expression of interleukin 8 (il-8) and substance

7/29/2019 Expression of Interleukin 8 (IL-8) and Substance

1/6

Expression of interleukin 8 (IL-8) and substanceP in human chronic pancreatitis

P Di Sebastiano, F F di Mola, C Di Febbo, G Baccante, E Porreca, P Innocenti, H Friess,M W Bchler

Abstract

BackgroundChanges in substance P

content and a relationship between the

degree of perineural inflammation and

pain has been demonstrated in chronic

pancreatitis. Whether a relationship exists

between neural alteration and pancreatic

inflammation (neurogenic inflammation)

is not known.

AimsIn the present study we evaluated

gene expression of preprotachykinin A

(PPT-A), the gene encoding substance P,

and interleukin 8, a proinflammatory and

hyperalgesic mediator whose release is

co-regulated by substance P.PatientsPancreatic tissue specimens

obtained from 21 patients (16 male, five

female) with chronic pancreatitis and 18

healthy organ donors (nine male, nine

female) were analysed.

MethodsGene expression of PPT-A and

interleukin 8 was studied by northern blot

analysis. Respective proteins were local-

ised using immunohistochemistry.

ResultsNorthern blot analysis showed

that PTT-A mRNA expression levels were

present at comparable levels in normal

and chronic pancreatitis tissue samples.

In contrast, interleukin 8 mRNA was

expressed at very low levels in normalcontrols but was increased 41-fold

(p


2/6

from macrophages. Substance P, a proinflam-matory mediator mainly involved in acuteinflammatory responses, stimulates and trig-gers the release of other cytokines and growthfactors.13 13 Inasmuch as IL-8 release also gen-erates hyperalgesia by stimulation of postgan-glion sympathetic neurones,10 these observa-tions strongly suggest that neuropeptides andcytokines might be involved in the pathophysi-ology of inflammatory disorders and might

contribute to the ongoing inflammation anddestruction of the exocrine pancreas in CP.

To assess this hypothesis we used northernblot analysis and immunohistochemistry toevaluate expression of the preprotachykinin A(PPT-A) gene, which encodes substance P, andof IL-8 in CP and the normal human pancreas.

Materials and methodsPATIENTS AND TISSUE SAMPLING

CP tissue samples were obtained from 21patients (five female, 16 male; median age 43.6years; range 3449 years) who underwent pan-creatic resection. The aetiology of CP wasalcohol overconsumption in all patients, and

surgical procedures consisted of either a partialduodenopancreatectomy (Whipple operation,three patients) or a duodenum preserving pan-creatic head resection. Normal human pancre-atic tissue samples were obtained through anorgan donor programme from 18 previouslyhealthy individuals (nine female, nine male;median age 30.3 years; range 1756 years)from whom other organs were taken for trans-plantation and no recipients for the pancreaswere present. Histomorphological analysis ofCP graded disease severity as moderate tosevere in all patients. In all cases, freshlyremoved tissue samples were fixed in a parafor-maldehyde solution for 1224 hours, followedby paraYn embedding for histological analysis.

In addition, part of the tissue samples destinedfor RNA extraction was snap frozen in liquidnitrogen immediately on surgical removal andmaintained at 80C until use. The studieswere approved by the Human Subjects Com-mittee of the University of Bern, Switzerland.

PROBE SYNTHESIS

The PPT-A cRNA probe consisted of a 457base pair KspI/SacI fragment of human PPT-AcDNA corresponding to nucleotides202659.15 The IL-8 cDNA probe consisted ofa 220 base pair NdeI/XhoI fragment of humanIL-8 cDNA excised from pRSET-A plasmid(kind gift of Dr Sozzani, Istituto di Ricerche

Farmacologiche Mario Negri, Milano, Italy).To verify equivalent RNA loading and RNAtransfer in the northern blot analysis, a cDNAprobe for glyceraldehyde-3-phosphate dehy-drogenase (GAPDH) (ATCC, Rockville,Maryland, USA) was used. For northern blotanalysis, cDNA probes (IL-8 and GAPDH)were labelled with [alpha-32P]dCTP (Du PontInternational, Regensdorf, Switzerland) pre-pared by using a random primer labellingsystem (Boehringer-Mannheim, Mannheim,Germany). The PPT-A cRNA probe was radio-labelled with [alpha-32P]CTP (Du Pont Inter-national, Regensdorf, Switzerland).

