR,2HENRY F. FRIERSON, JR.,3AND STEVEN M. POWELL1*
Departments of Medicine1and Pathology,3University of Virginia Health Sciences Center, Charlottesville,
Virginia 22908, and Department of Clinical Pharmacology and Toxicology, Indiana
University School of Medicine, Indianapolis, Indiana 226022
Received 7 January 1999/Returned for modification 15 March 1999/Accepted 16 April 1999
Stool, gastric biopsy, and serum samples were collected from 22 subjects. DNA from stool was extracted,
amplified, and hybridized with primers specific for the 16S rRNA gene of Helicobacter pylori. DNA from gastric
biopsy specimens was analyzed similarly for comparison. Universal primers were used to confirm successful
extraction of DNA from samples. Histologic, serologic, and DNA analyses were scored in a blinded fashion.
Universal primer amplification verified successful DNA extraction from all stool and gastric tissue specimens.
The gastric tissue DNA assay was positive for H. pylori in 11 of the 22 subjects, correlating completely with
histologic and serologic results. Stool DNA was positive for H. pylori by our molecular assay in 8 of these 11
H. pylori-positive subjects. All subjects who were negative by histologic, serologic, and gastric tissue DNA
analyses were also negative by stool DNA analysis. Compared to histology, serology, and gastric tissue DNA
analyses, the sensitivity of our stool DNA assay was 73%, with a specificity of 100%.
Helicobacter pylori, first isolated by Warren and Marshall in
1983 (12), has been shown to play an important role in gastritis,
peptic ulcer disease, and gastric malignancies (14, 25). Recent
studies suggest that certain H. pylori genotypes, such as strains
possessing the cagA gene, may be more virulent than others
(2). Several diagnostic tests are available for determining the
presence of H. pylori infection (19). Tests that require endos-
copy include the biopsy urease test, histology, culture, PCR-
based methods, and phase-contrast microscopy of gastric tis-
sue. Diagnostic tests that do not require endoscopy include
[13C] and [14C]urea breath tests, serology, the string test, and
stool antigen enzyme immunoassay (EIA).
Molecular methods such as PCR and Southern blot hybrid-
ization have the capability to sensitively and accurately deter-
mine both the presence of infection and the genotype of bac-
teria. These techniques have been used successfully to detect
H. pylori DNA in gastric tissue by amplifying genes such as the
adhesin gene (7), the urease gene (5), and the 16S rRNA gene
(8). The 16S rRNA gene of H. pylori is a highly specific target
for amplification and has been used previously to help reclas-
sify the organism. Weiss et al. demonstrated the specificity of
unique H. pylori 16S rRNA gene primers to identify the organ-
ism in paraffin-embedded gastric biopsy specimens (24).
Stool analysis would provide a noninvasive means of detect-
ing H. pylori. PCR analysis of stool has been used successfully
to diagnose several infections, including rotovirus (22), micro-
sporidia (6), Vibrio cholerae (1), verotoxin-producing Esche-
richia coli (16), and Salmonella (4) infections. PCR analysis of
stool has even detected mutations of K-ras from tumor cells
shed from colonic neoplasms (18).
Previous reports of PCR analysis of stool for H. pylori have
shown low sensitivity (23). Culturing stool samples allowed
detection of the urease gene by PCR (9), but the sensitivity of
this assay was low and the ability to routinely culture stools for
this purpose was unproven. The difficulty in direct PCR am-
plification of DNA from stool samples is generally thought to
be related to the presence of enzyme inhibitors. We sought to
develop a novel stool DNA extraction process which could
consistently generate amplifiable DNA for detection purposes.
Our results herein provide evidence for the routinely success-
ful detection of H. pylori DNA in stool samples from the
majority of patients infected with this organism.
MATERIALS AND METHODS
Patients undergoing upper endoscopy were recruited consecutively between
August 1996 and December 1996 after giving informed consent according to our
institution’s internal review board approval. Esophagogastroduodenoscopy was
performed on all subjects with endoscopes that had been sterilized by a Steris
(Mentor, Ohio) machine. Autoclaved biopsy forceps were used in obtaining
gastric biopsy specimens from the antrum for rapid urease testing (CLOtest).
