Differential interaction of glimepiride and glibenclamide with the beta-cell sulfonylurea receptor. II. Photoaffinity labeling of a 65 kDa protein by [3H]glimepiride.

Glimepiride is a novel sulfonylurea for the treatment of type II-diabetic patients exhibiting different receptor binding kinetics to beta-cell membranes with 8-9-fold higher koff rate and 2.5-3-fold higher kon rate compared to glibenclamide (see accompanying paper (Müller, G. et al. (1994) Biochim. Biophys. Acta 1191, 267-277)). To elucidate the molecular basis for this differential behaviour of glimepiride and glibenclamide, direct photoaffinity labeling studies using beta-cell tumor membranes were performed. [3H]Glimepiride was specifically incorporated into a membrane polypeptide of M(r) = 65,000 under conditions, which led to predominant labeling of a 140 kDa protein by [3H]glibenclamide (Kramer, W. et al. (1988) FEBS Lett. 229, 355-359). Labeling of the 140 kDa protein by [3H]glibenclamide was inhibited by unlabeled glimepiride and, vice versa, glibenclamide inhibited labeling of the 65 kDa protein by [3H]glimepiride. The 65 kDa protein was also specifically photolabeled by the sulfonylurea [125I]35623, whereas an 4-azidobenzoyl derivative of glibenclamide, N3-[3H]33055, exclusively labeled a 33 kDa protein. Competitive Scatchard analysis of [3H]glimepiride-binding and [3H]glibenclamide-binding to RINm5F cell membranes using glibenclamide and glimepiride, respectively, as heterologous displacing compounds yielded non-linear plots. These findings may be explained by cooperative interactions between the 140 and 65 kDa sulfonylurea-binding proteins. The possibility that sulfonylureas of different structure have different access to the 140 and 65 kDa receptor proteins due to the beta-cell membrane barrier was investigated by photoaffinity labeling of solubilized beta-cell membrane proteins. Interestingly, solubilization of beta-cell tumor membranes led to a shift of specific [3H]glibenclamide binding from the 140 kDa to the 65 kDa binding protein, exclusively, and to an increased labeling of the 65 kDa protein by [3H]glimepiride. The labeling of a unique protein is in agreement with similar Kd values measured for both sulfonylureas upon solubilization of beta-cell tumor and RINm5F cell membranes (see accompanying paper). Furthermore, competitive Scatchard plots of [3H]glimepiride binding to solubilized RINm5F cell membrane proteins in the presence of glibenclamide and vice versa approximate linearity suggesting loss of cooperativity between the 140 kDa glibenclamide-binding and 65 kDa glimepiride-binding proteins upon solubilization. The physiological significance of the differential interaction of glimepiride and glibenclamide with different binding proteins was also substantiated by photoaffinity labeling of RINm5F cells leading to labeling of a 140 kDa protein by [3H]glibenclamide and of a 65 kDa protein by [3H]glimepiride.(ABSTRACT TRUNCATED AT 400 WORDS)

Direct photoaffinity labeling of the putative sulfonylurea receptor in rat beta-cell tumor membranes by [3H]glibenclamide.

The oral antidiabetic sulfonylurea [3H]glibenclamide specifically binds to plasma membranes from a rat beta-cell tumor indicating a receptor for sulfonylureas in these membranes. Irradiation of [3H]glibenclamide at 254 or 300 nm in the presence of albumin resulted in covalent labeling of the albumin molecule. Direct photoaffinity labeling of beta-cell membranes with [3H]glibenclamide resulted in the covalent modification of two membrane polypeptides with apparent molecular masses 140 and 33 kDa. The extent of labeling of the 140 kDa polypeptide was specifically decreased by sulfonylureas. This suggests that a membrane polypeptide of 140 kDa is a component of the sulfonylurea receptor in the beta-cell membrane.

The molecular interaction of sulfonylureas with beta-cell ATP-sensitive K(+)-channels.

