Leukocyte Biophysics An Invited Review GEERT W. SCHMID-SCHONBEIN
Department of AMES-Bioengineering, University of California, San Diego, La JoUa, CA 92093 Received April 2, 1990; Accepted May 9, 1990 ABSTRACT The biophysical properties of leukocytes in the passive and active state are discussed. In the passive unstressed state, leukocytes are spherical with numerous membrane folds. Passive leukocytes exhibit viscoelastic properties, and the stress is carried largely by the cell cytoplasm and the nucleus. The membrane is highly deformable in shearing and bending, but resists area expansion. Membrane tension can usually be neglected but plays a role in cases of large deformation when the membrane becomes unfolded. The constant membrane area constraint is a determinant of phagocytic capacity, spreading of cells, and passage through narrow pores. In the active state, leukocytes undergo large internal cytoplasmic deformation, pseudopod projection, and granule redistribution. Several different measurements for assessment of biophysical properties and the internal cytoplasmic deformation in form of strain and strain rate tensors are presented. The current theoretical models for active cytoplasmic motion in leukocytes are discussed in terms of specific rnacromolecular reactions. Index Entries: Morphology; passive and active rheology; pseudopod; cytoplasmic strains; cytoplasm; viscoelasticity; adhesion; phagocytosis.
INTRODUCTION Leukocytes fulfill not only a protective immunological f u n c t i o n , b u t are also key m e d i a t o r s of cardiovascular disease. A l t h o u g h t h e y h a v e l o n g b e e n the subject of morphological a n d biochemical investigation, their Cell Biophysics
Editor-in-Chief: Leonard Weiss
107
9 1990 The Humana Press Inc.
Schmid-Sch6nbein
108
biophysical properties have been investigated systematically only in the last decade. In addition to a number of classical questions, several new problems have been posed. For example, under what conditions would granulocytes obstruct capillaries or cause reperfusion injury, why would monocytes accumulate preferentially in the subendothelial space of aortic endothelium during artherosclerosis, or why would interleukine activated lymphocytes fail to migrate into a tumor. The answer to these questions requires a detailed picture of their biophysical properties. The important role of leukocytes in microvessels was recognized, and there is increasing evidence for cell activation as a determinant of the leukocyte kinetics in the circulation. A number of new techniques have been introduced that permit studies on macromolecular species in the cytoplasm, how they interact with each other and with the membrane, and the way cytoplasmic deformation leads to a resultant cell motion. Discrepancies exist between leukocyte behavior in vitro and in vivo, that requires the development of mathematical models of biophysical properties so that a full analysis of the experiments can be carried out. The investigation of the biophysical properties of leukocytes is still at an early phase. But it may be useful to provide a summary of the subject in addition to recent reviews that emphasize specific aspects, such as adhesion (1) or cytoplasmic rheology (2-4). The following discussion will be limited to the properties of single leukocytes, mostly those of man. We start with a look at the cell morphology, the cytoplasmic molecular constituents, and then focus on the passive and active cytoplasmic physics.
CELL M O R P H O L O G Y
Volume, Membrane Area, and Limiting Shapes The passive undeformed shape of leukocytes is spherical, with numerous membrane folds on their surface (Fig. 1A, 2A, 2C). Any small area of membrane is relatively inextensible after it has been unfolded, a feature that may be universal among mammalian cell membranes. Therefore, the surface folds serve as a reservoir of membrane area, that allows the ceil to deform passively and actively. The percent excess membrane area, (7, can be expressed as
c=(Sc- Ss)/Ss
(1)
where Ss is the actual plasma membrane area and Sc the area of a smooth sphere with equal cell volume, g Equals about 80-140% in normal circulating leukocytes (5). In the case of neutrophils, basophils, and monocytes, a considerable amount of additional membrane area is available in form
Leukocyte Biophysics
109
Fig. 1. Scanning electron micrograph of (A) passive h u m a n leukocytes, and (B) an active neutrophil. The passive cells are spherical with a folded cytoplasmic membrane. For comparison, a red cell is shown on the right side of (A) with its typical biconcave membrane and smooth membrane. The active neutrophil was in free suspension during fixation. The unfolding of the plasma membrane at the base of the neutrophils is visible. Note that leukocytes prepared for scanning electron microscopy are isotropically shrunken as a result of dehydration.
