Biochimica et Biophysica Acta, 1139(1992) 65-69
65
© 1992 Elsevier Science Publishers B.V. All rights reserved 0925-4439/92/$05.00
BBADIS 61150
Modifications in platelet membrane transport functions in insulin-dependent diabetes mellitus and in gestational diabetes Laura Mazzanti a, Roberto Staffolani a, Rosa A. Rabini a, Anna M. Cugini b, Nelvio Cester b, Carlo Romanini b, Emanuela Faloia c and Roberto D e Pirro c Istituto di Biochimica, Facolth di Medicina e Chirurgia, Ancona (Italy), b Istituto di Clinica Ostetrica e Ginecologica, Facolth di Medicina e Chirurgia, Ancona (Italy) and c Clinica di Endocrinologia, Facolth di Medicina e Chirurgia, Ancona (Italy)
(Received 26 September 1991)
Key words: Cell membrane; Platelet; Insulin therapy; Insulin-dependent diabetes mellitus; Gestational diabetes mellitus The pathogenesis of plasma membrane alterations present in diabetes mellitus is unclear. To add new insights to the question, platelet membrane properties were evaluated in 16 women presenting impaired glucose tolerance at the 28-29th week of gestation (GDM) and in 8 women with insulin-dependent diabetes mellitus (IDDM). 15 healthy pregnant women (HPW) and 21 healthy non-pregnant (HNPW) women were the control group for GDM and IDDM, respectively. Pregnancy (HPW vs. HNPW) provoked an increase in Ca2+-ATPase activity and a decrease in membrane fluidity; in contrast, Na+/K+-ATPase, intracellular free Ca 2+ concentrations, membrane cholesterol and phospholipid content did not vary. Both GDM and IDDM showed lower Na+/K+-ATPase activity and higher Ca 2÷ concentration, compared to HPW and HNPW, respectively, whereas Ca2+-ATPase activity was higher only in IDDM; furthermore, membrane fluidity was lower in GDM and higher in IDDM. Finally, GDM showed higher membrane cholesterol content. Both GDM and IDDM showed a very good metabolic control so that variations reported cannot be due to hyperglycemia; it is tempting to suggest that membrane variations are present before the clinical metabolic alteration. Furthermore, both GDM and IDDM were on insulin therapy, therefore: (i) insulin may be the pathogenetic factor of higher intracellular free Ca 2÷ concentrations and lower Na+/K+-ATPase activity since they both varied accordingly in GDM and IDDM, but not of (ii) changes in Ca2+-ATPase, membrane fluidity and cholesterol content which did not vary accordingly in GDM and IDDM.
Introduction Previous studies showed that cells from insulin-dependent ( I D D M ) and non-insulin-dependent diabetic ( N I D D M ) patients present alterations in N a + / K +ATPase activity, Ca2+-ATPase activity, intracellular free Ca 2+ concentrations and membrane fluidity [1-6]. It has been reported that these alterations may be induced by hyperglycemia both in vivo and in vitro [7-9]. Similar plasma membrane changes, on the other hand, have been found in I D D M patients on good metabolic control [10]; therefore, they might appear independently from the presence of overt hyperglycemia. Prompt treatment of pregnant women at
Abbreviations: GDM, gestational diabetes mellitus; IDDM, insulindependent diabetes mellitus; HPW, healthy pregnant women; HNPW, healthy non-pregnant women; NIDDM, non-insulin-dependent diabetes mellitus; PRP, platelet-rich plasma. Correspondence: L. Mazzanti, Istituto di Biochimica, Facolt~ di Medicina e Chirurgia, Via Ranieri 60131, Ancona, Italy.
presentation of altered glucose tolerance lead us to study a proper model to evaluate membrane modifications induced by diabetes mellitus, but not by hyperglycemia. Furthermore, in order to evaluate the influence of insulin therapy, a group of I D D M patients was also studied.
Materials and Methods Patients
16 women who satisfied criteria for the diagnosis of gestational diabetes mellitus (GDM) [11] (28 + 5 years), 15 healthy pregnant women (HPW) (29 + 5 years), 8 w o m e n with insulin-dependent-diabetes mellitus ( I D D M ) (35 + 6 years) and 21 healthy nonpregnant women (HNPW) (32 + 4 years) were studied. G D M showed impaired glucose tolerance at the 28th or 29th week of gestation. Immediately after the diagnosis, patients were treated with insulin therapy in order to obtain a permanent very good metabolic control during the third trimester of pregnancy, with glycosylated hemoglobin ( H b A l c ) in the normal range and
66 glycemia less than 120 m g / d l in the morning, at the time of delivery. All IDDM patients received insulin treatment and were in good metabolic control ( H b A l c in the normal range and glycemia less than 120 m g / d l in the morning). All the subjects studied were normotensive and had a negative family history for diabetes mellitus, hypertension and obesity. They were taking no drug but insulin for IDDM and G D M patients. Blood samples were collected at 8.00 a.m. after an overnight fasting; particularly, in G D M and H P W patients they were obtained during the 30th week of pregnancy. H b A l c was measured by high-performance liquid chromatography according to the method of Akai [12].
Platelet isolation procedure Platelets were prepared as previously described [5]. Briefly, platelet-rich plasma (PRP) was obtained after centrifugation at room temperature of whole blood mixed with 9:1 v / v CCD (CCD:citrate 0.1 M, citric acid 7 mM, dextrose 140 raM, pH 6.5) for 20 min at 500 × g . The supernatant (i.e., platelet rich plasma, PRP) was then centrifuged at 2000 × g in anti-aggregating buffer (Tris-HCl 10 mM, E D T A 1 mM, NaCI 150 mM, glucose 5 mM, pH 7.3) and centrifuged at 2000 X g for 10 min. The pellet representing platelets was suspended in a buffer containing NaC1 145 mM, KC1 5 mM, MgSO 4 1 raM, Hepes 10 mM, glucose 10 mM, pH 7.4.