The molecular interaction of glimepiride and glibenclamide with the beta-cell sulfonylurea receptor was investigated by kinetic and steady state binding as well as photoaffinity labeling. The novel sulfonylurea, glimepiride, exhibits a significantly higher exchange rate with the sulfonylurea receptor but a 2.5-3 fold lower binding affinity compared to glibenclamide. [3H]Glimepiride was specifically incorporated into a 65-kDa polypeptide under conditions which led to predominant labeling of a 140-kDa protein by [3H]glibenclamide. Labeling of the 140-kDa protein by [3H]glibenclamide was inhibited by unlabeled glimepiride and, vice versa, glibenclamide inhibited labeling of the 65-kDa protein by [3H]glimepiride. The 65-kDa protein was also specifically photolabeled by the sulfonylurea [125I]35623, whereas an 4-azidobenzoyl derivative of glibenclamide, N3-[3H]33055, exclusively labeled a 33-kDa protein. Solubilization of beta-cell tumor membranes led to a shift of specific [3H]glibenclamide-binding from the 140-kDa to the 65-kDa protein, exclusively and to an increased labeling of the 65-kDa protein by [3H]glimepiride. The labeling of a unique protein is in agreement with similar Kd-values for binding to the sulfonylurea receptor measured for both sulfonylureas upon solubilization of beta-cell membranes. Photoaffinity labeling of intact cultured beta-cells led also to labeling of a 140-kDa protein by [3H]glibenclamide and of a 65-kDa protein by [3H]glimepiride. These studies suggest that the beta-cell sulfonylurea receptor consists of at least two protein subunits of M(r) 140,000 and 65,000 which bind sulfonylureas of different structure with different binding affinities and kinetic parameters. Furthermore, the exchange rate of a sulfonylurea determines the insulin releasing activity in vitro more closely than the binding affinity.

[Exchange of A1-glycine in bovine insulin with L- and D-tryptophan]. / Austausch von A1-Glycin in Rinderinsulin gegen L- und D-Tryptophan.

Substitution of A1-glycine of insulin by L-amino acids yields in analogues with low biological activity. With D-amino acids in A1 biological activity is essentially retained. Synthesis of [A1-L-tryptophan]- and [A1-D-tryptophan]-insulin should provide information about the position of the side chains of L- and D-amino acids relative to A19-tyrosine, e.g. by evaluation of intramolecular resonance energy transfer between the fluorescent side chains. [A1-D-Tryptophan]-insulin exhibits full biological activity.

[(A1-D-alanine) insulin (author's transl)]. / (A1-D-alanin) Insulin

Starting from porcine insulin, A1-glycine was substituted by D-alanine and by L-alanine for comparison. Replacement of A1-glycine by L-alanine revealed the known decrease in the biological activity. [A1-D-Alanine]insulin, however, has the same blood sugar lowering activity as insulin and is slightly more active in its influence on the glucose uptake into the rat diaphragm. The specific binding to insulin receptors of rat liver is decreased as, compared to insulin, but increased as compared to the L-alanine analogue.

[Binding of insulin analogs to partially purified insulin receptor from rat liver membrane (author's transl)]. / Bindung von Insulinanaloga an partiell gereinigten Insulinrezeptor aus Rattenlebermembran