Gastric tissue was also obtained from the antrum, incisura, and body of the
stomach for histologic examination and for DNA analysis. Stool specimens were
collected within 2 weeks of the time of endoscopy in sterile containers and kept at
?80°C until analysis. Blood from all patients was collected, and the serum was
stored at ?20°C until the EIA was performed with a Food and Drug Adminis-
tration-approved, commercially available kit (HM-CAP EIA kit; Enteric Prod-
ucts, Stonybrook, N.Y.) which detects immunoglobulin G antibody to H. pylori.
Zinc formalin-fixed paraffin-embedded biopsy specimens were stained with he-
matoxylin and eosin and Giemsa. A single pathologist (H.F.F.) scored all gastric
biopsy specimens without knowledge of the results of the other tests. The num-
ber of H. pylori organisms was semiquantitatively scored as 0 (none), 1 (few;
organisms were present but difficult to find and rare in 400? fields), 2 (moderate;
organisms were readily identified upon microscopic examination and present in
most 400? fields), and 3 (numerous; organisms were present in virtually all 400?
fields).
DNA extraction. One gram of stool from each patient was dissolved in 100%
ethanol and chloroform and then centrifuged at 2,135 ? g and rinsed with
acetone. The sample was then mixed with 8 M urea containing 1% sodium
dodecyl sulfate, 20 mM Tris-HCl (pH 8.0), 100 mg of Chelex (Bio-Rad, Hercules,
Calif.) and 50 mg for of polyvinylpyrrolidone subsequent incubation at 60°C. The
samples were then boiled and centrifuged at 469 ? g. The supernatant was
organically extracted, precipitated with alcohol, and redissolved with 0.7 M
sodium chloride and 1% hexadecyltrimethylammoniumbromide (CTAB) (Sig-
ma) for incubation at 65°C. Organic extraction and alcohol precipitation were
performed for subsequent RNase A (1 mg/ml; Sigma) and proteinase K (0.5
mg/ml; Bio-Rad) incubation at 58°C for 2 h. Another round of organic extraction
and alcohol precipitation was preformed with reconstitution in a solution of 3
mM Tris-HCl (pH 7.5) and 0.2 mM EDTA.
Gastric tissue DNA extraction from paraffin-embedded specimens was per-
formed using xylene and 0.1% sodium dodecyl sulfate–proteinase K (0.5 mg/ml;
Bio-Rad) on two 5-?m thick sections as previously described (13). Cultured
H. pylori DNA extraction was conducted with an H. pylori isolate from a human
* Corresponding author. Mailing address: University of Virginia
Health Sciences Center, Box 10013, Charlottesville, VA 22906-0013.
Phone: (804) 243-2655. Fax: (804) 243-6169. E-mail: smp8n@virginia
.edu.
2236

Page 2

subject who was confirmed to have this infection. H. pylori cultured on horse
blood agar plates was scraped into 1 ml of phosphate-buffered saline. An aliquot
of this suspension was then incubated overnight with proteinase K (0.5 mg/ml;
Bio-Rad) prior to organic extraction and alcohol precipitation. The optical den-
sity was measured in the redissolved pellet for quantitation and subsequent serial
dilutions of H. pylori DNA. Concentrations as low as 1 fg of DNA per ?l were
generated. A single bacterial genome was considered equivalent to 1.6 fg of DNA
(21).
