- Email: [email protected]
Effects of gliadin-derived peptides from bread and durum wheats on in vitro cultures of human cell lines. Implications for coeliac disease pathogenesis
331
Toxicology Letters, 16 (1983) 331-338 Elsevier
EFFECTS OF GLLADIN-DERIVED PEPTIDES FROM BREAD AND DURUM WHEATS ON IN VITRO CULTURES OF HUMAN CELL LINES. IMPLICATIONS FOR COELIAC DISEASE PATHOGENESIS (Gliadin peptides; cell culture; cytotoxity; wheat)
coeliac disease; durum wheat; bread
ERMELINDA ROCCA, ANNALAURA PAGANUZZI and FLAVIA ZUCCO**
STAMMATI*, FRANC0 ZAMPAGLIONI
Department of Comparative Toxicology and Ecotoxicology, Istituto Superiore di Sanitci, Viale Regina Elena 299, 00161 Rome, and **Istituto di Tecnologie Biomediche - C.N.R., Via Morgagni 30/E - 00161 Rome (Italy) (Received and accepted November 18th, 1982)
SUMMARY digest, obtained from bread (hexaploid) wheat gliadins under experimental A peptic-tryptic-cotazyme conditions mimicking in vivo protein digestion, was found to reduce in vitro viability of human embryo (MRC-5) and tumor cell (Hep-2) lines. Time of onset and extent of cytotoxic effects were largely dependent on initial peptide concentrations in the culture medium. The presence of 2% fetal calf serum was capable of delaying, but not of preventing, the onset of cytotoxic effects only in MRC-5 cultures. A digest obtained from durum (tetraploid) wheat gliadins and tested under idenpeptic-tryptic-cotazime tical conditions did not show any cytotoxic activity on MRC-5 and Hep-2 cell lines. These results indicate that cell systems are useful to investigate pathogenetic mechanisms of coeliac disease (gluten-dependent enteropathy).
INTRODUCTION
Cell culture systems are a promising new tool in toxicological investigations, particularly to clarify action mechanisms of toxic agents [ 1,2]. In this preliminary study we suggest the use of human cell cultures to explore the aetiological mechanisms involved in coeliac disease. This gluten-dependent enteropathy is a malabsorption syndrome which derives from a combination of genetic and environmental factors. When present in the diet of susceptible subjects (approx. 0.5% of the general population), wheat and some other cereals induce subtotal atrophy of intestinal mucosa. Characterization of wheat components, toxic in coeliac disease and other forms of wheat intolerance [3] is difficult owing to the lack of suitable in vitro methods * To whom correspondence 0378-4274/83/$03.00
is to be sent at the first address.
0 Elsevier Science Publishers B.V.
332
for toxicity testing. Falchuk et al. [4, 51 proposed organ culture of human small intestine biopsies as an in vitro model of coeliac disease. A peptic-tryptic digest of a crude gliadin preparation from bread (hexaploid) wheat has been found to be toxic for cultured small intestine musosa from patients with active coeliac disease [4, 6-S]. Cornell and Townley [9] have reported on ion-exchange fractionation into 10 major peptide fractions of the peptic-tryptic-cotazym@ digest of a crude gliadin preparation from bread wheat. The fraction eluted with 0.02 M phosphate buffer (coded as fraction 9) was shown to be extremely toxic for cultured mucosa from patients with active coeliac disease [lo], as well as for intestinal cells and other human embryo and tumor cell types cultured in vitro [ 11, 121. Under the test conditions used, this cytotoxicity was inhibited by rabbit and calf serum [l 1, 121. Rusconi et al. [ 131 were unable to show any variation in biochemical and morphological parameters of skin-derived fibroblasts from normal or coeliac donors incubated in the presence of 10% calf serum with a peptic-tryptic digest of bread wheat gliadin obtained by Frazer’s method [14]. In this paper, the in vitro effects of peptic-tryptic-cotazym@ digest prepared from durum (tetraploid) and bread wheat gliadins, were tested on two human cell lines by the method of Auricchio et al. [15]. One of these was the Hep-2 cell line already tested by Hudson et al. [ 1l] and the other was a diploid fibroblast line from embryo lung. The selection of these two cell lines is due, among other reasons, to the fact that gliadin digest seems to exert a direct non-specific cytotoxicity on various types of immature or malignant cells [l 11. Moreover, Weiser and Douglas [16] claim that gliadin would behave like lectin Con-A by recognising immature and transformed cells. These authors, while stressing the possible importance of cell-membrane glycoproteins for the aetiology of coeliac disease, pointed, out that incomplete glycoproteins are present on membranes of mitotically undifferentiated crypt cells and intestinal malignant cells [ 171. The aim of this work is also to explore to what extent human cell lines could be used to reproduce previous in vitro and in vivo results for coeliac disease, and to verify whether cytotoxic effects could be prevented by adding calf serum to the culture medium. MATERIALS
