Download Altered Behaviour of Erythrocytes in Scleroderma

Clinical Science (1983) 6 5 , 515-519
515
Altered behaviour of erythrocytes in scleroderma
I. B . KOVACS*, S. 0. SOWEMIMO-COKERt, J. D. T. KIRBY*
*Departments of Dermatology and
AND
P. T U R N E R t
t Clinical Pharmacology, St Bartholomew’s Hospital, London
(Received 11 January130 March 1983; accepted 1 3 May 1983)
Summary
1. Filterability (deformability) of erythrocytes
of patients with Raynaud‘s phenomenon together
with progressive systemic sclerosis (PSS) was decreased compared with the filtration of erythrocytes from normal subjects.
2. PSS erythrocytes showed lower electrophoretic velocity and about 23% less neuraminidaseremovable sialic acid density on their surface than
the normal erythrocytes.
3. PSS erythrocytes showed more adherence to
cultured endothelial cells than the control normal
erythrocytes.
4. It is concluded that the increased rigidity
and adherence of PSS erythrocytes may have
pathological significance in the mechanism of
vascular abnormalities in PSS.
Key words: adhesion, deformability , erythrocyte,
scleroderma, surface charge.
Abbreviations: PBS, phosphate-buffered saline;
PSS, progressive systemic sclerosis.
of connective tissue components and vascular
involvement is unknown. Hyperviscosity of whole
blood, increase in acute phase reactants, reduced
fibrinolytic activity and minor activation of platelets have been found in a heterogeneous group of
patients who exhibited Raynaud’s phenomenon
[5] . Scleroderma serum was cytotoxic to cultured
endothelial cells [ 6 ] . To our knowledge, no data
have been published on the behaviour of erythrocytes.
Altered haemo-rheological properties is thought
to be causally related to different thrombotic
diseases [7] and abnormalities in erythrocyte
behaviour have been found in diseases with severe
vascular complications, such as sickle-cell anaemia
[8-101 and diabetes mellitus [ l l ] . We have therefore studied those functional characteristics of
erythrocytes which are major determinants of the
rheological properties of the blood, i.e. deformability and endothelial adherence in patients with
PSS and Raynaud’s disease. Our present findings
raise the possibility that changes in behaviour of
erythrocytes might be related to the vascular
complications in scleroderma.
Introduction
Subjects and methods
Scleroderma (progressive systemic sclerosis, PSS)
is a connective tissue disorder with characteristic
excessive collagen deposition. The abnormal metabolism of all components of connective tissue has
been revealed by extensive studies [l-41. Besides
connective tissue involvement, vascular abnormalities, both morphological (intimal proliferation)
and functional (microcirculatory insufficiency),
are distinctive features of the disease [ l ] . The
mechanism of these vascular abnormalities in PSS
or the link between the unbalanced metabolism
General
Correspondence: Dr Iren B. Kovacs, Department of Dermatology, St Bartholomew’s Hospital,
London EClA 7BE.
Blood samples were obtained from patients
with Raynaud’s phenomenon together with PSS,
aged 28-78 years (mean 38). They had symptoms
of the disease for 1-24 years (mean 8). Control
blood samples were obtained from normal healthy
volunteers aged from 21 to 65 years (mean 40).
All PSS patients fulfilled the preliminary diagnostic
criteria set for PSS [12].
