Mechanisms of Ageing and Development, 59 ( 1991) 253--262

253

Elsevier ScientificPublishers Ireland Ltd.

AGE-RELATED STUDIES OF Slg, LEU-4 AND CONCANAVALIN A RECEPTOR DENSITIES AND CAPPING IN HUMAN LYMPHOCYTES

HARVEY J. COHEN, KATHY M. BOLAND and K. MURALI KRISHNA RAO Geriatric Research, Education and Clinical Center, VA Medical Center and Center for the Study of Aging and Human Development, Duke University Medical Center, Durham. NC 27705 (U.S.A.)

(Received November 9th, 1990) SUMMARY We compared the cell surface antigen density and capping of three antigens in lymphocytes obtained from healthy, young (mean age 27 years) and elderly (mean age 76), population. There were no differences in the expression of surface immunoglobulin (SIg), concanavalin A (con A) receptors and Leu-4 antigen between the two groups. Kinetic analysis of these molecules revealed a slight decrease in capping in the older population, but the differences were not statistically significant. In order to test the possibility that subjecting the cells to metabolic stress might bring out the differences, we performed a kinetic analysis of SIg and con A capping in the presence of various concentrations of the metabolic inhibitor sodium azide. Although the capping in cells from elderly subjects was slightly more sensitive to azide, no statistical difference was found. Analysis of con A capping by a flow cytometric method yielded similar results, confirming the data obtained by visual capping experiments. We conclude that although a trend toward decreased capping was observed, there is little alteration in the surface molecule capping phenomenon in the age-groups studied.

K ey words: Kinetics of capping; Metabolic inhibitors; Flow cytometry

INTRODUCTION Investigations into the distribution of cell surface receptors have resulted in the description of the 'capping phenomenon', which involves the temperature- and energy-dependent redistribution of cell surface molecules [1,2]. The capping phenomenon has been explained on the basis of 'fluid mosaic' concept of surface memCorrespondence to: HarveyJ. Cohen, Box 182, V.A. Medical Center, Durham, NC 27705, U.S.A.

0047-6374/91/$03.50 Printed and Published in Ireland

© 1991 ElsevierScientificPublishers Ireland Ltd.

254 brane structure which may apply to many normal cell types and surface markers [3,4]. This hypothesis suggests that protein moieties of the cell membrane float freely within the lipid components of the membrane and are free to move randomly in lateral motion about the cell surface. Attempts to characterize the factors which contribute to an age-dependent decline in immune response have largely focused on changes in lymphocytes [5]. Multiple studies have shown functional decrements in these cells over time, including decreased proliferative capacity of individual cells to mitogens, and decreased production of important mediators such as interleukin-2 [6,7]. In general, there appears to be a defective ability of senescent lymphocytes to respond to appropriate stimuli. It has been suggested that age-dependent changes in lymphocyte function are due, in part, to structural alterations in the membrane [8]. For instance, qualitative analysis of lymphocyte membrane antigens have shown a decrease in the density of some membrane receptors with age [9]. Other studies have suggested that intramembrane movement of the receptors is altered with increasing age [10--13]. The alterations in receptor movement may be related to changes in the content of membrane cholesterol and phospholipid in lymphocytes [14]. Ultimately, both of these findings could bear upon the membrane fluidity of senescent lymphocytes and this in turn could have a negative impact upon critical cell functions such as receptor exposure or membrane-associated activity [15,16]. In this study we compared the cell surface antigen density and capping of three different surface antigens - - surface membrane immunoglobulin (SIg), concanavalin A (con A) receptors and Leu-4 antigen - - in lymphocytes obtained from healthy young and elderly population. Further, the kinetics of capping was assessed in the presence of various concentrations of a metabolic inhibitor, sodium azide. In addition to the usual visual method for assessing the capping we have used a flow cytometric method [17,18], which allows quantitation of capping in a large number of cells, to confirm the observations made by the visual inspection of the caps. MATERIALS AND METHODS

Donor population Elderly volunteers were recruited from Human Subject Registry of the Center for the Study of Aging and Human development, Duke University. The age range varied from 69 to 86 with a mean age of 76. Young donors were recruited from Duke University community; the age range was 21--35 with a mean age of 27. Information regarding disease conditions, medications and smoking habits was obtained. All old donors were community-dwelling, active persons who could drive to the laboratory for donating the blood samples. Donors were drawn in paired sets of young/old for most of the studies to minimize differences due to day-to-day variation on the experimental conditions. All blood samples were drawn in A.M. with in one hour time period between the young and the old pair.

