Helicobacter pylori cag pathogenicity island is associated with enhanced Interleukin-8 expression in human gastric mucosa

ALIMENTARYTRACT

OIGES7LlUER DIS 2000;32:458-67

Helicobacter pylori tag pathogenicity island is associated with enhanced Interleukin-8 expression in human gastric mucosa B. G. S. E. G. S. H. A. C.

Orsini Ciancio Censinil Surrenti Pellegrini Milani Herbst2 Amorosi3 Surrenti

Background. In vitro studies showed that Helicobacter pylori strains carrying the tag pathogen&y island are able to induce epithelial secretion of Interleukin-8. Aims. To evaluate the assessment of tag pathogenicity island and the expression of Interleukin-8 in the gastric mucosa of Helicobacter pylori-infected patients and correlate these data with the activity of gastritis and Helicobacter pylori density Methods. tag status was determined by polymerase chain reaction directly on gastric biopsies from 13 Helicobacter pylori+ patients with non-ulcer dyspepsia and 13 Helicobacter pylori+ with duodenal ulcer Interleukin-8 gene transcription and protein expression were analysed by in situ hybridization and immunofluorescence, respectively Gastritis activity and Helicobacter pylori density were also investigated. Results. tag was present in 20/26 of Helicobacter pylori+ patients: in 7/13 non-ulcer dyspepsia [53.8%) and in 13/13 duodenal ulcer patients (100%-l, (pc0. OS]. Interleukin-8 mRNA and protein expression in epithelial and inflammatory cells was higher in tag+ than in tag- patients [p
Digest

Thisstudy is .wppwted by 8 pnt from the Ministen, de#‘Llniver&tt e de48 Riaiwc8 G%%Mio8 and from Boahringar MarnWm, It&y. The Authors 8re petefbi to A. cwmi, Chirim Vkcckas, Siana, &e/y, for ewxwagam@r% s~~~~s 8nd discussiion.

458

Liver

Key words:

Dis 2000;32:458-67 tag

pathogenicity

island;

gastric

biopsies;

Helicobacter

py/ori;

Interleukin-8

Introduction Helicobacter pylori (H. pylori) infection is strongly associated with chronic gastritis ‘, peptic ulcer disease I, athropic gastritis *, gastric adenocarcinoma 3 4, and mucosa-associated lymphoid tissue (MALT) lymphoma s. Although the infection is widespread in the world’s population 6, only a subset of individuals develop severe clinical diseases. A possibility relies on strains with enhanced virulence also associated with host genetics and environmental factors 7 * . Of the H. pylori strains, specific viru-

6. Orsini et al.

lence factors have been identified as playing a role in the clinical outcome of H. pylori infection 9 lo. H. pylori strains expressing cagA (cytotoxin-associated gene A) cause, in the gastric mucosa, extensive inflammation that is associated with bacterial density which, in turn, has been found to be enhanced “. Several studies suggest that cagA+ infection is more frequently found in peptic ulcer l2 18, atrophic gastritis 19, and gastric adenocarcinoma *O. It has been shown that cagA is a part of tag, a pathogenicity island derived from horizontal transfer, inserted into the chromosoma1 glutamate racemase and encoding a type IV secretion system apparatus responsible for increased virulence 9*’ **. The structure of tag is not identical in all H. pylori strains and can be functionally split into two regions (cag1” and cagI1”) separated by a transposable element (IS605) or, in a minority of strains, by an intervening sequence, flanked by two IS605 elements ?’ 23. IS605 encodes for two putative transposases (tnpA and tnpB) and may also be present elsewhere in the chromosome of H. pylori, promoting tag fragmentation and generation of strains with reduced virulence 24. It has been shown in vitro that tag is involved in many cellular responses, like: the tyrosine phosphorylation of a putative host protein in gastric cells, recently demonstrated by Stein et al. 25 to be the protein CagA after tag-driven host cell translocation, the reorganization of host cell actin and the induction in gastric epithelial cell lines 9 *’ 26?’ of Interleukin-8 (IL-8), a potent mediator of the migration and activation of neutrophils 28, associated with mucosal damage 29. Exposure of gastric epithelial cell lines to H. pylori has been shown to cause the secretion of IL-8 and, furthermore, cugA+ strains induce significantly higher IL-8 levels than cagA- strains 303’. In addition, cugA isogenic mutants induce IL-8 to the same degree as wild strains 3’ 32. Furthermore, the synthesis of proinflammatory cytokine IL-8 is strongly induced in gastric epithelial cell lines *’ 26and in human gastric epithelial cell primary cultures 33 by infection with tug+ strains but not with tag- strains. Mutational analyses have revealed that many genes in tug (both in cug1” and cugI1” regions) are needed to elicit IL-8 synthesis in gastric epithelial cells 3 ?’ I6 34 and also that some genes (such as cagA and cugN in the downstream cug1” region) are not involved *’ 26. With the aid of isogenic mutants, it has been found that the ability of tag+ H. pylori strains to elicit IL-8 secretion from gastric epithelial cells in vitro is positively regulated at the transcriptional level by NFkB 3536, depends on protein tyrosine kinase activa-tion 34 and is modulated by tag encoded Agrobacterium virD4 homologue 37. Independently of which tag genes are implicated in this phenomenon, any mutation of these results in no IL-8 production, since this ability de-