PCR amplification. (i) Universal primers. PCR amplification with nonspecific,
universal primers was performed in 25-?l reaction volumes containing 10 mM
Tris-HCl (pH 8.3), 1.5 mM MgCl2, 50 mM KCl, a 200 ?M concentration of each
deoxynucleoside triphosphate, 10 pmol of each primer, 2.5 U of Taq polymerase
(Perkin-Elmer Cetus, Norwalk, Conn.), and 1.5 ?l of DNA template prepared as
described above. The universal primers for stool analysis consisted of two non-
specific 16S rRNA oligonucleotides, designated USA-1 (AGAATGCCACGGT
GAATACG) and USA-2 (CCTACGGTTACCTTGTTACG). Forty thermo-
cycles were performed, with each cycle consisting of a 30-s denaturation step at
95°C, a 60-s annealing step at 50°C, and a 60-s extension step at 70°C. The
universal primers for gastric tissue DNA analysis consisted of two oligonucleo-
tides designed to amplify exon 7 of the Smad4 gene. The forward primer TGA
AAGTTTTAGCATTAGACAAC and the reverse primer TGTACTCATCTCA
GAAGTGAC were used in a PCR similar to that described above with an
annealing temperature of 50°C. These primers have previously been proven to
successfully amplify this exon for subsequent sequencing (15).
(ii) H. pylori-specific primers. PCR amplification with H. pylori-specific prim-
ers was performed in a manner similar to the universal primer amplifications in
25-?l reaction volumes, with the same buffer, deoxyribonucleoside triphosphate,
and Taq polymerase concentrations. The thermocycle conditions were similar,
with the exception of an annealing temperature of 60°C. The primers consisted
of two specific 16S rRNA oligonucleotides, designated HPF (5? GCG ACC TGC
TGG AAC ATT AC 3?) and HPR (5? CGT TAG CTG CAT TAC TGG AGA
3?), which generated a 138-bp product. In all PCR amplifications, the final cycle
was a 5-min extension step at 70°C to allow full product extension. Each exper-
iment included a negative control template consisting of water and a positive
control consisting of 100 fg of cultured H. pylori DNA.
Southern blot hybridization. Half of the PCR products were electrophoresed
on 2% agarose gels, transferred to nylon membranes (Bio-Rad), and hybridized
with an end-labeled (106cpm/ml) probe in a standard Southern blot fashion. The
probe was a 16-bp oligonucleotide (CGCTGATTGCGCGAAA) designed spe-
cifically for a region within the 16S rRNA gene of H. pylori as previously de-
scribed (24) and was labeled in a [32P]dATP T4 kinase (10 U/ml; New England
Biolab) reaction according to the manufacturer’s instructions. Autoradiographs
generated clear signals after an overnight exposure at room temperature. The
autoradiographs were scored by two independent observers, each of whom was
blinded to the results of the other tests, with a signal in the expected location
recorded as either present or absent.
RESULTS
Detection of Helicobacter pylori DNA. We first assessed the
sensitivity of our molecular assay by determining the threshold
for H. pylori DNA detection. A clinical isolate of H. pylori was
grown in culture, and DNA was extracted and quantitated.
Dilutions of this cultured H. pylori DNA were PCR amplified
with specific H. pylori 16S rRNA primers prior to subsequent
hybridization. Figure 1 shows specific H. pylori signals with
DNA amounts as low as 10 fg, which corresponds to less than
seven genome equivalents of H. pylori (21).
H. pylori status in human subjects. From 22 individuals
presenting for upper endoscopy examination, the following
specimens were collected: gastric biopsy tissue for CLOtest,
histology, and PCR amplification; serum for EIA; and stool for
PCR amplification. Endoscopic findings, medication use, and
FIG. 1. Molecular assay sensitivity. DNA was extracted from cultured H. py-
lori and then diluted to amounts ranging from 5,000 to 1 fg of bacterial DNA and
amplified with primers specific for the 16S rRNA gene of H. pylori. The PCR
products were then hybridized with a32P-labeled H. pylori-specific oligomer. The
lowest amount that produced a signal was 10 fg, which corresponds to less than
seven bacterial genomes.
TABLE 1. Clinical features of subjects in this study
Subject no.