255

Lymphocyte isolation Fifty milliliters of heparinized venous blood was obtained and the mononuclear cells were separated by Ficoll/Hypaque density gradient technique [19]. The mononuclear cells were washed three times in RPMI 1640 medium containing 10% heat inactivated fetal calf serum and 100 units/ml penicillin/streptomycin (GIBCO). The mononuclear cells were greater than 90% viable as assessed by trypan blue exclusion test.

Materials Fluorescein-labeled anti-human polyvalent (total) immunoglobulins and F(ab)2 antibodies were obtained from Kallestad (Chaska, MN), FITC-conjugated concanavalin A from Miles Scientific (Naperville, IL) and FITC-anti-Leu-4 was obtained from Becton and Dickinson (Mountainview, CA). All reagents used in the capping studies were dialyzed x 2 in phosphate buffered saline (pH 7.2) to reduce the concentration o f the sodium azide used as preservative.

Surface antigen density assessment One million cells in 50/~1 of RPMI medium were incubated with the appropriate reagent for 1 h in ice, washed three times in cold RPMI to remove the unbound reagent, then fixed with 1 ml of 1% paraformaldehyde in PBS. The cells were stored at 4"C, in the dark, until analysis by flow cytometry. Surface antigen density was measured on an Ortho Cytofluorograph model 50-H, equipped with a Lexel (model 95---04) argon laser; and the histograms were generated on a 2140 computer. A concentration of 200 #g/ml (final) o f con A was found optimal for determination of the antigen density. FITC-Ig was used at a final concentration of 1:6 based on previous capping studies in our laboratory [20]. FITC-Leu-4 was used at a final concentration of 1:20. PMT settings and flow rates were kept standardized throughout the study. Lymphocytes were gated on the basis of forward vs. right angle scatter. The fluorescence intensity was expressed as mean channel fluorescence on a 1000 channel scale. FITC-goat-anitmouse-IgG was used to assess non-specific binding.

Surface antigen redistribution (Capping) Two million cells in 0.5 ml RPMI were incubated with the appropriate reagent for 1 h on ice, then washed three times with cold RPMI. The cells were then incubated in a 37"C water bath and aliquots were removed at 0, 2, 5 and 15 min and the antigen redistribution was measured live on a Leitz Ortholux 2 fluorescent microscope equipped with HB50 mercury lamp, Ploem vertical epi-illumination and FITC excitor filters. At least 100 cells were counted for each time point and cells exhibiting less than one third surface fluorescence were considered 'capped'.

Sodium azide inhibition studies The cells were pre-incubated for 30 min at 37"C with various concentrations of sodium azide, washed once and suspended in a small amount of RPMI and in-

256 cubated with 50 #1 of either F I T C - i g or FITC-con A and the cells were processed as indicated above in the capping studies. The concentration of the azide was maintained at the indicated level throughout the incubation and washing steps.

Flow ~Ttometric determination of capping Capping was also assessed by flow cytometry as described previously [ 17,18]. The detection is based on the principle that the width of the fluorescent signal is narrower when the fluorescent ligand is capped and wider in uncapped cells. The narrowing of the width of the signal is taken as a measure of capping process. An Ortho cytofluorograph 50-H, equipped with 2140 computer and Lexel argon laser (model 95-4) was used for analysis. Statistical analysis Comparisons between the young and the old were done by paired t-test. Estimation for differences following azide treatment was done using ANOVA. RESULTS

Surface antigen density Table I summarizes the density of three different surface antigens, SIg, con A receptors and Leu-4 antigen. No significant difference was detectable between the 25 young and 30 old individuals studied. Surface antigen redistribution (capping) Capping studies were conducted with reagents directed against three surface antigens, namely, Slg, con A receptors, and Leu-4 antigen (Table II). Capping was assessed at three different time points 2, 5 and 15 min. In the case of SIg capping, the percent capped ceils were slightly lower at all three time points in the older donors, compared to the young donors. In the case of con A receptors and Leu-4 antigen the values were virtually the same at the 2-min time point, but slightly lower at 5 and 15 min. However, in none of the cases were the differences statistically different (P > 0.05). TABLE 1 DENSITY OF DIFFERENT SURFACE ANTIGENS ON LYMPHOCYTES FROM YOUNG AND OLD SUBJECTS

Slg Con A Leu-4

Young (n = 25)

Old (n = 30)

448 4- 22a 480 ± 20 301 4- 28

420 4- 14 505 4- 18 250 ± 26

aThe values represent mean channel fluorescence 4-

S.E.M.