Filterability of erythrocytes
Erythrocytes were separated from heparinized
venous blood (Heparin sodium without preserva-
516
I. B. Kovacs et al.
tive, Weddel Pharmaceuticals, London; 20 unitslml
of blood), the buffy coat was carefully removed
and the cells were washed in Dulbecco's phosphatebuffered sodium chloride solution (1 54 mmol/l:
saline), pH 7.4, containing 0.25% (v/v) human
albumin. Packed cell volume was determined from
a concentrated suspension (about 50%) and then
adjusted to 10% (v/v) with phosphate-buffered
saline (PBS)-albumin. This final erythrocyte suspension was passed through a polycarbonate sieve
(Nucleopore Corp., Pleasanton, CA, U.S.A.; 13 mm
diameter) of pore diameter 3 pm. The calculated
cell/pore ratio, per ml of suspension filtered, was
800/ml. A vacuum of -10cm water pressure was
used. After l m l of cell suspension had flowed
through the filter, the filtration time per millilitre
volume was registered up to 10 ml and the slope
was calculated from the linear segment of the
curve plotted. A new filter from the same batch
was used for each measurement and the filters
were selected by measuring the flow rate of PBSalbumin (only filters with a flow rate of 9-9.5 ml/
min were used). Triplicate determination were
performed on each sample of blood and the difference between the triplicate measurements was
usually less than 5%. The assay was performed
within 2 h of venepuncture at room temperature.
Elimination of leucocytes
By repeated measurements the leucocyte count
was around lOO/pl of the washed erythrocyte
suspension. In addition, in one experiment, blood
was centrifuged against Hypaque-Ficoll mixture
[13]. This resulted in an erythrocyte suspension
practically free of leucocytes (less than lO/pl).
Electrophoretic velocity of erythrocytes
Electrophoretic motility of washed erythrocytes
was measured by the moving boundary electrophoresis technique described in detail by Streichman et al. [ 141.
Removal and measurement of sialic acid from
erythrocytes
To 200pl of washed erythrocyte suspension
(15%, v/v) 200 of neuraminidase (0.1 unit/ml;
Clostridium perfringens, type V, Sigma, dissolved
in Dulbecco's PBS, pH 7.4) was added and incubated for 60 min at 37°C. A portion (100 p1) of
the supernatant was used for sialic acid determination.
Sialic acid was determined by a fluorimetric
procedure described by Hess & Rolde [ 151.
Endothelial adherence measurement
Adhesion of erythrocytes to cultured endothelium was measured by the technique described
by Hebbel et al. [lo]. Endothelial cells were
obtained from pig aorta and grown to confluency
by the method of Pearson et al. [16]. A portion
(0.2 ml) of "Cr-labelled erythrocyte suspension
(2576, v/v) layered on endothelium which was
grown on a plastic plate closely fitted into a 6 mm
diameter tissue culture well and was allowed to
undergo static incubation at 37°C for 30 min.
Non-adherent cells were decanted and, after being
removed from the well with a fine forceps, the
plastic plate with the endothelium was rinsed six
times in saline and finally dropped into distilled
water to lyse the remaining cells. The radioactivity
of this final lysate (two plates combined) was
measured. Erythrocyte suspensions from individual
patient were tested in 6-12 wells and simultaneously
normal erythrocytes were also tested under identical conditions. Adherence to endothelium was
expressed as a ratio (adherence ratio) = radioactivity from PSS erythrocytes/radioactivity from
normal erythrocytes.
Statistical treatment of data
The values are given as means f SEM and statistical comparisons were made by Student's paired
t-test, except the filterability of erythrocytes,
where a two-tailed signed-rank (non-parametric)
test was used.
Results
Filterability of PSS erythrocytes was significantly
lower than that of cells from normal, healthy
subjects (Table 1). The difference in the flow rate
between the PSS and control erythrocytes was less
striking, when filters of 5 p pore diameter
instead of 3 pm were used. There was a considerable variation among the patients, but for individual patients the measured filtration rates were
surprisingly constant and reproducible at different
times. The low leucocyte count was a reassurance
that 'clogging' was not a significant factor in the
measurements. Additionally, in one experiment we
used erythrocyte suspension, separated with density
gradient centrifugation practically free of leucocytes. By using this suspension, flow rates similar
to those from the washed cells of normal subjects
were obtained.
Moving boundary electrophoretic velocity of
PSS erythrocytes was significantly lower than that
of normal cells (Table 1). The difference in the
velocity between the PSS and normal erythrocytes
Erythrocytes in s c l e r o d e m
517
.
TABLE 1 Filterability, surface sialic acid density and moving boundary electrophoretic
velocity of erythrocytes of normal subjects and patients with Raynaud's phenomenon
and progressive systemic sclerosis (PSS)
*
Mean values SEM are shown.
