Denis R. Stanworth

The incidence of IgE and IgG antibodies to chlorhexidine

IgE antibodies to the antiseptic agent chlorhexidine have recently been detected in the majority of sera from a small group of predominantly Japanese individuals showing anaphylactic‐type adverse reactions towards chlorhexidine. In this study the prevalence of IgE and IgG antibodies with specificity for chlorhexidine was investigated in groups of Japanese and British individuals. The RAST data, using a better defined semi‐chlorhexidine‐HSA antigen than previously employed, revealed that chlorhexidine‐specific IgE was only detected in Japanese individuals who had experienced anaphylactic‐type reactions and was not detected in groups of Japanese nurses and patients, or in groups of British nurses and hospital staff, all in regular contact with chlorhexidine. A group of British blood donors was also negative. In contrast, IgG antibodies were detected not only in sera from chlorhexidine‐sensitive Japanese patients, but also in several sera from Japanese nurses, non‐sensitive Japanese patients and several British individuals. The possible reasons for these observations are discussed.

Investigation of the Allergenicity of β-Lactoglobulin and Its Cleavage Fragments

Fragments of β-lactoglobulin were produced by proteolytic cleavage with trypsin, and chemical cleavage with cyanogen bromide, followed by gel filtration on Sephadex G-50. The antigenicity and allergenicity of the products were studied, before and after reductive cleavage by treatment with 2-mercaptoethanol. The former was determined by inhibition ELISA using IgG anti-β-lactoglobulin raised in rabbits, whilst the latter was determined by inhibition ELISA and mast cell challenge, using respectively the sera and peritoneal cells of rats experimentally sensitised to β-lactoglobulin. The findings raise interesting points about the structural basis of allergenicity in relation to antigenicity.

The mast cell response to substance P: Effects of neuraminidase, limulin, and some novel synthetic peptide antagonists

We have investigated certain aspects of the mechanism whereby substance P triggers secretion of 5-hydroxytryptamine (5-HT) from rat peritoneal mast cells in vitro. Substance P-induced release of 5-HT was inhibited following pretreatment of rat peritoneal cells with 0.01-1.0 units/ml neuraminidase; secretion induced by anti-IgE antibody was inhibited by pretreatment with 1.0 units/ml but not by lower concentrations of enzyme. Addition of the sialic acid-rich substances N-acetyl-neuraminlactose (up to 1.0 mM) and mucin (up to 1.0 mg/ml) to substance P in free solution failed to block the activity of the neuropeptide. Limulin, a sialic acid-specific lectin, failed to block substance P-induced secretion of 5-HT, but was found to possess intrinsic non-lytic secretory activity (at 5-20 micrograms/ml). Release of 5-HT induced by limulin was independent of that induced by substance P. A range of octapeptides incorporating the C-terminal sequence Gly-Ser-Phe-Phe, but differing in degree of cationicity and positioning of cationic residues in the four N-terminal positions, were tested for their capacity to antagonise the mast cell-triggering activity of substance P. A peptide incorporating two lysine residues at the N-terminus was found to have partial substance P antagonist activity; no effects on IgE-mediated secretion were observed.

Relevance of Milk- and Egg-Specific IgG4 in Atopic Eczema

In a substantial proportion of atopic eczema patients clinical sensitivity to food is not confirmable by either skin prick test or IgE RAST. The diagnosis of provocative food factors in this condition is, therefore, highly dependent on dietary manipulations which are time consuming and sometimes dangerous. In an attempt to find an alternative laboratory procedure we have questioned the clinical relevance of specific IgG4 measurement in the sera of these patients in the light of previous reports demonstrating raised total serum IgG4 in patients with atopic eczema. Thus, a highly reliable ELISA was developed to measure IgG4 specific to 3 cows milk proteins and 2 hens egg proteins in the sera of milk- and egg-allergic eczema patients and non-eczematous controls. The study provides clear evidence showing that the detection of milk- and egg-specific IgG4 in atopic eczema patients has no clinical value.