257 TABLE 11 KINETICS OF L1GAND-INDUCED CAPPING IN LYMPHOCYTES FROM YOUNG AND OLD DONORS

Ligand

n

% Capped cells Time (ram) 2

Young

conA Slg Leu-4 con A Slg Leu-4

Old

25 25 14 30 30 14

15 45 23 !7 42 23,

5 ± ± ± 4± 4-

2a 5 7 I 4 4

23 57 20 19 54 13

15 ± ± ± 4± 4-

2 3 3 1 3 1

28±2 70± 2 8±1 24± 1 64±5 64-2

~The values represent mean ± S.E.M.

T h e r e w a s a g r a d u a l i n c r e a s e in t h e n u m b e r o f c a p s f o r m e d o v e r t i m e , w i t h c o n A a n d Slg; in c o n t r a s t t h e n u m b e r o f c a p s f o r m e d d e c r e a s e d w i t h t i m e in t h e c a s e o f Leu-4. This indicates that this antigen m a y be shed f r o m the surface rapidly following cap formation.

Effect o f sodium azide on surface antigen redistribution (Table III) The effect of the metabolic inhibitor sodium azide on surface antigen redistribut i o n w a s s t u d i e d a t 4 d i f f e r e n t c o n c e n t r a t i o n s v a r y i n g f r o m 100 # M t o 10 m M , a n d a t t w o t i m e p o i n t s , 2 a n d 5 m i n . A t all c o n c e n t r a t i o n s a n d a t b o t h t i m e p o i n t s ( e x c e p t w i t h c o n A a t 2 m i n w i t h 10 m M c o n c e n t r a t i o n ) , w i t h all t h r e e s u r f a c e a n t i g e n s t h e

TABLE I11 LIGAND-INDUCED CAPPING IN LYMPHOCYTES FROM YOUNG AND OLD DONORS IN THE PRESENCE OF NA AZIDE

Time (rain)

% Inhibition of capping Sodium a:ide concentration (raM) I0

5

1

0.1

1. Con A Caps Young

2 5 2 5

I1 12 I1 13

81 50 80 65

4± ± ±

Young

2

Old

5 2 5

12 14 14 12

72 53 77 61

Old

5a 6 4 6

55:1:9 27 4- 8 65 ± 8 33 4- 8

28 27 38 34

± 4± ±

I1 9 10 8

± 6

48 4- 9

42 4-

8

4- 5 ± 5 4- 8

32 4- 5 54 4- 9 38 ± 6

15 ± 3 45 4- 6 24 -4- 5

22 19 42 23

± 44±

8 9 10 7

2. SIg Caps

aThe values represent mean ± S.E.M,

17 +

5

8 4- 4 22 + 6 20 ± 5

258

T A B L E IV CON A CAPPING STUDIES IN YOUNG AND OLD USING FLOW CYTOMETRIC T I O N 1N T H E P R E S E N C E O F S O D I U M A Z I D E

Sodium azide (mM)

DETERMINA-

Width of the Fluorescent Signal Time (min) 0

2

5

Young (n = 10) 0

64 4- 7 a

49 4- 8

42 + 8

0.1 1.0 5.0

66 :~ 7 69 4- 6 69 4- 7

55 4. 8 58 4. 8 64 4. 8

47 4. 9 46 4. 7 49 4- 7

0 0.1 1.0

66 -~ 8 67 4- 7 70 4- 7

54 + 9 57 ± 9 61 + 8

45 ~ 7 48 + 8 48 + 9

5.0

70 4- 6

65 4. 6

53 4- 8

Old (n = 10)

a T h e values r e p r e s e n t m e a n ± S . E . M .

percent inhibition of capping was more in the lymphocytes obtained from the older individuals compared to those from the younger individuals. However, again the differences were small and statistically not significant.

