- Email: [email protected]
Nested primers improve sensitivity in the detection of Helicobacter pylori by the polymerase chain reaction
Journal of Infection (1998) 36, 105-110
Nested Primers Improve Sensitivity in the Detection of Helicobacter pylori by the Polymerase Chain Reaction K. B. Bamford *~, D. A. Lutton ~, B. O'Loughlin ~, W. A. Coulter 2 and J. S. A. Collins 3 ~Department of Microbiology and Immunobiology, 2The School of Dentistry, The Queen's University of Belfast, Grosvenor Road, Belfast BT12 6BN and 3The Royal Victoria Hospital, Belfast, U.K. To investigate potential routes of spread of infection by the polymerase chain reaction (PCR) it is important that the technique is effective in the types of specimen to be investigated. To establish the limits of detection of Helicobacter pylori by PCR in clinical material from the gastric mucosa, faeces, dental plaque and oral rinses, samples were seeded with known numbers of bacteria. DNA extraction was followed by amplification with primers from the urease C gene. Nested primers were used to amplify the PCR product which was detected using a digoxigenin-labeiled probe. Faeces or plaque inhibited the single reaction 102-106 fold. A second amplification using nested primers and probing increased the sensitivity to a level similar to that obtained with pure culture. This method is potentially useful with less likelihood of false negative results when trying to detect H. pylori by PCR in highly contaminated, clinical material.
Introduction Helicobacter pylori is the commonest cause of inflammation of the gastric mucosa and is fundamental to the pathogenesis of peptic ulceration. I-3 Infection with H. pylori may also be a risk factor for the subsequent development of gastric carcinoma and this bacterium has been linked with MALT lymphoma. *' ~ This provides a strong argument for the eradication of H. pylori and the use of antimicrobial agents in the treatment of duodenal ulceration. This is particularly so because elimination of the infection greatly reduces the risk of ulcer relapse, thus reducing the need for anti-ulcer maintenance therapy. 6 7 However, in a significant proportion of cases the organism is not completely eliminated at the first attempff and the occurrence of relapse/reinfection is well documented. 9 Helicobacter pylori may not be detected by histology, culture, rapid urease test or by urea breath tests immediately following a course of treatment; however, patients may be found to be H. pylori-positive when assessed 6 - 1 2 months later. ~° Early follow-up studies after treatment can be therefore be misleading. It is unclear whether this is due to recurrence of bacteria which are partially inhibited and may have persisted in a coccoid form, M or whether reinfection from another source has occurred. Reinfection of the gastric mucosa
*Address correspondence to: Dr. K. B. Bamford. Accepted for publication 7 May 1997. 0163-4453/98/010105 +06 $12.00/0
may occur via bacteria located within the oral cavity, ~2 or via an external source, e.g. another person. The mode of transmission is unclear, although there is evidence for person-to-person transmission, l~ The same strain of H. pylori has been found in more than one individual within several family settings, suggesting that intrafamilial crossinfection may occur. 1. Part of the problem in determining the source of infection is that it is difficult to culture H. pylori from heavily contaminated clinical specimens, and few descriptions exist of isolation of this bacterium from such samples, is This is partly due to the fastidious nature of H. pylori, but may also be influenced by the occurrence of metabolically quiescent non-culturable forms. ~ Thus, accurate methods of determining both the response to therapy, without long-term follow-up, and for assessing possible routes of transmission of H. pylori are required. The polymerase chain reaction (PER) can be used to specifically detect minute quantities of impure DNA by chemical rather than biological means. ~6 In order to apply this technique to a routine diagnostic laboratory setting, a simplified standard procedure is necessary where the limitations of the method are known. The specificity of the PeR is dependent on the choice of oligonucleotide primers, and, in theory, it is sensitive enough to detect one copy of target DNA. In practice, however, interfering material within many samples may inhibit the PCR, thus reducing the sensitivity of the method and the significance of a negative result/7 To this end, we have developed a PeR assay to detect H. © 1998 The British ;~nIection Society
K. B. Bamford et a t
106
pylori within a range of clinical samples. Oligonucleotide primers, specific for tt. pylori, were selected based on a conserved region of the urease C gene 18 and nested primers were used to increase the sensitivity of the technique.
Materials and Methods To determine the sensitivity of the assay and to investigate the possible effects of inhibitory impurities from clinical specimens, H. pylori-negative clinical samples of antral biopsy tissue, faeces, and dental plaque were artiflcally seeded with serial logarithmic dilutions of I-I. pylori prior to PCR analysis.
Clinical samples Antral biopsy specimens were obtained during routine examinations of the gastric mucosa at the time of oesophagogastrodnodenoscopy from consenting patients. Samples from patients who were CLO test (Delta West, Australia) and H. pylori culture negative were selected for sensitivity determination. Extra care was taken to decontaminate thoroughly both the endoscope and biopsy forceps between patients. Specimens for PCR analysis were snap frozen in liquid nitrogen directly following excision, to minimize the risk of contamination. For routine culture, biopsy specimens were transported to the laboratory in 0.01 M phosphate buffered saline (PBS) pH 7.4 and inoculated directly onto Colombia base agar (Oxoid) containing 7% defibrinated horse blood, plus a second agar with antibiotic supplement (vancomycin 10 mg/ml, trimethoprim lactate 5 mg/ml, amphotericin B 5 mg/ml and nalidixic acid 20 mg/ml). Plates were cultured at 37°C for up to 5 days in a microaerophillic atmosphere of 5% O2, 1% H2, 7% CO2 and 87% N2 (Don Whitley variable atmosphere incubator). Isolates were identified as/4. pylori if they were Gram-negative with a typical morphology, hydrolysed urea rapidly and were catalase and oxidase positive. 19 Subgingival plaque samples were obtained by curetting the depth of the gingival sulcus or pocket and were immediately placed in transport buffer (PBS). Samples were stored at - 2 0 °C prior to analysis. Oral rinse fluid was obtained by rinsing 15 ml PBS in the mouth for 10 s before spitting into a sterile universal container. Faeces samples were obtained from the diagnostic microbiology laboratories of the Royal Hospitals Trust, where they had been submitted for the investigation of gastrointestinal symptoms.
Titration of bacteria for sample inoculation Columbia base agar (Oxoid) containing 7% defibrinated horse blood was used to culture the bacteria at 37 °C for up to 5 days in a microaerophillic atmosphere as above. Bacteria were harvested, washed and suspended in sterile distilled water. Counts were determined using the method of Miles and Misra 2° and verified by dark field microscopy. Serial 10-fold dilutions of the bacterial culture were prepared in distilled water. Aliquots (300 pl) of bacterial culture, containing a known concentration of bacteria, were added to plaque, faeces, oral rinse fluid or biopsy specimens.
DNA extraction Plaque and faeces samples were suspended in sterile distilled water to an optical density of 0.2 and 0.3, respectively. Oral rinse fluids were used at the concentration provided from the clinic. Plaque, faeces and oral rinse fluids were sonicated (3 x 5 s bursts at 5 ~tm) to dissociate aggregates. Suspensions of plaque, faeces or oral rinse fluid (200 bd) were transferred to microcentrifuge tubes (Perkin-Elmer Cetus) containing 300 gl aliquots of a known concentration of bacterial culture. Bacterial control samples (pure culture) were processed in the absence of clinical material. The samples were centrifuged (12 000 x g, 3 min) in a microcentrifuge (Eppendorf 5414), the supernatant decanted and 300 btl of extraction buffer (Tris (20 mM) pH 8 and 0.5% v/v Tween 20) added. Biopsy specimens were removed from liquid nitrogen and known concentrations of bacterial culture added. The specimen was then suspended in extraction buffer (300 btl). All specimens were vortexed and proteinase K (Sigma) was added to a final concentration of 0.Smg/ml. Following incubation ( l h , 37°C, the proteinase K was inactivated by heating the sample to 98 °C for 10 rain. 21 Twenty microlitres of DNA template were added per 100 btl of PCR reaction mix. Negative control samples (sterile, double distilled deionised water) were processed in parallel for each extraction.