pends on the secretion machinery being intact y 38. No in vivo data have been reported. The present study aimed to evaluate the assessment of the tug pathogenicity island and the expression of IL8 directly in gastric biopsy specimens from H. pyZori+ patients together with activity of gastritis, H. pylori density and presence of duodenal ulcer (DU).

Patients and methods Patients

A total of 26 consecutive H. pyZori+ patients, undergoing endoscopy for upper gastrointestinal tract symptoms at the Gastroenterology Unit, University of Florence, were enrolled in this study. Exclusion criteria were as follows: previous history of gastric surgery, active gastrointestinal bleeding, active systemic infection, malignancy, treatment with steroids or other immunomodulating drugs, active alcohol consumption, chronic intake of non-steroidal anti-inflammatory drugs (NSAIDs), recent (in the previous three months) treatment with eradicating therapy, antibiotics or omeprazole. The presence of ulcer was based on endoscopic examination of the stomach and duodenum, and the patient’s history was investigated for an earlier diagnosis of peptic ulcer. If neither the patient’s documented history nor endoscopic examination showed peptic ulcer disease, the patient was considered to have non-ulcer dyspepsia (NUD). Four healthy volunteers were also recruited as controls. Patients and controls were all inhabitants of Northern Italy. H. pylori infection was established by the simultaneous positivity for the rapid urease test (RUT) (CL0 test, Delta West PtyLtd, Bentley, Australia), histology, “C-Urea Breath Test (‘“C-UBT) (Helicobucter Test INFAI, Bochum, Germany) and polymerase chain reaction (PCR) for Urease (ureA). Routine upper gastrointestinal endoscopy was performed and biopsies were collected from the gastric antrum and the corpus. Paraffin embedded sections were stained with modified Giemsa stain 39 to detect H. pylori and with haematoxylin and eosin (H & E) to grade the severity of gastritis. The activity of inflammation was analysed in accordance with the updated Sydney system and graded on a scale of O-3 40. Additional antrum specimens were immediately frozen in liquid nitrogen and maintained at -70°C until PCR for ureA and a panel of tug genes. Frozen sections were used for in situ hybridization (ISH) for IL-8 mRNA, immunofluorescence staining (IF) for IL-8 protein expression and Urease or CagA for H. pylori density, respectively. Informed consent was obtained from all those taking part in the study which was approved by the Local Ethics Committee. 459

In viva f/. pylon’ tag and Interleohin-8

DNA isolation from gastric antrum biopsy specimens: sample processing Gastric antrum biopsy specimens were ground with a sterile micropestle into ice-cold STE buffer (0.1 M NaCl, 10 mM Tiis-HCl pH=8, 1 mM EDTA pH=8) and DNA extraction was performed with lysozime, proteinase K, and phenol/chloroform, as previously described for DNA isolation from H. pylori bacterial colonies 41. Polymerase chain reaction (PCR) PCRs were performed in a volume of 100 ul containing 10 mM Tris-HCl (pH=8.3), 50 mM KCL, 2.5 mM MgCl*, 10 mM dNTPs (Pharmacia Biotech, Piscataway, USA), 2.5 U of AmpliTaq DNA polymerase (PerkinElmer, Norwalk, USA), 30 pmol of primers, BSA 0.1 mg/ml (Boehringer Mannheim Laboratories, Germany) and 5-10 ul of test sample as DNA template. PCRs were performed in a Gene Amp PCR System 2400 thermocycler (Perkin-Elmer) for ureA and tag genes. Two criteria were used to select tag genes to amplify: representative spacing alongside tag and the ability to induce IL-8 secretion in gastric epithelial cells. Three genes were selected in cag1” region (cagA, cagE, and cag1) and two were chosen in cagI1” (cagT and Agrobacterium virBl1 homologue). The presence of ZS605, the presence/absence and genetic organization of tag were also determined. PCR primers used in this study are listed in Table I. Gene amplification for ureA (411 bp) and cagA (298 bp) genes was carried out as previously described 41. PCRs for the other genes were run through 40 consecutive cycles, as for cagA, and the annealing temperature was 58°C for cagE (223 bp) and cagI(255 bp); 55°C for cagT (201 bp), virB 11 homologue (184 bp), IS605 (374 bp) and cagII”ZIS605 (23 1 bp). The cagII”lcagI” amplification was done with 35 cycles of 30 set of denaturation at 94°C 20 set of annealing at 55°C and 2 min of elongation at 72°C. Since tag can be disrupted by one or more IS605 and, in some strains, by an intervening sequence of variable length *l, amplimers of 600, 2502 bp or no products may be expected. In our in vivo condition, PCR for cagII”/cagI” provided only a 600 bp product, or no products, if an intervening sequence was present. In order to verify the absence of tag, PCR amplification (391 bp) for “empty site” (ES), the sequences flanking the tag region within the glutamate racemase gene, was performed and carried out as for cagA 4’. DNA extracts from G33, CCUG 17874, G21 bacterial strains and a test tube containing water in place of DNA were assayed in the appropriate PCR run as positive and negative controls, respectively. In situ hybridization for IL-8 mRNA Serial cryostat sections of gastric antrum biopsy specimens were used for IL-8 mRNA in situ hybridization 460

