Transglutaminases in Crohn's Disease Gut: First Published As 10.1136/Gut.37.5.690 on 1 November 1995

Home , Avidin

690 Gut 1995; 37: 690-695 Transglutaminases in Crohn's disease Gut: first published as 10.1136/gut.37.5.690 on 1 November 1995. Downloaded from G D'Argenio, L Biancone, V Cosenza, N Della Valle, F P D'Armiento, M Boirivant, F Pallone, G Mazzacca

Abstract organs,3 and a membrane bound epidermal Transglutaminases are a family of Ca- transglutaminase4 present in keratinocytes have dependent enzymes involved in various been described. As the final enzyme in the co- biological events. Circulating transglutam- agulation cascade, activated FXIII (FXIIIa) inase (factor XIIIa) is decreased in blood of catalyses the intermolecular cross linking of fib- patients with inflammatory bowel diseases. rine chains to each other and to other haemo- There is evidence that factor XIIIa and static proteins.5 The enzyme activity can be tissue type transglutaminase, present in detected in Ca-activated plasma (FXIIIa)6 and cell cytosol, bind to various proteins of the in serum after clot formation (serum transgluta- extracellular matrix. This study examined minase).7 FXIII circulates in blood as a hetero- the value of serum transglutaminase assay tetrameric zymogen (a2b2) composed of two in the treatment and follow up of enzymatically active a-subunits and two b-sub- Crohn's disease and then investigated the units as carrier proteins.8 Tissue transglutami- intestinal location of both forms of trans- nase participates in various processes such as glutaminases by immunohistochemistry programmed cell death,9 wound healing,10 and in normal and abnormal tissues. Serum cell growth and differentiation." 12 Evidence is transglutaminase activity was assayed in mounting that both tissue transglutaminase and 36 patients with active Crohn's disease FXIIIa bind to fibronectine and other extra- (CDAI>150). Eighteen patients were cellular proteins contributing to the wound studied prospectively from relapse into healing process.'0 13-15 We found serum trans- remission. A significant inverse correla- glutaminase activity reduced in various intesti- tion (p<0.001) was found between cir- nal disorders: coeliac disease with relation to the culating transglutaminase and Crohn's active and remission phases,'6 and widespread disease activity index; a correlation was intestinal malignancies (for example, intestinal also found between serum transglutami- lymphoma and a-chain disease).'7 Circulating nase and serum orosomucoid (p<0.01) and enzyme activities have also been found C reactive protein (p<0.01). Patients were decreased during the acute phase of inflam- http://gut.bmj.com/ prospectively studied until clinical remis- matory bowel disease and they were strongly sion showed improvement in both their related to the activity indices.67 Recently we CDAI score mean (SD) (230 (46) to 72 (34), showed in a rat model of chronic colitis that p

TABLE I Clinical and laboratory summary ofstudy group COen CO Crohn's patients Ileitis Ileocolitis/colitis E *-1

Patients (n) 36 15 21 Gut: first published as 10.1136/gut.37.5.690 on 1 November 1995. Downloaded from Male 16 9 13 ,0.75 a a Female 20 6 8 CODZC)- Age (mean (SD)) 40 40.7 (8-1) 38-3 (20.3) ' E 0.5 iL CDAI 233 (49) 228 (37) 236 (57) A_ Serum transglutaminase mean (SD) 0-62 (0-11) 0-72 (0-14) 0.50 (0.08) Packed cell volume mean (SD) (%) 37 (8) 38-8 (6.2) 37-5 (3.5) E 0.25 White blood count mean (SD) (mm3) 10-1 (3.2) 9 9 (2-1) 10-3 (4.2) Albumin mean (SD) (g/d) 3-7 (0 50) 3-6 (0 40) 3-5 (0.7) (I() Orosomucoid mean (SD) (mg/dl) 134 (56) 147 (60) 103 (33) n Sedimentation rate mean (SD) (mm/h) 31 (19) 42 (25) 38 (28) 150 250 350 C reactive protein mean (SD) (mg/dl) 3.0 (3.9) 2-8 (3.5) 3-9 (4-1) CDAI Figure 2: Correlation between serum transglutaminase and CDAI score in active Crohn's disease patients (r=-062, the University 'La Sapienza' of Ronne were n=36, p<0 001). included in the study. The disease wras con- fined to the distal ileum in 15 patientss and to the colon with or without ileal involvement in Serum transglutaminase and haematological 21 patients. Four patients with Crohn's disease parameters were also evaluated in 18 patients of the ileum had bowel resection bec-ause of studied prospectively during the active phase failure of medical treatment. The surgical (CDAI> 150) and subsequent inactive phase procedure was ileocaecal resection wit}i side to (CDAI< 150) of the disease. side ileocolonic anastomosis in all patiients. In In the four patients undergoing surgery, no patients did histological assessmexnt show serum transglutaminase concentrations and inflammation at the resection margirns. The haematological parameters were evaluated Crohn's disease group included 20 wonnen and before and 15 and 60 days after bowel resec- 16 men with a mean age of 40 yearss (range tion. 19-72). The control group included 21 patients with Disease activity was assessed by the 1Crohn's disease other than inflammatory bowel disease Disease Activity Index (CDAI),19 supple- but also leading to debility (liver cirrhosis, mented by laboratory measurementis. The cardiovascular disorders, non-intestinal malig- CDAI score was evaluated in each paltient on nancies). Serum transglutaminase from 38 the day ofblood collection for transglut;aminase healthy subjects was also assayed. activity assay. Serum C reactive protzein was measured by an electroimmunodiffusicmn tech- nique,22 orosomucoid by nephelometrry23; the Transglutaminase activity