110
Schrnid.SchSnbein A
=_ ~
c
I)
,.~w,c:,..mmk~.,t '..~ a: ~
1.V_~ Fig. 2. Transmission electron micrographs of (A) passive, (B) active lymphocyte, (C) passive, and (D) active neutrophil. All cells were fixed when in free suspension, and activation occurred spontaneously in autologous plasma. The typical exclusion of granules from the tip of the pseudopod is visible in (D). of the granules (6), an important reservoir during phagocyt0sis. Some average morphological parameters are summarized in Table 1. Earlier observations on blood lymphocytes with the scanning electron microscope have suggested that a subclass (T-cells) may have a smooth membrane area, suggesting no excess surface area. These observations are owing to the fact that such cells were swollen during isolation, and fixed when the membrane was fully unfolded with reduced numbers of surface folds. Since the dehydration for scanning electron microscopy causes isotropic shrinkage of the cells, surface folds that were present before fixation are preserved, whereas cells with smooth surfaces remain smooth during cell preparation (5). Cells without excess membrane area are unable to migrate or phagocytose. The degree to which the leukocytes can be deformed, even without degranulation, is largely determined by the available membrane area. Two limiting cases can be computed by assuming a cylindrical geometry with volume V, surface S, thickness h and radius a, such that a3 + V_
x
a S = 0
2x
(2)
Table 1 Leukocyte M o r p h o l o g y a Erythrocyte Neutrophil
Eosinophil
Lymphocyte
Monocyte
Cell volume O~m3) Cell membrane area (/~m2) Cell excess surface area (%)
90
190
206
120
230
140
300
324
270
430
44
84
92
130
137
Nuclear v o l u m e (#m3) (Percent of cell volume)
-
35 (21%)
50 (44%)
52 (23%)
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
37 (18%) .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Mitochondrial volume (#m3) (Percent of cell volume)
-
Cytoplasmic volume (/~m3) (Percent of cell volume)
-
120 (63%)
Granule volume ~m3) (Percent of cell volume)
-
30 (16%)
Number of granules per cell
-
3700
420
45
1850
-
925
512
5
222
Granule membrane: Area per cell (/~m2) (Percent plasma membrane) Average cell diameter (/~m): Based on cell mass Including surface folds Minimum diameter in a capillary 0~m)b Maximum diameter in a blood smear c
0.6 (0.3%)
-
(308%)
0.7 (0.3%)
.
3.8 (13%)
3.5 (2%)
120 (58%)
62 (52%)
150 (68%)
49 (24%)
0.23 (0.19%)
6.3 (2.7%)
(158%)
(1.9%)
(51%)
7.1 7.8
7.3 8.0
6.2 6.9
7.5 8.3
2.7
2.4
2.6
1.8
2.0
8.1
12.2
12.8
11.8
15.4
8 (max) 2 (min)
aAverage values derived from random electron microscopic micrographs, using stereological techniques (5, 6). bComputed with constant cell volume and surface area, and ignoring cell nucleus and granules. CFrom Ref. (7).
