British journal o/Haernatolog~,1991,77,209-214
ADONIS
000710489100034s
Electrokinetic properties and surface membrane sialic acid status of platelets in idiopathic thrombocytopenic purpura (ITP) M . C R O O K S, . M A C H I N *A N D N . CRAWFORD? Department of Clinical Chemistry, Guy’s Hospital, London, *Department 01Haematology, University College and Middlesex Hospital Medical School, London, and tDepartment of Biochemistry and Cell Biology, Hunterian Institute, Royal College of Surgeons of England, Lincoln’s Inn Fields. London
Received 12 March 1990; accepted for publication 2 October 1990
Summary. Continuous flow electrophoresis has been used to investigate heterogeneity in the circulating platelets from patients with idiopathic thrombocytopenia (ITP) and from age- and sex-matched control donors. The ITP patients’ platelet counts ranged from 1 5 to 80 x 10”/1 and they had much higher mean volumes and significantly higher (twofold) total cell and neuraminidase-labile surface sialic acid
contents expressed per unit cell than normal. After volume normalization the mean difference between the two groups was not statistically significant. The relationships of these findings to abnormal thrombopoiesis and to molecular recognition for reticuloendothelial system (RE)removal are discussed.
We have previously reported the use of continuous flow electrophoresis (CFE) as an alternative to density gradient separations for the study of the population heterogeneity in the circulating platelet pool (Crook & Crawford, 1988, 1989). Preparative electrophoresis in a free-flowing buffer film is a relatively mild, reproducible and high resolution procedure for the fractionation of cells on the basis of differences in surface membrane electrical charge. It is particularly relevant for studying platelet heterogeneity since their major functional expressions in vivo involve cell/cell and cell/matrix interactions which have surface electrokinetic features. In the aforementioned heterogeneity studies the larger and more electronegative cells in human platelet separation profiles were shown to be significantly richer in total and surface neuraminidase-labile sialic acid than the smaller and less electrophoretically mobile platelets. Moreover, a number of surface membrane-mediated functions were better expressed in the smaller, less electronegative platelets, than in the larger platelets (Crook & Crawford, 1989). The earlier platelet heterogeneity literature is extensive and largely derived from density gradient studies. It is difficult to cross correlate the findings from the many different groups, however, because of the different operational conditions Correspondence: Dr M. Crook, Department of Clinical Chemistry, Guy’s Hospital, Iandon SEl 9RT.
(continuous, discontinuous and self-generating density gradients) and the variety of density gradient media used. In consequence the background is replete with controversy and conflicting interpretations particularly about the origin, life span and functional competence of different subpopulations in the circulation and also their size/buoyant density relationships. For example, whilst some researchers have claimed that senescent-dependent changes to the platelets are the major contributor to any observed heterogeneity (Karpatkin. 1972: Corash et al. 1977: Ginsburg & Aster, 1969), others have proposed with equal conviction that density-derived subpopulation differences are a feature of the megakaryocyte class from which the platelets proliferate after maturation (Penington et al. 1967a. b: Martinet al. 1983). Since the size of the platelet and its intracellular granule content are the two most influential factors in separations by rate sedimentation, the observations that platelet membranes can be modified in vivo by the cell making transient and reversible encounters with vessel walls and by granule release events (George, 1985) adds a further dimension of complexity to the interpretation of such density-derived subpopulations. In our earlier studies of platelet electrophoretic heterogeneity (Crook & Crawford. 1988.1989) we established baseline information (analytical data, electrokinetic properties and membrane functional responses) across the separation
209
210
M. Crook, S. Machin and N . Crawford
