Genotypes of Helicobacter pylori obtained from gastric ulcer patients taking or not taking NSAIDs

THE AMERICAN JOURNAL OF GASTROENTEROLOGY © 1999 by Am. Coll. of Gastroenterology Published by Elsevier Science Inc.

Vol. 94, No. 6, 1999 ISSN 0002-9270/99/$20.00 PII S0002-9270(99)00189-6

Genotypes of Helicobacter pylori Obtained From Gastric Ulcer Patients Taking or Not Taking NSAIDs Li Li, Ph.D., Lisa K. Kelly, M.S., Kamran Ayub, M.D., David Y. Graham, M.D., and Mae F. Go, M.D. Department of Medicine, Digestive Diseases Section, Veterans Affairs Medical Center and Baylor College of Medicine, Houston, Texas

OBJECTIVE: Whether Helicobacter pylori infection and use of nonsteroidal antiinflammatory drugs (NSAIDs) are independent risk factors for ulcerogenesis remains unclear. We undertook this study to evaluate H. pylori isolates from gastric ulcer patients to determine whether the genotype of the infecting isolate could be correlated with the use or nonuse of NSAIDs. METHODS: Fifty-two patients presenting with gastric ulcer and infected with H. pylori were included; 26 patients were taking NSAIDs or aspirin (ASA) regularly at the time of ulcer diagnosis. Polymerase chain reaction (PCR) was employed to assess the presence and mosaicism of the following H. pylori genes: cagA, vacA, iceA, and picB. RESULTS: We found no statistical differences in the presence of these genes in H. pylori isolates from gastric ulcer patients taking or not taking prescription NSAIDs or ASA. A 297-bp fragment of the cagA gene was detected in 96% of the isolates from the NSAID and ASA users and 100% from the non-NSAID users (p 5 1.0). A larger and more variable region of the cagA gene was detected more frequently among the isolates from non-NSAID users than those from NSAID users (p 5 0.05). Ninety-two percent of the isolates were identified as vacA genotype s1. The dominant vacA subtype was s1b, 76.9% and 65.4% in isolates from non-NSAID–taking or NSAID–taking patients, respectively (p 5 0.4). iceA1 genotype was not correlated with gastric ulcer as this allele was only detected in 17.3% of all isolates. CONCLUSIONS: No significant differences in the presence of the candidate virulence genes vacA, cagA, picB, or iceA were detected in isolates from gastric ulcer patients taking prescription NSAIDs or ASA, compared with those not taking these drugs, indicating that single gene presence does not allow discrimination of isolates that may be important in NSAID-induced ulcerogenesis. A variable region of the cagA gene was more frequently detected in isolates from patients not taking NSAIDs or ASA, suggesting that this gene may be modified by NSAID- or ASA-related factors or that certain strains may be selected for in patients taking

these medications. (Am J Gastroenterol 1999;94: 1502–1507. © 1999 by Am. Coll. of Gastroenterology)

INTRODUCTION It is well recognized that the most common causes of ulcer disease are Helicobacter pylori infection and use of nonsteroidal antiinflammatory drugs (NSAIDs), including aspirin (1). Previous studies reported that approximately 50% of patients with NSAID-induced ulcers are also H. pylori infected (2– 4). However, it remains unclear whether H. pylori infection and NSAID ingestion are synergistic, additive, or independent risk factors for ulcer pathogenesis, or whether H. pylori infection may actually provide some protection against the ulceration caused by NSAIDs, possibly due to the microorganism’s ability to stimulate mucosal synthesis of prostaglandins (5–7). Better prognosis of NSAID-associated ulcers and increased effectiveness of omeprazole were detected in H. pylori-infected patients (6). No studies have been reported comparing the genotypic differences between H. pylori isolates obtained from gastric ulcer (GU) patients who are taking NSAIDs or ASA on a regular basis with those GU patients not taking NSAIDs or ASA. Studies have shown that strains of H. pylori differ in their association with gastrointestinal diseases and there is tremendous genetic diversity in the H. pylori species population (8, 9). Recently, several H. pylori genes, including cagA, vacA genotype s1a, and iceA1, have been suggested as conferring predisposition for the development of ulcer disease (10, 11). The cagA gene (cytotoxin-associated gene) encodes a 120- to 140-K protein of unknown function and is the marker gene for the presence of the H. pylori pathogenicity island, which contains multiple virulence factors such as those conferring increased inflammation (12). It has been suggested that the cagA gene is more prevalent in H. pylori isolates from patients with duodenal ulcers than in those from symptomatic patients with histological gastritis but without ulcer (13–15). The vacA gene encodes a protein inducing vacuolation of epithelial cell cultures (16). Within the vacA gene two variable segments have been found, the signal or s region