(Rmi n) and in the presence of calcium {R ...... ): Sf2 and
Sb2 = fluorescence intensities at 380 nm in the absence of (Sf2) and in the presence of calcium tSb2); and R ...... and Sf2 were measured after cell lysis with 5c~ Triton X-100 and addition of E G T A 10 mM (pH 8.3): R ...... and Sb2 were measured after cell lysis and addition of CaC1, 10 mM.
Platelet membrane preparation The preparation of platelet plasma membranes was performed according to the method of Enouf et al. [15]. Platelets were pelleted by centrifugation of the platelet-rich plasma for 15 min at 3000 × g. The pellet was washed twice in modified Tyrode's buffer (pH 7.5) containing 130 mM NaCI, 5 mM KC1, 1 mM NaH2PO 4, 24 mM NaHCO3, 2 mM Na2EDTA, 10 mM glucose, 12.5 mM sucrose and 0.35% bovine serum albumin (w/v). The cells were then lysed by ultrasonication and the lysate was centrifuged at 19000 x g to eliminate unlysed platelets, mitochondria and granules. The supernatant was centrifuged at 100 000 × g and the pellet obtained was then layered over a 40% (w/v) sucrose solution and centrifuged again at 100000 × g for 12(I min. The interface membrane subfraction obtained after the centrifugation consisted of plasma membrane, according to the biochemical and immunological characterization performed by Enouf et al. [15]. This subfraction was then resuspended in phosphate buffer (pH 7.2). Ca 2 +-ATPase
Cytosolic free calcium concentrations Ionized calcium (Ca 2+) in blood platelets was measured according to the method of Rao [13]. For the loading of the Ca2+-sensitive probe, Fura 2 AM, the platelets were incubated for 45 min with the acetoxy methyl ester of the dye at 1 /xM in a solution containing NaC1 145 mM, KC1 5 mM, MgSO 4 1 mM, Hepes 10 mM, glucose 10 mM. Cells were then washed in the same solution to remove the excess dye. The determination of intracellular Ca 2+ levels was performed in a Perkin Elmer MPF-66 fluorescence spectrophotometer at 37°C according to the method of Grynkiewicz et al. [14]. The fluorescence intensity was read at a constant emission wavelength (510 nm) with changes in the excitation wavelength (340 and 380 nm). The calibration was carried out as described by Grynkiewicz et al. [14] using the following equation: [Ca 2+ ]i = Kd
R-Rmi n Rmax _ R
Sf2 Sb2
where K d (dissociation constant of Fura 2) = 224 nM; R = ratio of the fluorescence intensities at the excitation wavelengths of 340 and 380 nm; Rmi n and Rma x = ratio of the fluorescence intensities in the absence
assay
The activity of the Ca2+-ATPase was determined according to the method of Davis et al. [16] by measuring inorganic phosphate (Pi) hydrolysed from Na2ATP 1 mM at 37°C in the presence and absence of Ca 2+ 0.15 mM. The reaction medium contained E G T A 0.1 mM, NaCI 75 mM, KC1 25 mM, MgC12 1 mM and Tris-HCl 25 mM (pH 7.4). The ATPase activity assayed in the absence of Ca 2+ was subtracted from the total ATPase activity to calculate the activity of Ca 2+ATPase. Results are expressed as p.mol Pi/mg membrane proteins per 90 min. Inorganic phosphate was measured according to Fiske and SubbaRow [17]. Protein concentration was determined by the Lowry method [18], using albumin as standard.
Na + / K +-A TPase assay N a + / K + - A T P a s e activity was determined by a modification of the Kitao method [19], as previously described [20]. The ATPase activity was assayed by incubating membranes at 37°C in 1 ml of medium (MgCI 2 5 mM, NaC1 140 mM, KC1 14 mM, in Tris-HC1 40 mM, pH 7.7). The ATPase reaction was started by the addition of 3 mM N a z A T P and stopped 20 min later by the addition of 1 ml of trichloroacetic acid 15%. Inorganic phosphate (Pi) hydrolyzed from reaction was
67
measured as previously described [17]. The ATPase activity assayed in the presence of 10 mM ouabain was subtracted from the total Mg2+-dependent ATPase activity to calculate the activity of the ouabain-sensitive Na+/K+-ATPase. Results are expressed as ~mol Pi/mg membrane protein per 60 min.
Membrane cholesterol and phospholipids Lipids were extracted from platelet membranes according to the method of Folch et al. [21] with 10 ml chloroform/methanol (2:1 v / v ) / m l membrane suspension. Cholesterol concentration was then determined by the method of Zak [22] and total phospholipids by the method of Bartlett [23].
*
C
-r- y
e
ill
F O
E =k ,D 0.150.
II
o~ w it. I--
"-m U
0
Membrane fluidity Membrane fluidity was determined by means of the measurement of the fluorescence polarization (P) of the lipophilic probe 1,6 diphenyl-l,3,5-hexatriene (DPH), according to the method of Schachter et al. [24]. The fluorescence polarization measurements were made in a Perkin-Elmer spectrofluorimeter MPF 66 equipped with two quartz prism polarizers, with exciting light at 365 nm. The P level is a quantitative index of the freedom of rotation of the fluorescent probe; a decrease in the P value indicates a higher mobility of the DPH in the deeper part of the membrane hydrophobic bilayer (i.e., increased membrane fluidity) and vice versa. Statistical analysis Statistical analysis were performed by using the Student's t-test for unpaired data.