profile for normal human platelets. In the present studies we have used exactly the same electrophoretic separation procedure to investigate a platelet disorder associated with disturbed thrombopoiesis. idiopathic thrombocytopenic purpura (ITP). In this disorder, platelet survival times are shortened due to accelerated destruction and megakaryocyte ploidy histograms and platelet production rates show abnormalities. However, the influence of various humoral and nonhumoral factors on the impaired thrombopoiesis have yet to be clearly defined and the clinical and laboratory picture can be quite different in acute and chronic ITP. In the present paper we compare the electrophoretic distribution profile of the platelets from five patients with ITP with platelet samples from age- and sex-matched control subjects. The total platelet sialic acid content and the surface neuraminidase-labile sialic acid was measured in the ITP and control platelets. MATERIALS AND METHODS Five patients with chronic ITP have been studied. Platelet counts lay in the range 15-80 x 10y/l(mean 3 5 f 26) but all were in a stable clinical condition with no recent bleeding events and were not receiving any specific therapy at the time of the investigations. For comparison, each patient's blood sample was accompanied by a sample from an age- and sexmatched normal donor and both were processed and analysed at the same time and under exactly the same conditions. These age- and sex-matched control donors (mean platelet count 3 1 3 f 69) had also been free from any medication for at least 10 d prior to the studies. Informed consent concordant with the Declaration of Helsinki 1964 was obtained for all subjects. Platelet counts and mean platelet volumes were determined on CDP-adenineanticoagulated samples within a few hours of collection using a Coulter STKR automated counter or a Coulter Model ZF with Channelyzer. Latex particles were used for calibration with lower and upper thresholds set at 3 and 36 fl respectively. Reproducibility on different days was assured by reference to a stored sample of fixed platelets of known concentration and size distribution. Platelet isolation jor electrophoreticprofiling. Blood samples were taken into CDP-adenine anticoagulant in the ratio of 6 volumes of blood to one of anticoagulant and processed within 2 h of donation. The platelets were isolated by a red cell layer recycling procedure and washed exactly as described earlier by Crook & Crawford (1988). Recoveries of platelets from the whole blood samples were always greater than 90% for both the patient samples and matched donor control samples. The bulk of the washed platelets were formaldehyde fixed for ease of transit to the electrophoresis apparatus and for short-term storage. Sialic acid measurements were made on unfixed platelets. For fixation, the freshly isolated and washed platelet suspension was mixed with an equal volume of the citrate wash buffer, pH 6.4, containing 0.74%(v/v) of formaldehyde. The mixture was incubated for 1 5 min at 37OC and then allowed to stand at room temperature for 12 h: after which the platelets were spun out of solution, at 750 g for 1 5 min, and then gently resuspended in electrophoretic chamber buffer. Continuousflow electrophoresis. The formaldehyde-treated
platelets suspended in chamber buffer of composition 10 mM triethanolamine, 280 mM glycine (adjusted to pH 7.35 with glacial acetic acid) were applied through a sample syringe to the chamber of a Bender and Hobein VAP5 continuous flow apparatus at a concentration of 1.5-1.8 x lo9platelets/ml. The electrode buffer compartment was filled with a 10-fold concentrated chamber buffer without added glycine. The optimal conditions for separation were found to be 1 1 80 V, 130 mA, temperature of 6-7OC with the chamber buffer flow rate set at 500 ml/h and a rate of 1 . 5 ml/h for the injected platelet suspension. Between each run of the paired samples the electrophoresis chamber was carefully cleaned with a 10% (v/v) Teepol solution followed by at least 2 litres of distilled water. Prior to separation. the chamber was filled with 3% (w/v) bovine serum albumin and held for at least 45 min. The albumin was then flushed out first with distilled water (5-8 chamber volumes) and then chamber buffer for 2-3 chamber volumes. The separated platelet fractions were collected into plastic tubes and the full electrophoretic profile constructed by either counting the platelets in each tube by Coulter Counter, or by measuring the absorbance at 280 nm ofeach tube fraction as described by Crook & Crawford (1988. 1989). Electron microscopy. The platelets from all donors and patients were also examined by scanning electron microscopy. For this the platelets were treated with 2.5% (v/v) isotonic glutaraldehyde in 0.1 M cacodylate buffer, pH 7.3.at room temperature for 1 h. The platelets were washed three times in the buffer without the glutaraldehyde. and dehydrated for 5 min periods in 30. 50, 70, 80 and 100%acetone, in a sequential manner. The samples were critical point dried in a Polaron E 3000 apparatus and then mounted and gold coated for viewing. Sialic acid assay. The procedure was that of Warren ( 19 59) as modified by Skoza & Mohos (1976) and adapted for platelets by Jung et al (1982). For total sialic acid determinations the platelets were hydrolysed at 80°C for 90 min with 0.5 M sulphuric acid. Neuraminidase-labile sialic acid was measured using neuraminidase (affinity purified type X. from Clostridium perjringens. Sigma Chemical Co.. Poole, Dorset). The platelets were incubated with the enzyme at a concentration of 1 U/ml at 3 7OC for 60 min as earlier described (Crook & Crawford. 1988. 1989). As defined by the suppliers one unit liberates 1.0 pmol of N acetyl neuraminic acid (NANA) per min at pH 5.0 using NAN-lactose as substrate. Correction for interfering ribose in the assay was performed as recommended by Warren (1959) by measuring the absorbance at two wavelengths (532 and 549 nm). RESULTS
