METHOD FOR THE DETOXIFICATION OF GLUTEN PROTEINS FROM CEREAL GRAINS AND USES THEREOF IN MEDICAL FIELD

Abstract

An improved method of detoxifying gluten proteins from cereal grains produces detoxified cereal grains with a reduction of antigenicity of toxic epitopes of gluten proteins by up to a range between 0 and 20 ppm. The detoxified cereal grains can be advantageously used for preparation of of food products (e.g. bakery products, pasta or dairy products) having a manifest preventive and/or therapeutic effect for gut dysbiosis caused by bacterial or viral infective agents or by pathologies with a strong inflammatory or autoimmune component such as celiac disease, ulcerative colitis, Crohn's disease and irritable intestine syndrome.

Claims

1-16. (canceled)

17. A method of producing detoxified cereal grains, comprising: hydrating cereal grains with water up to a humidity degree of the cereal grains between 15% and 18%; treating the hydrated cereal grains by electromagnetic waves for a time and with a power to reach a first temperature of the cereal grains between 60? C. and 70? C.; suspending the electromagnetic waves until the hydrated cereal grains reach a second temperature between 50? C. and 60? C. and a first humidity loss between 14% and 16%; treating the hydrated cereal grains by electromagnetic waves for a time and with a power to reach a third temperature of the hydrated cereal grains between 80? C. and 90? C.; suspending the electromagnetic waves until the hydrated cereal grains reach a fourth temperature between 70? C. and 80? C. and a second humidity loss between 40% and 44%; treating the hydrated cereal grains by electromagnetic waves for a time and with a power to reach a fifth temperature of the hydrated cereal grains between 110? C. and 120? C.; suspending the electromagnetic waves until the hydrated cereal grains reach a sixth temperature between 80? C. and 90? C. and a third humidity loss between 50% and 60%; and cooling the hydrated cereal grains at room temperature to produce the detoxified cereal grains.

18. The method of claim 17, wherein said electromagnetic waves are microwaves or infrared.

19. The method of claim 17, wherein when the electromagnetic waves are microwaves, the steps of treating the hydrated grains are carried out in a microwave oven.

20. The method of claim 17, further comprising milling the detoxified cereal grains to obtain detoxified flours or semolina.

21. The method of claim 20, further comprising extracting gluten from the detoxified flours or semolina with solvents.

22. The method of claim 17, wherein the cereal grains are selected from the group consisting of wheat, barley, orzo, rye and oat.

23. The method of claim 17, wherein the detoxified cereal grains have a level of toxic gluten epitopes reduced to a range between 0 and 20 ppm.

24. Detoxified cereal grains obtained by the method of claim 1, wherein the detoxified cereal grains have a level of toxic gluten epitopes reduced to a range between 0 and 20 ppm.

25. A food product comprising the detoxified cereal grains of claim 24, the food product selected from the group consisting of bread, pasta, bakery products, breakfast cereals, beer, ice-cream, dairy products, sauce, juice, baby foods, and salami.

26. A method of preventing or treating gut dysbiosis in a subject, comprising: administering to the subject an effective amount of the detoxified cereal grains of claim 24.

27. A method of preventing or treating in a subject inflammatory or autoimmune intestinal chronic diseases selected from the group consisting of celiac disease, ulcerative colitis, Crohn's disease and irritable intestine syndrome, the method comprising: administering to the subject an effective amount of the detoxified cereal grains of claim 24.

28. A method of preventing or treating in a subject systemic metabolic diseases selected from the group consisting of obesity, type 1 diabetes and type 2 diabetes, the method comprising: administering to the subject an effective amount of the detoxified cereal grains of claim 24.

29. An antimicrobial agent against Gram-negative and Gram-positive bacteria, comprising the detoxified cereal grains of claim 24.

30. The antimicrobial agent of claim 29, wherein the Gram-negative bacteria belong to Salmonella genus and the Gram-positive bacteria belong to the Staphylococcus genus.

31. The antimicrobial agent of claim 29, wherein the Gram-negative bacteria belong to the species Salmonella Typhimurium.

32. The antimicrobial agent of claim 29, wherein the Gram-positive bacteria belong to the species Staphylococcus aureus.

33. A protective agent towards probiotic species, comprising the detoxified cereal grains of claim 24.

34. The protective agent of claim 33, wherein the probiotic species belong to the Lactobacilli genus.

35. The protective agent of claim 33, wherein the probiotic species belong to the species Lactobacillus acidophilus.

36. A thickening agent for the preparation of foodstuff, comprising the detoxified cereal grains of claim 24.

Description

[0075] The present invention shall now be described by way of non-limiting illustration, on the basis of the results indicated in the following examples and in the accompanying figures, in which:

[0076] FIG. 1 shows an illustrative diagram of the pathogenesis of the celiac disease (MC).

[0077] FIG. 2 shows the wheat sections (1 ?M) related to the steps of the method of treatment after marking with antibody 06010 which recognizes the protein fractions of gliadins and of low molecular weight glutenins (LMW-GS). Panel A: Step b; panel B: Step c; panel C: Step d; panel D: Step e; panel E: Step f; panel F: Step g of the method according to the invention.

[0078] FIG. 3 represents the structure of the protein before and after the heat treatment according to the method of the present invention.

[0079] FIG. 4 shows the wheat sections (1 ?m) control and after treatment of the method of the present invention. (1) Control wheat section, with a non-homogeneous protein matrix (Pb 1); (2) Wheat section after treatment, with a homogeneous and confluent protein matrix (Pb 2). The non-parametric Friedman test applied to 6 different seeds (4 sections for each seed) detected highly significant differences between the two types of protein bodies (Pb1 and Pb2) in the seeds of the control and after treatment samples.

[0080] FIG. 5 shows the extraction of the protein fractions and the separation through SDS-PAGE in reducing conditions. Lane 1, gliadins extracted from flour coming from control seeds, which have not undergone the heat treatment of the present invention; lane 2, gliadins extracted from flours coming from seeds treated with the method of the present invention; lane 3, HMW-GS and gliadins extracted from flour coming from control seeds that have not undergone the heat treatment of the present invention; lane 4, HMW-GS and gliadins extracted from flours coming from seeds treated with the method of the present invention; lane 5, glutenins extracted from flour coming from seed control that have not undergone the heat treatment of the present invention; lane 6, glutenins extracted from flours coming from seeds treated with the method of the present invention; lane 7, total proteins extracted from flour coming from control seeds that have not undergone the heat treatment of the present invention; lane 8, total glutenin proteins extracted from flours coming from seeds treated with the method of the present invention.