The insulin binding fraction was solubilized from rat liver membrane vesicles by triton and partially purified up to the specific binding activity of 2.1 pmol/mg. A further characteristic of the partially purified soluble receptor is a decrease in irreversible binding to 0.36% (+/- 0.28) with regard to total iodine insulin and to (+/- 1.8) in comparison to reversible binding. From Scatchard plots a high affinity binding site (KD = 5 X 10(-10)M) with low capacity (5 pmol/mg) and a low affinity binding site (KD = 3 X 10(-8) M) with high capacity (30 pmol/mg) can be seen. With carefully prepared liver membrane vesicles, the dissociation constant of the high affinity binding site from Scatchard plot is only 2 X 10(-9)M. With liver membrane vesicles, isolated for preparative purification procedure, the high affinity binding site could not be demonstrated. Displacement studies with insulin analogs were performed with [A1-D-alanine] insulin, [A1-L-alanine] insulin, [des-Gly-A1, NB1, NB29-(Msc)2]-insulin, proinsulin and [desoctapeptid B23-B30]-insulin. Results of binding measurements are presented in half-maximal iodo-insulin binding, in determination of inhibitor- and dissociation constants from Dixon-, Scatchard- and Lineweaver-Burk plots. There are equal relative binding potencies of analogs, observed with crude membrane vesicles and partially purified soluble receptor, although there is a 50-fold difference in specific binding activity. Biologically active insulins are characterized by strong binding to the high affinity binding site. The binding to the low affinity binding site is not correlated to the biological activity of the insulin analog. With insulin and biologically responsive analogs a non-linear curve in the double-reciprocal Lineweaver-Burk plot can be observed. Analogs with low biological activity show a linear dependency. Functional interactions of insulin with the receptor can be demonstrated in a high affinity binding with the first binding site of the Scatchard plot and in a non-linear hyperbolic Lineweaver-Burk plot.

[On the inhibitory effect of chloramphenicol on mitochondrial protein synthesis as a possible cause of its selective toxic side effects (author's transl)]. / Uber die Hemmwirkung von Chloramphenicol auf die Proteinsynthese von Mitochondrien als Ursache der selektiven toxischen Nebenwirkung

The inhibition of protein synthesis by tetracycline and chloramphenicol in mitochondria isolated from rat liver or rabbit bone marrow was investigated in vitro. It could be demonstrated that there is but little difference between the inhibitory effect of both substances in regard to mitochondria protein synthesis. Therefore, a selective interference of chloramphenicol with bone marrow mitochondria cannot be concluded. In in vitro tests the concentrations required for inhibiting the mitochondria protein synthesis are comparable to those found in blood levels measured under usual therapeutic conditions. It can be assumed that in the living organism the mitochondria are protected against tetracycline or chloramphenicol interference. This hypothesis is supported by the fact that due to the increase in the substrate concentration (phenylalanine), the expected inhibition of mitochondrial protein synthesis does not occur. There may be an inhibition of influx of the antibiotic by competition with substrate for transport through mitochondrial membrane. The reason for myelotoxicity of chloramphenicol is not an increased sensitivity of mitochondrial ribosomes located in the bone marrow to chloramphenicol. But the existence of special conditions for chloramphenicol influx into mitochondria of bone marrow must be concluded.

[(A1-beta-Alanine) insulin]. / [A1-beta-Alanin]Insulin

Starting from porcine insulin, A1-glycine was substituted by beta-alanine. The blood sugar lowering effect of the new analogue in the rabbit is about 45% of that of insulin. The half-maximal binding to partially purified rat liver receptors is about 46%, to transformed human lymphocytes about 54%, compared to insulin.

Inhibition of 3H-glibenclamide binding to sulfonylurea receptors by oral antidiabetics.

Specific binding of 3H-glibenclamide (HB 419) is observed with membranes from rat cerebral cortex and rat beta-cell tumor. 3H-Glibenclamide binding is of high affinity, reversible, saturable, and can be displaced by oral antidiabetics of different structural types. Half maximal inhibition of binding correlates with the hypoglycemic action of the compounds.

[Insulin analogues with substitution of A1-glycine by D-amino acids and omega-amino acids (author's transl)]. / Insulinanaloga durch Austausch von A1-Glycin gegen D-Aminosäuren und omega-Aminosäuren.

Substitution of A1-glycine of insulin by L-amino acids yields in analogues with low biological activity. With D-amino acids in A1 biological activity is essentially retained. The influence of aliphatic, aromatic, acidic and basic alpha-D-amino acids as well as omega-amino acids in A1 on the biological effects in different test systems is studied and discussed.