American Journal of Hematology 20:337-343 (1985)

Abnormal Erythrocyte Fragmentation and Membrane Deformability in Paroxysmal Nocturnal Hemoglobinuria Brian D. Smith

University of Rochester, School of Medicine and Dentistry, Rochester, New York

Hemolysis in paroxysmal nocturnal hemoglobinuria (PNH) is considered to be a result of an intrinsic membrane defect. This defect may result in abnormal material properties of PNH erythrocytes. To examine this hypothesis, fragmentation failure, and membrane deformability were assessed in the absence of complement by micro- pipette techniques. Membrane viscosity was determined by observing relaxation of deformed cells. Results show a bimodal distribution of force for membrane failure, membrane viscoelasticity, and elastic shear modulus. One population requires signif- icantly less force for fragmentation, mean 0.56 x dyne; has increased mem- brane viscosity, mean 0.205 X dyne seckm; and has decreased elastic shear modulus, mean 0.56 X dyne/cm. A second population resembles control with fragmentation force, mean 1.19 x dyne; membrane viscosity, mean 0.112 X dyne/cm, control 0.102 X lop2 dyne sec/cm; elastic shear modulus, mean 0.70 X dynekm. The percent of cells with abnormal material properties corresponds to the percent of PNH I11 cells determined by complement lysis. Thus, the hemolysis attributed to an abnormal clone of erythrocytes in PNH is associated with an intrinsic membrane abnormality which predisposes to lysis.

dyne, control 1.05 X

dyne/cm, control 0.78 X

Key words: paroxysmal nocturnal hemoglobinuria, erythrocyte membrane deformability

IN T RO D U CT I0 N Paroxysmal nocturnal hemoglobinuria (PNH) is an acquired clonal disorder

having a sensitive population of erythrocytes with increased susceptibility to comple- ment lysis [l-61. Three subpopulations of PNH cells have been described. PNH I11 cells are 15 to 25 times more sensitive to complement lysis than normal red cells. PNH I1 cells are intermediate between PNH III and PNH I erythrocytes with 3 to 5 times greater sensitivity to complement-induced hemolysis, whereas PNH I cells cannot be distinguished from normal red cells [2,5,7,8]. PNH patients have various combinations of these subpopulation cell types. The relationship of in vitro comple- ment sensitivity and decreased red cell survival is not understood. It is known that normal red cell survival depends in part on membrane elasticity [9-121. This charac- teristic elasticity facilitates passage through the microcirculation where red cells must deform to accommodate to the restricted geometry. This study was designed to evaluate the potential role of abnormal membrane viscoelastic properties in reducing PNH erythrocyte survival.

Received for publication February 5, 1985; accepted April 25, 1985.

Address reprint requests to Brian D. Smith, MD, Department of Medicine, Highland Hospital, South Avenue at Bellevue Drive, Rochester, NY 14620.

0 1985 Alan R. Liss, Inc.

338 Smith

To investigate such a potential membrane defect in the abnormal clone of PNH erythrocytes, membrane material properties, force for membrane fragmentation, membrane viscosity, and surface elastic shear modulus were compared for PNH and control erythrocytes. The abnormal cells in the absence of complement were more fragile and less elastic, supporting the hypothesis that the acquired PNH defect alters the material properties and makes these erythrocytes more susceptible to membrane failure. For each of the three patients studied, the proportion of abnormal cells correlated with that of PNH I11 cells as determined by complement lysis techniques.

MATERIALS AND METHODS

After obtaining informed consent, venous blood samples were collected in heparin from three patients in whom the diagnosis of PNH was confirmed by complement lysis. Samples were collected in the same manner from three controls. Cells were washed and resuspended in isotonic phosphate-buffered saline at physio- logic pH with 0.1 % albumin added to minimize echinocyte formation. Final hemato- crit was adjusted to less than 1 % and experiments were performed at 23 k 2°C.

Glass micropipettes with an inside diameter of 0.8-0.9 pm and filled with physiologic buffer were utilized to assess membrane deformability . Elastic behavior of individual cells was observed and recorded on video tape to allow frame-by-frame analysis. A minimum of 100 cells per sample were randomly selected and individually studied.

Based on the complement lysis curves, these three patients had only type I and I11 cells. Thus, type I1 cells, if present, were insufficient in number to be recognized by current methodology. Subtypes of PNH cells cannot be distinguished by light microscopy, thus eliminating observer bias in cell selection.

Membrane properties were assessed by three material constants: force at yield point, sheer modulus of elasticity, and membrane viscosity.