This cubic equation for the radius a has two physical solutions. One solution predicts the elongated cylindrical shape for the m i n i m u m radius (without regard to the nucleus) at constant membrane area, and without water or granule loss, e.g., in a narrow capillary. The other solution predicts the limiting diameter of leukocytes on blood smears after the cells have been deformed to pancakeshapes (7). Average values for both parameters are listed in Table 1. 111
Schmid-SchSnbein
112
The undeformed initial ceil diameter is an important basic quantity in microvascular studies. If the ceil diameter is computed based on the cell mass, its value is typically 0.5 to 0.7 tzm less than if the diameter is measured from the tip of the surface folds. The amount of plasma trapped between the surface folds makes a considerable contribution to the apparent cell diameter, a consideration that is significant during automated cell sizing. If the cells are swollen by adjustment of the plasma osmolality, membrane area is preserved so that the ceil, the granules, mitochondria, and the nucleus assume the limiting shape of a sphere. Membrane area extension is possible only over a narrow range (about 2-%). Thereafter, the membrane ruptures. This feature is closely associated with the tight molecular packing of the membrane phospholipids and the inability to reduce the thickness of the double layer membrane. This behavior is in contrast to the membrane of a balloon that can expand to several times its initial area by reduction of its thickness while preserving membrane mass and volume. Membrane
Limitation During P h a g o c y t o s i s
Another manifestation of the limitation imposed by the available membrane area occurs during phagocytosis. If a microorganism with volume VM and surface area SM is internalized by a cell, its plasma membrane is reduced by SMwhen the cell volume increases by VM. The overall cell surface-volume ratio is reduced, so that internalization cannot proceed indefinitely. On the other hand, if a granule is discharged across the plasma membrane during phagocytosis, the cell loses a small volume whereas the plasma membrane area is enlarged by the granule membrane area. The limiting number of microorganisms,n, that can be internalized into a freely suspended leokocyte during these two processes, a number also denoted as the phagocytic capacity, is given by the polynomial n 3 b3 + n 2 b2 + n bl + bo = 0
(3)
with the coefficients bo = 36g (Vo - AVG)2 - (So+ ASG)3 bl = 72rc (Vo - AVG)VM + 3 (So+ ASG)2 SM b2 = 36rc VM2 - 3 (S O + ASG) SM2 b3 = SM3 where AVc is the total granule volume released and ASc the total additional cell membrane area from the granules. Vo and So are the initial leukocyte volume and surface, respectively, before phagocytosis and degranulation. So = Ss in Eq. (1).
Leukocyte Biophysics
113
13 LU
U ._.,I
Department of AMES-Bioengineering, University of California, San Diego, La JoUa, CA 92093 Received April 2, 1990; Accepted May 9, 1990 ABSTRACT The biophysical properties of leukocytes in the passive and active state are discussed. In the passive unstressed state, leukocytes are spherical with numerous membrane folds. Passive leukocytes exhibit viscoelastic properties, and the stress is carried largely by the cell cytoplasm and the nucleus. The membrane is highly deformable in shearing and bending, but resists area expansion. Membrane tension can usually be neglected but plays a role in cases of large deformation when the membrane becomes unfolded. The constant membrane area constraint is a determinant of phagocytic capacity, spreading of cells, and passage through narrow pores. In the active state, leukocytes undergo large internal cytoplasmic deformation, pseudopod projection, and granule redistribution. Several different measurements for assessment of biophysical properties and the internal cytoplasmic deformation in form of strain and strain rate tensors are presented. The current theoretical models for active cytoplasmic motion in leukocytes are discussed in terms of specific rnacromolecular reactions. Index Entries: Morphology; passive and active rheology; pseudopod; cytoplasmic strains; cytoplasm; viscoelasticity; adhesion; phagocytosis.
INTRODUCTION Leukocytes fulfill not only a protective immunological f u n c t i o n , b u t are also key m e d i a t o r s of cardiovascular disease. A l t h o u g h t h e y h a v e l o n g b e e n the subject of morphological a n d biochemical investigation, their Cell Biophysics
Editor-in-Chief: Leonard Weiss
107
9 1990 The Humana Press Inc.
