JOURNAL
OF BIOLUMINESCENCE AND CHEMILUMINESCENCE VOL 5
161-164
(1990)
Influence of Gluten-derived Fractions on ChemiIum inescence Production by Human Neutrophils D. Roccatello*, M. Claudia Amprimo, Rosanna Coppo, G.Cavalli, G. Quattrocchio, B. Gianoglio, A. Ferrero, Clelia di Mauro, L. M. Sena and G. Piccoli Laboratorio Nefrologia e lmmunoematologia NAM, Cattedra di Nefrologia Medica e Dipartimento di Medicina ed Oncologia Sperimentale, Sezione di Patologia Generale dell'llniversita di Torino, Torino, Italy
The effects of gliadin and glyc-gli on leukocyte chemiluminescence response were assessed in vitro. A dose-dependent increase in chemiluminescence response of neutrophils stimulated by zymosan was observed by using gliadin at concentrations ranging between 1 and 20 pg. By increasing glyc-gli concentration, a bimodal response was observed with an enhancement up t o 50 pg/ml, followed by suppressive effects, which were again dose-dependent. The possible implications of these findings in human pathology are discussed. Keywords: Gliadin; glyc-gli; gluten; chernilurninescence; IgA nephropathy
INTRODUCTION
Cellular sensitization to wheat proteins has been described in coeliac disease (Haeney and Asquith, 1978; Holmes et al., 1976; Sikora et al., 1976). Serum antibodies to gliadin and other cereal proteins were reported in patients with coeliac disease, dermatitis herpetiformis (reviewed by Kieffer, 1985) and IgA nephropathy for which a defect in antigen exclusion at the mucosal level, allowing an increased antigen or antigen-antibody entry, was recently proposed (Clarkson et al., 1984). Despite these observations, the role of gluten-derived antigens in stimulating effector cell populations is still under study. We have investigated the ability of two wheat protein antigens, gliadin and glyc-gli, to modulate cell oxidative metabolism which can be crucial in
initiating permeability changes in the intestinal epithelium and tissue damage (Sjolander et al., 1984; Sugahara et al., 1986) in diseases characterized by gluten intolerance. METHODS Reagents
Gliadin and N-acetylglucosamine (GlcN-Ac) were purchased from Sigma Chemical Co., St Louis, MO, USA. The Gluten fraction glyc-gli was prepared as described by Douglas (1976). Briefly, 25g of gluten was stirred from 12h at 22°C in 800ml of O.lmol/l acetic acid in 2mol.4 ethanol, pH 6.5. The supernatant collected after centrifugation at 2500g for 15min was left on ice for 1h after addition of 125ml of 2.2mol/l
*Author for correspondence 0884-3996/90/030161-04$05 .OO @ 1990 by John Wiley & Sons, Ltd
Received October 1988
162
trichloroacetic acid. The precipitate obtained by centrifugation at 5000g for 1 h at 4°C was redissolved in 0.1 mol/l acetic acid and dialysed for 48 h against 0.15 mol/l phosphate buffered saline (PBS), pH 7.5. The soluble fraction was cleared by centrifugation at 5000g for 30min. Since wheat lectins such as gluten fractions are able to bind carbohydrates, particularly GlcN-Ac residues, GlcN-Ac was tested as competitive sugar for inhibitory action. Leukocyte isolation
Luekocytes were isolated from heparinized blood samples obtained from normal adult volunteers of both sexes, aged 24-47 years under no medication and fasting for at least 8 hours before venepuncture. Blood samples were immediately processed. Neutrophils were separated by adding a 15% volume of 3% dextran to peripheral blood, allowing it to settle at room temperature for 45 min. The supernatant containing leukocytes was aspirated and a half volume of FicollHypaque (Pharmacia, Uppsala, Sweden) was added to the bottom of the supernatant. The mixture was centrifuged at 400g for 20min at 4°C. The erythrocytes remaining in the cell pellet were lysed by hypotonic shock with distilled water. After washing the cell pellet three times with phosphate buffered saline, cells were resuspended in the appropriate medium. The percentage of neutrophils harvested with these techniques ranged between 95% and 98%. Cell viability after isolation and exposure to the reagents under study was examined by the Trypan blue exclusion method. Cell viability always exceeded the 95% value. Chemiluminescence
The instrumentation equipment included a LKBWallac 1251 luminometer coupled to either display unit, potentiometric recorder (2210-032) and computer (Olivetti M 24) using a specific PHAGO-TEST program (LKB) for data integration. The luminol stock solution was prepared by dissolving 1.77 mg of luminol (LKB-Wallac 1243, 216) in 1 ml dimethylsulphoxide to obtain a concentration of molil. Immediately before use it was further diluted to 10-'mol/l in Hanks' balanced salt solution (HBSS) without phenol
