Interactions between proteins and polymers have been

Interaction of a-Gliadin with Poly(HEMA-co-SS): Structural Interaction of a-Gliadin with Poly(HEMA-co-SS): Structural Characterization and Biological ...
Author: Norma Quinn
1 downloads 0 Views 1MB Size
Interaction of a-Gliadin with Poly(HEMA-co-SS): Structural Interaction of a-Gliadin with Poly(HEMA-co-SS): Structural Characterization and Biological Implication Characterization and Biological Implication Li Liang,1 Maud Pinier,2 Jean-Christophe Leroux,2 Muriel Subirade1 1

Chaire de Recherche du Canada sur les Prote´ines, les Bio-syste`mes et les Aliments Fonctionnels, Institut de Recherche sur les Nutraceutiques et les Aliments Fonctionnels (INAF/STELA), Universite´ Laval, Que´bec, QC, Canada 2

Canada Research Chair in Drug Delivery, Faculty of Pharmacy, University of Montreal, C.P. 6128 Succ. Centre-ville, Montreal, QC H3C 3J7, Canada Received 26 June 2008; revised 2 October 2008; accepted 7 October 2008 Published online 29 October 2008 in Wiley InterScience ( DOI 10.1002/bip.21109


the mechanism of poly(HEMA-co-SS)/a-gliadin

The wheat gluten protein a-gliadin, a well known trigger

interaction and the polymer as a-gliadin sequestering

of celiac disease, can be complexed by random copolymers

agents in the supportive treatment of celiac disease.

of hydroxyethyl methacrylate (HEMA) and sodium 4-styrene sulfonate (SS). In this work, influence of


2008 Wiley Periodicals, Inc. Biopolymers 91: 169–178,


a-gliadin and poly(HEMA-co-SS) concentrations on

Keywords: a-gliadin; celiac disease; structural change;

a-gliadin structure was studied using spectroscopic

polymer; complex

techniques and dynamic light scattering. In 70% ethanol or 0.06M HCl (pH 1.2), a-gliadin was found to self-associate upon increasing its concentrations and displayed decreased a-helical content and increased b-turn and b-sheet contents. At pH 1.2, a-gliadin interacted with poly(HEMA-co-SS) to form supra-molecular complex particles. Poly(HEMA-co-SS) induced a-gliadin structural changes that mimicked those obtained by varying the protein concentration in pure solution. At pH 6.8, a-gliadin was poorly soluble and formed large particles but a-helix is still main secondary structure. The influence of the polymer on protein structure was weaker at neutral than acidic pH. Interaction with poly(HEMA-co-SS) disrupted a-gliadin conformation and self-association to form new complex particles at neutral pH. This study provides insight into

Correspondence to: Muriel Subirade; e-mail: [email protected] Contract grant sponsors: Fonds Que´be´cois de la Nature et de la Technologie (FQRNT) and Natural Sciences and Engineering Research Council of Canada (NSERC) C 2008 Wiley Periodicals, Inc. V

Biopolymers Volume 91 / Number 2

This article was originally published online as an accepted preprint. The ‘‘Published Online’’ date corresponds to the preprint version. You can request a copy of the preprint by emailing the Biopolymers editorial office at biopolymers@wiley. com



nteractions between proteins and polymers have been investigated widely in the context of protein separations,1,2 enzyme immobilization,3 protein structural transition,4,5 and protein encapsulation and release.6 These interactions can lead to the formation of complexes, coagula or precipitates, depending on the concentration of the protein and polymer and on solution pH and ionic strength. For interaction of a protein with an oppositely charged polymer, precipitation occurs when the positive and negative charges approach stoichiometric equivalence, while excess polymer net charge tends to stabilize protein-polymer solutions.7,8 Gluten, the major storage protein fraction of wheat and other cereals, is a mixture of glutenins and gliadins and plays a key role in human nutrition. Although ingestion of gluten does not normally elicit an immune response, it causes an immune-mediated food intolerance called celiac disease in



Liang et al.

genetically susceptible individuals with human leukocyte antigens DQ2 and DQ8. The pathology of this disease is characterized by inflammation, villous atrophy, crypt hyperplasia in the small intestine,9,10 and a broad range of symptoms such as chronic diarrhea, abdominal distension and pain, failure to thrive, weight loss, anemia, fatigue, and depression.9–11 It has been shown that a gluten sub-fraction called a-gliadin (Mw 31,000 Da12) is the predominant factor in gluten toxicity. This protein resists complete digestion into amino acids by gastric and intestinal proteases, due to the abundance and location of proline residues in its polypeptide chain. Specific proline-containing peptide fragments of a-gliadin have been identified as possible triggers for the response to gluten in celiac disease patients.11,13,14 The expression of celiac disease depends strictly on dietary exposure to wheat gluten and similar cereal proteins. Patients recover if they follow a gluten-free diet but relapse when gluten is reintroduced. The only current effective treatment for celiac disease is lifelong, strict adherence to a gluten-free diet.9 This is difficult to achieve because there are many hidden sources of gluten in processed foods. We have recently shown that a-gliadin could be complexed by random copolymers of hydroxyethyl methacrylate (HEMA) and sodium 4-styrene sulfonate (SS) (see Figure 1). SS homo-polymer destroyed on the integrity of the cytoskeleton of IEC-6 intestinal epithelial cells whereas its copolymer with HEMA had no influence on the cell-cell contacts and possessed better biocompatibility. Poly(HEMA-co-SS) could inhibit alterations in morphology and cell-cell contacts of intestinal cells and hinder immunogenic peptides formation through a-gliadin digestion by gastro-intestinal enzymes. These copolymers have a potential role as polymer binder in the treatment of celiac disease.15,16 However, the mechanisms involved in a-gliadin interactions with poly(HEMA-co-SS) are still unknown. Many studies have shown that synthetic polymers can induce the structural transitions of proteins.5,17–20 The gliadin structure contains a nonrepetitive domain rich in a-helical structure and a heterogeneous repetitive domain rich in b-reverse turns.21 It has been found that interactions of a-gliadin with polysaccharides such as dextrin and gum arabic affect its secondary structure.21,22 It is widely recognized that the structure of a protein is the result of a delicate balance between the interactions within the molecule and with its surroundings.4 The a-gliadin polypeptide sequence contains nine basic amino acid residues plus a large number of nonpolar and glutamine residues.23 They respectively provide potential for electrostatic interaction, hydrophobic interaction, and hydrogen bonding of the protein with itself and with poly(HEMA-co-SS). In this study, the complexes formed between a-gliadin and