AND METHODS
Gliadin fractions were extracted with 70% aqueous ethanol from pure bread (var. Mentana) and durum (var. Azizia) wheat kernels, after extraction of globulin and albumin fractions, by the method described in Fig. 1. Peptic-tryptic-cotazym digests were prepared from bread and durum wheat gliadins, following the three step procedure of Bronstein et al. [18], as reported by De Ritis et al. [3]. All the enzymes used were previously purified by gel filtration on Sephadex G-100 [3]. 100 g protein fraction was digested in 1.1 of 0.2 N HCl (pH 1.8) with 2 g purified pepsin at 37°C
333
whole wheat flour (100 g)
supernatant
extraction with 1 I of 0.04 M Na2HP04 containing 1.8 M (NH& pH 7; centrifugation for 15 min at 16 OOOxg
4-1
dyalisis against water at 4°C for 48 h; centrifugation as above discard the supernatant
I
sediment extraction as above with 0.04 M Na2HP04 containing 0.4 M (NH&S04, pH 7; centrifugation as above
precipitate (globulin) supernatant dialysis as above and liophylisation sediment
I
albumin
extraction as above with 300 ml ethanol: water (70: 30 v/v); centrifugation as above
s”;d;;;;~~
I
sediment (discard)
Fig. 1. Differential extraction of globulin, albumin and gliadin fractions from whole wheat flour.
for 2 h. The peptic digest was further digested by addition of 2 g purified trypsin after pH adjustment to 8.0 with 2 N NaOH. The reaction mixture was vigorously stirred at 37’C for 4 h at pH 8.0. Then, the peptic-tryptic digest was treated with 2 g of purified cotazym@ and mechanically stirred for 2 h at pH 8.0. During the entire digestion procedure the pH was checked periodically and, when needed, adjusted with HCl or NAOH. At the end of the whole digestion procedure, the digest was submitted to gel filtration and the peptide fractions eluted after cytochrome c had been collected. Peptides with M, values < 2000 were removed by ultrafiltration on UM-2 Amicon membranes. These freeze-dried enzyme-free peptide pools, with M, values in the range 2000-12 000 have been coded as PTC protein digest. Hep-2 is an epithelial cell line derived from human carcinoma of the larynx, MRC-5 is a diploid fibroblast line from human embryo lung. Cell monolayers were cultured in falcon plastic dishes with Eagle Basal Medium supplemented with Earle salts and 10% fetal bovine serum (Flow Laboratories) at 37°C in 5% COzair. Cytotoxicity tests were performed by plating about lo4 and 3 x lo3 (0.05 ml) of MRC-5 and Hep-2 cells, respectively, in microtest falcon plates. After the cells had formed a confluent monolayer (approx. 48-72 h after plating), the incubation
334
medium was removed and substituted with the Eagle Basal Medium with or without 25% fetal calf serum and with or without the peptic-tryptic cotazym@ digest of gliadins from bread or durum wheat. The cells were exposed to a number of concentrations of peptides from both bread and durum wheat gliadins (100, 200, 500, 700, pg peptides/ml incubation medium for Hep-2 and 700, 1000, 1500 and 2000 ,ug peptides/ml incubation medium for MRCJ). All tests were run in triplicate. Observations were made every 24 h up to 7 days of culture. Toxicity scoring was based on the microscope appearances of cells and monolayers, usually at a magnification of x 100.
Fig. 2. Effect of peptic-tryptic-cotazim@ (PTC) digests from bread (hexaploid and durum (tetraploid) wheat gliadins on Hep-2 cell-line cultures in absence (A, B, C) or in presence (a, b, c) of 2% fetal calf serum. Time of culture: % h (A, a) control cultures; (B, b) bread wheat PTC digest, 700 fig/ml; (C, c) durum wheat PTC digest, 700 pg/ml.
335
RESULTS
Toxic effects were only seen after a 48-h culture of confluent monolayers of Hep-2 cells in the presence of a 700 pg concentration of peptic-tryptic-cotazym@ digest of bread wheat gliadin/ml of incubation medium (Fig. 2.). Observed cytotoxic effects consisted of morphological alterations in cell polygonal shape and the appearance of lysis areas in the monolayer. After 96 h of incubation, cell damage was clearly observed even at the lowest peptide concentration tested (100 hg/ml). Cell
Fig.