Measurement
Subjects
PSS
Control
Filterability (ml/min)
Sialic acid (m0l/lO9 cells)
Electroohoretic mobilitv
(mmimin)
~
No.
Mean
No.
Mean
30
15
15
0.471 i 0.02
42.6 i 1.1
(AH 0.601 0.08
(D)t 0.689 0.10
30
15
15
0.220 i 0.01*
35.6 i 1.2*
0.459 f 0.06*
0.535 f 0.07*
*
*
*Difference from the controls is significant: P < 0.001.
t (A) Ascending, (D) descending boundaries.
was apparoximately the same, when either descending boundaries which move faster, or the ascending
ones, were compared.
In PSS erythrocytes 23% less neuraminidasesusceptible sialic acid density per cell was found,
than in the control cells (Table 1).
Added in Hank's solution (containing albumin)
PSS erythrocytes showed more adherence to pig
confluent endothelial cells than cells from healthy
volunteers (Fig. 1). In two experiments, addition
of a small amount of the patient's own plasma
further increased the adherence of PSS erythrocytes (data are not shown as there were insufficient
for statistical comparison).
relatively large area. The surface properties of the
membranes therefore play a decisive role in determining the transit times in the microcirculation.
Electrostatic repulsion between the negatively
0
0
Discussion
Earlier, in a limited number of patients, we reported
a decreased filterability of erythrocytes in PSS and
the coincidence of the clinical improvement with
the increased filterability of cells in response to
prostaglandin infusions [17]. This paper is an
extension of our preliminary findings and in a
larger number of patients the decreased filterability
of erythrocytes is confirmed.
Because of its simplicity, the filtration technique is widely used to assess flexibility of erythrocytes and has proved to be sensitive enough to
detect changes in behaviour bf cells in a number of
diseases with vascular involvement [ 17,181. However, the filtration technique, even when used as
recommended [19] (using washed cells and minimizing the error of pore obstruction by leucocytes),
is merely an approximation of the more complex
phenomenon, deformability, which occurs when
the cell passes through capillaries [ 2 0 ] .During this
passage, the erythrocyte not only changes its shape
but the adjacent membranes of erythrocytes and
endothelial cell come into close contact over a
0
0
0
0
0
0
FIG. 1. Adhesion of "Cr-labelled erythrocytes
from sclerodermic patients (n = 10) to cultured
endothelial cells. Distribution of adherence ratios
(adhesion ratio =radioactivity of PSS erythrocytes/
radioactivity of normal erythrocytes).
518
I. B. KOVQCS
et al.
charged membranes of erythrocytes and endothelial
cells is an important determinant of the interaction.
Thus decrease of surface charge could induce cell
adhesion and, hence, microvascular insufficiency.
N-Acetylneuraminic (sialic) acid moieties of the
membrane glycoproteins are responsible for the
bulk net negative charges at the cell surface.
Measurement of the enzymatically removable
sialic acid content from PSS erythrocytes showed
significantly less density than the matched control
cells, and this was in accordance with a decreased
electrophoretic mobility of PSS erythrocytes.
Surface sialic acid density of different cells in
the circulation was found t o be a critical determinant of the life-span. As low as 5% decrease of
total surface sialic acid density was an efficient
signal for the specific receptors to remove thrombocytes from the circulation [21]. In subjects with
sialo-glycoprotein deficient erythrocytes, however,
there was no evidence of reduced life-span [22] and
hence it was concluded that changes in surface
charge do not determine the survival of erythrocytes [23]. This agrees with the lack of reports of
altered erythrocyte life-span in PSS. However,
removal of sialic acid from the surface of blood
cells [24,25] or endothelial cells [26] did induce
cell adhesion, a phenomenon of great pathological
significance. PSS erythrocytes with lower sialic
acid and charge densities adhered in much higher
numbers to the cultured endothelial cells than cells
from normal controls. Increased adhesion of erythrocytes t o endothelial cells has been reported in
sickle-cell anaemia [9, 101 and diabetes mellitus
[l I]. In both diseases filterability of erythrocytes
was decreased [7,18], and adherence of cells to
the endothelium was considered to be related to
the vascular complications.