D. ROCCATELLO H A L .
red. Zymosan was opsonized by incubating 50 mg of it in 3 ml fresh human serum and 1 ml HBSS for 30 min. After washing, zymosan was resuspended in HBSS at the concentration of 12.5 mgiml. The following reagents were subsequently pipetted into duplicate cuvettes: 6OOyl of lo-' mol/l luminol, 50y1 of opsonized zymosan (or HBSS in controls) and 50yl of peripheral phagocyte suspension ( 10' cells/ml) under study. Phagocyte samples were previously incubated with various concentrations of gliadin and glyc-gli for 30min at 4°C. In other experiments various concentrations of GlcN-Ac were pre-incubated or co-incubated with the gluten fractions. Cuvette contents were gently mixed and conveyed to the measurement position. Duplicate samples were usually measured for 60min at 3min intervals as millivolts (mV) of light emission. In some experiments measurements were prolonged up to 120 min. Results were automatically plotted on a graph having light intensity units on ordinate and time units on abscissa. The chemiluminescence response was determined both as a height of the curve (peak value, expressed in mV) and as the integral under the curve (mV X min). Statistics
Statistical significance of differences in the values of the diverse cell activities recorded under different experimental conditions was determined by Wilkoxon test. The linear coefficient of regression was used when positive or negative relationships were analysed. RESULTS The peak chemiluminescence response of neutrophils exposed to various concentrations of gliadin and glyc-gli and stimulated by opsonized zymosan occurred at the same time as untreated cell suspensions. However, the chemiluminescence generation was found to be markedly affected by prior incubation of neutrophils with the two gluten fractions as regards height (peak value) and area (integral) of the curve. These two parameters were strictly related by a positive correlation (Y = 0.9, p < 0.01). Fig. 1 shows the dose-response relationship for the two gluten fractions under study. Data represent the mean percentage (F SD) of the
163
GLUTEN FRACTIONS AND CHEMILUMINESCENCE
**
**
120
120
m
@ 90
90
.
0
.
1.2
.
2.5
.
5.0
.
10
.
20
80
.
.
.
.
0
12
25
50
.
*:
100 200
W m l Figure 1. Chemiluminescence generation by neutrophils preincubated with increasing concentrations of gliadin (A). glyc-gli (B) and stimulated with opsonized zymosan Results are expressed as mean percentage (kSD) of the values obtained by cell suspensions unexposed to gluten fractions and stimulated by zymosan * = p < 0 05, * * = p < 0 01
values obtained by treated cells as compared to untreated preparations stimulated by zymosan. The increase in chemiluminescence response due to gliadin was found to be almost linear by increasing its concentration in the incubation medium. The dose of glyc-gli required to influence the chemiluminescence induced by opsonized zymosan, was found to be tenfold larger than that of gliadin. However, by increasing glyc-gli concentration, a bimodal response was observed with an enhancement up to 50yg/ml followed by suppressive effects which were again dose-dependent. Washing cells after incubation of neutrophils with the gluten fractions did not abolish these effects. Co-incubation of GlcN-Ac with gliadin or glyc-gli resulted in a complete abolition of these effects. The same results were obtained by prior incubation of GlcN-Ac before exposure to gliadin or glyc-gli. When chemiluminescence response was followed for longer time (up to 120 min) no further information was collected. DISCUSSION
Binding on neutrophils of some lectins, as concanavalin A (Romeo et al., 1973) or WGA (Magnusson et nl., 1988) results in metabolic activation and respiratory burst, yielding a series of reactive oxygen metabolites which can elicit chemiluminescence. We found gliadin and glycgli capable of modifying the chemiluminescence pattern of zymosan-stimulated neutrophils. Of particular interest, gliadin increased the chemiluminescence generation in a dose-dependent manner at the concentrations employed. It is worth noting that neutrophils phagocytosing opsonized
