Elevated Microviscosity in Membranes of Erythrocytes Affected by Hereditary Spherocytosis B. ALONI,M. SHINITZKY,* S. Mom? AND A. LIVNE Research and Development Authority and Department of Biology, Ben-Curion University ofthe Negev, Beer Sheva, *Laboratory of Membranes and Bioregulation, Weizmann Institute of Science, Rehovot, and ?Department of Pediatrics, Soroka Medical Center, Beer Sheva, Israel (Received 25 November 1974; acceptedfor publication 20 Febrriary 1975)
SUMMARY.Erythrocytes affected by hereditary spherocytosis (HS), obtained from several splenectoniized patients, showed a varying degree of elevated osmotic fragility. In order to evaluate a possible role of the erythrocyte membrane lipids in HS, microviscosity of the membrane lipid core was measured by a fluorescencepolarization technique. Intact HS-affected red cells, as well as their ghost membranes and liposomes prepared from their lipid extract, all showed a distinctly higher microviscosity than the respective normal control. The increased microviscosity correlated with the severity of HS. The data support the proposition that the defect in HS-affected red cells is associated, at least in part, with alterations in the membrane lipids. Hereditary spherocytosis (HS) is the most prevalent of the hereditary blood diseases in noncoloured people (Jandl, 1968; Jacob et al, 1972). The disorder apparently arises from an intrinsic defect in the red cell membrane, but the chemical nature of the lesion is not yet clear. While several studies have suggested that some proteins in the erythrocyte membranes are either altered (Jacob et al, 1971) or missing (Hayashi et al, 1974) in HS, other studies reported electrophoretic (Kitao et al, 1973) or immunological (Gomperts et al, 1971) homogeneity of membrane proteins from HS and normal red cells. Essentially no alteration could be detected in the profile of the various phospholipids and in the molar ratio of cholesterol to phospholipids (Phillips & Roome, 1962; de Gier et al, 1964; Reed & Swisher, 1966; Cooper & Jandl, 1969). However, some profound differences were recently observed in the phospholipid acyl groups of HS-affected cells (Kuiper & Livne, 1972). In addition, linolenoyl sorbitol, a synthetic lipid, was found to decrease the osmotic fragility of HS red cells in a striking manner: the greater the osmotic fragility of red cells from different patients, the greater the protective effect afforded by the acquisition of this lipid. These observations indicated a causal relationship between the lipid composition and the increased fragility of red cells in hereditary spherocytosis (Livne et al, 1973). Differencesin the hydrocarbon chains of the phospholipids should induce changes in the Correspondence: Professor A. Livne, Department of Biology, Ben-Gurion University, P.O. Box 1025,Beer Sheva, Israel. 117
118
B. Aloni et al
dynamics of the lipid core in the HS-affected erythrocyte membranes. A highly sensitive and accurate technique for monitoring dynamic properties of the lipid regions is based on fluorescence polarization of an embedded hydrocarbon probe. This technique was successfully applied to various membrane systems, including red blood cells (Shinitzky et al, 1971;Rudy & Gitler, 1972; Cogan et al, 1973 ; Shinitzky & Barenholz, 1974; Aloni el al, 1974; Shinitzky & Inbar, 1974). The use of this method was therefore applied in the present study as a probe for changes in the dynamic properties of the hydrocarbon chains in HS-affected cells. MATERIALS AND METHODS Heparinized blood was collected from five HS patients (listed in serial numbers: HSI, HS,, etc.) and from five normal donors. A splenectomized healthy person was included among the normal donors, since all the HS patients were splenectomized. The red cells were washed three times in 155 m~ NaCl buffered with z mM sodium phosphate, pH 7.4. Reticulocytes did not exceed 1% in all samples. For measurement of osmotic fragility, an aliquot ofzopl ofstock cell suspension (30%) was rapidly mixed at 25°C with 3 ml of z mM sodium phosphate, pH 7.4, containing NaCl at various concentrations. After 10 min at 25°C the cells were centrifuged at 2000 g for 5 min. The yo haemolysis was determined by measuring the absorbance of haemoglobin in the supernatant at 540 nm. Haemoglobin-free ghosts were prepared according to Dodge et a1 (1963). Lipids were extracted according to a procedure adapted to erythrocyte lipids (Kuiper et al, 1971). Liposomes were prepared as follows: A sample of 10 mg lipid extract dissolved in chloroform was dried under reduced pressure in a round-bottom flask and suspended in 0.5 ml of an aqueous solution containing 75 mM KCl, 75 mM NaCl and 0.1 M Tris-HC1, pH 7.6. The suspension was vigorously shaken under N2 at 37°C for 10 min and then incubated for 30 min at room temperature.
