Einleitung: Interleukin-8 (IL-8) ist ein chemotaktisches Zytokin, das eine wichtige Rolle in der Rekrutierung und Aktivierung von Neutrophilen in der intestinalen Mukosa spielt. Es konnte kürzlich gezeigt werden, dass intestinale Epithelzellen IL-8 produzieren. Wir wollten daher untersuchen, ob intestinale Epithelzellen IL-8 Rezeptoren exprimieren und ob diese Rezeptoren funktionelle Eigenschaften von IL-8 auf intestinale Epithelzellen vermitteln.

Background: CXCL8 (previously known as Interleukin-8), a member of the α-chemokine family of chemotactic cytokines, stimulates intestinal neutrophil activation and chemotaxis. As intestinal epithelial cells have been recently shown to produce CXCL8, the aim of this study was to identify functional activities of CXCL8 on intestinal epithelial cells. Methods: The expression of CXCL8 receptors CXCR1 and CXCR2 was assessed by RT-PCR and FACS analysis in human Caco-2 and HT-29 cells. The effects of CXCL8 on intestinal epithelial proliferation were assessed with colorimetric MTT assays and the effects on epithelial restitution with an in vitro migration model using Caco-2 and HT-29 cells. Results: While the expression of both CXCR1 mRNA and protein could be demonstrated by RT-PCR and FACS analysis in human Caco-2 and HT-29 cells, no expression of CXCR2 was observed in these cell lines. Colorimetric MTT assays revealed that CXCL8 does not modulate cell proliferation in HT-29 and Caco-2 cells. In contrast, CXCL8 significantly enhanced intestinal epithelial migration in an in vitro migration model of HT-29 and Caco-2 cells. Enhancement of intestinal epithelial cell migration by CXCL8 was partially CXCR1-dependent and TGFβ-independent. Conclusion: CXCL8 exerts functional effects on intestinal epithelial cells that may be relevant for intestinal inflammation and mucosal healing.

The heptapeptide Leu-Arg-Arg-Ala-Ser-Leu-Gly (Kemptide) is a synthetic construct of a substrate for cAMP-dependent protein kinase (PK). In this work we show that Kemptide has all the properties of a cytophilic substrate, i.e. it is a molecule preserving cell membrane intactness when added to cultured cells. Kemptide thus satisfies the prerequisites for employment in assays for cell surface-located ecto-PK activity. Different types of intact cells catalyze the phosphorylation of Kemptide in the presence of extracellular ATP and cAMP with Km values of 3-4 microM for Kemptide. Kemptide phosphorylation was influenced by PKI, the inhibitory protein specific for cAMP-PK. The results of comparative experiments with intact cells and with cell extracts demonstrate the ectoenzyme nature of this cAMP-PK. Further, the possibility was ruled out of a transfer of enzyme activity from damaged cells to the surface of intact cells. The anchorage of the surface cAMP-PK activity to the plasma membrane appears to be relatively stable since (i) cell supernatants, obtained after preincubation of intact cells with cAMP or Kemptide, did not show Kemptide phosphorylation, and (ii) the cAMP-dependent PK activity remained with cells even after five consecutive washes with cAMP or Kemptide. This is in contrast to the ecto-cAMP-independent phosvitin/casein type PK (Kübler, D., Pyerin, W., Burow, E., and Kinzel, V. (1983) Proc. Natl. Acad. Sci. U.S.A. 80, 4021-4025) which is released from intact cells through the addition of substrate. Data are presented which show that both ectokinase activities are exhibited independently. In conjunction with published evidence for an active export of cAMP from cells as well as for the appearance of extracellular ATP the demonstration of an ecto-cAMP-PK further supports the potential of PK for intercellular regulation. The potential of ecto-cAMP-PK is demonstrated by its ability to phosphorylate biologically active forms of atrial natriuretic peptide, the atrial natriuretic peptide, which possesses the specific sequence for a cAMP-PK-catalyzed phosphorylation.

Protein kinase; Cyclic AMP; (Bovine)

The catalytic subunit (C) of cAMP-dependent protein kinase holoenzyme type II from bovine cardiac muscle was separated by isoelectric focusing in Immobiline polyacrylamide gels into 9 protein forms. The major forms (i) appeared at pH 7.1, 7.4, 7.5, and 7.7, (ii) exhibited protein kinase activity and were inhibited by heat and acid stable inhibitor, (iii) represented approx. 30%, 4%, 64%, and 1% of the protein respectively, (iv) refocused in the same position from which they had been eluted from the first gel. Antibodies against C detected additional proteins at approx. pH 7.55, 7.75, and 7.8. Two more bands became detectable at approx. pH 7.3 and 7.45 by application of antibody against C beta (Uhler, M.D. & McKnight G.S. 1987, J.Biol.Chem. 262, 15202-15207). The relation of the different forms of C to the fractions CA and CB (Kinzel V. et al. 1987 Arch. Biochem. Biophys. 253, 341-349) is demonstrated.