FIGURE 1 Chemical structure of poly(HEMA-co-SS).

poly(HEMA-co-SS) were characterized at physiological pH— that is, gastric (1.2) and intestinal (6.8) pH—to gain insight into the mechanism of the interaction between these two molecular species in the gastro-intestinal tract.

MATERIALS AND METHODS Materials a-Gliadin was kindly supplied by professor Popineau from the Institut National de la Recherche Agronomique, (Nantes, France). Briefly, it was obtained by extracting crude gliadin from gluten (isolated from soft wheat flour) and then separating gliadin subgroups successively by ion exchange chromatography, size exclusion chromatography, and hydrophobic interaction chromatography.24–28 The random copolymer of HEMA and SS with Mw of 40,000 Da was synthesized by atom transfer radical polymerization, as described previously.16,29 The copolymer contains 131 SS and 99 HEMA repeat units. Pyrene (purity  99%) was purchased from Fluka (Buchs, Switzerland). Eppendorf tubes and pipette tips (Maximum Recovery) were purchased from Axygen Scientific (Union City, CA).

Sample Preparation Stock solutions of a-gliadin in 70% ethanol or 0.06M HCl (pH 1.2) were prepared at a protein concentration of 5000 mg/L. Stock solutions of poly(HEMA-co-SS) at pH 1.2 or 6.8 were also prepared at a concentration of 5000 mg/L, in 0.06M HCl and 10 mM sodium phosphate buffer, respectively. The a-gliadin concentration was adjusted by diluting stock solutions with the corresponding solvent. Mixtures of a-gliadin with poly(HEMA-co-SS) at pH 1.2 were prepared by adding undiluted acidic protein stock solution to diluted acidic polymer solution. They were held for about 2–3 h at room temperature and then stored at 48C before spectral analysis. Mixtures at pH 6.8 were prepared by adding undiluted alcoholic protein stock solution to diluted phosphate buffer polymer solution. These were always freshly prepared due to the low solubility of a-gliadin around its


Interaction of a-Gliadin with Poly(HEMA-co-SS) isoelectric point (pH 6).15 They were held for about 2–3 h at room temperature before spectral analysis.

Circular Dichroism Circular Dichroism (CD) spectra were recorded on a Jasco J-710 spectropolarimeter (Jasco, Easton, MD). The path length was 0.1 cm and the 190–250 nm region was scanned. Ellipticity was recorded at a speed of 100 nm/min, 0.2-nm resolution, 10 accumulations, and 1.0-nm bandwidth. Buffer background was subtracted from the raw spectra.

Fourier Transform Infrared Spectroscopy Infrared spectra were recorded with a Nicolet Magna 560 Spectrometer (Nicolet, Madison, WI) equipped with a mercury-cadmium-telluride (MCT) detector. The spectrometer was continuously purged with dry, carbon-dioxide-free air. A solution of 5% (w/v) a-gliadin in 70% ethanol was placed between two CaF2 windows separated with 6-lm polyethylene terephthalate film spacer. The spectrum was collected by co-adding 128 scans at a spectral resolution of 2 cm21 and was apodized with a Happ-Genzel function. The 70% ethanol background was subtracted and Fourier self-deconvolution was done with a bandwidth of 18 cm21 and resolution enhancement by a factor of 2 using the software provided with the spectrophotometer (Omnic 3.1 software). Curve fitting was performed using GRAMS/AI (version 7.02, Thermo Galactic) software with a Gaussian type function.

Turbidity Measurement Turbidity (100-%T) of a-gliadin-polymer mixtures was determined from the apparent absorbance at 600 nm measured using a HP 8453 UV-Visible spectrophotometer (Agilent Technologies, Santa Clara, CA). Shake the tubes for 1 min to completely disperse the samples before each measurement.