erythrocyte sedimentation rate was mteasured Venous blood samples were collected after an http://gut.bmj.com/ by the Westergren method.24 Blood pac-ked cell overnight fast and left at room temperature for volume and white blood cell counts w,ere per- two hours to avoid the influence of coagula- formed by a Coulter counter and serum albu- tion. Serum samples were stored at -20°C min concentrations by a colorimetric method.25 until the assay. Transglutaminase activity on serum was assayed using a modified method of Lorand et a126 27: 30 ,Il of the sample were

A p<0-os added to 45 ,l of reaction mixture containing on September 28, 2021 by guest. Protected copyright. 1 a final concentration of 0.25 mM of 14C- E putrescine (Amersham, UK), 50 mM of a)E I. dithiothreitol, 10 mM of CaCl2, and 40/o (w/v) C_ 0-75 _- dimethylcasein in TRIS-HCI buffer (50 mM) .u pH 9 0 with 0.1 % Triton X100 and incubated 0.5 _ in a shaking bath at 37°C for 20 minutes. C, 0 Twenty .lI were spotted onto 3MM Whatman round paper filters (2 cm) and

CO 0325 _- immediately plunged into 10% ice cold trichloroacetic acid for 15 minutes. Two cm were 0 consecutive 15 minutes washings lleitis lleocolitis;/colitis performed in 5% ice cold trichloroacetic acid followed by a brief washing in ethanol-acetone B NS --l (50% v/v) and then in acetone. The dried 350 paper filters were counted in 6 ml of Aquasure scintillant (Dupont-NEN). A similar procedure was adopted for blanks, standards, 300 _ and controls. Transglutaminase units were expressed as 1 mU= 1 nmol of putrescine into 0 250 _ acceptor protein at 37°C, pH 9.

200 Figure 1: Serum Western blotting transglutaminase (A) and 150 Tissue transglutaminase from guinea pig liver CDAI score (B) in Crohn's (Sigma, St Louis, MO) and purified FXIIIa disease with ileal or patients (gift from Behring, Marburg, Germany) were ileocolonic/colonic disease 100 _ involvement. lleitis lleocolitis/colitis run on SDS/polyacrylamide gels according to 692 D'Argenio, Biancone, Cosenza, Della Valle, D'Armiento, Boirivant, Pallone, Mazzacca

the method ofLaemmli,28 and than transferred Immunohistochemistry to nitrocellulose with Biorad transblot Routinely processed, formalin fixed, and apparatus. The nitrocellulose was blocked by paraffin wax embedded specimens were taken

