Dipeptidyl peptidase 4 (DP4)/CD26 regulates the biological function of various peptide hormones by releasing dipeptides from their N-terminus. The enzyme is a prominent target for the treatment of type-2 diabetes and various DP4 inhibitors have been developed in recent years, but their efficacy and side effects are still an issue. Many available crystal structures of the enzyme give a static picture about enzyme-ligand interactions, but the influence of amino acids in the active centre on binding and single catalysis steps can only be judged by mutagenesis studies. In order to elucidate their contribution to inhibitor binding and substrate catalysis, especially in discriminating the P 1 amino acid of substrates, the amino acids R125, N710, E205 and E206 were investigated by mutagenesis studies. Our studies demonstrated, that N710 is essential for the catalysis of dipeptide substrates. We found that R125 is not important for dipeptide binding but interacts in the P 1 ’position of the peptide backbone. In contrast to dipeptide substrates both amino acids play an essential role in the binding and arrangement of long natural substrates, particularly if lacking proline in the P 1 position. Thus, it can be assumed that the amino acids R125 and N710 are important in the DP4 catalysed substrate hydrolysis by interacting with the peptide backbone of substrates up- and downstream of the cleavage site. Furthermore, we confirmed the important role of the amino acids E205 and E206. However, NP Y, displaying proline in P 1 position, is still processed without the participation of E205 or E206.

Abstract In the present studies we resolved the post-translational microheterogeneity of purified porcine dipeptidyl peptidase IV (DP 4) from kidney cortex. Applying SDS-homogeneous DP 4 onto an analytical agarose isoelectric focusing (IEF) gel, pH 4–6, activity staining resulted in at least 17 isoforms between pH 4.8–6.0. These could be separated into fractions with only two to six isoforms by means of preparative liquid-phase IEF, using a Rotofor cell. Starting off with three parallel Rotofor runs under the same conditions at pH 5–6, the fractions were pooled according to the specific activity of DP 4, pH and analytical IEF profile, and further refractionated without any additional ampholytes. Since excessive dilution of ampholytes and proteins was kept to the minimum, a second refractionation step could be introduced, resulting in pH gradients between 0.022 and 0.028 pH increments per fraction. By performing two consecutive refractionation steps, the high resolution necessary for the separation of DP 4 isoforms could be achieved. This represents an alternative method if isolation of isoforms with similar pI's results in precipitation and denaturation in presence of a narrow pH range. Furthermore, it demonstrates that preparative IEF is a powerful tool to resolve post-translational microheterogeneity of a purified protein required for crystallization processing.

Posttranslational modifications influence the structure, stability and biological activity of proteins. Most of the reactions are enzyme-catalyzed, but some, such as asparagine (Asn) and glutamine (Gln) deamidation and the isoaspartate (isoAsp) formation within peptide chains, occur spontaneously. It has been previously shown that certain peptide sequences form isoAsp quite fast if the Asp stretches are exposed to the protein surface, thereby potentially changing susceptibility to proteolysis at these sites. This tempted us to investigate the activity of exo- and endopeptidases against Asp- or isoAsp-containing substrates. Members of the prolyl oligopeptidase family were unable to cleave substrates after proline if isoAsp was placed in the P2-position. Caspases, usually accepting Asp at P1-position of their substrates, did not cleave isoAsp-containing sequences. Similarly, the metal-dependent aminopeptidase amino peptidase N did not turnover N-terminal isoAsp-containing substrates, nor could the endopeptidase matrix metalloproteinase 3 (MMP 3) hydrolyze a serum amyloid A protein-like substrate if the sequence contained isoAsp instead of Asp. Also, the highly specific enterokinase, usually clipping after a stretch of four Asp residues and a lysine in the P1 position, could not turnover substrates if the P2 amino acid was replaced by isoAsp. In contrast, acylamino acid-releasing enzyme and dipeptidyl peptidases 1, 2 and 4 hydrolyzed substrates containing the isoAsp-Ala motif.

The data presented, demonstrate for the first time PEP turnover of HN by a limited postcysteine as well as the expected post-proline proteolysis demonstrated in cell extract resulting in the inactivation of this potentially apoptosis-related factor. These findings lead to the hypothesis of a PEP-mediated control of HN homeostasis maintaining neuronal cell survival. This implicates a novel use of PEP inhibitors potentially preventing intracellular HN digestion. Consequently, PEP-inhibition might be a new target for apoptosis prevention.