Steady-State Fluorescence Fluorescence emission spectra were recorded with a Cary Eclipse Fluorescence spectrophotometer (Varian, Mississauga, ON, Canada). Spectral resolution was 5 nm for both excitation and emission. Poly(HEMA-co-SS) intrinsic fluorescence emission spectra were recorded from 270 to 350 nm with an excitation wavelength of 260 nm. Pyrene, 2 3 1025 g/mL in acetone, was added (1% by volume) to a-gliadin solutions as an extrinsic probe. These samples were incubated for 24 h at room temperature and extrinsic fluorescence emission spectra were then recorded from 450 to 550 nm with an excitation wavelength of 335 nm. The intensity ratio of the first to third band (I1/I3) was calculated and served to monitor the formation of hydrophobic domains.

Dynamic Light Scattering Size measurements were performed by quasi-elastic light scattering using a Nicomp 370 Submicron Particle Sizer (Pacific Scientific Division, Hiac-Royce Instruments, Menlo Park, CA). All measurements were done at 238C and at a scattering angle of 908. The measured time correlation functions were analyzed by volume-weighted Gaussian analysis mode.



RESULTS Influence of Concentration and Solvent on a-Gliadin Structure

Before analyzing the interactions between a-gliadin and poly (HEMA-co-SS) at pH 1.2 and 6.8, the protein structure was characterized at various concentrations in the solvents (i.e., 0.06M HCl and 70% ethanol) used to prepare the stock solutions. It is known that at acidic pH far from the isoelectric point of a-gliadin (pH 6), side-chain repulsion among positively charged basic amino acid residues results in protein unfolding.22,30 Moreover, a-gliadin is poorly soluble in water at intestinal pH,15 and must be first dissolved in a waterethanol mixture.31,32 Since alcohols shield amide groups from water, ethanol could disrupt the native structure of protein and increase a-helical structure.33,34 It has been reported that addition of methanol increased the a-helical content of a-gliadin.35 Conformational changes of a-gliadin due to the two solvents were therefore first analyzed using CD spectroscopy, which is widely used to observe the secondary structure of proteins in solution.36 Figure 2 shows the far-UV CD spectra of a-gliadin at different concentrations in 70% ethanol (A) and at pH 1.2 (B). The effect of a-gliadin concentration on its secondary structure appears to be relatively independent of the solvent. At protein concentrations lower than 500 mg/L, spectra typical of a-helix conformation are seen in both solvents, characterized by two partly overlapping negative bands around 222 and 208 nm with the 222 nm band appearing as a shoulder on the stronger 208 nm band.30 These two bands, especially 208 nm, decrease with increasing protein concentration. At a-gliadin concentrations above 1000 mg/L, a new band appears between 224 and 230 nm. Further increases in the protein concentration shift this new band to a longer wavelength and decrease its ellipticity. These results suggest that a-helical content decreases and the secondary structure shifts from a-helix to another structure as protein concentration increases. However, small spectral differences were noted between the two solvents. These may be attributed to the effect of solvent-protein interactions on the protein structure as discussed above. At the lowest protein concentrations (100 and 250 mg/L), the two negative bands around 222 and 208 nm are slightly less intense in aqueous solution at pH 1.2 than in 70% ethanol. In addition, the structural transition occurs at a lower protein concentration in 70% ethanol than in the acidic aqueous solutions. At 500 mg/L in 70% ethanol, the negative band around 208 nm appears to decrease and its ellipticity is similar to that around 222 nm, whereas in the acidic solution, the band around 222 nm is still a shoulder


Liang et al.

FIGURE 3 FTIR spectrum in the amide I region of 5% (w/v) agliadin in 70% ethanol.

FIGURE 2 Far-UV CD spectra of a-gliadin at different concentrations in 70% ethanol (A) and at pH 1.2 (B).

on the 208 nm band. Furthermore, the negative band around 208 nm almost disappears at a protein concentration of 1000 mg/L in 70% ethanol, compared with 2500 mg/L in acidic solution. At the highest a-gliadin concentrations (2500 and 5000 mg/L), the far-UV CD spectra exhibit only a weak negative band around 230 nm, which does not resemble the characteristic spectra of proteins with typical secondary structures (i.e., a-helix, b-sheet, b-turn and random coil).37 However, this weak band has been observed in the far-UV CD spectra of b-turn or random-coil-rich proteins in previous studies and its ellipticity was stronger for b-turn than random coil.38,39 It has also been associated with tryptophan residues and has been suggested to represent mainly b-turns in tryptophan-rich structures.40–42 To examine a-gliadin structure at high concentrations, 5% a-gliadin in 70% ethanol was analyzed by Fourier Transform Infrared (FTIR) spectroscopy (see Figure 3). Since a-gliadin is highly soluble in water-ethanol mixtures but not in water,43 this spectroscopy was only used to investigate the

structure of a-gliadin in 70% ethanol. The amide I band was deconvoluted into its main components. The peak positions, their assignment to secondary structural elements,44,45 and the proportions of the various secondary structures are reported in Table I. The secondary structure of 5% a-gliadin contains 29% a-helix, 26% b-sheet, 3% poly-L-proline II helix, 30% b-turn, and 12% random coil. Although high content of b-structures, zero ellipticity was observed at wavelength lower than 220 nm in far-UV CD spectra of a-gliadin (see Figure 2), suggesting that the spectra were very likely distorted due to high absorbance, resulting in difficult analysis of the protein secondary structure by CD. The content of bturn is apparently higher than that of random coil. Hence, the weak band around 230 nm in the far-UV CD spectra (see Figure 2) may be attributed to b-turn at high concentrations. The bands at 1919 and 1680 cm21 (see Figure 3) have been associated with intermolecular b-sheet aggregates45–48 and their presence suggests that a-gliadin self-associates at high concentrations. One may thus hypothesize that a-helical content decreases and both b-turn and b-sheet contents increase with concentration as a result of protein self-association. Self-association of a-gliadin at pH 1.2 and the influence of protein concentration on a-gliadin structure were studied Table I Characterization of a-Gliadin Amide Secondary Structure Based on FTIR Spectrum Wave Number (cm21) 1668 1654 1639 1630 1619 1680