AJG – June, 1999

H. pylori Genotypes in Gastric Ulcer Patients

Table 1. Demographics of Patient Study Groups GU2ASA (n 5 26)

GU1ASA* (n 5 26)

55.69 37–73

59.15 44–79

12 11 2 1

13 8 5 0

25 1

26 0

Age, yr Mean Range Race Black White White Hispanic Asian Gender Men Women

* GU1ASA includes patients taking ASA or NSAIDs.

(s1: subtype s1a, s1b, or s2) and the middle or m region (m1, m2) (17–19). Specific mosaicism of these two regions of the vacA gene has been associated with pathogenicity of the bacterium. Strains with vacA type s1a/m1 are more frequently detected in isolates from patients with duodenal ulcer disease (18, 20, 21). However, Warburton and colleagues did not find any statistical correlation between clinical disease and vacA gene subtype (22). iceA (induced by contact with epithelium) gene is reportedly always present and has two allelic variants (iceA1, iceA2); iceA1 is speculated to be associated with peptic ulcers (23, 24). The picB gene is present in the pathogenicity island and is suggested to be essential for inflammation (25). In this study we evaluated H. pylori isolates from gastric ulcer patients to determine whether the genotype of the infecting isolate could be correlated with the use or nonuse of NSAIDs or ASA. Patients using NSAIDs or ASA were studied as one group because the mechanism of ulcerogenesis of these drugs is thought to be similar, via cyclooxygenase inhibition.

MATERIALS AND METHODS Patient Population H. pylori isolates were obtained from 52 patients with gastric ulcer disease at the Veterans Affairs Medical Center in Houston, Texas. Twenty-six patients were taking prescription ASA or NSAIDs at the time of their ulcer presentation (GU1ASA), whereas 26 patients were not taking prescription ASA/NSAIDs (GU2ASA). Demographic information on these patients is summarized in Table 1. Ten patients were taking ASA alone whereas 16 patients were using various NSAIDs. The commonest NSAIDs were naproxen and ibuprofen. All patients were taking these medications for .6 months. The commonest indication for NSAID use was degenerative joint disease, whereas the commonest indication for ASA use was coronary artery disease. None of the patients had rheumatoid arthritis, and no patient was taking immunosuppressive medication or disease-modifying antirheumatic drugs.

