Atherosclerosis, 83 (1990) 231-237 Elsevier Scientific Publishers Ireland,
ATHERO
231 Ltd
04509
Rheological properties and membrane fluidity of red blood cells and platelets in primary hyperlipoproteinemia S. Muller I, 0. Ziegler ’INSERM
*, M. Donner
I, P. Drouin
* and J.F. Stoltz ’
U. 284, CO IO, Plateau de Brabois, 54511 Vandoeuvre-l&Nancy Cedex (France), and ’ Service de M&de&e Nancy, H6pital Jeanne d’Arc, Dommartin-l&Toul (France)
G, CHUde
(Received 20 October, 1989) (Revised, received 19 April, 1990) (Accepted 20 April, 1990)
Summary Lipid fluidity of the erythrocyte membrane and intact platelets was examined in 32 male patients affected by types IIA, IIB and IV primary hyperlipoproteinemia and 15 control subjects. Lipid fluidity was determined by fluorescence polarization using two probes: DPH and TMA-DPH which are localized in different lipid areas of the cell membrane. Classical haemorheological tests were also performed including plasma viscosity, whole blood viscosity and erythrocyte aggregation. As compared to a control group, plasma viscosity and whole blood viscosity at low shear rate was significantly increased in types IIB and IV, but not in type IIA patients. In contrast, the increase in erythrocyte aggregation was significant in all HLP types. Concerning lipid fluidity, the results recorded with red cells and platelets were not significantly different for type IIA HLP compared to the control group. In contrast, erythrocyte membranes from patients with types IIB and IV HLP had a significantly higher level of fluidity in lipid regions characterized by TMA-DPH. Using DPH as a fluorescent probe, identical results were only noted in type IIB patients. Regarding intact platelets of IIB and IV patients, an increase in lipid fluidity was noted for two fluorescent probes. These findings suggest that HLP associated erythrocyte and platelet fluidity alterations are not related to hypercholesterolemia but to the triglyceride level.
Key words:
Hyperlipoproteinemia; fluidity; Erythrocyte
Plasma membrane;
viscosity; Platelet;
Introduction Patients with hyperlipoproteinemia risk for atherosclerosis and occlusive
are at high vascular dis-
Correspondence to: Dr. S. Muller, INSERM, U. 284, CO 10, Plateau de Brabois, F-54511 Vandoeuvre-l&-Nancy, France. 0021-9150/90/$03.50
0 1990 Elsevier Scientific
Publishers
Ireland,
Blood viscosity; Erythrocyte Fluorescence polarization
aggregation;
Lipid
eases [l]. Platelets are assumed to play an important part in the development of atherosclerosis and its complications [2-51. An altered platelet function has been reported in hyperlipidemic subjects together with modifications in physical properties of platelet membranes [6,7]. In particular, in an experimental model, hyperlipidemia was related to an increase in platelet membrane lipid Ltd.
232 fluidity [8], a biophysical parameter which controls various cellular functions [9-111. Reports are also available pertaining to the role of red blood cells in the cardiovascular risk of hyperlipidemic patients. Increased plasma lipid levels have been related to alterations in macrorheological parameters: plasma viscosity and blood viscosity [12-161. In contrast, the relations of hyperlipidemia with membrane dependent rheological parameters like lipid fluidity are not well established. Besides in vitro studies on the effect of cholesterol and plasma lipoproteins on platelet fluidity [17-191, only scarce data refer to type IIA-HLP and platelets or red cell membrane fluidity [20]. Therefore, the aim of the present study was to investigate the erythrocyte membrane and platelet fluidity of patients with primary hyperlipoproteinemia. The results were compared to macro and microrheological data. Red blood cell aggregation and membrane fluidity have been investigated in patients with primary hyperlipoproteinemia (phenotypes IIA, IIB and IV). Erythrocyte aggregation was assessed using a photometric method based on the analysis of light transmission by blood suspension in the course of the aggregation process. A parallel study included determinations of classical rheological parameters, plasma and blood viscosity. The lipid fluidity of red blood cell membranes and intact platelets has been determined using the fluorescence polarization technique and two fluorescent probes, 1,6-diphenyl-1,3,5hexatriene (DPH) and its cationic by-product l-[Ctrimethylammoniumphenyll-6-phenyl-1,3,5-hexatriene (TMA-DPH) which are embedded in the deep and surface regions of the membrane, respectively [21]. Materials and methods Patient selection The investigation was carried out in 32 Caucasian male patients (mean age 43.9 f 10.8 years, range 19-61) affected by primary hyperlipoproteinemia (HLP). Secondary hyperlipoproteinemia due to hypothyroidism, diabetes, nephrotic syndrome, biliary obstruction, pancreatitis or dysglobulinemia had been excluded before. HLP was classified according to WHO criteria [22]. We
studied 12 patients with type IIA (familial hypercholesterolemia), 13 patients with type IIB (combined hyperlipidemia) and 7 patients with type IV (endogenous hypertriglyceridemia). These patients recruited in the lipid clinic had not taken any medication known to modify rheological parameters in the 3 months prior to the study. Fifteen normal Caucasian male healthy volunteers (mean age 33.1 + 5.1, range 24-45) with a normal blood lipid profile and a normal fasting blood glucose were selected from the medical staff of the hospital. All had neither symptoms nor a family history of hyperlipidemia or diabetes mellitus. Both patients and controls were non smokers. The study was approved by the “Centre Hospitalier et Universitaire de Nancy” Ethical Committee. Blood lipid determinations Blood samples were obtained by venipuncture after a 12 h overnight fast. Total serum cholesterol and triglycerides were enzymatically determined in a Technicon Autoanalyser. Total apo AI and apo B were determined by laser immunonephelometry using specific antibodies from Behring (Behring, Rueil-Malmaison, France). High density lipoprotein (HDL) cholesterol was measured in serum after precipitation of VLDL and LDL with phosphotungstic acid [23]. Rheological parameters Plasma viscosity was determined at 37 o C using an automatic capillary viscometer (AMTEC, Villeneuve-Loubet, France). Blood viscosity at a corrected hematocrit (Ht 40%) was determined at 37 o C for 7 shear rates (y = 0.50 to 128 s-l) using a coaxial cylinder rotational viscometer (LS 30, Contraves, Zurich, Switzerland). Erythrocyte aggregation at a corrected hematocrit (Ht = 30%) was determined at 37” C with a Myrenne aggregometer (Myrenne, GmbH, Aachen, F.R.G.) based on the analysis of transmitted light by a blood suspension in a cone-plan viscometer [24]. Platelet preparation Platelets were isolated from venous blood by a metrizamide gradient centrifugation as previously described [25]. After isolation, platelets were resuspended in a phosphate buffer at a concentra-
233 tion of 3 x 10’ cells/ml rescent labelling.