[0081] FIG. 6 shows a hypothesis of aggregation between the proteins present in the same protein body and with different charge.

[0082] FIG. 7 shows the summary histograms of the ELISA assay with monoclonal antibody R5 Ridascreen Gliadin carried out on samples of control flour and after treatment according to the method of the international patent application WO2014/053891 compared with the method of the present invention.

[0083] FIG. 8 shows the sections of control wheat (Control) and after treatment (Treated) of the method described in the present invention, cut transversely, and examined through SEM-Immunogold, immunomarked with 0610 antibody and ?-gliadin. 1. Control marked with 0610 antibody; 2. Control after marking with antibody anti ?-gliadin; 3. Treated marked with 0610 antibody; 4. Treated marked with antibody anti ?-gliadin. The arrow in the figure represent the silver particles (AgNp) detected through EDS analysis (Energy Dispersive Spectroscopy).

[0084] FIG. 9 shows the sections of wheat (1 ?m) control (Control) and after treatment (Treated) according to the method of the present invention, marked with the 0610 antibody, HMW-G e ?-gliadin. 1. Control marked with 0610 antibody; 2. Control after marking with ?-gliadin antibody; 3. Control marked with antibody HMW-GS; 4. Treated marked with 0610 antibody; 5. Treated after marking with the antibody ?-gliadin; 6. Treated after marking with the antibody HMW-GS.

[0085] FIG. 10 shows the colorimetric analysis carried out with the monoclonal antibody R5-HRP conjugated, in sections of control and after treatment seeds. 1. Sub-aleurone of the control seed; 2. Sub-aleurone of the seed after treatment according to the method of the present invention; 3. Crease of the control seed; 4. Crease of the seed after treatment according to the method of the present invention. The bars in the figure correspond to 100 ?m.

[0086] FIG. 11 shows the kinetic analysis of death of Lactobacillus acidophilus in saline solution after the addition both of the control bread and of the modified bread (0.8 g/l). The lines represent the best fit through the Weibull distribution.

[0087] FIG. 12 shows the vital count of Staphylococcus aureus in saline solution with the addition of the control bread or of the treated bread (0.2, 0.4 or 0.8 g/l). The mean values?standard deviation. The symbols * and ** identify the significant differences (one-way ANOVA and Tukey's test).

[0088] FIG. 13 shows the vital count of Salmonella sp. in saline solution with 0.8 g/l of control bread or modified bread added. The mean values?standard deviation.

[0089] FIG. 14 shows the analysis of the main components relating to the results of the SCFA and FISH after 6 hours of fermentation. Panel A) Projection of the variables; panel B) Projection of the cases. Samples: A, negative control healthy donors; B, healthy donors+control bread; C, healthy donors+modified bread; D, negative control of celiac donors; E, celiac donors+control bread; F, celiac donors+modified bread. Variables: 1, Bif164; 2, Erec482; 3, Bac; 5, lab158; AC: acetic acid; BUT: butyric acid; PROP: propionic.

[0090] FIG. 15 shows the analysis of the main components relating to the results of the SCFA and FISH after 24 hours of fermentation. Panel A) Projection of the variables; panel B) Projection of the cases. Samples: A, negative control healthy donors; B, healthy donors+control bread; C, healthy donors+modified bread; D, negative control of celiac donors; E, celiac donors+control bread; F, celiac donors+modified bread. Variables: 1, Bif164; 2, Erec482; 3, Bac; 5, lab158; AC: acetic acid; BUT: butyric acid; PROP: propionic.

[0091] FIG. 16 shows the analysis of the main components relating to the results of the SCFA and FISH after 48 hours of fermentation. Panel A) Projection of the variables; panel B) Projection of the cases. Samples: A, negative control healthy donors; B, healthy donors+control bread; C, healthy donors+modified bread; D, negative control of celiac donors; E, celiac donors+control bread; F, celiac donors+modified bread. Variables: 1, Bif164; 2, Erec482; 3, Bac; 5, lab158; AC: acetic acid; BUT: butyric acid; PROP: propionic.

[0092] FIG. 17 shows the bacteria groups recovered from the culture broth of three different vessels (V1, V2 and V3) of the model system simulating the parts of the colon before (SS1) and after (SS2) addition of (A) control bread and (B) modified bread in healthy volunteers. Results are reported as data mean of two model systems?SEM (n=2).

[0093] FIG. 18 shows the bacteria groups recovered from the culture broth of three different vessels (V1, V2 and V3) of the model system simulating the parts of the colon before (SS1) and after (SS2) addition of modified bread in (A) healthy patients and (B) celiacs. Results are reported as data mean of two model systems?SEM (n=2).

[0094] FIG. 19 shows short-chain fatty acids SCFA recovered from the culture broth of three different vessels (V1, V2 and V3) of the model system simulating the parts of the colon before (SS1) and after (SS2) addition of (A) control bread and (B) modified bread in healthy volunteers. Results are reported as data mean of two model systems?SEM (n=2).

[0095] FIG. 20 shows short-chain fatty acids SCFA recovered from the culture broth of three different vessels (V1, V2 and V3) of the model system simulating the parts of the colon before (SS1) and after (SS2) addition of modified bread in (A) healthy patients and (B) celiacs. Results are reported as data mean of two model systems?SEM (n=2).

[0096] FIG. 21, panel a) shows soft wheat sections (1 ?m) relative to the treatment steps after conjugation with 0610 antibody, which recognizes the protein fractions of gliadins and LMW-GS. (A) Step b; (B) Step c; (C) Step d; (D) Step e; (E) Step f; (F) Step g; panel b) shows the image thresholding analysis of the above steps through the use of Image J software; panel c) shows SDS-PAGE analysis of the protein fractions of gliadinsand related weight control flour (CWF) extracted in 70% EtOH. (A) gliadin from step b; (B) gliadin from step c; (C) gliadin from step d; (D) gliadin from step e; (E) gliadin from step f; (F) gliadin from step g; panel d) illustrates a summary Table of the % decrease relative to the values of MGV (Mean Grey Value) obtained from the analysis of microscopy image steps and optical density (OD) relative to SDS-PAGE expressed gliadin protein fractions.