GDM
HPW
HNPW
IDDM
Fig. 1. Ca2+-ATPase activity in platelet plasma membranes obtained from patients affected by gestational diabetes mellitus (GDM), healthy pregnant women (HPW), healthy non-pregnant women (HNPW) and insulin-dependent diabetic patients (IDDM). The effect of pregnancy was evaluated comparing HPW with HNPW; the effect of diabetes mellitus was evaluated comparing GDM and IDDM with HPW and HNPW, respectively. Results are expressed as mean + S.D. * P < 0.001.
parison with HPW (Fig. 1). Comparing IDDM with HNPW, IDDM showed higher membrane fluidity (Fig. 2), but unchanged cholesterol and phospholipid content (Table I). GDM showed lower fluidity than HPW (P < 0.001) (Fig. 2), higher cholesterol content (P < 0.001) (Table I), but unchanged phospholipid content.
4(" 4f
Results 0.300,
The effect of pregnancy on platelet properties was evaluated comparing healthy pregnant women (HPW) with healthy non-pregnant women (HNPW). Platelet membranes obtained from HPW showed higher Ca 2÷ATPase activity (P < 0.001) (Fig. 1) and lower membrane fluidity (P < 0.001) (Fig. 2) in comparison to HNPW. No difference was observed between the two groups in platelet Ca 2÷ concentration (Fig. 3), Na+/K+-ATPase activity (Fig. 4), cholesterol and phospholipid content (Table I). The effect of diabetes mellitus on platelet properties was evaluated comparing women presenting gestational diabetes mellitus (GDM) with HPW, and women affected by insulin-dependent diabetes mellitus (IDDM) with HNPW. Both GDM and IDDM showed lower Na+/K+-ATPase activity (P < 0.001) (Fig. 4) and higher intracellular free Ca 2÷ concentration (P < 0.001) than healthy pregnant and non-pregnant controls, respectively (Fig. 3). The Ca2+-ATPase activity was higher in IDDM than in HNPW (P < 0.001), but unchanged in GDM in com-
I-Ill
T -r"
Z
(2_
-I"
I--
"i"
~N 0.150. o
L
GDM
HPW
HNPW
IDDM
Fig. 2. Membrane fluidity in platelet plasma membranes obtained from patients affected by gestational diabetes mellitus (GDM), healthy pregnant women (HPW), healthy non-pregnant women (HNPW) and insulin-dependent diabetic patients (IDDM). The effect of pregnancy was evaluated comparing HPW and HNPW; the effect of diabetes mellitus was evaluated comparing GDM and IDDM with HPW and HNPW, respectively. A decrease in the parameter value indicates an increase in membrane fluidity and vice versa. Results are expressed as mean + S.D., * P < 0.001.
6~ TABLE 1 *
Membrane choh, sterol and phospholipid content
200-
.
-r
0
E
"I"
I--
z "' I-Z
100. "r"
Platelets were obtained from healthy pregnant women (HPW), patients with gestational diabetes mellitus (GDM), healthy non-pregnant women (HNPW) and insulin-dependent diabetes patients (IDDM). Results are expressed as mean _+S.D. The difference in the m e m b r a n e cholesterol content between t-IPW and G D M patients was statistically significant ( * P < (1.001).
-Ir-
o
HPW GDM HNPW IDDM
U U
Cholesterol ( n m o l / l )
Phospholipids (nmol/1)
102.(I _+4.6 137.0_+5.1 * 115.2+_8.4 125.0+4.7
171.4 _+3.2 172.3_+3.t 173.1 _+5.8 192.3+7.(/
0
GDM
HPW
HNPW
IDDM
Fig. 3. Ca 2+ concentration in platelets obtained from patients affected by gestational diabetes mellitus (GDM), healthy pregnant women (HPW), healthy non-pregnant women ( H N P W ) and insulindependent diabetic patients (IDDM). The effect of pregnancy was evaluated comparing H P W with H N P W ; the effect of diabetes meUitus was evaluated comparing G D M and I D D M with H P W and H N P W , respectively. Results are expressed as mean _+S.D. * P < 0.001.
Discussion
Comparing the membrane platelets from H P W with HNPW, H P W showed increased Ca2+-ATPase, but similar N a + / K + - A T P a s e and similar intracellular free Ca 2÷ concentrations. The lack of change in intracellular free Ca 2+ concentrations led us to suggest that
1.5
2_± n
T
o.
_L
w o
E =k
g g
0.5
t,
0
GDM
HPW
HNI~/
IDDM
Fig. 4. N a + / K + - A T P a s e activity in platelet plasma m e m b r a n e s obtained from patients affected by gestational diabetes mellitus (GDM), healthy pregnant w o m e n (HPW), healthy non-pregnant women (HNPW) and insulin-dependent diabetic patients (IDDM). The effect of pregnancy was evaluated comparing H P W with H N P W ; the effect of diabetes mellitus was evaluated comparing G D M and I D D M with H P W and H N P W , respectively. Results are expressed as mean _+S.D. * P < 0.001.
