Molecular interactions in beta-lactoglobulin. X. The stoichiometry of the beta-lactoglobulin mixed tetramerization.
Explore the source record for details and available documents.
SEARCH · PubMed Health
Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
BACKGROUND: Cow's milk proteins secreted in human milk may cause cow's milk allergy (CMA) even during exclusive breast-feeding. We studied beta-lactoglobulin levels in human milk of mothers of infants with CMA. We also studied intestinal absorption of macromolecules in the same mothers to see whether it is related to the secretion of beta-lactoglobulin in human milk. METHODS: CMA was verified with oral challenge in 46 of 55 infants assessed. beta-Lactoglobulin levels were assessed in human milk from 53 of 55 mothers of the infants before (basal sample) and 1 and 2 hours after an oral cow's milk load, which was given after a 24-hour milk-free diet. beta-Lactoglobulin was determined by an ELISA with a detection limit of 0.002 microgram/L. The 6-hour urine recovery of a high-molecular-weight polyethylene glycol (PEG) 3000 was assessed after an oral load of PEG in 45 of 55 mothers. RESULTS: beta-Lactoglobulin was found in the 1- or 2-hour samples in 75% of the mothers. beta-Lactoglobulin levels were increased in the 1- or 2-hour samples as compared with the basal levels in about half of the mothers. The respective levels were decreased in one third of the mothers whose basal beta-lactoglobulin levels were higher than in the others. beta-Lactoglobulin was found in none of the three human milk samples in 15% of the mothers. After an oral load of a high-molecular-weight PEG 3000, the 6-hour urine recovery of PEG was similar in the mothers of the infants with CMA and the mothers of infants without CMA. Neither was the urinary recovery of PEG related to the beta-lactoglobulin levels in human milk. CONCLUSIONS: The results support the view that beta-lactoglobulin in human milk may contribute to, but does not alone explain, the development of CMA in breast-fed infants.
Beta-lactoglobulin, the main whey protein in bovine milk, exists in several isoforms of which the most abundant are isoforms A and B. We have previously reported the denaturation of beta-lactoglobulin A by hydrostatic pressure [Valente-Mesquita, V.L., Botelho, M.M. & Ferreira, S.T. (1998) Biophys. J. 75, 471-476]. Here, we compare the pressure stabilities of isoforms A and B. These isoforms differ by two amino-acid substitutions: Asp64 and Val118 in isoform A are replaced by glycine and alanine, respectively, in isoform B. Replacement of the buried Val118 residue by the smaller alanine side-chain is not accompanied by significant structural rearrangements of the neighbouring polypeptide chain and creates a cavity in the core of beta-lactoglobulin. Pressure denaturation experiments revealed different stabilities of the two isoforms. Standard volume changes (DeltaVunf) of - 49 +/- 8 mL.mol-1 and -75 +/- 3 mL.mol-1, and unfolding free energy changes (DeltaGunf) of 8.5 +/- 1.3 kJ.mol-1 and 11.3 +/- 0.4 kJ.mol-1 were obtained for isoforms A and B, respectively. The volume occupied by the two methyl groups of Val118 removed in the V118A substitution is approximately 40 A3 per monomer of beta-lactoglobulin, in excellent agreement with the experimentally measured difference in DeltaVunf for the two isoforms (DeltaDeltaVunf = 26 mL.mol-1, corresponding to approximately 43 A3 per monomer). Thus, the existence of a core cavity in beta-lactoglobulin B may explain its enhanced pressure sensitivity relative to beta-lactoglobulin A. beta-Lactoglobulin undergoes a reversible pH-induced conformational change around pH 7, known as the Tanford transition. We have compared the pressure denaturation of beta-lactoglobulin A at pH 7 and 8. Unfolding free energy changes of 8.5 +/- 1.3 and 8.3 +/- 0.3 kJ.mol-1 were obtained at pH 7 and 8, respectively, showing that the thermodynamic stability of beta-lactoglobulin is identical at these pH values. Interestingly, DeltaVunf was dependent on pH, and varied from -49 +/- 8 mL.mol-1 to -68 +/- 2 mL.mol-1 at pH 7 and 8, respectively. The large increase in DeltaVunf at pH 8 relative to pH 7 appears to be associated with an overall expansion of the protein structure and could explain the increased pressure sensitivity of beta-lactoglobulin at alkaline pH.