Secondary Structure

Content (%)

b-turn a-helix Random coil Poly-L-proline II helix b-sheet

30 29 12 3 26


Interaction of a-Gliadin with Poly(HEMA-co-SS)


to the protein self-association. a-Gliadin association induces changes in secondary structure such as decreased a-helical content and increased b-turn and b-sheet contents. However, ethanol likely enhances electrostatic interaction54 but weakens hydrophobic interaction due to its lower polarity, and weakens hydrogen bonding by competition between the alcohol OH group and the protein NH group for bonding with the amide C¼ ¼O.55 Moreover, all carboxyl groups are protonated and would no longer engage in electrostatic self-interaction of a-gliadin in aqueous solution at pH 1.2. These protein-solvent interactions may thus lead to small spectral differences of agliadin structural changes in the two solvents (see Figure 2). FIGURE 4 I1/I3 of pyrene fluorescence emission spectra as a function of a-gliadin concentration at pH 1.2.

using pyrene as an extrinsic fluorescence probe. Pyrene has been widely used to monitor the association of macromolecules because its low solubility in water allows it to transfer to hydrophobic regions once the hydrophobic association occurs, with conspicuous photophysical character changes, such as its emission spectrum I1/I3 intensity ratio.7,49,50 This ratio decreases as the environment surrounding the pyrene molecules becomes less polar. Figure 4 shows an a-gliadin concentration dependence of the pyrene I1/I3 intensity ratio in aqueous solution at pH 1.2. This ratio for pure pyrene is approximately 1.5 and decreases with increasing a-gliadin concentration. At an a-gliadin concentration of 5000 mg/L, the I1/I3 ratio drops to about 1.0, indicating that most of the pyrene molecules have transferred from water to a hydrophobic environment. Such hydrophobic domains could form through the self-assembly of a-gliadin at high concentrations. At 5000 mg/L and pH 1.2, a-gliadin solution is opaque with a turbidity of about 50% of about 0.3, suggesting the occurrence of phase separation. However, these solutions at 2500 mg/L and lower as well as alcoholic solutions at all investigated concentrations are transparent with zero turbidity. It has been reported previously that protein oligomers and insoluble aggregates form at increased concentrations.51,52 Generally, protein oligomers are precursors to the aggregates.52,53 We therefore speculate that a-gliadin may form only soluble oligomers in acidic solutions at concentrations of 2500 mg/L or less and in alcoholic solutions. As a-gliadin concentration increases, intermolecular distance in solution decreases, which may bring an increase in the protein interaction with itself. The numerous nonpolar and glutamine residues of a-gliadin likely provide a high potential for hydrophobic interaction and hydrogen bonding, respectively.23 Electrostatic interaction between the negatively and positively charged residues also likely contributes Biopolymers

Influence of Poly(HEMA-co-SS) Concentration on a-Gliadin Structure

To investigate the mechanism of a-gliadin-poly(HEMAco-SS) interaction, the influence of poly(HEMA-co-SS) on a-gliadin structure was studied at gastric and intestinal pH, 1.2 and 6.8, respectively. Figure 5 shows far-UV CD spectra

FIGURE 5 Far-UV CD spectra of 150 mg/L a-gliadin in the presence of poly(HEMA-co-SS) at pH 1.2. (A) a-gliadin alone (a) and with 1, 2, 3, 7, 13 mg/L poly(HEMA-co-SS) (b–f); (B) a-gliadin with 25, 37, 50, 70, 100, 125, and 250 mg/L poly(HEMA-co-SS) (g–m).


Liang et al.

gated above 250 mg/L because the far-UV CD spectrum became too noisy. Figure 6 shows the influence of increasing amounts poly (HEMA-co-SS) on the secondary structure of a-gliadin at pH 6.8. As the polymer concentration increases to 2 mg/L, the ellipticities around 208 and 222 nm decrease (Figure 6A) but their ratio (y208/y222) remains constant (1.4, Figure 6B). Ellipticities then increase between 3 and 13 mg/L. Within this concentration range, y208/y222 first decreases and then increases with a minimum of about 1.0 at 7 mg/L poly (HEMA-co-SS). At 25 and 50 mg/mL, the spectra are superimposable; the ellipticities around 208 and 222 nm are apparently greater than for pure a-gliadin and y208/y222 is close to that of the pure protein. No further change in the far-UV CD spectrum of a-gliadin was observed with increasing polymer concentration up to 250 mg/L (data not shown).