Schmid-Sch6nbein
108
biophysical properties have been investigated systematically only in the last decade. In addition to a number of classical questions, several new problems have been posed. For example, under what conditions would granulocytes obstruct capillaries or cause reperfusion injury, why would monocytes accumulate preferentially in the subendothelial space of aortic endothelium during artherosclerosis, or why would interleukine activated lymphocytes fail to migrate into a tumor. The answer to these questions requires a detailed picture of their biophysical properties. The important role of leukocytes in microvessels was recognized, and there is increasing evidence for cell activation as a determinant of the leukocyte kinetics in the circulation. A number of new techniques have been introduced that permit studies on macromolecular species in the cytoplasm, how they interact with each other and with the membrane, and the way cytoplasmic deformation leads to a resultant cell motion. Discrepancies exist between leukocyte behavior in vitro and in vivo, that requires the development of mathematical models of biophysical properties so that a full analysis of the experiments can be carried out. The investigation of the biophysical properties of leukocytes is still at an early phase. But it may be useful to provide a summary of the subject in addition to recent reviews that emphasize specific aspects, such as adhesion (1) or cytoplasmic rheology (2-4). The following discussion will be limited to the properties of single leukocytes, mostly those of man. We start with a look at the cell morphology, the cytoplasmic molecular constituents, and then focus on the passive and active cytoplasmic physics.
CELL M O R P H O L O G Y
Volume, Membrane Area, and Limiting Shapes The passive undeformed shape of leukocytes is spherical, with numerous membrane folds on their surface (Fig. 1A, 2A, 2C). Any small area of membrane is relatively inextensible after it has been unfolded, a feature that may be universal among mammalian cell membranes. Therefore, the surface folds serve as a reservoir of membrane area, that allows the ceil to deform passively and actively. The percent excess membrane area, (7, can be expressed as
c=(Sc- Ss)/Ss
(1)
where Ss is the actual plasma membrane area and Sc the area of a smooth sphere with equal cell volume, g Equals about 80-140% in normal circulating leukocytes (5). In the case of neutrophils, basophils, and monocytes, a considerable amount of additional membrane area is available in form
Leukocyte Biophysics
109
Fig. 1. Scanning electron micrograph of (A) passive h u m a n leukocytes, and (B) an active neutrophil. The passive cells are spherical with a folded cytoplasmic membrane. For comparison, a red cell is shown on the right side of (A) with its typical biconcave membrane and smooth membrane. The active neutrophil was in free suspension during fixation. The unfolding of the plasma membrane at the base of the neutrophils is visible. Note that leukocytes prepared for scanning electron microscopy are isotropically shrunken as a result of dehydration.
110
Schrnid.SchSnbein A
=_ ~
c
I)
,.~w,c:,..mmk~.,t '..~ a: ~
1.V_~ Fig. 2. Transmission electron micrographs of (A) passive, (B) active lymphocyte, (C) passive, and (D) active neutrophil. All cells were fixed when in free suspension, and activation occurred spontaneously in autologous plasma. The typical exclusion of granules from the tip of the pseudopod is visible in (D). of the granules (6), an important reservoir during phagocyt0sis. Some average morphological parameters are summarized in Table 1. Earlier observations on blood lymphocytes with the scanning electron microscope have suggested that a subclass (T-cells) may have a smooth membrane area, suggesting no excess surface area. These observations are owing to the fact that such cells were swollen during isolation, and fixed when the membrane was fully unfolded with reduced numbers of surface folds. Since the dehydration for scanning electron microscopy causes isotropic shrinkage of the cells, surface folds that were present before fixation are preserved, whereas cells with smooth surfaces remain smooth during cell preparation (5). Cells without excess membrane area are unable to migrate or phagocytose. The degree to which the leukocytes can be deformed, even without degranulation, is largely determined by the available membrane area. Two limiting cases can be computed by assuming a cylindrical geometry with volume V, surface S, thickness h and radius a, such that a3 + V_
x
a S = 0
2x
(2)
Table 1 Leukocyte M o r p h o l o g y a Erythrocyte Neutrophil
Eosinophil
Lymphocyte
Monocyte
Cell volume O~m3) Cell membrane area (/~m2) Cell excess surface area (%)
90
190
206
120
230
140
300
324
270
430
44
84
92
130
137
Nuclear v o l u m e (#m3) (Percent of cell volume)
-
35 (21%)
50 (44%)
52 (23%)
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
37 (18%) .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Mitochondrial volume (#m3) (Percent of cell volume)
-
Cytoplasmic volume (/~m3) (Percent of cell volume)
-
120 (63%)
Granule volume ~m3) (Percent of cell volume)
-
30 (16%)
Number of granules per cell
-
3700
420
45
1850
-
925
512
5
222
Granule membrane: Area per cell (/~m2) (Percent plasma membrane) Average cell diameter (/~m): Based on cell mass Including surface folds Minimum diameter in a capillary 0~m)b Maximum diameter in a blood smear c
0.6 (0.3%)
-
(308%)
0.7 (0.3%)
.