Acid suppression medication
(dosage [mg])a
Endoscopic finding(s)
Result of:
CLOtestSerology Histology (grade)b
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
None
Ranitidine (150 bid)
Omeprazole (20 bid)
None
Famotidine (20 qd)
Nizatidine (150 bid)
Nizatidine (150 qd)
Ranitidine (150 bid)
None
Ranitidine (150 bid)
Omeprazole (20 qd)
None
Nizatidine (150 bid)
Nizatidine (150 bid)
None
Nizatidine (150 bid)
Famotidine (20 qd)
None
None
None
None
Omeprazole (20 bid)
Normal
Gastritis
Normal
Duodenal dilatation
Esophagitis
Duodenitis
Antritis
Duodenitis
Duodenitis
Normal
Antritis, esophagitis
Esophageal nodule
Gastritis
Duodenitis
Schatzki’s ring, antritis
Duodenal ulcer
Erosive antritis, duodenitis
Erosive esophagitisitis
Prepyloric ulcer
Gastric ulcer
Gastritis, duodenitis
Esophagitis, antritis
Positive
Negative
Negative
Negative
Negative
Negative
Positive
Positive
Negative
Positive
Negative
Negative
Negative
Positive
Positive
Negative
Negative
Positive
Positive
Positive
Positive
Negative
Positive
Negative
Negative
Negative
Negative
Negative
Positive
Positive
Negative
Positive
Negative
Negative
Negative
Positive
Positive
Negative
Negative
Positive
Positive
Positive
Positive
Positive
Positive (1)
Negative (0)
Negative (0)
Negative (0)
Negative (0)
Negative (0)
Positive (2)
Positive (1)
Negative (0)
Positive (2)
Negative (0)
Negative (0)
Negative (0)
Positive (2)
Positive (3)
Negative (0)
Negative (0)
Positive (2)
Positive (2)
Positive (2)
Positive (2)
Positive (2)
aMedication taken prior to specimen accrual. Abbreviations: qd, once daily; bid, twice daily.
bSemiquantitation of the number of H. pylori organisms: O ? none; 1 ? few; 2 ? moderate; 3 ? numerous.
VOL. 37, 1999DETECTION OF H. PYLORI IN FECES2237

Page 3

results of conventional tests for H. pylori (i.e., serology, histol-
ogy, and CLOtest) are summarized in Table 1. Half of the
subjects tested positive by serology and histology, whereas the
CLOtest failed to give a positive result for one of these indi-
viduals (subject 22) who was on 40 mg of omeprazole daily.
Gastric biopsy DNA analysis. Amplifiable DNA extracted
from the gastric biopsy specimens was obtained from each of
the 22 study subjects. A 224-bp PCR product from exon 7 of
Smad4 was successfully generated and seen on ethidium bro-
mide-stained agarose gels for each subject (Fig. 2). The same
DNA templates were amplified and hybridized with oligonu-
cleotides specific for H. pylori’s 16S rRNA gene, generating
positive signals in 11 of the 22 patients (Fig. 2). Independent
PCR amplification and hybridization experiments confirmed
these results. The autoradiographic signals were easily scored
with 100% agreement by each of the two observers. These
DNA analysis results correlated perfectly with the histologic
and serologic findings for these same subjects.
The sensitivity of this molecular assay was further illustrated
by gastric tissue testing of 24 additional subjects. For two
subjects, the molecular assay detected a clear but weaker signal
when chronic gastritis was present, histologically consistent
with infection, but no demonstrable organisms were found
even on analysis of additional stained sections. Both of these
subjects were H. pylori positive by serology testing but had
negative CLOtest results. In fact, both of these subjects had
been treated with antiacid secretory therapy prior to examina-
tion, one with omeprazole (20 mg daily) and the other with
ranitidine (300 mg daily). No prior antibiotic use was noted.
Complete agreement on histology, serology, and DNA analysis
for the other 22 gastric biopsy subjects was found (data not
shown).
Stool DNA analysis. Several methods of stool DNA extrac-
tion from subjects with and without H. pylori infection as shown
by conventional testing were tested. The amount of DNA re-
covered varied depending on the protocol used. A novel meth-
od of extraction using lipid solubilizers, ionic and nonionic
detergents, chelators, and organic solvents, was developed and
routinely produced amplifiable DNA (see Materials and Meth-
ods). One gram of stool consistently generated approximately
10 ?g of DNA. Identifiable signals were repeatedly generated
when 200 ng of DNA was used as a PCR template in our assay.