and processed
for fluo-
Red cell membrane preparations The Heidrich and Leutner method [26] was used for preparing the red cell ghosts. Following lysis in a 5 mM phosphate buffer, pH 8, the suspensions were washed in the same buffer until every trace of hemoglobin was removed. The membranes obtained are referred to as “unsealed” ghosts. Heven and Solomon’s method [27] was applied for preparing “resealed” ghosts. The protein concentration in the membrane suspensions was then assessed using the Lowry test and was systematically adjusted to a concentration of 100 pg of protein per ml. The fluorescence assessments were carried out within 5 h after blood sampling. Determination of erythrocyte membrane and platelet lipid fluidity The fluorescent probes DPH (Aldrich-Chimie, Strasbourg, France) and TMA-DPH (Molecular probes, Eugene, OR, U.S.A.) were dissolved in tetrahydrofuran (THF) and N, N-dimethyl formamide (DMF), respectively, at a concentration of 2 x lop3 M. The solutions were stored in the darkness at 4°C. Labelling was carried out by adding 3 ~1 of the above solutions to 3 ml of both membrane and the cell suspension. Measurements were carried out at 25 o C. Fluorescence polarization was assessed by means of a continuous excitation instrument (Fluofluidimeter, Sefam, Vandoeuvre-l&s-Nancy, France). In all cases, non-labelled membrane sus-
TABLE
pensions were measured before each series of experiments in order to calculate the correcting factors. After determining the signals due to “blanks” (non-labelled samples), Ib,,, Ib, , the fluorescence intensities of the samples were assessed, If,,, If i Fluorescence steady-state anisotropy (r ) and fluorescence intensity were assessed, based on the following relationships: (r)
= (If,, - Ib,,) - (Ifl
-Ib,)
I, with Jr = (If,, - Ib,,) + 2(If,
-Ib,)
Statistical analysis of the results The results were expressed as mean + standard deviation. They were analysed using the non-parametric Mann-Whitney test. Statistical significance was assumed at P -C 0.05. Any relation between the different biochemical, rheological and molecular parameters of this study was tested by means of Spearman’s multiple correlation test. Results The plasma lipids and lipoprotein profiles of the studied subjects are reported in Table 1. As expected, mean plasma total cholesterol and triglyceride levels were higher in patients compared with the normal subjects. HDL-cholesterol levels were lower in types IIB and IV hyperlipoproteinemias.
1
PLASMA
Phenotype
LIPID
VALUES
Cholesterol
Triglycerides
Apo Al
Apo B
HDL-
(mg/dl)
@g/d11
@g/d11
@g/dl)
cholesterol
182+19
**
55+16
177+17
**
47ill
*
49+42
(w/d0 IIA
335i44
**
101+39
*
166 * 34
IIB
293+42
**
207*62
**
175*46
IV
257i42
*
425f20
**
123+42
145k34
Controls
195&27
137+30
91+27
* P < 0.05. * *P c 0.001 versus controls.
71+22
*
58+
8
234 TABLE 2 BLOOD VISCOSITY IN NORMAL
SUBJECTS AND HYPERLIPOPROTEINEMIC
Group
Blood viscosity (mPa/s)
Shear rate:
0.5 s-1
0.9 s-1
HA IIB IV Controls
18.6*3.0 21.353.7 20.6+2.0 18.9+1.9
153~2.7 15.9k2.2 * 15.8*2.5 * 13.8 + 1.2
** *
3.2 s-’ 9.2 f 1.0 9.9kl.l * 10.3 + 1.2 * 8.6 f 0.5
PATIENTS
8.1 s-t
20.4 s-l
128 s-1
6.9 f 0.5 7.3 f 0.7 7.5+0.7 * 6.7kO.3
5510.3 5.9 & 0.4 5.8kO.3 5.4kO.2
3.8iO.l 3.9kO.2 3.9 * 0.3 3.8kO.l
* P < 0.05, * * P < 0.01versus control.
TABLE 3 PLASMA VISCOSITY AND ERYTHROCYTE AGGREGATION IN NORMAL SUBJECTS AND HYPERLIPOPROTEINEMIC PATIENTS Group
Plasma viscosity (cSt)
Erythrocyte aggregation index
IIA IIB IV
1.24k 0.05 1.29kO.08 ** 1.27kO.06 **
17.2k2.0 18.0+4.0 17.8*2.5
Controls
1.23 + 0.04
15.3 * 3.1
** ** **
* * P i 0.01versus controls.
The values of rheological parameters are summarized in Tables 2 and 3. Significant differences in blood viscosity at low shear rates and plasma viscosity were only noted between control subjects paand types IIB and IV hyperlipoproteinemic erythrocyte aggregation was tients. In contrast, significantly higher in the 3 groups of patients
with hyperlipoproteinemia compared with the normal subjects. The fluorescence anisotropy of TMA-DPH and DPH embedded in erythrocyte ghosts and intact platelets was determined in both controls and patients. The results recorded on erythrocyte membranes are summarized in Table 4. Concerning both unsealed and resealed membranes and the two fluorescent probes TMA-DPH and DPH, no significant differences were observed between type IIA patients and the control group. In contrast, in types IIB and IV patients, fluorescence anisotropy values of TMA-DPH were significantly lower in unsealed and resealed membranes. For DPH, a significant decrease in fluorescence anisotropy values was only observed in unsealed and resealed erythrocyte membranes of type IIB patients. The results on platelets are summarized in Table 5. As for erythrocyte membranes, no differences between the type IIA group and the control group were observed with respect to TMA-
TABLE 4 FLUORESCENCE ANISOTROPY VALUES OF TMA-DPH AND DPH EMBEDDED NORMAL SUBJECTS AND HYPERLIPIDEMIC PATIENTS
IN ERYTHROCYTE
MEMBRANES
TMA-DPH
Group
DPH Unsealed membranes
Resealed membranes
Unsealed membranes
Resealed membranes
HA IIB IV Controls
0.238 + 0.002 0.236 + 0.002 * 0.238 +0.003 0.239 +O.OOl
0.238 0.236 0.239 0.242
0.258 0.255 0.254 0.260
0.266 0.266 0.261 0.270
* P < 0.05vs controls.
f f f f
0.002 0.001 * 0.002 0.001
+ + + +
0.001 0.001 * 0.002 * 0.001
f + f +
0.002 0.001 * 0.002 * 0.001
OF
235 TABLE
5
FLUORESCENCE ANISOTROPY VALUES OF TMA-DPH AND DPH EMBEDDED IN PLATELETS OF NORMAL SUBJECTS AND HYPERLIPIDEMIC PATIENTS
Group
DPH
TMA-DPH
ITA IIB
0.224 * 0.002 0.208 * 0.004 * * 0.1@3*0.005 * * 0.223 + 0.003
0.287 + 0.002 0.285 & 0.002 * 0.285 +0.002 * 0.291+ 0.002
IV Controls
* P < 0.05, * * P i 0.01vs.controls.