All the ITP patients showed severe thrombopenia with counts on the day of investigation in the range 15-80 x 10y/l and the counts for the age- and sex-matched control subjects all lay within the normal range (Table I). Scanning electron micrographs showed that the ITP platelets were generally slightly larger than the control platelets consistent with the higher Coulter volumes. A proportion of the platelets from all the ITP patients showed some processes arising from their
Electrokinetic Properties of ZTP Platelets
2 11
Table 1. Platelets from patients with idiopathic thrombocytopenic purpura (ITP) and from age- and sex-matched control subjects
ITP patients Property
1
Age (yr)
2
90
Sex
Matched control subjects
48
F
4
3
40
45
M
F
18
F
20
80
40
20
.Mean platelet volumes (fl) Fresh Formaldehyde fixed
12.9 13.3
10.3 11.2
9.8 10.5
9.6 10.3
15 8.9 9.5
1'
-
35f26
3'
50
4'
48
Mean f S D
5'
43
20
F
F
M
E
M
240
250
316
400
360
10.3f1.5' 1 1 .Of 1.4'
115.8 113.8 114.8 181.8 121.3 129.Sf 29.4
Neuraminidase-labile surface sialic acid (nmol/l0' platelets)
2'
74
-
M
Platelet counts ( x lOY/l)
Total platelet sialic acid (nmol/10' platelets)
Mean fSD
5
-
313f69
7.6 7.8
7.0 7.4
6.3 7.2
6.7 7.4
6.1 7.2
6.7f0.6' 7.4f0.2.
77.1
59.6
66.4
63.6
63.1
65.9f6.7'
40.5
44.3
46.4
79.5
52.2
29.2
24.3
19.8
48.4
31.7
30.7*11.0t
9.0
1 1 .0
1 1.7
18.9
13.6
10.1
8.5
10.5
9.5
10.3
9.8f0.8
3.1
4.3
4.7
8.3
5.9
3.8
3.5
3.1
7.2
5.2
4.6f1.7
Total platelet sialic acid (nmol/pl platelet volume$)
Neuraminidase labile surface sialic acid
(nmol/pl platelet volume$)
5.3f2.0
Significance: * P < 0 . 0 1 . tP
ADONIS
000710489100034s
Electrokinetic properties and surface membrane sialic acid status of platelets in idiopathic thrombocytopenic purpura (ITP) M . C R O O K S, . M A C H I N *A N D N . CRAWFORD? Department of Clinical Chemistry, Guy’s Hospital, London, *Department 01Haematology, University College and Middlesex Hospital Medical School, London, and tDepartment of Biochemistry and Cell Biology, Hunterian Institute, Royal College of Surgeons of England, Lincoln’s Inn Fields. London
Received 12 March 1990; accepted for publication 2 October 1990
Summary. Continuous flow electrophoresis has been used to investigate heterogeneity in the circulating platelets from patients with idiopathic thrombocytopenia (ITP) and from age- and sex-matched control donors. The ITP patients’ platelet counts ranged from 1 5 to 80 x 10”/1 and they had much higher mean volumes and significantly higher (twofold) total cell and neuraminidase-labile surface sialic acid
contents expressed per unit cell than normal. After volume normalization the mean difference between the two groups was not statistically significant. The relationships of these findings to abnormal thrombopoiesis and to molecular recognition for reticuloendothelial system (RE)removal are discussed.
We have previously reported the use of continuous flow electrophoresis (CFE) as an alternative to density gradient separations for the study of the population heterogeneity in the circulating platelet pool (Crook & Crawford, 1988, 1989). Preparative electrophoresis in a free-flowing buffer film is a relatively mild, reproducible and high resolution procedure for the fractionation of cells on the basis of differences in surface membrane electrical charge. It is particularly relevant for studying platelet heterogeneity since their major functional expressions in vivo involve cell/cell and cell/matrix interactions which have surface electrokinetic features. In the aforementioned heterogeneity studies the larger and more electronegative cells in human platelet separation profiles were shown to be significantly richer in total and surface neuraminidase-labile sialic acid than the smaller and less electrophoretically mobile platelets. Moreover, a number of surface membrane-mediated functions were better expressed in the smaller, less electronegative platelets, than in the larger platelets (Crook & Crawford, 1989). The earlier platelet heterogeneity literature is extensive and largely derived from density gradient studies. It is difficult to cross correlate the findings from the many different groups, however, because of the different operational conditions Correspondence: Dr M. Crook, Department of Clinical Chemistry, Guy’s Hospital, Iandon SEl 9RT.