PCR primers Two regions from each end of a highly conserved 294 base pair region of the urease C (ureC) gene of H. pylori were used for primer 1 (5'AAGCTTTTAGGGGTGTTAGGGGTTT-3') and primer 2 (5"-AAGCTTACTTTCTAACACTAACGC-3'), as previously described by Labigne et al. 18 A second set of primers; primer 3 (5'-CTTTCTTCTCAAGCAATTGTC-3') and primer 4 (5"-CAAGCCATCGCCGGTTTTAGC-3"), were
Helicobacter pylori: nested PCR sensitivity used to amplify a 252 base pair region located 21 base pairs internal to primers 1 and 2. Primer pairs 1 and 2, and 3 and 4, were purchased from The Department of Biochemistry, Queen's University of Belfast and British Bio-technology Products Ltd., U.K., respectively. A 44-mer oligonucleotide probe (5'A G A A T T G A A GC ATTGCGCG ATTGGGGATAA GTTTGTGAGCGAAT-3'), described by Labigne, is which recognized a region central to both primer sets was used to confirm the identity of the amplified product.
!
107
2 3 4 5 6 7 8 9 1011121314
15181718192021
22232425262728
29
Figure l. Results of PCR using primers 1 and 2. Pure culture (lanes 1-7), faeces I (lanes 8-14), faeces 2 (lanes 15-21, faeces 3 (lanes 22-28), molecular weight markers (lane 29) X (352 bases) and Y (212 bases). Ten-fold reductions in number of cfu in sample, 10 ~ (lanes 1, 8,15,22), 10 s (lanes 2,9,16,23), 104 (lanes 3,10,17,24), 103 (lanes 4,11,18,25), 10 z (lanes 5,12,19,26), 10 (lanes 6,13,20,27), 10 -t (lanes 7, 14.21,28).
DNA Amplification The PCR reaction mixture contained KC1 (lOmM), (NH4)2SO4 (10mm), Tris-HCl pH 8.8 (20mM), MgSO4 (2 raM), Triton X-100 (0.1% v/v), deoxynucleotide triphosphates (200 laM of each dATER dCTER dGTER and dTTP), the relevant primers (1 t~g/lO0 ~tl reaction volume), Vent DNA polymerase (New England Biolabs) at 1 unit/100 btl reaction volume and I)NA template (typically 20 lal per 100 t~1 of PCR reaction mixture). The DNA polymerase was added immediately prior to PCR incubation to reduce 3' to 5' exonuclease activity. Negative control samples, in which the DNA template was replaced by sterile distilled water, were included in each PCR assay to eliminate the possibility of false-positive reactions. In addition, samples containing positive control DNA were included to check that samples were amplifiable, and to exclude the possibility that a negative result was due to a fault in the reaction process. The reaction mixture was overlaid with mineral oil, and placed in a Perkin Elmer 480 Thermal Cycler. Denaturation, annealing, and extension temperatures and times were 95 °C for 2 min, 55 °C for 2 m i n and 75 °C for 1 min, respectively. A maximum of 35 cycles was run. The primary PeR reaction product was subsequently incubated with primers 3 and 4:(2 btl PCR product/100 bd PCR reaction mixture) for a further 35 cycles.
1 2 3
4 5 6 7 8 9 1011121314151617181920212223242526272829
Figure 2. Results of PCR using primers 1 and 2. Pure culture (lanes 1-7), plaque 1 (lanes 8-t4), plaque 2 (lanes 15-21, plaque 3 (lanes 22-28). molecular weight markers (lane 29). Ten-fold reductions in number of cfu in sample, as in Fig. 1.
fixed by baking (120 °C, 30 min). The membrane was prehybridized (68°C, l h) in a hybridization buffer (5 × SSC (20 x SSC is NaCI (3 M), Na citrate (0.3 g), pH 7.0), Boehringer Mannheim Biochemica blocking reagent (1%), N-lauroylsarkosine (0.1% w/v), and sodium dodecyl sulfate (0.02% 5v/v)). Hybridization of the labelled oligonucleotide probe (50 pM/ml in hybridization buffer) to target DNA was carried out at 37 °C for 5 h. Unhybridized label was removed by washing in tetramethylammonium chloride (2 x 70 °C, 20 min). DIG-labelled DNA was detected using an enzyme-linked immunoassay incorporating an anti-digoxigenin linked alkaline phosphatase conjugate (Boehringer Mannheim Biochemica).
Detection of amplified DNA
Results
PCR-amplified DNA was resolved by electrophoresis on 2% agarose gels (Sigma), stained with ethidium bromide (1 gg/ml) and visualized using a UV transilluminator. The specificity of the PeR product was verified by Southern blotting using the 4Zt-mer oligonucleotide probe described above. Briefly, the probe was non-isotopically 3' end labelled using terminal transferase, by incorporation of a single digoxigenin (DfG)-labelled dideoxyuridine-triphosphate (Boehringer Mannheim Biochemica). Amplified DNA fragments were transferred to a Hybond-N nylon membrane (Amersham) by Southern blotting and
The limits of detection achieved using the single set of primers, primers 1 and 2, was of the order of 10 colony forming units (cfu) of H. pylori when a suspension of pure culture was used as the original sample (Figs 1 and 2, lanes 1-7). The sensitivity of detection was reduced when the first set of primers only was used in PCR to detect H. pylori in the prepared samples of faeces (n = 3) (Fig. 1, lanes 8-14, lanes 15-21, and lanes 22-28). A reduction in the limit of detection of H. pylori was also seen when different samples of dental plaque (n = 3) were examined (Fig. 2 lanes 8-14, lanes 15-21, and lanes
108
1 2
K.B. Bamford et aL
34
5 6 7 8 9 101112131415161718192021222324252627282930
Figure 3. Results of PCR using nested primers (1 and 2 for reaction 1, followed by 3 and 4 for reaction 2). Gastric mucosaI biopsy (lanes 1-7), oral rinse (lanes 8-14), plaque 1 (lanes 15-21, plaque 2 lanes 22-28), negative control (lane 29), molecular weight markers (lane 30). Tenfold reductions in number of cfu in sample, as in Fig, 1.
1
2
3
4
5
6
7
8
9
10 11
12 13
14 15 16
Figure 5. Results of PCR using nested primers (1 and 2 for reaction l, followed by 3 and 4 for reaction 2) (top) and hybridization with the 44 base DNA probe with immunodetection (bottom). Faeces 4 (lanes 1-7), faeces 5 (lanes 8-14), negative control (lane 15), molecular weight markers (lane 16). Ten4old reductions in number of cfu in sample, as in Fig. 4.
~X
~Y
1
2
3
4
5
6
7
8
9
10 11
12 13
14 1516
Figure 4. Results of PCR using nested primers (1 and 2 for reaction 1, followed by 3 and 4 for reaction 2) (top) and hybridization with the 44 base DNA probe with immunodetection (bottom). Faeces 1 (lanes 1-7), faeces 2 (lanes 8-14), negative control (lane 15), molecular weight markers (lane 16). Ten-fold reductions in number of cfu in sample, 106 (lanes 1, 8,), 10 s (lanes 2,9), 104 (lanes 3,10). 103 (lanes 4.11), 102 (lanes 5,12), 10 (lanes 6,13), 10 -1 (lanes 7,14).