Fluorescence Polarization Determination Membranes were labelled with the fluorescence polarization probe I ,bdipheny1-1,3,5o f 2 x I O - ~M DPH in tetrahydrofuran was hexatriene (DPH), as follows: An aliquot (10~1) injected into vigorously stirred 10 ml of 155 mM NaCl pH 7.4 containing either red cells (0.05%),ghosts (30 p g proteinlnd) or liposomes (50 pl of stock suspension). The suspensions of intact cells and ghosts were fortified with 10mM glucose prior to the addition of DPH. The suspensions of labelled intact cells and ghosts were incubated for 10 h at 4°C and the excess of DPH was removed by washing the cells and the ghosts twice in the label-free medium. The labelled liposomes were measured after I h incubation at room temperature. The contribution of the excess DPH was regarded as negligible since aqueous dispersions of DPH are virtually non-fluorescent (Shinitzky & Barenholz, 1974). Fluorescence polarization and intensity were measured with an instrument which was previously described (Teichberg& Shinitzky, 1973). A 366 nm band generated from a 500 W mercury arc, which was passed through a Glan-Thompson polarizer, was used for excitation. The emitted light was detected in two independent cross polarized channels equipped with Glan-Thompson polarizers passing through a z M sodium nitrite solution used as a cut-off and llperpendicular ) (I,) to the directions filter. The emission intensities polarized parallel (I of polarization of the excitation beam, were obtained by a simultaneous measurement of
Microviscosity in Erythrocyte Membranes
119
i l l /and L1,. These intensities relate to the degree of fluorescence polarization, P, and to the total fluorescence intensity, F, by the following equations:
With the aid of P and F the microviscosity, q, was determined (Shinitzky & Barenholz, 1974; Shinitzky & Inbar, 1974). In order to detect possible phase transitions in the lipid layer of the studied systems, q was determined at a series of temperatures in the range of 0-40" and the results were plotted as log f vs I/ T, where T is the absolute temperature. A system of an invariant phase obeys the relation In fi = A + A E / R T where AE is the fusion activation energy, R is the gas constant, and A is a constant relating to the system (Shinitzky et a!, 1971). For invariant phases such plots should thus yield straight lines from which slopes AE can be derived.
120
110
I00
90
80
70
60
NaCl (mu)
FIG I . Osmotic fragility of normal and HS-affected erythrocytes at 2s°C. The dotted area represents the range of the osmotic fragility in samples from five normal donors, including a donor after splenectomy.
RESULTS Elevated osmotic fragility is a cardinal feature of HS, commonly used for diagnostic purposes. Fig I represents the osmotic fragility curves of erythrocytes from five HS patients. Their
B. Aloni et a1
I20
Temperoture ("C) 25 15
35
I
I
*I
I
I
3.2
3.3
5
I
I
I
I
3.4 3.5 I /Tx 10' (" K - ' I
I
3.6
:
I
FIG 2. Temperature dependence of the microviscosity of ghost membranes derived from normal red cells ( 0 ) and from HSz (O), HS3 (A) and HSS ( A ) .