IMMUNOHISTOCHEMISTRY

ParaYn embedded tissue sections 24 m thickwere subjected to immunostaining using thestreptavidin-peroxidase technique. Tissue sec-tions were submerged for 15 minutes in TrisbuVered saline (TBS) (10 mM Tris HCI,0.85% NaCI, pH 7.4) containing 0.1% (vol/vol) Triton X-100, and washed for five minutesin TBS solution, as previously reported.1619

Endogenous peroxidase activity was quenched

by incubating the slides in methanol and inmethanol/0.6% hydrogen peroxide, followedby washings in methanol and TBS containing0.1% bovine serum albumin. Following treat-ment with hyaluronidase (1 mg/ml in 100 mMsodium acetate, 0.85% NaCI), sections wereincubated for 30 minutes at 23C with 10%normal goat serum, and incubated overnight at4C with polyclonal antibodies against IL-8(Endogen,Inc., Woburn, Massachusetts, USA)(1:25 dilution) and substance P (Serotec, Kid-lington, Oxford, UK) (1:5000 dilution), or amonoclonal CD-68 antibody (DAKO Diag-nostics AG, Zrich, Switzerland) (1:200 dilu-tion) that labels human monocytes and macro-

phages. The antibodies were diluted in 5%normal goat serum. Bound antibody wasdetected with a biotinylated goat antirabbit orantimouse IgG secondary antibody and astreptavidin-peroxidase complex (Kirkegaardand Perry Laboratories, Inc., Gaithersburg,Maryland, USA), followed by incubation withdiaminobenzidine tetrahydrochloride (0.05%)(brown colour reaction) as substrate and coun-terstaining with Mayers haematoxylin.

To ensure antibody specificity, control slideswere incubated in the absence of primary anti-body or incubated with non-specific IgGantibody. In both cases no immunostaining wasdetected. All slides were analysed by two inde-pendent observers blinded to patient status,

followed by resolution of any diVerences byjoint review and consultation with a thirdobserver.

NORTHERN BLOT ANALYSIS

Total RNA was extracted by the guanidine iso-thiocyanate method, size fractionated on 1.2%agarose/1.8 M formaldehyde gels, and stainedwith ethidium bromide for verification of RNAintegrity and loading equivalency.1618 20 RNAwas electrotransferred onto nylon membranes(Gene Screen, Du Pont, Boston, Massachu-setts, USA) and cross linked by UV irradiation.The filters were then prehybridised, hybrid-ised, and washed under conditions appropriate

for specific antisense cRNA riboprobes(PPT-A) or cDNA probes (IL-8 and GAPDH).In the case of the antisense riboprobe, blots

were prehybridised overnight at 65C. Theblots were then hybridised for 18 hours at 68Cin the presence of 1106 cpm/ml of the [alpha-32P]CTP labelled antisense riboprobe, washedtwice at 65C in a solution containing 1SSPE(150 mM NaCI, 10 mM NaH2PO4, and 1 mMEDTA) and 0.5% SDS, and washed twice at65C in a solution containing 0.1SSPE and0.5% SDS.1618 20 In the case of the cDNAprobes, blots were prehybridised overnight at42C. Hybridisation was carried out at 42C

424 Di Sebastiano, di Mola, Di Febbo,et al

www.gutjnl.com
http://gut.bmj.com/


3/6

for 18 hours with the [alpha-32P]dCTP labelledcDNA probe (1106 cpm/ml for IL-8, 1105

cpm/ml for GAPDH) and then washed twice(50C) in 2SSC and three times at 55C in0.2SSC and 2% SDS.1618 20

Membranes were then exposed at 80C toFuji x ray films with intensifying screens for112 days. All membranes were rehybridisedwith the GAPDH cDNA probe to assessequivalent RNA loading. The autoradiogramswere scanned and peak areas were measuredfor relative mRNA levels (IL-8/GAPDH andPPT-A/GAPDH) in the samples tested. Valueswere stored in spreadsheet software producinga graphic representation of the signals.

RELATIONSHIP BETWEEN IL-8 mRNA EXPRESSION

AND HISTOPATHOLOGICAL ANALYSISIL-8 mRNA expression data obtained bynorthern blot analysis were correlated with his-topathological parameters. In this analysis theinflammation score, presence of ductal meta-plasia, and degree of fibrosis were included. Toassess these parameters in the CP samples,haematoxylin and eosin stained tissue sectionswere analysed. The inflammation score(graded 03), presence of ductal metaplasia(graded 03), and percentage of fibrosis(graded 0100%) were measured on fivediVerent tissue sections in each CP tissue sam-ple and the mean value calculated, as previ-ously reported.17 20

STATISTICAL ANALYSIS

Results are expressed as median and range orupper and lower quartiles. For statistical analy-sis the Mann-Whitney U test and Spearmansrank correlation test were used. Significancewas defined as p< 0.05.