2 2 - 2 8 ) . The detection limit r a n g e d from 104-102 in different faecal a n d p l a q u e specimens b u t was in all cases less t h a n t h a t for the p u r e culture. (Figs 1 a n d 2). I n c r e a s e d sensitivity w a s achieved by e m p l o y i n g a nested PCR, w h e r e the second r e a c t i o n used primers 3 a n d 4, w i t h the r e a c t i o n p r o d u c t of the first r e a c t i o n as the target template. In d e n t a l p l a q u e p r e p a r a t i o n s less t h a n 1 0 c f n were consistently detected (Fig. 3, lanes 1 5 - 2 1 a n d lanes 2 2 - 2 8 ; Fig. 4, lanes 8 - 1 4 ) . A similar level of detection w a s o b t a i n e d from oral rinse fluid (Fig. 3, lanes 8 - 1 4 ) , a n d biopsy specimens (Fig. 3, lanes 1 - 7 ) . S o u t h e r n blotting a n d DNA-DNA h y b r i d i z a t i o n w i t h the oligonucleotide probe confirmed the specificity of the amplified p r o d u c t (Figs 4 - 6 ) . Less t h a n 10 cfu could be detected from m o s t faeces samples by using the nested
1
2
3
4
5
6
7
8
9
10 11 12 13 14 1516
Figure 6. Results of PCR using nested primers (1 and 2 for reaction 1, folIowed by 3 and 4 for reaction 2) (top) and hybridization with the 44 base DNA probe with immunodetection (bottom). Gastric biopsy (lanes 1-7), Plaque (lanes 8-14), negative controI (lane 15), molecular weight markers (lane 16). Ten-fold reductions in number of cfu in sample, as in Fig. 4.
p r i m e r r e a c t i o n (Figs 5 a n d 6). The h i g h e s t limit of detection achieved with a faecal s a m p l e was 10 2 cfu (Fig. 5), even w i t h the lO-fold increase in sensitivity achieved as a result of hybridization w i t h the probe. Vent DNA p o l y m e r a s e (New E n g l a n d Biolabs) was used for t h e PCR, as the sensitivity w i t h this e n z y m e was consistently 10fold g r e a t e r t h a n t h a t achieved w i t h AmpliTaq DNA p o l y m e r a s e (Perkin Elmer-Cetus).
Discussion We h a v e described a rapid, sensitive a n d specific PCR a s s a y for 17. pylori w h i c h is suitable for use w i t h i n a clinical laboratory, This m e t h o d does n o t require extensive e x t r a c t i o n a n d purification of t h e DNA prior to amplification. We h a v e assessed the sensitivity of this t e c h n i q u e w h e n applied to different clinical s a m p l e types; t h a t is, p u r e culture, gastric m u c o s a l biopsies, d e n t a l plaque, oral rinse fluid, a n d faeces. This a s s a y relies
Helicobacter pylori: nested PCR sensitivity solely on the presence of target DNA rather than viable, metabolizing cells. ~6 Thus, it is suitable for detection of 17. pylori within biopsy specimens where viability m a y have been lost due to recent antibiotic therapy or delays in transportation to the laboratory. It is also appropriate for samples such as faeces and dental plaque, from which it is difficult to isolate H. pylori because of the presence of very large numbers of other, less fastidious, bacteria. In our experiments we have been able to detect tt. pylori in highly contaminated samples which contain large numbers of both prokaryotic and eukaryotic cells as well as m a n y organic impurities. The urease gene region of H. pylori is a suitable target site for quantitative PCR because, despite a very high level of expression, the DNA is present as a single chromosomal copy per bacterial cell. = The urease region contains nine genes; four of which (ureA, ureB, ureC, and ureD) confer a urease positive phenotype upon Campylobacter jejuni, 22 and five accessory genes (ureE, ureE ureG, uteri, and ureI), three of which (ureF, ureG and uteri) are required for urease expression in Escherichia coll. 23 UreA and ureB code for the structural A and B subunits of the urease enzyme, 23 but the functions of the other genes associated with urease expression are, as yet, less clear. Urease is a highly conserved enzyme, and there is considerable homology between amino acid sequences of naturally occurring ureases from both plant and bacterial sources including that of 11. pglori. 22 Detection of H. pglori by PCR using primer sites on the ureA and ureB genes has previously been described. 24-2s Selected primer sites from these genes have also been used to amplify large variable regions of DNA, the products of which have subsequently been analysed by restriction fragment length polymorphism and used as a typing method. 2<27 In the present study, the ureC gene which is unique to H. p!lIori was selected as a suitable target site for PCR. Bickley et al. demonstrated the specificity of this gene by examining a wide range of bacteria, 28 and the PCR technique has been applied to gastric biopsies. 29 Our PCR using primers 1 and 2, and the intragenic probe, was highly sensitive when applied to pure culture and gastric biopsies. Helicobacter pylori was detected when as few as 10 organisms were present (Figs 1 and 2). However, when applied to faeces and plaque, the PCR was inhibited, with the sensitivity of the technique in some cases decreased to 10 4 CFU. The degree of inhibition of the PCR varied between individual specimens but was in all cases lower than that with pure culture (Figs 1 and 2). By using nested primers and a small quantity of the original reaction mixture as the template, the sensitivity of the reaction was considerably increased (Fig. 3). A nested polymerase chain reaction has previously been applied to the amplification of 16S ribosomal RNA genes
109
to identify H. pylori in the mouths and stomachs of patients with gastritis. It was suggested by this group that the organism was present in the mouths of subjects, and as such they identified a potential source of reinfection. 3° Inhibition of PCR by materials in highly contaminated specimens is well documented. Wilson 3I described a reduction in sensitivity of PCR for the amplification and detection of staphylococcal genes associated with virulence. In this setting ca. l 0 s cfu/ml of artificially contaminated dried skimmed milk could be detected using nested primers. 31 Others have described variations as wide as 1000-fold in the sensitivity of PCR when faeces samples from different individuals are processed for the detection of E. coli genes, j7 It is thought that a variety of substances found in foods, faeces, soil and other organic matter m a y inhibit polymerase enzymes 17, 31.32 and it has been documented that specific molecules such as urea, haemoglobin, heparin and traces of phenol can inhibit 1hq polymerase. 17 An increase in the sensitivity of PCR m a y be obtained by time-consuming extraction methods using phenol, but in our hands this has not been found to be necessary. The method we have described, while still exhibiting some variation in sensitivity when applied to faeces, does provide a method which will detect H. pylori in clinical samples containing between 10 and 100 cfu. The extraction procedure is relatively simple, making this method suitable for both epidemiological and transmission studies. Investigation of a variety of food sources and environmental samples which m a y act as vehicles for transmission should also be possible. It is important to assess the sensitivity of the PCR protocol specifically for each sample type to be investigated, otherwise negative results are inconclusive. This method m a y have a number of applications in the clinical setting as well as in epidemiological studies. It is suitable for the investigation of faeces or oral samples of patients known to be infected with H. pylori. It is also suitable as either a diagnostic tool or as an instrument for post-treatment follow-up. It would be particularly interesting to assess PCR as a method for detecting very small numbers of t-1. pylori or non-culturable forms in the time period immediately following treatment. This would allow early post-treatment evaluation and assessment of the need for further therapy. In conclusion, we have found considerable differences in the sensitivity of PCR for the detection of tt. pylori in clinical samples of different types when compared with pure culture. This m a y be due to specific and non-specific inhibitors of polymerase enzymes. It is also known that use of the 'Hot-start' method is likely to improve the limits of detection and reduce interference from nonspecific effects. 32 In our hands the sensitivity of PCR
K. B. Bamford e t al.
110
can be increased by lO-lO00-fold by using a second amplification procedure with nested primers, and by a further lO-fold when the amplified product is hybridized with an intragenic probe. It is important that the sensitivity of any PCR technique which is to be applied to the detection of H. pylori is assessed for each sample type to be investigated.