DPH or DPH. In types IIB and IV groups, a significant decrease in anisotropy values was noted for the two fluorescent probes. When patients and control data are grouped together, a negative relationship was observed between the fluorescence anisotropy values of TMADPH and DPH embedded in platelets and plasma triglyceride amounts (r = -0.31, P < 0.01, and Y = - 0.74, P -c 0.001, respectively). A weak negative relationship was observed between the fluorescence anisotropy of TMA-DPH incorporated in unsealed erythrocyte membranes and triglyceride levels (r = -0.33 P < 0.05). There was no significant relationship between classical rheological parameters and molecular parameters except for the fluorescence anisotropy of DPH embedded in unsealed and resealed ghosts which were negatively related to plasma viscosity (r = - 0.32 P 80 years). Ditzel and Kampmann [28] and Voisin et al. [29] did not find any difference in blood viscosity and in erythrocyte aggregation before 60-65 years. Moreover, the determination of normal values in our laboratory showed no significant difference between two groups with ages corresponding to the controls and patients of the present study. Given the fact that erythrocyte aggregation is considered to be involved in the increase in blood viscosity at low shear rates [30], it may seem paradoxical that the blood viscosity of IIA patients at low shear rate is not significantly different from that of healthy controls. However, one must keep in mind that erythrocyte aggregation is a complex phenomenon which depends on both the level of plasma proteins (immunoglobulins, fibrinogen, albumin, etc.) and the intrinsic properties of red cells. Erythrocyte aggregation can be approached in terms of kinetic, structural and rheological parameters and it is likely that blood viscosity at low shear rates only reflects one component of erythrocyte aggregation [31]. Moreover, from a technical point of view, the measurement of erythrocyte aggregation may be a more sensitive approach than blood viscosity determinations. The hypothesis of an alteration of intrinsic properties of erythrocytes is supported by the data on the fluorescence anisotropy values of fluorescent probes: DPH and TMA-DPH embedded in red cell membranes. It is well admitted that the fluorescence anisotropy of such probes is inversely related to the lipid fluidity of the probe immediate surroundings [ll]. Our results provide evidence that type IIB and IV-HLP are significantly associated with an increase in the fluidity near the polar heads of the phospholipids where the fluorescent probe TMA-DPH is located. No significant difference is observed for type IIA-HLP. By contrast,
236 a significant increase in the fluidity of the central hydrophobic core characterized by DPH as a fluorescent probe is only recorded in type IIB-HLP. A significant increase in the fluidity of intact platelets is also noted in the superficial and deep areas of cell lipid regions from types IIB and IV patients. There are no significant differences between type IIA-HLP and normal subjects in any of the molecular parameters measured with TMADPH and DPH. The fact that the mean ages of type IIA and control groups are lower than that of types IIB and IV groups cannot explain the results. The process of tissue and cell aging is essentially associated to a progressive decrease in lipid fluidity [32-341. The erythrocyte findings are probably related to previous results showing a decrease in cholesterol/phospholipid (C/PL) ratio of erythrocyte membrane in IV-HLP patients [35,36]. They may be relevant to the high amounts of plasma triglycerides, Indeed, it has been shown that triglyceride-emulsion particles could promote the efflux or block the influx of cholesterol in cultured macrophages [37]. Whether a similar mechanism exists for erythrocytes remains to be elucidated. Nevertheless, 80% of erythrocyte membrane cholesterol can be removed by lipoproteins of reduced cholesterol content [38]. Since added amounts of cholesterol decrease the fluidity of membranes, it may be suggested that the presence of high amounts of plasma triglycerides could enhance cholesterol efflux from the cell membrane resulting in an increase in the lipid fluidity of erythrocytes. Where platelets are concerned, the results are not fully understood. Despite the fact that an inverse relationship was shown in vitro between C/PL ratio and fluidity [17], a free cholesterol/phospholipid ratio was normal in type IV-HLP patients [39]. Therefore, it is not possible to assume for platelets that the higher fluidity in IIB and IV HLP patients is due to a reduced C/PL ratio. Our observations suggest that hypercholesterolemia does not induce marked changes in the lipid fluidity of erythrocytes and platelets. In another work, no change in red cell C/PL ratio was also observed in IIA-HLP patients [20]. In platelets an increase in C/PL ratio of plasma membrane
(+7%) was previously reported [40]. It may be suggested that this weak increase in C/PL ratio only induces discrete modifications in membrane fluidity. The alterations are perhaps not detected by the fluorescent probes TMA-DPH and DPH which label both membrane and cytoplasmic lipid regions. By contrast, it appears that this molecular rheological parameter could be affected by the concentration of plasma triglycerides. Further studies are needed to determine the exact relations between HLP associated lipid fluidity alterations, plasma triglycerides and classical haemorheological patterns. Acknowledgement We are grateful for the expert technical assistance of Mrs. S. Droesch, G. Cauchois and M. Gentils. References of atherosclerosis. An update. 1 Ross, R., The pathogenesis New Engl. J. Med., 314 (1986) 488. B., Poindexter, B.J. and Schaeffer, 2 Corash, L., Andersen, E.J., Platelet functions and survival in patients with severe hypercholesterolemia. Arteriosclerosis, 1 (1981) 443. a problem of the biology of 3 Ross, R., Atherosclerosis: arterial wall cells and their interaction with blood components. Arteriosclerosis, 1 (1981) 293. 4 Mustard, J.F. and Packham, M.A., The role of blood and platelets in atherosclerosis and the complications of atherosclerosis. Thromb. Diath. Haemorh., 33 (1975) 444. in atherosclerosis and 5 Schwartz, S.M., Cellular proliferation hypertension. Proc. Sot. Exp. Biol. Med., 173 (1983) 1. 6 Tremoli, E., Madema, P., Colli, S., Morazzoni, G., Sirtori, M. and Sirtori, C.R., Increased platelet sensitivity and thromboxane B, formation in type-11 hyperlipoproteinaemic patients. Eur. J. Clin. Invest., 14 (1984) 329. I Carvalho, A., Carvalho, M., Colman, R.W. and Lees, R.S., Platelet function in hyperlipoproteinemia. N. Engl. J. Med., 290 (1974) 434. M., Platelet mem8 Berlin, E., Shapiro, S.G. and Friedland, brane fluidity and aggregation of rabbit platelets. Atherosclerosis, 51 (1984) 223. fluidity 9 Shiga, T. and Maeda, N., Influence of membrane on erythrocyte functions. Biorheology, 17 (1980) 485. polarization applied 10 Stoltz, J.F., Donner, M., Fluorescence to cellular microrheology. Biorheology, 22 (1985) 227. 11 Stoltz, J.F. and Donner, M., Relations between molecular rheology and red blood cell structure: methods and clinical approaches. Clinical Haemorheology, 5 (1985) 813. H., Arntz, H.R. and Klemens, U., Studies of 12 Leonhardt, plasma viscosity in primary hyperlipoproteinemia. Atherosclerosis, 28 (1977) 29.