3. Effect
(tetraploid)
of
peptic-tryptic-cotazim@
wheat gliadins
of 2% fetal calf serum. digest,
2000 pg/ml;
on MRC-5
(PTC) digests from bread (hexaploid) and durum cell-line cultures in absence (A, B, C) or in presence (a, b, c)
Time of culture:
(C, c) durum
wheat
7 days.
(A, a) control
PTC digest,
2000 pg/ml.
cultures;
(B, b) bread
wheat
PTC
336
damage severity was seen to increase with the concentration of added bread wheat peptides for each incubation period. No significant difference in the time of onset and extent of cytotoxic effects was observed between cultures performed in the presence of bread wheat peptides with or without 2% fetal calf serum. The peptic-tryptic-cotazym Q digest of wheat bread gliadins was also found to be toxic for MRC-5 cell cultures (Fig. 3). In this case, the earliest onset of a cytotoxic effect was observed in the absence of serum after a 72-h incubation with a concentration of 2000 pg peptides/ml culture medium. Observed effects were similar to those mentioned above for the Hep-2 cell line with the exception of the appearance of a decreased MRC-5 cell layer density in the presence of bread wheat gliadinderived peptides. After longer incubation times (up to 7 days), toxic effects were also detectable at lower peptide concentrations (down to 500 pg/ml) and, for a given concentration, toxic effects were more severe than those observed after shorter incubations. The onset of cytotoxic effects was delayed (about 2 days) when fetal calf serum was also present in the culture medium. None of the above-reported effects was observed when Hep-2 or MRC-5 cells were incubated, with or without serum, under identical experimental conditions with the peptide mixture obtained from durum (tetraploid) wheat gliadin. DISCUSSION
The data reported show that the peptic-tryptic-cotazym@ digest of bread wheat gliadins reduces in vitro viability of human embryo and tumor cell lines. The time of onset and extent of toxic effects were dependent on initial peptide concentration. The presence of serum was capable of delaying the onset of cellular toxic effects in only one of the two cell lines tested. Consequently, it may be seen that these results only partially confirm the findings reported by Hudson et al. [l 11. We tend to believe that the Hudson et al. report [l l] on the capacity of serum to prevent cytotoxicity induction by a proteolytic digest of wheat gliadin was mainly due to the fact that they performed observation only after 24 h of incubation and not after longer periods of time. At present, we have no explanation for the transient protective effects of fetal calf serum. Furthermore, it would seem that Hep-2 epithelial cells are more sensitive to the toxic action of bread wheat gliadin peptides that MRC-5 fibroblasts cells. However, this may depend on the fact that, for any given peptide concentration, the number of exposed Hep-2 cells was smaller than that of MRC-5 cells. The cytotoxic effects of bread wheat gliadin peptides reported herein are related to the inhibition of in vitro development and morphogenesis of fetal rat intestine [3] and to the inhibition of the in vitro epithelial recovery of cultured specimens of flat intestinal mucosa from patients with active coeliac disease [3]. The peptic-tryptic-cotazym@ digest of durum wheat gliadins showed no cytotoxic activity on Hep-2 and MRC-5 cell lines. Moreover, durum wheat gliadin peptides did not affect the in vitro development and
331
morphogenesis of fetal rat intestine and were much less active than bread wheat gliadin peptides in preventing in vitro epithelial recovery of culture specimens of flat intestinal mucosa from active coeliac patients [3]. There is also evidence that bread wheat gliadin peptides do not exhibit any in vitro toxic activity on adult human fibroblast cells [13; preliminary unpublished results from our laboratory], on the jejunum from 21-day-old rat fetuses or newborn rats or on intestinal mucosa from normal individuals [3, 19, 201. All these data are consistent with the hypothesis that bread wheat gliadin peptides only interact with immature and/or relatively undifferentiated cells. Therefore, the presence of immature enterocytes, well known to occur on the surface of coeliac mucosa, could play a major role in making the intestinal mucosa of these susceptible individuals highly sensitive to bread wheat gliadin. A similar hypothesis was put forward by Douglas [16, 211, who showed that a carbohydrate-containing component of wheat gluten was capable of effective binding to membrane components of coeliac intestinal cells, and of only poor binding to normal intestinal cell membranes. Further investigations with different types of embryo and adult cells in the presence and absence of bread and durum wheat gliadin peptides might be useful in confirming this hypothesis. ACKNOWLEDGEMENTS