It has been previously demonstrated that the combination of pure preparations of the inhibitor protein of the cyclic AMP-dependent protein kinase and the catalytic subunit of this enzyme resulted in the formation of multiple complexes [Van Patten, Fletcher & Walsh (1986) J. Biol. Chem. 261, 5514-5523]. In the present study it is demonstrated that these multiple species occur because the bovine heart protein kinase preparation contains multiple forms of catalytic subunit [Kinzel, Hotz, König, Gagelmann, Pyerin, Reed, Köbler, Hofmann, Obst, Gensheimer, Goldblatt & Shaltiel (1987) Arch. Biochem. Biophys. 253, 341-349].

Electrophoretically homogeneous preparations of catalytic subunit (C) of cAMP-dependent protein kinase isolated according to two different procedures from holoenzyme type I and type II from rabbit and from holoenzyme type II from rat skeletal muscle and from bovine cardiac muscle can be separated on carboxymethyl cellulose or on a Mono S column (Pharmacia) by salt gradient elution into two enzymatically active peaks called A and B, which do not interconvert on rechromatography. Cochromatography of peak A fractions or of peak B fractions derived from both holoenzymes respectively yields single enzyme peaks in each case, thus indicating that both represent different entities, which were named CA and CB. The separate character of both enzyme forms is supported by the fact that CB under all conditions is degraded faster by the C-specific protease (E. Alhanaty et al. (1981) Proc. Natl. Acad. Sci. USA 78, 3492-3495) than CA, a phenomenon which is enhanced in both enzyme forms by substrate (Kemptide). The separation of both subtypes from each other is probably based on differences in isoelectric values (delta pH less than or equal to 0.5 units). The reason for the charge difference is not presently known. CA and CB do not differ significantly in their phosphate content. No differences between CA and CB have been detectable so far with respect to their migration in SDS gels, kinetic behavior regarding both substrates and cosubstrate, pH dependence, inhibition by regulatory subunits of holoenzyme type I (rabbit skeletal muscle) and of type II (bovine cardiac muscle), and inhibition by specific-heat and acid-stable inhibitor-modulator. The peptide pattern of both forms after limited proteolysis exhibits small differences.

Glucocorticoid receptors of wild-type lymphoid cells and of two classes of glucocorticoid-resistant variants of "nuclear transfer deficient" (nt-) and "increased nuclear transfer" (nti) phenotypes, respectively, were investigated. Photoaffinity labeling of receptor complexes with a radiolabeled glucocorticoid of high affinity was used to analyze these receptor types by electrophoresis in sodium dodecyl sulfate containing gels. Wild-type and nt- -variant receptors yielded radiolabeled polypeptide bands of Mr 94 000 +/- 5000 while nti-variant receptors had a molecular weight of 40 000 +/- 2000. Partial proteolysis of wild-type and nt- receptors with alpha-chymotrypsin resulted in steroid-labeled receptor fragments of Mr 37 000-38 000 while nti-variant receptors remained unchanged. In the case of wild-type receptors, the chymotryptic fragment had increased affinity for DNA indistinguishable from that of native nti-variant receptors. Depending on the nt- cell clone, the chymotryptic receptor fragments containing the steroid binding site had either the same low affinity for DNA as the undigested receptors or a slightly increased affinity. Partial proteolysis with trypsin of wild-type, nt-, and nti receptors resulted in steroid-labeled fragments of Mr 29 000 as major products and some fragments of Mr 27 000. These tryptic receptor fragments were devoid of DNA binding ability regardless of the original receptor types. With a lysine-specific protease, similar fragments were obtained from wild-type, nt-, and nti receptors. In contrast, a protease specific for arginine residues did not produce receptor fragments detectable by our techniques. A model of the wild-type receptor is discussed.

[3H]Triamcinolone acetonide was used to tag covalently specific glucocorticoid receptors by photoaffinity labelling at lambda greater than or equal to 320 nm. Receptors of wild-type mouse lymphoma cells and two glucocorticoid resistant mutants of "nuclear transfer deficient" (nt-) and "increased nuclear transfer" (nti) phenotypes, respectively, were used. Wild-type and nt- receptors yielded radiolabelled polypeptide bands of mol. wt. 98 000 as revealed by gel electrophoresis under denaturing conditions and fluorography. In contrast, the nti receptor had a mol. wt. of 42 000. Partial proteolysis of the wild-type receptor with alpha-chymotrypsin resulted in a fragment of mol. wt. 39 000 which still contained the steroid binding site but had increased affinity for DNA indistinguishable from that of the nti receptor. Chymotrypsin thus removed a domain from the wild-type receptor polypeptide which is involved in modulating DNA binding. The same domain is missing from the nti receptor.