Investigation of the allergenicity of beta-lactoglobulin and its cleavage fragments

Specific effects of denaturation, hydrolysis and exposure to Lactococcus lactis on bovine beta-lactoglobulin transepithelial transport, antigenicity and allergenicity

BACKGROUND

Food allergy in developed countries represents a growing concern as reflected by epidemiological studies, indicating that up to 4% of the overall population is affected. Reduction of symptoms takes place following eviction or processing of some allergens. However, it cannot be predicted which structural changes will be associated with significant effects on the allergenicity.

OBJECTIVE

To determine how various treatments of bovine beta-lactoglobulin (BLG) used as a model antigen alters its immunoreactivity and transepithelial transport, and whether this correlates with reduced allergenicity using an in vitro basophil activation assay.

METHODS

BLG was subjected to reduction/alkylation, trypsin digestion or exposed to Lactococcus lactis. The remaining immunoreactivity toward IgG raised against native BLG was assessed by ELISA. Transepithelial transport of BLG and derivatives was examined using polarized Caco-2 cell monolayers mimicking the intestinal epithelium. Selective passage of tryptic peptides was determined using colchicine and cytochalasin D. Basophil activation was measured following stimulation with BLG and derivatives.

RESULTS

Reduction/alkylation, trypsin digestion or incubation with L. lactis was associated with decreased BLG recognition by IgG antibodies raised against the native protein. All treatments also resulted in a more efficient transepithelial transport of BLG. BLG crossed the Caco-2 monolayer through passage across the cell, whereas tryptic peptides followed both the para- and transcellular routes. With the exception of denaturation by reduction/alkylation, cross-linking of IgE antibodies by BLG derivatives led to lower basophil degranulation.

CONCLUSION

In vitro dissection of antigenicity and allergenicity may be a valid and convenient alternative to evaluate the effects of biotechnological processing on dietary proteins. In addition, it can help to define the molecular and cellular mechanisms that will provide improved means of diagnosis and possibly therapy of food-allergic disorders.

Identification of epitopes within beta lactoglobulin recognised by polyclonal antibodies using phage display and PEPSCAN

Two different epitope mapping techniques were used to identify linear epitopes recognised by polyclonal IgG antibodies from rabbits immunised with bovine beta lactoglobulin (BLG), which is generally regarded as a major allergen in milk. The first, PEPSCAN, was used to investigate the binding of several rabbit polyclonal antisera to sequential overlapping peptides (12-mers) across the sequence of BLG. Each peptide was synthesized on a different polypropylene PIN, and a standard ELISA procedure was used to locate which of these peptides bound the antibodies under investigation. Comparisons of PEPSCANs for antisera from six different rabbits showed that each rabbit recognized a similar set of epitopes within BLG. PEPSCAN analysis also showed that polyclonal antibodies from the mouse recognize a set of epitopes similar to those recognized by the rabbit. The second epitope mapping technique is known as phage display and utilizes libraries of randomized short peptides fused to the coat proteins of filamentous phage as a source of epitopes for analysis. A gene VIII phage display library was used in this study with constrained nonapeptides, which were screened for epitopes recognized by affinity purified rabbit anti-BLG IgG. Immobilised rabbit anti-BLG IgG was screened in two separate experiments, each consisting of three rounds of panning. For each separate experiment, a sensitive phage ELISA was used to screen several hundred single phage clones for binding to anti-BLG IgG immobilised on microtiter plates. As a result, a number of positive phage were identified from the two separate screens of the library (19 different peptides were isolated, which resembled four different regions of BLG). The identified sequences were found to constitute a subset of the linear epitopes recognized by the PEPSCAN technique. The coordinates of the crystal structure of BLG were used to display mapped epitopes on its structure. This study has permitted detailed mapping of the major linear antigenic regions within BLG recognised by IgG antibodies from immunised rabbits and mice.

