Volker Gnau

Protein engineering of lantibiotics

Whereas protein engineering of enzymes and structural proteins nowadays is an established research tool for studying structure-function relationships of polypeptides and for improving their properties, the engineering of posttranslationally modified peptides, such as the lantibiotics, is just coming of age. The engineering of lantibiotics is less straightforward than that of unmodified proteins, since expression systems should be developed not only for the structural genes but also for the genes encoding the biosynthetic enzymes, immunity protein and regulatory proteins. Moreover, correct posttranslational modification of specific residues could in many cases be a prerequisite for production and secretion of the active lantibiotic, which limits the number of successful mutations one can apply. This paper describes the development of expression systems for the structural lantibiotic genes for nisin A, nisin Z, gallidermin, epidermin and Pep5, and gives examples of recently produced site-directed mutants of these lantibiotics. Characterization of the mutants yielded valuable information on biosynthetic requirements for production. Moreover, regions in the lantibiotics were identified that are of crucial importance for antimicrobial activity. Eventually, this knowledge will lead to the rational design of lantibiotics optimally suited for fighting specific undesirable microorganisms. The mutants are of additional value for studies directed towards the elucidation of the mode of action of lantibiotics.

Peptide motifs of HLA-A3, -A24, and -B7 molecules as determined by pool sequencing.

As a source of HLA molecules, the mouse mastocytoma P815 transfected with the respective HLA gene was used. P815-A3 and P815-B7 cells were a gift from J. Maryanski, P815-A24 a gift from E LemonrLier (Layet et al. 1984; Maryanski et al. 1986; Pala et al. 1988). Transfectants were expanded in tissue culture with Dulbeccos modified Eagle medium/high glucose supplemented with 25 mM HEPES, 10% FCS and antibiotics. Two times 10 l0 cells were pelleted and lysed with 1% NP40 in phosphate buffered saline. HLA molecules were precipitated, using solid phase bound PA2.6 antibodies (specific for HLA-A, -B, -C; Parham and Bodmer 1978) as described (Falk et al. 1991). Precipitates were treated with trifluoroacetic acid in order to dissociate peptides from MHC molecules. The supernatant, containing the peptides, was separated by reversed phase HPLC, using Pharmacia equipment (Pharmacia, Freiburg, Germany) and a 4.0 minx250 mm C2/C18 column. The fractions known from previous experience (Falk et al. 1991) to contain peptides were pooled and sequenced by Edman degradation, using either a model 477 or 476 protein sequencer (Applied Biosystems, Weiterstadt, Germany). Interpretation of pool sequencing data was done as described earlier (Falk

Natural ligand motifs of closely related HLA-DR4 molecules predict features of rheumatoid arthritis associated peptides

Abstract Rheumatoid arthritis (RA), one of the most common autoimmune disorders, is believed to be mediated via T lymphocytes and genetic studies have shown that it is strongly associated with HLA-DR4. The DR4 subtypes DR4Dw4, DR4Dw14 and DR4Dw15 represent increased risk factors for RA, whereas DR4Dw10 is not associated with the disorder. Our study determines and compares the natural ligand motifs of these MHC class II molecules and identifies 60 natural ligands. At relative position 4 (P4), only the RA-associated DR4 molecules allow, or even prefer, negatively charged amino acids, but do not allow those which are positively charged (Arg, Lys). In the case of DR4Dw10 the preference for these amino acids is reversed. The results predict features of the putative RA-inducing peptide(s). A remarkable specificity, almost exclusively for negative charges (Asp, Glu), is found at P9 of the DR4Dw15 motif. This specificity can be ascribed to amino acid β57 of the DR β chain, and gives an important insight into the β57-association of another autoimmune disease, insulin-dependent diabetes mellitus type I.