We wish to thank Prof. V. Silano for critical reading of the manuscript and for helpful discussions; Dr. M. De Vincenzi for supplying the gliadin PTC-digest. REFERENCES 1 R.M. Nardone, Toxicity testing in vitro, in G.H. Rothblat and V.J. Cristofalo (Eds.), Growth, Nutrition and Metabolism of Cells in Culture, Vol. 3, Academic Press, New York, 1977, pp. 478-495. 2 A. Paganuzzi Stammati, V. Silano and F. Zucco, Toxicology investigations with cell cultures systems, Toxicology, 20 (1981) 91-153. 3 G. De Ritis, P. Occorsio, S. Auricchio, F. Gramenzi, G. Morisi and V. Silano, Toxicity of wheat flour proteins and protein derived peptides for in vitro developing intestine from rat fetus, Pediat. Res., 13 (1979) pp. 1255-1261. 4 Z.M. Falchuk, R.L. Gebhard, C. Sessams and W. Strober, An in vitro model of gluten-sensitive enteropathy., J. Clin. Invest., 53 (1974), 487-500. 5 Z.M. Falchuk and A.J. Katz, Organ culture model of gluten-sensitive enteropathy, in B. McNicholl, C.F. MC Carthy and P.F. Fottrel (Eds.), Perspective in Coeliac Disease, MTP Press, Lancaster, 1978, pp. 65-72. 6 Z.M. Falchuk, R.L. Gebhard and W. Strober, The pathogenesis of sensitive enteropathy (coeliac sprue): organ culture studies, in W.J.T.M. Hekkens and A.S. Pena (Eds.), Coeliac Disease, Stenfert Kroese, Leiden, (1974), pp. 107-117. 7 J. Jos, G. Lenoir, G. De Ritis and J. Rey, In vitro culturing of biopsies from children, in W.J.T.M. Hekkens and A.S. Pena (Eds.), Coeliac Disease, Stenfert Kroese, Leiden, (1974), pp. 91-105. 8 J. Jos, G. Lenoir, G. De Ritis and J. Rey, In vitro pathogenetic studies of coeliac disease. Effects of protein digest on coeliac intestinal biopsy specimens maintained in culture for 48 hours, Stand. J. Gastroent. 10 (1975) 121-128.
338 9 H.J. Cornell and R.R.W. Townley, Investigation of possible intestinal peptidase deficiency in coeliac disease, Clin. Chim. Acta (1973) 113-125. 10 R.R.W. Townley, P.S. Bhathal, H.J. Cornell and J.D Mitchell, Toxicity of wheat gliadin fractions in coeliac disease, Lancet i (1973) 1363-1364. 11 D.A. Hudson, H.J. Cornell, D.R. Purdham and C.J. Rolles, Non-specific cytotoxicity of wheat gliadin components towards cultured human cells, Lancet i (1976) 339-341. 12 D.L.J. Freed and R.J. Cooper, Cytotoxicity of bread and soya protein in tissue culture, Lancet i (1977) 371-372. 13 R. Rusconi, M. Borgonovo, C. Bianchi Porro and P. Roggero, Ricerche sulla tossicita in vitro di una preparazione enzimatica di gliadina su fibroblasti di pazienti celiaci e normali, Minerva Ped., 30 (1978) 15-18. 14 A.C. Frazer, R.F. Fletcher, C.A.C. Ross, B. Shaw, H.C. Sammons and R. Schneider, Gluteninduced entheropathy. The effect of partially digested gluten, Lancet ii (1959) 252-255. 15 S. Auricchio, G. De Ritis, M. De Vincenzi, P. Occorsio and V. Silano, Effect of gliadin derived peptides from bread and durum wheats on small intestine culture from rat fetus and coeliac children, Pediat. Res., (1982) in press. 16 M.M. Weiser and A.P. Douglas, An alternative mechanism for gluten toxicity in coeliac disease, Lance& i (1976) 567-569. 17 M.M. Weiser, Intestinal epithelial cell surface membrane glycoprotein synthesis, II. Glycosyl transferases and endogenous acceptors of the undifferentiated cell surface membrane, J. Biol. Chem. 248 (1973) 2542-2548. 18 H.D. Bronstein, L.J. Haeffner and O.D. Kowlessar, Enzymatic digestion of gliadin: the effect of the resultant peptides in adult coeliac disease, Clin. Chim. Acta 14 (1966) 141-155. 19 Z.M. Falchuk, C. Sessoms and W. Strober, An in vitro model of gluten-sensitive enteropathy: evidence that gluten is not directly toxic to gastrointestinal epithelium, J. Clin. Invest. 51 (1972) 28a. 20 A.J. Katz, Z.M. Falchuk, W. Strober and H. Schwachman, Gluten-sensitive enteropathy. Inhibition by cortisol of the effect of gluten protein in vitro, N. Engl. J. Med. 295 (1976) 131-135. 21 A.P. Douglas, The binding of a glycopeptide component wheat gluten to intestinal mucosa of normal and coeliac human subjects, Clin. Chim Acta 73 (1976) 357-361.