The expression and purification of soluble human DP IV was established in P. pastoris on a 5 liter fermentation scale (3 mg/fermentation, 32.27 U/mg). Thus it represents an alternate approach to previous published methods were DP IV was expressed in cell culture (Tanaka et al.7) with lower production rates and to the production in insect cells (Dobers et al.8) Biochemical and kinetic characterization demonstrated that the soluble recombinant DP IV displayed similar properties as DP IV purified from porcine kidney regarding size, activity, isoelectric point and glycosylation. Furthermore, the new expression method enables future structure-function related studies of DP IV.

The incretins, glucose-dependent insulinotropic peptide (GIP1–42) and glucagon-like peptide 1 (GLP-17–36), are involved in regulation of gastric emptying, glucose homeostasis, body fat regulation and the glucose-induced insulin secretion from the endocrine pancreas. After release in the circulation both peptides are rapidly degraded by the exopeptidase dipeptidyl peptidase IV (DP IV) to the inactive polypeptides GIP3–42 and GLP-19–36. In vivo stabilization of the active incretins by orally available DP IV-inhibitors is now widely accepted as a new therapeutic approach in antidiabetic treatment. In order to demonstrate the pharmacodynamic effect of DP IV-inhibitors, it is necessary to measure the plasma levels of active and inactive forms of GIP and GLP-1. We previously described an immunoprecipitation method as sample preparation and concentration in combination with a LC–MS analysis for determination of active and inactive GIP. We could improve the efficiency and suitability of this method by reduction of the necessary sample volume to 1.0 ml and simultaneous measurement of GIP1–42, GIP3–42 and GLP-17–36, GLP-19–36, without loss of sensitivity. An LOQ of approximately 5 and 11 pmol/l was maintained for GIP and GLP-1, respectively.

Six GIP(1-30NH2) analogs were synthesized with modifications (deprotonation, Nmethylation, reversed chirality, and substitution) at positions 1, 3, and 4 of the Nterminus, and additionally, a cyclized GIP derivative was synthesized. The relationship between altered structure to biological activity was assessed by measuring receptor binding affinity and ability to stimulate adenylyl cyclase in CHOK1 cells transfected with the wildtype GIP receptor (wtGIPR). These structureactivity relationship studies demonstrate the importance of the GIP Nterminus and highlight structural constraints that can be introduced in GIP analogs. These analogs may be useful starting points for design of peptides with enhanced in vivo bioactivity.

The synthesis and complete characterization by multinuclear NMR, infrared, and Mossbauer spectroscopy, by single crystal X-ray analysis, as well as by electrospray mass spectrometry of the new soluble triorganotin fluoride Me2PhSnF (1) is reported. The crystal structure of 1 reveals a rod-like polymeric structure in the solid state. Solutions of 1 in apolar solvents, such as toluene, contain mixtures of interconvertible oligomers. Ab initio MO calculations on model compounds H3SnF, [H3SnFSnH3](+), and [FH3SnFSnH3F](-) indicate that the Sn-F bonds are substantially ionic in character, and suggest open-chain species, rather than cyclic species. In donor solvents, such as pyridine, 1 forms complexes with the solvent, such as Me2PhSnF. pyridine. The solid-state structure of (Me3SiCH2)(3)SnF (2) is reformulated as monomeric with a weak intermolecular (SnF)-F-. . . interaction that gives rise to a [4+1] coordination. The relative short Sn-Sn separation enables the fluorine atoms to oscillate (flip-flop) between two neighboring (Me3SiCH2)(3)Sn-groups, which is expressed in the X-ray experiment by a dynamic disorder. Ab initio MO calculations on a model compound, [H3SnFSnH3+](+), suggest only a small energy barrier for the flip-flop motion.

Although type 2 diabetic patients exhibit resistance to GIP when the peptide is administered in doses that result in circulating levels approximating those found physiologically, it is likely that DP IV-resistant forms of the peptide administered in pharmacological doses will prove to be effective in improving glucose tolerance. Additionally, in view of recent studies showing that GIP receptor knockout mice are resistant to diet induced obesity25, it is possible that GIP-antagonists will prove useful in obesity treatment.