incubation with 3% bovine serum albumine in from the ileum offour Crohn's disease patients Gut: first published as 10.1136/gut.37.5.690 on 1 November 1995. Downloaded from TTBS (50 mM TRIS, pH 7.9, 150 mM and from the colon of four Crohn's disease NaCI and 0-05% TWEEN 20). Primary anti- patients who were operated on during 1993 bodies, anti-tissue transglutaminase (gener- and were drawn from the files of the Institute ously given by Dr Vittorio Gentile 2nd of Pathologic Anatomy at the School of University ofNaples) or anti-FXIIIa (Behring) Medicine, Naples. Uninvolved bowel of were added and incubation continued patients undergoing surgery for carcinoma or overnight at room temperature. After the nitro- large polyps were used as a 'normal' control cellulose had been washed three times in tissue. Immunohistochemical examination was TTBS, the appropriate avidin conjugated sec- performed on normal and abnormal ileum and ondary antibody was added in TTBS for 60 colon using either anti-tissue transglutaminase minutes. After washing, immunoreactive or anti-FXIIIa antibodies. The ematoxilin- proteins were detected by development with eosine stained specimens were immunoprobed the ABC Vectastain kit, according to the with the antibodies and visualised using a per- manufacturer's directions. oxidase anti-peroxidase system,29 according to the manufacturer's directions.

p < 0.01 A Statistics Student's t test and linear regression were used 1 to perform statistical evaluations. Results, expressed as mean (SD), were considered statistically significant when p<0 05.

E 0.75 V- Results E

enCO, Patients CoCO Table I shows the clinical and haematological E 0.5 K characteristics of Crohn's disease patients. As cn shown, the mean CDAI score in the total (I) group of patients was 233 (range 169-366), while the mean (SD) serum transglutaminase http://gut.bmj.com/ activity was 0.62 (0 11). When patients were 0.25 e- grouped according to localisation ofthe disease, serum transglutaminase concentrations were found to be significantly higher in patients with ileitis (transglutaminase=0.72 (0 14) mU/ml) compared with patients with ileocolitis/colitis

0 on September 28, 2021 by guest. Protected copyright. Active disease Inactive disease (transglutaminase=0.50 (0.08); p<0 05). No differences were found, however, between p < 0.01 CDAI scores from patients with ileitis versus B ileocolitis/colitis (228 (37) v 236 (57) respec- tively) (Fig 1). When the whole group of patients was considered, a strong correlation (r= -0'60; p

TABLE II CDAI and serum transglutaminase (mU/ml) in patients who had a resection FXIIIa antibody - the positivity to FXIIIa antibody found in normal ileum (Fig 6 (A)) Resection Before 15 Days after 60 Days after and colon Fig 6 (C)) was similar to that CDAI 265 (57)t 64 (23)* 85 (31)* obtained with tissue transglutaminase anti- Serum transglutaminase 0-64 (0.08)§ 0-74 (0.10) 1-05 (0.09)4 body. Immunostaining in ileal (Fig 6 (B)) and Gut: first published as 10.1136/gut.37.5.690 on 1 November 1995. Downloaded from *P

enzyme values in the control and healthy volunteer groups fell to within the normal range (1.80 (0.57) mU/ml).7

Western blotting Antiserum raised against FXIIIa recognised purified FXIIIa and cross reacted with tissue transglutaminase on western blots as shown in Fig 4 (lane 1 and 2). Anti-tissue transglutaminase antiserum showed a specific immunoreac- tivity for tissue transglutaminase (lane 4), while did not recognise FXIIIa (lane 3).

Immunohistochemical studies Tissue transglutaminase antibody - the staining pattern showed that the enzyme is present in the basal region of the crypts in normal ileum (Fig 5 (A)) while strongly positive areas involving the whole crypt and to a lesser extent the extracellular matrix appeared in ileal Crohn's disease (Fig 5 (B)). In normal colon the positivity to tissue transglutaminase antibody was found along the crypt surface (Fig 5 (C)). In colonic Crohn's disease the staining pattern showed that the enzyme is mainly Figure SB localised to the extracellular matrix but also within the crypts (Fig 5 (D)). http://gut.bmj.com/

1 2 3 4

Transgl.taf 'a''' -O's 85 kO) on September 28, 2021 by guest. Protected copyright.