Influence of a-Gliadin on Poly(HEMA-co-SS) Fluorescence

FIGURE 6 A: Far-UV CD spectra of 150 mg/L a-gliadin in the presence of poly(HEMA-co-SS) at pH 6.8. (a) a-gliadin alone; (b–j) a-gliadin with 1, 2, 3, 7, 10, 13, 25, and 50 mg/L poly(HEMA-coSS). B: Ratio of ellipticities around 208 and 222 nm as a function of poly(HEMA-co-SS) concentration.

of a-gliadin at 150 mg/L and pH 1.2 with increasing amounts of poly(HEMA-co-SS). As polymer concentrations increase up to 3 mg/L, the two negative bands around 208 and 222 nm gradually flatten and the negative band around 208 nm is blue-shifted by 3 nm. At polymer concentrations between 7 and 25 mg/L, the negative bands around 208 and 222 nm practically disappear. Upon further increase of polymer concentration (Figure 5B), a-gliadin experiences a structural transition opposite to that obtained for the pure protein (Figure 2B). The negative band around 230 nm stops increasing in ellipticity and the two strong bands around 208 and 222 nm become dominant again. This implies that the a-helical content increases and b-turn and b-sheet contents decrease as polymer concentration goes from 25 to 250 mg/L. At 250 mg/L, the ellipticity around 222 nm is almost the same as for the pure protein, suggesting similar a-helical contents, whereas the ellipticity around 208 nm appears to be less than for pure protein. The influence of poly(HEMA-coSS) on the secondary structure of a-gliadin was not investi-

Vinyl aromatic polymers are known to emit fluorescence.56–58 Polystyrene fluorescence emission intensity increases with degree of sulfonation, due to the conjugation of the p electrons of the sulfonated group with the six p electrons of the phenyl ring.59 Figure 7 shows the fluorescence emission spectra at pH 1.2 and 6.8 of poly(HEMA-co-SS) and a poly (HEMA-co-SS)/a-gliadin mixture after subtraction of a-gliadin fluorescence. At acidic pH, the kmax of poly(HEMA-coSS) fluorescence is around 289 nm (curve a), which is assigned to monomer emission of SS. Upon interacting with a-gliadin, the fluorescence intensity of the polymer decreases and kmax is red-shifted by 3 nm (curve b). At this pH, a-glia-

FIGURE 7 Fluorescence emission spectra of 50 mg/L poly (HEMA-co-SS) at pH 1.2 (a) and 6.8 (c) and in presence of 150 mg/L a-gliadin, minus the fluorescence due to protein at pH 1.2 (b) and 6.8 (d).


Interaction of a-Gliadin with Poly(HEMA-co-SS)

din carries nine positive charges (one Lys 1 four His 1 four Arg),23 while each poly(HEMA-co-SS) chain contains 131 negatively charged sulfonate groups. Electrostatic interaction between the sulfonates and the positively charged amino acids may weaken the conjugation of the sulfonate p electrons with the phenyl p-electrons, decreasing polymer fluorescence. The electrostatic interaction between protein and polymer is corroborated by precipitation and the disappearance of the negative bands around 208 and 222 nm at a poly (HEMA-co-SS) concentration of 13 mg/L (Figure 5A), at which the ratio of protein positive charges and polymer negative charges is about 1.0, that is, at stoichiometric equilibrium. It has been observed in numerous studies that electrostatic interaction is important for the complexation of polyanions to proteins in the vicinity of and above the protein isoelectric point, where protein charge heterogeneous distribution plays a dominant role.60–62 At pH 6.8 (i.e., near pH 6, the isoelectric point), the net charge on a-gliadin molecules is close to zero. Figure 7 shows that the fluorescence emission spectra of pure poly(HEMA-co-SS) are almost identical at pH 1.2 and 6.8 (curves a and c). Interaction with a-gliadin at pH 6.8 (curve d) also decreases fluorescence intensity but to a lesser extent than at pH 1.2 (curve b). This may be explained in terms of decreased electrostatic interaction between the polymer and a-gliadin at pH 6.8. These findings might also explain the lower polymer concentration required to neutralize the protein positive charges and the weaker effect of the polymer on a-gliadin secondary structure at pH 6.8 (Figure 6A) than at pH 1.2 (see Figure 5).

Analysis of a-Gliadin-Poly(HEMA-co-SS) Mixtures by Turbidity The a-gliadin/poly(HEMA-co-SS) interaction was shown to occur at gastric and intestinal pH by CD and fluorescence analysis. The change in a-gliadin turbidity with increasing amounts of poly(HEMA-co-SS) was then monitored to analyze phase separation due to interaction. Figure 8 shows that pure a-gliadin solution is transparent at pH 1.2. As the poly (HEMA-co-SS) concentration increases, the turbidity gradually increases and reaches a maximum at 13 mg/L, at which the ratio of protein positive charges and polymer negative charges is at stoichiometric equilibrium. The maximal turbidity can be explained by phase separation arising from charge neutrality. Turbidity then begins to decrease and reaches about 12% at 250 mg/L, suggesting that excess negatively charged polymer re-dissolves the complex particles suspended in the solution.8,63,64 The breadth of the peak suggests that phase separation occurs over a wide concentration range around stoichiometric equilibrium. Biopolymers


FIGURE 8 Turbidity of 150 mg/L a-gliadin mixtures with increasing amount of poly(HEMA-co-SS) at pH 1.2 and 6.8.