Glutaminyl cyclase (QC, EC 2.3.2.5) catalyzes the formation of pyroglutamate residues from glutamine at the N-terminus of peptides and proteins. In the current study, human QC was functionally expressed in the secretory pathway of Pichia pastoris, yielding milligram quantities after purification from the supernatant of a 5 L fermentation. Initial characterization studies of the recombinant QC using MALDI-TOF mass spectrometry revealed correct proteolytic processing and N-glycosylation at both potential sites with similar 2 kDa extensions. CD spectral analysis indicated a high alpha-helical content, which contrasts with plant QC from Carica papaya. The kinetic parameters for conversion of H-Gln-Tyr-Ala-OH by recombinant human QC were almost identical to those previously reported for purified bovine pituitary QC. However, the results obtained for conversion of H-Gln-Gln-OH, H-Gln-NH2, and H-Gln-AMC were found to be contradictory to previous studies on human QC expressed intracellularly in E. coli. Expression of QC in E. coli showed that approximately 50% of the protein did not contain a disulfide bond that is present in the entire QC expressed in P. pastoris. Further, the enzyme was consistently inactivated by treatment with 15 mM DTT, whereas deglycosylation had no effect on enzymatic activity. Analysis of the fluorescence spectra of the native, reduced, and unfolded human QC point to a conformational change of the protein upon treatment with DTT. In terms of the different enzymatic properties, the consequences of QC expression in different environments are discussed.

Glucagon is a 29-amino acid polypeptide released from pancreatic islet α-cells that acts to maintain euglycemia by stimulating hepatic glycogenolysis and gluconeogenesis. Despite its importance, there remains controversy about the mechanisms responsible for glucagon clearance in the body. In the current study, enzymatic metabolism of glucagon was assessed using sensitive mass spectrometric techniques to identify the molecular products. Incubation of glucagon with purified porcine dipeptidyl peptidase IV (DP IV) yielded sequential production of glucagon3–29 and glucagon5–29. In human serum, degradation to glucagon3–29 was rapidly followed by N-terminal cyclization of glucagon, preventing further DP IV-mediated hydrolysis. Bioassay of glucagon, following incubation with purified DP IV or normal rat serum demonstrated a significant loss of hyperglycemic activity, while a similar incubation in DP IV-deficient rat serum did not show any loss of glucagon bioactivity. Degradation, monitored by mass spectrometry and bioassay, was blocked by the specific DP IV inhibitor, isoleucyl thiazolidine. These results identify DP IV as a primary enzyme involved in the degradation and inactivation of glucagon. These findings have important implications for the determination of glucagon levels in human plasma.

The gastrointestinal peptide glucose-dependent insulinotropic polypeptide (GIP1-42) is one of the incretin hormones regulating glucose-induced insulin secretion from the endocrine pancreas. GIP1-42 is a substrate of the circulating enzyme dipeptidyl peptidase IV, which removes the N-terminal peptide Tyr-Ala resulting in the inactive polypeptide GIP3-42. Hither to existing immunoassays do not enable a separate quantification of active and inactive forms, respectively. Therefore, we developed a highly specific and sensitive LC-MS assay for the identification and quantification of GIP1-42 and GIP3-42. Total GIP was immunoprecipitated from crude plasma samples using a C-terminally directed antibody. Thus, peptides were purified and concentrated prior to LC-MS analysis. The present immunoprecipitation-LC-MS assay enables the quantification of active and inactive GIP over a concentration range from 5 to 350 pmol/l in human plasma samples. Since this range covers the basal and postprandial levels of GIP the method is applicable to the determination of concentration changes and changes in the ratio of active and inactive forms of GIP in human plasma.

Currently used methods for investigating kinetics of peptide degradation such as refractive index monitoring, radioimmunoassay (RIA), high-performance liquid chromatography (HPLC), or capillary electrophoresis (CE) are time consuming, need large amounts of substrate, and are often too insensitive. Moreover, as in the case of RIA, HPLC, and CE, it is often impossible to interpret the observed results with confidence in the integrity of the analyte. To circumvent such obstacles, we found matrix-assisted laser desorption/ionization (MALDI) used with time-of-flight mass spectrometry (TOFMS) not only useful for qualitative analysis of reaction pathways but also for quantification. In the following chapter, we give two examples of kinetic reaction course evaluation, one non-enzymatic and one enzymatic.