(ISH). The IL-8 cDNA was generated by oligo-(dT)primed reverse transcription of total cell RNA prepared from human placenta with subsequent amplification using specific oligodeoxyribonucleotide primers. After sequencing 42, a 264 bp DNA fragment was subcloned into the Sal 1 (Life Technologies, Paisley, PA, USA) site of a pGEM1 plasmid (Promega Biotech, Heidelberg, Germany). After linearization of the plasmid with either Pstl or BamHl (Life Technologies), T7 or SP6 RNA polymerase (Promega) was used to obtain run-off transcripts of the antisense (complementary to mRNA) or sense (anticomplementary to mRNA, negative control) strands, respectively. Transcription and labelling of RNA probes were performed with 60 ,uCi of 35S-uridine-5’-(thio)-triphosphate (1250 uCi/mmol) (New England Nuclear, Boston, MA, USA). The specific activity routinely obtained was around 1.3 x lo9 cpm/ug. Prehybridization, hybridization, removal of non-specific bound probe by RNAse A (Boehringer Mannheim Laboratories) digestion and further washing procedures were performed for positive and negative strand RNA probes on frozen sections of gastric mucosa, as previously described 43. The autoradiographic signal for IL-8 mRNA (assessed by two of the authors, one of which was unaware of the diagnosis) was analysed on 3 serial sections by a computerized video image system Leica Quantimet Q5OOMC (Leica Cambridge Ltd, Cambridge, England). A total of 6 fields for each section, 3 on superficial epithelial cells facing the gastric lumen and 3 on non-epithelial cells below the surface epitheliurn in the foveolar region in the lamina propria, were observed under a bright field microscope equipped with a 40x lens. The specific signal was acquired by a CCD video camera connected to the microscope. The threshold of specific detection was automatically calibrated on control sections hybridized with the sense probe. The mean&D values of the number of pixels corresponding to the specific IL-8 mRNA signal was counted separately on epithelial and non-epithelial cells. An arbitrary and calibrated grid of 0.01 mm* tissue area was used for the estimation of number of pixels for all the analyses. Zmmunojluorescence staining for IL-8 and urease Serial frozen sections of each gastric antrum specimen were analysed by IF. Briefly, after preincubation with normal goat serum (Sigma), sections were incubated simultaneously with primary mouse monoclonal antibody against hIL-8 (IgG 1K subclass) (Genzyme, Cambridge, MA, USA) and polyclonal rabbit anti-Urease. After washing, sections were incubated with FITC-conjugated goat anti-mouse IgG (Fab specific) (Sigma) together with TRITC-conjugated goat antirabbit IgG (whole molecule) (Sigma). Semiquantitative evaluation of IL-8 antigen expression was performed by two independent observers using a fluores-

B. Orsini et al.

lkbk I. Polymarase chain reaction primers used in this study. Primer

Nucbotide coordinatf?s

rrnll HPUI HPU2

(411 bp, nt. 360-7701”

4

[2B8 bp, nt. 1751-20481b

z 1223 bp, nt. 16697-16B19P TCAAGCCJATGAACCJ~G GTCllCTCJGTfTGGATlTAAG [255 bp, nt. 10652-l OB061c AG~~A~G~TGA~TAGC TAGCAAJGTl7CJTG!ZTGJCGT 1201 bp, nt. 845-10451e

4 c@-1 cegJ-2

Jl-GCATGGCTATGATGTGAMG CJGCCTCAlTTGAJAACCTGCA

e&m ORFI I-1 ORFI 1-2

AACCAT’ITGATGJGAGAGACAG GTl”fCATCAJTA4~TAACAG

I6655 IS-F IS-R

AAGJGCAAATGATCTATAJCGC GJTATACACGACCCTAGAACAG

o~n6ssos ORG-F ORG-R

lTGllTCAA~~TcR;4TATATC G~CGATC~GGCGTATATAG

Awg#rl~” E64T HR6

GATAACAGCDXATAATCTIXAT G~GATG~~~G~~TAGAT

ht. 1755-l 7781” ht. 6136-61131”