deviation from normal blood varies within a considerable range. Ghost membranes from HS, and HS, (showing an intermediate level of fragility) and of HS, (being most fragile) were further examined. Fig z displays the temperature dependence of the microviscosity in these membranes and in membranes from normal subjects. The hydrocarbon region in both normal and HS membranes consists of an invariant phase between o and 40°C, which is characterized by a fusion activation energy of AE = 7.ofo.z kcallmole. The features of invariant phase and AE- 7 kcallmole are typical of virtually all mammalian membranes which were determined by this method so far (Rudy & Gitler, 1972; Aloni et a!, 1974; Shinitzky & Inbar, 1974). The membranes derived from HS erythrocytes, however, exhibit distinctly higher microviscosity than normal, in apparent correlation with the increased osmotic fragility of the intact cells (Fig I). This correlation holds also for the microviscosity of intact erythrocytes, as shown in Table I. It is noteworthy that, in general, microviscosity
(MM NaClfor 50% haemolysir)
Minoviscosity (poise)
76 (74-78) 74 78 88 97
4.0 (3.8-4.2) 4.1 4.3 6.8 7.4
Osmotic fragility Normal (four donors) Normal, splenectomized
HS1 HS2 HSs
Microviscosity in Erythrocyte Membranes
I21
values in intact cells are lower than in the ghost membranes (Aloni et al, 1974). The osmotic fragility and microviscosity of erythrocytes from a splenectomized person without HS were the same as found in other normal donors. Fig 3 demonstrates that liposomes composed of lipids from HS erythrocytes possess higher values of microviscosity throughout the entire temperature range of 0-40°C. 35
I
I
Temperature (“c) 25 15 I
I
5 I
FIG 3. Temperature dependence of the microviscosity of liposomes prepared from lipid extracts of normal and HSs erythrocytes.
DISCUSSION Solomon (1974) computed the apparent viscosity of human red cell membranes fiom the diffusion coefficient of methanol across the membrane and reported a value of 1.7 poise at 21OC. Evidently the absolute value of membrane viscosity depends on the method of estimation. The exact reasons for the variance between the values obtained by Solomon’s method and the method used in this study are not clear. The increased microviscosity of the HS erythrocyte membrane may be associated with at least some ofthe typical features in the affected cell membranes. Thus, in analogy, the osmotic fragility of mycoplasma cells was elevated when the membrane fluidity was decreased by an altered lipid composition (Razin et al, 1966). Furthermore, lowering the temperature (which, in turn, elevates membrane microviscosity !) affects the erythrocyte membranes, so that the osmotic fragility of the cells is increased (Murphy, 1969; Livne & Raz, 1971). HS red cell ghosts are intrinsically stiffer than normal, resisting passage in glass microcapillary tubes (LaCelle & Weed, 1969). It is possible that this feature may be associated with the increased microviscosity of the membrane lipid core. Membranes of HS-affected erythrocytes contain decreased quantities of long-chain fatty acid conjugates of sphingomyellin, lecithin and
I22
B. Aloni et a2
phosphatidylserine (Kuiper & Livne, 1972). Furthermore, cells have a decreased ratio of surface area to cell volume, while the lipid content after splenectomy is similar to normal cells (Cooper & Jandl, 1969). Thus, a tighter packing of the membrane lipids in HS-erythrocytes is anticipated, as indeed the microviscosity data illustrate. Our results support the claim (Livne et a!, 1973) that the membrane lesion in HS is associated, at least in part, with modified lipid properties. The reasons for the variable severity of HS among different patients are not clear. The present data lend further support to the proposition that the variability is associated with the extent of the modification of lipid composition, of protein components, or both. ACKNOWLEDGMENTS