237 H. and Amtz, H.R., Blutviskositat und Etyth13 Leonhardt, rozytenflexibilitat bei prim&en Hyperlipoproteinlmien. Klin. Wschr., 56 (1978) 271. 14 Pfeiffer. M. and Tilsner, V.. Der Einfluss von Clofibrat auf die Plasmaviskositat bei Hyperlipoproteinlmien. Med. Klin., 73 (1978) 60. S., Rousselle, D., Voisin, P. and 15 Stoltz. J.F., Gaillard, Drouin, P., Rheological study of blood during primary hyperlipoproteinemia. Clin. Hemorheol., 3 (1981) 227. 16 Seplowitz, A.H.. Chien, S. and Smith, F.R., Effects of lipoproteins on plasma viscosity. Atherosclerosis, 38 (1981) 89. microviscosity 17 Shattil, S.J. and Cooper, R.A., Membrane and human platelet function. Biochemistry, 15 (1976) 4832. R., Marenah, 18 Hassall. D.G., Forrest, L.A., Bruckdorfer, C.B., Turner, P.. Cortese, C., Miller, N.E. and Lewis, B., Influence of plasma lipoproteins on platelet aggregation in a normal male population. Arteriosclerosis, 3 (1983) 332. 19 Malle. E., Cries, A., Kostner, G.M., Pfeiffer, K., Nimpf, J. and Hermetter. A.. Is there any correlation between platelet aggregation, plasma lipoproteins, apoproteins and membrane fluidity of human blood platelets? Thromb. Res., 53 (1989) 181. of cell-membrane fluidity in 20 Cooper, R.A.. Abnormalities the pathogenesis of disease. New Engl. J. Med., 297 (1977) 371. 21 Donner. M. and Stoltz, J.F., Comparative study on fluorescent probes distributed in human erythrocytes and platelets. Biorheology. 22 (1985) 385. 22 Beaumont. J.L., Carlson, I.A., Cooper, G.R., Fejfar, Z., Fredrickson, D.S. and Strasser, T., Classification of hyperlipidemia and hyperlipoproteinemia. Bull. WHO, 43 (1970) 891. M., Scholcnick, H.R. and Morfin, R.. Rapid 23 Burstein. method for isolation of lipoproteins from serum by precipitation with polyanions. J. Lipid Res., 11 (1970) 583. 24 Schmid-Schiinbein. H., Volger, E., Teitel, P.. Kiesewetter, H.. Daver, V. and Heilmann, L., New hemorheological techniques for the routine laboratory. Clin. Hemorheol., 2 (1982) 93. S.. Bredoux, R., Rendu, F. Jeanneau, C., 25 Levy-Toledano, Savanau, E. and Dassin, E., Isolement et fonction des plaquettes. Nouv. Rev. Fr. Hemat., 16 (1976) 367. 26 Heidrich. H.G. and Leutner, G., Two types of vesicles from the erythrocyte ghost membrane differing in surface charge. Eur. J. Biochem., 41 (1974) 37.
27 Heven, S. and Solomon, A.K., Permeability of human erythrocyte membrane vesicles to alkali cations. Biochim. Biophys. Acta, 550 (1979) 393. 28 Ditzel, J. and Kampmann. J., Whole-blood viscosity, hematocrit and plasma protein in normal subjects at different ages. Acta Physiol. Stand., 81 (1971) 264. 29 Voisin. P.. Chevillard, F.. Penin, F. and Stoltz. J.F.. Influence of patients age in the determination of hemorheological parameters. In: G. Potron, J.F. Stoltz, M. Gueguen, M. Boisseau (Eds.), Standardization in Clinical Haemorheology, Laboratoires Boots-Dacourt, 1987, 9. R.J. and Gregersen. M.I., 30 Chien, S., Usami, S., Dellenback, Shear dependent interactions of plasma protein with erythrocytes in blood. Am. J. Physiol., 219 (1970) 143. 31 Stoltz. J.F., Paulus, F. and Donner, M.. Experimental approaches to erythrocyte aggregation. Clin. Hemorheol., 7 (1987) 33. fluidity and cellular functions. 32 Shinitzky, M., Membrane In: Meir Shinitzky (Ed.), Physiology of Membrane Fluidity, Vol. 1. CRC Press, Boca Raton, FL, 1984. p. 1. 33 Rivray. B., Bergman, S., Shinitzky, M. and Globerson, A.. Correlations between membrane viscosity serum cholesterol. lymphocyte activation and aging in man. Mech. Ageing Dev.. 12 (1980) 119. 34 Terranova, R.. Alberghina, M. and Camazzo. G.. Composizione lipidica e fluidita della membrana eritrocitaria nell’anziano. Ric. Clin. Lab., 15 (1985) 327. 35 Vakakis. N., Redgrave, T.G., Small, D.M. and Castelli, W.P., Cholesterol content of red blood cells and low-density lipoproteins in hypertriglyceridemia. Biochim. Biophys. Acta, 751 (1983) 280. 36 Neerhout. R.C., Erythrocyte stromal lipids in hyperlipemic states. J. Lab. Clin. Med., 71 (1968) 448. 37 Aviram, M., Williams, K.J. and Deckelbaum, R.J., Macrophage cholesterol removal by triglyceride phospholipid emulsions, Biochem. Biophys. Res. Commun.. 155 (1988) 709. M.H., Rates of cholesterol exchange between 38 Gottlieb, human erythrocytes and plasma lipoproteins. Biochem. Biophys. Acta, 600 (1980) 530. 39 Shastri, K.M.. Carvalho, A.C.A. and Lees. R.S., Platelet function and platelet lipid composition in the dislipoproteinemia. J. Lipid Res., 21 (1980) 467. 40 Shattil. S.J., Bennett, J.S., Colman, R.W. and Cooper. R.A.. Abnormalities of cholesterol-phospholipid composition in platelets and low-density lipoproteins of human hyperbetalipoproteinemia. J. Lab. Clin. Med., 89 (1977) 341.
