Richard Pumphrey

Computer models of the human immunoglobulins shape and segmental flexibility

At present there is interest in the design and deployment of engineered biosensor molecules. Antibodies are the most versatile of the naturally occurring biosensors and it is important to understand their mechanical properties and the ways in which they can interact with their natural ligands. Two dimensional representations are clearly inadequate, and three dimensional representations are too complicated to manipulate except as numerical abstractions in computers. Recent improvements in computer graphics allow these coordinate matrices to be seen and more easily comprehended, and interactive programs permit the modification and reassembly of molecular fragments. The models which result have distinct advantages both over those of lower resolution, and those showing every atom, which are limited to the few fragments(2-5) or mutant molecules for which the X-ray crystallographic coordinates are known. In this review Richard Pumphrey describes the shape and flexibility of immunoglobulin molecules in relation to the three dimensional structure.

Computer models of the human immunoglobulins Binding sites and molecular interactions.

In last months issue(1) Richard Pumphrey showed how low resolution computer models could be used to investigate the shape and segmental flexibility of human immunoglobulin molecules. In this article, he extends the use of these models to the study of interactions between immunoglobulins and other molecules, such as their receptors, J chain, secretory component and complement fragments. Although precise studies of molecular interaction require computation in atomic detail, there is still much to be learnt from low resolution modelling, in which atomic details are glossed over but general principles become more obvious.

Identification of peanut and hazelnut allergens by native two-dimensional gel electrophoresis.

A procedure for the native two‐dimensional electrophoresis of peanut and hazelnut proteins is described. Proteins were solubilised after acetone treatment using a combination of 3‐[(3‐cholamidopropyl)dimethylammonio]‐1‐propanesulfonate (CHAPS) and tetramethylene sulphone. These extracts were analysed by a combination of isoelectric focusing in the presence of lactose in immobilized pH gradients followed by charge shift electrophoresis. Immunoblot analysis, using sera from nut allergic patients, allowed the identification of a peanut and hazelnut allergen with identical isoelectric point and apparent molecular mass. These proteins were recovered from duplicate gels using a mixture of formic acid, acetonitrile (ACN) and isopropanol. The molecular masses for both proteins, determined by matrix assisted laser desorption/ionisation‐mass spectrometry (MALDI‐MS), were 4826 Da.

Food allergy—towards predictive testing for novel foods

The risks associated with IgE-mediated food allergy highlight the need for methods to screen for potential food allergens. Clinical and immunological tests are available for the diagnosis of food allergy to known food allergens, but this does not extend to the evaluation, or prediction of allergenicity in novel foods. This category includes foods produced using novel processes, genetically modified (GM) foods, and foods that might be used as alternatives to traditional foods. Through the collation and analysis of the protein sequences of known allergens and their epitopes, it is possible to identify related groups which correlate with observed clinical cross-reactivities. 3-D modelling extends the use of sequence data and can be used to display eptiopes on the surface of a molecule. Experimental models support sequence analysis and 3-D modelling. Observed crossreactivities can be examined by Western blots prepared from native 2-D gels of a whole food preparation (e.g. hazelnut, peanut), and common proteins identified. IgEs to novel proteins can be raised in Brown Norway rat (a high IgE responder strain), and the proteins tested in simulated digest to determine epitope stability. Using the CSL serum bank, epitope binding can be examined through the ability of an allergen to cross-link the high affinity IgE receptor and thereby release mediators using in vitro cell-based models. This range of methods, in combination with data mining, provides a variety of screening options for testing the potential of a novel food to be allergenic, which does not involve prior exposure to the consumer.

Conservation of immunoglobulin variable and joining region structure and the design of universal anti-immunoglobulin antibodies reactive with antigen-binding T cell receptors.

The immunoglobulins comprise a multigenic family of recognition molecules that occur as receptors on the surfaces of lymphocytes and as antibodies in circulation in the serum of all vertebrate species [l]. Circulating antibodies (classical immunoglobulins) are produced by lymphocytes of the B cell lineage, and T lymphocytes contain and express on their surface the products of rearranging immunoglobulin genes [2, 31. Extensive data are currently available on the protein sequence of immunoglobulins of mammalian species and on the gene sequences of A chains of the chicken [4] and of heavy chain