increased influx of Ca 2+ is balanced by the increased Ca2+-ATPase. Platelets from pregnant women showed lower membrane fluidity without modifications in cholesterol a n d / o r phospholipid content. The reason for changes in membrane fluidity is unknown. These data point out that pregnancy may cause variations in plasma membrane structure and function. G D M presented an inverse situation in which intracellular free Ca z+ concentration was increased, but CaZ+-ATPase was unchanged. Both G D M and IDDM showed decreased N a + / K + - A T P a s e thus confirming previous results obtained on platelets and erythrocytes [5,9,10,20]. Previously, in non-pregnant diabetic patients, membrane fluidity was reported decreased [2527], increased [28,10,5] or not changed with respect to healthy subjects [29]. In the present study membrane fluidity was found decreased in G D M compared to HPW, but increased in IDDM compared to HNPW. The decreased N a + / K + - A T P a s e observed in both G D M and IDDM cannot derive from changed membrane fluidity but, probably, from a more specific effect on the enzyme induced by diabetes mellitus. The present data point out that platelet plasma membranes show various alterations in G D M despite the absence of an overt period of hyperglycemia or altered metabolic control; in fact, G D M patients were euglycemic in basal conditions and had a really short period of impaired glucose tolerance which in turn did not produce any alteration in both the length of pregnancy a n d / o r the metabolic control during the third trimester (see Materials and Methods). This would suggest that membrane alterations may be present before the onset of impaired glucose tolerance or may be induced by insulin treatment or other unclear factors. In this respect, it is noteworthy that both IDDM and G D M were treated with insulin, but showed different variations in some membrane parameters (CaZ+-ATPase, membrane fluidity, cholesterol content); therefore, insulin cannot have a key-role in inducing those alterations. On the contrary, a relationship between the intracellular free Ca 2+ concentrations and the
69 Na+/K+-ATPase may be possible. Finally, both diabetic groups were on good metabolic control so that alterations described cannot be produced by hyperglycemia. In conclusion, these data suggest that diabetes and pregnancy may affect platelet plasma membrane function. The differences between GDM and IDDM would support the thesis that those are two different disease states [11].
Acknowledgements The present work was supported by a Grant of Regione Marche (LM and NC); C.N.R. No. 91.00345. PF 40 (RDP) and progetto Ricerca Scientifica 40% 1991 (RDP).
References 1 Finotti, P. and Palatini, P. (1987) Diabetologia 36, 991-995. 2 Rahmani-Jourdheuil, D., Mourayre, Y., Vague, P., Boyer, J. and Juhan-Vague, I. (1987) Diabetes 36, 991-995. 3 Cohen, M.P. and Klepser, H. (1988) Diabetes 37, 558-562. 4 Testa, I., Rabini, R.A., Fumelli, P., Bertoli, E. and Mazzanti, L. (1988) J. Clin. Endoerinol. Metab. 67, 1129-1133. 5 Mazzanti, L., Rabini, R.A., Faloia, E., Fumelli, P., Bertoli, E. and De Pirro, R. (1990) Diabetes 39, 850-854. 6 Bergh, C.H., Hjialmarson, A., Holm, G., Anguald, E. and Jacobsson, B. (1988) Eur. J. Clin. Invest. 18, 92-97. 7 Watala, C. (1988) Exp. Pathol. 33, 233-238. 8 Jain, S.K., McVie, R., Duett, J. and Herbst, J.J. (1989) Diabetes 38, 1539-1543.
9 Jain, S.K. (1989) J. Biol. Chem. 264, 21340-21345. 10 Mazzanti, L., Rabini, R.A., Testa, I. and Bertoli, E. (1989) Eur. J. Clin. Invest. 19, 84-89. 11 National Diabetes Data Group (1979) Diabetes 28, 1039-1057. 12 Akai, T. (1982) Glycosylated hemoglobin vensonic seminar proceeding 1. Tokyo, Daichii Kagaku, p. 17. 13 Rao, G.H.R. (1988) Anal. Bioehem. 169, 400-404. 14 Grynkiewicz, G., Poenie, M. and Tsien, R.Y. (1985) J. Biol. Chem. 6, 3440-5034. 15 Enouf, J., Bredoux, R., Bourdeau, N., Sarkadi, B. and LevyToledano, S. (1989) Bioehem. J. 263, 547-552. 16 Davis, F.B., Davis, P.J., Nat, G., Bias, S.D., Mac Gillivray, M., Gutman, S. and Feldman, M.J. (1985) Diabetes 34, 639-646. 17 Fiske, C., SubbaRow, Y. (1925) J. Biol. Chem. 193, 375-400. 18 Lowry, O.H., Rosenburg, M.Y., Farr, A.L. and Randall, R.T. (1951) J. Biol. Chem. 193, 265-275. 19 Kitao, T. and Hattori, K. (1983) Experientia 39, 1362-1364. 20 Testa, I., Rabini, R.A., Danieli, G., Tranquilli, A.L., Cester, N., Romanini, C., Bertoli, E. and Mazzanti, L. (1988) Scand. J. Clin. Lab. Invest. 48, 7-14. 21 Folch, J., Less, M. and Sloane-Stanley, G.H. (1957) J. Biol. Chem. 226, 466-468. 22 Zak, B. (1957) Am. J. Clin. Pathol. 27, 583-588. 23 Bartlett, G.R. (1959) J. Biol. Chem. 234, 466-468. 24 Schachter, D. and Shinitzky, M. (1977) J. Clin. Invest. 59, 536-548. 25 Bryszewska, M., Watala, C. and Torzzcka, W. (1986) Br. J. Haematol. 62, 111-116. 26 Baba, Y., Kai, M., Kamada, T. Setoyama, S. and Otsuji, S. (1979) Diabetes 28, 1138-1140. 27 Winocour, P.D., Bryszewska, M., Watala, C., Rand, M.L., Epand, R.M., Kinlough-Rathbone, R.U, Packham, M.A. and Mustard, J.F. (1990) Diabetes 39, 241-244. 28 Freyburger, G., Gin, H., Dousseau, F., Lorient-Roudault, M.F. and Boisseau, M.R. (1988) Clin. Hemorheol. 8, 159-164. 29 Baldini, P., lncerpi, S., Lambert-Gardini, S., Spinedi, A. and Luly, P. (1989) Diabetes 38, 825-831.