ATHERO
231 Ltd
04509
Rheological properties and membrane fluidity of red blood cells and platelets in primary hyperlipoproteinemia S. Muller I, 0. Ziegler ’INSERM
*, M. Donner
I, P. Drouin
* and J.F. Stoltz ’
U. 284, CO IO, Plateau de Brabois, 54511 Vandoeuvre-l&Nancy Cedex (France), and ’ Service de M&de&e Nancy, H6pital Jeanne d’Arc, Dommartin-l&Toul (France)
G, CHUde
(Received 20 October, 1989) (Revised, received 19 April, 1990) (Accepted 20 April, 1990)
Summary Lipid fluidity of the erythrocyte membrane and intact platelets was examined in 32 male patients affected by types IIA, IIB and IV primary hyperlipoproteinemia and 15 control subjects. Lipid fluidity was determined by fluorescence polarization using two probes: DPH and TMA-DPH which are localized in different lipid areas of the cell membrane. Classical haemorheological tests were also performed including plasma viscosity, whole blood viscosity and erythrocyte aggregation. As compared to a control group, plasma viscosity and whole blood viscosity at low shear rate was significantly increased in types IIB and IV, but not in type IIA patients. In contrast, the increase in erythrocyte aggregation was significant in all HLP types. Concerning lipid fluidity, the results recorded with red cells and platelets were not significantly different for type IIA HLP compared to the control group. In contrast, erythrocyte membranes from patients with types IIB and IV HLP had a significantly higher level of fluidity in lipid regions characterized by TMA-DPH. Using DPH as a fluorescent probe, identical results were only noted in type IIB patients. Regarding intact platelets of IIB and IV patients, an increase in lipid fluidity was noted for two fluorescent probes. These findings suggest that HLP associated erythrocyte and platelet fluidity alterations are not related to hypercholesterolemia but to the triglyceride level.
Key words:
Hyperlipoproteinemia; fluidity; Erythrocyte
Plasma membrane;
viscosity; Platelet;
Introduction Patients with hyperlipoproteinemia risk for atherosclerosis and occlusive
are at high vascular dis-
Correspondence to: Dr. S. Muller, INSERM, U. 284, CO 10, Plateau de Brabois, F-54511 Vandoeuvre-l&-Nancy, France. 0021-9150/90/$03.50
0 1990 Elsevier Scientific
Publishers
Ireland,
Blood viscosity; Erythrocyte Fluorescence polarization
aggregation;
Lipid
eases [l]. Platelets are assumed to play an important part in the development of atherosclerosis and its complications [2-51. An altered platelet function has been reported in hyperlipidemic subjects together with modifications in physical properties of platelet membranes [6,7]. In particular, in an experimental model, hyperlipidemia was related to an increase in platelet membrane lipid Ltd.
232 fluidity [8], a biophysical parameter which controls various cellular functions [9-111. Reports are also available pertaining to the role of red blood cells in the cardiovascular risk of hyperlipidemic patients. Increased plasma lipid levels have been related to alterations in macrorheological parameters: plasma viscosity and blood viscosity [12-161. In contrast, the relations of hyperlipidemia with membrane dependent rheological parameters like lipid fluidity are not well established. Besides in vitro studies on the effect of cholesterol and plasma lipoproteins on platelet fluidity [17-191, only scarce data refer to type IIA-HLP and platelets or red cell membrane fluidity [20]. Therefore, the aim of the present study was to investigate the erythrocyte membrane and platelet fluidity of patients with primary hyperlipoproteinemia. The results were compared to macro and microrheological data. Red blood cell aggregation and membrane fluidity have been investigated in patients with primary hyperlipoproteinemia (phenotypes IIA, IIB and IV). Erythrocyte aggregation was assessed using a photometric method based on the analysis of light transmission by blood suspension in the course of the aggregation process. A parallel study included determinations of classical rheological parameters, plasma and blood viscosity. The lipid fluidity of red blood cell membranes and intact platelets has been determined using the fluorescence polarization technique and two fluorescent probes, 1,6-diphenyl-1,3,5hexatriene (DPH) and its cationic by-product l-[Ctrimethylammoniumphenyll-6-phenyl-1,3,5-hexatriene (TMA-DPH) which are embedded in the deep and surface regions of the membrane, respectively [21]. Materials and methods Patient selection The investigation was carried out in 32 Caucasian male patients (mean age 43.9 f 10.8 years, range 19-61) affected by primary hyperlipoproteinemia (HLP). Secondary hyperlipoproteinemia due to hypothyroidism, diabetes, nephrotic syndrome, biliary obstruction, pancreatitis or dysglobulinemia had been excluded before. HLP was classified according to WHO criteria [22]. We
studied 12 patients with type IIA (familial hypercholesterolemia), 13 patients with type IIB (combined hyperlipidemia) and 7 patients with type IV (endogenous hypertriglyceridemia). These patients recruited in the lipid clinic had not taken any medication known to modify rheological parameters in the 3 months prior to the study. Fifteen normal Caucasian male healthy volunteers (mean age 33.1 + 5.1, range 24-45) with a normal blood lipid profile and a normal fasting blood glucose were selected from the medical staff of the hospital. All had neither symptoms nor a family history of hyperlipidemia or diabetes mellitus. Both patients and controls were non smokers. The study was approved by the “Centre Hospitalier et Universitaire de Nancy” Ethical Committee. Blood lipid determinations Blood samples were obtained by venipuncture after a 12 h overnight fast. Total serum cholesterol and triglycerides were enzymatically determined in a Technicon Autoanalyser. Total apo AI and apo B were determined by laser immunonephelometry using specific antibodies from Behring (Behring, Rueil-Malmaison, France). High density lipoprotein (HDL) cholesterol was measured in serum after precipitation of VLDL and LDL with phosphotungstic acid [23]. Rheological parameters Plasma viscosity was determined at 37 o C using an automatic capillary viscometer (AMTEC, Villeneuve-Loubet, France). Blood viscosity at a corrected hematocrit (Ht 40%) was determined at 37 o C for 7 shear rates (y = 0.50 to 128 s-l) using a coaxial cylinder rotational viscometer (LS 30, Contraves, Zurich, Switzerland). Erythrocyte aggregation at a corrected hematocrit (Ht = 30%) was determined at 37” C with a Myrenne aggregometer (Myrenne, GmbH, Aachen, F.R.G.) based on the analysis of transmitted light by a blood suspension in a cone-plan viscometer [24]. Platelet preparation Platelets were isolated from venous blood by a metrizamide gradient centrifugation as previously described [25]. After isolation, platelets were resuspended in a phosphate buffer at a concentra-
233 tion of 3 x 10’ cells/ml rescent labelling.