At least three independent experiments of stool DNA extrac-
tion, PCR amplification, and hybridization for 10 subjects (5
H. pylori positive and 5 H. pylori negative) consistently identi-
fied the presence or absence of H. pylori DNA, confirming the
reproducibility and accuracy of our molecular assay.
We then sought to determine the sensitivity and specificity of
this assay in identifying H. pylori infection. Frozen samples of
stool collected from 22 study subjects were subjected to DNA
extraction according to our newly developed protocol and
tested for H. pylori DNA to compare the results of conven-
tional testing and gastric tissue DNA analysis. Stool sample
extracts were first amplified with universal primers for the 16S
rRNA gene to demonstrate the presence of amplifiable DNA.
A clearly visible PCR product of expected size (148 bp) was
generated in all samples, as seen by ethidium bromide staining
of agarose gels (Fig. 3). Subsequent amplification and hybrid-
ization of the same stool DNA templates with primers specific
for the H. pylori 16S rRNA gene resulted in unambiguous
FIG. 2. Gastric tissue DNA analysis. The ethidium bromide-stained agarose gel in the top panel demonstrates amplification of exon 7 of Smad4 in gastric tissue
DNAs of all 22 subjects, confirming the presence of amplifiable DNA. Lane M displays DNA size markers (Bethesda Research Laboratories 1-kb ladder). Specific
H. pylori amplification produced an expected 139-bp PCR product with samples from infected subjects, seen as hybridization signals in the middle panel, with a
32P-labeled H. pylori-specific oligomer used as a probe. The results of histologic analysis (?, positive; ?, negative) are presented in the bottom panel for comparison.
The positive control lane (C) used DNA extracted from cultured H. pylori as a template.
FIG. 3. Stool DNA analysis. The ethidium bromide-stained agarose gel in the top panel illustrates amplification from the stool DNA of all 22 subjects of an expected
148-bp PCR product with universal 16S rRNA gene primers, confirming the presence of amplifiable DNA. Lane M displays DNA size markers (Bethesda Research
Laboratories 1-kb ladder). Specific H. pylori amplification primers generated an expected 139-bp PCR product with samples from infected subjects, seen as hybridization
signals in the middle panel, with a32P-labeled H. pylori-specific oligomer used as a probe. The results of histologic analysis (?, positive; ?, negative) are presented in
the bottom panel for comparison. The positive control lane (C) used DNA extracted from cultured H. pylori as a template.
2238GRAMLEY ET AL.J. CLIN. MICROBIOL.

Page 4

positive signals in 8 subjects and negative signals in the other
14 subjects (Fig. 3). Independent PCR amplification and hy-
bridization of all DNA samples confirmed these results. Com-
pared with histology, serology, and gastric DNA analysis, the
molecular stool DNA assay had a sensitivity of 73% and a
specificity of 100%.
DISCUSSION
Currently available tests for the diagnosis of H. pylori infec-
tion have relatively high sensitivities and specificities, but each
has its limitations in clinical application. Urease-based biopsy
tests require endoscopy and are not reliable in patients taking
proton pump inhibitors. Histologic examination also follows
endoscopy and is subject to sampling error, and its accuracy is
dependent on the stain selected and on the pathologists’ skill.
Serology is inexpensive but is not reliable in determining the
presence of active infection.
Our successful amplification and specific detection of H. py-
lori DNA directly from stool samples in the majority of in-
fected subjects indicates that this approach is feasible and
demonstrates it has true potential in aiding the diagnosis and
management of patients with H. pylori infection. The sensitivity
of H. pylori DNA detection of 73% from uncultured stool
specimens exceeds that in previous attempts (9, 23). This
higher sensitivity most likely is the result of our extraction
method, which removes from the stool samples impurities that
are known to contain PCR inhibitors (18).