1503

Culture Conditions H. pylori isolates were cultured from gastric biopsies and identified based on typical cell morphology, Gram stain, and positive reactions for catalase, urease, and oxidase activity. The isolates were cultured at 37°C on brain heart infusion (BHI) agar plates supplemented with 7% horse blood under 12% CO2 high humidity. DNA Extraction Chromosomal DNA was extracted from cell lysate and purified using phenol and chloroform and overnight ethanol precipitation, as described previously (26). The DNA concentration was measured using the minifluorometer and the integrity determined using 0.7% agarose gels stained with ethidium bromide. Polymerase Chain Reaction (PCR) Genomic DNA was amplified by PCR in an automated thermal cycler (Perkin Elmer, Norwalk, CT) for four different genes, including vacA, cagA, iceA, and picB. The PCR amplifications were performed in 25 ml containing 100 ng of genomic DNA as template, 50 pmol of each oligonucleotide (Table 2), 0.625 mmol/L of dNTPs (Pharmarcia, Piscataway, NJ), and 2 units of AmpliTaq DNA polymerase (Perkin-Elmer). For vacA, five different primer sets were used to assess the genotype of the s-region (s1, s1a, s1b, s2) and the m-region (m1 or m2) (17). All vacA amplifications were performed with 30 cycles of denaturation (94°C, 1 min), annealing (56°C, 2 min), extension (72°C, 3 min), and one final extension (72°C, 10 min). For cagA, two sets of primers were used to amplify a 297-bp region from the middle region (lcagA; nucleotide 1751–2048) (14) and an approximately 1.4-kb region further downstream (bcagA; nucleotides 2549 –3967) (27). The lcagA amplifications were performed with 30 cycles of denaturation (94°C, 1 min), annealing (60°C, 2 min), extension (72°C, 3 min), and one final extension (72°C, 5 min); for bcagA, with 30 cycles of denaturation (94°C, 1 min), annealing (58°C, 2 min), extension (72°C, 3 min), and one final extension (72°C, 10 min). For iceA PCR two sets of primers were used to assess for the allelic variants, iceA1 or iceA2 (24). Both iceA amplifications were performed with 40 cycles of denaturation (95°C, 30 s), annealing (50°C, 45 s), extension (72°C, 45 s), and one final extension (72°C, 10 min). For picB, PCR amplifications were performed with 30 cycles of denaturation (95°C, 1 min), annealing (46°C, 2 min), extension (72°C, 3 min), and one final extension (72°C, 5 min) (25). The H. pylori reference strains NCTC 11638 and ATCC 49503 were used as positive controls; PCR mix without DNA was used as negative control. Each reaction was performed at least twice. PCR products were resolved in 1.5% agarose gels, stained with ethidium bromide, and visualized under ultraviolet light. PCR was determined as positive if any amplification product was detected, compared with negative control.

1504

Li et al.

AJG – Vol. 94, No. 6, 1999

Table 2. PCR Primer Sets for Amplification of vacA, cagA, iceA, and picB Sequences Primer

Sequence (5'–3')

vacA s-region VA1-F VA1-R SS1-F SS3-F vacA m-region VA3-F VA3-R VA4-F VA4-R cagA lcagAD008 lcagAR008 bcagA1 bcagA2 iceA iceA1F iceA1R iceA2F iceA2R picB picB1 picB2

Product (size, bp)

ATGGAAATACAAACAACAC CTGCTTGAATGCGCCAAAC GTCAGCATCACACCGCAAC AGCGCCATACCGCAAGAG

s1 (259)/s2 (289) s1a (190) s1b (187)

GGTCAAAATGCGGTCATGG CCATTGGTACCTGTAGAAAC GGAGCCCCAGGAAACATTG CATAACTAGCGCCTTGCAC

m1 (290) m2 (352)

ATAATGCTAAATTAGACAACTTGAGCGA TTAGAATAATCAACAAACATCACGCCAT ACGCTCAGAATCTTAAAGGC GGAAACCACCTTTTGTATTAACG

lcagA (297) bcagA (;1418)

GTGTTTTTAACCAAAGTATC CTATAGCCASTYTCTTTGCA GTTGGGTATATCACAATTTAT TTRCCCTATTTTCTAGTAGGT

iceA1 (246) iceA2 (229 or 334)

TGTTTGGTTTCCCTG ACGCATTCCTTAACG

Statistics Data were analyzed using Fisher’s exact test (two-tailed).

RESULTS The genotypes of vacA and iceA genes and the presence of cagA and picB genes were determined using PCR in 52 clinical isolates of H. pylori obtained from GU patients. Among these patients, 26 were taking prescription ASA or NSAIDs at the time of their ulcer presentation (GU1ASA), whereas 26 were not taking prescription ASA or NSAIDs (GU2ASA). The PCR genotyping results are summarized in Tables 3 and 4 . The GU1ASA group was further divided into aspirin and NSAIDs groups and no statistical differences between the genotypes were detected (data not shown). Most of the isolates (48/52; 92%) were detected as vacA type s1. Only eight isolates possessed vacA type s1a: 4/26 (15.4%) in the GU2ASA group and 4/26 (15.4%) in the GU1ASA group (p 5 1.0). Seventy-one percent (37/52) of the isolates had vacA type s1b: 76.9% (20/26) in the GU2ASA group, and 65.4% (17/26) in the GU1ASA group (p 5 0.4). One isolate from each patient group had both s1a and s1b genotypes. One s1 isolate in the GU1ASA group could not be identified as either subtype s1a or s1b. In