Serum concentrations of alpha-lactalbumin and beta-lactoglobulin in first pregnancy, parturition, lactation, involution, and second parturition in 37 Holstein cattle were determined and used as an index of mammary status and in predicting milk yield. During first pregnancy, serum alpha-lactalbumin increased in the last 3 mo and reached a peak at parturition (approximately 1100 ng/ml). Changes in alpha-lactalbumin could not be described by a simple exponential equation, whereas changes in serum beta-lactoglobulin were described by a single exponential from second trimester until 4 wk prepartum and reached a peak at parturition (approximately 460 ng/ml). By 2 wk after parturition, alpha-lactalbumin had dropped to approximately 140 ng/ml, and beta-lactoglobulin dropped to approximately 25 ng/ml. In late lactation, alpha-lactalbumin was approximately 70 ng/ml and beta-lactoglobulin approximately 20 ng/ml. Short-term elevations were found after cessation of milking in both alpha-lactalbumin and beta-lactoglobulin in serum. The concentrations of alpha-lactalbumin and beta-lactoglobulin at second parturition were similar to those at first parturition with no differences found between parity. Both alpha-lactalbumin and beta-lactoglobulin in serum were functionally associated with mammary growth and development. In heifers late in pregnancy, both serum concentrations of alpha-lactalbumin and beta-lactoglobulin were positively correlated with mature equivalent milk and fat yields in the subsequent lactation. Serum beta-lactoglobulin concentrations at 16 wk prepartum in heifers were highly correlated with the sum of first and second lactation milk (r = .60) and fat (r = .60) yields. The potential value of using serum beta-lactoglobulin as an index for prescreening of heifers for lactation potential is discussed.
Depending on solution conditions, beta-lactoglobulin can exist in one of its six pH-dependent structural states. We have characterized the acid and basic-induced conformational transitions between these structural states over the pH range of pH 1 to pH 13. To this end, we have employed high-precision ultrasonic and densimetric measurements coupled with fluorescence and CD spectroscopic data. Our combined spectroscopic and volumetric results have revealed five pH-induced transitions of beta-lactoglobulin between pH 1 and pH 13. The first transition starts at pH 2 and is not completed even at pH 1, our lowest experimental pH. This transition is followed by the dimer-to-monomer transition of beta-lactoglobulin between pH 2.5 and pH 4. The dimer-to-monomer transition is accompanied by decreases in volume, v degrees (-0.008(+/-0.003) cm3 x g(-1)), and adiabatic compressibility, k degrees (S) (-(0.7(+/-0.4))x10(-6) cm3 x g(-1) x bar(-1)). We interpret the observed changes in volume and compressibility associated with the dimer-to-monomer transition of beta-lactoglobulin, in conjunction with X-ray crystallographic data, as suggesting a 7 % increase in protein hydration, with the hydration changes being localized in the area of contact between the two monomeric subunits. The so-called N-to-Q transition of beta-lactoglobulin occurs between pH 4.5 and pH 6 and is accompanied by increases in volume, v degrees (0.004(+/-0.003) cm3 x g(-1)), and compressibility, k degrees (S) ((0.7(+/-0.4))x10(-6) cm3 x g(-1) x bar(-1)). The Tanford transition of beta-lactoglobulin is centered at pH 7.5 and is accompanied by a decrease in volume, v degrees (-0.006(+/-0.003) cm3 x g(-1)), and an increase in compressibility, k degrees (S) ((1.5(+/-0.5))x10(-6) cm3 x g(-1) x bar(-1)). Based on these volumetric results, we propose that the Tanford transition is accompanied by a 5 to 10 % increase in the protein hydration and a loosening of the interior packing of beta-lactoglobulin as reflected in a 12 % increase in its intrinsic compressibility. Finally, above pH 9, the protein undergoes irreversible base-induced unfolding which is accompanied by decreases in v degrees (-0.014(+/-0.003) cm3 x g(-1)) and k degrees (S) (-(7.0(+/-0.5))x10(-6) cm3 x g(-1) x bar(-1)). Combining these results with our CD spectroscopic data, we propose that, in the base-induced unfolded state of beta-lactoglobulin, only 80 % of the surface area of the fully unfolded conformation is exposed to the solvent. Thus, in so far as solvent exposure is concerned, the base-induced unfolded states of beta-lactoglobulin retains some order, with 20 % of its amino acid residues remaining solvent inaccessible.