The sensitivity of PCR amplification is well known and is
exhibited by our ability to easily detect DNA from less than
seven H. pylori bacterial organisms. Other investigators have
used PCR of gastric tissue specimens to identify H. pylori DNA
(17). We demonstrated the extreme sensitivity and specificity
of our molecular assay in analyzing gastric tissue biopsy sam-
ples in our study subjects as well as in an additional set of
subjects. In two of these additional subjects in which infection
was felt to be present serologically and histologically by the
pattern of gastritis, H. pylori DNA was detectable, but organ-
isms could not be identified microscopically. Notably, these
two patients had had prior antiacid secretory therapy, but no
prior antibiotic use was noted.
Our findings indicate that this molecular assay is specific for
H. pylori DNA in human stool and gastric specimens. The
H. pylori 16S rRNA primers used were designed from those
tested by Weiss et al. (24), who demonstrated the specificity of
their primers in detecting H. pylori DNA in paraffin-embedded
gastric tissue. Furthermore, they found that these primers did
not cross-react with many other common microorganisms, such
as E. coli, various Campylobacter species, and other Helicobac-
ter species, including H. cinnaedi, H. fenelliae, and H. muselae.
Recently, Scholte et al. (17) have tested similar primers de-
signed to amplify this unique region of 16S rRNA on 38 dif-
ferent bacteria, including 10 Helicobacter species, and found
100% sensitivity and specificity.
Our DNA analyses of human stool and gastric biopsy sam-
ples using these specific oligonucleotides correlated well with
conventional tests for H. pylori infection. We chose histology as
the “gold standard” in this study due to the expertise of our
pathologist (H.F.F.). These results also suggest that sampling
error is not a major obstacle in the detection of H. pylori DNA
in stool samples. Small portions of spontaneously passed stool
yielded consistent results on repeated analyses and identified
the majority of infections. The reason for the failure to identify
H. pylori DNA in three stool samples from individuals known
to have H. pylori organisms and DNA in their gastric tissue is
uncertain. Sampling error certainly could have played a role in
these three cases, as could have degradation of DNA and
organisms during intestinal transit. Additional studies to fur-
ther analyze these samples are planned.
The molecular detection of H. pylori DNA has the added
benefits of being able to genotype the infecting strains and to
provide useful information about the presence or absence of
potential virulence factors for a particular infection. For in-
stance, once virulent components of H. pylori are character-
ized, PCR primers can be designed to specifically identify these
important variants. Potentially important virulence factors in-
clude vacA gene variants and the cagA gene. The presence of
certain genotypes of vacA and cagA in infecting strains has
been shown to be associated with a more dense inflammatory
gastritis and peptic ulcer disease (2).
The finding of detectable H. pylori DNA in the stool of the
majority of infected subjects has important implications for
transmission of this microorganism. These data support a fe-
cal-oral route of transmission, although viable bacteria were
not sought in this study. While it is known that H. pylori can be
found in and cultured from the stool of infected individuals (9,
11, 20), the present study provides further supporting evidence
that this may be a frequent occurrence. Makristathis et al.
identified H. pylori’s species-specific protein antigen in stool
samples by a seminested PCR method in 93.7% of patients
with duodenal ulcers (10). A study of Bangladeshi children
found 60% had H. pylori in their stool upon testing with im-
munomagnetic separation and ureA gene PCR amplification
(3).
Our noninvasive assay to detect H. pylori DNA in spontane-
ously passed stool is currently labor-intensive, mostly due to
the extensive DNA extraction procedure. Further develop-
ment of the assay will focus on simplifying the DNA extraction
process. Significant advances in technology to neutralize or
remove impurities from DNA efficiently are eagerly awaited to
make this assay more clinically applicable. Due to the cumber-
some technical procedures required to handle radioactivity,
the incorporation of chemiluminescence detection methods
will be explored. Further testing will be required to determine
if the method can be modified for stool card tests analogous to
guaiac testing. Moreover, additional testing of stools after
treatment of H. pylori infection is planned to determine if the
assay is helpful in determining eradication of the microorgan-
ism.