E@W##flizIWV EMPl EMP2

ACA~~T~T~C~TG~~~GCC ACACGCATTCCCTAAAGJGGAAlTTCATGCGA

[nt. 34660-34621 If fnt. 23662-235311”

1184 bp, nt. 11136-113191d

(374 bp, nt. 3C10-673?e

(231 bp, nt. 2133-23631c

aAccession NI X17079, bA~sshn no. X70039, GAccesskinffo. U60176, dAccsssio~ no. AEWS, BA~ssh no. usO177, fAcws~~n no. ACQWc18. * Thegmti~ a&wniWw of Gagcanbe d&w@edby onear mu@& 66U5 insertions.An &wvenhg ~c~uenee of wM&? Ien@ L4’7LJKbh CGOG178741 can also be prwent &snwW by sn event of ixomhgw rscomWiwlrl. pMdw& of 600. 25‘02 bp or no p&&s at sll n-mybe sxpsctd by Pars [ref. 211. “PCR smplifK;stn from capstrahs [primws EYWF7and EW2 flanking csgl gsnerstes 8 unilbnn bsnd of 391 bp.

cence microscope equipped with a 40x lens with FITC filter (3 fields on superficial epithelial and 3 on non-epithelial cells of the foveolar region in the lamina propria for each serial section). Results were expressed semiquantitatively and scored on a scale of O-3. A standardized tissue area was considered for the estimation of IF staining for all examined sections. Computerized image analysis was used to measure the number of anti-Urease-positive H. pylori microorganisms (3 fields near superficial and foveolar epithelial cells for each serial section with TRITC filter). The signal values corresponding to H. pylori immunoreactivity were expressed as mean + SD of pixel units determined on an arbitrary tissue area for each section.

Statistical analysis Age of patients, IL-8 mRNA signal expression and bacterial density values were expressed as the mean (M) with the standard deviation (SD) and analysed by Student’s t-test (two tailed) for independent variables. Chi-square analysis was used to examine the presence of tag in the analysed population, to compare mean score values of IL-8 immunoreaction and the activity of gastritis between tag+ and tag- infection. Linear regression analysis was utilized to correlate the IL-8 signal value with the activity of gastritis and H. pylori density. The inter-observer variability for the computer image analysis and semiquantitative detection of IL-8 immunofluorescence 461

In Go L pyhri

tag and Interleukin-8

staining was also determined using the paired t-test (number of observations = 3). A p value of co.05 was considered to be significant.

Results Polyrnerase chain reaction for ureA and tag genes in gastric antrum biopsies At endoscopy, 13 out of 26 H. pyZori+ patients (50%) had no evidence of peptic ulcer and were classified as patients with NUD while the other 13 patients showed DU. In all H. pyZori+ patients, the histological examination of gastric antrum mucosa specimens revealed chronic active gastritis. There was no significant difference between the age of NUD and DU patients (M&SD: 42.2+10.00 and 5 1.85k14.52, respectively). Four healthy age-matched volunteers were recruited as controls. They were H.pylori- and showed no macroscopic or histological abnormalities. ureA and the presence/absence of tag with its organization were analysed by PCR directly in gastric antrum biopsy specimens. We amplified the genes cagA, cagE, cag1, cagT, virB11 homologue and IS605, cagII”lIS60.5 and cagII”lcagI” regions to assess the presence and organization of tag (Fig. 1). No amplification for ureA and tag genes was observed in control subjects, while all H. pyZori+ patients resulted positive for ureA (data not shown). Regarding the presence of tag, independently from its organization, 20126 (76.9%) Hp+ patients resulted tag+: in particular, only 7/13 NUD (53.8%) and all DU patients 1303 (100%) were tag+ (~~0.05). The genes cagA, cagE, cag1, cagT and virB 11 homologue were simultaneously present in all tag+ patients. A detailed scheme of tag characterization results in the study population is shown in Table II. cag1” and cagI1” were fused in lo/20 tag+ patients, particularly in 3/7 NUD patients and in 7/13 DU patients (p=NS). No amplification for cagA, cagE, cag1, cagT, virB 11 homologue, IS605 or cagII”/IS605, cagII”/cagI” was obtained in tag- patients, which, in turn, resulted positive for ES (data not shown).