We wish to thank Dr I. Ben-Basat who provided HS blood samples and to Dr R. A. Cooper for critically reading the manuscript. This work was supported by Grant No. 1248 from the National Council for Research and Development to A. Livne and B. Aloni and by Grant No. 49340 from the Israel National Academy of Sciences to M. Shinitzky. REFERENCES ALONI, B., SHINITZKY, M. & LIVNE, A. (1974) Dynamics of erythrocyte lipids in intact cells, in ghost membranes and in liposomes. Biochimica et Biophysica Acfa, 348, 438. U., SHINITZKY, M., WEEIER, G. & NISHIDA, T. COGAN, (1973) Microviscosity and order in the hydrocarbon region of phospholipid and phopholipid-cholesterol dispersions determined with fluorescent probe. Biochemisfry, 12, 521. COOPER,R.A. & JANDL. J.H. (1969) The role of membrane lipids in the survival of red cells in hereditary spherocytosis. Journal of Clinical Investigation, 48. 736. C. & HANAHAN, D.J. (1963) DODGE, J.T., MITCHELL, The preparation and chemical characteristics of hemoglobin-free ghosts of human erythrocytes. Archives ofBiochemistry and Biophysics, ZOO, 1x9. M.C. & GIER,J. DE,DEENEN,L.L.M. VAN, VERLOOP, GASTEL, C. VAN (1964) Phospholipid and fatty acid characteristics of erythrocytes in some cases of anaemia. British journal of Haernatology, 10, 246. GOMPERTS, E.D., METZ,J. & ZAIL, S.S. (1971) Immunological homogeneity of membrane proteins from hereditary spherocytic, hereditary elliptocytic, and normal red cells. BrifishJournal of Haemutology, 20, 443. HAYASHI, S., KOOMOTO, R.,YANO,A., ISHIGAMI,S., TSUJINO, G., SAEKI, S. & TANAKA,T. (1974) Abnormality in a specific protein of the erythrocyte membrane in hereditary spherocytosis. Biochemical and Biophysical Research Communications, 57, 103 8 . JACOB,H., AMSDEN, T. & WHITE, J. (1972) Membrane microfilaments of erythrocytes: alteration in intact
cells reproduces the hereditary spherocytosis syndrome. Proceedings ofthe National Academy of Sciences ofthe United States ofAmerica, @, 471. JACOB,H.S., RUBY,A., OVERLAND, E.S. & MAZIA,D. (1971) Abnormal membrane protein of red blood cells in hereditary spherocytosis. Journal af Clinical Invesfigation, 50, 1800. J.H. (1968) Hereditary spherocytosis. Hereditary JANDL, Disorders of Erythrocyte Metabolism (Ed. by E. Beutler), p zog. Grune & Stratton, New York. KITAO, T., HATTORI, K., TAKESHITA, M. (1973) Electrophoretic analysis of the major proteins of the human red cell membrane. Clinica Chimica Acfa, 49. 353. KUIPER,P.J.C. & LIVNE,A. (1972) Differences in fatty acid composition between normal human erythrocytes and hereditary spherocytosis affected cells. Biochimica et Biophysicu Acta, 260, 755. KUIPBR, P.J.C., L m , A. & MEYBRSTEIN, N. (1971) Changes in lipid composition and osmotic fragility of erythrocytes of hamster induced by heat exposure. Biochimica ef Biophysica Acta, 248, 300. LACELLE, P.L. & WEED,R.I. (1969) Reversibility of abnormal deformability and permeability of the hereditary spherocyte. (Abstract). Blood, 34, 858. LIVNE,A., ALONI,B., MOSES,S. & KUIPER,P.J.C. (1973) Linolenoyl sorbitol and the fragility ofhereditary spherocytes. British journal of Haematology, 25, 429. LIVNB,A. & RAZ,A. (1971) Erythrocyte fragility and potassium efflux as affected by temperature and hemolyzing rate. FEBS Letters, 16,99. MURPHY, J.R. (1969) Erythrocyte osmotic fragility and
Microviscosity in Erythrocyte Membranes cell water--influence of pH and temperature. Journal of Laboratory and Clinical Medicine, 74, 319. PHILLIPS, G.B. & R o o m , N.S. (1962) Quantitative chromatographic analysis of the phospholipids of abnormal human red blood cells. Proceedings of the Society for Experimental Biology and Medicine, IW, 360.
RAZIN,S,,TOURTELLOTTE, M.E., MCELHANBY, R.N. & POLLACK, J.D. (1966) Mluence of lipid components of Mycoplasma laidlawii membrane on osmotic fragility of cells. Journal OfBaCteriology, 91, 609. RRED,C.F. & SWISHER, S.N.(1966) Erythrocyte lipid loss in hereditary spherocytosis. journal of Clinical Investigation, 45, 777. RUDY,B. & GITLER,C. (1972) Microviscosity of the cell membrane. Biochimica et Biophysica Acta, 288, 231. SHINITZKY,
M. & BARENHOLZ, Y. (1974) Dynamics of hydrocarbon layer in liposomes of lecithin and
123
sphingomyelin containing dicetylphosphate.Journal ofBiological Chemistry, 49. 2652. SHINITZKY, M., DIANOUX.A&., GITLFZ~~ C. & WELIER, G. (1971) Microviscosity and order of the hydrocarbon region of micelles and membrane determined with fluorescent probes. I. Synthetic micelles. Biochemistry, 10,2106. SHINITZKY,M. & INBAR, M. (1974) Difference in microviscosity induced by different cholesterol levels in the surface membrane lipid layer of normal lymphocytes and malignant lymphoma cells. Journal of Molecuhr Biology, 85, 603. SOLOMON, A.K. (1974) Apparent viscosity of human red cell membranes. Biochimica et Biophysica A d a , 373, 145.