and processed
for fluo-
Red cell membrane preparations The Heidrich and Leutner method [26] was used for preparing the red cell ghosts. Following lysis in a 5 mM phosphate buffer, pH 8, the suspensions were washed in the same buffer until every trace of hemoglobin was removed. The membranes obtained are referred to as “unsealed” ghosts. Heven and Solomon’s method [27] was applied for preparing “resealed” ghosts. The protein concentration in the membrane suspensions was then assessed using the Lowry test and was systematically adjusted to a concentration of 100 pg of protein per ml. The fluorescence assessments were carried out within 5 h after blood sampling. Determination of erythrocyte membrane and platelet lipid fluidity The fluorescent probes DPH (Aldrich-Chimie, Strasbourg, France) and TMA-DPH (Molecular probes, Eugene, OR, U.S.A.) were dissolved in tetrahydrofuran (THF) and N, N-dimethyl formamide (DMF), respectively, at a concentration of 2 x lop3 M. The solutions were stored in the darkness at 4°C. Labelling was carried out by adding 3 ~1 of the above solutions to 3 ml of both membrane and the cell suspension. Measurements were carried out at 25 o C. Fluorescence polarization was assessed by means of a continuous excitation instrument (Fluofluidimeter, Sefam, Vandoeuvre-l&s-Nancy, France). In all cases, non-labelled membrane sus-
TABLE
pensions were measured before each series of experiments in order to calculate the correcting factors. After determining the signals due to “blanks” (non-labelled samples), Ib,,, Ib, , the fluorescence intensities of the samples were assessed, If,,, If i Fluorescence steady-state anisotropy (r ) and fluorescence intensity were assessed, based on the following relationships: (r)
= (If,, - Ib,,) - (Ifl
-Ib,)
I, with Jr = (If,, - Ib,,) + 2(If,
-Ib,)
Statistical analysis of the results The results were expressed as mean + standard deviation. They were analysed using the non-parametric Mann-Whitney test. Statistical significance was assumed at P -C 0.05. Any relation between the different biochemical, rheological and molecular parameters of this study was tested by means of Spearman’s multiple correlation test. Results The plasma lipids and lipoprotein profiles of the studied subjects are reported in Table 1. As expected, mean plasma total cholesterol and triglyceride levels were higher in patients compared with the normal subjects. HDL-cholesterol levels were lower in types IIB and IV hyperlipoproteinemias.
1
PLASMA
Phenotype
LIPID
VALUES
Cholesterol
Triglycerides
Apo Al
Apo B
HDL-
(mg/dl)
@g/d11
@g/d11
@g/dl)
cholesterol
182+19
**
55+16
177+17
**
47ill
*
49+42
(w/d0 IIA
335i44
**
101+39
*
166 * 34
IIB
293+42
**
207*62
**
175*46
IV
257i42
*
425f20
**
123+42
145k34
Controls
195&27
137+30
91+27
* P < 0.05. * *P c 0.001 versus controls.
71+22
*
58+
8
234 TABLE 2 BLOOD VISCOSITY IN NORMAL
SUBJECTS AND HYPERLIPOPROTEINEMIC
Group
Blood viscosity (mPa/s)
Shear rate:
0.5 s-1
0.9 s-1
HA IIB IV Controls
18.6*3.0 21.353.7 20.6+2.0 18.9+1.9
153~2.7 15.9k2.2 * 15.8*2.5 * 13.8 + 1.2
** *
3.2 s-’ 9.2 f 1.0 9.9kl.l * 10.3 + 1.2 * 8.6 f 0.5
PATIENTS
8.1 s-t
20.4 s-l
128 s-1
6.9 f 0.5 7.3 f 0.7 7.5+0.7 * 6.7kO.3
5510.3 5.9 & 0.4 5.8kO.3 5.4kO.2
3.8iO.l 3.9kO.2 3.9 * 0.3 3.8kO.l
* P < 0.05, * * P < 0.01versus control.
TABLE 3 PLASMA VISCOSITY AND ERYTHROCYTE AGGREGATION IN NORMAL SUBJECTS AND HYPERLIPOPROTEINEMIC PATIENTS Group
Plasma viscosity (cSt)
Erythrocyte aggregation index
IIA IIB IV
1.24k 0.05 1.29kO.08 ** 1.27kO.06 **
17.2k2.0 18.0+4.0 17.8*2.5
Controls
1.23 + 0.04
15.3 * 3.1
** ** **
* * P i 0.01versus controls.
The values of rheological parameters are summarized in Tables 2 and 3. Significant differences in blood viscosity at low shear rates and plasma viscosity were only noted between control subjects paand types IIB and IV hyperlipoproteinemic erythrocyte aggregation was tients. In contrast, significantly higher in the 3 groups of patients
with hyperlipoproteinemia compared with the normal subjects. The fluorescence anisotropy of TMA-DPH and DPH embedded in erythrocyte ghosts and intact platelets was determined in both controls and patients. The results recorded on erythrocyte membranes are summarized in Table 4. Concerning both unsealed and resealed membranes and the two fluorescent probes TMA-DPH and DPH, no significant differences were observed between type IIA patients and the control group. In contrast, in types IIB and IV patients, fluorescence anisotropy values of TMA-DPH were significantly lower in unsealed and resealed membranes. For DPH, a significant decrease in fluorescence anisotropy values was only observed in unsealed and resealed erythrocyte membranes of type IIB patients. The results on platelets are summarized in Table 5. As for erythrocyte membranes, no differences between the type IIA group and the control group were observed with respect to TMA-
TABLE 4 FLUORESCENCE ANISOTROPY VALUES OF TMA-DPH AND DPH EMBEDDED NORMAL SUBJECTS AND HYPERLIPIDEMIC PATIENTS
IN ERYTHROCYTE
MEMBRANES
TMA-DPH
Group
DPH Unsealed membranes
Resealed membranes
Unsealed membranes
Resealed membranes
HA IIB IV Controls
0.238 + 0.002 0.236 + 0.002 * 0.238 +0.003 0.239 +O.OOl
0.238 0.236 0.239 0.242
0.258 0.255 0.254 0.260
0.266 0.266 0.261 0.270
* P < 0.05vs controls.
f f f f
0.002 0.001 * 0.002 0.001
+ + + +
0.001 0.001 * 0.002 * 0.001
f + f +
0.002 0.001 * 0.002 * 0.001
OF
235 TABLE
5
FLUORESCENCE ANISOTROPY VALUES OF TMA-DPH AND DPH EMBEDDED IN PLATELETS OF NORMAL SUBJECTS AND HYPERLIPIDEMIC PATIENTS
Group
DPH
TMA-DPH
ITA IIB
0.224 * 0.002 0.208 * 0.004 * * 0.1@3*0.005 * * 0.223 + 0.003
0.287 + 0.002 0.285 & 0.002 * 0.285 +0.002 * 0.291+ 0.002
IV Controls
* P < 0.05, * * P i 0.01vs.controls.