The solubility of a-gliadin at pH 6.8 (near its isoelectric point) is about 40 mg/L.15 When diluting alcoholic a-gliadin stock solution in 10 mM phosphate buffer at pH 6.8 to a final concentration of 150 mg/L, the solution becomes turbid (55%, Figure 8). In comparison with transparent protein solution at pH 1.2, these results indicate a huge increase in the self-aggregation of a-gliadin as the pH approaches the protein isoelectric point (pH 6). At pH 6.8, with increasing poly (HEMA-co-SS) concentration, turbidity first increases and then decreases much like at pH 1.2, but with maximal absorbance appearing at 2 mg/L. a-Glaidin/poly(HEMA-co-SS) mixtures at neutral pH are less turbid than pure protein solution at polymer concentration of 3 mg/L or more, dropping to 3% at 50 mg/L. This result indicates that a-gliadin aggregation can be suppressed by interaction with poly(HEMAco-SS) at neutral pH.

Analysis of Poly(HEMA-co-SS) and a-Gliadin Complexes by Dynamic Light Scattering Dynamic light scattering has been widely used to investigate the interaction of proteins with polymers.7,65 Figure 9 shows that the apparent diameter of poly(HEMA-co-SS)-a-gliadin complex particles decreased rapidly at copolymer concentrations of 2 and 7 mg/L and pH 1.2, indicating that the particle dispersion is unstable. Precipitation was observed at 13 mg/L poly(HEMA-co-SS), which corresponds to stoichiometric equilibrium. These observations could explain the loss of negative intensity around 208 and 222 nm as the copolymer concentration increases from 0 to 13 mg/L (Figure 5A). Figure 10A shows the apparent diameter of polymer-protein complex particles at higher poly(HEMA-co-SS) concentrations at pH 1.2. The complex diameter decreases from 2800


Liang et al.

interaction has been found to occur between poly(HEMAco-SS) and a-gliadin but lesser at pH 6.8 than at acidic pH. At pH 1.2, a-gliadin self-associated upon increasing its concentrations and displayed decreased a-helical content and increased b-turn and b-sheet contents. It has been reported that adding NaCl facilitated interaction among gliadin proteins themselves and induced protein aggregation, for which electrostatic interaction of anions with specific amino acid residues may be necessary.48 The electrostatic attraction of agliadin to poly(HEMA-co-SS) may weaken intra- and/or intermolecular electrostatic repulsion among the positive charges of the protein and thus trigger the protein self-association. As the poly(HEMA-co-SS) concentration increases, FIGURE 9 The change in apparent diameter of a-gliadin-poly (HEMA-co-SS) particles over time at pH 1.2. Concentrations are 150 mg/L a-gliadin and 2 and 7 mg/L poly(HEMA-co-SS).

to 230 nm at poly(HEMA-co-SS) concentration between 25 and 250 mg/L. As polymer concentration increases, each polymer chain might interact with less protein molecules.66 Over this concentration range, the polymer negative charges are in excess. These charges may provide the stabilizing effect on the smaller complex particles. At pH 6.8, the apparent diameter of pure a-gliadin particles is about 2600 nm (Figure 10B). At the lowest concentration of poly(HEMA-co-SS), complex diameter is about 3800 nm. As the polymer concentrations increase, the diameter decreases and ends up smaller than that of pure protein particles. The diameter of complex particles is about 200 nm at 50 mg/L poly(HEMA-co-SS). These results support that poly(HEMA-co-SS) suppresses a-gliadin self-aggregation.


a-Gliadin contains numerous nonpolar and glutamine residues, which provide the potential respectively for hydrophobic interaction and hydrogen bonding with poly(HEMA-coSS). At pH 1.2, precipitation at stoichiometric equivalence of protein positive charges to polymer negative charges suggests that HEMA, highly soluble in water (up to 81% by weight),67 does not act as a stabilizer by attaching to the periphery of protein-polymer complex particles but interacts directly with a-gliadin by hydrophobic interaction and/or hydrogen bonding. Furthermore, hydrophobic interactions might occur between the polymer backbone chain and phenyl ring and the protein nonpolar residues.17,20 Hydrophobic interaction and hydrogen bonding may be more important for poly (HEMA-co-SS)/a-gliadin interaction at pH 6.8, at which the protein net charges is almost zero. In addition, electrostatic

FIGURE 10 Apparent diameter of complexes formed by 150 mg/ L a-gliadin with different concentrations of poly(HEMA-co-SS) at pH 1.2 (A) and 6.8 (B).


Interaction of a-Gliadin with Poly(HEMA-co-SS)

the size of the protein-polymer complexes decreases (Figure 10A). The amount of bound protein per polymer chain decreases with increasing molar ratio of polymer to protein,66 resulting in weaker interactions of a-gliadin with itself. Hence, poly(HEMA-co-SS) can induce structural changes in a-gliadin that simulate those observed by varying the protein concentration. Although a-gliadin is poorly soluble in water at pH 6.8 and forms large particles, the secondary structure of these is mainly a-helix (Figure 6A). We therefore speculate that these particles form by simple agglomeration of un-dissolved protein, which may have no significant influence on a-gliadin structure. However, its interaction with poly(HEMA-co-SS) results in a-gliadin structural change (see Figure 6) and protein-polymer complexes that are smaller than pure protein (Figure 10B), indicating that this interaction disrupts the conformation and aggregated structure of pure a-gliadin to form new complex particles. The influence of a-gliadin/poly(HEMA-co-SS) interaction on protein structure is likely attributed to two factors. As discussed above, the structural changes were attributed to protein self-association induced by electrostatic attraction with poly(HEMA-co-SS). On the other hand, interactions between polymer and protein could directly affect a-gliadin structure. This might be the reason for the less negative ellipticity around 208 nm at 250 mg/L polymer relative to pure protein at pH 1.2 (see Figure 5) and the downward shift of the farUV CD curves with increasing poly(HEMA-co-SS) concentration at pH 6.8 (Figure 6A). Poly(HEMA-co-SS) and a-gliadin could interact to form supra-molecular complex particles at gastric and intestinal pH. We speculate that a-gliadin may be trapped into the core of protein-polymer particles, which are stabilized by the excess negative charges associated with the styrene sulfonate at the surface. Hence, complexation by poly(HEMA-co-SS) could prevent a-gliadin contact with and digestion into immunogenic peptides by gastro-intestinal enzymes and also inhibit the influence of a-gliadin on morphology and integrity of intestinal cells.16