BACKGROUND: The immunomodulating potential residing in cow's milk proteins is currently receiving increasing attention because of growing interest in functional foods and the complex problem of cow's milk allergy. One of the major cow's milk allergens, whey protein beta-lactoglobulin, has previously been shown to mediate cellular activation in both human and murine immune cells. OBJECTIVE: We examined the response to different beta-lactoglobulin preparations in naive immune cells. METHODS: Splenocytes and cells from mesenteric lymph nodes derived from BALB/c mice bred and maintained on a milk-free diet were cultured in vitro with different beta-lactoglobulin preparations. Cell proliferation, cytokine production, and increases in intracellular glutathione were used as cellular activation markers. Moreover, the effect of beta-lactoglobulin on cytokine production in murine bone-marrow-derived dendritic cells was examined. RESULTS: We observed that some commercial beta-lactoglobulin preparations induced pronounced proliferation of both spleen cells and cells from mesenteric lymph nodes; production of TNF-alpha, IL-6, IL-1beta, and IL-10; and an increased level of intracellular glutathione in spleen cell cultures. Furthermore, TNF-alpha, IL-6, IL-1beta, and IL-10 production was induced in murine bone-marrow-derived dendritic cells. Purification of beta-lactoglobulin from raw milk using nondenaturating conditions, however, revealed that the beta-lactoglobulin per se did not possess the immunomodulatory activity. Eventually, the immunostimulatory effect was found to be caused by endotoxin contamination. CONCLUSION: These results identify endotoxin as the main immunostimulatory component present in some commercial beta-lactoglobulin preparations. Moreover, the present study makes it evident that immunomodulatory effects attributed to beta-lactoglobulin need to be reassessed.
The nucleotide sequence of ovine beta-lactoglobulin mRNA has been determined by chemical sequencing of two cDNA recombinant plasmids and primer extension products. Ovine beta-lactoglobulin mRNA consists of a 540 nucleotide coding region, flanked by 39 nucleotide 5' and 206 nucleotide 3' non-coding regions including a 20 nucleotide poly A tail. The deduced 180 amino acid sequence of pre-beta-lactoglobulin is in agreement with the previously published amino acid sequence of signal peptide and mature protein. Northern blot analysis of poly A+ RNAs from the lactating mammary glands of porcine, rabbit and rat species, allowed us to identify a homologous RNA to beta-lactoglobulin mRNA solely in the porcine species. We also detected a mRNA transcript of a size similar to that of beta-lactoglobulin mRNA in hepatic poly A+ RNA from female rat liver treated by estrogens. Furthermore, we have examined the levels of beta-lactoglobulin mRNA during the functional differentiation of the mammary gland and after hormonal stimulation. During the last third of pregnancy, the expression of beta-lactoglobulin gene is significantly more elevated than that of alpha s1- or beta-casein whose mRNA levels were found to change very slightly during this period. Both beta-lactoglobulin and casein mRNAs showed a rapid response and a wide range of change in response to cortisol treatment. However, there was a significant difference in the rate at which these processes occurred, suggesting that beta-lactoglobulin gene expression is regulated independently of the casein genes.
Changes in protein secondary structure and conformation of ovalbumin and beta-lactoglobulin (15% protein w/w) were investigated by Fourier transform Raman spectroscopy and self-deconvolution. The amounts of alpha-helix, beta-sheets, random coil, and beta-turns in native beta-lactoglobulin were 15, 54, 6, and 25%, respectively, and those for ovalbumin (41, 34, 13, and 12%) compared well with published values obtained by X-ray crystallography. The proteins were heated at 90 degrees C for 30 min and high-pressure-treated at 600 MPa for 20 min. Heating increased beta-sheet structures in both proteins at the expense of alpha-helix; for beta-lactoglobulin beta-sheet structures increased from 54 to 70% and for ovalbumin, from 34 to 54%. Random coil increased from 6% in the native protein to 30% in high-pressure-treated beta-lactoglobulin. However, for ovalbumin, the contribution from beta-turns doubled in high-pressure-treated samples, with little change in random coil. Further examination of the deconvoluted amide I band in heated samples revealed several component bands. Bands at 1626 and 1682 cm(-1) for ovalbumin and at 1625 and 1680 cm(-1) for beta-lactoglobulin were observed and are associated with aggregated, intermolecular beta-sheet (beta-aggregation), indicative of heat denaturation. The band seen at 1632-1640 cm(-1) corresponded to intramolecular beta-sheet structures, whereas the band at 1625 cm(-1) is associated with exposed beta-sheets (for example, beta-strands with strong hydrogen bonding that are not part of the core of beta-sheets). In high-pressure-treated samples bands were also observed at 1628 and 1680 cm(-1) for ovalbumin and at 1626 and 1684 cm(-1) for beta-lactoglobulin, suggesting involvement of beta-sheet structures in protein aggregation. Raman bands were observed at 1665-1670 cm(-1) for ovalbumin and at 1663-1675 cm(-1) for beta-lactoglobulin due to random coil structures. The bands at 1650-1660 cm(-1) due to alpha-helices were observed in both heated and high-pressure-treated samples. In addition, in heated samples of both ovalbumin and beta-lactoglobulin, peak intensity increased for beta-sheet in the amide III region, 980-990 cm(-1), and decreased for helix structures (900-960 cm(-1)). In contrast, there was no peak at 1240 cm(-1) (amide III beta-sheet structures) in either high-pressure-treated ovalbumin or beta-lactoglobulin, suggesting that high-pressure denaturation at 600 MPa for 20 min is less extensive than heat denaturation at 90 degrees C for 30 min.