I

Cl 1. Exampteof de&Man aSPGRproductsfor tag H pyrariinfection in gaatricrsitrum bto~sy~peahans. ~~~t~ was obtained from originalF+cWuidfilms for 1% and 1.6% agarasf3get electraphoresisand ethidiumbromide staining.A: Lane 1 corrtwtns Ma 174 Ma&l DNAend km 2 amplikstian of 6CiUbp ka# fused to cagtl”l.Et Lane1 cor&sinsmufeix& weight markerDNA100 bp; product for mgA=298 bp): Lane3 k&=223 Lane2 ~emplification bpl; Lane4 k&G!!55 bpl; Lane 5 fcagT=X11bp1;Lane6 Ivk811 homotogue=lM bpl; Lane 7 (156Q5=374 bpl; Lane 8

rc&Wls&0~~31 bpl.

IL-8 (mRNA and protein) expression and H. pylori density on gastric antrum biopsy specimens In all H. pyZori+ patients, IL-8 mRNA expression was observed in gastric cells of the surface epithelium and pits (Fig. 2A, B). The intensity of the autoradiographic signal was higher in tag+ patients as compared with tag-, independently of the presence of DU (Fig. 2C, D). The quantitative evaluation of the signal by image analysis confirmed that IL-8 mRNA in gastric superficial epithelial cells was significantly higher in tag+ patients (257.68k31.88 pixel units) as compared R ‘+k tag- patients (165.45228.50 pixel units) (p
TWe II. ceg characterization in the study population.

-N. pts

cidr

[2NUD, 6 WI ‘l0[4NUD, 6 DUI 2 [I NUD, 1 DLtI 8

6

+ + +

md

@MT

+ + +

* * +

[6NUDl

1 IV pts : number

462

+ + +

*

of patients;

M/D: nun ulcer diyspapsia;

DU: duudaoal ulcer.

cwll + + +

B#lM + *

c#@61c@

kds +

+ +

8. Orsini et al.

cells was higher in tag+ than in tag- patients (22852237.54 vs 100.00~28.50 pixel units) (pcO.005) (Fig. 2C, D). In contrast, no specific hybridization sig-

nal was detected in controls either on epithelial or nonepithelial cells (data not shown). No significant differences for inter-observer variability were noted for the

In uivo /I. pyhri

tag end Interleukin-8

computer image analysis of ISH IL-8 mRNA. The pattern of distribution and intensity of staining for immunoreactive IL-S (Table III) reflected the data obtained by ISH. IL-8 staining in gastric superficial epithelial cells was stronger in tag+ patients when compared with tag- (p
with the pattern of distribution of IL-8 mRNA, a large number of positive non-epithelial cells of the lamina propria with prominent IL-8 immunostaining was evident in tag+ patients (Fig. 3A), while only a minority of positive non-epithelial cells with a weak IL-8 reaction was shown in tag- patients (pcO.005) (Fig. 3B).

No appreciable IL-8 antigen expression was documented in uninfected controls (data not shown). The density of H. pylori infection, analysed by immunostaining for Urease, was enhanced in tag+ (Fig. 3C) than in tag- infections (Fig. 3D) (152.91k35.29 vs 77.4Q10.61 pixel units) (pcO.005). In occasional sites, where H. pylori density was similar both in tag+ and tag- infections, IL-8 staining was always stronger in tag+ than in tag- patients on superficial and foveolar epithelial cells (data not shown). No significant differences in inter-observer variability were found for the analysis of bacterial density and IL-8 IF. The analysis of the relationship between H. pylori bacterial density and IL-8 expression showed that, in general, a significant correlation existed between the density of Urease-positive organisms and IL-8 protein expression on epithelial (t=O.749, p
Discussion Studies in the literature suggested that H. pylori is able to induce IL-8 gene expression and protein secretion from gastric cancer cell lines 31 and primary culture of human gastric epithelial cells 33. In addition, evidence has been provided showing that this ability in vitro is due to the presence of the fully functional tag 9 38. In the present study, we investigated tag status in vivo directly in gastric antrum biopsy specimens and IL-8 gene transcription and protein expression in H. pylori+ patients together with activity of gastritis, H. pylori density and presence of DU. We showed that IL-8 expression in the gastric mucosa of patients affected by H. pylori+ chronic active gastritis was associated with the presence of tag. We analysed the tag structure in clinical samples by amplification of eight sequences (four of these specific for the induction of IL-8 synthesis from gastric cell lines in vitro) of bacterial DNA directly extracted from gastric mucosal biopsies. In our