zymosan were found to induce an increase in permeability of a monolayer of epithelial cells (Sugahara et af., 1986). This effect was attributed to the release of oxygen radicals and was partially abolished by catalase, which inactivates H 2 0 2 . Moreover, an activator of neutrophil oxidative metabolism, such as N-formylmethionyl-leucylphenylalanine (Dahlgren, 1987; Schiffman el al., 1978), was reported to increase in vitro the epithelial permeability of rat small intestine. It is therefore conceivable that gliadin itself, or a gliadin-derived factor, may favour tissue damage and permeability changes with the same mechanism(s). The effects of glyc-gli appear more intriguing. Glyc-gli is a WGA-like lectin of gluten, able to agglutinate only papain-treated erythrocytes. WGA mediates a variety of divergent effects on leukocytes. At high concentrations it suppresses DNA synthesis and cell division in mitogenstimulated lymphocytes (Greene and Waldmann, 1980) and inhibits amino-acid transport and phospholipid turnover (Greene et al., 1976). Conversely, at low concentrations, it stimulates DNA synthesis and blast transformation (Gordon et al., 1980) in T lymphocytes and activates B cell immunoglobulin production (Greene et al., 1981). Therefore the bimodal influence of glyc-gli o n chemiluminescence generation by neutrophils is not surprising. These effects are rather exciting when considering that glyc-gli binds to intestinal cell membranes (Douglas, 1976). The storage proteins of several cereals, particularly wheat gliadin, have an aetiological role in gluten-sensitive enteropathy. Coeliac disease is the best example of this disorder. Most patients with dermatitis herpetiformis also have glutensensitive enteropathy similar to but less severe
D. ROCCATELLO ETAL.
164
than that of coeliac disease (Marks et a [ . , 1966). Finally numerous studies indicate an abnormal intestinal permeability (Davin et al., 1988) and high levels of serum IgA to dietary antigens, particularly gliadin (Laurent ef al., 1987; Nagy et al., 1987), in IgA nephropathy, a glomerulonephritis characterized by mesangial deposition of IgA. The possible occurrence of IgA nephropathy in diseases associated with gluten-sensitive enteropathy (Clarkson et al., 1984) and the favourable effect of gluten-free diet in decreasing the levels of circulating IgA immune complexes in a group of IgA nephropathy patients (Coppo et al., 1986) substantiate the hypothesis that gluten compounds act both as toxic lectins, enhancing the intestinal permeability, and as antigens promoting a mucosal immune response, leading to high levels of immune complexes. The present study provides information on some mechanisms possibly involved in these pathological events. By enhancing generation of oxygen reactive species, gliadin and glyc-gli could contribute, at least in predisposed subjects, to both permeability changes of intestinal epithelium and tissue damage. This could promote penetration of either free or complexed antigens into the body, eventually resulting in renal deposition of immune material. REFERENCES Clarkson, A . R.. Woodroffe, A . J . , Bannister, K. M., Lomax-Smith, J . D . and Aarons, I . (1984). The syndrome of IgA nephropathy. Clin. Nephrol., 21, 7-14. Coppo, R . , Basolo, B., Rollino, C . , Roccatello, D . , Martina, G . , Amore, A., Bongiorno, G . and Piccoli, G . (1986). Mediterranean diet and primary IgA nephropathy. Clin. Nephrol., 26, 72-82. Dahlgren, C. (1987). Polymorphonuclear leukocyte chemiluminescence induced by formyl-methionyl-leucilphenylalanine and phorbol myristate acetate. Effects of catalase and superoxide dismutase. Agents Action, 21, 1-9. Davin, J. C., Forget, P. and Mahieu, P. R. (1988). Increased intestinal permeability to (5'Cr) E D T A is correlated with IgA immune complex-plasma levels in children with IgA-associated nephropathies. Acta Paedriatr. Scand., 77. 11S124. Douglas, A. P. (1976). The binding of a glycopeptide component of wheat gluten to intestinal mucosa of normal and coelic human subjects. Clin. Chem. Acta., 73, 357-361. Gordon, L. K., Hamill, B. and Parker, C. W . (1980). The activation of blast transformation and DNA synthesis in
