The effect of hypoglycemic sulfonylureas on human red blood cell transglutaminase activity

Diabetes Research and Clinical Practk'e, 5 (1988) 31-36

Elsevier

31

DRC 00200

The effect of hypoglycemic sulfonylureas on human red blood cell transglutaminase activity R. G o m i s , A. C a s a n o v a s , R. C a s a m i t j a n a , A. Sarto, J. A r r o y o , M.J. Coves, F. R i v e r a and E. Vilardell Endocrhudogy and Diahetes Unit. Hm'mone Laboratory. Hospital Clinic. Schmd c~/"Medich~e. Barcehma University. 08036 Barcehma. Spain

(Received 16 December 1986, revision received 2 December 1987, accepted 3 December 1987)

Key words: Hypoglycemicsulfonylureas,Transglutaminase activity: Diabetes mellilus type 2: Insulin receptors: Blood red cells

Summary We have examined the effect of glipizide, a hypoglycemic sulfonylurea, upon transglutaminase activity in human red blood cells. In a first series of experiments the in vitro effect of the drug was assessed. The results obtained showed that glipizide inhibits transglutaminase activity in human red blood cells. In a second approach, glipizide was administered orally to six type 2 diabetic patients during 3 months, in order to evaluate the long-term effect upon transglutanlinase activity. Again, glipizide induced a significant decrease in the enzyme activity in blood red cells (P < 0.01). We suggest that treatment of type 2 diabetes mellitus with hypoglycemic sulfonylureas could improve insulin effects by inhibiting cellular transglutaminase activity.

Introduction Hypoglycemic sulfonylureas are an effective means of lowering glucose plasma levels in patients with type 2 diabetes mellitus [1]. However, their exact mechanism of action is still a matter of discussion. It is believed that they could act primarily by stimulating insulin release from the pancreatic fl-cell and,

Address lbr correspondence: Dr. R. Gomis, Endocrinology and Diabeles Unit, Hospital Clinic. School of Medicine, Villarroel 170, 98036 Barcelona. Spain.

at the same time, by enhancing insulin binding in the target tissues. In fact, an initial transitory rise in plasma insulin level at the beginning of treatment has been frequently described: continued therapy shows a return of plasma insulin concentrations to levels that are equal to or lower than pretreatment values [1,2]. In addition, many studies have shown an improvement in cellular insulin binding in patients treated with hypoglycemic sulfonylureas [3]. To explain this improvement it has been suggested that these drugs may inhibit the internalization of ligand-bound receptors [4].

0168-8227/88/$03.50 :(', 1988 ElsevierScience Publishers B.V. (Biomedical Division)

32 Recently, transglutaminase, an ubiquitous calcium responsive enzyme, was implicated in the mechanism of receptor mediated endocytosis [5-8]. Also tolbutamide, a hypoglycemic sulfonylurea, was found to inhibit transglutaminase activity in fibroblasts. In addition, high concentrations of this agent were able to block receptor mediated c~z-macroglobulin internalization [9]. The present study was undertaken to assess the effect of a continued therapy with glipizide, a hypoglycemic sulfonylurea, upon transglutaminase activity in human red blood cells.

Material and methods

Slt/gj£,cls Six patients with diabetes mellitus type 2 were studied. Their age averaged 49.6 + 2.8 years, and the clinical evolution of the disease was 1.8 + 0.5 years. All subjects had a body mass index of 22.3-26.0 (kg/m-') and prior to the study were being managed only with isocaloric, weight maintaining diet, comprised of 45% carbohydrate, 40% fat, and 15% protein. None of them was taking any medication nor did any of them present any specific complication of diabetes mellitus. Patients were followed up to assess metabolic control. They received the diet during 48 h prior to initiation of the study. Venous samples were drawn in basal conditions, and every 15 days throughout the glipizide treatment (5 mg/day per os, at breakfast) for 3 months. At the end of this period, patients returned to their isocaloric diet for 15 days, after which a new blood sample was obtained. In every sample, transglutaminase activity was measured, its values expressed as a percentage of the values obtained from seven normal subjects in the same assay.