EXAMPLES

Example 1

Reduction of the Antigenicity of the Gluten Proteins After Treatment With the Method According to the Invention

[0097] After subjecting the wheat grains to the detoxification method according to the invention, the reduction of the antigenicity of the toxic epitopes of the gluten protein on the flours was tested using the official method (ELISA assay with antibody R5) adopted by the barred ear laboratories for the recognition of gluten in flours and products intended for celiac patients.

[0098] In particular, the gluten proteins previous denatured with the Mendez cocktail were extracted from the flours by means of alcohol solution and tested according to the R5 sandwich ELISA method (see FIG. 7) using the monoclonal antibody R5 that recognizes the toxic peptide sequence QQPFP, which is found repeated in the gluten proteins. FIG. 7, (panel A) shows the summary histograms of the ELISA assay with R5 Ridascreen Gliadin, of the samples treated with the method described in the international patent application WO2014/053891 compared to the method of the present invention (panel B). Incidentally, the flour samples coming from the grains treated with the method described in the international patent application W02014/053891 exhibited a reduction in the antigenicity of the toxic epitope QQPFP in the range between 60 and 40 ppm (panel A, FIG. 7; Lamacchia C. et al. 2016). Instead, the flours coming from the grains treated with the method of the present invention show a significant reduction in the antigenicity of the toxic epitope to 13.83?7.22 ppm, enabling these flours to be considered, for all purposes of the law, gluten free flours, although the gluten is still present within them.

[0099] The improvement of the detoxification of the gluten protein from the cereal grains according to the present invention consists of the reduction in the antigenicity of the toxic epitopes of gluten to a range between 0 and 20 ppm, making them far safer for celiac patients. This reduction was not achievable through the method of the international patent application WO2014/053891 because in this method the microwave step for 120 seconds, followed by slow cooling at ambient temperature, does not assure the complete modification of the proteins to the plastic form, which instead is reached completely thanks to the steps of the method described in the present invention.

[0100] The changes induced by the method according to the present invention enable the reduction in the antigenicity of the gluten proteins so that they are no longer recognizable even by their own antibodies. To demonstrate this, three samples of control seeds (CWS) and treated seeds (TWS) were cut transversely, and examined through immunogold (FIG. 8), immunofluorescence (FIG. 9) and colorimetric microscopy (FIG. 10) using three specific monoclonal antibodies for the gliadin fraction: [0101] IFRN 0610, monoclonal antibody that recognizes an epitope (QQSF) common to many gliadins; [0102] LMW-GS, murine monoclonal antibodies, which recognizes a repetitive domain present in the fraction of ? gliadin (PEQPFPQGC); [0103] R5, monoclonal antibodies R5 recognizes the highly toxic sequence QQPFP, which is present repeated in gluten proteins.

[0104] FIG. 8 shows the sections of control wheat (Control) and after treatment (Treated) of the method according to the present invention, cut transversely, and examined through SEM-Immunogold, immunomarked with 0610 antibody and ?-gliadin. The values obtained in 5 sections coming from 5 different Control and Treated seeds were compared with Student's T-test. The differences observed with the two types of antibody were found to be highly significant (p<0.001). With the antibody 0610, a decrease of 89% was obtained in the treated seeds compared to the control seeds and a decrease of 87.5% was obtained compared to the antibody ?-gliadin.

[0105] FIG. 9 shows the sections of wheat (1 ?m) control (Control) and after treatment (Treated) according to the method of the present invention, marked with the 0610 antibody, HMW-G and ?-gliadin. Three different samples (3 sections for each seed) coming from samples of control and after treatment seed were analyzed using the Imagej software. For each image, converted into grey scale, the respective MGVs (mean grey values) were obtained. With the 0610 antibody, a decrease of 91.7% was observed in the treated seeds compared to control seeds and a decrease of 90.6% was observed compared to the antibody ?-gliadin. Thereafter, the two-way Anova test was carried out, with decomposition hypothesis for two variables (type of sample and type of treatment). Among the parameters analyzed, the type of treatment undergone by the seed was the most important factor. The data obtained were found to be highly significant (p<0.001).

[0106] FIG. 10 shows the colorimetric analysis carried out with the monoclonal antibody R5-IIRP conjugated, in sections of control and after treatment seeds. Three different samples (3 sections for each seed) coming from samples of control and after treatment seed were analyzed using the Imagej software. For each image, converted into grey scale, in binary format, the respective MGVs (mean grey values) were obtained. A decrease of 89.2% was observed in the seed after treatment compared to the control seed, at the level of the sub-aleuronic region, and a decrease of 82% was observed at the fold level. Thereafter, the two-way Anova test was carried out, with decomposition hypothesis for two variables (type of sample and type of treatment). Among the parameters analyzed, the type of treatment undergone by the seed was the most important factor. The data obtained were found to be highly significant (p<0.001).

Example 2

In Vitro Study on the Protective Effect of the Digested Bread Prepared With the Flours Treated According to the Method With Respect to Lactobacillus acidophilus and on the Antimicrobial Effect With Respect to Staphylococcus aureus and of Salmonella Typhimurium

[0107] Two different series of experiments were carried out, as shown in the following Table 1.

[0108] In particular, the aliquots of physiological solution (NaCl 0.9%) (50 ml) were supplemented with different aliquots of control bread or bread whose flour derives from the milling of the seeds whose gluten was modified with the method described above, digested in vitro in appropriate conditions according to the procedures described by Maccaferri S. et al. (2012) dehydrated and inoculated at 8 log ufc/ml; the samples were then analyzed to periodically assess the vital count by plating on MRS agar (Lactobacillus acidophilus and Bifidobacterium animalis) or TSA (pathogens) and incubated at 37? C. for 2-4 days. The lactic bacteria were analyzed in anaerobic conditions.

TABLE-US-00001 TABLE 1 Death Lactobaccilus Saline solution and samples 7 days (microorganism kinetics acidophilus with the addition of 0.4 or count every 6-10 hours) Bifidobacterium 0.8 g/l of control bread or animalis bread prepared with flour from seeds treated according to the method of the present invention Effect of the Lactobaccilus Saline solution and samples 24 hour concentration acidophilus with the addition of 0.8 or 5 Bifidobacterium g/l of control bread or bread animalis prepared with flour from seeds treated according to the method of the present invention Pathogens Salmonella Saline solution and samples 7 days (microorganism Typhimurium with the addition of 0.2, 0.4, count after 1 and 7 Staphylococcus or 0.8 g/l of control bread or days) aureus bread prepared with flour from seeds treated according to the method of the present invention

[0109] Table 2 shows the fitness parameters for the Weibull distribution for the death kinetics of Lactobacillus acidophilus (mean?standard values). For each parameter, the letters indicate the significant differences (ANOVA and Tukey's test, P<0.05). The death kinetics showed a downward curve with a shape parameter >1.