Figure 5C

Figure 4: Western blots ofpurifiedfactor XIIIa (lane 1 and 3) and tissue transglutaminase4_Ire (lane 2 and 4) after denaturing SDS-gel electrophoresis. (A) Blot immunoprobed with factor XIIIa antibody; (B) blot immunoprobed with tissue transglutaminase antibody. AJ. "|

Figure SD Figure 5: Immunohistochemistry ofspecimens from normal controls and Crohn 's disease: tissue transglutaminase staining using tissue transglutaminase antibody and peroxidase-antiperoxidase system; (A) normal ileum; (B) Crohn 's of the ileum; (C) normal colon; (D) Crohn 's of Figure 5A the colon. 694 D'Argenio, Biancone, Cosenza, Della Valle, D'Armiento, Boinivant, Pallone, Mazzacca Gut: first published as 10.1136/gut.37.5.690 on 1 November 1995. Downloaded from

Figure 6A Figure 6C

Figure 6D Figure 6: Immunohistochemistry ofspecimensfrom normal controls and Crohn's disease: FXIIIa staining using FXIIIa antibody and peroxidase-antiperoxidase system; (A) normal ileum; (B) Crohn's of the ileum; (C) normal colon; (D) Crohn's of the colon.

The decreased serum transglutaminase activity led us to look for a possible enzyme require- ment in mucosal recovery. Besides FXIII, http://gut.bmj.com/ which can be delivered to the damaged tissue by the blood flow, tissue transglutaminase seems to be another eligible form of transglutaminase that can be found in the damaged mucosa because of cell lysis or accelerated programmed Figure 6B cell death (apoptosis) with a subsequent release of the enzyme in the extracellular region. To on September 28, 2021 by guest. Protected copyright. obtain a clearer insight into the mechanisms the same pattern of section stained with tissue through which circulating transglutaminase and transglutaminase antibody but FXIIIa anti- tissue transglutaminase are related to mucosal body produce the highest positivity in the status in Crohn's disease, we undertook an extracellular matrix. immunohistochemical study using specific antibodies. Antibody raised against FXIIIa was found to be useful for detecting both tissue Discussion transglutaminase and FXIIIa on western blots Our data confirm and extend previous studies while tissue transglutaminase antibody was able showing the presence of a correlation between to recognise only the protein derived from circulating transglutaminase values and tissue. These immunoreactivities allowed us to Crohn's disease activity as assessed by CDAI investigate the intestinal pattern of distribution score.7 30 A strong correlation was found of tissue transglutaminase and FXIIIa both in between serum transglutaminase values and normal and abnormal mucosa. The same inflammatory mediators as serum orosomu- immunostaining pattern was seen in normal coid and C reactive protein. When patients tissues using either FXIIIa or tissue transgluta- were followed up, serum transglutaminase minase antibodies. In Crohn's disease the two values increased during the quiescent phase antibodies showed a similar staining pattern in according to the improved CDAI. the crypts but when FXIIIa antibody was used Furthermore, patients with ileocolitis/colitis the highest positivity was detected in extra- showed serum transglutaminase values lower cellular matrix. These findings, for the first than the subgroup with ileitis, suggesting that time, show the presence of both types of trans- this parameter could be helpful not only in the glutaminases in extracellular matrix of damaged assessment of the severity of inflammation but tissues. Tissue transglutaminase activity has also in providing some information about the been seen in wound healing process in the cell extension/localisation of the disease. monolayer in vitrol and in mucosal repair in Transglutaminases in Crohn 's disease 695