(continuous, discontinuous and self-generating density gradients) and the variety of density gradient media used. In consequence the background is replete with controversy and conflicting interpretations particularly about the origin, life span and functional competence of different subpopulations in the circulation and also their size/buoyant density relationships. For example, whilst some researchers have claimed that senescent-dependent changes to the platelets are the major contributor to any observed heterogeneity (Karpatkin. 1972: Corash et al. 1977: Ginsburg & Aster, 1969), others have proposed with equal conviction that density-derived subpopulation differences are a feature of the megakaryocyte class from which the platelets proliferate after maturation (Penington et al. 1967a. b: Martinet al. 1983). Since the size of the platelet and its intracellular granule content are the two most influential factors in separations by rate sedimentation, the observations that platelet membranes can be modified in vivo by the cell making transient and reversible encounters with vessel walls and by granule release events (George, 1985) adds a further dimension of complexity to the interpretation of such density-derived subpopulations. In our earlier studies of platelet electrophoretic heterogeneity (Crook & Crawford. 1988.1989) we established baseline information (analytical data, electrokinetic properties and membrane functional responses) across the separation
209
210
M. Crook, S. Machin and N . Crawford
profile for normal human platelets. In the present studies we have used exactly the same electrophoretic separation procedure to investigate a platelet disorder associated with disturbed thrombopoiesis. idiopathic thrombocytopenic purpura (ITP). In this disorder, platelet survival times are shortened due to accelerated destruction and megakaryocyte ploidy histograms and platelet production rates show abnormalities. However, the influence of various humoral and nonhumoral factors on the impaired thrombopoiesis have yet to be clearly defined and the clinical and laboratory picture can be quite different in acute and chronic ITP. In the present paper we compare the electrophoretic distribution profile of the platelets from five patients with ITP with platelet samples from age- and sex-matched control subjects. The total platelet sialic acid content and the surface neuraminidase-labile sialic acid was measured in the ITP and control platelets. MATERIALS AND METHODS Five patients with chronic ITP have been studied. Platelet counts lay in the range 15-80 x 10y/l(mean 3 5 f 26) but all were in a stable clinical condition with no recent bleeding events and were not receiving any specific therapy at the time of the investigations. For comparison, each patient's blood sample was accompanied by a sample from an age- and sexmatched normal donor and both were processed and analysed at the same time and under exactly the same conditions. These age- and sex-matched control donors (mean platelet count 3 1 3 f 69) had also been free from any medication for at least 10 d prior to the studies. Informed consent concordant with the Declaration of Helsinki 1964 was obtained for all subjects. Platelet counts and mean platelet volumes were determined on CDP-adenineanticoagulated samples within a few hours of collection using a Coulter STKR automated counter or a Coulter Model ZF with Channelyzer. Latex particles were used for calibration with lower and upper thresholds set at 3 and 36 fl respectively. Reproducibility on different days was assured by reference to a stored sample of fixed platelets of known concentration and size distribution. Platelet isolation jor electrophoreticprofiling. Blood samples were taken into CDP-adenine anticoagulant in the ratio of 6 volumes of blood to one of anticoagulant and processed within 2 h of donation. The platelets were isolated by a red cell layer recycling procedure and washed exactly as described earlier by Crook & Crawford (1988). Recoveries of platelets from the whole blood samples were always greater than 90% for both the patient samples and matched donor control samples. The bulk of the washed platelets were formaldehyde fixed for ease of transit to the electrophoresis apparatus and for short-term storage. Sialic acid measurements were made on unfixed platelets. For fixation, the freshly isolated and washed platelet suspension was mixed with an equal volume of the citrate wash buffer, pH 6.4, containing 0.74%(v/v) of formaldehyde. The mixture was incubated for 1 5 min at 37OC and then allowed to stand at room temperature for 12 h: after which the platelets were spun out of solution, at 750 g for 1 5 min, and then gently resuspended in electrophoretic chamber buffer. Continuousflow electrophoresis. The formaldehyde-treated