Toxicology Letters, 16 (1983) 331-338 Elsevier
EFFECTS OF GLLADIN-DERIVED PEPTIDES FROM BREAD AND DURUM WHEATS ON IN VITRO CULTURES OF HUMAN CELL LINES. IMPLICATIONS FOR COELIAC DISEASE PATHOGENESIS (Gliadin peptides; cell culture; cytotoxity; wheat)
coeliac disease; durum wheat; bread
ERMELINDA ROCCA, ANNALAURA PAGANUZZI and FLAVIA ZUCCO**
STAMMATI*, FRANC0 ZAMPAGLIONI
Department of Comparative Toxicology and Ecotoxicology, Istituto Superiore di Sanitci, Viale Regina Elena 299, 00161 Rome, and **Istituto di Tecnologie Biomediche - C.N.R., Via Morgagni 30/E - 00161 Rome (Italy) (Received and accepted November 18th, 1982)
SUMMARY digest, obtained from bread (hexaploid) wheat gliadins under experimental A peptic-tryptic-cotazyme conditions mimicking in vivo protein digestion, was found to reduce in vitro viability of human embryo (MRC-5) and tumor cell (Hep-2) lines. Time of onset and extent of cytotoxic effects were largely dependent on initial peptide concentrations in the culture medium. The presence of 2% fetal calf serum was capable of delaying, but not of preventing, the onset of cytotoxic effects only in MRC-5 cultures. A digest obtained from durum (tetraploid) wheat gliadins and tested under idenpeptic-tryptic-cotazime tical conditions did not show any cytotoxic activity on MRC-5 and Hep-2 cell lines. These results indicate that cell systems are useful to investigate pathogenetic mechanisms of coeliac disease (gluten-dependent enteropathy).
INTRODUCTION
Cell culture systems are a promising new tool in toxicological investigations, particularly to clarify action mechanisms of toxic agents [ 1,2]. In this preliminary study we suggest the use of human cell cultures to explore the aetiological mechanisms involved in coeliac disease. This gluten-dependent enteropathy is a malabsorption syndrome which derives from a combination of genetic and environmental factors. When present in the diet of susceptible subjects (approx. 0.5% of the general population), wheat and some other cereals induce subtotal atrophy of intestinal mucosa. Characterization of wheat components, toxic in coeliac disease and other forms of wheat intolerance [3] is difficult owing to the lack of suitable in vitro methods * To whom correspondence 0378-4274/83/$03.00
is to be sent at the first address.
0 Elsevier Science Publishers B.V.
332
for toxicity testing. Falchuk et al. [4, 51 proposed organ culture of human small intestine biopsies as an in vitro model of coeliac disease. A peptic-tryptic digest of a crude gliadin preparation from bread (hexaploid) wheat has been found to be toxic for cultured small intestine musosa from patients with active coeliac disease [4, 6-S]. Cornell and Townley [9] have reported on ion-exchange fractionation into 10 major peptide fractions of the peptic-tryptic-cotazym@ digest of a crude gliadin preparation from bread wheat. The fraction eluted with 0.02 M phosphate buffer (coded as fraction 9) was shown to be extremely toxic for cultured mucosa from patients with active coeliac disease [lo], as well as for intestinal cells and other human embryo and tumor cell types cultured in vitro [ 11, 121. Under the test conditions used, this cytotoxicity was inhibited by rabbit and calf serum [l 1, 121. Rusconi et al. [ 131 were unable to show any variation in biochemical and morphological parameters of skin-derived fibroblasts from normal or coeliac donors incubated in the presence of 10% calf serum with a peptic-tryptic digest of bread wheat gliadin obtained by Frazer’s method [14]. In this paper, the in vitro effects of peptic-tryptic-cotazym@ digest prepared from durum (tetraploid) and bread wheat gliadins, were tested on two human cell lines by the method of Auricchio et al. [15]. One of these was the Hep-2 cell line already tested by Hudson et al. [ 1l] and the other was a diploid fibroblast line from embryo lung. The selection of these two cell lines is due, among other reasons, to the fact that gliadin digest seems to exert a direct non-specific cytotoxicity on various types of immature or malignant cells [l 11. Moreover, Weiser and Douglas [16] claim that gliadin would behave like lectin Con-A by recognising immature and transformed cells. These authors, while stressing the possible importance of cell-membrane glycoproteins for the aetiology of coeliac disease, pointed, out that incomplete glycoproteins are present on membranes of mitotically undifferentiated crypt cells and intestinal malignant cells [ 171. The aim of this work is also to explore to what extent human cell lines could be used to reproduce previous in vitro and in vivo results for coeliac disease, and to verify whether cytotoxic effects could be prevented by adding calf serum to the culture medium. MATERIALS