Immunogenicity of casein phosphopeptides derived from tryptic hydrolysis of beta-casein

The immunogenicity of phosphopeptides derived from tryptic hydrolysis of beta-casein (CN) was investigated in a rat model system. The titers of specific immunoglobulin (Ig)G and IgE antibodies made in response to intraperitoneal sensitization to beta-CN, casein phosphopeptides, and skim milk proteins were examined using indirect and amplified indirect ELISA, respectively. Serum IgG antibodies from rats injected with beta-CN were significantly more reactive to beta-CN, casein phosphopeptides, and skim milk proteins coated on microtiter plate wells than were the IgG antibodies generated in rats that had been subjected to other treatments. A significant difference in titers because of the time of sampling (14 or 21 d postinjection) was noted for IgE but not for IgG. Rats that were injected with casein phosphopeptides did not produce IgG antibodies that crossreacted with either skim milk proteins or beta-CN. Specific antibody levels for the IgE class rarely exceeded those of unimmunized controls. The findings suggest that immunogenicity of the phosphopeptides was reduced compared with that of native beta-CN and skim milk proteins.

Expression in Escherichia coli of Sin a 1, the major allergen from mustard

Sin a 1, the major yellow mustard allergen, is a seed storage protein that belongs to the 2S albumin family. It is composed of two disulfide-bonded polypeptide chains. The cloning of this allergen has been carried out by means of the polymerase chain reaction using non-degenerate oligonucleotides encoding the N-terminal and C-terminal regions of the mature protein as primers. Five genomic nucleotide sequences have been analyzed, encoding both mature polypeptide chains linked by the internal processed fragment. The sequence data show the existence of microheterogeneities at ten positions, demonstrating the polymorphism exhibited by the natural protein. One of the genomic clones was expressed in Escherichia coli by fusion to glutathione S-transferase from Schistosoma japonicum. The resulting chimeric protein was purified by affinity chromatography on a glutathione-Sepharose 4B matrix, and digested with thrombin to release the recombinant allergen. The recombinant Sin a 1 is recognized by rabbit polyclonal and mouse monoclonal antisera raised against natural Sin a 1, as well as by the IgE of mustard-sensitive human sera. In addition, recombinant Sin a 1 possesses a high resistance to trypsin digestion, like the native mustard allergen.

A major continuous allergenic epitope of bovine beta-lactoglobulin recognized by human IgE binding

Hexapeptides of sequential overlapping sequences of beta-lactoglobulin (BLG) were used to probe serum from children with immediate-type cow milk allergy for IgE binding to continuous epitopes of BLG in an enhanced enzyme-linked immunosorbent assay (ELISA). Six regions of IgE binding were identified on the BLG molecule and these were synthesized as dodecapeptides. Inhibition of IgE binding to whole BLG was used to confirm the BLG-specific binding of IgE to each of the synthesized peptides. One of the peptides, peptide 4, showed inhibition in an IgE anti-BLG radioimmunoassay to all 16 sera tested. The patterns of inhibition with the native BLG molecule and peptide 4 were significantly correlated (P = 0.005), suggesting that this peptide contains a major continuous IgE binding epitope of BLG.

Effect of bovine milk antigens and egg lysozyme on the binding of 59Fe-lactoferrin to platelet plasma membranes

1. Platelets bind specifically to lactoferrin. A significant similarity between human lactoferrin and some bovine milk proteins has been established. 2. Because of the structural homology of lactoferrin and cows milk proteins they are able to influence lactoferrins regulatory function on the level of its binding to membrane receptors on platelets. 3. An inhibitory effect of bovine alpha-lactalbumin and of beta-lactoglobulin on lactoferrin-receptor interaction was shown. 4. Bovine alpha-lactalbumin competes with lactoferrin for the binding sites. 5. Scatchard plot analysis of data shows one binding site for lactoferrin in the presence of alpha-lactalbumin with an affinity constant, Ka = 0.46 x 10(9) mol/l and 335 receptors/cell. 6. The inhibitory effect of beta-lactoglobulin reaches 62% and is different for the common fraction beta-lactoglobulin and the genetic variants beta-lactoglobulin A and B. 7. beta-lactoglobulin does not compete with lactoferrin for the membrane receptors. 8. Bovine casein and egg lysozyme stimulate 59Fe-lactoferrin binding to the receptors. The mechanism of these effects is still unknown. 9. Tested alimentary antigens are able to interact with lactoferrin and also with some platelet membrane structures. 10. Established changes in lactoferrin binding to the platelet membrane might be in relation to lactoferrins regulatory function and (or) eliminating mechanisms of these alimentary antigens.

