University of Amsterdam

IgE antibodies specific to alpha-gal: an example of clinically relevant cross-reactive anti-carbohydrate antibodies.

IgE production to α-gal is accompanied by elevated levels of specific IgG1 antibodies and low amounts of IgE to blood group B.

IgE antibodies to gal-α-1,3-gal-β-1,4-GlcNAc (α-gal) can mediate a novel form of delayed anaphylaxis to red meat. Although IgG antibodies to α-gal (anti-α-gal or anti-Gal) are widely expressed in humans, IgE anti-α-gal is not. We explored the relationship between the IgG and IgE responses to both α-gal and the related blood group B antigen. Contradicting previous reports, antibodies to α-gal were found to be significantly less abundant in individuals with blood group B or AB. Importantly, we established a connection between IgE and IgG responses to α-gal: elevated titers of IgG anti-α-gal were found in IgE-positive subjects. In particular, proportionally more IgG1 anti-α-gal was found in IgE-positive subjects against a background of IgG2 production specific for α-gal. Thus, two types of immune response to α-gal epitopes can be distinguished: a 'typical' IgG2 response, presumably in response to gut bacteria, and an 'atypical', Th2-like response leading to IgG1 and IgE in addition to IgG2. These results suggest that IgE to a carbohydrate antigen can be formed (probably as part of a glycoprotein or glycolipid) even against a background of bacterial immune stimulation with essentially the same antigen.

Cyclic analogues of Thr6-bradykinin, N epsilon-Lys-bradykinin and endo-Lys8a-vespulakinin 1.

Syntheses are described of the endo-Lys8a-vespulakinin 1 and of cyclo-Thr6- and cyclo-N epsilon-Lys-bradykinin. The linear peptides covering the entire sequences of endo-Lys8a-VSK-1 and Thr6-BK, and the decapeptide containing all residues constituting Lys-BK, with a Arg-Lys peptide bond involving the epsilon-amino function of lysine, were prepared by the solid-phase procedure based on Fmoc chemistry. Cyclization was carried out by the diphenylphosphorazide method. The amino-terminal octapeptide sequence of vespulakinin 1, Fmoc-Thr(tBu)-Ala-Thr(tBu)-Thr(tBu)-Arg(Pmc)-Arg(Pmc)-Arg(Pmc)-Gly-OH, and its N alpha-Boc-[(Gal beta)Thr3, (Gal beta)Thr4]-analogue, were used to prepare N alpha-(1-8 VSK 1)-cyclo-N epsilon-kallidin and N alpha-[(Gal beta)Thr3, (Gal beta)Thr4, 1-8 VSK 1]-cyclo-N epsilon-kallidin. Peptides and glycopeptides were characterized by amino-acid analysis, optical rotation, analytical HPLC and FAB-MS. Consistent with previous findings, preliminary pharmacological experiments on smooth muscle preparations showed that the cyclic, or partially cyclic, analogues were significatively less potent than the linear ones.

Effect of a pmr 1 disruption and different signal sequences on the intracellular processing and secretion of Cyamopsis tetragonoloba alpha-galactosidase by Saccharomyces cerevisiae.

We fused the yeast-derived sequences encoding the invertase, acid phosphatase and alpha-factor pre- and prepro-signal peptides (SP) to the Cyamopsis tetragonoloba (guar plant) alpha-galactosidase(alpha Gal)-encoding gene and expressed these gene fusions in yeast. Whereas the amount of fusion protein produced by each of the constructs did not vary significantly, the secretion efficiency of the fusion protein that carried the SP of the prepro-alpha-factor (MF alpha 1) was consistently found to be about 10% higher than that of the other fusions (99% vs. 90%). Furthermore, when the secretion of alpha Gal was directed by the invertase (SUC2) SP, the intracellular enzyme localized to the endoplasmic reticulum (ER), whereas use of the MF alpha 1 SP caused the intracellular enzyme to be outer-chain-glycosylated and processed by the KEX2 endoproteinase, implying that it had passed the ER. These results suggest that the pro-peptide of MF alpha 1 stimulates the efflux of the heterologous protein from the ER. Null mutants of PMR1 (encoding a Ca(2+)-dependent ATPase) are known to give higher secretion efficiencies for a number of different heterologous proteins. Therefore, we also studied the secretion of alpha Gal in a pmr 1 disruption mutant. Structural analysis of the enzyme secreted by the mutant cells showed that it was completely processed by KEX2 and outer-chain-glycosylated, although the length of the outer-chain carbohydrate moiety was reduced when compared with the enzyme secreted by wild-type cells. These results contradict the hypothesis advanced by Rudolph et al. [Cell 58 (1989) 133-145] that disruption of PMR1 causes the secretory pathway to bypass the Golgi apparatus.