Acknowledgements We would like to acknowledge the support of this work by grants NG 37 and CG13 from the DHSS (NI). We would also like to thank Dr. P. Winter for advice regarding PCR and Prof. D.I.H. Simpson for advice and encouragement.
References 1 BlaserMJ. Helicobacterpyloriandthepathogenesisofgastroduodenal inflammation. J InfDis 1990; 161: 626-633. 2 Soll AH. Pathogenesis of peptic ulcer and implications for therapy. N E n g l ] M e d 1990; 322: 909-916. 3 Lee A, Fox J, Hazell S. Pathogenicity of Helicobacter pylori: a perspective. Infect Immun 1993; 61: 1601-1610. 4 Parsonnet J, Freidman GD, Vandersteen DP et al. Helieobacter pylori infection and the risk of gastric carcinoma. N Engl J Med 1991; 325: 1127-1131. 5 Wotherspoon AC, Doglioni C, Diss TC et al. Regression of primary low grade B-cell gastric lymphoma of mucosa-associated lymphoid tissue type after eradication of Helicobacterpylori. Lancet 1993; 342: 575-577. 6 Hentschel E, Brandstatter Get al. Effect ofranitidine and amoxycillin plus metronidazole on the eradication of Helicobaeter pylori and the recurrence of duodenal ulcer. N Eng f Med 1993; 328: 308-312. 7 Rauws EAJ, Tytgat GNJ. Cure of duodenal ulcer associated with eradication of Helicobacter pylori. Lancet 1990; 335: 1233-1235. 8 Chiba N, Rao BV, Rademaker JW, Hunt RH. Meta-analysis of the efficacy of antibiotic therapy in eradicating Helicobaeter pylori. Am J Gastroenterol 1992; 87: 1716-1727. 9 0 r m a n d JE, Talley NJ. Helicobacter pylori: controversies and an approach to management. Mayo Clin Proc 1990; 65: 414-426. 10 Rauws EAJ. Therapeutic attempts at erradication of Campylobaeter pylori. Eur.J.Gastroenterol HepatoI 1989;1: 34-41. 11 Bode G, Mauch F, Malfertheiner P. The coccoid forms of Helicobacter pylori. Criteria for their viability. Epidemiol Infect 1993; 111: 483490. 12 Shames B, Krajden S, Fuksa M, Babida C, Penner JL. Evidence for the occurrence of the same strain of Campylobacter pylori in the stomach and dental plaque. J Clin Microbiol 1989; 27:2849-2850. 13 Drumm B, Perez-Perez GI, Blaser MJ, Sherman PM. Intrafamilial clustering of Helicobacter pylori infection. N Engl J Med 1990; 322: 359-363. 14 Bamford KB, Bickley J, Collins JS et al. Helieobacterpylori: comparison of DNA finger prints provides evidence for intrafamilial infection. Gut 1993; 34: 1348-1350.
15 Thomas IE, Gibson GR, Darboe MK, Dale A, Weaver LT. Isolation of Helieobacter pylori from human faeces. Lancet 1992; 340: 11941195. 16 Eisenstein BI. The polymerase chain reaction. A new method of using molecular genetics for medical diagnosis. N Engl J Med 199 O: 322: 178-183. 17 Saulnier P, Andremont A. Detection of genes in feces by booster polymerase chain reaction. ] Clin Microbiol 1992; 30: 2080-2083. 18 Labigne A, Cussac V, Courcoux P. Development of the genetic and molecular approaches for the diagnosis and study of the pathogenicity of Helicobacter pylori, agent of gastric inflammatory diseases. Bulletin de l'Academie Nationale de Medecine 1991; 175: 791-802. 19 Barow GI, Feltham RKA, eds. Cowan and Steel's Manual for the determination of medical bacteria, 3rd edition. Cambridge University Press, 1993. 20 Miles AA, Misra SS. The estimation of bactericidal power of blood. J Hyg (Lond) 1938; 38: 732. 21 Hammar M, TyszkiewiczT, Wadstrom T, O'Toole PW. Rapid detection of Helicobacterpylori in gastric biopsy material by polymerase chain reaction. ] Clin Micmbiol 1992; 30: 54-58. 22 Labigne A, Cussac V, Courcoux P. Shuttle cloning and nucleotide sequences of Helicobacterpylori genes responsible for urease activity. ] Bacteriol 1991; 173: 1920-1931. 23 Cussac V, Ferrero RL, Labigne A. Expressior~ of Helicobacter pylori urease genes in Esehericia eoli grown under nitrogen-limiting conditions. ] Bacteriol 1992; 174: 2466-2473. 24 Clayton CL, Kleanthous H, Coates PJ, Morgan DD, Tabaqchali S. Sensitive detection of Helicobacter pylori by using polymerase chain reaction. J Clin Microbiol 1992; 30: 192-200. 25 Westblom TU, Phadnis S, Yang P, Czinn SJ. Diagnosis of Hefieobacter pylori infection by means of a polymerase chain reaction assay for gastric juice aspirates. Clin InfDis 1993; 16: 367-371. 26 Clayton CL, Kleanthous H, Morgan DD, Puckey L, Tabaqchali S. Rapid fingerprinting of Helicobaeter pylori by polymerase chain reaction and restriction fragment length polymorphism analysis. ] Clin Microbiol 1993; 31: 1420-1425. 27 Lopez CR, Owen RJ, Desai M. Differentiation between isolates of Helicobacter pylori by PCR-RFLP analysis of urease A and B genes and comparison with ribosomal RNA gene patterns. FEMS Mierobiol Letts 1993; 110: 37-43. 28 Bickley J, Owen RJ, Fraser AG, Pounder RE. Evaluation of the polymerase chain reaction for detecting the urease C gene of Helicobacter pylori in gastric biopsy samples and dental plaque. ] Med Mierobiol 1993; 39: 388-44. 29 Wang IT, Lin JT, Sheu JC, Yang JC, Chen DS, Wang TH. Detection of Helicobacter pylori in gastric biopsy tissue by polymerase chain reaction. Eur ] Clin Microbiol lnf Dis 1993; 12: 367-371. 30 Nguyen AMH, Egstrand L, Genta RM, Graham DY, el-Zaatari FAK. Detection of Helicobaeter pylori in dental plaque by reverse transcription-polymerase chain reaction. J Clin Microbiol 1993; 31: 783-787. 31 Wilson IG, Cooper JE, Gilmour A. Detection of enterotoxigenic Staphylococcus aureus in dried skimmed milk: use of the polymerase chain reaction for amplification and detection of staphylococcal enterotoxin genes entB and entCI and the thermonuclease gene nuc. Appl Environ Microbiol 1991; 57: 1793-1798. 32 Tsai YL, Olson BH. Detection of low numbers of bacterial cells in soils and sediments by polymerase chain reaction. Appl Environ Microbiol 1992; 58: 754-757.