platelets suspended in chamber buffer of composition 10 mM triethanolamine, 280 mM glycine (adjusted to pH 7.35 with glacial acetic acid) were applied through a sample syringe to the chamber of a Bender and Hobein VAP5 continuous flow apparatus at a concentration of 1.5-1.8 x lo9platelets/ml. The electrode buffer compartment was filled with a 10-fold concentrated chamber buffer without added glycine. The optimal conditions for separation were found to be 1 1 80 V, 130 mA, temperature of 6-7OC with the chamber buffer flow rate set at 500 ml/h and a rate of 1 . 5 ml/h for the injected platelet suspension. Between each run of the paired samples the electrophoresis chamber was carefully cleaned with a 10% (v/v) Teepol solution followed by at least 2 litres of distilled water. Prior to separation. the chamber was filled with 3% (w/v) bovine serum albumin and held for at least 45 min. The albumin was then flushed out first with distilled water (5-8 chamber volumes) and then chamber buffer for 2-3 chamber volumes. The separated platelet fractions were collected into plastic tubes and the full electrophoretic profile constructed by either counting the platelets in each tube by Coulter Counter, or by measuring the absorbance at 280 nm ofeach tube fraction as described by Crook & Crawford (1988. 1989). Electron microscopy. The platelets from all donors and patients were also examined by scanning electron microscopy. For this the platelets were treated with 2.5% (v/v) isotonic glutaraldehyde in 0.1 M cacodylate buffer, pH 7.3.at room temperature for 1 h. The platelets were washed three times in the buffer without the glutaraldehyde. and dehydrated for 5 min periods in 30. 50, 70, 80 and 100%acetone, in a sequential manner. The samples were critical point dried in a Polaron E 3000 apparatus and then mounted and gold coated for viewing. Sialic acid assay. The procedure was that of Warren ( 19 59) as modified by Skoza & Mohos (1976) and adapted for platelets by Jung et al (1982). For total sialic acid determinations the platelets were hydrolysed at 80°C for 90 min with 0.5 M sulphuric acid. Neuraminidase-labile sialic acid was measured using neuraminidase (affinity purified type X. from Clostridium perjringens. Sigma Chemical Co.. Poole, Dorset). The platelets were incubated with the enzyme at a concentration of 1 U/ml at 3 7OC for 60 min as earlier described (Crook & Crawford. 1988. 1989). As defined by the suppliers one unit liberates 1.0 pmol of N acetyl neuraminic acid (NANA) per min at pH 5.0 using NAN-lactose as substrate. Correction for interfering ribose in the assay was performed as recommended by Warren (1959) by measuring the absorbance at two wavelengths (532 and 549 nm). RESULTS
All the ITP patients showed severe thrombopenia with counts on the day of investigation in the range 15-80 x 10y/l and the counts for the age- and sex-matched control subjects all lay within the normal range (Table I). Scanning electron micrographs showed that the ITP platelets were generally slightly larger than the control platelets consistent with the higher Coulter volumes. A proportion of the platelets from all the ITP patients showed some processes arising from their
Electrokinetic Properties of ZTP Platelets
2 11
Table 1. Platelets from patients with idiopathic thrombocytopenic purpura (ITP) and from age- and sex-matched control subjects
ITP patients Property
1
Age (yr)
2
90
Sex
Matched control subjects
48
F
4
3
40
45
M
F
18
F
20
80
40
20
.Mean platelet volumes (fl) Fresh Formaldehyde fixed
12.9 13.3
10.3 11.2
9.8 10.5
9.6 10.3
15 8.9 9.5
1'
-
35f26
3'
50
4'
48
Mean f S D
5'
43
20
F
F
M
E
M
240
250
316
400
360
10.3f1.5' 1 1 .Of 1.4'
115.8 113.8 114.8 181.8 121.3 129.Sf 29.4
Neuraminidase-labile surface sialic acid (nmol/l0' platelets)
2'
74
-
M
Platelet counts ( x lOY/l)
Total platelet sialic acid (nmol/10' platelets)
Mean fSD
5
-
313f69
7.6 7.8
7.0 7.4
6.3 7.2
6.7 7.4
6.1 7.2
6.7f0.6' 7.4f0.2.
77.1
59.6
66.4
63.6
63.1
65.9f6.7'
40.5
44.3
46.4
79.5
52.2
29.2
24.3
19.8
48.4
31.7
30.7*11.0t
9.0
1 1 .0
1 1.7
18.9
13.6
10.1
8.5
10.5
9.5
10.3
9.8f0.8
3.1
4.3
4.7
8.3
5.9
3.8
3.5
3.1
7.2
5.2
4.6f1.7
Total platelet sialic acid (nmol/pl platelet volume$)
Neuraminidase labile surface sialic acid
(nmol/pl platelet volume$)
5.3f2.0
Significance: * P < 0 . 0 1 . tP