Force at yield point (F). If deformation exceeds the yield point of an elastic material additional deforming force results in irreversible plastic deformation and finally membrane fragmentation [ 131.

By using video recording and playback, individual cells were first evaluated at low-magnitude force to determine elastic shear modulus, then the same cells were stressed by applying progressively larger forces until local membrane failure oc- curred. Shear modulus of elasticity (p ) is the relationship between applied negative pressure (P) through the micropipette with radius (Rp) and corresponding length of erythrocyte membrane aspirated into the pipette (D) [9,14].

Membrane viscosity (17) reflects the time-dependent membrane elastic behavior. To measure this parameter, cells were attached at one point to a glass surface and stretched by applying a negative pressure through a micropipette. Upon release the perpendicular cell major and minor diameters were measured successively as the cell relaxed to the biconcave disc shape. The least squares fit of the data as a function of

Material Properties in PNH Erythrocytes 339

time to determine the time constant (Tc) was computed [ 15-17]. The product of the time constant and shear modulus yields membrane viscosity.

RESULTS

The material properties of a subset of PNH erythrocytes from each patient are abnormal. They are more susceptible to mechanical fragmentation failure and have decreased elasticity. The mean force for fragmentation is shown in Table I with a distribution depicted in Figure 1. In the table, erythrocytes were divided into A or B subpopulations solely on the basis of the force required for fragmention. Cells that fragmented at a force less than the most fragile control cell are subpopulation B, while cells that resemble control are subpopulation A. There is a linear relationship between population B and PNH I11 cells with Pearson correlation of 0.998; however, the confidence limits are broad because of the low number of patients.

Additional parameters of membrane elasticity, including time constant for mem- brane relaxation and elastic shear modulus, were also abnormal for a subset of PNH erythrocytes as shown in Table 11. Time constants for group B cells were greater than the longest time constant reported for a control cell. Group A cells had time constants in the control range. The distribution for controls in each of the three patients is shown in Figure 2. Approximately 40% of each of the PNH patients' erythrocytes had prolonged time constants reflecting decreased membrane elasticity.

The mean elastic shear modulus shown in Table I1 for selected cells (B) that fragmented at subnormal force is less than the elastic shear modulus for cells (A) that required normal range force for fragmentation. The membrane viscosity shown in Table I1 is much greater for the abnormal subset of erythrocytes in each PNH patient.

DISCUSSION

The cell membrane abnormality in paroxysmal nocturnal hemoglobinuria is known to influence the functional expression of cell-bound complement. The struc- tural and biochemical basis of this acquired stem cell defect is not defined but

TABLE I. Fragmentation Force for PNH and Control Erj throcytes' ~~

Fragmentation force dyne

Sample (mean 5 SD)

EC 0.62 f 0.08 0.99 & 0.19

GM 0.57 * 0.13 1.23 f 0.33

AF 0.49 0.18 1.34 0.36

Control 1 1.01 * 0.24 Control 2 1.05 * 0.23

Number of cells

47 53 54 49 52 40

100 100

Subpopulation cell type (%)

B (47) A (53)

A (48) B (57) A (43)

B (52)

Complement lysis %

38 PNH I11 62 PNH I 43 PNH 111 57 PNH I 47 PNH I11 53 PNH I

Control 3 1.10 & 0.21 100

'PNH cells are divided into subpopulations based on force for fragmentation. B cells are more fragile than the most fragile control cell, A cells resemble control. PNH cells divided into subpopulations based on complement lysis, PNH I11 and PNH I results in a distribution similar to B and A cells.

340 Smith

95

5 0

v, _I

40 V

LL 0

t- z W

3 0

g 2 0 W a

10

0

n r

0.0-0.4 0.4-0.8 0.8- 1.2 ) 1.2

FRAGMENTATION FORCE. DYNE

Fig. 1. Distribution of force required for erythrocyte membrane fragmentation. A significant percent of each of the PNH patient’s erythrocytes fragmented at forces less than control. Control: solid; patient EC: open; patient AF: dots; patient GM: crosshatch.