Nested Primers Improve Sensitivity in the Detection of Helicobacter pylori by the Polymerase Chain Reaction K. B. Bamford *~, D. A. Lutton ~, B. O'Loughlin ~, W. A. Coulter 2 and J. S. A. Collins 3 ~Department of Microbiology and Immunobiology, 2The School of Dentistry, The Queen's University of Belfast, Grosvenor Road, Belfast BT12 6BN and 3The Royal Victoria Hospital, Belfast, U.K. To investigate potential routes of spread of infection by the polymerase chain reaction (PCR) it is important that the technique is effective in the types of specimen to be investigated. To establish the limits of detection of Helicobacter pylori by PCR in clinical material from the gastric mucosa, faeces, dental plaque and oral rinses, samples were seeded with known numbers of bacteria. DNA extraction was followed by amplification with primers from the urease C gene. Nested primers were used to amplify the PCR product which was detected using a digoxigenin-labeiled probe. Faeces or plaque inhibited the single reaction 102-106 fold. A second amplification using nested primers and probing increased the sensitivity to a level similar to that obtained with pure culture. This method is potentially useful with less likelihood of false negative results when trying to detect H. pylori by PCR in highly contaminated, clinical material.
Introduction Helicobacter pylori is the commonest cause of inflammation of the gastric mucosa and is fundamental to the pathogenesis of peptic ulceration. I-3 Infection with H. pylori may also be a risk factor for the subsequent development of gastric carcinoma and this bacterium has been linked with MALT lymphoma. *' ~ This provides a strong argument for the eradication of H. pylori and the use of antimicrobial agents in the treatment of duodenal ulceration. This is particularly so because elimination of the infection greatly reduces the risk of ulcer relapse, thus reducing the need for anti-ulcer maintenance therapy. 6 7 However, in a significant proportion of cases the organism is not completely eliminated at the first attempff and the occurrence of relapse/reinfection is well documented. 9 Helicobacter pylori may not be detected by histology, culture, rapid urease test or by urea breath tests immediately following a course of treatment; however, patients may be found to be H. pylori-positive when assessed 6 - 1 2 months later. ~° Early follow-up studies after treatment can be therefore be misleading. It is unclear whether this is due to recurrence of bacteria which are partially inhibited and may have persisted in a coccoid form, M or whether reinfection from another source has occurred. Reinfection of the gastric mucosa
*Address correspondence to: Dr. K. B. Bamford. Accepted for publication 7 May 1997. 0163-4453/98/010105 +06 $12.00/0
may occur via bacteria located within the oral cavity, ~2 or via an external source, e.g. another person. The mode of transmission is unclear, although there is evidence for person-to-person transmission, l~ The same strain of H. pylori has been found in more than one individual within several family settings, suggesting that intrafamilial crossinfection may occur. 1. Part of the problem in determining the source of infection is that it is difficult to culture H. pylori from heavily contaminated clinical specimens, and few descriptions exist of isolation of this bacterium from such samples, is This is partly due to the fastidious nature of H. pylori, but may also be influenced by the occurrence of metabolically quiescent non-culturable forms. ~ Thus, accurate methods of determining both the response to therapy, without long-term follow-up, and for assessing possible routes of transmission of H. pylori are required. The polymerase chain reaction (PER) can be used to specifically detect minute quantities of impure DNA by chemical rather than biological means. ~6 In order to apply this technique to a routine diagnostic laboratory setting, a simplified standard procedure is necessary where the limitations of the method are known. The specificity of the PeR is dependent on the choice of oligonucleotide primers, and, in theory, it is sensitive enough to detect one copy of target DNA. In practice, however, interfering material within many samples may inhibit the PCR, thus reducing the sensitivity of the method and the significance of a negative result/7 To this end, we have developed a PeR assay to detect H. © 1998 The British ;~nIection Society
K. B. Bamford et a t
106
pylori within a range of clinical samples. Oligonucleotide primers, specific for tt. pylori, were selected based on a conserved region of the urease C gene 18 and nested primers were used to increase the sensitivity of the technique.
Materials and Methods To determine the sensitivity of the assay and to investigate the possible effects of inhibitory impurities from clinical specimens, H. pylori-negative clinical samples of antral biopsy tissue, faeces, and dental plaque were artiflcally seeded with serial logarithmic dilutions of I-I. pylori prior to PCR analysis.
Clinical samples Antral biopsy specimens were obtained during routine examinations of the gastric mucosa at the time of oesophagogastrodnodenoscopy from consenting patients. Samples from patients who were CLO test (Delta West, Australia) and H. pylori culture negative were selected for sensitivity determination. Extra care was taken to decontaminate thoroughly both the endoscope and biopsy forceps between patients. Specimens for PCR analysis were snap frozen in liquid nitrogen directly following excision, to minimize the risk of contamination. For routine culture, biopsy specimens were transported to the laboratory in 0.01 M phosphate buffered saline (PBS) pH 7.4 and inoculated directly onto Colombia base agar (Oxoid) containing 7% defibrinated horse blood, plus a second agar with antibiotic supplement (vancomycin 10 mg/ml, trimethoprim lactate 5 mg/ml, amphotericin B 5 mg/ml and nalidixic acid 20 mg/ml). Plates were cultured at 37°C for up to 5 days in a microaerophillic atmosphere of 5% O2, 1% H2, 7% CO2 and 87% N2 (Don Whitley variable atmosphere incubator). Isolates were identified as/4. pylori if they were Gram-negative with a typical morphology, hydrolysed urea rapidly and were catalase and oxidase positive. 19 Subgingival plaque samples were obtained by curetting the depth of the gingival sulcus or pocket and were immediately placed in transport buffer (PBS). Samples were stored at - 2 0 °C prior to analysis. Oral rinse fluid was obtained by rinsing 15 ml PBS in the mouth for 10 s before spitting into a sterile universal container. Faeces samples were obtained from the diagnostic microbiology laboratories of the Royal Hospitals Trust, where they had been submitted for the investigation of gastrointestinal symptoms.
Titration of bacteria for sample inoculation Columbia base agar (Oxoid) containing 7% defibrinated horse blood was used to culture the bacteria at 37 °C for up to 5 days in a microaerophillic atmosphere as above. Bacteria were harvested, washed and suspended in sterile distilled water. Counts were determined using the method of Miles and Misra 2° and verified by dark field microscopy. Serial 10-fold dilutions of the bacterial culture were prepared in distilled water. Aliquots (300 pl) of bacterial culture, containing a known concentration of bacteria, were added to plaque, faeces, oral rinse fluid or biopsy specimens.
DNA extraction Plaque and faeces samples were suspended in sterile distilled water to an optical density of 0.2 and 0.3, respectively. Oral rinse fluids were used at the concentration provided from the clinic. Plaque, faeces and oral rinse fluids were sonicated (3 x 5 s bursts at 5 ~tm) to dissociate aggregates. Suspensions of plaque, faeces or oral rinse fluid (200 bd) were transferred to microcentrifuge tubes (Perkin-Elmer Cetus) containing 300 gl aliquots of a known concentration of bacterial culture. Bacterial control samples (pure culture) were processed in the absence of clinical material. The samples were centrifuged (12 000 x g, 3 min) in a microcentrifuge (Eppendorf 5414), the supernatant decanted and 300 btl of extraction buffer (Tris (20 mM) pH 8 and 0.5% v/v Tween 20) added. Biopsy specimens were removed from liquid nitrogen and known concentrations of bacterial culture added. The specimen was then suspended in extraction buffer (300 btl). All specimens were vortexed and proteinase K (Sigma) was added to a final concentration of 0.Smg/ml. Following incubation ( l h , 37°C, the proteinase K was inactivated by heating the sample to 98 °C for 10 rain. 21 Twenty microlitres of DNA template were added per 100 btl of PCR reaction mix. Negative control samples (sterile, double distilled deionised water) were processed in parallel for each extraction.