ACKNOWLEDGMENTS
We are grateful to Richard Zoltec for facilitating the H. pylori
serology testing and Mark Clem for technical assistance in processing
histologic specimens.
This work was supported in part by NIH grant CA67900-04.
REFERENCES
1. Albert, M. J., D. Islam, S. Nahar, F. Qadri, S. Falklind, and A. Wintraub.
1997. Rapid detection of Vibrio cholerae O139 Bengal from stool specimens
by PCR. J. Clin. Microbiol. 35:1633–1635.
2. Atherton, J. C., R. M. Peek, Jr., K. T. Tham, T. L. Cover, and M. J. Blaster.
1997. Clinical and pathological importance of heterogeneity in VacA, the
vacuolating cytotoxin gene of helicobacter pylori. Gastroenterology 112:92–
99.
3. Casswal, T. H., H. Nilsson, M. Bergstrom, P. Aleljung, T. Wadstrom, A. K.
Dahlstrom, M. J. Albert, and S. A. Sarker. 1999. Evaluation of serology,
13C-urea breath test, and polymerase chain reaction of stool samples to
detect Helicobacter pylori in Bangladeshi children. J. Pedicatr. Gastroenterol.
Nutr. 28:31–36.
4. Chiu, C., and J. T. Ou. 1996. Rapid identification of Salmonella serovars in
feces by specific detection in virulence genes, invA and spvC, by an enrich-
ment broth culture-multiplex PCR combination assay. J. Clin. Microbiol.
34:2619–2622.
5. Clayton, C. L., H. Kleanthous, P. J. Coates, D. D. Morgan, and S.
Tabaqchali. 1992. Sensitive detection of Helicobacter pylori by using poly-
VOL. 37, 1999 DETECTION OF H. PYLORI IN FECES2239

Page 5

merase chain reaction. J. Clin. Microbiol. 30:192–200.
6. Da Silva, A. J., F. J. Bornay-Linares, C. A. De la Puente, H. Moura, J. M.
Peralta, I. Sobottka, D. A. Schwartz, G. S. Visvesvara, S. B. Slemenda, and
N. J. Pieniazek. 1997. Diagnosis of Enterocytozoon bieneusi (Microsporidia)
invections by polymerase chain reaction in stool samples using primers based
on the region coding for small-subunit ribosomal RNA. Arch. Pathol. Lab.
Med. 121:874–879.
7. Evans, D. E., D. J. Evans, H. C. Lampert, and D. Y. Graham. 1995. Restric-
tion fragment length polymorphism in the adhesin gene hpaA of Helicobacter
pylori. Am. J. Gastroenterol. 90:1282–1288.
8. Ho, S., J. Hoyle, F. Lewis, A. Secker, D. Cross, N. Mapstone, M. Dixon, J.
Wyatt, D. Tompkins, G. Taylor, and P. Quirke. 1991. Direct polymerase
chain reaction test for detection of Helicobacter pylori in humans and ani-
mals. J. Clin. Microbiol. 29:2543–2549.
9. Kelly, S., M. Pitcher, S. Farmery, and G. Gibson. 1994. Isolation of Helico-
bacter pylori from feces of patients with dyspepsia in the United Kingdom.
Gastroenterology 107:1671–1674.
10. Makristathis, A., E. Pasching, K. Schutze, M. Wimmer, M. L. Rotter, and
A. M. Hirschl. 1998. Detection of Helicobacter pylori in stool specimens by
PCR and antigen enzyme immunoassay. J. Clin. Microbiol. 36:2772–2774.
11. Mapstone, N., F. Lewis, D. Tompkins, D. Lynch, A. Axon, M. Dixon, and P.
Quirke. 1993. PCR identification of Helicobacter pylori in faeces from gas-
tritis patients. Lancet 341:447.
12. Marshall, B., and J. Warren. 1983. Unidentified curved bacillus on gastric
epithelium in active chronic gastritis. Lancet i:1233–1235.