picB (1200)

the midregion of the vacA gene, the majority of the isolates (75%) were typed as m1 subtype: 76.9% (20/26) in the GU2ASA group and 73.1% (19/26) in the GU1ASA group (p 5 1.0). Three isolates from GU1ASA patients and one isolate from a GU2ASA patient had positive amplification for both m1 and m2 PCR. lcagA PCR resulted in the expected 297-bp fragment from the middle region of the cagA gene in all but one isolate: 26/26 (100%) in the GU2ASA group and 25/26 (96.2%) in the GU1ASA group. The presence of the cagA gene indicated that all but one isolate possessed the cag pathogenicity island. However, using PCR with the second primer set for a cagA gene region (bcagA) further downstream, 45 isolates (45/52; 86.5%) were positive for the 1.4-kb cagA PCR amplicon. There was a significant difference in the presence of the 1.4-kb fragment between GU2ASA isolates (25/26; 96.2%) and GU1ASA isolates (20/25; 80%) (p 5 0.05). Size polymorphism was observed in the amplified cagA PCR products using the bcagA primer set. Polymerase chain reaction amplification with the two primer sets for the iceA gene showed that a minority (9/52; 17.3%) of the isolates were detected as iceA1 genotype: 19.2% (5/26) in the GU2ASA and 15.4% (4/26) in the GU1ASA groups. Two isolates in the GU1ASA group

Table 3. Mosaicism of the vacA Gene in the Signal and Mid gene Regions of the H. pylori Clinical Isolates

GU2ASA (n 5 26) GU1ASA (n 5 26) p

s1

s1a

s1b

s1a1s1b

Neither s1a nor s1b

s2

m1

m2

m11m2

25 23 0.6

4 4 1.0

20 17 0.4

1 1 1.0

0 1 1.0

1 3 0.6

20 19 1.0

5 4 1.0

1 3 0.6

AJG – June, 1999

H. pylori Genotypes in Gastric Ulcer Patients

1505

Table 4. cagA, picB, and iceA Allelic Variants in the H. pylori Clinical Isolates GU2ASA (n 5 26) GU1ASA (n 5 26) p

lcagA

bcagA

iceA1

iceA2

iceA11A2

iceA Negative

picB

26 25 1.0

25 20 0.05

5 4 1.0

19 18 1.0

0 2 0.3

2 2 1.0

26 24 0.3

were positive for both iceA1 and iceA2 PCR, whereas two isolates from each group were negative for both iceA1 and iceA2. picB was detected in 26/26 GU1ASA isolates (100%) and 24/26 GU2ASA isolates (92.3%). All picBpositive isolates were also cagA positive. However, one picB-negative isolate was cagA positive. Genetic profiles were constructed based on the combinations of the vacA, cagA, iceA, and picB genotypes. Five genetic profiles were classified among these clinical isolates (Table 5). The predominant genetic profile was vacAs1/ cagA1/iceA2/picB1: 80.8% (21/26) in the GU2ASA group and 69.2% (18/26) in the GU1ASA group (p 5 0.36). The second most common genetic profile was vacAs1/cagA1/ iceA1/picB1: 15.4% (4/26) in the GU2ASA group and 19.2% (5/26) in the GU1ASA group (p 5 1.0). Three rare genetic profiles included vacAs2/cagA1/iceA1/picB1 (1/26 in GU2ASA, 1/26 in GU1ASA), vacAs2/cagA1/iceA2/ picB2 (1/26 in GU1ASA), and vacAs2/cagA2/iceA2/ picB2 (1/26 in GU1ASA).