TABLE-US-00002 TABLE 2 Samples log N.sub.0* ? p d.t. R Control 8.43 ? 0.14A 17.99 ? 0.90A 1.62 ? 0.15A 67.46 ? 2.06A 0.995 0.4 g/l Detox 8.38 ? 0.13A 17.43 ? 2.06A 1.40 ? 0.14A 80.53 ? 2.03B 0.994 0.4 g/l Control 8.19 ? 0.12A 23.40 ? 2.00B 1.94 ? 0.20A 70.28 ? 2.63A 0.993 0.8 g/l Detox 8.56 ? 0.14A 17.57 ? 2.70A 1.27 ? 0.17A 93.96 ? 4.00C 0.990 0.8 g/l

[0110] The addition of the saline solution both in the control bread and in the modified bread had no impact on the shape of the curve. On the other hand, the type of bread had a significant effect on the death kinetics of the bacterial population, which was prolonged from 67.46 to 80.53 at 0.4 g/l and from 70.28 to 93.96 at 0.8 g/l when using the bread prepared with flour whose seeds were treated with the method described above.

[0111] The effect of the bread prepared with flour whose seeds were treated with the previously described method, on the death kinetics, but not on the shape parameter, is a consequence of a probable reduction in mortality in the last part of the death curve, as suggested by the death kinetics of Lactobacillus acidophilus in saline solution after the addition both of the control bread and of the treated bread (0.8 g/l) shown in FIG. 11. The lines represent the best fit through Weibull's equation. A second test was carried out to determine whether the concentration of modified bread could cause or exercise a harmful effect both on Lactobacillus acidophilus and on Bifidobacterium animalis; the saline solution was added with the quantity used for the first experiment (0.8 g/l) and with a higher concentration (5.0 g/l) to simulate a local increase of the bread due to a slow transit in the intestine. The vital count was not influenced by the concentration of the digested bread and the death kinetics showed a similar trend to the one shown in FIG. 11.

[0112] Lastly, the saline solution was inoculated with a Gram-positive or Gram-negative pathogen (Staphylococcus aureus and Salmonella Typhimurium); the results for Staphylococcus aureus are shown in FIG. 12 which shows the vital counts in saline solution with the addition of the control bread or of the treated bread (0.2, 0.4 or 0.8 g/l). The mean values?standard deviation. The symbols * and ** identify the significant differences (one-way ANOVA and Tukey's test).

[0113] A significant difference was observed for the sample with the addition of 0.8 g/l of bread prepared with flour whose seeds were treated according to the described method, which showed a lower vital count rate by 1-log compared to the sample to which the control bread was added. In the presence of Salmonella sp. a reduction of 3-log was observed in the same sample after 7 days, while the control bread determined a reduction of 1-log (FIG. 13).

Example 3

Study on the Therapeutic Effect in the Restoration of the Balance of the Intestinal Flora of the Celiac Patient in Model Systems That Simulate the Distal Part of the Colon

[0114] An assessment was made of both the control bread and of the bread prepared with flour whose seeds were treated according to the method of the present invention, in the batch fermentation cultures (model systems with controlled pH that simulate the distal part of the colon that allow to study the effect of single compounds or of fibers).

[0115] The fecal samples were obtained from three healthy human volunteers (two males, one female, aged between 30 and 38 years; BMI: 18.5-25) exempt from known metabolic and gastrointestinal diseases (e.g., diabetes, ulcerative colitis, Crohn's disease, irritable bowel syndrome, peptic ulcer and cancer). All healthy donors were administered a standard questionnaire to collect information about health condition, drug use, clinical anamnesis, and the life style before the donor was asked to provide a fecal sample. For celiac donors (two females, one male, aged between 30 and 38; BMI: 18.5-25), a written informed consent was obtained in each case and the study was approved by the Research Ethics Committee of the University of Reading, UK (UREC 15/20: donated fecal sample collection center for the in vitro model systems of the human colon). All fecal samples collected from healthy and celiac donors were collected on site, preserved in an anaerobic cabinet (10% H.sub.2-10% CO.sub.2-80% N.sub.2) and used no later than 15 minutes after collection. The samples were diluted 1:10 (w/v) in an anaerobic PBS solution (0.1 M solution of phosphate buffer, pH 7.4) and homogenized for 2 minutes. The containers for the fermentation of cultures in batch culture (280 ml) previously sterilized were filled with 45 ml of a model complex growth medium of the colon (Tejero-Sarinena S., et al., 2012).

[0116] Thereafter, the containers were connected to a bath of circulating water at 37? C. and the N.sub.2 gas lacking O.sub.2 was injected to make them anaerobic before inoculation. The pH was buffered to 6.7 and 6.9 using a pH-meter with NaOH or HCl solutions (Electrolab260; Electrolab Ltd, Tewkesbury, United Kingdom). To the culture medium were then added 5 ml of fecal homogenate, prepared as described above, and 1 ml of digested bread.

[0117] For each donor, 3 different containers were prepared: [0118] negative control (in which the digested bread was not added) called A for the healthy subjects and D for the celiac subjects; [0119] container with the addition of control bread called B for the healthy subjects and E for the celiac subjects; [0120] container with the addition of bread prepared with flour whose seeds were treated with the method described above called C for the healthy subjects and F for the celiac subjects.

[0121] The batch cultures were analyzed for 48 hours, drawing at the time of inoculation and after 6, 24 and 48 hours of time the samples necessary for the assessment of the microbiota through fluorescence in situ hybridization (FISH) and the determination of short chain fatty acids (SCFA) using high performance liquid chromatography (HPLC). FIGS. 14, 15 and 16 show the results of the analyses of the main components relating to the results of the SCFA and FISH after 6, 24 and 48 hours of fermentation.

[0122] The results obtained from the FISH and SCFA experiments were standardized as increase/decrease referred to tO (inoculation) of the negative control, to exclude the variability due to the type of donor; therefore, the results show the modification of the system with respect to the start of the experiment and should be read as the increase (positive value) or decrease (negative values) of the microbial population or of the products of microbial metabolism. The increase/decrease is referred to the inoculation of the negative control (log cells/ml).