AND METHODS
Gliadin fractions were extracted with 70% aqueous ethanol from pure bread (var. Mentana) and durum (var. Azizia) wheat kernels, after extraction of globulin and albumin fractions, by the method described in Fig. 1. Peptic-tryptic-cotazym digests were prepared from bread and durum wheat gliadins, following the three step procedure of Bronstein et al. [18], as reported by De Ritis et al. [3]. All the enzymes used were previously purified by gel filtration on Sephadex G-100 [3]. 100 g protein fraction was digested in 1.1 of 0.2 N HCl (pH 1.8) with 2 g purified pepsin at 37°C
333
whole wheat flour (100 g)
supernatant
extraction with 1 I of 0.04 M Na2HP04 containing 1.8 M (NH& pH 7; centrifugation for 15 min at 16 OOOxg
4-1
dyalisis against water at 4°C for 48 h; centrifugation as above discard the supernatant
I
sediment extraction as above with 0.04 M Na2HP04 containing 0.4 M (NH&S04, pH 7; centrifugation as above
precipitate (globulin) supernatant dialysis as above and liophylisation sediment
I
albumin
extraction as above with 300 ml ethanol: water (70: 30 v/v); centrifugation as above
s”;d;;;;~~
I
sediment (discard)
Fig. 1. Differential extraction of globulin, albumin and gliadin fractions from whole wheat flour.
for 2 h. The peptic digest was further digested by addition of 2 g purified trypsin after pH adjustment to 8.0 with 2 N NaOH. The reaction mixture was vigorously stirred at 37’C for 4 h at pH 8.0. Then, the peptic-tryptic digest was treated with 2 g of purified cotazym@ and mechanically stirred for 2 h at pH 8.0. During the entire digestion procedure the pH was checked periodically and, when needed, adjusted with HCl or NAOH. At the end of the whole digestion procedure, the digest was submitted to gel filtration and the peptide fractions eluted after cytochrome c had been collected. Peptides with M, values < 2000 were removed by ultrafiltration on UM-2 Amicon membranes. These freeze-dried enzyme-free peptide pools, with M, values in the range 2000-12 000 have been coded as PTC protein digest. Hep-2 is an epithelial cell line derived from human carcinoma of the larynx, MRC-5 is a diploid fibroblast line from human embryo lung. Cell monolayers were cultured in falcon plastic dishes with Eagle Basal Medium supplemented with Earle salts and 10% fetal bovine serum (Flow Laboratories) at 37°C in 5% COzair. Cytotoxicity tests were performed by plating about lo4 and 3 x lo3 (0.05 ml) of MRC-5 and Hep-2 cells, respectively, in microtest falcon plates. After the cells had formed a confluent monolayer (approx. 48-72 h after plating), the incubation
334
medium was removed and substituted with the Eagle Basal Medium with or without 25% fetal calf serum and with or without the peptic-tryptic cotazym@ digest of gliadins from bread or durum wheat. The cells were exposed to a number of concentrations of peptides from both bread and durum wheat gliadins (100, 200, 500, 700, pg peptides/ml incubation medium for Hep-2 and 700, 1000, 1500 and 2000 ,ug peptides/ml incubation medium for MRCJ). All tests were run in triplicate. Observations were made every 24 h up to 7 days of culture. Toxicity scoring was based on the microscope appearances of cells and monolayers, usually at a magnification of x 100.
Fig. 2. Effect of peptic-tryptic-cotazim@ (PTC) digests from bread (hexaploid and durum (tetraploid) wheat gliadins on Hep-2 cell-line cultures in absence (A, B, C) or in presence (a, b, c) of 2% fetal calf serum. Time of culture: % h (A, a) control cultures; (B, b) bread wheat PTC digest, 700 fig/ml; (C, c) durum wheat PTC digest, 700 pg/ml.
335
RESULTS
Toxic effects were only seen after a 48-h culture of confluent monolayers of Hep-2 cells in the presence of a 700 pg concentration of peptic-tryptic-cotazym@ digest of bread wheat gliadin/ml of incubation medium (Fig. 2.). Observed cytotoxic effects consisted of morphological alterations in cell polygonal shape and the appearance of lysis areas in the monolayer. After 96 h of incubation, cell damage was clearly observed even at the lowest peptide concentration tested (100 hg/ml). Cell
Fig.