65
© 1992 Elsevier Science Publishers B.V. All rights reserved 0925-4439/92/$05.00
BBADIS 61150
Modifications in platelet membrane transport functions in insulin-dependent diabetes mellitus and in gestational diabetes Laura Mazzanti a, Roberto Staffolani a, Rosa A. Rabini a, Anna M. Cugini b, Nelvio Cester b, Carlo Romanini b, Emanuela Faloia c and Roberto D e Pirro c Istituto di Biochimica, Facolth di Medicina e Chirurgia, Ancona (Italy), b Istituto di Clinica Ostetrica e Ginecologica, Facolth di Medicina e Chirurgia, Ancona (Italy) and c Clinica di Endocrinologia, Facolth di Medicina e Chirurgia, Ancona (Italy)
(Received 26 September 1991)
Key words: Cell membrane; Platelet; Insulin therapy; Insulin-dependent diabetes mellitus; Gestational diabetes mellitus The pathogenesis of plasma membrane alterations present in diabetes mellitus is unclear. To add new insights to the question, platelet membrane properties were evaluated in 16 women presenting impaired glucose tolerance at the 28-29th week of gestation (GDM) and in 8 women with insulin-dependent diabetes mellitus (IDDM). 15 healthy pregnant women (HPW) and 21 healthy non-pregnant (HNPW) women were the control group for GDM and IDDM, respectively. Pregnancy (HPW vs. HNPW) provoked an increase in Ca2+-ATPase activity and a decrease in membrane fluidity; in contrast, Na+/K+-ATPase, intracellular free Ca 2+ concentrations, membrane cholesterol and phospholipid content did not vary. Both GDM and IDDM showed lower Na+/K+-ATPase activity and higher Ca 2÷ concentration, compared to HPW and HNPW, respectively, whereas Ca2+-ATPase activity was higher only in IDDM; furthermore, membrane fluidity was lower in GDM and higher in IDDM. Finally, GDM showed higher membrane cholesterol content. Both GDM and IDDM showed a very good metabolic control so that variations reported cannot be due to hyperglycemia; it is tempting to suggest that membrane variations are present before the clinical metabolic alteration. Furthermore, both GDM and IDDM were on insulin therapy, therefore: (i) insulin may be the pathogenetic factor of higher intracellular free Ca 2÷ concentrations and lower Na+/K+-ATPase activity since they both varied accordingly in GDM and IDDM, but not of (ii) changes in Ca2+-ATPase, membrane fluidity and cholesterol content which did not vary accordingly in GDM and IDDM.
Introduction Previous studies showed that cells from insulin-dependent ( I D D M ) and non-insulin-dependent diabetic ( N I D D M ) patients present alterations in N a + / K +ATPase activity, Ca2+-ATPase activity, intracellular free Ca 2+ concentrations and membrane fluidity [1-6]. It has been reported that these alterations may be induced by hyperglycemia both in vivo and in vitro [7-9]. Similar plasma membrane changes, on the other hand, have been found in I D D M patients on good metabolic control [10]; therefore, they might appear independently from the presence of overt hyperglycemia. Prompt treatment of pregnant women at
Abbreviations: GDM, gestational diabetes mellitus; IDDM, insulindependent diabetes mellitus; HPW, healthy pregnant women; HNPW, healthy non-pregnant women; NIDDM, non-insulin-dependent diabetes mellitus; PRP, platelet-rich plasma. Correspondence: L. Mazzanti, Istituto di Biochimica, Facolt~ di Medicina e Chirurgia, Via Ranieri 60131, Ancona, Italy.
presentation of altered glucose tolerance lead us to study a proper model to evaluate membrane modifications induced by diabetes mellitus, but not by hyperglycemia. Furthermore, in order to evaluate the influence of insulin therapy, a group of I D D M patients was also studied.
Materials and Methods Patients
16 women who satisfied criteria for the diagnosis of gestational diabetes mellitus (GDM) [11] (28 + 5 years), 15 healthy pregnant women (HPW) (29 + 5 years), 8 w o m e n with insulin-dependent-diabetes mellitus ( I D D M ) (35 + 6 years) and 21 healthy nonpregnant women (HNPW) (32 + 4 years) were studied. G D M showed impaired glucose tolerance at the 28th or 29th week of gestation. Immediately after the diagnosis, patients were treated with insulin therapy in order to obtain a permanent very good metabolic control during the third trimester of pregnancy, with glycosylated hemoglobin ( H b A l c ) in the normal range and
66 glycemia less than 120 m g / d l in the morning, at the time of delivery. All IDDM patients received insulin treatment and were in good metabolic control ( H b A l c in the normal range and glycemia less than 120 m g / d l in the morning). All the subjects studied were normotensive and had a negative family history for diabetes mellitus, hypertension and obesity. They were taking no drug but insulin for IDDM and G D M patients. Blood samples were collected at 8.00 a.m. after an overnight fasting; particularly, in G D M and H P W patients they were obtained during the 30th week of pregnancy. H b A l c was measured by high-performance liquid chromatography according to the method of Akai [12].
Platelet isolation procedure Platelets were prepared as previously described [5]. Briefly, platelet-rich plasma (PRP) was obtained after centrifugation at room temperature of whole blood mixed with 9:1 v / v CCD (CCD:citrate 0.1 M, citric acid 7 mM, dextrose 140 raM, pH 6.5) for 20 min at 500 × g . The supernatant (i.e., platelet rich plasma, PRP) was then centrifuged at 2000 × g in anti-aggregating buffer (Tris-HCl 10 mM, E D T A 1 mM, NaCI 150 mM, glucose 5 mM, pH 7.3) and centrifuged at 2000 X g for 10 min. The pellet representing platelets was suspended in a buffer containing NaC1 145 mM, KC1 5 mM, MgSO 4 1 raM, Hepes 10 mM, glucose 10 mM, pH 7.4.