The formation of conjugates between beta-lactoglobulin and acacia gum based on electrostatic complexes formed at pH 4.2 was investigated upon dry-state incubation for up to 14 days at 60 degrees C and 79% relative humidity (RH). By means of SEC-HPLC and RP-HPLC, it was shown that the beta-lactoglobulin incubated alone was able to form polymers with molecular masses higher than 200 kDa until 50% of the initial monomeric protein disappeared after 14 days. In the presence of acacia gum at initial protein to polysaccharide weight mixing ratios of 2:1 and 1:2, only 35% of the initial beta-lactoglobulin monomers disappeared after 14 days. Using RP-HPLC, an apparent reaction order of 2 was found for the disappearance of monomeric beta-lactoglobulin both in the presence or absence of acacia gum. However, the reaction rate was faster in the absence of acacia gum. SDS-PAGE electrophoresis with silver staining confirmed the formation of beta-lactoglobulin/acacia gum conjugates. The solubility curves of the incubated beta-lactoglobulin showed a minimum around pH 4-5. By contrast, the minimum of solubility of the beta-lactoglobulin/acacia gum incubated mixtures shifted to lower pH values compared to initial mixtures. The conjugates exhibited higher foam capacity than the incubated protein as well as lower equilibrium air/water surface tension. Conjugation at ratio 1:2 led to increased interfacial viscosity (300 mN s m(-1) at 0.01 Hz) compared to beta-lactoglobulin alone (100 mN s m(-1) at 0.01 Hz), but similar interfacial elasticity (30-40 mN m(-1)). The foam capacity of the conjugates was significantly higher than that of the incubated beta-lactoglobulin as well as foam expansion and drainage time, especially at pH 5.3, i.e., higher than the pH of formation of the conjugates.
BACKGROUND: Oral tolerance against food proteins has been achieved in different animal models with use of native or moderately hydrolyzed proteins as inducers. However, native proteins remain highly allergenic, although it has been demonstrated that protein hydrolyzates and resulting peptides can lose their allergenicity. OBJECTIVE: This study was designed to evaluate the ability of beta-lactoglobulin hydrolyzate and peptides to induce oral tolerance to native beta-lactoglobulin and to identify tolerogenic beta-lactoglobulin peptides with low allergenicity. METHODS: beta-Lactoglobulin was hydrolyzed by trypsin and fractionated by ion exchange chromatography. Peptide enrichment of fractions was evaluated. Balb/c mice were fed beta-lactoglobulin hydrolyzate or fractions by single gavage at day 1. Five days later animals were challenged intraperitoneally with native beta-lactoglobulin. At day 27 delayed-type hypersensitivity was performed. Twenty-four hours later mice were bled, and intestinal contents and spleens were collected. Oral tolerance was measured by titrating specific IgE in sera and intestinal samples. Specific T-cell responses were analyzed by splenocyte proliferation. Antigenicity of hydrolyzate and fractions was evaluated by specific ELISA inhibition. RESULTS: Mice fed either beta-lactoglobulin hydrolyzate or 2 fractions of the hydrolyzate were tolerized against beta-lactoglobulin. Specific serum and intestinal IgE were suppressed. Delayed-type hypersensitivity and proliferative responses were inhibited. One tolerogenic fraction was found to be 50 times less antigenic than the total beta-lactoglobulin hydrolyzate was. CONCLUSION: These findings support the strategy of inducing oral tolerance in "at-risk" patients by means of tolerogenic cow's milk peptides or hydrolyzate.