Integrin α5 during early development of Xenopus laevis

Abstract The full length sequence of the Xenopus integrin α 5 subunit is reported. Analysis of cloned cDNA fragments reveals that alternative polyadenylation of α 5 mRNA occurs in the embryo. Furthermore, a variant form of the α 5 mRNA is expressed which encodes an integrin α 5 subunit with a truncated cytoplasmic domain. Integrin α 5 mRNA and protein are expressed in oocytes, eggs and throughout development. Spatial expression of α 5 mRNAs is first detected by whole mount in situ hybridization in presumptive neural crest cells and in the somitic mesoderm from the midgastrula stage onwards. In contrast, the α 5 protein is present on newly formed plasma membranes beginning at first cleavage. During neurulation, the integrin α 5 subunit disappears from the outer layer of the ectoderm, the notochord and the neural tube and accumulates in the sensorial layer of the ectoderm, the somites and the neural crest cells. These results provide evidence for the position specific regulation of α subunit expression in early vertebrate embryos.

Distribution of parasite cysteine proteinases in lesions of mice infected with Leishmania mexicana amastigotes

It is well established that Leishmania mexicana amastigotes contain large amounts of cysteine proteinases in their extended lysosomes. In this study it is shown that the cell-free supernatant of homogenized lesion tissue from infected mice contains large amounts of acid proteinases. The majority of this enzymatic activity also corresponds to cysteine proteinases from L. mexicana amastigotes. Immunoelectron microscopy of mouse lesion sections suggests, that frequently amastigotes lyse and release lysosomal cysteine proteinases into the parasitophorous vacuole of infected macrophages. The cysteine proteinases are also found extracellularly in the tissue presumably as a result of macrophage rupture and appear to persist in the lesion tissue, where they may damage host cells and the extracellular matrix.

Motif of HLA-B*3503 peptide ligands

HLA-B*3501 and HLA-B*3503 molecules differ by a single amino acid located in the [~-pleated sheet of the c~2 domain. In order to clarify the basis of the strict distinction between these highly related class I molecules by several alloreactive T cell clones (Steinle et al. 1993), we determined the motif of HLA-B*3503 peptide ligands and compared it with the previously published HLA-B*3501 motif (Hill et al. 1992; Falk et al. 1993). HLA-B*3503 full-length cDNA (1.35 kilobases) was amplified by HLA-B locus-specific polymerase chain reaction from cDNA of donor HF and cloned in pUC18 as described (Steinle et al. 1993). A fully sequenced cDNA clone devoid of Taq-DNA pol, ymerase incorporation errors was cloned into the eucaryotic expression vector pFM92.1 (kindly provided by G. H~immerling, Heidelberg), a derivative of pHI3APr-1 neo (Gunning et al. 1987). Cotransfection with the human gene for ~2-microglobulin (Perarnau et al. 1988) of the mouse mastocytoma cell line P815 was performed using Lipofectin (Gibco BRL) according to manufacturers instructions. Staining with the HLA-B-specific antibody 4E (Trapani et al. 1989) revealed good surface expression of HLA-B*3503 on G418-resistant P815 cells (data not shown). These were expanded to obtain about 30 ml of cell pellet, lysed and HLA-B*3503 mole-

A 16MER PEPTIDE OF THE HUMAN AUTOANTIGEN CALRETICULIN IS A MOST PROMINENT HLA-DR4DW4-ASSOCIATED SELF-PEPTIDE

The human Ca(2+)-binding (storage) protein calreticulin, located in the lumen of the endoplasmic reticulum, is proposed to play a role as autoantigen: anticalreticulin autoantibodies occur in the sera of patients with SLE and patients with onchocerciasis (calreticulin shows a high sequence homology to the Onchocerca volvulus antigen RAL-1). Here we present sequencing data of a HLA-DR4Dw4-associated calreticulin peptide fragment, Cal(295-310), purified from a DR4Dw4 self-peptide pool. Cal(295-310) proved to be one of three commonest self-peptides associated with DR4Dw4 molecules that were isolated from the EBV-transformed B-cell line BSM (DR4Dw4, DRw53). We tested the binding of Cal(295-309) and the analogous RAL-1 peptide to HLA-DR molecules: Cal(295-309) exhibited specific binding characteristics for DR4Dw4. Binding assays using self-peptide analogues with replaced amino acids led us to a DR4Dw4-binding motif with anchor residues at relative positions 1 and 6. The sequencing data suggest that calreticulin is a frequently processed intracellular protein. The abundance of calreticulin makes the presentation of different calreticulin peptides associated with HLA-D molecules likely to occur, supporting the immunologic relevance of this molecule.