The cause of the increased endothelial adherence
of PSS erythrocytes may be caused by an increased
activity of lysosomal glycosidase in scleroderma
serum [2], an enzyme which can split off the
carbohydrate chains of membrane glycoproteins.
Furthermore, decrease in neuraminidase-removable
sialic acid does not necessarily mean less sialic acid
moieties on the cell surface. It has been pointed
out that adsorbed plasma proteins, mainly yglobu!ins, mask negatively charged sites on the
erythrocyte glycocalyx [27]. Hypergammaglobulinaemia has been a frequent finding in patients
with PSS and increases in y-globulins were reported
to run parallel with the extent of cutaneous and
visceral disease [28]. Being masked by the adsorbed
plasma proteins, those sialic acid moieties lose their
contribution to the surface charge and are not
necessarily accessible to the enzyme. This possibility is supported by our present observation that,
in two cases, addition of a small quantity of the
patient’s own plasma increased the endothelial
adhesiveness of erythrocytes. A similar observation
was made on sickled erythrocytes by Hebbel et QI.
[lo].
Increased rigidity (decreased filterability),
increased endothelial adhesiveness and decreased
surface charge of PSS erythrocytes are related to
the disease and they may be of pathological
importance by providing a possible explanation of
the rheological abnormalities and vascular manifestations of PSS.
Acknowledgments
We express our appreciation to A. R. MacKay,
Department of Experimental Pathology, St
Bartholomew’s Hospital Medical College, London,
for supplying endothelial cells. This study was
supported by a grant from the North East Thames
Regional Research Committee.
References
1 . LeRoy, E.C. (1982) Pathogenesis of scleroderma.
Journal of Investigative Dermatology, 79 (Suppl. l ) ,
87-89.
2. Herrmann, K.,Haustein, U.F., Bohme, H.J. & Lohrisch,
1. (1982) Acid lysosomal hydrolases in systemic
sclerosis and other connective tissue diseases. British
Journal of Dermatology, 106,523528.
3. Uitto, J., Tan, E.M.L. & Ryhanen, L. (1982) Inhibition of collagen accumulation in fibrotic processes:
review of pharmaceutical agents and new approaches
with amino acids and their analogues. Journal of Investigative Dermatology, 79 (Suppl. l ) , 113-120.
4. Rennard, S.I., Stier, L.E. & Crystal, R.G. (1982)
Intra-cellular degradation of newly synthesized collagen. Journal of Investigative Dermatology, 79
(Suppl. l ) , 77-82.
5. Blunt, R.J., George, A.J., Hurlow, R.A., Strachan,
C.J.L. & Stuart, J. (1980) Hyperviscosity and thrombotic changes in idiopathic and secondary Raynaud‘s
syndrome. British Journal of Haematology, 45,651658.
6. Kahaleh, M.B., Sherer, G.K. & LeRoy, E.C. (1979)
Endothelial injury in scleroderma. Journal of Experimental Medicine, 149,1326-1335.
7. Dormandy, J.A. (1980) Haemorrheological aspects
of thrombosis. British Journal of Haematology, 45,
519-522.
8. Rice-Evans, C.A. & Dunn, M.J. (1982) Erythrocyte
deformability and disease. Dends in Biochemical
Science, 7,282-286.
9. Hebbel, R.P., Yamada, O., Moldow, C.F., Jacob, H.S.,
White, J.G. & Eaton, J.W. (1980) Abnormal adherence of sickle erythrocytes to cultured vascular
endothelium. Journal of Clinical Investigation, 65,
154-160.
10. Hebbel, R.P., Moldow, C.F. & Steinberg, M.H. (1981)
Modulation of erythrocyte-endothelial interactions
and the vasocclusive severity of sickling disorders.
Blood, 58,947-952.