the intestine,31 using proteins of the extracellu- 8 Ichinose A, Bottenus RE, Davie EW. Structure of transglutaminases. J Biol Chem 1990; 265: 13411-4. lar matrix such as fibronectin and collagen, 9 Fesus L, Thomazy V, Falus A. Induction and activation of which are suitable substrates for the enzyme.15 tissue transglutaminase during programmed cell death. FEBS Lett 1987; 244: 104-8. 32 33 On the other hand, it has been suggested 10 Upchurch HF, Conway E, Patterson MK Jr, Maxwell MD. that FXIIIa and plasma fibronectin may localise Localization of cellular transglutaminase on the extra- Gut: first published as 10.1136/gut.37.5.690 on 1 November 1995. Downloaded from cellular matrix after wounding: characteristics ofthe matrix to the extracellular matrix contributing to the bound the enzyme. J CellPhysiol 1991; 1249: 375-82. healing process.3' In 1989 Allan et a134 showed 11 Birckbichler PJ, Carter HA, Orr GR, Conway E, Patterson MK. E-(-y-glutamyl) lysine isopeptide bonds in normal that plasma fibronectin concentrations were low and virus transformed human fibroblast. Biochem Biophys in patients with extensive or severe Crohn's Res Commun 1978; 84: 233-7. 12 Multaugh MP, Mehta K, Johnsons J, Myers M, Juliano RL, disease while increased tissue deposition has Davies PJA. Induction of tissue transglutaminase in mouse been shown in experimental animal models in a peritoneal macrophages. JBiol Chem 1983; 258: 11074-81. 13 Barry E, Mosher F. Factor XIII cross-linking of fibronectin variety of circumstances.35 36 It is probable that at cellular matrix assembly styes. J Biol Chem 1988; 263: plasma transglutaminase (FXIII) in its active 10464-9. 14 Achyuthan KE, Mary A, Greenberg CS. The binding sites form FXIIIa is locally utilised for tissue repair of fibrin(ogen) for guinea pig liver transglutaminase are with a subsequent reduced circulating concen- similar to those of blood coagulation factor XIII: charac- terization ofthe binding ofliver transglutaminase to fibrin. tration. Recent studies6 also showed a signifi- Jf Biol Chem 1988; 263: 14296-301. cant decrease of FXIIIa in ulcerative colitis 15 Juprelle-Soret M,- Wattiaux-de Coninck S, Wattiaux R. Subcellular localization of transglutaminase: effects of suggesting that the enzymatically active fraction collagen. BiochemJ' 1988; 250: 421-7. plays a part in haemostatic events during the 16 D'Argenio G, Sorrentini I, Ciacci C, Spagnuolo S, Ventriglia R, De Chiara A, et al. Human serum transglut- active phase of the disease. We previously aminase and coeliac disease: correlation between serum showed that transglutaminase activity signifi- and mucosal activity in an experimental model ofrat small bowel enteropathy. Gut 1989; 30: 950-4. cantly decreases in serum being closely related 17 D'Argenio G, Sorrentini I, Ciacci C, Mazzacca G. Low serum to the severity of inflammation in experimental transglutaminase in patients with intestinal limphoma and a-chain disease. Am J Gastroenterol 1990; 85: 1654-5. colitis in rats.'8 In the same model, FXIII intra- 18 D'Argenio G, Sorrentini I, Cosenza V, Gatto A, Iovino P, venous treatment, improved the induced coli- D'Armiento FP, et al. Serum and tissue transglutaminase correlates with the severity of inflammation in induced tis.3' These findings show that circulating colitis in the rat. Scand J Gastroenterol 1992; 27: 111-4. transglutaminase values are related to the 19 Best WR, Beckten JM, Singleton JW, Kern F. Development of the Crohn's disease activity index. Gastroenterology presence of intestinal inflammatory lesions. 1976; 70: 439-44. Even if serum transglutaminase cannot be con- 20 Harvey RF, Bradshaw JM. A simple index of Crohn's disease activity. Lancet 1980; i: 514. sidered a specific test for the diagnosis ofinflam- 21 Modigliani R, Mary JY, Simon JF, Cortot A, Soule JC, matory bowel disease, our data suggest that Gendre JP, et al. Groupe D'etude therapeutique des affec- tions inflammatories digestives. Clinical, biological, and serum transglutaminase could represent a very endoscopic picture of attacks of Crohn's disease. easy test for monitoring the intestinal inflamma- Gastroenterology 1990; 98: 811-8. 22 Aimbender E, Cabatu EE, Guzman DM, Sweet AY. Serum tory process in Crohn's disease. Patients under- C-reactive protein and problems of newborn infants. Jf going bowel resection, showed an Pediatr 1982; 101: 438-40. improved 23 Bienvenu J, Sann L, Bienvenu F, Lahet C, Divry P, Cotte J. CDAI 15 days after surgery while serum trans- Laser nephelometry of orosomucoid in serum of newborn: glutaminase increased toward control values by reference intervals and relation to bacterial infections. Clin http://gut.bmj.com/ Chem 1981; 27: 721-6. 