Essential structural requirements for triggering of mast cells by a synthetic peptide comprising a sequence in the Cε4 domain of human IgE

A range of synthetic analogues of the peptides Lys-Thr-Lys-Gly-Ser-Gly-Phe-Phe-Val-PheNH2 (human IgE epsilon-chain 497-506 decapeptide) and Lys-Thr-Lys-Gly-Ser-Gly-Phe-PheNH2 (epsilon-chain 497-504 octapeptide) were tested for activity as releasers of 5-hydroxytryptamine from rat peritoneal mast cells. The following structural modifications were found to abrogate activity: N-acetylation of the alpha-amino group of the N-terminal lysine residue; substitution of the two lysine residues by either serine or glutamine; depletion of the two C-terminal hydrophobic residues (Val-Phe) of the decapeptide; and substitution of phenylalanine by alanine in the C-terminal position of the octapeptide. These observations point to a requirement for positively charged amino acids and hydrophobic amino acids at the N- and C-terminus respectively for triggering of mast cells by these short-chain peptides. Releasing activity was also found to depend on the stereospecific conformation of the positively charged region, since substitution of L-isomeric amino acids by D-isomeric forms in the three N-terminal positions of the decapeptide led to loss of potency. Inactive analogues of the decapeptide and octapeptide, at concns up to 10(-4) M, failed to antagonise the mediator-releasing effects of the active decapeptide at concns of 3 X 10(-6) - 10(-4) M.

Regulation of human natural cytotoxicity by IgG—I. Characterization of the structural site on monomeric IgG responsible for inhibiting natural killer cell activity

Inhibition of human natural killer (NK) cell activity upon exposure of peripheral blood lymphocytes (PBL) to IgG in monomeric form (mIgG) was found to be dose-, time- and temp-dependent. PBL incubated for 2 hr at 37 degrees C in the presence of myeloma protein of a certain class or subclass had a significant reduction of their NK activity when exposed to IgG, but not to IgM or IgD, and the IgG-induced inhibition of NK cells was observed only when IgG1 or IgG3 paraproteins were used. IgG3 isolated from normal serum had a higher inhibitory property than that of total mIgG. The cytophilic activity of the IgG molecules was confined entirely to the Fc region and seemed to be localized in the CH3 domain, since human and rabbit Facb fragments had a reduced ability to inhibit NK activity. When synthetic peptides representing various sequences of the human gamma-chain were tested for inhibition of NK activity, only treatment of effector cells with a peptide comprising the sequence Tyr407-Arg416 of the CH3 domain showed a reduction of NK cell function comparable to the inhibition obtained following incubation of cells in the presence of mIgG. However, on a molar basis, this peptide was 20 times less active than mIgG. In contrast, peptides derived from sequences in the CH2 domain lacked this inhibitory capacity. Our data indicate that the structural site responsible for inhibiting NK cell activity is located in the C-terminal domain of the IgG molecule.

Use of synthetic peptides in the production and characterization of antibodies directed against predetermined specificities in rat immunoglobulin E

Two peptides, P123 and P124, representing amino acid sequences His 542-Lys 557 and Tyr 459-Arg 472, respectively, of the CH4 domain of rat IgE and predicted to be located on accessible regions of the protein were synthesized by a solid-phase procedure. Rabbits were immunized with the peptides conjugated to KLH and their antisera were tested for reactivity with free peptide and rat IgE by inhibition-ELISA. Each animal produced antibodies which reacted specifically with its immunizing peptide (titre greater than 1/62,500), but not with other synthetic peptides of similar chain-length and composition. Antisera directed against peptides P123 and P124 specifically bound purified rat IgE (IR 162) and IgE in whole myeloma serum (greater than 1/6400), but showed no reaction with normal rat serum proteins and only very low binding to purified human IgE. In addition the binding of anti-peptide sera to rat IgE could be completely inhibited with either homologous peptide or purified rat IgE, but not by other peptides or purified human IgE. Heating rat IgE for 1 hr at 56 degrees C enhanced its binding to anti-peptide antibodies by between 4- and 60-fold, but markedly reduced its reactivity with a rabbit anti-rat IgE (Fc) serum. These results suggest that antibodies directed against the synthetic peptides employed recognize and specifically bind to sites within the CH4 domain of rat IgE represented by their respective immunizing peptides. Furthermore, these antibodies are capable of detecting subtle alterations in structural conformation resulting from heating at 56 degrees C. Epitopes represented by peptides P123 and P124 may contribute to the heat-sensitive cytophilic region of rat IgE.