TEICHBERG, V.I. & SHINITZKY.M. (1973) Fluorescence polarization studies of lysozyme and lysozymesaccaride complexes.Journal of Molecular Biology, 74, s19.
SUMMARY.Erythrocytes affected by hereditary spherocytosis (HS), obtained from several splenectoniized patients, showed a varying degree of elevated osmotic fragility. In order to evaluate a possible role of the erythrocyte membrane lipids in HS, microviscosity of the membrane lipid core was measured by a fluorescencepolarization technique. Intact HS-affected red cells, as well as their ghost membranes and liposomes prepared from their lipid extract, all showed a distinctly higher microviscosity than the respective normal control. The increased microviscosity correlated with the severity of HS. The data support the proposition that the defect in HS-affected red cells is associated, at least in part, with alterations in the membrane lipids. Hereditary spherocytosis (HS) is the most prevalent of the hereditary blood diseases in noncoloured people (Jandl, 1968; Jacob et al, 1972). The disorder apparently arises from an intrinsic defect in the red cell membrane, but the chemical nature of the lesion is not yet clear. While several studies have suggested that some proteins in the erythrocyte membranes are either altered (Jacob et al, 1971) or missing (Hayashi et al, 1974) in HS, other studies reported electrophoretic (Kitao et al, 1973) or immunological (Gomperts et al, 1971) homogeneity of membrane proteins from HS and normal red cells. Essentially no alteration could be detected in the profile of the various phospholipids and in the molar ratio of cholesterol to phospholipids (Phillips & Roome, 1962; de Gier et al, 1964; Reed & Swisher, 1966; Cooper & Jandl, 1969). However, some profound differences were recently observed in the phospholipid acyl groups of HS-affected cells (Kuiper & Livne, 1972). In addition, linolenoyl sorbitol, a synthetic lipid, was found to decrease the osmotic fragility of HS red cells in a striking manner: the greater the osmotic fragility of red cells from different patients, the greater the protective effect afforded by the acquisition of this lipid. These observations indicated a causal relationship between the lipid composition and the increased fragility of red cells in hereditary spherocytosis (Livne et al, 1973). Differencesin the hydrocarbon chains of the phospholipids should induce changes in the Correspondence: Professor A. Livne, Department of Biology, Ben-Gurion University, P.O. Box 1025,Beer Sheva, Israel. 117
118
B. Aloni et al
dynamics of the lipid core in the HS-affected erythrocyte membranes. A highly sensitive and accurate technique for monitoring dynamic properties of the lipid regions is based on fluorescence polarization of an embedded hydrocarbon probe. This technique was successfully applied to various membrane systems, including red blood cells (Shinitzky et al, 1971;Rudy & Gitler, 1972; Cogan et al, 1973 ; Shinitzky & Barenholz, 1974; Aloni el al, 1974; Shinitzky & Inbar, 1974). The use of this method was therefore applied in the present study as a probe for changes in the dynamic properties of the hydrocarbon chains in HS-affected cells. MATERIALS AND METHODS Heparinized blood was collected from five HS patients (listed in serial numbers: HSI, HS,, etc.) and from five normal donors. A splenectomized healthy person was included among the normal donors, since all the HS patients were splenectomized. The red cells were washed three times in 155 m~ NaCl buffered with z mM sodium phosphate, pH 7.4. Reticulocytes did not exceed 1% in all samples. For measurement of osmotic fragility, an aliquot ofzopl ofstock cell suspension (30%) was rapidly mixed at 25°C with 3 ml of z mM sodium phosphate, pH 7.4, containing NaCl at various concentrations. After 10 min at 25°C the cells were centrifuged at 2000 g for 5 min. The yo haemolysis was determined by measuring the absorbance of haemoglobin in the supernatant at 540 nm. Haemoglobin-free ghosts were prepared according to Dodge et a1 (1963). Lipids were extracted according to a procedure adapted to erythrocyte lipids (Kuiper et al, 1971). Liposomes were prepared as follows: A sample of 10 mg lipid extract dissolved in chloroform was dried under reduced pressure in a round-bottom flask and suspended in 0.5 ml of an aqueous solution containing 75 mM KCl, 75 mM NaCl and 0.1 M Tris-HC1, pH 7.6. The suspension was vigorously shaken under N2 at 37°C for 10 min and then incubated for 30 min at room temperature.