(Rmi n) and in the presence of calcium {R ...... ): Sf2 and
Sb2 = fluorescence intensities at 380 nm in the absence of (Sf2) and in the presence of calcium tSb2); and R ...... and Sf2 were measured after cell lysis with 5c~ Triton X-100 and addition of E G T A 10 mM (pH 8.3): R ...... and Sb2 were measured after cell lysis and addition of CaC1, 10 mM.
Platelet membrane preparation The preparation of platelet plasma membranes was performed according to the method of Enouf et al. [15]. Platelets were pelleted by centrifugation of the platelet-rich plasma for 15 min at 3000 × g. The pellet was washed twice in modified Tyrode's buffer (pH 7.5) containing 130 mM NaCI, 5 mM KC1, 1 mM NaH2PO 4, 24 mM NaHCO3, 2 mM Na2EDTA, 10 mM glucose, 12.5 mM sucrose and 0.35% bovine serum albumin (w/v). The cells were then lysed by ultrasonication and the lysate was centrifuged at 19000 x g to eliminate unlysed platelets, mitochondria and granules. The supernatant was centrifuged at 100 000 × g and the pellet obtained was then layered over a 40% (w/v) sucrose solution and centrifuged again at 100000 × g for 12(I min. The interface membrane subfraction obtained after the centrifugation consisted of plasma membrane, according to the biochemical and immunological characterization performed by Enouf et al. [15]. This subfraction was then resuspended in phosphate buffer (pH 7.2). Ca 2 +-ATPase
Cytosolic free calcium concentrations Ionized calcium (Ca 2+) in blood platelets was measured according to the method of Rao [13]. For the loading of the Ca2+-sensitive probe, Fura 2 AM, the platelets were incubated for 45 min with the acetoxy methyl ester of the dye at 1 /xM in a solution containing NaC1 145 mM, KC1 5 mM, MgSO 4 1 mM, Hepes 10 mM, glucose 10 mM. Cells were then washed in the same solution to remove the excess dye. The determination of intracellular Ca 2+ levels was performed in a Perkin Elmer MPF-66 fluorescence spectrophotometer at 37°C according to the method of Grynkiewicz et al. [14]. The fluorescence intensity was read at a constant emission wavelength (510 nm) with changes in the excitation wavelength (340 and 380 nm). The calibration was carried out as described by Grynkiewicz et al. [14] using the following equation: [Ca 2+ ]i = Kd
R-Rmi n Rmax _ R
Sf2 Sb2
where K d (dissociation constant of Fura 2) = 224 nM; R = ratio of the fluorescence intensities at the excitation wavelengths of 340 and 380 nm; Rmi n and Rma x = ratio of the fluorescence intensities in the absence
assay
The activity of the Ca2+-ATPase was determined according to the method of Davis et al. [16] by measuring inorganic phosphate (Pi) hydrolysed from Na2ATP 1 mM at 37°C in the presence and absence of Ca 2+ 0.15 mM. The reaction medium contained E G T A 0.1 mM, NaCI 75 mM, KC1 25 mM, MgC12 1 mM and Tris-HCl 25 mM (pH 7.4). The ATPase activity assayed in the absence of Ca 2+ was subtracted from the total ATPase activity to calculate the activity of Ca 2+ATPase. Results are expressed as p.mol Pi/mg membrane proteins per 90 min. Inorganic phosphate was measured according to Fiske and SubbaRow [17]. Protein concentration was determined by the Lowry method [18], using albumin as standard.
Na + / K +-A TPase assay N a + / K + - A T P a s e activity was determined by a modification of the Kitao method [19], as previously described [20]. The ATPase activity was assayed by incubating membranes at 37°C in 1 ml of medium (MgCI 2 5 mM, NaC1 140 mM, KC1 14 mM, in Tris-HC1 40 mM, pH 7.7). The ATPase reaction was started by the addition of 3 mM N a z A T P and stopped 20 min later by the addition of 1 ml of trichloroacetic acid 15%. Inorganic phosphate (Pi) hydrolyzed from reaction was
67
measured as previously described [17]. The ATPase activity assayed in the presence of 10 mM ouabain was subtracted from the total Mg2+-dependent ATPase activity to calculate the activity of the ouabain-sensitive Na+/K+-ATPase. Results are expressed as ~mol Pi/mg membrane protein per 60 min.
Membrane cholesterol and phospholipids Lipids were extracted from platelet membranes according to the method of Folch et al. [21] with 10 ml chloroform/methanol (2:1 v / v ) / m l membrane suspension. Cholesterol concentration was then determined by the method of Zak [22] and total phospholipids by the method of Bartlett [23].
*
C
-r- y
e
ill
F O
E =k ,D 0.150.
II
o~ w it. I--
"-m U
0
Membrane fluidity Membrane fluidity was determined by means of the measurement of the fluorescence polarization (P) of the lipophilic probe 1,6 diphenyl-l,3,5-hexatriene (DPH), according to the method of Schachter et al. [24]. The fluorescence polarization measurements were made in a Perkin-Elmer spectrofluorimeter MPF 66 equipped with two quartz prism polarizers, with exciting light at 365 nm. The P level is a quantitative index of the freedom of rotation of the fluorescent probe; a decrease in the P value indicates a higher mobility of the DPH in the deeper part of the membrane hydrophobic bilayer (i.e., increased membrane fluidity) and vice versa. Statistical analysis Statistical analysis were performed by using the Student's t-test for unpaired data.