PCR primers Two regions from each end of a highly conserved 294 base pair region of the urease C (ureC) gene of H. pylori were used for primer 1 (5'AAGCTTTTAGGGGTGTTAGGGGTTT-3') and primer 2 (5"-AAGCTTACTTTCTAACACTAACGC-3'), as previously described by Labigne et al. 18 A second set of primers; primer 3 (5'-CTTTCTTCTCAAGCAATTGTC-3') and primer 4 (5"-CAAGCCATCGCCGGTTTTAGC-3"), were
Helicobacter pylori: nested PCR sensitivity used to amplify a 252 base pair region located 21 base pairs internal to primers 1 and 2. Primer pairs 1 and 2, and 3 and 4, were purchased from The Department of Biochemistry, Queen's University of Belfast and British Bio-technology Products Ltd., U.K., respectively. A 44-mer oligonucleotide probe (5'A G A A T T G A A GC ATTGCGCG ATTGGGGATAA GTTTGTGAGCGAAT-3'), described by Labigne, is which recognized a region central to both primer sets was used to confirm the identity of the amplified product.
!
107
2 3 4 5 6 7 8 9 1011121314
15181718192021
22232425262728
29
Figure l. Results of PCR using primers 1 and 2. Pure culture (lanes 1-7), faeces I (lanes 8-14), faeces 2 (lanes 15-21, faeces 3 (lanes 22-28), molecular weight markers (lane 29) X (352 bases) and Y (212 bases). Ten-fold reductions in number of cfu in sample, 10 ~ (lanes 1, 8,15,22), 10 s (lanes 2,9,16,23), 104 (lanes 3,10,17,24), 103 (lanes 4,11,18,25), 10 z (lanes 5,12,19,26), 10 (lanes 6,13,20,27), 10 -t (lanes 7, 14.21,28).
DNA Amplification The PCR reaction mixture contained KC1 (lOmM), (NH4)2SO4 (10mm), Tris-HCl pH 8.8 (20mM), MgSO4 (2 raM), Triton X-100 (0.1% v/v), deoxynucleotide triphosphates (200 laM of each dATER dCTER dGTER and dTTP), the relevant primers (1 t~g/lO0 ~tl reaction volume), Vent DNA polymerase (New England Biolabs) at 1 unit/100 btl reaction volume and I)NA template (typically 20 lal per 100 t~1 of PCR reaction mixture). The DNA polymerase was added immediately prior to PCR incubation to reduce 3' to 5' exonuclease activity. Negative control samples, in which the DNA template was replaced by sterile distilled water, were included in each PCR assay to eliminate the possibility of false-positive reactions. In addition, samples containing positive control DNA were included to check that samples were amplifiable, and to exclude the possibility that a negative result was due to a fault in the reaction process. The reaction mixture was overlaid with mineral oil, and placed in a Perkin Elmer 480 Thermal Cycler. Denaturation, annealing, and extension temperatures and times were 95 °C for 2 min, 55 °C for 2 m i n and 75 °C for 1 min, respectively. A maximum of 35 cycles was run. The primary PeR reaction product was subsequently incubated with primers 3 and 4:(2 btl PCR product/100 bd PCR reaction mixture) for a further 35 cycles.
1 2 3
4 5 6 7 8 9 1011121314151617181920212223242526272829
Figure 2. Results of PCR using primers 1 and 2. Pure culture (lanes 1-7), plaque 1 (lanes 8-t4), plaque 2 (lanes 15-21, plaque 3 (lanes 22-28). molecular weight markers (lane 29). Ten-fold reductions in number of cfu in sample, as in Fig. 1.
fixed by baking (120 °C, 30 min). The membrane was prehybridized (68°C, l h) in a hybridization buffer (5 × SSC (20 x SSC is NaCI (3 M), Na citrate (0.3 g), pH 7.0), Boehringer Mannheim Biochemica blocking reagent (1%), N-lauroylsarkosine (0.1% w/v), and sodium dodecyl sulfate (0.02% 5v/v)). Hybridization of the labelled oligonucleotide probe (50 pM/ml in hybridization buffer) to target DNA was carried out at 37 °C for 5 h. Unhybridized label was removed by washing in tetramethylammonium chloride (2 x 70 °C, 20 min). DIG-labelled DNA was detected using an enzyme-linked immunoassay incorporating an anti-digoxigenin linked alkaline phosphatase conjugate (Boehringer Mannheim Biochemica).
Detection of amplified DNA
Results
PCR-amplified DNA was resolved by electrophoresis on 2% agarose gels (Sigma), stained with ethidium bromide (1 gg/ml) and visualized using a UV transilluminator. The specificity of the PeR product was verified by Southern blotting using the 4Zt-mer oligonucleotide probe described above. Briefly, the probe was non-isotopically 3' end labelled using terminal transferase, by incorporation of a single digoxigenin (DfG)-labelled dideoxyuridine-triphosphate (Boehringer Mannheim Biochemica). Amplified DNA fragments were transferred to a Hybond-N nylon membrane (Amersham) by Southern blotting and
The limits of detection achieved using the single set of primers, primers 1 and 2, was of the order of 10 colony forming units (cfu) of H. pylori when a suspension of pure culture was used as the original sample (Figs 1 and 2, lanes 1-7). The sensitivity of detection was reduced when the first set of primers only was used in PCR to detect H. pylori in the prepared samples of faeces (n = 3) (Fig. 1, lanes 8-14, lanes 15-21, and lanes 22-28). A reduction in the limit of detection of H. pylori was also seen when different samples of dental plaque (n = 3) were examined (Fig. 2 lanes 8-14, lanes 15-21, and lanes
108
1 2
K.B. Bamford et aL
34
5 6 7 8 9 101112131415161718192021222324252627282930
Figure 3. Results of PCR using nested primers (1 and 2 for reaction 1, followed by 3 and 4 for reaction 2). Gastric mucosaI biopsy (lanes 1-7), oral rinse (lanes 8-14), plaque 1 (lanes 15-21, plaque 2 lanes 22-28), negative control (lane 29), molecular weight markers (lane 30). Tenfold reductions in number of cfu in sample, as in Fig, 1.
1
2
3
4
5
6
7
8
9
10 11
12 13
14 15 16
Figure 5. Results of PCR using nested primers (1 and 2 for reaction l, followed by 3 and 4 for reaction 2) (top) and hybridization with the 44 base DNA probe with immunodetection (bottom). Faeces 4 (lanes 1-7), faeces 5 (lanes 8-14), negative control (lane 15), molecular weight markers (lane 16). Ten4old reductions in number of cfu in sample, as in Fig. 4.
~X
~Y
1
2
3
4
5
6
7
8
9
10 11
12 13
14 1516
Figure 4. Results of PCR using nested primers (1 and 2 for reaction 1, followed by 3 and 4 for reaction 2) (top) and hybridization with the 44 base DNA probe with immunodetection (bottom). Faeces 1 (lanes 1-7), faeces 2 (lanes 8-14), negative control (lane 15), molecular weight markers (lane 16). Ten-fold reductions in number of cfu in sample, 106 (lanes 1, 8,), 10 s (lanes 2,9), 104 (lanes 3,10). 103 (lanes 4.11), 102 (lanes 5,12), 10 (lanes 6,13), 10 -1 (lanes 7,14).