Fluorescence Polarization Determination Membranes were labelled with the fluorescence polarization probe I ,bdipheny1-1,3,5o f 2 x I O - ~M DPH in tetrahydrofuran was hexatriene (DPH), as follows: An aliquot (10~1) injected into vigorously stirred 10 ml of 155 mM NaCl pH 7.4 containing either red cells (0.05%),ghosts (30 p g proteinlnd) or liposomes (50 pl of stock suspension). The suspensions of intact cells and ghosts were fortified with 10mM glucose prior to the addition of DPH. The suspensions of labelled intact cells and ghosts were incubated for 10 h at 4°C and the excess of DPH was removed by washing the cells and the ghosts twice in the label-free medium. The labelled liposomes were measured after I h incubation at room temperature. The contribution of the excess DPH was regarded as negligible since aqueous dispersions of DPH are virtually non-fluorescent (Shinitzky & Barenholz, 1974). Fluorescence polarization and intensity were measured with an instrument which was previously described (Teichberg& Shinitzky, 1973). A 366 nm band generated from a 500 W mercury arc, which was passed through a Glan-Thompson polarizer, was used for excitation. The emitted light was detected in two independent cross polarized channels equipped with Glan-Thompson polarizers passing through a z M sodium nitrite solution used as a cut-off and llperpendicular ) (I,) to the directions filter. The emission intensities polarized parallel (I of polarization of the excitation beam, were obtained by a simultaneous measurement of
Microviscosity in Erythrocyte Membranes
119
i l l /and L1,. These intensities relate to the degree of fluorescence polarization, P, and to the total fluorescence intensity, F, by the following equations:
With the aid of P and F the microviscosity, q, was determined (Shinitzky & Barenholz, 1974; Shinitzky & Inbar, 1974). In order to detect possible phase transitions in the lipid layer of the studied systems, q was determined at a series of temperatures in the range of 0-40" and the results were plotted as log f vs I/ T, where T is the absolute temperature. A system of an invariant phase obeys the relation In fi = A + A E / R T where AE is the fusion activation energy, R is the gas constant, and A is a constant relating to the system (Shinitzky et a!, 1971). For invariant phases such plots should thus yield straight lines from which slopes AE can be derived.
120
110
I00
90
80
70
60
NaCl (mu)
FIG I . Osmotic fragility of normal and HS-affected erythrocytes at 2s°C. The dotted area represents the range of the osmotic fragility in samples from five normal donors, including a donor after splenectomy.
RESULTS Elevated osmotic fragility is a cardinal feature of HS, commonly used for diagnostic purposes. Fig I represents the osmotic fragility curves of erythrocytes from five HS patients. Their
B. Aloni et a1
I20
Temperoture ("C) 25 15
35
I
I
*I
I
I
3.2
3.3
5
I
I
I
I
3.4 3.5 I /Tx 10' (" K - ' I
I
3.6
:
I
FIG 2. Temperature dependence of the microviscosity of ghost membranes derived from normal red cells ( 0 ) and from HSz (O), HS3 (A) and HSS ( A ) .
deviation from normal blood varies within a considerable range. Ghost membranes from HS, and HS, (showing an intermediate level of fragility) and of HS, (being most fragile) were further examined. Fig z displays the temperature dependence of the microviscosity in these membranes and in membranes from normal subjects. The hydrocarbon region in both normal and HS membranes consists of an invariant phase between o and 40°C, which is characterized by a fusion activation energy of AE = 7.ofo.z kcallmole. The features of invariant phase and AE- 7 kcallmole are typical of virtually all mammalian membranes which were determined by this method so far (Rudy & Gitler, 1972; Aloni et a!, 1974; Shinitzky & Inbar, 1974). The membranes derived from HS erythrocytes, however, exhibit distinctly higher microviscosity than normal, in apparent correlation with the increased osmotic fragility of the intact cells (Fig I). This correlation holds also for the microviscosity of intact erythrocytes, as shown in Table I. It is noteworthy that, in general, microviscosity
(MM NaClfor 50% haemolysir)
Minoviscosity (poise)
76 (74-78) 74 78 88 97
4.0 (3.8-4.2) 4.1 4.3 6.8 7.4
Osmotic fragility Normal (four donors) Normal, splenectomized
HS1 HS2 HSs
Microviscosity in Erythrocyte Membranes
I21
values in intact cells are lower than in the ghost membranes (Aloni et al, 1974). The osmotic fragility and microviscosity of erythrocytes from a splenectomized person without HS were the same as found in other normal donors. Fig 3 demonstrates that liposomes composed of lipids from HS erythrocytes possess higher values of microviscosity throughout the entire temperature range of 0-40°C. 35
I
I
Temperature (“c) 25 15 I
I
5 I
FIG 3. Temperature dependence of the microviscosity of liposomes prepared from lipid extracts of normal and HSs erythrocytes.