patients, despite some diversities in the genomic organization of tag, cagA was present virtually in all tag+ strains. Within the limits of the small number of patients analysed, these data suggest that, in agreement with Censini et al. ” the presence of tag lacking cagA may be a rare occurrence in clinical settings. Therefore, the detection of cagA may be sufficient to identify the majority of tag+ strains. Nevertheless, Maeda et al. 44, by using different techniques, demonstrated, in a small percentage of Japanese H. pylori strains, that cagA gene does not necessarily indicate the presence of intact tag and cannot invariably be used as a marker of tag-related virulence in Japanese strains. In agreement with a recent in vitro study 45, we documented that tag, independently of its rearrangement, was significantly associated with DU disease. With regard to the genetic microvariability of the H. pylori genome and unlike the recent observations of Slater et al. 46 in the United Kingdom, we detected, in our Northern Italian tag+ population, a large number of subjects with IS605 within the tag pathogenicity island. As previously shown for cagA+ strains and with a different methodological approach 47-49,we found that tag+ infection was associated with higher levels of IL8 expression in gastric mucosa, regardless of the presence of DU. The microscopic analysis showed that IL8 was expressed in a wide range of cell types, including epithelial cells of the superficial mucosa and pits and non-epithelial cells of the lamina propria which, based on morphological appearance, were likely to be largely represented by neutrophils and other inflammatory cell types. In our study, tag+ patients showed a higher IL-8 gene and protein expression on gastric cells of the surface of the epithelium and on inflammatory cells in the lamina propria compared with tag- pa465

In viva L pyhri

tag and Interleukin-8

tients. It has been reported that all H. pylori strains may attract neutrophils to the site of infection 5o and it has been suggested that their recruitment may be followed by the induction of IL-8 production from the surface epithelium 29 and, in addition, that, tag+ strains, unlike tag-, are also able to elicit IL-8 production directly from gastric epithelial cells in vitro 9 21263138. This supports our result of a higher IL-8 expression on epithelial cells of the surface and pits that we found only in tag+ patients. Moreover, the finding of high levels of IL-8 expression in non-epithelial cells of patients with tag+ infection, suggests that the recruitment of neutrophils may be particularly effective in the presence of tag+ strains. In addition, neutrophils and inflammatory cells infiltrating the gastric mucosa are well known primary targets of the chemotactic effect of IL-8 29. Therefore, the high levels of IL-8 expression in tag+ infection might contribute to the further amplification of the mucosal inflammatory response by an autocrinefparacrine mechanism 29.Accordingly, we found a positive correlation between the high levels of IL-8 expression in the gastric mucosa and the score of the activity of gastritis, which largely reflects the extent of the inflammatory infiltration. We observed that the density of H. pylori infection in gastric antrum was enhanced in tag+ patients with an intact tag, regardless of the presence of DU. As suggested for cugA+ infection ‘I, a high bacterial density might be responsible for the amplification of the inflammatory response. However, we also found that at sites where tug+ and cugH. pylori density were similar, the IL-8 expression on gastric cells of the surface and pits was greater in tug+ patients. This is consistent with the hypothesis that the presence of tug elicits a higher IL-8 production in vivo and also affects the total bacteria number, as recently suggested 9 51. Therefore, it is possible that tug+ strains, not only because of their higher density but also of the presence of tug in their chromosome, might be able to amplify the cellular response to the infection by directly acting on the gastric epithelium. Moreover, the presence of an intact tug within a population of heterogeneous bacteria may be relevant in determining the enhanced polymorph response to infection. In conclusion, our results indicate that the integrity of tug is a good indicator of gastric mucosa damage, paralleled by enhanced IL-8 expression.

References ’ Blaser MJ. Helicobacter pylori: microbiology of a slow bacterial infection. Trends Microbial 1993;1:255-60. ? Sipponen P. Natural course of Helicobacter pylori gastric infection. Ital J Gastroenterol Hepatol 1998;3O(Supp1.3):270-2. i Parsonnet J, Friedman GD, Vandersteed DP, Chang Y, Vogelman JH, Orentreich N, et al. Helicobacter pylori infection and the risk