[0123] In addition, each parameter was analyzed through the ANOVA test to identify the significant differences; use of the approach for homogeneous groups was applied as an additional instrument to establish a possible trend over time. Table 3 below shows the results of the one-way ANOVA test for homogeneous groups on the FISH data of bifidobacteria after 6, 24 and 48 hours of fermentation.

TABLE-US-00003 TABLE 3 Homogeneous groups Sample FISH I II III 6 hours E 0.147100 **** F 0.264242 **** D 0.489640 **** B 0.604836 **** A 0.632162 **** C 0.700706 **** 24 hours E 0.371967 **** F 0.490137 **** D 0.507300 **** A 0.716912 **** B 0.734684 **** C 0.909206 **** 48 hours E 0.273558 **** A 0.654479 **** F 0.681301 **** **** D 0.707120 **** **** B 0.715355 **** **** C 0.925907 **** Samples: A, negative control healthy donors; B, healthy donors + control bread; C, healthy donors + modified bread; D, negative controls celiac donors; E, celiac donors + control bread; F, celiac donors + modified bread

[0124] The differences between the samples were not significant either after 6 hours, or after 24 hours.

[0125] Instead, after 48 hours two statistical groups were observed: the first group consisted only of sample E (celiac donor with control bread) and the second group consisted of all other samples.

[0126] Sample E did not exhibit any significant increase in the population of bifidobacteria (increase 0.27-log cells/ml), probably due to a negative effect exercised by the bread on the microflora, while an increase from 0.7 to 0.9 log cells/ml occurred in the other samples. The interesting data resided, in fact, in the inclusion of the sample F with the samples of the healthy subject, suggesting a beneficial effect of the bread prepared with flour whose seeds had been treated with the method described above, able to restore a normal trend in the bifidobacteria population.

[0127] Tables 4 and 5 show the results of the one-way ANOVA tests for homogeneous groups on the FISH data relating to the bacterial groups Erec482 (Franks A. H. et al., 1998), Bac303 (Manz W. et al., 1996) after 6, 24 and 48 hours of fermentation (log cells/ml). The bacterial groups were identified using synthetic oligonucleotide probes intended for specific regions of 16S RNA (Langendijk P. S. et al., 1995) marked with the fluorescent dye Cy3 as reported in probeBase (http://www.microbial-ecology.net/probebase).

TABLE-US-00004 TABLE 4 Homogeneous groups Sample FISH I II III 6 hours B 0.001945 **** A 0.087913 **** C 0.130655 **** E 0.159812 **** D 0.370365 **** F 0.382277 **** 24 hours F ?0.065175 **** A 0.077214 **** **** E 0.169113 **** **** C 0.286015 **** **** D 0.443906 **** B 0.481756 **** 48 hours D ?0.017101 **** C 0.051435 **** **** F 0.064366 **** **** **** A 0.150569 **** **** **** E 0.223303 **** **** B 0.267762 **** Samples: A, negative control healthy donors; B, healthy donors + control bread; C, healthy donors + modified bread; D, negative controls celiac donors; E, celiac donors + control bread; F, celiac donors + modified bread.

TABLE-US-00005 TABLE 5 Homogeneous groups Sample FISH I II III IV 6 hour A ?0.094164 **** D ?0.043412 **** E 0.080924 **** B 0.106672 **** C 0.133569 **** F 0.176720 **** 24 hours D ?0.189282 **** E ?0.172388 **** F 0.028873 **** **** A 0.414786 **** **** B 0.433302 **** **** C 0.636738 **** 48 hours B ?0.330564 **** F ?0.313381 **** **** E ?0.307379 **** **** D ?0.193349 **** **** C ?0.110862 **** **** A ?0.034976 **** Samples: A, negative control healthy donors; B, healthy donors + control bread; C, healthy donors + modified bread; D, negative controls celiac donors; E, celiac donors + control bread; F, celiac donors + modified bread.

[0128] The statistical analysis highlighted a continuous distribution of the samples, with 2-4 superposed homogeneous groups, depending on time and on the type of microorganisms. The statistical distribution of the samples changed over time; however, the increase/decrease in the vital count (-0-33-0.26 log cells/ml) were of moderate size in absolute values. The effects of the addition of bread on the bacterial groups Chis150 (Franks A. H. et al., 1998) is shown in Table 6 below.

TABLE-US-00006 TABLE 6 Homogeneous groups Sample FISH I II III 6 hours B ?0.266858 **** A ?0.180315 **** C ?0.103523 **** **** D 0.153936 **** **** F 0.171644 **** **** E 0.316956 **** 24 ore C ?0.162934 **** F ?0.120933 **** A ?0.083551 **** B ?0.030945 **** E 0.072539 **** D 0.096110 **** 48 ore A ?0.305986 **** C ?0.234457 **** B 0.060428 **** D 0.166901 **** F 0.190838 **** E 0.216414 **** Samples: A, negative control healthy donors; B, healthy donors + control bread; C, healthy donors + modified bread; D, negative controls celiac donors; E, celiac donors + control bread; F, celiac donors + modified bread.

[0129] After 6 hours, a continuous distribution of the samples was observed with 2 well defined groups (1.sup.st group with the A and B samples; 2.sup.nd group containing the sample E) and an intermediate class (samples C, D, F). Lastly, the sample E (celiac donor with control bread) was not statistically different from the samples D and F (negative control and celiac donor with modified bread) also statistically different from the samples of healthy donors. However, in the samples F and D a statistical shift towards the sample C was observed. This change was not observed after 24 and 48 hours. The lactic bacteria exhibited a characteristic trend over time, as shown in Table 7 below, which illustrates the results of the one-way ANOVA test for homogeneous groups on the FISH data of Lab 158 after 6, 24 and 48 hours of fermentation (log cells/ml).

TABLE-US-00007 TABLE 7 Homogeneous groups Sample FISH I II III 6 hours F ?0.639714 **** E ?0.565338 **** D ?0.327822 **** **** C ?0.122414 **** A 0.001010 **** B 0.038547 **** 24 hours E ?0.591904 **** F 0.014006 **** D 0.015039 **** A 0.165791 **** C 0.267343 **** B 0.288811 **** 48 hours E ?0.526397 **** D ?0.289074 **** **** F ?0.022714 **** **** **** A 0.135032 **** **** B 0.188054 **** **** C 0.304061 **** Samples: A, negative control healthy donors; B, healthy donors + control bread; C, healthy donors + modified bread; D, negative controls celiac donors; E, celiac donors + control bread; F, celiac donors + modified bread.