pended in 800 ,ul of the washing buffer (Tris-HC1 50 mM, pH 7.4, containing 225 mM sucrose) and lysed by sonication (5 x 5 s). The hornogenate was centrifuged (for 2 min at 800 x g), and an aliquot (30 itl) of the supernatant was mixed with 70 ,ul of the reaction mixture to yield the following final concentrations: 50 mM Tris-HC1 (pH 7.4), 67.5 mM sucrose, 1.0% N-N-dimethylcasein (w/v; Calbiochem, La Jolla, CA 92037) and [2,5-3H]histamine providing 5 llCi/rnl. The concentration of ionized Ca 2+ in the reaction mixture was calculated from its content in CaCI2 and EDTA (0 0.5 mM). After incubation at 37°C for 30 rain, the reaction was stopped by the addition of 1.0 ml trichloroacetic acid (10% w/v); after centrifugation (4°C, for 5 rain at 800 x g) the precipitate was washed five times with trichloroacetic acid (1 ml), resuspended in 0.5 rnl N a O H (0.5 M), transferred together with 1.0 ml H 2 0 to a counting vial, and examined for its radioactive content by liquid scintillation. All results

b

CO

2

.D

-1-

0 J~ "1"

9

E

E m 0

E

0

E

6 (J 0

E ~-£ °~ 0

>

Transglutaminase assay All experiments were performed with human red blood cells obtained from fresh heparinized venous blood, and samples were kept chilled in iced water until centrifugation (800 x g for 10 min at 4°C). For measuring transglutaminase (E.C. 2.3.2.13) activity, 200 I~1 of human red blood cells were sus-

12

O~

I

I

I

i

20

15

10

5

1/Histamine

(ram -~)

I

I

I

I

I

I

0.5

1

2

3

4

5

Histamine

mM

Fig. I. Double reciprocal plot for tile incorporation of [2.53H]laistamineinto N.N-dirnethylcasein.Mean wtlues are derived from two individual experinaents.

33 were corrected for the 'blank value" obtained in the presence of red blood cell lysate but in the absence of Ca -,+ ( < 0 . 0 5 I~M). The incorporation of histamine into N-N-dimethylcasein was proportional to both the arnount of tissue added to each sample (1.36-5.44 mg Hb, 4.52-18.10 × 107 red blood cells) and to the length of the incubation period (10-40 rain). Experimental results are expressed as the mean value (X + SEM) with the number of individual determinations. The statistical significance was assessed by Student's t-test.

% Inhibition 25" 2015-

10-

s/

0-

Results

Transglutamhutse activi O' in human red blood cells In the presence of 0.5 mM Ca 2 + human red blood cell lysates catalyzed the incorporation of [2,53H]histamine into N-N-dimethylcasein. As depicted in Fig. 1, K~ for histamine was 0.95 mM and V m a x reached 2.5 nmol/30 rain per mg Hb. The reaction was Ca 2 + dependent and no significant incorporation could be detected at concentrations close to 0.05/~M or less. At increasing Ca-' + concentrations the reaction velocity yielded a hyperbolic dose-response relationship (Fig. 2).

20 b~.

0.5

ot ~

-r

1-

15 E

c0

E

O

~.0.3

¢Io E

0

E

t-

5~, >

~ oj 4b i i

i

i

i

0.25

0.5

3'0

2'0

1/Ca2, mM 4

10

~ N % , %

,

j

j

0.25.5251

i

i

i

3 5 10 rGlipizid¢3(mM)

Fig. 3. Transglutaminase activity in human red blood cell lysates. Logarithmic dose-response curve for inhibition by glipizide. Mean values arc derived from three individual experiments.

All experiments were conducted in the presence of 0.5 mM [2.5-3H]histamine and 0.5 mM Ca-'+.

In the presence of 0.5 mM [2,5-3H]histamine the half-maximal stimulatory effect of Ca 2+ was reached at 0.125 mM.

h7 vitro effect o/'glipizide upon transglutamhmse activity Glipizide inhibited the incorporation of [2,53H]histamine into N-N-dimethylcasein catalyzed by human red blood cell lysates. Fig. 3 illustrates the dose action relationship for the effect of glipizide upon reaction velocity. At fixed concentrations of 0.5 mM of Ca-" + and 0.5 mM of [2,5-3H]histamine (near the Km for histamine and the half-maximal stimulatory effect for Ca z +), the relative extent of inhibition was maximal at glipizide concentrations close to 1 mM. At this drug concentration the inhibitory effect upon enzyme activity was unrelated to the Ca 2+ presence (0.1-2 mM) in the reaction mixture (data not shown).

i

1 Ca2"mM

Fig. 2. Ca 2+ dependency of the incorporation of [2,53H]histamine (0.5 raM) into N.N-dimethylcasein. Mean values refer to three individual measurements.