3. Effect
(tetraploid)
of
peptic-tryptic-cotazim@
wheat gliadins
of 2% fetal calf serum. digest,
2000 pg/ml;
on MRC-5
(PTC) digests from bread (hexaploid) and durum cell-line cultures in absence (A, B, C) or in presence (a, b, c)
Time of culture:
(C, c) durum
wheat
7 days.
(A, a) control
PTC digest,
2000 pg/ml.
cultures;
(B, b) bread
wheat
PTC
336
damage severity was seen to increase with the concentration of added bread wheat peptides for each incubation period. No significant difference in the time of onset and extent of cytotoxic effects was observed between cultures performed in the presence of bread wheat peptides with or without 2% fetal calf serum. The peptic-tryptic-cotazym Q digest of wheat bread gliadins was also found to be toxic for MRC-5 cell cultures (Fig. 3). In this case, the earliest onset of a cytotoxic effect was observed in the absence of serum after a 72-h incubation with a concentration of 2000 pg peptides/ml culture medium. Observed effects were similar to those mentioned above for the Hep-2 cell line with the exception of the appearance of a decreased MRC-5 cell layer density in the presence of bread wheat gliadinderived peptides. After longer incubation times (up to 7 days), toxic effects were also detectable at lower peptide concentrations (down to 500 pg/ml) and, for a given concentration, toxic effects were more severe than those observed after shorter incubations. The onset of cytotoxic effects was delayed (about 2 days) when fetal calf serum was also present in the culture medium. None of the above-reported effects was observed when Hep-2 or MRC-5 cells were incubated, with or without serum, under identical experimental conditions with the peptide mixture obtained from durum (tetraploid) wheat gliadin. DISCUSSION
The data reported show that the peptic-tryptic-cotazym@ digest of bread wheat gliadins reduces in vitro viability of human embryo and tumor cell lines. The time of onset and extent of toxic effects were dependent on initial peptide concentration. The presence of serum was capable of delaying the onset of cellular toxic effects in only one of the two cell lines tested. Consequently, it may be seen that these results only partially confirm the findings reported by Hudson et al. [l 11. We tend to believe that the Hudson et al. report [l l] on the capacity of serum to prevent cytotoxicity induction by a proteolytic digest of wheat gliadin was mainly due to the fact that they performed observation only after 24 h of incubation and not after longer periods of time. At present, we have no explanation for the transient protective effects of fetal calf serum. Furthermore, it would seem that Hep-2 epithelial cells are more sensitive to the toxic action of bread wheat gliadin peptides that MRC-5 fibroblasts cells. However, this may depend on the fact that, for any given peptide concentration, the number of exposed Hep-2 cells was smaller than that of MRC-5 cells. The cytotoxic effects of bread wheat gliadin peptides reported herein are related to the inhibition of in vitro development and morphogenesis of fetal rat intestine [3] and to the inhibition of the in vitro epithelial recovery of cultured specimens of flat intestinal mucosa from patients with active coeliac disease [3]. The peptic-tryptic-cotazym@ digest of durum wheat gliadins showed no cytotoxic activity on Hep-2 and MRC-5 cell lines. Moreover, durum wheat gliadin peptides did not affect the in vitro development and
331
morphogenesis of fetal rat intestine and were much less active than bread wheat gliadin peptides in preventing in vitro epithelial recovery of culture specimens of flat intestinal mucosa from active coeliac patients [3]. There is also evidence that bread wheat gliadin peptides do not exhibit any in vitro toxic activity on adult human fibroblast cells [13; preliminary unpublished results from our laboratory], on the jejunum from 21-day-old rat fetuses or newborn rats or on intestinal mucosa from normal individuals [3, 19, 201. All these data are consistent with the hypothesis that bread wheat gliadin peptides only interact with immature and/or relatively undifferentiated cells. Therefore, the presence of immature enterocytes, well known to occur on the surface of coeliac mucosa, could play a major role in making the intestinal mucosa of these susceptible individuals highly sensitive to bread wheat gliadin. A similar hypothesis was put forward by Douglas [16, 211, who showed that a carbohydrate-containing component of wheat gluten was capable of effective binding to membrane components of coeliac intestinal cells, and of only poor binding to normal intestinal cell membranes. Further investigations with different types of embryo and adult cells in the presence and absence of bread and durum wheat gliadin peptides might be useful in confirming this hypothesis. ACKNOWLEDGEMENTS