Evaluation of capping by flow cytometry In ten separate experiments con A-induced capping was evaluated using the flow cytometric technique (Table IV). The data are presented as width of the fluorescent signal; the decrease in fluorescence width represents capping. In cells treated with the ligand only the mean signal widths were 64, 49 and 42 at 0, 2 and 5 min respectively, in lymphocytes from young individuals; the corresponding numbers for lymphocytes from old donors were 66, 54 and 45. The mean signal width at 5 min increased from 42 to 49 as the concentration of the sodium azide was increased from 0 to 5 mM in the lymphocytes obtained from the young donors. The mean channel width at 5 min with 5 mM sodium azide was 53 in the older donors compared to the value of 49 for the young donors - - the higher channel width representing greater inhibition of capping. Once again the differences were small, but they confirm the trend seen with visual counting of the caps. DISCUSSION

The age-related decrease in the immune response may reflect a basic change in the lymphocyte, among the several other components of the immune system. An early event in the immune response is the recognition of the various ligands involved in the immune response by the cell surface molecules. It has been suggested that the

259 redistribution of the cell surface receptors observed after ligand-receptor interactions at the cell surface of lymphocytes might play a role in the initiation of the immune response [21,22]. In murine models the density of certain cell surface receptors in lymphocytes has been shown to decrease with age [9]. However, in human ieukocytes, the density of the beta-adrenergic receptors does not show any age-related changes, although the coupling of the receptors to adenylate cyclase is impaired [23]. Our observations indicate that there is no age-related decrease in the density of the three surface molecules we studied, when cells are in their basal state. Others have reported a decrease in the expression of cell surface receptors following activation, in humans [24]. It is certainly possible that under conditions in which cells are subjected to greater stress, as during activation process, defects not observed at rest would become evident. Capping studies in relation to age have also yielded conflicting results. In murine studies an age-related decline in capping was described in three surface molecules including Thy-1. Naiem and Walford [12] reported a decrease in con A- and antihuman immunoglobulin-induced capping in lymphocytes obtained from older individuals. However, in another study no age-related differences were found in con A-induced caps, though differences could be elicited following treatment with zinc [25]. In this report we studied the kinetics of capping of three different cell surface molecules o n lymphocytes and failed to detect any age-related differences. Though cells from older people tended to have lower capping values, the range was broad and differences were not statistically significant. In order to test the hypothesis that an abnormality in capping, even if not present in cells in their basal condition might become evident under conditions of stress we studied cells subjected to various concentrations of sodium azide. Capping is known to be dependent upon glycolysis and inhibited by azide [20]. We reasoned that while high concentrations of azide would inhibit capping in cells from both young and old subjects, perhaps cells from older subjects would have an increased susceptibility which would become evident at lower concentrations of the inhibitor. In fact there was a consistent trend towards greater inhibition of capping at lower concentrations of azide in older subjects, but due to the considerable variability these differences were not statistically significant. Though the trend would be consistent with our previous findings that the enzymes of the glycolytic pathway are altered with age resulting in decreased metabolic flux [26], the resulting effects are subtle at best, under the conditions studied. The differences in data reported from various laboratories regarding age-related changes may be due to a number of factors. First is the issue of the health status of the older population under study. It is well recognized that older people have a higher incidence of multiple illnesses. Such disorders could influence the immune status and introduce an additional variable to the aging status. In fact a number of diseases have been associated with defects in lymphocyte capping. These include