human peripheral blood lymphocytes by wheat germ agglutinin. J . Imtnunol.,125, 814-823. Greene, W. C. and Waldmann. T. A . (1980). Inhibition of human lymphocyte proliferation by the nonmitogenic lectin wheat germ agglutinin. J . Immwiol., 124, 29792986. Greene, W. C . . Parker. C. M . and Parker, C. W. (1976). Opposing effects of mitogenic and nonmitogcnic lectins on lymphocyte activation. J . Biol. Chetn. 251. 4017-4024. Greene, W. C . , Goldnian. C. K.. Marshall. 7'. A , , Fleisher, T. A . and Waldmann, T. A . (1981). Stimulation of immunoglobulin biosynthesis in human B cells by wheat germ agglutinin. 1. Evidence that WGA can produce both ;I positive and negative signal f o r activation o r human lymphocytes. J . Immunol.. 127, 799-809. Haeney, M. R . and Asquith, P. (1978). In-vitro immunological assay for diagnosis of coeliac disease. Lancer. i , 876. Holmes, G . K. T., Asquith, P. and Cooke, W. T. (1976). Cell-mediated immunity to gluten fraction 111 in adult coeliac disease. Cliti. Exp. Immunol., 24, 259-265. Kieffer, M. (1985). Serum antibodies to gliadin and other cereal proteins in patients with coeliac disease and dermatitis herpetiformis. Danish Med. Bull.. 32, 251-262. Kolberg, J . and Sollid, L. (1985). Lcctin activity of gluten identified as wheat germ agglutinin. Biocherrr. Biophys. Res. Commun., 130, 867-872. Laurent, J . , Branellac, A . , Heslan, J . M.. Rostoker. G.. Bruneau, C., Andrk, C.. Intrator. L. and Lagrue. G . (1987). An increase in circulating IgA antibodies to gliadin in IgA mesangial glomerulonephritis. A m . J . Med., 7, 178-183. Magnusson, K . E., Dahlgren, C. and Sjolander, A . (1988). Distinct patterns of granulocyte luminol-dependent chemiluminescence response to leetins WGA and RCA-I. Inflammation, 12, 17-22. Marks, J., Shuster, S. and Watson. A. J . (1966). Small bowel changes in dermatitis herpetiformis. Lancer, ii, 128@1282. Nagy, J . , Scott, H . and Brandtzaeg, P. (1987). Food antigens in the pathogenesis of IgA nephropathy. (Abstr) XXIVth Congress ofthe EDTA-ERA. Berlin 1987, p. 39. Romeo, D., Zabucci, G . and Rossi, F. (1973). Reversible metabolic stimulation of polymorphonuclear leukocytes and macrophages by concanavalin A . Nature (London) New Biol., 243, 11 1-1 12. Schiffman, E., Corcoran, B. A . and Ashwanikumai, S . (1978). Molecular events in the responses of neutrophils to synthetic N - M e t chemotactic peptides. Demonstration of a specific receptor. In Leukocyte Chemotaxis, Gallin, J . I . and Quie, P. G . (Eds), Raven Press, New York, p. 97. Sikora, K . , Anand, B. S . , Truelove, S. C., Ciclitira, P. J . and Offord, R. E. (1976). Stimulation of lymphocytes from patients with coeliac disease by a subfraction of gluten. Lancet, ii, 389-391, Sjolander, A , , Magnusson, K. E. and Latcovic, S. (1984). The effect of concanavalin A and wheat germ agglutinin on the ultrastructure and permeability of rat intestine. Inr. Arch. Allergy. Appl. Immunol., 75, 23C236. Sugahara, K . , Cott, G . R., Parsons, E . , Mason, R. J . , Sandhaus, R. A. and Henson, P. M. (1986). Epithelial permeability produced by phagocytosing neutrophils in vitro. A m . J . Respir. Dis., 133, 875-881. ~
OF BIOLUMINESCENCE AND CHEMILUMINESCENCE VOL 5
161-164
(1990)
Influence of Gluten-derived Fractions on ChemiIum inescence Production by Human Neutrophils D. Roccatello*, M. Claudia Amprimo, Rosanna Coppo, G.Cavalli, G. Quattrocchio, B. Gianoglio, A. Ferrero, Clelia di Mauro, L. M. Sena and G. Piccoli Laboratorio Nefrologia e lmmunoematologia NAM, Cattedra di Nefrologia Medica e Dipartimento di Medicina ed Oncologia Sperimentale, Sezione di Patologia Generale dell'llniversita di Torino, Torino, Italy