a-Gliadin self-association occurs as its concentration increases, which leads to decreased a-helical content and increased b-turn and b-sheet contents in 70% ethanol and pH 1.2 aqueous solutions. At pH 1.2, the addition of poly (HEMA-co-SS) can induce structural changes in a-gliadin that mimic those observed by varying the protein concentration. Interaction with poly(HEMA-co-SS) might weaken intra- and/or intermolecular electrostatic repulsion of proBiopolymers


tein to form complex particles. At neutral pH, a-gliadin aggregates to form particles with a-helix as main secondary structure. Interaction with poly(HEMA-co-SS) interaction disrupts the protein structure to form new complex particles. This study provides insight into the mechanism of poly (HEMA-co-SS)/a-gliadin interaction and the polymer as a-gliadin sequestering agents in the supportive treatment of celiac disease. The authors thank Dr. Voyer (CREFSIP, Universite´ Laval) for access to the CD instrument.

REFERENCES 1. Fisher, R. R.; Glatz, C. E. Biotechnol Bioeng 1988, 32, 777–785. 2. Wang, Y. F.; Gao, J. Y.; Dubin, P. L. Biotechnol Prog 1996, 12, 356–362. 3. Sheldon, R. A. Adv Synth Catal 2007, 349, 1289–1307. 4. Liang, L.; Yao, P.; Jiang, M. Biomacromolecules 2005, 6, 2748– 2755. 5. Liang, L.; Yao, P.; Jiang, M. Biomacromolecules 2006, 7, 1829– 1835. 6. Harada, A.; Kataoka, K. J Am Chem Soc 1999, 121, 9241–9242. 7. Liang, L.; Yao, P.; Gong, J.; Jiang, M. Langmuir 2004, 20, 3333– 3338. 8. Carlsson, F.; Malmsten, M.; Linse, P. J Am Chem Soc 2003, 125, 3140–3149. 9. Sollid, L. M. Annu Rev Immunol 2000, 18, 53–81. 10. Alaedini, A.; Green, P. H. R. Ann Intern Med 2005, 142, 289– 298. 11. McManus, R.; Kelleher, D. N Engl J Med 2003, 348, 2573–2574. 12. Ornebro, J.; Nylander, T.; Eliasson, A. C.; Shewry, P. R.; Tatham, A. S.; Gilbert, S. M. J Cereal Sci 2003, 38, 147–156. 13. Shan, L.; Molberg, O.; Parrot, I.; Hausch, F.; Filiz, F.; Gray, G. M.; Sollid, L. M.; Khosla, C. Science 2002, 297, 2275–2279. 14. Maiuri, L.; Ciacci, C.; Ricciardelli, I.; Vacca, L.; Raia, V.; Auricchio, S.; Picard, J.; Osman, M.; Quaratino, S.; Londei, M. Lancet 2003, 362, 30–37. 15. Nasser-Edine, M.; Leroux, J. C. US Pat Appl 2005, 60/735.820. 16. Pinier, M.; Verdu, E. F.; Nasser-Eddine, M.; David, C. S.; Ve´zina, A.; Rivard, N.; Leroux, J. C. Gastroenterology, in press. 17. Gong, J.; Yao, P.; Duan, H. W.; Jiang, M.; Gu, S. H.; Chunyu, L. J. Biomacromolecules 2003, 4, 1293–1300. 18. Cleland, J. L.; Hedgepeth, C.; Wang, D. I. C. J Biol Chem 1992, 267, 13327–13334. 19. Chen, Y. J.; Huang, L. W.; Chiu, H. C.; Lin, S. C. Enzyme Microb Technol 2003, 32, 120–130. 20. Sedlak, E.; Antalik, M. Biopolymers 1998, 46, 145–154. 21. Secundo, F.; Guerrieri, N. J Agric Food Chem 2005, 53, 1757– 1764. 22. Chourpa, I.; Ducel, V.; Richard, J.; Dubois, P.; Boury, F. Biomacromolecules 2006, 7, 2616–2623. 23. Bernardin, J. E.; Kasarda, D. D.; Mecham, D. K. J Biol Chem 1967, 242, 445–450. 24. Popineau, Y.; Godon, B. Cereal Chem 1982, 59, 55–62. 25. Godon, B.; Leblanc, M. P.; Popineau, Y. Qual Plant Foods Hum Nutr 1983, 33, 161–168.


Liang et al.