3.8 (13%)
3.5 (2%)
120 (58%)
62 (52%)
150 (68%)
49 (24%)
0.23 (0.19%)
6.3 (2.7%)
(158%)
(1.9%)
(51%)
7.1 7.8
7.3 8.0
6.2 6.9
7.5 8.3
2.7
2.4
2.6
1.8
2.0
8.1
12.2
12.8
11.8
15.4
8 (max) 2 (min)
aAverage values derived from random electron microscopic micrographs, using stereological techniques (5, 6). bComputed with constant cell volume and surface area, and ignoring cell nucleus and granules. CFrom Ref. (7).
This cubic equation for the radius a has two physical solutions. One solution predicts the elongated cylindrical shape for the m i n i m u m radius (without regard to the nucleus) at constant membrane area, and without water or granule loss, e.g., in a narrow capillary. The other solution predicts the limiting diameter of leukocytes on blood smears after the cells have been deformed to pancakeshapes (7). Average values for both parameters are listed in Table 1. 111
Schmid-SchSnbein
112
The undeformed initial ceil diameter is an important basic quantity in microvascular studies. If the ceil diameter is computed based on the cell mass, its value is typically 0.5 to 0.7 tzm less than if the diameter is measured from the tip of the surface folds. The amount of plasma trapped between the surface folds makes a considerable contribution to the apparent cell diameter, a consideration that is significant during automated cell sizing. If the cells are swollen by adjustment of the plasma osmolality, membrane area is preserved so that the ceil, the granules, mitochondria, and the nucleus assume the limiting shape of a sphere. Membrane area extension is possible only over a narrow range (about 2-%). Thereafter, the membrane ruptures. This feature is closely associated with the tight molecular packing of the membrane phospholipids and the inability to reduce the thickness of the double layer membrane. This behavior is in contrast to the membrane of a balloon that can expand to several times its initial area by reduction of its thickness while preserving membrane mass and volume. Membrane
Limitation During P h a g o c y t o s i s
Another manifestation of the limitation imposed by the available membrane area occurs during phagocytosis. If a microorganism with volume VM and surface area SM is internalized by a cell, its plasma membrane is reduced by SMwhen the cell volume increases by VM. The overall cell surface-volume ratio is reduced, so that internalization cannot proceed indefinitely. On the other hand, if a granule is discharged across the plasma membrane during phagocytosis, the cell loses a small volume whereas the plasma membrane area is enlarged by the granule membrane area. The limiting number of microorganisms,n, that can be internalized into a freely suspended leokocyte during these two processes, a number also denoted as the phagocytic capacity, is given by the polynomial n 3 b3 + n 2 b2 + n bl + bo = 0
(3)
with the coefficients bo = 36g (Vo - AVG)2 - (So+ ASG)3 bl = 72rc (Vo - AVG)VM + 3 (So+ ASG)2 SM b2 = 36rc VM2 - 3 (S O + ASG) SM2 b3 = SM3 where AVc is the total granule volume released and ASc the total additional cell membrane area from the granules. Vo and So are the initial leukocyte volume and surface, respectively, before phagocytosis and degranulation. So = Ss in Eq. (1).
Leukocyte Biophysics
113
13 LU
U ._.,I