[0130] After 6 hours of fermentation, a decrease was observed in the lactic population in the samples E and F (0.57-0.64 log cells/ml). After 24 hours, this negative trend was observed in the sample E, but not in the sample F, in which the lactic population increased and showed a similar trend to that of the healthy subjects, suggesting an interesting and beneficial effect of the bread prepared with flour whose gluten proteins were modified.

[0131] After 48 hours, their distribution was continuous; the sample F, in particular, was positioned in an intermediate group between the healthy subjects and the sample E.

[0132] The statistical results for the bacterial groups Eu (Eub338 I, Eub338 II, Eub338 III (used together) (Daims H. et al., 1999), showed a constant distribution, without significant differences between the different samples.

[0133] Table 8 below shows the results of the one-way ANOVA test for homogeneous groups on the FISH data of Eu after 6, 24 and 48 hours of fermentation (log cells/ml).

TABLE-US-00008 TABLE 8 Homogeneous groups Sample FISH I II 6 hours B ?0.132923 **** A ?0.061032 **** C 0.056311 **** F 0.238798 **** D 0.336604 **** E 0.435467 **** 24 hours A 0.274960 **** B 0.488056 **** **** C 0.496600 **** **** F 0.599021 **** **** D 0.720836 **** E 0.825880 **** 48 hours C ?0.197966 **** B ?0.005315 **** **** A 0.102244 **** **** D 0.265949 **** **** F 0.345989 **** E 0.434101 **** Samples: A, negative control healthy donors; B, healthy donors + control bread; C, healthy donors + modified bread; D, negative controls celiac donors; E, celiac donors + control bread; F, celiac donors + modified bread.

[0134] The same approach was used to analyses the results of the SCFA (short chain fatty acids). SCFAs generally showed a discrete distribution of the results with well-defined statistical groups and significant differences. The results are illustrated in Tables 9, 10 and 11 below.

[0135] Table 9 shows the one-way ANOVA test for homogeneous groups of butyric acid after 24 and 48 hours; the mean increase compared to the negative control is indicated (mM).

TABLE-US-00009 TABLE 9 Homogeneous groups Sample SCFA I II III IV V VI 24 hours C 43.3736 **** A 45.3854 **** D 52.3644 **** F 52.3767 **** E 62.2977 **** B 174.0981 **** 48 hours A 43.8577 **** F 52.9641 **** E 57.6583 **** D 61.8410 **** C 62.4645 **** B 258.4700 **** Samples: A, negative control healthy donors; B, healthy donors + control bread; C, healthy donors + modified bread; D, negative controls celiac donors; E, celiac donors + control bread; F, celiac donors + modified bread.

[0136] Table 10 shows the one-way ANOVA test for homogeneous groups of propionic acid after 24 and 48 hours; the mean increase compared to the negative control is indicated (mM).

[0137] With regard to the propionic acid, the increase was small in the samples of healthy donors (both after 24 hours and 48 hours), while its concentration increased by 23-37 mM in the samples of the celiac donors.

TABLE-US-00010 TABLE 10 Homogeneous groups Sample SCFA I II III IV V VI 24 hours B ?4.30662 **** A ?0.75728 **** C 0.96521 **** D 23.15887 **** E 31.18544 **** F 32.70900 **** 48 hours A 0.39286 **** B 1.71661 **** C 4.68863 **** D 22.58833 **** F 37.13872 **** E 37.45659 **** Samples: A, negative control healthy donors; B, healthy donors + control bread; C, healthy donors + modified bread; D, negative controls celiac donors; E, celiac donors + control bread; F, celiac donors + modified bread.

[0138] Table 11 shows the one-way ANOVA test for homogeneous groups of butyric acid after 24 and 48 hours; the mean increase compared to the negative control is indicated (mM).

TABLE-US-00011 TABLE 11 Homogeneous groups Sample SCFA I II III IV V VI 24 hours C ?3.63255 **** B ?0.38556 **** A 2.43478 **** F 4.07283 **** E 7.59356 **** D 17.42258 **** 48 hours A 2.24822 **** B 4.01809 **** F 4.27836 **** C 6.58688 **** E 10.66568 **** D 15.03661 **** Samples: A, negative control healthy donors; B, healthy donors + control bread; C, healthy donors + modified bread; D, negative controls celiac donors; E, celiac donors + control bread; F, celiac donors + modified bread.

[0139] After 24 hours, butyric acid increased by 17 mM in the negative control D, which recorded the greatest increase, followed by the other two samples of the celiac donors (respectively E, 7.6 mM and F, 4.1 mM); the results after 48 hours showed an interesting trend, inasmuch as the sample F showed a similar profile to the samples of the healthy donors, with a net increase of 4.28 mM in butyric acid.

[0140] FIGS. 15, 16 and 17 show the study of the global differences relating to the batch cultures coming from the healthy and celiac donors through an analysis of the main components; the results of the SCFA and of the FISH were all used as inputs; for every time of analysis, a different analysis was carried out.

[0141] After 6 hours, two statistical groups could be identified in the multifactorial space: the first one consisted of the samples from the healthy subjects (A, B and C) and the second one of the samples E and F.

[0142] The negative control of the celiac donors (sample D) was situated in a different region of the space (FIG. 14).

[0143] After 24 hours, the distribution of the space changed drastically (FIG. 15); the group coming from the healthy donors divided into two sub-groups, because the sample B shifted to a different region of the space, but the interesting result involved the sample F which shifted from the factorial region occupied by the samples from the celiac subjects and moved towards the region of the samples of the healthy subjects A and C. A similar effect was observed after 48 hours (FIG. 16).

Example 4

Study on the Therapeutic Effect in the Improvement of the Composition and Metabolism of Gut Flora in Healthy Patients and Celiacs in Model Systems Simulating the Proximal, Transverse and Distal Part of the Colon

[0144] The effect of control and modified bread whose grains have been treated according to the gluten epitopes detoxification method of the invention has been evaluated, in a three-steps continuos fermentation culture simulating the proximal, transverse and distal part of human colon (vessel 1, 2 and 3, respectively).