An increased level of antibodies to beta-lactoglobulin is a risk determinant for early-onset type 1 (insulin-dependent) diabetes mellitus independent of islet cell antibodies and early introduction of cow's milk

Using a case-control design we have studied whether antibodies to cow's milk proteins are risk determinants for childhood-onset Type 1 (insulin-dependent) diabetes mellitus independent of early exposure to cow's milk formula and islet cell antibodies. Sera from 116 recent-onset diabetic children and 112 age- and sex-matched control children were analysed for cow's milk protein IgA, IgG and IgM antibodies, beta-lactoglobulin IgA and IgM antibodies and islet cell antibodies. The titres were compared to questionnaire data on duration of breast-feeding and introduction of formula feeding. Most antibody levels tended to be increased among diabetic compared to control children. This was statistically significant for cow's milk protein IgA antibodies (p less than 0.001) and beta-lactoglobulin IgA antibodies (p less than 0.01) as well as for islet cell antibody-positivity which was found among 92% of the diabetic and 3% of control children. The differences in cow's milk protein antibodies as well as beta-lactoglobulin antibodies were more pronounced among children with an early onset of Type 1 diabetes. Breast-feeding duration was significantly inversely related to the log of beta-lactoglobulin IgG (r = -0.16, p = 0.04) and the log of cow's milk protein IgA antibodies (r = -0.17, p less than 0.001). A positive correlation was found between formula feeding and the logarithm of beta-lactoglobulin IgG antibodies (r = 0.22, p = 0.01) and the log of cow's milk protein IgA antibodies (r = 0.16, p = 0.04).(ABSTRACT TRUNCATED AT 250 WORDS)

Milk proteins in the etiology of insulin-dependent diabetes mellitus (IDDM)

The etiology of insulin-dependent diabetes mellitus (IDDM) is multifactorial. The final cause of the disease, the specific destruction of the islet beta-cells, is the result of a cellular/humoral autoimmune process that operates in individuals with a particular genetic background in response to an external triggering factor(s). The most likely environmental triggers are virus infections and dietary factors. Among the latter group dietary proteins, mainly cow milk proteins, have been found to be important. Elimination of intact cow milk proteins from the diet significantly reduced the incidence of IDDM in the spontaneously diabetic BB rat, the elimination being most effective when it occurs during the pre-weaning period. Conversely, in newly discovered diabetics (both rats and children) increased levels of antibodies to cow milk proteins as compared with non-diabetic controls were found. These higher titres of antibodies were against beta-lactoglobulin and anti-bovine serum albumin. In further studies we found that antibodies to bovine serum albumin cross-react with a beta-cell membrane protein of Mr 69,000 and that this protein is likely induced by interferon. At the molecular level, a region of the bovine serum albumin has distinct homology to the beta-subunits of the MHC class II proteins Ia, DQ and DR, and antibodies raised against this bovine serum albumin region identified the same 69K beta cell membrane protein, in the same manner as antibodies to the third hypervariable region of DR-beta did. Our hypothesis is that bovine milk proteins (mainly bovine serum albumin) might be an important environmental factor providing specific peptides that share antigenic epitopes with host cell proteins.(ABSTRACT TRUNCATED AT 250 WORDS)

Human IgE-binding synthetic peptides of bovine beta-lactoglobulin and alpha-lactalbumin. In vitro cross-reactivity of the allergens

The allergenicity of cow's milk whey proteins, purified by high performance liquid chromatography (HPLC), was examined by the radio-allergosorbent test (RAST) against the sera of children immediately hypersensitive to milk. beta-lactoglobulin and alpha-lactalbumin bound specific IgE in the sera of 63% and 75% of these patients respectively. These allergens were tested for cross reactivity with each other by RAST inhibition. Both inhibited the binding of IgE, in the sera of allergic patients, to the other protein. Two possible determinant peptides, one from beta-lactoglobulin and one from alpha-lactalbumin, were selected by computer prediction of antigenic sites and synthesized by the fluorenylmethoxycarbonyl (FMOC)-polyamide method. The peptides were adsorbed to nitrocellulose discs and used in further RAST studies with sera from the allergic children. Both peptides bound specific IgE in the RAST assay.