Affinity-purified soluble FcϵRII/CD23 derived from RPMI-8866 cells induces histamine release from human nasal polyp mast cells through a non-IgE-mediated mechanism

Soluble Fc epsilon RII/CD23 (IgE-binding factor) is released spontaneously from activated B cells and most EBV-immortalised B cell lines. We have purified soluble Fc epsilon RII/CD23 from culture supernatants of RPMI-8866 cells on an IgE Sepharose column, and studied its ability to release histamine from human nasal polyp mast cells. Soluble Fc epsilon RII/CD23 induces release of a significant amount of histamine from nasal polyp mast cells in a dose-dependent manner. IgE, and a monoclonal antibody specific for the soluble form of this receptor, were shown to neutralise this effect. It was found that soluble Fc epsilon RII/CD23 was still capable of triggering histamine release from nasal polyp mast cells from which IgE had been eluted by incubation in a low pH buffer, suggesting that a non-IgE mediated mechanism was responsible for this effect.

Changes in cellular levels of cyclic AMP in rat mast cells during secretion of histamine induced by immunoglobulin E decapeptide and ACTH(1–24) peptide: Comparison with immunological and ionophore triggers

The role of cyclic AMP in the secretory mechanism of mast cells has been investigated by comparing the time course of changes in cellular levels of this cyclic nucleotide with the kinetics of secretion induced by basic peptides, antigen, anti-IgE and calcium ionophore. ACTH(1-24) peptide and a synthetic decapeptide representative of the sequence 497-506 within the C epsilon 4 domain of human IgE induced a transient rise in cyclic AMP which reached approx. 150% of the resting levels by 10 s. Peptide-induced secretion of histamine was also rapid, reaching a maximum after 5-10 s. Immunological triggering of mast cells with antigen and anti-IgE raised levels of cyclic AMP to 150% of resting levels within 15 s, accompanying secretion of histamine which reached a maximum after 30 s. A relatively slower release of histamine induced by the calcium ionophore A23187 was paralleled by a significant reduction in cyclic AMP to 50% of the resting levels after 300 s. These data suggest a relationship between the accumulation of cyclic AMP in mast cells and secretion of histamine mediated by the C epsilon 4 decapeptide and the ACTH(1-24) peptide as well as by IgE-dependent mechanisms. However, the simultaneous increase in cyclic AMP and secretion of histamine suggests that the two events may not be causally related.

Application of synthetic peptides representative of immunoglobulin sequences to the delineation of receptor binding and signalling processes

Since the classical work of Porter (1959), which defined the Fab and Fc regions of the IgG molecule, proteolytic cleavage has been widely adopted as a basic method in the localization and characterization of immunoglobulin effector sites [see, for example, Stanworth & Stewart (1976) and Stanworth & Turner (1978)]. This approach has been extended by use of other types of protease and cyanogen bromide. which cleaves only at methionine residues and thus provides a relatively small number of fragments (as is depicted in Fig. 1. which summarizes the results of application of these methods in my own laboratory to the cleavage of IgG. in studies aimed at delineating the site of rheumatoid factor (RF) and protein A reactivity). Application of similar methods to the cleavage of human myeloma IgE enabled us to obtain the first evidence that the mast cell site is located in the Fc region of the reaginic antibody molecule (Stanworth et al., 1968). There is, however, a limitation to the extent to which this approach can be applied owing, in some instances, to the sacrifice of the active site, either as a result of its direct dissociation or as a consequence of a conformational change originating in another part of the immunoglobulin polypeptide chain. This problem has been encountered in our attempts to characterize the auto-antigenic site on the IgG molecule against which the socalled ‘general’ RF reactivity is directed. The lack of RF reactivity shown by the pF’ peptic cleavage fragment (comprising the rabbit C;.3 domain) suggested that auto-antigenic activity would be found to be associated with a C;.2 domain site. Surprisingly, however, neither the Facb nor the pF’ fragments produced by plasmin cleavage of rabbit IgG showed reactivity with rheumatoid factor (Stewart et al., 1973); nor, interestingly, with the bacterial cell wall constituent protein A (Stewart et al., 1978). Significantly, circular dichroism studies on the proteoytic cleavage products indicated an interaction between the C,.2 and C.,3 domains of rabbit IgG (Stewart et al., 1977), which has also been revealed by X-ray crystallographic analysis, which has provided important independent evidence that protein A binds to sites within both the C;.2 and C;.3 domains of the human IgG molecule (Deisenhofer et al., 1978). This would seem, therefore, to offer an explanation for our failure to further exploit successfully Fc subfragmentation procedures, in attempts to locate more precisely the positions of the autoantigenic and protein A reactive sites. As, however, will be indicated in the next section, we have gone closer to achieving this goal by resort to the alternative strategy of synthesizing peptides which are supposedly representative of sequences comprising the Fc effector sites.