TABLE 11. Elastic Material Properties for PNH and Control Erythrocytes*

Elastic shear Time Constant modulus (p )

Subpopulation (set) lo-’ dyne/cm Viscosity (9 ) Sample cell type (mean f SD) (mean f SD) lo-* dyne sec/cm

EC B 0.35 f 0.17 0.58 f 0.09 0.203 A 0.16 f 0.05 0.73 + 0.18 0.117

GM B 0.43 f 0.10 0.55 f 0.15 0.237 A 0.16 -f 0.03 0.67 f 0.07 0.107

AF B 0.32 * 0.05 0.55 + 0.17 0.176 A 0.16 + 0.02 0.70 0.20 0.112

Control 0.14 f 0.05 0.73 f 0.06 0.102

‘Elastic material properties of PNH cells are divided into subpopulations based on force required for fragmentation and compared to control cells. Subpopulation B cells have longer time constants, reduced elastic shear modulus and increased membrane viscosity all of which reflect abnormal deformability .

phenotypically results in both increased sensitivity to complement lysis and altered erythrocyte membrane material properties. The capacity of the erythrocyte to deform reversibly is thought essential for repeated passage through the microcirculation. The normal membrane is thin and thus behaves as a two-dimensional material that can accommodate large elastic deformations while maintaining essentially constant sur- face area [ 13,18,19], thus permitting reversible change in cell shape. Small alterations in these membrane mechanical properties are predicted to result in marked reduction in red cell survival by reducing capacity for deformation and for withstanding shear forces tending to disrupt the membrane structural integrity.

Material Properties in PNH Erythrocytes 341

25-0 I5 116-025 026-035 ) 0 3 5

TIME CONSTANT (Tc) SECONDS

Fig. 2. Distribution of time constants for stressed erythrocyte membranes to return to normal biconcave disc shape. Approximately 40% of PNH erythrocytes required longer times to relax reflecting altered deformability with decreased elasticity. Control: solid; patient EC: open; patient AF: dots; patient GM: crosshatch.

The capability of measuring viscoelastic properties of red cell membranes permits identification of disorders that alter membrane structural protein. Acquired membrane alterations due to metabolic changes [ 121, erythrocyte senescence [20] or cross-linkage of membrane proteins [ 111 result in significant alterations in biophysical properties of red cells. Inherited membrane defects caused by molecular abnormalities in the erythrocyte membrane skeleton [21] and spectrin deficiency [22] result in abnormal material properties in hereditary spherocytosis [9], while secondary effects on membrane proteins caused by polymerization of sickle hemoglobin in sickle cell anemia also alter membrane mechanical properties [ 101. However, when structural changes do not occur in spite of multiple other red cell changes such as diabetes mellitus [23], membrane deformability is normal [24].

Although there are several reports of PNH erythrocyte membrane protein abnormalities [25-281 as well as abnormal membrane lipids [29], this study demon- strates an intrinsic erythrocyte structural abnormality in PNH that results in reduced force required for membrane disruption and in decreased elasticity. Shear forces encountered in the microcirculation would be sufficient to cause membrane failure in susceptible PNH cells. Furthermore, these cells are less distensible and slower to recover from deformation reflected by decreased elastic shear modulus and increased membrane viscosity decreasing their competence to flow through channels with restricted geometry. For each patient the percent of cells with increased susceptibility to complement lysis parallels the percent with altered material properties, apparently reflecting the same subpopulation, which has a shorter survival. However, deform- ability measurements that were made in the absence of complement indicate that an

342 Smith

abnormality exists irrespective of the potential role of complement. It is unlikely that the change in membrane properties was caused by prior exposure to complement activation, which is not known to influence the relationship of membrane structural proteins. Furthermore, if complement altered deformability by successive membrane fragmentation, abnormal cells might appear spherocytic or show other morphologic change when viewed by light microscopy. Conversely, complement fixation may be facilitated by the structural abnormality. It may be postulated that both altered deformability and complement may be important in reducing erythrocyte survival.

REFERENCES

1. 2.

3.

4.

5 .

6.

7.

8.

9.

10.

11.

12.

13.

14.

15. 16.

17.

18. 19.

20.

21.