We wish to thank Prof. V. Silano for critical reading of the manuscript and for helpful discussions; Dr. M. De Vincenzi for supplying the gliadin PTC-digest. REFERENCES 1 R.M. Nardone, Toxicity testing in vitro, in G.H. Rothblat and V.J. Cristofalo (Eds.), Growth, Nutrition and Metabolism of Cells in Culture, Vol. 3, Academic Press, New York, 1977, pp. 478-495. 2 A. Paganuzzi Stammati, V. Silano and F. Zucco, Toxicology investigations with cell cultures systems, Toxicology, 20 (1981) 91-153. 3 G. De Ritis, P. Occorsio, S. Auricchio, F. Gramenzi, G. Morisi and V. Silano, Toxicity of wheat flour proteins and protein derived peptides for in vitro developing intestine from rat fetus, Pediat. Res., 13 (1979) pp. 1255-1261. 4 Z.M. Falchuk, R.L. Gebhard, C. Sessams and W. Strober, An in vitro model of gluten-sensitive enteropathy., J. Clin. Invest., 53 (1974), 487-500. 5 Z.M. Falchuk and A.J. Katz, Organ culture model of gluten-sensitive enteropathy, in B. McNicholl, C.F. MC Carthy and P.F. Fottrel (Eds.), Perspective in Coeliac Disease, MTP Press, Lancaster, 1978, pp. 65-72. 6 Z.M. Falchuk, R.L. Gebhard and W. Strober, The pathogenesis of sensitive enteropathy (coeliac sprue): organ culture studies, in W.J.T.M. Hekkens and A.S. Pena (Eds.), Coeliac Disease, Stenfert Kroese, Leiden, (1974), pp. 107-117. 7 J. Jos, G. Lenoir, G. De Ritis and J. Rey, In vitro culturing of biopsies from children, in W.J.T.M. Hekkens and A.S. Pena (Eds.), Coeliac Disease, Stenfert Kroese, Leiden, (1974), pp. 91-105. 8 J. Jos, G. Lenoir, G. De Ritis and J. Rey, In vitro pathogenetic studies of coeliac disease. Effects of protein digest on coeliac intestinal biopsy specimens maintained in culture for 48 hours, Stand. J. Gastroent. 10 (1975) 121-128.
338 9 H.J. Cornell and R.R.W. Townley, Investigation of possible intestinal peptidase deficiency in coeliac disease, Clin. Chim. Acta (1973) 113-125. 10 R.R.W. Townley, P.S. Bhathal, H.J. Cornell and J.D Mitchell, Toxicity of wheat gliadin fractions in coeliac disease, Lancet i (1973) 1363-1364. 11 D.A. Hudson, H.J. Cornell, D.R. Purdham and C.J. Rolles, Non-specific cytotoxicity of wheat gliadin components towards cultured human cells, Lancet i (1976) 339-341. 12 D.L.J. Freed and R.J. Cooper, Cytotoxicity of bread and soya protein in tissue culture, Lancet i (1977) 371-372. 13 R. Rusconi, M. Borgonovo, C. Bianchi Porro and P. Roggero, Ricerche sulla tossicita in vitro di una preparazione enzimatica di gliadina su fibroblasti di pazienti celiaci e normali, Minerva Ped., 30 (1978) 15-18. 14 A.C. Frazer, R.F. Fletcher, C.A.C. Ross, B. Shaw, H.C. Sammons and R. Schneider, Gluteninduced entheropathy. The effect of partially digested gluten, Lancet ii (1959) 252-255. 15 S. Auricchio, G. De Ritis, M. De Vincenzi, P. Occorsio and V. Silano, Effect of gliadin derived peptides from bread and durum wheats on small intestine culture from rat fetus and coeliac children, Pediat. Res., (1982) in press. 16 M.M. Weiser and A.P. Douglas, An alternative mechanism for gluten toxicity in coeliac disease, Lance& i (1976) 567-569. 17 M.M. Weiser, Intestinal epithelial cell surface membrane glycoprotein synthesis, II. Glycosyl transferases and endogenous acceptors of the undifferentiated cell surface membrane, J. Biol. Chem. 248 (1973) 2542-2548. 18 H.D. Bronstein, L.J. Haeffner and O.D. Kowlessar, Enzymatic digestion of gliadin: the effect of the resultant peptides in adult coeliac disease, Clin. Chim. Acta 14 (1966) 141-155. 19 Z.M. Falchuk, C. Sessoms and W. Strober, An in vitro model of gluten-sensitive enteropathy: evidence that gluten is not directly toxic to gastrointestinal epithelium, J. Clin. Invest. 51 (1972) 28a. 20 A.J. Katz, Z.M. Falchuk, W. Strober and H. Schwachman, Gluten-sensitive enteropathy. Inhibition by cortisol of the effect of gluten protein in vitro, N. Engl. J. Med. 295 (1976) 131-135. 21 A.P. Douglas, The binding of a glycopeptide component wheat gluten to intestinal mucosa of normal and coeliac human subjects, Clin. Chim Acta 73 (1976) 357-361.