DISCUSSION NSAIDs are one of the most commonly prescribed classes of drugs. In 1991, more than 70 million prescriptions for NSAIDs were filled in the United States. Risk of developing a serious, life-threatening ulcer complication is an estimated 1.3–2% of a population of NSAID users per year of use, and may be as high as 4% (28). Approximately 50% of patients with NSAID-related ulcers also have H. pylori infection (2– 4). Although there is an extensive literature concerning ulcers in H. pylori infection and in NSAID users, there is almost no information about whether there is a clinically significant interaction between the two. Such an interaction could be synergistic or additive towards ulcer pathogenesis, or H. pylori may actually provide some protection against NSAID-related ulcers due to the microorganism’s ability to stimulate mucosal synthesis of prostaglandins (5–7). It is also possible that any of these interactions could take place, depending on the genetic make-up of H. pylori. H. pylori strains with certain genetic profiles might have a synergistic effect on NSAIDs to induce ulceration, but strains with

different genetic profiles might actually have a negative correlation with NSAID-induced ulceration. It is well known that there is tremendous genetic diversity in the H. pylori population and some genes confer enhanced pathogenicity to the bacterium. For example, the cagA gene is strongly associated with the development of peptic ulcer disease, but not with nonulcer dyspepsia (13–15). The genetic profiles of H. pylori in patients with NSAID-related ulcers have not been reported before our study. We examined genetic differences among H. pylori isolates from gastric ulcer patients with or without prescription NSAIDs or ASA at their ulcer diagnosis, using PCR genotyping for multiple candidate H. pylori risk factors such as vacA, cagA, iceA, and picB. No statistically significant differences were detected in the single gene presence or the combination of these genes (genetic profiles) among the two groups of H. pylori isolates. The predominant genetic profile for these clinical isolates was vacAs1/cagA1/iceA2/picB1. The cagA gene was detected in 100% of the GU2ASA isolates and 96% of the GU1ASA isolates when assessing the 297-bp region using the lcagA primer set, indicating high conservation of this gene in all strains. However, when assessing a further downstream region using the bcagA primer set, the cagA gene was detected in 96.3% of the GU2ASA isolates but in only 76.9% of the GU1ASA isolates (p 5 0.05). This indicates that the true presence of the cagA gene can be artificially underestimated due to targeting different regions of the cagA gene in PCR reactions. Similar observations were also made by Miehlke and coworkers that PCR with the lcagA primer set gave a higher detection rate (98.3%) of the cagA gene than the bcagA primer set (1.7%) in H. pylori isolates from Korea (27). Our data also suggest that there is more sequence variation in certain regions of the cagA gene (e.g., bcagA primer targeting region) in the GU1ASA isolates than in the GU2ASA isolates. The reason for this difference remains to be clarified. Possible explanations include that NSAIDs may modify or effect genetic variation in H. pylori strains or that certain strains may be selected for in the presence of NSAID-related factors, such as induction of prostaglandins. Numerous recent studies have focused on identifying H.

Table 5. Distribution of Genetic Profiles Among H. pylori Clinical Isolates Genetic Profiles 1

1

vacAs1/cagA /iceA1/picB vacAs1/cagA1/iceA2/picB1 vacAs2/cagA1/iceA1/picB1 vacAs2/cagA1/iceA2/picB2 vacAs2/cagA2/iceA2/picB2

GU2ASA (n 5 26)

GU1ASA (n 5 26)

p

4 (15.4%) 21 (80.8%) 1 (3.8%) 0 (0%) 0 (0%)

5 (19.2%) 18 (69.2%) 1 (3.8%) 1 (3.8%) 1 (3.8%)

1.0 0.36 1.0 1.0 1.0

1506

Li et al.