2 2 - 2 8 ) . The detection limit r a n g e d from 104-102 in different faecal a n d p l a q u e specimens b u t was in all cases less t h a n t h a t for the p u r e culture. (Figs 1 a n d 2). I n c r e a s e d sensitivity w a s achieved by e m p l o y i n g a nested PCR, w h e r e the second r e a c t i o n used primers 3 a n d 4, w i t h the r e a c t i o n p r o d u c t of the first r e a c t i o n as the target template. In d e n t a l p l a q u e p r e p a r a t i o n s less t h a n 1 0 c f n were consistently detected (Fig. 3, lanes 1 5 - 2 1 a n d lanes 2 2 - 2 8 ; Fig. 4, lanes 8 - 1 4 ) . A similar level of detection w a s o b t a i n e d from oral rinse fluid (Fig. 3, lanes 8 - 1 4 ) , a n d biopsy specimens (Fig. 3, lanes 1 - 7 ) . S o u t h e r n blotting a n d DNA-DNA h y b r i d i z a t i o n w i t h the oligonucleotide probe confirmed the specificity of the amplified p r o d u c t (Figs 4 - 6 ) . Less t h a n 10 cfu could be detected from m o s t faeces samples by using the nested
1
2
3
4
5
6
7
8
9
10 11 12 13 14 1516
Figure 6. Results of PCR using nested primers (1 and 2 for reaction 1, folIowed by 3 and 4 for reaction 2) (top) and hybridization with the 44 base DNA probe with immunodetection (bottom). Gastric biopsy (lanes 1-7), Plaque (lanes 8-14), negative controI (lane 15), molecular weight markers (lane 16). Ten-fold reductions in number of cfu in sample, as in Fig. 4.
p r i m e r r e a c t i o n (Figs 5 a n d 6). The h i g h e s t limit of detection achieved with a faecal s a m p l e was 10 2 cfu (Fig. 5), even w i t h the lO-fold increase in sensitivity achieved as a result of hybridization w i t h the probe. Vent DNA p o l y m e r a s e (New E n g l a n d Biolabs) was used for t h e PCR, as the sensitivity w i t h this e n z y m e was consistently 10fold g r e a t e r t h a n t h a t achieved w i t h AmpliTaq DNA p o l y m e r a s e (Perkin Elmer-Cetus).
Discussion We h a v e described a rapid, sensitive a n d specific PCR a s s a y for 17. pylori w h i c h is suitable for use w i t h i n a clinical laboratory, This m e t h o d does n o t require extensive e x t r a c t i o n a n d purification of t h e DNA prior to amplification. We h a v e assessed the sensitivity of this t e c h n i q u e w h e n applied to different clinical s a m p l e types; t h a t is, p u r e culture, gastric m u c o s a l biopsies, d e n t a l plaque, oral rinse fluid, a n d faeces. This a s s a y relies
Helicobacter pylori: nested PCR sensitivity solely on the presence of target DNA rather than viable, metabolizing cells. ~6 Thus, it is suitable for detection of 17. pylori within biopsy specimens where viability m a y have been lost due to recent antibiotic therapy or delays in transportation to the laboratory. It is also appropriate for samples such as faeces and dental plaque, from which it is difficult to isolate H. pylori because of the presence of very large numbers of other, less fastidious, bacteria. In our experiments we have been able to detect tt. pylori in highly contaminated samples which contain large numbers of both prokaryotic and eukaryotic cells as well as m a n y organic impurities. The urease gene region of H. pylori is a suitable target site for quantitative PCR because, despite a very high level of expression, the DNA is present as a single chromosomal copy per bacterial cell. = The urease region contains nine genes; four of which (ureA, ureB, ureC, and ureD) confer a urease positive phenotype upon Campylobacter jejuni, 22 and five accessory genes (ureE, ureE ureG, uteri, and ureI), three of which (ureF, ureG and uteri) are required for urease expression in Escherichia coll. 23 UreA and ureB code for the structural A and B subunits of the urease enzyme, 23 but the functions of the other genes associated with urease expression are, as yet, less clear. Urease is a highly conserved enzyme, and there is considerable homology between amino acid sequences of naturally occurring ureases from both plant and bacterial sources including that of 11. pglori. 22 Detection of H. pglori by PCR using primer sites on the ureA and ureB genes has previously been described. 24-2s Selected primer sites from these genes have also been used to amplify large variable regions of DNA, the products of which have subsequently been analysed by restriction fragment length polymorphism and used as a typing method. 2<27 In the present study, the ureC gene which is unique to H. p!lIori was selected as a suitable target site for PCR. Bickley et al. demonstrated the specificity of this gene by examining a wide range of bacteria, 28 and the PCR technique has been applied to gastric biopsies. 29 Our PCR using primers 1 and 2, and the intragenic probe, was highly sensitive when applied to pure culture and gastric biopsies. Helicobacter pylori was detected when as few as 10 organisms were present (Figs 1 and 2). However, when applied to faeces and plaque, the PCR was inhibited, with the sensitivity of the technique in some cases decreased to 10 4 CFU. The degree of inhibition of the PCR varied between individual specimens but was in all cases lower than that with pure culture (Figs 1 and 2). By using nested primers and a small quantity of the original reaction mixture as the template, the sensitivity of the reaction was considerably increased (Fig. 3). A nested polymerase chain reaction has previously been applied to the amplification of 16S ribosomal RNA genes
109
to identify H. pylori in the mouths and stomachs of patients with gastritis. It was suggested by this group that the organism was present in the mouths of subjects, and as such they identified a potential source of reinfection. 3° Inhibition of PCR by materials in highly contaminated specimens is well documented. Wilson 3I described a reduction in sensitivity of PCR for the amplification and detection of staphylococcal genes associated with virulence. In this setting ca. l 0 s cfu/ml of artificially contaminated dried skimmed milk could be detected using nested primers. 31 Others have described variations as wide as 1000-fold in the sensitivity of PCR when faeces samples from different individuals are processed for the detection of E. coli genes, j7 It is thought that a variety of substances found in foods, faeces, soil and other organic matter m a y inhibit polymerase enzymes 17, 31.32 and it has been documented that specific molecules such as urea, haemoglobin, heparin and traces of phenol can inhibit 1hq polymerase. 17 An increase in the sensitivity of PCR m a y be obtained by time-consuming extraction methods using phenol, but in our hands this has not been found to be necessary. The method we have described, while still exhibiting some variation in sensitivity when applied to faeces, does provide a method which will detect H. pylori in clinical samples containing between 10 and 100 cfu. The extraction procedure is relatively simple, making this method suitable for both epidemiological and transmission studies. Investigation of a variety of food sources and environmental samples which m a y act as vehicles for transmission should also be possible. It is important to assess the sensitivity of the PCR protocol specifically for each sample type to be investigated, otherwise negative results are inconclusive. This method m a y have a number of applications in the clinical setting as well as in epidemiological studies. It is suitable for the investigation of faeces or oral samples of patients known to be infected with H. pylori. It is also suitable as either a diagnostic tool or as an instrument for post-treatment follow-up. It would be particularly interesting to assess PCR as a method for detecting very small numbers of t-1. pylori or non-culturable forms in the time period immediately following treatment. This would allow early post-treatment evaluation and assessment of the need for further therapy. In conclusion, we have found considerable differences in the sensitivity of PCR for the detection of tt. pylori in clinical samples of different types when compared with pure culture. This m a y be due to specific and non-specific inhibitors of polymerase enzymes. It is also known that use of the 'Hot-start' method is likely to improve the limits of detection and reduce interference from nonspecific effects. 32 In our hands the sensitivity of PCR
K. B. Bamford e t al.
110
can be increased by lO-lO00-fold by using a second amplification procedure with nested primers, and by a further lO-fold when the amplified product is hybridized with an intragenic probe. It is important that the sensitivity of any PCR technique which is to be applied to the detection of H. pylori is assessed for each sample type to be investigated.