ResultsNORTHERN BLOT ANALYSIS

PPT-A mRNA signals (transcript size 1.2 kb)were detectable in both the normal pancreasand CP tissue samples (fig 1). Densitometricanalysis of the northern blot signals revealed no

significant diVerences between the normal andCP tissue samples (p>0.05) (fig 2). IL-8mRNA expression (transcript size 1.8 kb) wasonly weakly visible on the original autoradio-graphs in the normal pancreatic tissue samples(fig 1). In contrast, 86% of the CP tissue sam-ples exhibited markedly enhanced IL-8 mRNAlevels (fig 1). Densitometric analysis of thenorthern blots indicated that IL-8 mRNA lev-els were 41-fold increased (p


4/6

staining was detected in the remaining acinarcells of the exocrine parenchyma, in metaplas-tic ductal cells, and in areas where the atrophicacinar cells appeared to dediVerentiate intotubular and duct-like structures (fig 3B, C,arrows). Moreover, consecutive CP tissue sec-tions were stained with both IL-8 (fig 3H) andCD-68 antibodies (fig 3G), a specific marker ofhuman monocytes and macrophages, thusindicating that most of the IL-8 immunoposi-

tive cells were also CD-68 immunoreactive (fig3G, H, arrowheads).

CORRELATION BETWEEN NORTHERN BLOT

RESULTS AND HISTOMORPHOLOGICAL ANALYSIS

CP tissue specimens were evaluated to deter-mine if there exists a relationship betweeninflammation score, presence of ductal meta-plasia or degree of fibrosis, and IL-8 mRNAexpression levels (table 1). There was a signifi-

Figure 3 Immunohistochemical analysis. Interleukin 8 (IL-8) immunohistochemical analysis in the nor mal pancreas (A)and in chronic pancreatitis (CP) (BD, F, H). In the normal pancreas (A),no IL-8 immunostaining was detectable. Incontrast, in CP tissue sections (BD, F, H), moderate to intense IL-8 immunoreactivity was found, mainly in theinflammatory cells (D,F, arrows) surrounding enlarged pancreatic nerves. In some cases those nerves were substance Pimmunoreactive (E).Panels (G) and (H) (arrowheads) show staining of consecutive tissue sections of CP for CD-68 (G)and IL-8 (H). The inserts in (G) and (H) show that IL-8 immunoreactive cells (H) are most probably also stainedCD-68 positive (G). In addition, in CP tissue samples, moderate to strong IL-8 immunoreactivity was also found inmetaplastic ductal cells (B,C, arrows). n=nerve; d=duct. (A) IL-8 immunostaining;(B D,F, H) IL-8 immunostaining(patients Nos 14,16, see table 1); (E) substance P immunostaining (patient No 14); (G) CD-68 immunostaining (patient

No 16). Original magnification 200.


www.gutjnl.com
http://gut.bmj.com/


5/6

cant positive relationship between the presenceof ductal metaplasia and IL-8 mRNA expres-sion levels (r=0.795; p


6/6

2 Besson JM, Chaouch A. Peripheral and spinal mechanismsof nociception. Physiol Rev 1987;67:67186.

3 Gates TS, Zimmerman RP, Mantyh CR, et al. Calcitoningene-related peptide-receptor binding sites in the gastro-intestinal tract. Neuroscience 1989;31:75770.

4 Holzer P. Local eVector functions of capsaicin-sensitive sen-sory nerve endings: involvement of tachykinins, calcitoningene-related peptide and other neuropeptides. Neuroscience1988;24:73968.

5 Maggi CA, Meli A. The sensory-eVerent function ofcapsaicin-sensitive sensory neurons. Gen Pharmacol 1988;19:111.

6 Mantyh PW, Catton MD, Boehmer CG, et al. Receptors forsensory neuropeptides in human inflammatory diseases:implications for the eVector role of sensory neurons.

Peptides 1989;10:62745.7 Fink T, Di Sebastiano P, Bchler MW, et al. Growth-

associated protein-43 and protein gene product 9.5 inner-vation in human pancreas: changes in chronic pancreatitis.Neuroscience 1994;63:24966.

8 Di Sebastiano P, Fink T, Weihe E, et al. Immune cellinfiltration and growth associated protein-43 correlate withpain in chronic pancreatitis. Gastroenterology 1997;112:164855.

9 Bchler M, Weihe E, Friess H, et al. Changes in peptidergicinnervation in chronic pancreatitis. Pancreas 1992;7:18392.

10 Lotz M, Vaughan JH, Carson DA. EVects of neuropeptideson production of inflammatory cytokines by human mono-cytes. Science 1988;241:121821.

11 Walsh JH, Dockray GJ. Gut peptides:Biochemistry and physi-ology. New York: Raven Press, 1994.

12 Kimball ES, Fischer MC. Potentiation of IL-1-inducedBALB/3T3 fibroblast proliferation by neuropeptides. JImmunol1988;141:42038.