[0145] Foecal samples were obtained from two healthy and two celiacs volunteers (men and women of age between 30 and 50 yrs; BMI: 18,5-25) without known metabolic or gastrointenstinal diseases (such as diabetes, ulcerative colitis, Crohn's disease, irritable colon syndrome, peptic ulcer and cancer) who did not take any probiotic or prebiotic supplement, and antibiotics 6 months before the graft of the foecal sample.

[0146] A standard questionnaire has been submitted to the healthy donors to collect information on healthy status, drug consumption, case history, and life style before requiring the foecal sample. The study has been approved by The University of Roehampton Research Ethics Committee (UREC 15/20).

[0147] Foecal samples have been stored in anaerobic jar (AnaeroJar? 2.5 L, Oxoid Ltd) including a gas regeneration kit (AnaeroGen?, Oxoid) in order to reproduce anaerobic conditions inside the room. A 20 g aliquot of each sample has been diluted in 100 ml of anaerobic PBS solution (0.1 M phosphate solution, pH 7.4, w/w) and homogenized for 2 minutes (Stomacher 400, Seward, West Sussex, UK).

[0148] Samples have been added to the anaerobic fermenters within 15 minutes from their preparation. Physical-chemical colon conditions have been repeated in a three-step continuous system made by three glass fermenters with increasing volume and serially connected. For the first time in this study a small scale version of the system validated by Macfarlane et al. (1998) has been used, wherein the proximal part of the colon was represented by vessel 1 (V1, 80 ml), the transverse part of the colon by vessel 2 (V2, 100 ml), and the distal part by vessel 3 (V3, 120 ml) inoculated with 20% (w/v) of foecal homogenate of healthy and celiacs volunteers in a growth medium. The growth medium contained the following ingredients: starch, 5 g/l; mucin, 4 g/l; casein, 3 g/l; peptone water, 5 g/l; tryptone water, 5 g/l; biliary salts, 0.4 g/L; yeast extract, 4.5 g/l; FeSO.sub.4, 0.005 g/l; NaCl, 4.5 g/l; KCl, 4.5 g/l; KH.sub.2PO.sub.4, 0.5 g/l; MgSO.sub.4?7H.sub.2O, 1.25 g/l; CaCl.sub.2?6H.sub.2O, 0.15 g/l; NaHCO.sub.3, 1.5 g/l; Tween 80, 1 mL; hemin, 0.05 g/l; and cysteine HCl, 0.8 g/l.

[0149] Following inoculum, bacterial population have been stabilized as batch culture for 24 hrs. After 24 hrs (T0), the model system runs for 8 complete volume rounds to enable the achievement of the steady state (SS1) (verified through the stabilization of SCFA profiles (+/?5%).

[0150] Keeping in mind the working volume (300 ml) and retention time (48 hrs, flow rate 6.25 ml/hr) of the model system, in vitro digested control or modified bread (3.75 ml) in suitable conditions according to the procedure disclosed by Maccaferri S. et al. (2012) have been daily added in vessel V1. The bread has been added to the system for further 8 complete volume rounds until the achievement of the steady state 2 (SS2).

[0151] 4.5 mL aliquots have been removed and analyzed at SS1 (day T.sub.0) and SS2 (day T.sub.30).

[0152] Changes in the bacterial compositions of the model system simulating the three parts of the colon have been evaluated through FISH analysis (FIGS. 17 and 18) whilst changes of the microflora metabolism have been evaluated by the determination of short chain fatty acids (SCFA) (FIGS. 19 and 20) by high performance liquid chromatography (HPLC).

[0153] Results of the effect of control bread on healthy volunteers depicted in FIG. 17 showed a meaningful decrease in the number of Lactobacillus/Enterococcus spp. (detected by the probe Lab158) (vessel V1 and V2), Bacteroides-Prevotella group (V2) (detected by the probe Bac303) and Clostridium clusters XIVa+b (V1) (detected by the probe Erec482).

[0154] A total bacteria decrease trend has been observed in all the steps of the model system also if such differences have not been resulted as significant. Then, the control bread had not positive impact on the modulation and composition of foecal microflora.

[0155] Instead, the administration of the bread treated according to the method of the invention, led to a significant increase of bifidobacteria (detected by the probe Bif164) both in celiacs and healthy volunteers.

[0156] Particularly, in celiacs subjects a significant increase of bifidobacteria from 8.42 to 8.90 Log CFU/ml has been observed (P<0.05) in the second step of model system (vessel 2) and from 8.60 to 9.20 Log CFU/ml (P<0.05) in vessel 3, respectively.

[0157] In healthy subjects, a significant increase in the number of bifidobacteria from 7.90 to 8.40 Log CFU/ml (P<0.05) has been observed in vessel 3 (FIG. 18).

[0158] Furthermore, in celiacs volunteers it has been observed a significant increase of the Clostridium cluster from 8.85 to 9.50 Log CFU/ml (P<0.05); from 9.1 to 9.60 Log CFU/ml (P<0.01) and from 9 to 9.50 Log CFU/ml (P<0.05) in all vessels, respectively.

[0159] The general trend of the enhancement in all the bacterial groups and in all vessels has been detected in both healthy and celiacs subjects, without any significant differences.

[0160] SCFAs have been measured by HPLC at SS1 and SS2 in all the three different vessels of the model system (FIGS. 19 and 20). The administration of control bread induces a significant decrease of the acetate (V1 and V2) and propionate (V1), and an increase of butyrate in all vessels (FIG. 19).

[0161] In healthy subjects, the fermentation of the modified bread led to a significant production of acetate from 28.80 to 22.10 mM (P<0.01) in V1, from 44.40 to 56.94 mM (P<0.01) in V2 ad from 46 to 76.50 mM (P<0.001) in V3, respectively. Furthermore, a significant increase of propionate concentration from 70.46 to 89.81 mM (P<0.05) in V1, and of butyrate concentration from 40.35 to 77.09 mM (P<0.05) in V3, has been observed. In celiacs volunteers, a significant increase of propionate levels from 45.10 to 69.20 mM (P<0.01) in vessel 1 and from 50.80 to 70.20 mM (P<0.05) in vessel 2, has been observed, respectively.

[0162] Moreover, a significant increase of acetate concentration in vessel 1 from 41.20 to 89 mM (P<0.01) has been detected (FIG. 20).

[0163] From the results, it is inferred that in vitro fermentation of the modified bread induced a modulation of the colon microbiota with an increase of the acetate and propionate concentration that, has not been observed with the control bread in healthy subjects.