260 chronic lymphocytic leukemia [20], systemic lupus erythematosus [27], Duchenne Muscular Dystrophy [28], and Down's Syndrome [12,29]. Other chronic diseases have not specifically been studied. Thus, if some studies had included people with more chronic illnesses it is possible that some of the alterations seen could have been related to those diseases. Moreover in the studies showing altered capping some but not all individuals exhibited the alterations [11]. Furthermore, some recent studies employing the Senieur Protocol [30] to systematically select only very healthy older individuals for study have shown only minimal alterations in measures of immune function [10,31]. Though we did not employ the Senieur protocol we did try to exclude people with chronic and/or acute illnesses that might have confounded the results, and thus may have had a healthier group than in some of the previous reports. Previous studies have shown subjects from this particular population to have normal lymphocyte membrane protein organization at rest, by electron spin resonance analysis [32]. On the other hand the average age of our older group was 76 whereas in Naeim's study [12] the subjects were 80--98 and in Whisler's study [11] the average was 88 years. Thus it is possible that with more advanced age the subtle defects noted would have become more prominent. Another factor contributing to the variability of the results may involve technical factors. Capping is a time- and temperature-dependent phenomenon and capping experiments in different studies have been done at different temperatures and incubation conditions. We have done our studies at early time-points. Longer incubation times may lead to internalization of ligands, or shedding, further confounding the capping data. None of the methods can be considered 'standard' at the present time, but these methodological differences have to be considered in interpreting the data. We conclude then that for at least the healthy 'old-old' of average age of 75 years, there is only a slight change in the ability of their lymphocytes to redistribute their surface membrane antigens (Slg, con A and Leu-4). This is accentuated slightly by inhibitors of glycolysis suggesting that other stresses on cellular function might uncover subtle defects in this aspect of cellular function. Moreover other studies suggest that in the 'oldest-old' such defects may become more obvious. This points out the need to stratify patient populations under study by both age and health status. ACKNOWLEDGMENTS This work was supported by funds from Veterans Affairs Medical Research Program and Grant 2 P30 AG00371-11 and Cancer Center Core Grant 2-P30-CA-14326-17 from NIH. The flow cytometric analysis was performed at the Duke Comprehensive Cancer facility. REFERENCES R.B. Taylor, P.H. Duffus, M.C. Raffand S. de Petris, Redistributionand lainocytosisof lymphocyte surface immunoglobulin molecules induced by anti-immunoglobulin antibody. Nature New Bh~l.. 233 (1971) 225--229.

261 2 3

4 5 6 7

8

9 10

I1

12

13 14 15 16

17 18 19 20 21

22 23

E.R. Unanue and M. J: Karnovsky, Redistribution and fate of Ig complexes on surface of B lymphocytes: functional implications and mechanisms. Transplant. Rev.. 14 (1973) 184~210. S.J. Singer and G.L. Nicolson, The fluid mosaic model of the structure of cell membranes. Cell membranes are viewed as two-dimensional solutions of oriented globular proteins and lipids. Science. 175 (1972) 720---731. L.D. Frye and M. Edidin, The rapid intermixing of cell surface antigens after formation of mousehuman heterokaryons. J. Cell Sci., 7 (1970) 319--335. J.E. Nagel, R.H. Yanagihara and W.H. Alder, Cells of the immune response. In V.J. Cristafalo (ed.), CRC Handbook of Cell Biology o f Aging, CRC Press, Boca Raton, 1985, 341--363. H.T. Cheung, J-S. Twu and A. Richardson, Mechanism of the age-related decline in lymphocyte proliferation: role of IL-2 production and protein synthesis. Exp. GerontoL, 18 (1983) 4 5 1 4 6 0 . J.M. Hefton, G.J. Darlington, B.A. Casazza and M.E. Weksler, Immunological studies of aging. V. Impaired proliferation of PHA responsive human lymphocytes in culture. J. lmmunoL, 125 (1985) 1007--1010. F. Naeim and R.L. Walford, Aging and cell membrane complexes: the lipid layer, integral proteins, and cytoskeleton. In C.E. Finch and E.L. Schneider (eds.), Handbook of the Biology of Aging, Van Nostrand Reinhold, New York, 1985, 272--289. B.A. Woda, J. Yguerabide and J.D. Feldman, Mobility and density of AgB, "ia', and Fc receptors on the surface of lymphocytes from young and old rats. J. lmmunol.. 123 (1979) 2161 2167. K.N. Traill, D. Schonitzer, G. Jurgens, G. Bock, R. Pfeilschifter, M. Hilchenbach, A. Holasek, O. Forster and G. Wick, Age-related changes in lymphocyte subset proportions, surface differentiation antigen denisty and plasma membrane fluidity:application of the eurage SENIEUR protocol admission criteria. Mech. Ageing Dev., 33 (1985) 39--66. R.L. Whisler and Y.G. Newhouse, Functions of T cells from elderly humans: reduction of membrane events and proliferative responses mediated via T3 determinants and diminished elaboration of soluble T-cell factors for B-cell growth. Cell. lmmunol.. 99 (1986) 4 2 2 4 3 3 . F. Naeim and R.L. Walford, Disturbance of redistribution of surface membrane receptors on peripheral mononuclear cells of patients with Down's syndrome and of aged individuals. J. Gerontol., 35 (1980) 650~655. A.B.C. Noronha, J.P. Antel, R.P. Roos and B.G.W. Arnason, Changes in concanavalin A capping of human lymphocytes with age. Mech. Ageing Dev., 12 (1980) 331--337. B. Rivnay, S. Bergman, M. Shinitzky and A. Globerson, Correlations between membrane viscosity. serum cholesterol, lymphocyte activation and aging in man. Mech. Ageing Dev.. 12 (1980) 119 126. H. Borochov and M. Shinitzky, Vertical displacement of membrane proteins mediated by changes in microviscosity. Proc. Natl. Acad. Sci. U.S.A., 73 (1989) 4526~4530. R.N. Farias, B. Bloj, R.D. Morero, F. Sineriz, and R.E. Trucco, Regulation ofallosteric membrane bound enzymes through changes in membrane lipid composition. Bioehim. Biophys. Aeta, 415 (1975) 231--251. H.J. Cohen, K. Boland, F. Cornew and A. Balber, Flow cytometry analysis and sorting of human B lymphocytes during slg capping. Blood, 5 (suppl. 1) (1987) 99a. M.A. Cuchens and T.M. Buttke, A kinetic study of membrane immunoglobulin capping by flow cytometry. Cytometry, 5 (1984) 601 609. A. B6yum, Separation of leukocytes from blood and bone marrow. Seand. J. Clin. Lab. Invest., 21 (Suppl. 97) (1968) 77--89. H.J. Cohen, Human lymphocyte surface immunoglobulin capping: Normal characteristics and anomalous behaviour of chronic lymphocytic leukemia cells. J. Clin. Invest., 55 (1975) 84~93. M.C. Raft and S. de Petris, Movement of lymphocyte surface antigens and receptors: the fluid nature of the lymphocyte plasma membrane and its immunological significance. Fed. Proc., 32 (1973) 48--54. K.M.K. Rao, Capping and mitogenesis: a model implicating microfilaments in lymphocyte activation. J. Theoret. Biol., 98 (1982) 61--71. R.D. Feldman, L.E. Limbird, J. Nadeau, D. Robertson and A.J.J. Wood, Alterations in leukocyte 0-receptor affinity with aging. A potential explanation for altered /3-adrenergic sensitivity in the elderly. New Engl. J. Med., 310 (1984) 815--819.