466

of gastric cancer. N Engl J Med 1991;325: 1127-31. 4 McNamara D, O’Morain C. Helicobacter pylori and gastric cancer. Ital J Gastroenterol Hepatol 1998;32(Supp1.3):294-8. 5 Bayerdorffer E, Miehlke S, Neubauer A, Stolte M. Gastric MALT-lymphoma and Helicobacter pylori infection. Aliment Pharmacol Ther 1997;l l(Supp1 1):89-94. 6 Pounder RE, Ng D. The prevalence of Helicobacter pylori infection in different countries. Aliment Pharmacol Ther 1995;9(Suppl 2):33-9. 7 Go MF, Graham DY. How does Helicobacter pylori cause duodenal ulcer disease: The bug, the host, or both? J Gastroenterol Hepat01 1994;9(Suppl 1):8-12. ’ Malaty HM, Engstrand L, Pedersen NL, Graham DY. Helicobacter pylori infection: Genetic and environmental influences. A study of twins. Ann Intern Med 1994;120:982-6. ’ Covacci A, Telford JL, Del Giudice G, Parsonnet J, Rappuoli R. Helicobacter pylori, virulence, and genetic geography. Science 1999;284: 1328-33. lo Covacci A. Biological marker for diseases substained by Helicobacter pylori infection. Digest Liver Dis 2000;32:384-5. ‘I Atherton JC, Tham LT, Peek RM, Cover TL, Blaser MJ. Density of Helicobacter pylori infection in vivo as assessed by quantitative culture and histology. J Infect Dis 1996:174;552-6. ‘* Covacci A, Censini S, Bugnoli M, Petracca R, Burroni D, Macchia G, et al. Molecular characterization of the 128-kDa immuno-dominant antigen of Helicobacter pylori associated with cytotoxicity and duodenal ulcer. Proc Nat1 Acad Sci USA 1993;90:5791-5. I3 Xiang Z, Censini S, Bayeli PF, Telford JL, Figura N, Rappuoli R, et al. Analysis of expression of CagA and VacA virulence factors in 43 strains of Helicobacter pylori reveals that clinical isolates can be divided into major types and CagA is not necessary for expression of the vacuolating cytotoxin. Infect Immun 1995;63:94-8. I4 Blaser MJ. Role of vacA and the cagA locus of Helicobacter pylori in human disease. Aliment Pharmacol Ther 1996;lO (Suppl 1):73-g. I5 Xiang Z, Bugnoli M, Ponzetto A, Morgando A, Figura N, Covacci A, et al. Detection in an enzyme immunoassay of an immune response to a recombinant fragment of the I28 kDa protein (CagA) of Helicobacter pylori. Eur J Clin Microbial Infect Dis 1993;12:739-45. Ih Cover TL, Glupczynski Y, Lage AP, Burette A, Tummuru MKR, Perez-Perez GI, et al. Serologic detection of infection with CagA+ Helicobacter pylori strains. J Clin Microbial 1995;33: 1496-500. I7 Wee1 JF, van der Hulst RW, Gerrits Y, Roorda P, Feller M, Dankert J, et al. The interrelationship between cytotoxin-associated gene A, vacuolating cytotoxin, and Helicobacter pylori-related diseases. J Infect Dis 1996;173: 117 l-5. I8 Orsini B, Ciancio G, Surrenti E, Macil G, Biagini MR, Milani S, et al. Serologic detection of CagA positive Helicobacter pylori infection in a Northern Italian population: its association with peptic ulcer disease. Helicobacter 1998;3: 15-20. I’) Kuipers EJ, Perez-Perez GI, Meuwissen SG, Blaser MJ. Helicobacter pylori and atrophic gastritis: importance of the cagA status. J Nat1 Cancer Inst 1995;87:1777-80. 20 Parsonnet J, Friedman GD, Orentreich N, Vogelman H. Risk for gastric cancer in people with CagA positive or CagA negative Helicobacter pylori infection. Gut 1997;40:297-301. *’ Censini S, Lange C, Xiang Z, Crabtree JE, Ghiara P, Borodovsky M, et al. tag, a pathogenicity island of Helicobacter pylori, encodes type j-specific -and disease-associated virulence factors. Proc Nat1 Acad Sci USA 1996;93:14648-53. ” Akopyants NS, Clifton SW, Kersulyte D, Crabtree JE, Youree BE, Reece A, et al. Analyses of the tag pathogenicity island of Helicobacter pylori. Mol Microbial 1998;28:37-53.