It has been demonstrated using CD that ethanol induces important secondary structure changes of beta-lactoglobulin. CD spectra indicate that beta-lactoglobulin secondary structure, which is mainly composed of beta-strands, becomes mostly alpha-helical under the influence of the solvent polarity changes. The midpoint of beta-strand/alpha-helix transition in beta-lactoglobulin is observed at dielectric constant approximately 60 (35% ethanol; v/v). According to CD measurements, the ethanol-dependent secondary structure changes are reversible. The alkylation of lysines epsilon-NH2 in beta-lactoglobulin weakens the central beta-barrel structure, since the beta-strand/alpha-helix transition midpoint of alkylated beta-lactoglobulin is shifted to lower ethanol concentration (25% ethanol; v/v). beta-Lactoglobulin structural changes are triggering the dissociation of the beta-lactoglobulin-retinol complex as judged from complete quenching of its fluorescence in ethanol concentration greater than 30% (v/v). However, in 20% ethanol (v/v), beta-lactoglobulin still retains most of its native secondary structure as shown by CD and, in this condition, one beta-lactoglobulin molecule binds an additional second retinol molecule. This suggests that the highly populated species observed around 20% ethanol (v/v) might represent an intermediate state able to bind two molecules of retinol.
beta-Lactoglobulin-like proteins were detected in horse colostrum and normal milk using immunological techniques. In contrast to the beta-lactoglobulins sequenced so far these proteins are monomeric and genetically not homogenous. In this paper we report the first primary structure of a monomeric beta-lactoglobulin from horse colostrum. By means of an automatic liquid-phase sequenator the sequence of peptides obtained by tryptic digestion and by cyanogen bromide cleavage was determined. A limited tryptic digestion and hydrolysis with chymotrypsin provided the necessary overlapping peptides. The horse beta-lactoglobulin I consists of 162 amino acids, among these four cysteine, six methionine residues and one tryptophan residue. Homologous comparison with bovine beta-lactoglobulin A shows an unexpectedly great difference of 72 amino acids (or 44%). Thirteen of these exchanges are explained as two-point mutations. We found that the free thiol group, localized at position 121 or in equal amounts at positions 119 and 121 in bovine beta-lactoglobulin, is absent in beta-lactoglobulin I from horse colostrum. In position 121 a tyrosine substitution for cysteine was found. The amino-acid exchanges of the horse beta-lactoglobulin I as compared to the other beta-lactoglobulins are discussed.
beta-Lactoglobulins from pooled milk (Sus scrofa domestica) are isolated and characterized. The complete primary structure of the major beta-lactoglobulin component I is presented. The amino-acid sequence was elucidated by automated Edman degradation of tryptic peptides and cyanogen bromide cleavage products in a liquid phase sequencer. The tryptic and cyanogen bromide peptides were separated by reverse-phase (RP-2) or size exclusion (TSK 2000 SW) high performance liquid chromatography. Pig beta-lactoglobulin is composed of only 159 amino acids in contrast to other beta-lactoglobulins which contain 162 or 166 amino acids. Sequence alignment with previously sequenced beta-lactoglobulins was obtained by introducing two gaps at positions 115 and 151-152. Thus bovine beta-lactoglobulin A reveals 62 amino-acid substitutions. The phylogenetic distance from horse beta-lactoglobulin I and II is indicated by 49.4% and 62% amino-acid exchanges, respectively. Pig beta-lactoglobulin is a mixture of two chains with Gln or Thr at position 119. The free thiol group is localized at position 59. The structural and functional aspects of beta-lactoglobulins and its role in vitamin A (retinol) transport are discussed.