60 days. It is possible that the delayed normali- 24 Nelson DA. Basic metodology. In: Henry JB, ed. Clinical sation of serum transglutaminase when com- diagnosis and management by laboratory methods. Philadelphia: WB Saunders, 1979: 858-917. pared with CDAI scores could be related to a 25 Doumas BT, Watson WA, Biggs HG. Albumin standards slow reconstitution of the normal and the measurement of serum albumin with bromocresol transglutami- green. Clin ChimActa 1971; 31: 87-96. nase circulating pool or to the local need of the 26 Bruce SE, Bjarnason I, Peters TJ. Human jejunal transglut- enzyme for the mucosal morphofunctional aminase: demonstration of activity, emzyme kinetics and substrate with to specificity special relation gliadin and on September 28, 2021 by guest. Protected copyright. recovery after intestinal resection, or both.37 celiac disease. Clin Sci 1985; 68: 573-9. In conclusion our data further confirm the 27 Lorand L, Campbell-Wilkes LK, Cooperstein L. A filter paper assay for transamidating enzyme using radioactive usefulness of serum transglutaminase assay as amine substrate. Ann Biochem 1972; 50: 623-31. a marker of Crohn's disease activity, extend 28 Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage. Nature 1970; 227: the utility of serum transglutaminase assay to 680-5. the follow up of the disease, and underline the 29 Stermberger LA, Hardy PH, Cuculis JJ, Meyer HG. The unlabeled antibody-enzyme method of immuno-histo- role of different types of transglutaminases in chemistry. Preparation and properties of soluble the extracellular matrix assembly in the antigen-antibody complex (horseradish peroxidase- antihorseradish peroxidase) and its use in identification of damaged tissue. spirochetes. 7 Histochem Cytochem 1970; 18: 315-9. 30 Hudson M, Wakefield AJ, Hutton RA, Sankey EA, Dhillon The authors thank Dr H Metzner (Behring) for generously AP, More L, et al. Factor XIIIa subunit and Crohn's providing purified factorXIII and Dr V Gentile (2nd University of disease. Gut 1993; 34: 75-9. Naples) for kindly supplying the tissue transglutaminase antibody. 31 D'Argenio G, Cosenza V, Sorrentini I, De Ritis F, Gatto A, Delle Cave M, et al. Butyrate, mesalamine and factor XIII 1 Folk JE. Transglutaminases. Annu Rev Biochem 1980; 49: in experimental colitis in the rat: effects on transglutami- 517-31. nase activity. Gastroenterology 1994; 106: 399-404. 2 D'Argenio G, Sorrentini I, Cosenza V, De Ritis F, Gatto A, 32 Upchurch HF, Conway E, Patterson MK Jr, Maxwell MD. Mazzacca G. Transglutaminase and factor XIII in intesti- Cellular transglutaminase has affinity for extracellular nal diseases. Am Jr Gastroenterol 1991; 10: 1529-30. matrix. In Vitro Cell Dev Biol 1987; 23: 795-9. 3 Ikura K, Nasu T, Yokota Y, Sasakj R, Chiba H. Amino acid 33 Lorand L, Daily JE, Turner PM. Fibronectin as a carrier for sequence of guinea pig liver transglutaminase from its the transglutaminase from human erythrocytes. Proc Nad cDNA sequence. Biochemistry 1988; 27: 2898-905. Acad Sci USA 1988; 85: 1057-9. 4 Floyd EE, Jetten AM. Regulation of type I (epidermal) 34 Allan A, Wyke J, Allan RN, Morel P, Robinson M, Scott transglutaminase mRNA levels during squamous dif- DL, et al. Plasma fibronectin in Crohn's disease. Gut ferentiation: down regulation by retinoids. Mol Cell Biol 1989; 30: 627-33. 1989; 9: 4846-51. 35 Robbins AB, Doran JE, Reese AC, Mansberger AR. Cold 5 Folk JE, Finlayson JS. The e (y glutamyl) lysine cross-link insoluble globulin levels in operative trauma: serum deple- and the catalytic role of transglutaminases. Adv Protein tion, wound sequestration and biological activity: an experi- Chem 1977; 31: 1-133. mental and clinical study. Am Surg 1980; 46: 663-72. 6 Standnicki A, Kloczko J, Nowak A, Sierka E, Slwiniski Z. 36 Reese AC, Doran JEi, Callaway BD, Mansberger AR. Factor XIII subunits in relation to some other hemostatic Sequestration of fibronectin at the site of any injury. Adv parameters in ulcerative colitis. Am Jf Gastroenterol 1991; Shock Res 1982; 8: 119-27. 86: 690-3. 37 D'Argenio G, Ciacci C, Sorrentini I, Ventriglia R, 7 D'Argenio G, Giacci C, Sorrentini I, Iovino P, Gatto A, Spagnuolo 5, Matters D, et at. Transglutaminase changes Cosenza V, et al. Serum transglutaminase in inflammatory in intestinal mucosa after experimental small bowel resec- bowel disease. J Clin Gastroenterol 1990; 12: 400-4. tion in the rat. Clin Physiol Biochem 1992; 9: 74-7.