GDM
HPW
HNPW
IDDM
Fig. 1. Ca2+-ATPase activity in platelet plasma membranes obtained from patients affected by gestational diabetes mellitus (GDM), healthy pregnant women (HPW), healthy non-pregnant women (HNPW) and insulin-dependent diabetic patients (IDDM). The effect of pregnancy was evaluated comparing HPW with HNPW; the effect of diabetes mellitus was evaluated comparing GDM and IDDM with HPW and HNPW, respectively. Results are expressed as mean + S.D. * P < 0.001.
parison with HPW (Fig. 1). Comparing IDDM with HNPW, IDDM showed higher membrane fluidity (Fig. 2), but unchanged cholesterol and phospholipid content (Table I). GDM showed lower fluidity than HPW (P < 0.001) (Fig. 2), higher cholesterol content (P < 0.001) (Table I), but unchanged phospholipid content.
4(" 4f
Results 0.300,
The effect of pregnancy on platelet properties was evaluated comparing healthy pregnant women (HPW) with healthy non-pregnant women (HNPW). Platelet membranes obtained from HPW showed higher Ca 2÷ATPase activity (P < 0.001) (Fig. 1) and lower membrane fluidity (P < 0.001) (Fig. 2) in comparison to HNPW. No difference was observed between the two groups in platelet Ca 2÷ concentration (Fig. 3), Na+/K+-ATPase activity (Fig. 4), cholesterol and phospholipid content (Table I). The effect of diabetes mellitus on platelet properties was evaluated comparing women presenting gestational diabetes mellitus (GDM) with HPW, and women affected by insulin-dependent diabetes mellitus (IDDM) with HNPW. Both GDM and IDDM showed lower Na+/K+-ATPase activity (P < 0.001) (Fig. 4) and higher intracellular free Ca 2÷ concentration (P < 0.001) than healthy pregnant and non-pregnant controls, respectively (Fig. 3). The Ca2+-ATPase activity was higher in IDDM than in HNPW (P < 0.001), but unchanged in GDM in com-
I-Ill
T -r"
Z
(2_
-I"
I--
"i"
~N 0.150. o
L
GDM
HPW
HNPW
IDDM
Fig. 2. Membrane fluidity in platelet plasma membranes obtained from patients affected by gestational diabetes mellitus (GDM), healthy pregnant women (HPW), healthy non-pregnant women (HNPW) and insulin-dependent diabetic patients (IDDM). The effect of pregnancy was evaluated comparing HPW and HNPW; the effect of diabetes mellitus was evaluated comparing GDM and IDDM with HPW and HNPW, respectively. A decrease in the parameter value indicates an increase in membrane fluidity and vice versa. Results are expressed as mean + S.D., * P < 0.001.
6~ TABLE 1 *
Membrane choh, sterol and phospholipid content
200-
.
-r
0
E
"I"
I--
z "' I-Z
100. "r"
Platelets were obtained from healthy pregnant women (HPW), patients with gestational diabetes mellitus (GDM), healthy non-pregnant women (HNPW) and insulin-dependent diabetes patients (IDDM). Results are expressed as mean _+S.D. The difference in the m e m b r a n e cholesterol content between t-IPW and G D M patients was statistically significant ( * P < (1.001).
-Ir-
o
HPW GDM HNPW IDDM
U U
Cholesterol ( n m o l / l )
Phospholipids (nmol/1)
102.(I _+4.6 137.0_+5.1 * 115.2+_8.4 125.0+4.7
171.4 _+3.2 172.3_+3.t 173.1 _+5.8 192.3+7.(/
0
GDM
HPW
HNPW
IDDM
Fig. 3. Ca 2+ concentration in platelets obtained from patients affected by gestational diabetes mellitus (GDM), healthy pregnant women (HPW), healthy non-pregnant women ( H N P W ) and insulindependent diabetic patients (IDDM). The effect of pregnancy was evaluated comparing H P W with H N P W ; the effect of diabetes meUitus was evaluated comparing G D M and I D D M with H P W and H N P W , respectively. Results are expressed as mean _+S.D. * P < 0.001.
Discussion
Comparing the membrane platelets from H P W with HNPW, H P W showed increased Ca2+-ATPase, but similar N a + / K + - A T P a s e and similar intracellular free Ca 2÷ concentrations. The lack of change in intracellular free Ca 2+ concentrations led us to suggest that
1.5
2_± n
T
o.
_L
w o
E =k
g g
0.5
t,
0
GDM
HPW
HNI~/
IDDM
Fig. 4. N a + / K + - A T P a s e activity in platelet plasma m e m b r a n e s obtained from patients affected by gestational diabetes mellitus (GDM), healthy pregnant w o m e n (HPW), healthy non-pregnant women (HNPW) and insulin-dependent diabetic patients (IDDM). The effect of pregnancy was evaluated comparing H P W with H N P W ; the effect of diabetes mellitus was evaluated comparing G D M and I D D M with H P W and H N P W , respectively. Results are expressed as mean _+S.D. * P < 0.001.