13. Moskaluk, C. A., and S. E. Kern. 1997. Microdissection and polymerase
chain reaction amplification of genomic DNA from histological tissue sec-
tions. Am. J. Pathol. 150:1547–1549.
14. Parsonnet, J., G. Friedman, D. Vandersteen, Y. Chang, H. Vogelman, N.
Orientreich, and R. Sibley. 1991. Helicobacter pylori infection and the risk
of gastric carcinoma. N. Engl. J. Med. 325:1127–1131.
15. Powell, S. M., J. Harper, S. Hamilton, C. Robinson, and O. W. Cummings.
1997. Inactivation of Smad4 in gastric carcinomas. Cancer Res. 57:4221–
4224.
16. Ramotar, K., B. Waldhart, D. Church, R. Szumski, and T. J. Louie. 1995.
Direct detection of verotoxin-producing Escherichia coli in stool samples by
PCR. J. Clin. Microbiol. 33:519–524.
17. Scholte, G. H., L. J. van Doorn, W. G. Quint, and J. Lindeman. 1997.
Polymerase chain reaction for the detection of Helicobacter pylori in form-
aldehyde-sublimate fixed, paraffin-embedded gastric biopsies. Diagn. Mol.
Pathol. 6:238–243.
18. Sidransky, D., T. Tokino, S. Hamilton, K. Kinzler, B. Levin, P. Frost, and B.
Vogelstein. 1992. Identification of ras oncogene mutations in the stool of
patients with curable colorectal tumors. Science 256:102–105.
19. Thijs, J. C., A. A. van Zwet, W. J. Thijs, H. B. Oey, A. Karrenbeld, F.
Stellaard, D. S. Luijt, B. C. Meyer, and J. H. Kleibeuker. 1996. Diagnostic
tests for Helicobacter pylori: a prospective evaluation of their accuracy, with-
out selecting a single test as the gold standard. Am. J. Gastroenterol. 91:
2125–2129.
20. Thomas, J., G. Gibson, M. Darboe, A. Dale, and L. Weaver. 1992. Isolation
of Helicobacter pylori from human feces. Lancet 340:1194–1195.
21. Tomb, J. F., O. White, A. R. Kerlavage, R. A. Clayton, G. G. Sutton, R. D.
Fleischmann, K. A. Ketchum, H. P. Klenk, S. Gill, B. A. Dougherty, K.
Nelson, J. Quackenbush, L. Zhou, E. F. Kirkness, S. Peterson, B. Loftus, D.
Richardson, R. Dodson, H. G. Khalak, A. Glodek, K. McKenney, L. M.
Fitzgerald, N. Lee, M. D. Adams, and J. C. Venter. 1997. The complete
genome sequence of the gastric pathogen Helicobacter pylori. Nature 388:
539–547.
22. Ushijima, H., H. Koike, A. Mukoyama, A. Hasegawa, S. Nishimura, and
J. Gentsch. 1992. Detection and serotyping of totaviruses in stool specimens
by using reverse transcription and polymerase chain reaction amplification.
J. Med. Virol. 38:292–297.
23. van Zwet, A. A., J. C. Thijs, A. M. D. Kooistra-Smid, J. Schirm, and J. A. M.
Snijder. 1994. Use of PCR with feces for detection of Helicobacter pylori
infections in patients. J. Clin. Microbiol. 32:1346–1348.
24. Weiss, J., J. Mecca, E. Silva, and D. Gassner. 1994. Comparison of PCR and
other diagnostic techniques for detection of Helicobacter pylori infection in
dyspeptic patients. J. Clin. Microbiol. 32:1663–1668.
25. Wotherspoon, A. C., C. Ortiz-Hidalgo, M. F. Falzon, and P. G. Isaacson.
1993. Helicobacter pylori-associated gastritis and primary B-cell lymphoma.
Lancet 338:1175–1176.
2240 GRAMLEY ET AL. J. CLIN. MICROBIOL.