26. Popineau, Y. J Cereal Sci 1985, 3, 29–38. 27. Popineau, Y.; Pineau, F. Lebenson Wiss Technol 1985, 18, 133– 135. 28. Larre, C.; Popineau, Y.; Loisel, W. J Cereal Sci 1991, 14, 231–241. 29. Leroux, J. C.; Nasser-Edine, M. WO Pat Appl 2007, WO/2007/ 053935. 30. Kasarda, D. D.; Bernardin, J. E.; Gaffield, W. Biochemistry 1968, 7, 3950–3957. 31. De Deken, R. H.; Mortier, A.; Wiame, J. M. Biochim Biophys Acta 1953, 10, 488–488. 32. Frase, J. S.; Ciclitira, P. J. World J Gastroenterol 2001, 7, 772– 776. 33. Bychkova, V. E.; Dujsekina, A. E.; Klenin, S. I.; Tiktopulo, E. I.; Uversky, V. N.; Ptitsyn, O. B. Biochemistry 1996, 35, 6058–6063. 34. Bhakuni, V. Arch Biochem Biophys 1998, 357, 274–284. 35. Blanch, E. W.; Kasarda, D. D.; Hecht, L.; Nielsen, K.; Barron, L. D. Biochemistry 2003, 42, 5665–5673. 36. Kelly, S. M.; Price, N. C. Biochim Biophys Acta 1997, 1338, 161–185. 37. Johnson, W. C., Jr. Annu Rev Biophys Chem 1988, 17, 145–166. 38. Brahms, S.; Brahms, J.; Spach, G.; Brack, A. Proc Natl Acad Sci USA 1977, 74, 3208–3212. 39. Bush, C. A.; Sarkar, S. K.; Kopple, K. D. Biochemistry 1978, 17, 4951–4954. 40. Vuilleumier, S.; Sancho, J.; Loewenthal, R.; Fersht, A. R. Biochemistry 1993, 32, 10303–10313. 41. Andrushchenko, V. V.; Vogel, H. J.; Prenner, E. J. Biochim Biophys Acta 2006, 1758, 1596–1608. 42. Ladokhin, A. S.; Selsted, M. E.; White, S. H. Biochemistry 1999, 38, 12313–12319. 43. Duclairoir, C.; Nakache, E.; Marchais, H.; Orecchioni, A. M. Colloid Polym Sci 1998, 276, 321–327. 44. Makarenko, S. P.; Trufanov, V. A.; Putilina, T. E. Russ J Plant Physiol 2002, 49, 326–331. 45. Meng, G. T.; Ma, C. Y. Int J Biol Macromol 2001, 29, 287–294. 46. Lefevre, T.; Subirade, M. Biopolymers 2000, 54, 578–586. 47. Dong, A.; Matsuura, J.; Manning, M. C.; Carpenter, J. F. Arch Biochem Biophys 1998, 355, 275–281.

48. Ukai, T.; Matsumura, Y.; Urade, R. J Agric Food Chem 2008, 56, 1122–1130. 49. Kalyanasundaram, K.; Thomas, J. K. J Am Chem Soc 1977, 99, 2039–2044. 50. Li, M.; Jiang, M.; Wu, C. J Polym Sci Part B: Polym Phys 1997, 35, 1593–1599. 51. Zhao, C.; Li, L. Z.; Wang, L.; Ji, H. W. Chin Sci Bull 2006, 51, 2581–2585. 52. Philo, J. S. Am Biotechnol Lab 2003, 21, 22–26. 53. Arakawa, T.; Philo, J. S.; Ejima, D.; Tsumoto, K.; Arisaka, F. Bioproc Intern 2006, 4, 42–49. 54. Van Oss, C. J. J Protein Chem 1989, 8, 661–668. 55. Subirade, M.; Kelly, I.; Gueguen, J.; Pezolet, M. Int J Biol Macromol 1998, 23, 241–249. 56. Morawetz, H. Science 1979, 203, 405–410. 57. Major, M. D.; Torkelson, J. M. Macromolecules 1986, 19, 2801– 2806. 58. Guenoun, P.; Lipshy, S.; Mays, J. W.; Tirrell, M. Langmuir 1996, 12, 1425–1427. 59. Li, G.; Xi, S. P.; Liu, Z. X.; Huang, Y. E. Spectrosc Spect Anal 1999, 19, 289–292. 60. Cooper, C. L.; Dubin, P. L.; Kayitmazer, A. B.; Turksen, S. Curr Opin Colloid Interface Sci 2005, 10, 52–78. 61. De Vries, R. J Chem Phys 2004, 120, 3475–3481. 62. De Vries, R.; Weinbreck, F.; de Kruif, C. G. J Chem Phys 2003, 118, 4649–4659. 63. Ball, V.; Winterhalter, M.; Schwinte, P.; Lavalle, P.; Voegel, J. C.; Schaaf, P. J Phys Chem B 2002, 106, 2357–2364. 64. Jiang, J. W.; Prausnitz, J. M. J Phys Chem B 1999, 103, 5560– 5569. 65. Xia, J. L.; Dubin, P. L.; Dautzenberg, H. Langmuir 1993, 9, 2015–2019. 66. Matsudo, T.; Ogawa, K.; Kokufuta, E. Biomacromolecules 2003, 4, 1794–1799. 67. Chu, H. H.; Lin, C. S. J Polym Res 2003, 10, 283–287.

Reviewing Editor: C. Bush


Suggest Documents