DISCUSSION Solomon (1974) computed the apparent viscosity of human red cell membranes fiom the diffusion coefficient of methanol across the membrane and reported a value of 1.7 poise at 21OC. Evidently the absolute value of membrane viscosity depends on the method of estimation. The exact reasons for the variance between the values obtained by Solomon’s method and the method used in this study are not clear. The increased microviscosity of the HS erythrocyte membrane may be associated with at least some ofthe typical features in the affected cell membranes. Thus, in analogy, the osmotic fragility of mycoplasma cells was elevated when the membrane fluidity was decreased by an altered lipid composition (Razin et al, 1966). Furthermore, lowering the temperature (which, in turn, elevates membrane microviscosity !) affects the erythrocyte membranes, so that the osmotic fragility of the cells is increased (Murphy, 1969; Livne & Raz, 1971). HS red cell ghosts are intrinsically stiffer than normal, resisting passage in glass microcapillary tubes (LaCelle & Weed, 1969). It is possible that this feature may be associated with the increased microviscosity of the membrane lipid core. Membranes of HS-affected erythrocytes contain decreased quantities of long-chain fatty acid conjugates of sphingomyellin, lecithin and
I22
B. Aloni et a2
phosphatidylserine (Kuiper & Livne, 1972). Furthermore, cells have a decreased ratio of surface area to cell volume, while the lipid content after splenectomy is similar to normal cells (Cooper & Jandl, 1969). Thus, a tighter packing of the membrane lipids in HS-erythrocytes is anticipated, as indeed the microviscosity data illustrate. Our results support the claim (Livne et a!, 1973) that the membrane lesion in HS is associated, at least in part, with modified lipid properties. The reasons for the variable severity of HS among different patients are not clear. The present data lend further support to the proposition that the variability is associated with the extent of the modification of lipid composition, of protein components, or both. ACKNOWLEDGMENTS
We wish to thank Dr I. Ben-Basat who provided HS blood samples and to Dr R. A. Cooper for critically reading the manuscript. This work was supported by Grant No. 1248 from the National Council for Research and Development to A. Livne and B. Aloni and by Grant No. 49340 from the Israel National Academy of Sciences to M. Shinitzky. REFERENCES ALONI, B., SHINITZKY, M. & LIVNE, A. (1974) Dynamics of erythrocyte lipids in intact cells, in ghost membranes and in liposomes. Biochimica et Biophysica Acfa, 348, 438. U., SHINITZKY, M., WEEIER, G. & NISHIDA, T. COGAN, (1973) Microviscosity and order in the hydrocarbon region of phospholipid and phopholipid-cholesterol dispersions determined with fluorescent probe. Biochemisfry, 12, 521. COOPER,R.A. & JANDL. J.H. (1969) The role of membrane lipids in the survival of red cells in hereditary spherocytosis. Journal of Clinical Investigation, 48. 736. C. & HANAHAN, D.J. (1963) DODGE, J.T., MITCHELL, The preparation and chemical characteristics of hemoglobin-free ghosts of human erythrocytes. Archives ofBiochemistry and Biophysics, ZOO, 1x9. M.C. & GIER,J. DE,DEENEN,L.L.M. VAN, VERLOOP, GASTEL, C. VAN (1964) Phospholipid and fatty acid characteristics of erythrocytes in some cases of anaemia. British journal of Haernatology, 10, 246. GOMPERTS, E.D., METZ,J. & ZAIL, S.S. (1971) Immunological homogeneity of membrane proteins from hereditary spherocytic, hereditary elliptocytic, and normal red cells. BrifishJournal of Haemutology, 20, 443. HAYASHI, S., KOOMOTO, R.,YANO,A., ISHIGAMI,S., TSUJINO, G., SAEKI, S. & TANAKA,T. (1974) Abnormality in a specific protein of the erythrocyte membrane in hereditary spherocytosis. Biochemical and Biophysical Research Communications, 57, 103 8 . JACOB,H., AMSDEN, T. & WHITE, J. (1972) Membrane microfilaments of erythrocytes: alteration in intact