pylori microbial gene markers for different gastroduodenal diseases. In many studies vacA type s1/m1 was found to produce increased vacuolating toxin and to be associated with duodenal ulcer disease (18, 20, 24, 29). However, different findings have been reported in some other studies. For example, one histology study showed that there were no statistically significant links among vacA status and atrophy and intestinal metaplasia (22). In another study, no association was found between vacA type s1/m1 and clinical diseases among H. pylori isolates obtained from China, suggesting different H. pylori gene pools may exist in different geographical regions (30). In the present study, 92.3% of our isolates from gastric ulcer patients possessed vacA type s1 and 75% possessed vacA type m1, suggesting a possible link between vacA type s1/m1 and gastric ulcer disease. Early studies showed a strong correlation between vacA genotype s1a and duodenal ulcer disease, with .85% of the isolates from duodenal ulcer patients possessing s1a (18, 20). In contrast to these studies, only a small proportion (15.4%) of the isolates obtained from our gastric ulcer patients had vacA genotype s1a. The majority (71.1%) of the isolates had vacA genotype s1b. Similar vacA genotype patterns were also observed among H. pylori isolates from duodenal ulcer patients from the same geographic region (31). This suggests specific genetic strains may predominate in different geographic regions. iceA gene allelic type 1 (iceA1) has been suggested as a virulence factor associated with ulcer disease (23, 24). However, this allele was only detected in 17.3% of the isolates from our gastric ulcer patients, suggesting a lack of correlation between iceA1 and gastric ulcer disease in H. pylori isolates from our patient population. The picB gene is present within the cag pathogenicity island and associated with the cagA gene. In our study the picB gene was found in all but one cagA-positive isolate. In conclusion, after assessing the genotypes and the genetic profiles of the vacA, cagA, iceA, and picB genes of H. pylori isolates from gastric ulcer patients, we found no correlation between single gene characteristics and the use or nonuse of NSAIDs and aspirin. The only exception was the greater variability in a polymorphic region of the cagA gene in the GU1ASA population, compared with the GU2ASA population.

ACKNOWLEDGMENTS This work was supported by the National Institutes of Health/National Cancer Institute Grant 5R01CA67469-02 and by the Department of Veterans Affairs. We thank SunYoung Anderson, PA-C for her excellent clinical assistance. Reprint requests and correspondence: Mae F. Go, M.D., Veterans Affairs Medical Center (111D), 2002 Holcombe Blvd., Houston, TX 77030. Received Aug. 13, 1998; accepted Jan. 22, 1999.

AJG – Vol. 94, No. 6, 1999

REFERENCES 1. Soll AH. Peptic ulcer and its complications. In: Sleisenger M, Fordtran J, eds. Gastrointestinal and liver disease, 6th ed. Philadephia: WB Saunders, 1997:620 –78. 2. Kim JG, Graham DY. Helicobacter pylori infection and development of gastric or duodenal ulcer in arthritic patients receiving chronic NSAID therapy. Am J Gastroenterol 1994; 89:203–7. 3. Taha AS, Nakshabendi I, Lee FD, et al. Chemical gastritis and Helicobacter pylori related gastritis in patients receiving nonsteroidal anti-inflammatory drugs: Comparison and correlation with peptic ulceration. J Clin Pathol 1992;45:135–9. 4. Heresbach D, Raoul JL, Bretagne JF, et al. Helicobactor pylori: A risk and severity factor of non-steroidal anti-inflammatory drug induced gastropathy. Gut 1992;33:1608 –11. 5. Hudson N, Balsitis M, Filipowicz F, et al. Effect of Helicobacter pylori colonization on gastric mucosal eicosanoid synthesis in patients taking non-steroidal anti-inflammatory drugs. Gut 1993;34:748 –51. 6. Hawkey CJ, Karrasch JA, Szczepanski L, et al. Omeprazole compared with misoprostol for ulcers associated with nonsteroidal antiinflammatory drugs. N Eng J Med 1998;338:727– 34. 7. Malfertheiner P, Labenz J. Does Helicobacter pylori status affect nonsteroidal anti-inflammatory drug-associated gastroduodenal pathology? Am J Med 1998;104:35S– 40S. 8. Go MF, Kapur V, Graham DY, et al. Population genetic analysis of Helicobacter pylori by multilocus enzyme electrophoresis: Extensive allelic diversity and recombinational population structure. J Bacteriol 1996;178:3934 – 8. 9. Blaser MJ. Heterogeneity of Helicobacter pylori. Eur J Gastroenterol Hepatol 1997;9(suppl 1):53–7. 10. Blaser MJ. Not all Helicobacter pylori strains are created equal: Should all be eliminated? Lancet 1997;349:1020 –2. 11. Atherton JC. The clinical relevance of strain types of Helicobacter pylori. Gut 1997;40:701–3. 12. Censini S, Lange C, Xiang Z, et al. cag, a pathogenicity island of Helicobacter pylori, encodes type I-specific and diseaseassociated virulence factors. Proc Natl Acad Sci USA 1996; 93:14648 –53. 13. Owen RJ, Figura N, Moreno M. Biotypes and DNA fingerprints of cytotoxigenic Helicobacter pylori from patients with gastritis and peptic ulceration in Italy. Eur J Epidemiol 1992; 8:15–21. 14. Covacci A, Censini S, Bugnoli M, et al. Molecular characterization of the 128-kDa immunodominant antigen of Helicobacter pylori associated with cytotoxicity and duodenal ulcer. Proc Natl Acad Sci USA 1993;90:5791–5. 15. Crabtree JE, Taylor JD, Wyatt JI, et al. Mucosal IgA recognition of Helicobacter pylori 120 kDa protein, peptic ulceration, and gastric pathology. Lancet 1991;338:332–5. 16. Harris PR, Cover TL, Crowe DR, et al. Helicobacter pylori cytotoxin induces vacuolation of primary human mucosal epithelial cells. Infect Immun 1996;64:4867–71. 17. Atherton JC, Cao P, Peek PM, et al. Mosaicism in vacuolating cytotoxin alleles of Helicobacter pylori. J Biol Chem 1995; 270:17771–7. 18. Atherton JC, Peek RM Jr, Tham KT, et al. Clinical and pathological importance of heterogeneity in vacA, the vacuolating cytotoxin gene of Helicobacter pylori. Gastroenterology 1997;112:92–9. 19. Cover TL, Tummuru MKR, Cao P, et al. Divergence of genetic sequences for the vacuolating cytotoxin among Helicobacter pylori strains. J Biol Chem 1994;269:10566 –73. 20. Strobel S, Bereswill S, Balig P, et al. Identification and analysis of a new vacA genotype variant of Helicobacter pylori in