Acknowledgements We would like to acknowledge the support of this work by grants NG 37 and CG13 from the DHSS (NI). We would also like to thank Dr. P. Winter for advice regarding PCR and Prof. D.I.H. Simpson for advice and encouragement.
References 1 BlaserMJ. Helicobacterpyloriandthepathogenesisofgastroduodenal inflammation. J InfDis 1990; 161: 626-633. 2 Soll AH. Pathogenesis of peptic ulcer and implications for therapy. N E n g l ] M e d 1990; 322: 909-916. 3 Lee A, Fox J, Hazell S. Pathogenicity of Helicobacter pylori: a perspective. Infect Immun 1993; 61: 1601-1610. 4 Parsonnet J, Freidman GD, Vandersteen DP et al. Helieobacter pylori infection and the risk of gastric carcinoma. N Engl J Med 1991; 325: 1127-1131. 5 Wotherspoon AC, Doglioni C, Diss TC et al. Regression of primary low grade B-cell gastric lymphoma of mucosa-associated lymphoid tissue type after eradication of Helicobacterpylori. Lancet 1993; 342: 575-577. 6 Hentschel E, Brandstatter Get al. Effect ofranitidine and amoxycillin plus metronidazole on the eradication of Helicobaeter pylori and the recurrence of duodenal ulcer. N Eng f Med 1993; 328: 308-312. 7 Rauws EAJ, Tytgat GNJ. Cure of duodenal ulcer associated with eradication of Helicobacter pylori. Lancet 1990; 335: 1233-1235. 8 Chiba N, Rao BV, Rademaker JW, Hunt RH. Meta-analysis of the efficacy of antibiotic therapy in eradicating Helicobaeter pylori. Am J Gastroenterol 1992; 87: 1716-1727. 9 0 r m a n d JE, Talley NJ. Helicobacter pylori: controversies and an approach to management. Mayo Clin Proc 1990; 65: 414-426. 10 Rauws EAJ. Therapeutic attempts at erradication of Campylobaeter pylori. Eur.J.Gastroenterol HepatoI 1989;1: 34-41. 11 Bode G, Mauch F, Malfertheiner P. The coccoid forms of Helicobacter pylori. Criteria for their viability. Epidemiol Infect 1993; 111: 483490. 12 Shames B, Krajden S, Fuksa M, Babida C, Penner JL. Evidence for the occurrence of the same strain of Campylobacter pylori in the stomach and dental plaque. J Clin Microbiol 1989; 27:2849-2850. 13 Drumm B, Perez-Perez GI, Blaser MJ, Sherman PM. Intrafamilial clustering of Helicobacter pylori infection. N Engl J Med 1990; 322: 359-363. 14 Bamford KB, Bickley J, Collins JS et al. Helieobacterpylori: comparison of DNA finger prints provides evidence for intrafamilial infection. Gut 1993; 34: 1348-1350.
15 Thomas IE, Gibson GR, Darboe MK, Dale A, Weaver LT. Isolation of Helieobacter pylori from human faeces. Lancet 1992; 340: 11941195. 16 Eisenstein BI. The polymerase chain reaction. A new method of using molecular genetics for medical diagnosis. N Engl J Med 199 O: 322: 178-183. 17 Saulnier P, Andremont A. Detection of genes in feces by booster polymerase chain reaction. ] Clin Microbiol 1992; 30: 2080-2083. 18 Labigne A, Cussac V, Courcoux P. Development of the genetic and molecular approaches for the diagnosis and study of the pathogenicity of Helicobacter pylori, agent of gastric inflammatory diseases. Bulletin de l'Academie Nationale de Medecine 1991; 175: 791-802. 19 Barow GI, Feltham RKA, eds. Cowan and Steel's Manual for the determination of medical bacteria, 3rd edition. Cambridge University Press, 1993. 20 Miles AA, Misra SS. The estimation of bactericidal power of blood. J Hyg (Lond) 1938; 38: 732. 21 Hammar M, TyszkiewiczT, Wadstrom T, O'Toole PW. Rapid detection of Helicobacterpylori in gastric biopsy material by polymerase chain reaction. ] Clin Micmbiol 1992; 30: 54-58. 22 Labigne A, Cussac V, Courcoux P. Shuttle cloning and nucleotide sequences of Helicobacterpylori genes responsible for urease activity. ] Bacteriol 1991; 173: 1920-1931. 23 Cussac V, Ferrero RL, Labigne A. Expressior~ of Helicobacter pylori urease genes in Esehericia eoli grown under nitrogen-limiting conditions. ] Bacteriol 1992; 174: 2466-2473. 24 Clayton CL, Kleanthous H, Coates PJ, Morgan DD, Tabaqchali S. Sensitive detection of Helicobacter pylori by using polymerase chain reaction. J Clin Microbiol 1992; 30: 192-200. 25 Westblom TU, Phadnis S, Yang P, Czinn SJ. Diagnosis of Hefieobacter pylori infection by means of a polymerase chain reaction assay for gastric juice aspirates. Clin InfDis 1993; 16: 367-371. 26 Clayton CL, Kleanthous H, Morgan DD, Puckey L, Tabaqchali S. Rapid fingerprinting of Helicobaeter pylori by polymerase chain reaction and restriction fragment length polymorphism analysis. ] Clin Microbiol 1993; 31: 1420-1425. 27 Lopez CR, Owen RJ, Desai M. Differentiation between isolates of Helicobacter pylori by PCR-RFLP analysis of urease A and B genes and comparison with ribosomal RNA gene patterns. FEMS Mierobiol Letts 1993; 110: 37-43. 28 Bickley J, Owen RJ, Fraser AG, Pounder RE. Evaluation of the polymerase chain reaction for detecting the urease C gene of Helicobacter pylori in gastric biopsy samples and dental plaque. ] Med Mierobiol 1993; 39: 388-44. 29 Wang IT, Lin JT, Sheu JC, Yang JC, Chen DS, Wang TH. Detection of Helicobacter pylori in gastric biopsy tissue by polymerase chain reaction. Eur ] Clin Microbiol lnf Dis 1993; 12: 367-371. 30 Nguyen AMH, Egstrand L, Genta RM, Graham DY, el-Zaatari FAK. Detection of Helicobaeter pylori in dental plaque by reverse transcription-polymerase chain reaction. J Clin Microbiol 1993; 31: 783-787. 31 Wilson IG, Cooper JE, Gilmour A. Detection of enterotoxigenic Staphylococcus aureus in dried skimmed milk: use of the polymerase chain reaction for amplification and detection of staphylococcal enterotoxin genes entB and entCI and the thermonuclease gene nuc. Appl Environ Microbiol 1991; 57: 1793-1798. 32 Tsai YL, Olson BH. Detection of low numbers of bacterial cells in soils and sediments by polymerase chain reaction. Appl Environ Microbiol 1992; 58: 754-757.