ExperhTwnts with httact red blood cells Intact red blood cells were first incubated for 60 min at 37°C at Ca z+ concentration 1.0 mM and in the presence or absence of glipizide (0.05-1 mM);

34

%Transglutamjnase actiwty 100-

TABLE 1 IN VITRO EFFECT OF GLIPIZIDE UPON T R A N S G L U TAMINASE ACTIVITY IN INTACT H U M A N RED BLOOD CELLS Drug

Concentration (mmol/I)

Velocity (% of control)

Glipizide

0.05 0.1

106.0 ( I ) 97.2 4- 1.0(3)

1

802 602 2° 2

91.6 4- 3.0 (6)

All measurements were performed after preincubation of intact human red blood cells during 60 min in the presence of glipizide. Prior to being lysed, the red blood cells were washed twice with Tris-HCI buffer (50 mM, pH 7.4) containing 225 mM sucrose. The medium used in the preincubation period was: Hepes 25 mM, NaCI 115 mM, KCI 5 raM, MgCI2 1 raM, CaClz I mM, and glucose 5.6 mM (pH 7.4). Glipizide was dissolved in DMSO (DMSO in the preincubation medium never exceeded 1 ld/ml). Determination of transglutaminase activity was performed in the presence of 0.5 mM Ca 2* and 0.5 mM [2,5--~H]histamine. Mean values (Y~ 4- SEM) are expressed as percentage of the paired control value found in the absence of glipizide and are given together with the number of observations (in parentheses).

3b r:b s ies "

Ti me (days)

Fig. 4. The effects of glipizide treatment on red blood cell trmlsglutamJnase activity were tested in the presence ot"0.5 mM C a 2 + and 0.5 mM [2,5-3H]histamine. Mean values (X 4- SEM) are expressed as percentage of the values obtained from seven normal subjects in the same assay.

but rapidly reverted to basal values upon discontinuation of glipizide treatment.

Discussion

then the cells were washed twice with buffer (TrisHC1 50 raM, pH 7.4, sucrose 225 mM) and lysed by sonication (5 x 5 s). The transglutaminase activity in the lysates was inhibited by the presence of the sulfonylurea (Table 1).

Effect of glipizide upon transghaaminase activity h7 diabetic patients Before treatment with glipizide, the transglutaminase activity in the red blood cells was the same in patients with type 2 diabetes mellitus (X + SEM, 395.8 + 19.1 pmol/mg Hb/60 min) as in seven normal subjects (X + SEM, 432.0 + 15.5 pmol/mg Hb/60 min) used as controls. The values of transglutaminase activity in patients with type 2 diabetes mellitus at the beginning, during and after treatment are shown in Fig. 4. A significant decrease in transglutaminase activity was observed after the 45th day of glipizide treatment (P < 0.01). This decrease was sustained,

Transglutaminase is an ubiquitous enzyme which catalyzes the cross-linking of proteins and participates in a number of biological processes such as blood coagulation, post-ejaculatory clotting of sperm, insulin secretion and receptor mediated endocytosis [5-8]. The present study confirms the presence of transglutaminase in red blood cells [10], and indicates that glipizide inhibits transglutaminase activity from human blood red cell lysates. Schrier and Junga [10] have demonstrated that in vitro preincubation with tolbutamide does not inhibit transglutaminase activity in human red blood cells. One possible explanation for this discrepancy with our results could be the difference in in vitro concentrations of hypoglycemic sulfonylurea u~ed according to the detected low affinity of tolbutamide for transglutaminase in fibroblast. These authors investigated the effect of tolbutamide at high concentrations of the drug, namely 20 mM. How-

35 ever, in our experiments in red blood cell homogenates it can be observed that the maximum inhibitory effect exerted by glipizide was obtained at a concentration of 1 raM. Therefore, the preincubation of red blood cells was carried out at low glipizide concentrations finding little inhibition of transglutaminase activity (10% of the controls). On the other hand, the treatment of patients with diabetes mellitus type 2 with glipizide produced a significant decline in red blood cell transglutaminase values. This is striking since glipizide, in vitro, at equivalent concentrations to the therapeutic plasma levels (1 ,ttM) [l l] did not affect transglutaminase activity. To explain these contradictory results it could be conceivable that some glipizide metabolite causing a higher inhibitory effect by itself could exist or, more probably, that the hypoglycemic agent requires its continued presence to suppress transglutaminase activity. In this perspective, the fact that inhibition of human red blood cell transglutaminase by glipizide in patients with diabetes was observed after 30 days of treatment would suggest that the effect of the sulfonylurea could be related to the action upon red blood cell precursors. On the other hand, a recent article described that transglutaminase activity of rat hepatic tissue was unaffected by in vivo administration of glipizide [12]. However, this statement could be questioned since transglutaminase activity was measured at high Ca -,+ concentrations (close to 10 raM), near to the maximum velocity for the enzyme, making the observation of the inhibitory effect exerted by glipizide difficult. Obviously the question as to whether the extrapancreatic action of hypoglycemic sulfonylureas is related to the inhibitory effect of these pharmacological agents upon transglutaminase activity must be raised. Several studies have demonstrated the participation of transglutaminase in receptor mediated endocytosis, a process responsible for the cellular uptake of a variety ofpolypeptides including insulin [8-10]. In this way, it is conceivable that an inhibition of the normal internalization of [igandbound receptors could contribute to explain the en-