BACKGROUND: Although a number of studies have investigated the induction of oral tolerance to several proteins, relatively little is known about the induction of oral tolerance to beta-lactoglobulin, one of the major antigenic proteins in milk. OBJECTIVE: We investigated the influence of the timing of the initial beta-lactoglobulin exposure on oral tolerance induction and examined some characteristics of the tolerogenic immune response. METHODS: BALB/c mice were given beta-lactoglobulin prenatally or from the third or fifth postnatal week, bred for 17 weeks, and compared with unexposed control mice. Specific plasma anti-beta-lactoglobulin antibodies (total IgG, IgG subclasses, IgM, and IgE), antigen-specific splenocyte responses, frequencies of antibody-producing cells, and cytokine production by splenocytes, intestinal mucosal lymphocytes, and Peyer's patches were analyzed. RESULTS: Differences were observed among the 4 groups of mice in changes of plasma anti-beta-lactoglobulin antibody titers, antigen-specific T-cell proliferation, and frequencies of antibody-producing splenocytes, intestinal mucosal lymphocytes, and Peyer's patch cells after the first exposure to beta-lactoglobulin. The onset and duration of the immunologic responses were found to be dependent on the timing of antigen exposure. Prenatal exposure to antigen facilitated the induction of oral tolerance to beta-lactoglobulin, whereas delayed antigen exposure retarded tolerance. The induction of oral tolerance was associated with increased IL-4 and/or IL-10 production and decreased IL-12 production. CONCLUSION: Our results suggest that the timing of initial antigen exposure greatly influences the induction of oral tolerance to beta-lactoglobulin and that altered secretion of regulatory cytokines may be responsible for the differences in antibody production and oral tolerance induction.
The causes of the salting-in of beta-lactoglobulin by glycine and NaCl, a solubility behavior contrary to expectations, were probed by a detailed study of the interactions between these solvent components and the protein. The preferential interactions of beta-lactoglobulin with solvent components in aqueous glycine and NaCl systems have been compared with those of bovine serum albumin and lysozyme. At neutral pH, beta-lactoglobulin exhibited insignificant preferential interactions in glycine and NaCl at low cosolvent concentrations and an increasing preferential hydration at higher concentrations, the levels approaching the values expected from the other two proteins. These results indicate considerable binding of the electrolytes to beta-lactoglobulin, sufficient to compensate for the exclusion due to perturbation of the solvent surface tension. The difference between the preferential interactions of beta-lactoglobulin and the other proteins with these two solvent additives was shown to be the cause of the increase of beta-lactoglobulin solubility even at high concentrations of the additives, at which they have salting-out effects on the other proteins. The preferential interactions of NaCl with the three proteins were examined as a function of pH. The results showed no pH dependence of the preferential hydration for bovine serum albumin and lysozyme, while this parameter increased significantly for beta-lactoglobulin at lower pH. This suggests that the binding of electrolytes to beta-lactoglobulin is due to a unique charge distribution on the surface of the protein around neutral pH, which imparts to this protein a large dipole moment.
As a prelude to experimental and theoretical work on the mechanical properties of fibrillar beta-lactoglobulin gels, this paper reports the structural characterization of beta-lactoglobulin fibrils by electron and atomic force microscopy (AFM), infrared and Raman spectroscopy, and powder X-ray diffraction. Aggregates formed by incubation of beta-lactoglobulin in various alcohol-water mixtures at pH 2, and in water-trifluoroethanol (TFE) at pH 7, were found to be wormlike (approximately 7 nm in width and <500 nm in length), with a "string-of-beads" appearance. Longer (approximately 7 nm in width, and >1 microm in length), smoother, and seemingly stiffer fibrils formed on heating aqueous beta-lactoglobulin solutions at pH 2 and low ionic strength, although there was little evidence for the higher-order structures common in most amyloid-forming systems. Time-lapse AFM also revealed differences in the formation of these two fibril types: thermally induced aggregation occurring more cooperatively, in keeping with a nucleation and growth process. Only short stiff-rods (<20 nm in length) formed on heating beta-lactoglobulin at pH 7, and only complex three-dimensional "amorphous"aggregates in alcohols other than TFE at this pH. Studies of all of the pH 2 fibrils from beta-lactoglobulin, by Raman and infrared spectroscopy confirmed beta-sheet as mediating the aggregation process. Interestingly, however, some evidence for de novo helix formation for the solvent-induced systems was obtained, although it remains to be seen whether this is actually incorporated into the fibril-structure. In contrast to other amyloid systems, X-ray powder diffraction provided no evidence for extensive repeating "crystalline" structures for any of the pH 2 beta-lactoglobulin fibrils. In relation to amyloid, the lactoglobulin fibrils bear more resemblance to protofilaments than to higher-order fibril structures, these latter appearing more convincingly for thermally induced insulin fibrils (pH 2) also included in the AFM study.