Dacie JV: Paroxysmal nocturnal hemoglobinuria. Proc R SOC Med 56587, 1963. Rosse WF, Dacie JV: Immune lysis of normal human and paroxysmal nocturnal hemoglobinuria (PNH) red blood cells. I. The sensitivity of PNH red cells to lysis by complement and specific antibody. J Clin Invest 45:736, 1966. Oni SB, Osunkoya BO, Luzzatto L: Paroxysmal nocturnal hemoglobinuria: Evidence for monoclo- nal origin of abnormal red cells. Blood 36: 145, 1970. Rouault TA, Rosse WF, Bell S, Shelburne J: Differences in the terminal steps of complement lysis of normal and paroxysmal nocturnal hemoglobinuria red cells. Blood 51 :325, 1978. Packman CH, Rosenfeld SI, Jenkins DE, Thiem PA, Leddy JP: Complement lysis of human erythrocytes: differing susceptibility of two types of paroxysmal nocturnal hemoglobinuria cells to C5b-9. J Clin Invest 64:428, 1979. Parker CJ, Baker PJ, Rosse WF: Increased enzymatic activity of the alternative pathway convertase when bound to the erythrocytes of paroxysmal nocturnal hemoglobinuria. J Clin Invest 69:337, 1982. Rose WF: Variations in the red cells in paroxysmal nocturnal hemoglobinuria. Br J Haematol 24:327, 1973. Rosse WF, Adams JP, Thorpe AM: The population of cells in paroxysmal nocturnal hemoglobinuria of intermediate sensitivity to complement lysis: significance and mechanism of increased immune lysis. Br J Haematol28: 181, 1974. Waugh RE, LaCelle PL: Abnormalities in the membrane material properties of hereditary sphero- cytes. J Biomechan Engin 102:240, 1980. Nash GB, Johnson CS, Meiselman HJ: Mechanical properties of oxygenated red blood cells in sickle cell (HbSS) disease. Blood 63:73, 1984. Smith BD, LaCelle PL, Siefring GE, Lowe-Krentz L, Lorand L: Effects of the calcium-mediated enzymated cross-linking of membrane proteins on cellular deformability . J Membrane Biol 61 :75, 1981. Weed RI, LaCelle PL, Merill EW: Metabolic dependence of red cell deformability. J Clin Invest 48:795, 1969. LaCelle PL, Evans EA, Hochmuth RM: Erythrocyte membrane elasticity, fragmentation, and lysis. Blood Cells 3:335, 1977. Evans EA: New membrane concept applied to the analysis of fluid shear and micropipette-deformed red blood cells. Biophys J 13:941, 1973. Evans EA, Hochmuth RM: Membrane viscoplastic flow. Biophys J 16:13, 1976. Hochmuth RM, Buxbaum KL, Evans EA: Temperature dependence of the viscoelastic recovery of red cell membrane. Biophys J 29: 177, 1980. Hochmuth RM, Worthy PR, Evans EA: Red cell extensional recovery and the determination of membrane viscosity. Biophys J 26:101, 1979. Evans EA: A new material concept for the red cell membrane. Biophys J 13:926, 1973. Slalak R, Tozeren A, Zarda RP, Chien S : Strain energy functions of red blood cell membrane. Biophys J 13:245, 1973. Linderkamp 0, Meiselman HJ: Geometric, osmotic, and membrane mechanical properties of density-separated human red cells. Blood 59: 1121, 1982. Goodman SR, Shiffer KA, Casoria LA, Eyster ME: Identification of the molecular defect in the erythrocyte membrane skeleton of some kindreds with hereditary spherocytosis. Blood 60:772, 1982.

Material Properties in PNH Erythrocytes 343

22. Agre PC, Zinkham WH, Casella JF, Bennett V: Spectrin deficiency is common to all forms of hereditary spherocytosis (HS): The degree of deficiency correlates with osmotic fragility. Blood 62:42a, 1983.

23. Jones RL, Peterson CM: Hematologic alterations in diabetes mellitus. Am J Med 70:339-352, 1981. 24. LaCelle PL: Behavior of abnormal erythrocytes in capillaries. In Cokelet GR, Meiselman HJ,

Brooks DE (eds): "Erythrocyte Mechanics and Blood Flow." New York: Alan R. Liss, 1980, pp

25. Righetti PG, Perrella M, Zayella A, Sirchia G: The membrane abnormality of the red cell in paroxysmal nocturnal haemoglobinuria. Nature 245:273, 1973.

26. Dalmass AP, Pizzimenti MC, Vacs E, Diaz A: Abnormal solubilization by Triton X-100 of erythrocyte membranes from patients with paroxysmal nocturnal hemoglobinuria. Proc SOC Exp Biol Med 147:273, 1974.

27. Nicholson-Weller A, March JP, Rosenfeld SI, Austen KF: Affected erythrocytes of patients with paroxysmal nocturnal hemoglobinuria are deficient in the complement regulatory protein, decay accelerating factor. Proc Natl Acad Sci USA 803066, 1983.

28. Pangburn MK, Schreiber RD, Muller-Eberhard HJ: Deficiency of an erythrocyte membrane protein with complement regulatory activity in paroxysmal nocturnal hemoglobinuria. Proc Natl Acad Sci USA 805430, 1983.

29. Tedesco F, Kahane I, Zanella A, Giovanetti AM, Sirchia G: Functional study of lipids of PNH red cell membranes: Susceptibility of liposomes to reactive lysis. Blood 57:900, 1981.

195-209.