‘3 Tomb JE White 0, Kervalage AR, Clayton RA, Suttun GG, Fleischmann RD, et al. The complete genome sequence of the gastric pathogen Helicobacter pylori. Nature 1997;388:539- 47. 24Covacci A, Falkow S, Berg DE, Rappuoli R. Did the inheritance of a pathogenicity island modify the virulence of Helicobacter pylori? Trend Microbial 1997;5:205-8. 25 Stein M, Rappuoli R, Covacci A. Tyrosine phosphorylation of the Helicobacter pylori CagA antigen after tag-driven host cell translocation. Proc Nat1 Acad Sci USA 2000;97: 1263-S. 26Tummuru MK, Sharma SA, Blaser MJ. Helicobacter pylori picB, 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. 27Segal ED, Lange C, Covacci A, Tompkins LS, Falkow S. Induction of host signal transduction pathways by Helicobacter pylori. Proc Nat1 Acad Sci USA 1997;94:7595-9. 28Crabtree JE. Inflammatory responses to Helicobacter pylori infection. Stand J Gastroenterol (Suppl) 1996;215:3-10. 29 Crabtree JE, Spencer J. Immunologic aspects of Helicobacter pylori infection and malignant transformation of B cells. Semin Gastrointest Dis 1996;7: 1- 11. 30Crabtree JE, Farmery SM, Lindley LJD, Figura N, Peichl P, Tompkins DS. CagA cytotoxic strains of Helicobacter pylori and Interleukin-8 in gastric epithelial cell lines. J Clin Path01 1994;47:945-50. ” Sharma SA, Tummuru MK, Miller GG, Blaser MJ. Interleukin-8 response of gastric epithelial cell lines to Helicobacter pylori stimulation in vitro. Infect Immun 1995;63:1681-7. X2Crabtree JE, Xiang Z, Lindley IJD, Tompkins DS, Rappuoli R, Covacci A. Induction of interleukin-8 secretion from gastric epithelial cells by cagA negative isogenic mutant of Helicobacter pylori. J Clin Path01 1995;48:967-9. ” Ogura K, Takahashi M, Maeda S, Ikenoue T, Kanai F, Yoshida H, et al. Interleukin-8 production in primary cultures of human gastric epithelial cells induced by Helicobacter pylori. Dig Dis Sci 1998;43:2738-43. 34 Li SD, Kersulyte D, Lindley IJD, Neelam B, Berg D, Crabtree JE. Multiple genes in the left half of the tag pathogenicity island of Helicobacter pylori are required for tyrosine kinase-dependent transcription of interleukin-8 in gastric epithelial cells. Infect Immun 1999;67:3893- 9. j5 Glocker E, Lange C, Covacci A, Bereswill S, Kist M, Pahl HL. Proteins encoded by the tag pathogenicity island of Helicobacter pylon are required for NF-rB activation. Infect Immun 1998;66:2346-8. ” Sharma SA, Tummuru MK, Blaser MJ, Kerr LD. Activation of IL-8 gene expression by Helicobacter pylori is regulated by transcription factor nuclear factor-kappa B in gastric epithelial cells. J Immunol 1998;160:2401-7. ” Crabtree JE, Kersulyte D, Li SD, Lindley IJD, Berg D. Modulation of Helicobacter pylori induced Interleukin-8 synthesis in

gastric epithelial cells mediated by tag PA1 encoded virD4 homologue. J Clin Path01 1999;52:653-7. 38 Covacci A, et al. (personal communication). 3y Gray SF, Wyatt JI, Rathbone BJ. Simplified techniques for identifying Campylobacter pyloridis (letter). J Clin Path01 1986:39;1279. 4o Dixon MF, Genta RM, Yardley JH, Correa P. Classification and grading of gastritis. The updated Sydney System. International Workshop on the histopathology of gastritis, Houston 1994. Am J Surg Path01 1996;20:1161-81. 41 Covacci A, Rappuoli R. PCR amplification of gene sequences from H. pylori strains. In: Lee A, Megraud F, editors. Helicobacter pylori: techniques for clinical diagnosis and basic research. Philadelphia: Saunders, 1996. p. 224-33. 42 Mukaida N Shiroo M, Matsushima K. Genomic structure of the human monocyte-derived neutrophil chemotactic factor IL-S. J Immunol 1989;143:1366-71. 43 Orsini B, Calabro A, Milani S, Grappone C, Herbst H, Surrenti C. Localization of epidermal growth factor/transforming growth factor-a receptor in the human gastric mucosa. Virchows Archiv A Path01 Anat 1993;423:57-63. 44 Maeda S, Yoshida H, Ikenoue T, Ogura K, Kanai F, Kato N, et al. Structure of tag pathogenicity island in Japanese Helicobacter pylori isolates. Gut 1999;44:336-41. 45 Jenks PJ, Megraud F, Labigne A. Clinical outcome after infection with Helicobacter pylori does not appear to be reliably predicted by the presence of any of the genes of the tag pathogenicity island. Gut 1998;43:752-8. 46 Slater E, Owen RJ, Williams M, Pounder RE. Conservation of the tag pathogenicity island of Helicobacter pylori: Associations with vacuolating cytotoxin allele and IS605 diversity. Gastroenterology 2000;117:1308-15. 47 Peek RM Jr, Miller GG, Tham KT, Perez-Perez GI, Zhao X, Atherton JC, et al. Heightened inflammatory response and cytokine expression in vivo to cagA+ Helicobacter pylori strains. Lab Invest 1995;73:760-70. 48 Yamaoka Y, Kita M, Kodama T, Sawai N, Tanahashi T, Kashima K, et al. Chemokines in the gastric mucosa in Helicobacter pylori infection. Gut 1998;42:609- 17. 4y Shimoyama T, Everett SM, Dixon MF, Axon AT, Crabtree JE. Chemokine mRNA expression in gastric mucosa is associated with Helicobacter pylori cagA positivity and severity of gastritis. J Clin Path01 1998;5 1:765-70. 50 Crabtree JE. Gastric mucosal inflammatory responses to Helicobacter pylori. Aliment Pharmacol Ther 1996;lO(Suppl 1):29-37. ” Hamlet A, Thoreson AC, Nilsson 0, Svennerholm AM, Olbe L. Duodenal Helicobacter pylori infection differs in cagA genotype between asymptomatic subjects and patients with duodenal ulcers. Gastroenterology 1999;116:259-68.

467