11. Wautier, J.L., Paton, R.C., Wautier, M.P., Pintigny, D.,
Abadie, E., Passa, P. & Caen, J.P. (1981) Increased
Erythrocytes in sclerodemza
adhesion of erythrocytes to endothelial cells in
diabetes mellitus and its relation to vascular complications. New England Journal of Medicine, 305,
237-242.
12. Subcommittee for Scleroderma Criteria of the American Rheumatism Association Diagnostic and Therapeutic Criteria Committee (1980) Preliminary criteria
for the classification of systemic sclerosis. Arthritis
and Rheumatism, 23,581-590.
13. Ferrante, A. & Thong, Y.H. (1978) A rapid one-step
procedure for purification of mononuclear and polymorphonuclear leucocytes from human blood using a
modification of the Hypaque-Ficoll technique. Journal
of Immunological Methods, 24,389-393.
14. Streichman, S., Segal, E., Tatarsky, I. & Marmur, A.
(1981) Moving boundary electroporesis and sialic acid
content of normal and polycythaemic red blood cells.
British Journal of Haematology, 48,273-279.
15. Hess, H.H. & Rolde, E. (1964) Fluorometric assay of
sialic acid in brain gangliosides. Journal of Biological
Chemistry, 239,3215-3220.
16. Pearson, J.D., Carleton, J.S., Beesley, J.E., Hutchings,
A. & Gordon, J.L. (1979) Granulocyte adhesion to
andothelium in culture. Journal of Cell Science, 38,
225-235.
17.Dowd, P.M., Kovacs, I.B., Bland, C.J.H. & Kirby,
J.D.T.(1981) Effect of prostaglandins E, and I, on
red cell deformability in patients with Raynaud’s
phenomenon and systemic sclerosis. British Medical
Journal, 283,350.
18. Marcel, G.A. (1979) Red cell deformability: physiological, clinical and pharmacological aspects. Journal
of Medicine, 10,409-416.
19. Working Group Report (1981) Red Cell Deformability
(Methods and Terminology): II European Conference
on Clinical Haemorheology, 29th September, 1981,
London U.K. Russel Wilks, London.
519
20. Nash, G.B. & Meiselman, H.J. (1981) Red cell ageing:
changes in deformability and other possible determinants of in vim survival. Microcirculation, 1, 255284.
21. Greenberg, J., Packham, M.A., Cazenave, J., Riemers,
H.J. & Mustard, J.F. (1975) Effects on platelet function on removal of platelet sialic acid by neuraminidase. Luboratory Investigation, 32,476-484.
22. Anstee, D.J. (1980) Blood group MNSs-active sialogycoproteins of the human erythrocyte membrane.
In: Immunobiology o f the Human Erythrocytes, p.
67. Ed. Sandler, S.G., Nushbacher, J. & Scanfield,
M.S. Alan R. Liss, New York.
23. Seaman, G.V.F., Knox, R.J., Nordt, F.J. & Regan,
D.H. (1977) Red cell ageing. I. Surface charge density
and sialic acid content of density fractionated human
erythrocytes. Blood, 50,1001-1012.
24. Atherton, A. & Born, G.V.R. (1973) Effects of neuraminidase and N-acetylneuraminicacid on the adhesion
of circulating granulocytes and platelets in venules.
Journal of Physiology (London), 234,66P67P.
25. Hoover, R.L., Briggs. R.T. & Karnovsky, M.J. (1978)
The adhesive interactions between PMN leukocytes
and endothelial cells in vitro. Cell, 14,423-428.
26. Gorog, P., Schrafstatter, I. & Born, G.V.R. (1982)
Effect of removing sialic acids from endothelium on
the adherence of circulating platelets in arteries in
vim. Roceedings of the Royal Society of London,
Series B, 214,471-480.
27. Geyers, G., Linss, W. & Stibenz, D. (1977) Adsorbed
proteins mask negatively charged sites of the erythrocyte glycocalyx. Acta Histochemica, 60,312-316.
28. Rodnan, G.P. (1963) The natural history of progressive
systemic sclerosis (diffuse scleroderma). Bulletin of
the Rheumatic Diseases, 13,301-304.