DPH or DPH. In types IIB and IV groups, a significant decrease in anisotropy values was noted for the two fluorescent probes. When patients and control data are grouped together, a negative relationship was observed between the fluorescence anisotropy values of TMADPH and DPH embedded in platelets and plasma triglyceride amounts (r = -0.31, P < 0.01, and Y = - 0.74, P -c 0.001, respectively). A weak negative relationship was observed between the fluorescence anisotropy of TMA-DPH incorporated in unsealed erythrocyte membranes and triglyceride levels (r = -0.33 P < 0.05). There was no significant relationship between classical rheological parameters and molecular parameters except for the fluorescence anisotropy of DPH embedded in unsealed and resealed ghosts which were negatively related to plasma viscosity (r = - 0.32 P 80 years). Ditzel and Kampmann [28] and Voisin et al. [29] did not find any difference in blood viscosity and in erythrocyte aggregation before 60-65 years. Moreover, the determination of normal values in our laboratory showed no significant difference between two groups with ages corresponding to the controls and patients of the present study. Given the fact that erythrocyte aggregation is considered to be involved in the increase in blood viscosity at low shear rates [30], it may seem paradoxical that the blood viscosity of IIA patients at low shear rate is not significantly different from that of healthy controls. However, one must keep in mind that erythrocyte aggregation is a complex phenomenon which depends on both the level of plasma proteins (immunoglobulins, fibrinogen, albumin, etc.) and the intrinsic properties of red cells. Erythrocyte aggregation can be approached in terms of kinetic, structural and rheological parameters and it is likely that blood viscosity at low shear rates only reflects one component of erythrocyte aggregation [31]. Moreover, from a technical point of view, the measurement of erythrocyte aggregation may be a more sensitive approach than blood viscosity determinations. The hypothesis of an alteration of intrinsic properties of erythrocytes is supported by the data on the fluorescence anisotropy values of fluorescent probes: DPH and TMA-DPH embedded in red cell membranes. It is well admitted that the fluorescence anisotropy of such probes is inversely related to the lipid fluidity of the probe immediate surroundings [ll]. Our results provide evidence that type IIB and IV-HLP are significantly associated with an increase in the fluidity near the polar heads of the phospholipids where the fluorescent probe TMA-DPH is located. No significant difference is observed for type IIA-HLP. By contrast,
236 a significant increase in the fluidity of the central hydrophobic core characterized by DPH as a fluorescent probe is only recorded in type IIB-HLP. A significant increase in the fluidity of intact platelets is also noted in the superficial and deep areas of cell lipid regions from types IIB and IV patients. There are no significant differences between type IIA-HLP and normal subjects in any of the molecular parameters measured with TMADPH and DPH. The fact that the mean ages of type IIA and control groups are lower than that of types IIB and IV groups cannot explain the results. The process of tissue and cell aging is essentially associated to a progressive decrease in lipid fluidity [32-341. The erythrocyte findings are probably related to previous results showing a decrease in cholesterol/phospholipid (C/PL) ratio of erythrocyte membrane in IV-HLP patients [35,36]. They may be relevant to the high amounts of plasma triglycerides, Indeed, it has been shown that triglyceride-emulsion particles could promote the efflux or block the influx of cholesterol in cultured macrophages [37]. Whether a similar mechanism exists for erythrocytes remains to be elucidated. Nevertheless, 80% of erythrocyte membrane cholesterol can be removed by lipoproteins of reduced cholesterol content [38]. Since added amounts of cholesterol decrease the fluidity of membranes, it may be suggested that the presence of high amounts of plasma triglycerides could enhance cholesterol efflux from the cell membrane resulting in an increase in the lipid fluidity of erythrocytes. Where platelets are concerned, the results are not fully understood. Despite the fact that an inverse relationship was shown in vitro between C/PL ratio and fluidity [17], a free cholesterol/phospholipid ratio was normal in type IV-HLP patients [39]. Therefore, it is not possible to assume for platelets that the higher fluidity in IIB and IV HLP patients is due to a reduced C/PL ratio. Our observations suggest that hypercholesterolemia does not induce marked changes in the lipid fluidity of erythrocytes and platelets. In another work, no change in red cell C/PL ratio was also observed in IIA-HLP patients [20]. In platelets an increase in C/PL ratio of plasma membrane
(+7%) was previously reported [40]. It may be suggested that this weak increase in C/PL ratio only induces discrete modifications in membrane fluidity. The alterations are perhaps not detected by the fluorescent probes TMA-DPH and DPH which label both membrane and cytoplasmic lipid regions. By contrast, it appears that this molecular rheological parameter could be affected by the concentration of plasma triglycerides. Further studies are needed to determine the exact relations between HLP associated lipid fluidity alterations, plasma triglycerides and classical haemorheological patterns. Acknowledgement We are grateful for the expert technical assistance of Mrs. S. Droesch, G. Cauchois and M. Gentils. References of atherosclerosis. An update. 1 Ross, R., The pathogenesis New Engl. J. Med., 314 (1986) 488. B., Poindexter, B.J. and Schaeffer, 2 Corash, L., Andersen, E.J., Platelet functions and survival in patients with severe hypercholesterolemia. Arteriosclerosis, 1 (1981) 443. a problem of the biology of 3 Ross, R., Atherosclerosis: arterial wall cells and their interaction with blood components. Arteriosclerosis, 1 (1981) 293. 4 Mustard, J.F. and Packham, M.A., The role of blood and platelets in atherosclerosis and the complications of atherosclerosis. Thromb. Diath. Haemorh., 33 (1975) 444. in atherosclerosis and 5 Schwartz, S.M., Cellular proliferation hypertension. Proc. Sot. Exp. Biol. Med., 173 (1983) 1. 6 Tremoli, E., Madema, P., Colli, S., Morazzoni, G., Sirtori, M. and Sirtori, C.R., Increased platelet sensitivity and thromboxane B, formation in type-11 hyperlipoproteinaemic patients. Eur. J. Clin. Invest., 14 (1984) 329. I Carvalho, A., Carvalho, M., Colman, R.W. and Lees, R.S., Platelet function in hyperlipoproteinemia. N. Engl. J. Med., 290 (1974) 434. M., Platelet mem8 Berlin, E., Shapiro, S.G. and Friedland, brane fluidity and aggregation of rabbit platelets. Atherosclerosis, 51 (1984) 223. fluidity 9 Shiga, T. and Maeda, N., Influence of membrane on erythrocyte functions. Biorheology, 17 (1980) 485. polarization applied 10 Stoltz, J.F., Donner, M., Fluorescence to cellular microrheology. Biorheology, 22 (1985) 227. 11 Stoltz, J.F. and Donner, M., Relations between molecular rheology and red blood cell structure: methods and clinical approaches. Clinical Haemorheology, 5 (1985) 813. H., Arntz, H.R. and Klemens, U., Studies of 12 Leonhardt, plasma viscosity in primary hyperlipoproteinemia. Atherosclerosis, 28 (1977) 29.