262

24 25 26 27

28 29

30

31 32

A. Amadori, P. Zanovello, E. Cozzi, V. Ciminale, F. Borghesan, U. Fagiolo and G. Crepaldi, Study of some early immunological parameters in aging humans, Gerontology, 34 (1988) 277--283. K.M.K. Rao, Age-related differential effects of zinc on concanavalin A-induced capping of human lymphocytes. Exp. Gerontol., 17 (1982) 205--211. T.O. Tollefsbol and H.J. Cohen, Carbohydrate metabolism in transforming lymphocytes from the aged. J. Cell. Physiol., 123 (1985) 4 1 7 4 2 4 . G.M. Kammer and E. Mitchell, Impaired mobility of human T lymphocyte surface molecules during inactive systemic lupus erythematosis: relationship to a defective cAMP pathway. Arthritis Rheum., 31 (1988) 88--98. B.M. Goldsmith and H-D. Gruemer, Systemic membrane defect and inhibition of lymphocyte capping in Duchenne Muscular Dystrophy. Clinica Chemiea Aeta, 164 (1987) 3 3 4 6 . M. Chiricolo, L. Minelli, F. Licastro, P. Tabacchi, M. Zannotti and C. Franceschi, Alterations of the capping phenomenon on lymphocytes from aged and Down's syndrome subjects. Gerontology, 30 (1984) 145--152. G.J. Ligthart, J.X. Corberand, C. Fournier, P. Galanaud, W. Hijmans, B. Kennes, H.K. MullerHermelink and G.G. Steinmann, Admission criteria for immunogerontological studies in man: The Senieur Protocol. Mech. Ageing Dev., 28 (1984) 47--55. J.X. Corberand, P. Laharrague and G. Filliola, Neutrophils of healthy aged humans are normal. Mech. Ageing Dev., 36 (1986) 57 63. J.A. Jurivich, G.M. Rosen and H.J. Cohen, Membrane protein organization of peripheral blood lymphocytes from healthy young and aged adults. Mech. Ageing Dev.. 45 (1988) 65--74.