[0164] The most known metabolic pathway in gastrointestinal bacteria for the production of acetate and propionate concerns the polysaccharides metabolism.

[0165] Acetate production is mainly achieved through the metabolic pathway of fructose-6-phosphate phosphoketolase by bifidobacteria, and the main production of such acid is strictly correlated with the bacteria enhancement (Miller T. L. et al., 1996).

[0166] According to Hosseini E. et al. (2011), propionate may be produced by fermentable carbohydrates through two metabolic pathways. The first one foreses succinate decarboxylation in the presence of Bacteroides fragilis and Propioni bacterium spp., while the second one foresees the metabolic pathway of acrylate, wherein pyruvate is reduced to lactate by lactate dehydrogenase in the presence of some clusters of Clostridi. During the fermentation of the modified bread a significant increase of bifidobacteria, Bacteroides and E. rectale groups has been observed.

[0167] Modified bread showed a positive modulation of the composition of the microbiota as well as an increase of SCFAs concentration in both healthy and celiac donors.

[0168] After, the fermentation of the modified bread creates a positive modulation in terms of bifidogenic effect in both healthy and celiac subjects and in terms of growth number of Clostridium XIVa+b in celiacs subjects.

[0169] Although in healthy subjects acetic and propionic acid levels were reduced in vessel 1, acetic acid levels were considerably increased in vessels V2 and V3, and butyric acid levels were increased in vessel V3. Furthermore, as for celiac subjects high concentrations of acetic and propionic acid in vessel V1, and of propionate in vessel V2 have been observed.

BIBLIOGRAPHY

[0170] Shuppan, D., Yunker, Y., Barisani, D. 2009. Gastroenterology, 137(6):1912-33.

[0171] Rossi, M. Schwartz, K. B. J. Leukoc. Biol. 87, 749-751 (2010).

[0172] S?nchez, E., Donat, E., Ribes-Koninckx, C., Fern?ndez Murga, M. L. Sanz, Y. 2013. Appl. Environ. Microbiol. 79, 5472-5479.

[0173] Wacklin, P. et al.2013. Inflamm. Bowel Dis. 19, 934-941.

[0174] Collado, M. C., Donat, E., Ribes-Koninckx, C., Calabuig, M. & Sanz, Y. 2009. J. Clin. Pathol. 62, 264-269.

[0175] Collado, M., Donat, E., Ribes-Koninckx, C., Calabuig, M. & Sanz, Y. 2008. BMC Microbiol. 8, 232.

[0176] Di Cagno, R.et al. 2011.BMC Microbiol. 11, 219

[0177] De Palma, G. et al. 2010. BMC Microbiol. 10, 63.

[0178] Sanz, Y.et al. 2007. FEMS Immunol. Med. Microbiol. 51, 562-568.

[0179] Nadal, I., Donant, E., Ribes-Koninckx, C., Calabuig, M. & Sanz, Y. 2007. J. Med. Microbiol. 56, 1669-1674.

[0180] Schippa, S. et al. 2010. BMC Microbiol. 10, 175.

[0181] Wacklin, P. et al. 2014. Am. J. Gastroenterol. 109, 1933-1941.

[0182] Cenit, M. C.,Olivares, M., Codoner-Franch. P., Sanz, Y. 2015. Nutrients, 7, 6900-6923.

[0183] Smecuol, E.; Hwang, H. J.; Sugai, E.; Corso, L.; Cher?avsky, A. C.; Bellavite, F. P.; Gonz?lez, A.; Vod?novich, F.; Moreno, M. L.; Vazquez, H. 2013. J. Clin. Gastroenterol., 47, 139-147.

[0184] Olivares, M.; Castillejo, G.; Varea, V.; Sanz, Y. 2014. Br. J. Nutr., 112, 30-40.

[0185] Klemenak, M.; Dolin?ek, J.; Langerholc, T.; Di Gioia, D. Dig. Dis. Sci., 7. 3454-3460.

[0186] Lamacchia C., Landriscina L., D'Agnello P. 2016. Food Chemistry, 197, 634-640.

[0187] WO 2014/053891

[0188] Noel T R, Parker R, Ring S G, Tatham A S. 1995. Int J Biol Macromol, 17:81

[0189] Micard, V., and Guilbert, S. 2000. International Journal of Biological Macromolecules, 27, 229-236.

[0190] Tosi, P., Gritsch, C. S., He, J., Shewry, P. R. 2011. Annals of Botany, 108, 23-25.

[0191] Lamacchia C., Di Luccia, A., Baiano, A., Gambacorta, G., La Gatta, B., Pati, S., La Notte, E. 2007. Journal of Cereal Science, 46, 58-63.

[0192] Gerrard, J. A. 2000. Trends in Food Science and Technology, 13, 391-399.

[0193] Tilley, K. A., Benjamin, R. E., Bagorogoza, K. E., Okoy-Kotber, B. M., Prakash, O., Kwen, H. 2001. 49 (5), 2627-2632.

[0194] Maccaferri S, Klinder A, Cacciatore S, Chitarrari R, Honda H, et al. 2012. Molecular Nutrition and Food Research 56, 1342-1352.

[0195] Tejero-Sarinena, S., Barlow, J., Costabile, A. Gibson, G. R. and Rowland, I. 2012 Anaerobe, vol. 18, no. 5, pp. 530-538.

[0196] Franks A H, Harmsen H J, Raangs G C, Jansen G J, Schut F, et al. 1998. Applied and Environmental Microbiology 64,3336-3345.

[0197] Manz W, Amann R, Ludwig W, Vancanneyt M, Schleifer K H. 1996. Microbiology 142, 1097-1106.

[0198] Langendijk P S, Schut F, Jansen G J, Raangs G C, Kamphuis G R, et al. 1995. Applied Environmental Microbiology 61,3069-3075.

[0199] Maccaferri S, Klinder A, Cacciatore S, Chitarrari R, Honda H, Luchinat C, Bertini I, Carnevali P, Gibson G R, Brigidi P, Costabile A. Mol Nutr Food Res 2012, 56:1342-52.

[0200] Macfarlane G T, Macfarlane S, Gibson G R. Microb Ecol 1998, 35:180-7.

[0201] Hosseini E, Grootaert C, Verstraete W, Wiele T V. Nutr Rev 2011, 69(5):245-58.

[0202] Miller T L, Wolin M J. Appl Environ Microbiol 1996, 62 (5):1589-92.