The effects of gliadin and glyc-gli on leukocyte chemiluminescence response were assessed in vitro. A dose-dependent increase in chemiluminescence response of neutrophils stimulated by zymosan was observed by using gliadin at concentrations ranging between 1 and 20 pg. By increasing glyc-gli concentration, a bimodal response was observed with an enhancement up t o 50 pg/ml, followed by suppressive effects, which were again dose-dependent. The possible implications of these findings in human pathology are discussed. Keywords: Gliadin; glyc-gli; gluten; chernilurninescence; IgA nephropathy
INTRODUCTION
Cellular sensitization to wheat proteins has been described in coeliac disease (Haeney and Asquith, 1978; Holmes et al., 1976; Sikora et al., 1976). Serum antibodies to gliadin and other cereal proteins were reported in patients with coeliac disease, dermatitis herpetiformis (reviewed by Kieffer, 1985) and IgA nephropathy for which a defect in antigen exclusion at the mucosal level, allowing an increased antigen or antigen-antibody entry, was recently proposed (Clarkson et al., 1984). Despite these observations, the role of gluten-derived antigens in stimulating effector cell populations is still under study. We have investigated the ability of two wheat protein antigens, gliadin and glyc-gli, to modulate cell oxidative metabolism which can be crucial in
initiating permeability changes in the intestinal epithelium and tissue damage (Sjolander et al., 1984; Sugahara et al., 1986) in diseases characterized by gluten intolerance. METHODS Reagents
Gliadin and N-acetylglucosamine (GlcN-Ac) were purchased from Sigma Chemical Co., St Louis, MO, USA. The Gluten fraction glyc-gli was prepared as described by Douglas (1976). Briefly, 25g of gluten was stirred from 12h at 22°C in 800ml of O.lmol/l acetic acid in 2mol.4 ethanol, pH 6.5. The supernatant collected after centrifugation at 2500g for 15min was left on ice for 1h after addition of 125ml of 2.2mol/l
*Author for correspondence 0884-3996/90/030161-04$05 .OO @ 1990 by John Wiley & Sons, Ltd
Received October 1988
162
trichloroacetic acid. The precipitate obtained by centrifugation at 5000g for 1 h at 4°C was redissolved in 0.1 mol/l acetic acid and dialysed for 48 h against 0.15 mol/l phosphate buffered saline (PBS), pH 7.5. The soluble fraction was cleared by centrifugation at 5000g for 30min. Since wheat lectins such as gluten fractions are able to bind carbohydrates, particularly GlcN-Ac residues, GlcN-Ac was tested as competitive sugar for inhibitory action. Leukocyte isolation
Luekocytes were isolated from heparinized blood samples obtained from normal adult volunteers of both sexes, aged 24-47 years under no medication and fasting for at least 8 hours before venepuncture. Blood samples were immediately processed. Neutrophils were separated by adding a 15% volume of 3% dextran to peripheral blood, allowing it to settle at room temperature for 45 min. The supernatant containing leukocytes was aspirated and a half volume of FicollHypaque (Pharmacia, Uppsala, Sweden) was added to the bottom of the supernatant. The mixture was centrifuged at 400g for 20min at 4°C. The erythrocytes remaining in the cell pellet were lysed by hypotonic shock with distilled water. After washing the cell pellet three times with phosphate buffered saline, cells were resuspended in the appropriate medium. The percentage of neutrophils harvested with these techniques ranged between 95% and 98%. Cell viability after isolation and exposure to the reagents under study was examined by the Trypan blue exclusion method. Cell viability always exceeded the 95% value. Chemiluminescence
The instrumentation equipment included a LKBWallac 1251 luminometer coupled to either display unit, potentiometric recorder (2210-032) and computer (Olivetti M 24) using a specific PHAGO-TEST program (LKB) for data integration. The luminol stock solution was prepared by dissolving 1.77 mg of luminol (LKB-Wallac 1243, 216) in 1 ml dimethylsulphoxide to obtain a concentration of molil. Immediately before use it was further diluted to 10-'mol/l in Hanks' balanced salt solution (HBSS) without phenol