237 H. and Amtz, H.R., Blutviskositat und Etyth13 Leonhardt, rozytenflexibilitat bei prim&en Hyperlipoproteinlmien. Klin. Wschr., 56 (1978) 271. 14 Pfeiffer. M. and Tilsner, V.. Der Einfluss von Clofibrat auf die Plasmaviskositat bei Hyperlipoproteinlmien. Med. Klin., 73 (1978) 60. S., Rousselle, D., Voisin, P. and 15 Stoltz. J.F., Gaillard, Drouin, P., Rheological study of blood during primary hyperlipoproteinemia. Clin. Hemorheol., 3 (1981) 227. 16 Seplowitz, A.H.. Chien, S. and Smith, F.R., Effects of lipoproteins on plasma viscosity. Atherosclerosis, 38 (1981) 89. microviscosity 17 Shattil, S.J. and Cooper, R.A., Membrane and human platelet function. Biochemistry, 15 (1976) 4832. R., Marenah, 18 Hassall. D.G., Forrest, L.A., Bruckdorfer, C.B., Turner, P.. Cortese, C., Miller, N.E. and Lewis, B., Influence of plasma lipoproteins on platelet aggregation in a normal male population. Arteriosclerosis, 3 (1983) 332. 19 Malle. E., Cries, A., Kostner, G.M., Pfeiffer, K., Nimpf, J. and Hermetter. A.. Is there any correlation between platelet aggregation, plasma lipoproteins, apoproteins and membrane fluidity of human blood platelets? Thromb. Res., 53 (1989) 181. of cell-membrane fluidity in 20 Cooper, R.A.. Abnormalities the pathogenesis of disease. New Engl. J. Med., 297 (1977) 371. 21 Donner. M. and Stoltz, J.F., Comparative study on fluorescent probes distributed in human erythrocytes and platelets. Biorheology. 22 (1985) 385. 22 Beaumont. J.L., Carlson, I.A., Cooper, G.R., Fejfar, Z., Fredrickson, D.S. and Strasser, T., Classification of hyperlipidemia and hyperlipoproteinemia. Bull. WHO, 43 (1970) 891. M., Scholcnick, H.R. and Morfin, R.. Rapid 23 Burstein. method for isolation of lipoproteins from serum by precipitation with polyanions. J. Lipid Res., 11 (1970) 583. 24 Schmid-Schiinbein. H., Volger, E., Teitel, P.. Kiesewetter, H.. Daver, V. and Heilmann, L., New hemorheological techniques for the routine laboratory. Clin. Hemorheol., 2 (1982) 93. S.. Bredoux, R., Rendu, F. Jeanneau, C., 25 Levy-Toledano, Savanau, E. and Dassin, E., Isolement et fonction des plaquettes. Nouv. Rev. Fr. Hemat., 16 (1976) 367. 26 Heidrich. H.G. and Leutner, G., Two types of vesicles from the erythrocyte ghost membrane differing in surface charge. Eur. J. Biochem., 41 (1974) 37.
27 Heven, S. and Solomon, A.K., Permeability of human erythrocyte membrane vesicles to alkali cations. Biochim. Biophys. Acta, 550 (1979) 393. 28 Ditzel, J. and Kampmann. J., Whole-blood viscosity, hematocrit and plasma protein in normal subjects at different ages. Acta Physiol. Stand., 81 (1971) 264. 29 Voisin. P.. Chevillard, F.. Penin, F. and Stoltz. J.F.. Influence of patients age in the determination of hemorheological parameters. In: G. Potron, J.F. Stoltz, M. Gueguen, M. Boisseau (Eds.), Standardization in Clinical Haemorheology, Laboratoires Boots-Dacourt, 1987, 9. R.J. and Gregersen. M.I., 30 Chien, S., Usami, S., Dellenback, Shear dependent interactions of plasma protein with erythrocytes in blood. Am. J. Physiol., 219 (1970) 143. 31 Stoltz. J.F., Paulus, F. and Donner, M.. Experimental approaches to erythrocyte aggregation. Clin. Hemorheol., 7 (1987) 33. fluidity and cellular functions. 32 Shinitzky, M., Membrane In: Meir Shinitzky (Ed.), Physiology of Membrane Fluidity, Vol. 1. CRC Press, Boca Raton, FL, 1984. p. 1. 33 Rivray. B., Bergman, S., Shinitzky, M. and Globerson, A.. Correlations between membrane viscosity serum cholesterol. lymphocyte activation and aging in man. Mech. Ageing Dev.. 12 (1980) 119. 34 Terranova, R.. Alberghina, M. and Camazzo. G.. Composizione lipidica e fluidita della membrana eritrocitaria nell’anziano. Ric. Clin. Lab., 15 (1985) 327. 35 Vakakis. N., Redgrave, T.G., Small, D.M. and Castelli, W.P., Cholesterol content of red blood cells and low-density lipoproteins in hypertriglyceridemia. Biochim. Biophys. Acta, 751 (1983) 280. 36 Neerhout. R.C., Erythrocyte stromal lipids in hyperlipemic states. J. Lab. Clin. Med., 71 (1968) 448. 37 Aviram, M., Williams, K.J. and Deckelbaum, R.J., Macrophage cholesterol removal by triglyceride phospholipid emulsions, Biochem. Biophys. Res. Commun.. 155 (1988) 709. M.H., Rates of cholesterol exchange between 38 Gottlieb, human erythrocytes and plasma lipoproteins. Biochem. Biophys. Acta, 600 (1980) 530. 39 Shastri, K.M.. Carvalho, A.C.A. and Lees. R.S., Platelet function and platelet lipid composition in the dislipoproteinemia. J. Lipid Res., 21 (1980) 467. 40 Shattil. S.J., Bennett, J.S., Colman, R.W. and Cooper. R.A.. Abnormalities of cholesterol-phospholipid composition in platelets and low-density lipoproteins of human hyperbetalipoproteinemia. J. Lab. Clin. Med., 89 (1977) 341.