AMP deaminase in rat brain: Localization in neurons and ependymal cells

The purine nucleotide cycle enzyme AMP deaminase (AMPD) catalyzes the irreversible hydrolytic deamination of AMP. The physiological function of the purine nucleotide cycle in the brain is unknown. In situ hybridization and immunocytochemical studies were performed to identify the regional and cellular expression of AMPD in rat brain with the goal of elucidating the neural function of the purine nucleotide cycle. AMPD messenger RNA was detected in ventricular ependymal cells and cells of the choroid plexus and in neurons of distinct brain areas. Although only low antibody titers were obtained by immunization with the purified sheep brain AMPD, immunization of mice with synthetic lipopeptide vaccines containing oligopeptides derived from a known partial complementary DNA sequence of the enzyme yielded an antiserum suitable for immunocytochemistry. Immunostaining of cells in culture showed that neurons but not astroglial cells express appreciable amounts of the enzyme. Results of immunocytochemical staining performed on rat brain slices were in accord with the localization of AMPD messenger RNA, thus confirming the expression of AMPD in neurons of the brain stem, hippocampus, cerebellar nuclei and mesencephalic nuclei, as well as in ventricular ependymal cells and their cilia.

Identification of unusual amino acids in peptides using automated sequential Edman degradation coupled to direct detection by electrospray-ionization mass spectrometry.

The determination of the primary structure of peptides and proteins is routine in many laboratories; however, many of the obtained sequences are incomplete or can be misinterpreted when the samples contain unusual amino acids. Here we report the development of an automated peptide sequenator coupled to an electrospray-ionization (ESI) mass spectrometer (MS) that, in conjunction with minor modifications to the sequencing conditions and, in some cases, prior derivatization of amino acids, allows the detection of the phenylthiohydantoin (PTH) derivatives of a number of unusual amino acids. Using the coupled sequenator-ESI-MS system we were able to determine the complete sequence of the lantibiotic gallidermin, a partial sequence of the calcium-dependent peptide antibiotic CDA2 as well as the pool sequence of a mixture of synthetic peptides containing nonproteinogenic amino acids. In addition to the 20 proteinogenic amino acids, the procedure was able to detect PTH derivatives of hydroxyphenylglycine, 2,3-didehydroasparagine, 3-methylglutamic acid, oxytryptophan, ornithine, N-methylglycine, dihydroxyphenylalanine, and alpha-aminoisobutyric acid. Similarly, after a simple derivatization procedure, we were also able to correctly identify educts of 2,3-didehydroalanine, 2,3-didehydrobutyrine, lanthionine, and 3-methyllanthionine.

P408 - Natural ligand motifs of closely related HLA-DR4 molecules predict features of rheumatoid arthritis associated peptides

Rheumatoid arthritis (RA), one of the most common autoimmune disorders, is believed to be mediated via. T lymphocytes and genetic studies have shown that it is strongly associated with HLA-DR4. The DR4 subtypes DR4Dw4, DR4Dw14 and DR4Dw15 represent increased risk factors for RA, whereas DR4Dw10 is not associated with the disorder. Our study determines and compares the natural ligand motifs of these MHC class II molecules and identifies 60 natural ligands. At relative position 4 (P4), only the RA-associated DR4 molecules allow, or even prefer, negatively charged amino acids, but do not allow those which are positively charged (Arg, Lys). In the case of DR4Dw10 the preference for these amino acids is reversed. The results predict features of the putative RA-inducing peptide(s). A remarkable specificity, almost exclusively for negative charges (Asp, Glu), is found at P9 of the DR4Dw15 motif. This specificity can be ascribed to amino acid beta57 of the DR beta chain, and gives an important insight into the beta57-association of another autoimmune disease, insulin-dependent diabetes mellitus type I.