D. ROCCATELLO H A L .
red. Zymosan was opsonized by incubating 50 mg of it in 3 ml fresh human serum and 1 ml HBSS for 30 min. After washing, zymosan was resuspended in HBSS at the concentration of 12.5 mgiml. The following reagents were subsequently pipetted into duplicate cuvettes: 6OOyl of lo-' mol/l luminol, 50y1 of opsonized zymosan (or HBSS in controls) and 50yl of peripheral phagocyte suspension ( 10' cells/ml) under study. Phagocyte samples were previously incubated with various concentrations of gliadin and glyc-gli for 30min at 4°C. In other experiments various concentrations of GlcN-Ac were pre-incubated or co-incubated with the gluten fractions. Cuvette contents were gently mixed and conveyed to the measurement position. Duplicate samples were usually measured for 60min at 3min intervals as millivolts (mV) of light emission. In some experiments measurements were prolonged up to 120 min. Results were automatically plotted on a graph having light intensity units on ordinate and time units on abscissa. The chemiluminescence response was determined both as a height of the curve (peak value, expressed in mV) and as the integral under the curve (mV X min). Statistics
Statistical significance of differences in the values of the diverse cell activities recorded under different experimental conditions was determined by Wilkoxon test. The linear coefficient of regression was used when positive or negative relationships were analysed. RESULTS The peak chemiluminescence response of neutrophils exposed to various concentrations of gliadin and glyc-gli and stimulated by opsonized zymosan occurred at the same time as untreated cell suspensions. However, the chemiluminescence generation was found to be markedly affected by prior incubation of neutrophils with the two gluten fractions as regards height (peak value) and area (integral) of the curve. These two parameters were strictly related by a positive correlation (Y = 0.9, p < 0.01). Fig. 1 shows the dose-response relationship for the two gluten fractions under study. Data represent the mean percentage (F SD) of the
163
GLUTEN FRACTIONS AND CHEMILUMINESCENCE
**
**
120
120
m
@ 90
90
.
0
.
1.2
.
2.5
.
5.0
.
10
.
20
80
.
.
.
.
0
12
25
50
.
*:
100 200
W m l Figure 1. Chemiluminescence generation by neutrophils preincubated with increasing concentrations of gliadin (A). glyc-gli (B) and stimulated with opsonized zymosan Results are expressed as mean percentage (kSD) of the values obtained by cell suspensions unexposed to gluten fractions and stimulated by zymosan * = p < 0 05, * * = p < 0 01
values obtained by treated cells as compared to untreated preparations stimulated by zymosan. The increase in chemiluminescence response due to gliadin was found to be almost linear by increasing its concentration in the incubation medium. The dose of glyc-gli required to influence the chemiluminescence induced by opsonized zymosan, was found to be tenfold larger than that of gliadin. However, by increasing glyc-gli concentration, a bimodal response was observed with an enhancement up to 50yg/ml followed by suppressive effects which were again dose-dependent. Washing cells after incubation of neutrophils with the gluten fractions did not abolish these effects. Co-incubation of GlcN-Ac with gliadin or glyc-gli resulted in a complete abolition of these effects. The same results were obtained by prior incubation of GlcN-Ac before exposure to gliadin or glyc-gli. When chemiluminescence response was followed for longer time (up to 120 min) no further information was collected. DISCUSSION