increased influx of Ca 2+ is balanced by the increased Ca2+-ATPase. Platelets from pregnant women showed lower membrane fluidity without modifications in cholesterol a n d / o r phospholipid content. The reason for changes in membrane fluidity is unknown. These data point out that pregnancy may cause variations in plasma membrane structure and function. G D M presented an inverse situation in which intracellular free Ca z+ concentration was increased, but CaZ+-ATPase was unchanged. Both G D M and IDDM showed decreased N a + / K + - A T P a s e thus confirming previous results obtained on platelets and erythrocytes [5,9,10,20]. Previously, in non-pregnant diabetic patients, membrane fluidity was reported decreased [2527], increased [28,10,5] or not changed with respect to healthy subjects [29]. In the present study membrane fluidity was found decreased in G D M compared to HPW, but increased in IDDM compared to HNPW. The decreased N a + / K + - A T P a s e observed in both G D M and IDDM cannot derive from changed membrane fluidity but, probably, from a more specific effect on the enzyme induced by diabetes mellitus. The present data point out that platelet plasma membranes show various alterations in G D M despite the absence of an overt period of hyperglycemia or altered metabolic control; in fact, G D M patients were euglycemic in basal conditions and had a really short period of impaired glucose tolerance which in turn did not produce any alteration in both the length of pregnancy a n d / o r the metabolic control during the third trimester (see Materials and Methods). This would suggest that membrane alterations may be present before the onset of impaired glucose tolerance or may be induced by insulin treatment or other unclear factors. In this respect, it is noteworthy that both IDDM and G D M were treated with insulin, but showed different variations in some membrane parameters (CaZ+-ATPase, membrane fluidity, cholesterol content); therefore, insulin cannot have a key-role in inducing those alterations. On the contrary, a relationship between the intracellular free Ca 2+ concentrations and the
69 Na+/K+-ATPase may be possible. Finally, both diabetic groups were on good metabolic control so that alterations described cannot be produced by hyperglycemia. In conclusion, these data suggest that diabetes and pregnancy may affect platelet plasma membrane function. The differences between GDM and IDDM would support the thesis that those are two different disease states [11].
Acknowledgements The present work was supported by a Grant of Regione Marche (LM and NC); C.N.R. No. 91.00345. PF 40 (RDP) and progetto Ricerca Scientifica 40% 1991 (RDP).
References 1 Finotti, P. and Palatini, P. (1987) Diabetologia 36, 991-995. 2 Rahmani-Jourdheuil, D., Mourayre, Y., Vague, P., Boyer, J. and Juhan-Vague, I. (1987) Diabetes 36, 991-995. 3 Cohen, M.P. and Klepser, H. (1988) Diabetes 37, 558-562. 4 Testa, I., Rabini, R.A., Fumelli, P., Bertoli, E. and Mazzanti, L. (1988) J. Clin. Endoerinol. Metab. 67, 1129-1133. 5 Mazzanti, L., Rabini, R.A., Faloia, E., Fumelli, P., Bertoli, E. and De Pirro, R. (1990) Diabetes 39, 850-854. 6 Bergh, C.H., Hjialmarson, A., Holm, G., Anguald, E. and Jacobsson, B. (1988) Eur. J. Clin. Invest. 18, 92-97. 7 Watala, C. (1988) Exp. Pathol. 33, 233-238. 8 Jain, S.K., McVie, R., Duett, J. and Herbst, J.J. (1989) Diabetes 38, 1539-1543.
9 Jain, S.K. (1989) J. Biol. Chem. 264, 21340-21345. 10 Mazzanti, L., Rabini, R.A., Testa, I. and Bertoli, E. (1989) Eur. J. Clin. Invest. 19, 84-89. 11 National Diabetes Data Group (1979) Diabetes 28, 1039-1057. 12 Akai, T. (1982) Glycosylated hemoglobin vensonic seminar proceeding 1. Tokyo, Daichii Kagaku, p. 17. 13 Rao, G.H.R. (1988) Anal. Bioehem. 169, 400-404. 14 Grynkiewicz, G., Poenie, M. and Tsien, R.Y. (1985) J. Biol. Chem. 6, 3440-5034. 15 Enouf, J., Bredoux, R., Bourdeau, N., Sarkadi, B. and LevyToledano, S. (1989) Bioehem. J. 263, 547-552. 16 Davis, F.B., Davis, P.J., Nat, G., Bias, S.D., Mac Gillivray, M., Gutman, S. and Feldman, M.J. (1985) Diabetes 34, 639-646. 17 Fiske, C., SubbaRow, Y. (1925) J. Biol. Chem. 193, 375-400. 18 Lowry, O.H., Rosenburg, M.Y., Farr, A.L. and Randall, R.T. (1951) J. Biol. Chem. 193, 265-275. 19 Kitao, T. and Hattori, K. (1983) Experientia 39, 1362-1364. 20 Testa, I., Rabini, R.A., Danieli, G., Tranquilli, A.L., Cester, N., Romanini, C., Bertoli, E. and Mazzanti, L. (1988) Scand. J. Clin. Lab. Invest. 48, 7-14. 21 Folch, J., Less, M. and Sloane-Stanley, G.H. (1957) J. Biol. Chem. 226, 466-468. 22 Zak, B. (1957) Am. J. Clin. Pathol. 27, 583-588. 23 Bartlett, G.R. (1959) J. Biol. Chem. 234, 466-468. 24 Schachter, D. and Shinitzky, M. (1977) J. Clin. Invest. 59, 536-548. 25 Bryszewska, M., Watala, C. and Torzzcka, W. (1986) Br. J. Haematol. 62, 111-116. 26 Baba, Y., Kai, M., Kamada, T. Setoyama, S. and Otsuji, S. (1979) Diabetes 28, 1138-1140. 27 Winocour, P.D., Bryszewska, M., Watala, C., Rand, M.L., Epand, R.M., Kinlough-Rathbone, R.U, Packham, M.A. and Mustard, J.F. (1990) Diabetes 39, 241-244. 28 Freyburger, G., Gin, H., Dousseau, F., Lorient-Roudault, M.F. and Boisseau, M.R. (1988) Clin. Hemorheol. 8, 159-164. 29 Baldini, P., lncerpi, S., Lambert-Gardini, S., Spinedi, A. and Luly, P. (1989) Diabetes 38, 825-831.