hanced insulin-binding detected [4] after sulfonylurea treatment, In this perspective, the fact that administration of glipizide to patients with type 2 diabetes mellitus inhibits transglutaminase activity would suggest that the extrapancreatic effect of sulfonylureas is mediated by this enzyme. It could also be suggested that glipizide induced changes in transglutaminase activity were secondary to the normalization of glucose plasma levels. However, in another experimental study, we treated obese patients with hypoglycemic agents and found that the specific inhibition of transglutaminase was not related to changes in plasma glucose levels (Gomis et al., unpublished observation), On the other hand, Davies et al. [13] using high levels of tolbutamide ranging from 5 to 20 mM demonstrated inhibition of transglutaminase activity. Although at these concentrations tolbutamide is known to inhibit protein synthesis as well, in our experiments, the inhibitory effect of glipizide had already been observed at concentrations close to 0.5 mM, levels at which no toxic cellular effects have been described. In conclusion, the present work suggests that treatment of type 2 diabetes mellitus with hypoglycemic sulfonylureas may modify insulin binding by inhibiting the cellular transglutaminase levels. This proposal does not rule out the participation of other regulatory factors in the extrapancreatic effect of hypoglycemic sulfonylureas.

Acknowledgements This work was supported in part by grants from 'Fondo Investigaciones Sanitarias' of the Spanish Ministery of National Health. The authors are grateful to R.M. Viaplana for secretarial help.

References I Pfeifer, M.A., Halter, J,B., Judzewitsch, R.G. et al. (1984) Acute and chronic effectsofsulfonylureadrugs on pancreatic islet function in man. Diabetes Care 7 (Suppl. l), 25-34. 2 Feldman, J.M. and Lebovitz. H.E. (1971) Endocrine and

36 metabolic effects of glybenclamide. Diabetes 20, 745-755. 3 Kolterman, D.G. and Olefsky, J.M. (1984) The impact of sulfonylurea treatment upon the mechanism responsible for the insulin resistance in type II diabetes. Diabetes Care 7 (Suppl. I), 81-88. 4 Prince, M.J. and Olefsky, J.M. (1980) Direct in vitro effect of a sulfonylurea to increase human fibroblast insulin receptors. J. Clin. Invest. 66, 608-611. 5 Folk, J.E. (1980)Transglutaminases. Ann. Rev. Biochem. 49, 517-531. 6 Sener, A., Dunlop, M.E., Gomis, R., Mathias, P.C.F., Malaisse-Lagae, F. and Malaisse, W.J. (1985) Role of transglutaminase in insulin release. Study with glycine and sarcosine methylesters. Endocrinology 117, 237-242. 7 Paonessa, G., Metafora, S., Tajana, G. et al. (1984) Transglutaminase-mediated modifications of the rat sperm surface in vitro. Science 226, 852-855. 8 Levitzki, A., Willingham, M. and Pastan, I. (1980) Evidence for participation of transglutaminase in receptor-mediated endocytosis. Proc. Natl. Acad. Sci. U.S.A. 77, 2706-2710.

9 Davies, P.J.A., Davies, D.R., Levitzki, A. et al. (1980) Transglutaminase is essential in receptor-mediated endocytosis of 2-macroglobulin and polypeptide hormones. Nature 283, 162-167. 10 Schrier, S.L. and Junga, I. (1981) The role of transglutaminase in human erythrocyte endocytosis. Biochem. Biophys. Res. Commun. 99, 58-64. 11 Skillman, T.G. and Feldman, J.M. (1981) The pharmacology of sulfonylureas. In: J.S. Skyler and G.F. Cahill (Eds.), Diabetes Mellitus. Yorke Medical Books, Boston, MA, pp. 107-118. 12 Novak, S.M., McCaleb, M.L. and Lockwood, D.H. (1983) Extrapancreatic action of sulfonylureas: hypoglycaemic effects are not dependent on altered insulin binding or inhibition of transglutaminase. Metabolism 32, 398-401. 13 Davies, P.S.A., Cornwell, M.M., Johnson, J.D., Reggiamni, A., Myers, M. and Murthaugh, M.P. (1984) Studies on the effects of dansylcadaverine and related compounds on receptor-mediated endocytosis in cultured cells. Diabetes Care 7 (Suppl. 1), 35-41.