AJG – June, 1999

21.

22. 23.

24. 25.

26.

different patient groups in Germany. J Clin Microbiol 1998; 36:1285–9. van Doorn LJ, Figueiredo C, Rossau R, et al. Typing of Helicobacter pylori vacA gene and detection of cagA gene by PCR and reverse hybridization. J Clin Microbiol 1998;36: 1271– 6. Warburton VJ, Everett S, Mapstone NP, et al. Clinical and histological associations of cagA and vacA genotypes in Helicobacter pylori gastritis. J Clin Pathol 1998;51:55– 61. Peek RM, Thompson SA, Atherton JC, et al. Expression of a novel ulcer-associated Helicobacter pylori gene, iceA, following adherence to gastric epithelial cells. Gastroenterology 1996;110(suppl):A225 (abstract). van Doorn LJ, Figueiredo C, Sanna R, et al. Clinical relevance of the cagA, vacA, and iceA status of Helicobacter pylori. Gastroenterology 1998;115:58 – 66. Tummuru MKR, Sharma SA, Blaser MJ. Helicobacter pylori pic B, a homologue of the Bordetella pertussis toxin secretion protein, is required for induction of IL-8 in gastric epithelial cells. Mol Microbiol 1995;18:867–76. Go MF, Chan KY, Versalovic J, et al. Cluster analysis of

H. pylori Genotypes in Gastric Ulcer Patients

27.

28.

29.

30.

31.

1507

Helicobacter pylori genomic DNA fingerprints suggests gastroduodenal disease-specific associations. Scand J Gastroenterol 1995;30:640 – 6. Miehlke S, Kibler K, Kim JG, et al. Allelic variation in the cagA gene of Helicobacter pylori obtained from Korea compared to the United States. Am J Gastroenterol 1996;91: 1322–5. Graham DY. Nonsteroidal anti-inflammatory drugs, Helicobacter pylori, and ulcers: Where we stand. Am J Gastroenterol 1996;91:2080 – 6. Wang HJ, Kuo CH, Kuo CH, et al. Vacuolating toxin production in clinical isolates of Helicobacter pylori with different vacA genotypes. J Infect Dis 1998;178:207–12. Pan ZJ, Berg DE, van der Hulst RWM, et al. Prevalence of vacuolating cytotoxin production and distribution of distinct vacA alleles in Helicobacter pylori from China. J Infect Dis 1998;178:220 – 6. Go MF, Cissell L, Graham DY. Failure to confirm association of vacA gene mosaicism with duodenal ulcer disease. Scand J Gastroenterol 1998;33:132– 6.