Binding on neutrophils of some lectins, as concanavalin A (Romeo et al., 1973) or WGA (Magnusson et nl., 1988) results in metabolic activation and respiratory burst, yielding a series of reactive oxygen metabolites which can elicit chemiluminescence. We found gliadin and glycgli capable of modifying the chemiluminescence pattern of zymosan-stimulated neutrophils. Of particular interest, gliadin increased the chemiluminescence generation in a dose-dependent manner at the concentrations employed. It is worth noting that neutrophils phagocytosing opsonized
zymosan were found to induce an increase in permeability of a monolayer of epithelial cells (Sugahara et af., 1986). This effect was attributed to the release of oxygen radicals and was partially abolished by catalase, which inactivates H 2 0 2 . Moreover, an activator of neutrophil oxidative metabolism, such as N-formylmethionyl-leucylphenylalanine (Dahlgren, 1987; Schiffman el al., 1978), was reported to increase in vitro the epithelial permeability of rat small intestine. It is therefore conceivable that gliadin itself, or a gliadin-derived factor, may favour tissue damage and permeability changes with the same mechanism(s). The effects of glyc-gli appear more intriguing. Glyc-gli is a WGA-like lectin of gluten, able to agglutinate only papain-treated erythrocytes. WGA mediates a variety of divergent effects on leukocytes. At high concentrations it suppresses DNA synthesis and cell division in mitogenstimulated lymphocytes (Greene and Waldmann, 1980) and inhibits amino-acid transport and phospholipid turnover (Greene et al., 1976). Conversely, at low concentrations, it stimulates DNA synthesis and blast transformation (Gordon et al., 1980) in T lymphocytes and activates B cell immunoglobulin production (Greene et al., 1981). Therefore the bimodal influence of glyc-gli o n chemiluminescence generation by neutrophils is not surprising. These effects are rather exciting when considering that glyc-gli binds to intestinal cell membranes (Douglas, 1976). The storage proteins of several cereals, particularly wheat gliadin, have an aetiological role in gluten-sensitive enteropathy. Coeliac disease is the best example of this disorder. Most patients with dermatitis herpetiformis also have glutensensitive enteropathy similar to but less severe
D. ROCCATELLO ETAL.
164
than that of coeliac disease (Marks et a [ . , 1966). Finally numerous studies indicate an abnormal intestinal permeability (Davin et al., 1988) and high levels of serum IgA to dietary antigens, particularly gliadin (Laurent ef al., 1987; Nagy et al., 1987), in IgA nephropathy, a glomerulonephritis characterized by mesangial deposition of IgA. The possible occurrence of IgA nephropathy in diseases associated with gluten-sensitive enteropathy (Clarkson et al., 1984) and the favourable effect of gluten-free diet in decreasing the levels of circulating IgA immune complexes in a group of IgA nephropathy patients (Coppo et al., 1986) substantiate the hypothesis that gluten compounds act both as toxic lectins, enhancing the intestinal permeability, and as antigens promoting a mucosal immune response, leading to high levels of immune complexes. The present study provides information on some mechanisms possibly involved in these pathological events. By enhancing generation of oxygen reactive species, gliadin and glyc-gli could contribute, at least in predisposed subjects, to both permeability changes of intestinal epithelium and tissue damage. This could promote penetration of either free or complexed antigens into the body, eventually resulting in renal deposition of immune material. REFERENCES Clarkson, A . R.. Woodroffe, A . J . , Bannister, K. M., Lomax-Smith, J . D . and Aarons, I . (1984). The syndrome of IgA nephropathy. Clin. Nephrol., 21, 7-14. Coppo, R . , Basolo, B., Rollino, C . , Roccatello, D . , Martina, G . , Amore, A., Bongiorno, G . and Piccoli, G . (1986). Mediterranean diet and primary IgA nephropathy. Clin. Nephrol., 26, 72-82. Dahlgren, C. (1987). Polymorphonuclear leukocyte chemiluminescence induced by formyl-methionyl-leucilphenylalanine and phorbol myristate acetate. Effects of catalase and superoxide dismutase. Agents Action, 21, 1-9. Davin, J. C., Forget, P. and Mahieu, P. R. (1988). Increased intestinal permeability to (5'Cr) E D T A is correlated with IgA immune complex-plasma levels in children with IgA-associated nephropathies. Acta Paedriatr. Scand., 77. 