13 Taub DD, Oppenheim JJ. Chemokines, inflammation andthe immune system. Ther Immunol1994;1:22946.

14 Oppenheim JJ, Zachariae COC, Mukaida N, et a l .

Properties of the novel proinflammatory supergene inter-crine cytokine family. Ann Rev Immunol1991;9:61748.15 Marchand JE, Zaccheo TS, Connelly CS, et al. Selective in

situ hybridization histochemical analysis of alternativelys pl ic ed m RN As e nc od ing b et a- a nd g am ma -preprotachykinins in rat central nervous system. Mol BrainRes 1993;17:8394.

16 di Mola FF, Friess H, Scheuren A, et al. Transforminggrowth factor-s and their signalling receptors are co-expressed in Crohns disease. Ann Surg1999;229:6775.

17 Friess H, Cantero D, Graber H, et al. Enhanced urokinaseplasminogen activation in chronic pancreatitis suggests arole in its pathogenesis. Gastroenterology 1997;113:90413.

18 Friess H, Yamanaka Y, Bchler M, et al . Increasedexpression of acidic and basic fibroblast growth factors inchronic pancreatitis. Am J Pathol1994;144:11728.

19 Bishop AE, Polak JM, Bryant MG, et al. Abnormalities ofvasoactive intestinal polypeptide-containing nerves inCrohns disease. Gastroenterology 1980;79:85360.

20 di Mola FF, Friess H, Martignoni ME, et al. Connective tis-sue growth factor is a regulator for fibrosis in humanchronic pancreatitis. Ann Surg1999;230:6371.

21 Goldin E, Karmeli F, Selinger Z, et al. Colonic substance Plevels are increased in ulcerative colitis and decreased inchronic severe constipation. Dig Dis Sci 1989;34:7547.

22 Di Sebastiano P, Weihe E, di Mola FF, et al. Neuroimmuneappendicitis. Lancet1999;354:4616.

23 Mantyh CR, Gates TS, Zimmerman RP, et al. Receptorbinding sites for substance P, but not substance K or neu-romedin K, are expressed in high concentrations by arteri-oles, venules, and lymph nodules in surgical specimensobtained from patients with ulcerative colitis and Crohndisease. Proc Natl Acad Science USA 1988;85:32359.

24 Bchler MW, Friess H,Baer HU, et al. Surgical treatment ofchronic pancreatitis: new standards. Dig Surg 1996;13:63158.

25 Mallet-Guy PA. Late and very late results of resections ofthe nervous system in the treatment of chronic relapsingpancreatitis. Am J Surg1983;145:2348.

26 Adams DH, Lloyd AR. Chemokines: leukocyte recruitmentand activation cytokines. Lancet1997;349:4905.

27 Baggiolini M, Walz A, Kunkel S. Neutrophil-activatingpeptide-1/interleukin 8, a novel cytokine that activates neu-trophils. J Clin Invest 1989;84:10459.

28 Larsen CG, Anderson AO, Appella E, et al. The neutrophil-activating protein (NAP-1) is also chemotactic for Tlymphocytes. Science 1989;243:14646.

29 Schroder JM, Mrowietz U, Morita E, et al. Purification andpartial biochemical characterization of human monocyte-derived, neutrophil-activating peptide that lacksinterleukin-1 activity. J Immunol1987;139:347483.

30 Schroder JM, Mrowietz U,Christophers E. Purification andpartial biologic characterization of human lymphocyte-derived peptide with potent neutrophil-stimulating activity.

J Immunol 1988;140:353440.31 Bazzoni F, Cassatella MA, Rossi F, et al. Phagocytosingneutrophils produce and release high amounts of theneutrophil-activating peptide 1/interleukin-8. J Exp Med1991;173:7714.

32 Standiford TJ, Kunkel SL, Basha MA, et al. Interleukin-8gene expression by a pulmonary epithelial cell line. Amodel for cytokine networks in the lung. J Clin Invest 1990;86:194553.

33 Strieter RM, Kunkel SL, Showell HJ, et al. Endothelial cellgene expression of a neutrophil chemotactic factor byTNF-, LPS, and IL-1. Science 1989;243:14679.

34 Strieter RM,Phan SH, Showell HJ,et al. Monokine-inducedneutrophil chemotactic factor gene expression in humanfibroblasts. J Biol Chem 1989;264:106216.

35 Serra MC, Calzetti F, Ceska M, et al. EVect of substance Pon superoxide anion and IL-8 production by humanPMNL. Immunology 1994;82:639.


www.gutjnl.com
http://gut.bmj.com/

expression of interleukin 8 (il-8) and substance

Documents