cells reproduces the hereditary spherocytosis syndrome. Proceedings ofthe National Academy of Sciences ofthe United States ofAmerica, @, 471. JACOB,H.S., RUBY,A., OVERLAND, E.S. & MAZIA,D. (1971) Abnormal membrane protein of red blood cells in hereditary spherocytosis. Journal af Clinical Invesfigation, 50, 1800. J.H. (1968) Hereditary spherocytosis. Hereditary JANDL, Disorders of Erythrocyte Metabolism (Ed. by E. Beutler), p zog. Grune & Stratton, New York. KITAO, T., HATTORI, K., TAKESHITA, M. (1973) Electrophoretic analysis of the major proteins of the human red cell membrane. Clinica Chimica Acfa, 49. 353. KUIPER,P.J.C. & LIVNE,A. (1972) Differences in fatty acid composition between normal human erythrocytes and hereditary spherocytosis affected cells. Biochimica et Biophysicu Acta, 260, 755. KUIPBR, P.J.C., L m , A. & MEYBRSTEIN, N. (1971) Changes in lipid composition and osmotic fragility of erythrocytes of hamster induced by heat exposure. Biochimica ef Biophysica Acta, 248, 300. LACELLE, P.L. & WEED,R.I. (1969) Reversibility of abnormal deformability and permeability of the hereditary spherocyte. (Abstract). Blood, 34, 858. LIVNE,A., ALONI,B., MOSES,S. & KUIPER,P.J.C. (1973) Linolenoyl sorbitol and the fragility ofhereditary spherocytes. British journal of Haematology, 25, 429. LIVNB,A. & RAZ,A. (1971) Erythrocyte fragility and potassium efflux as affected by temperature and hemolyzing rate. FEBS Letters, 16,99. MURPHY, J.R. (1969) Erythrocyte osmotic fragility and
Microviscosity in Erythrocyte Membranes cell water--influence of pH and temperature. Journal of Laboratory and Clinical Medicine, 74, 319. PHILLIPS, G.B. & R o o m , N.S. (1962) Quantitative chromatographic analysis of the phospholipids of abnormal human red blood cells. Proceedings of the Society for Experimental Biology and Medicine, IW, 360.
RAZIN,S,,TOURTELLOTTE, M.E., MCELHANBY, R.N. & POLLACK, J.D. (1966) Mluence of lipid components of Mycoplasma laidlawii membrane on osmotic fragility of cells. Journal OfBaCteriology, 91, 609. RRED,C.F. & SWISHER, S.N.(1966) Erythrocyte lipid loss in hereditary spherocytosis. journal of Clinical Investigation, 45, 777. RUDY,B. & GITLER,C. (1972) Microviscosity of the cell membrane. Biochimica et Biophysica Acta, 288, 231. SHINITZKY,
M. & BARENHOLZ, Y. (1974) Dynamics of hydrocarbon layer in liposomes of lecithin and
123
sphingomyelin containing dicetylphosphate.Journal ofBiological Chemistry, 49. 2652. SHINITZKY, M., DIANOUX.A&., GITLFZ~~ C. & WELIER, G. (1971) Microviscosity and order of the hydrocarbon region of micelles and membrane determined with fluorescent probes. I. Synthetic micelles. Biochemistry, 10,2106. SHINITZKY,M. & INBAR, M. (1974) Difference in microviscosity induced by different cholesterol levels in the surface membrane lipid layer of normal lymphocytes and malignant lymphoma cells. Journal of Molecuhr Biology, 85, 603. SOLOMON, A.K. (1974) Apparent viscosity of human red cell membranes. Biochimica et Biophysica A d a , 373, 145.
TEICHBERG, V.I. & SHINITZKY.M. (1973) Fluorescence polarization studies of lysozyme and lysozymesaccaride complexes.Journal of Molecular Biology, 74, s19.