11S124. Douglas, A. P. (1976). The binding of a glycopeptide component of wheat gluten to intestinal mucosa of normal and coelic human subjects. Clin. Chem. Acta., 73, 357-361. Gordon, L. K., Hamill, B. and Parker, C. W . (1980). The activation of blast transformation and DNA synthesis in
human peripheral blood lymphocytes by wheat germ agglutinin. J . Imtnunol.,125, 814-823. Greene, W. C. and Waldmann. T. A . (1980). Inhibition of human lymphocyte proliferation by the nonmitogenic lectin wheat germ agglutinin. J . Immwiol., 124, 29792986. Greene, W. C . . Parker. C. M . and Parker, C. W. (1976). Opposing effects of mitogenic and nonmitogcnic lectins on lymphocyte activation. J . Biol. Chetn. 251. 4017-4024. Greene, W. C . , Goldnian. C. K.. Marshall. 7'. A , , Fleisher, T. A . and Waldmann, T. A . (1981). Stimulation of immunoglobulin biosynthesis in human B cells by wheat germ agglutinin. 1. Evidence that WGA can produce both ;I positive and negative signal f o r activation o r human lymphocytes. J . Immunol.. 127, 799-809. Haeney, M. R . and Asquith, P. (1978). In-vitro immunological assay for diagnosis of coeliac disease. Lancer. i , 876. Holmes, G . K. T., Asquith, P. and Cooke, W. T. (1976). Cell-mediated immunity to gluten fraction 111 in adult coeliac disease. Cliti. Exp. Immunol., 24, 259-265. Kieffer, M. (1985). Serum antibodies to gliadin and other cereal proteins in patients with coeliac disease and dermatitis herpetiformis. Danish Med. Bull.. 32, 251-262. Kolberg, J . and Sollid, L. (1985). Lcctin activity of gluten identified as wheat germ agglutinin. Biocherrr. Biophys. Res. Commun., 130, 867-872. Laurent, J . , Branellac, A . , Heslan, J . M.. Rostoker. G.. Bruneau, C., Andrk, C.. Intrator. L. and Lagrue. G . (1987). An increase in circulating IgA antibodies to gliadin in IgA mesangial glomerulonephritis. A m . J . Med., 7, 178-183. Magnusson, K . E., Dahlgren, C. and Sjolander, A . (1988). Distinct patterns of granulocyte luminol-dependent chemiluminescence response to leetins WGA and RCA-I. Inflammation, 12, 17-22. Marks, J., Shuster, S. and Watson. A. J . (1966). Small bowel changes in dermatitis herpetiformis. Lancer, ii, 128@1282. Nagy, J . , Scott, H . and Brandtzaeg, P. (1987). Food antigens in the pathogenesis of IgA nephropathy. (Abstr) XXIVth Congress ofthe EDTA-ERA. Berlin 1987, p. 39. Romeo, D., Zabucci, G . and Rossi, F. (1973). Reversible metabolic stimulation of polymorphonuclear leukocytes and macrophages by concanavalin A . Nature (London) New Biol., 243, 11 1-1 12. Schiffman, E., Corcoran, B. A . and Ashwanikumai, S . (1978). Molecular events in the responses of neutrophils to synthetic N - M e t chemotactic peptides. Demonstration of a specific receptor. In Leukocyte Chemotaxis, Gallin, J . I . and Quie, P. G . (Eds), Raven Press, New York, p. 97. Sikora, K . , Anand, B. S . , Truelove, S. C., Ciclitira, P. J . and Offord, R. E. (1976). Stimulation of lymphocytes from patients with coeliac disease by a subfraction of gluten. Lancet, ii, 389-391, Sjolander, A , , Magnusson, K. E. and Latcovic, S. (1984). The effect of concanavalin A and wheat germ agglutinin on the ultrastructure and permeability of rat intestine. Inr. Arch. Allergy. Appl. Immunol., 75, 23C236. Sugahara, K . , Cott, G . R., Parsons, E . , Mason, R. J . , Sandhaus, R. A. and Henson, P. M. (1986). Epithelial permeability produced by phagocytosing neutrophils in vitro. A m . J . Respir. Dis., 133, 875-881. ~
