Evaluation of natural killer cell activity in patients with persistent generalized lymphadenopathy and acquired immunodeficiency syndrome

CLINICAL

IMMUNOLOGY

AND

IMMUNOPATHOLOGY

36, 141- 150 ( 1985)

Evaluation of Natural Killer Cell Activity in Patients with Persistent Generalized Lymphadenopathy and Acquired lmmunodeficiency Syndrome’ PAULA C. Vancouver

CREEMERS,

DONALD

F. STARK,

AND WILLIAM

J. BOYKO’

Lymphadenopathy-AIDS Study Group, Depnrtments qf Laboratories, Fumiiy and Medicine, St. Parcl’s Hospital and the University of British Colrtmbia, Vancouver. British Columbia V6J I Y6. Canuda

Practice,

Natural killer (NK) cell activity was quantitated using 5’Cr release from the human erythroleukemia cell line KS62 in 39 heterosexual males, 60 asymptomatic homosexuals, 39 patients with persistent generalized lymphadenopathy (PGL), and 16 patients with acquired immunodeficiency syndrome (AIDS). PGL and AIDS patients showed a slight decrease in NK cell activity compared to control groups. Absolute numbers of Leu 1 lapositive cells were decreased in PGL and AIDS patients, and this decrease correlated with a decrease in absolute number of both the T4* and T8+ cell subsets. Autologous plasma inhibited NK cell activity in 48% of asymptomatic homosexuals, 63% of PGL patients, and 63% of AIDS patients, but in none of the heterosexual controls. NK cell responses in fetal calf serum, normal human plasma, or autologous plasma showed no correlation with absolute numbers of T4+ cells, or with T4/T8 ratio. We conclude that NK cell responses are not of prime importance in the pathogenesis of PGL and AIDS. kl 1985 Academic Press. Inc.

INTRODUCTION

Natural cytotoxicity, generally believed to be an important host defense mechanism against neoplastic and virus-infected cells (1) is mediated by nonadherent, nonphagocytic white cells which have been designated large granular lymphocytes (LGL) (2). It is known that natural killer (NK) cells are derived from bone marrow precursors (3) and their development is thymus independent (4). They have, however, several surface markers in common with T cells (5, 6) and have therefore been assigned to the T-cell lineage. T-Cell clones derived by micromanipulation have recently been shown to display both specific cytolytic activity and lysis of the commonly used NK-sensitive KS62 target cell (7). Two monoclonal antibodies have been developed against LGL: the Leu 7 antibody recognizes both NK cells and cells that display antibody-dependent cellular cytotoxicity (8). The Leu 1 la antibody recognizes essentially all functional NK cells and granulocytes (9). Double labeling experiments with Leu Ila and Leu 7 showed that the Leu 1 la-positive population displayed heterogeneous activity: Leu 1 la+, 7- cells were by far more efficient mediators of NK cell activity than Leu 1 1+ , 7+ cells (10). ’ This work is supported by B. C. Health Care Research Foundation. 2 To whom reprint requests should be addressed at: Department of Pathology, St. Paul’s Hospital. 1081 Burrard Street, Vancouver, B.C. V6J lY6, Canada. 141 0890-1229185 $1.50 Copyright 0 1985 by Academic Press. Inc. All rights of reproduction in any form reserved

142

CKEEMERS,

STARK,

AND BOYKO

The acquired immunodeficiency syndrome (AIDS) and prodrome persistent generalized lymphadenopathy (PGL) in homosexual males is associated with the retrovirus HTLV-III (11) which expresses tropism for T4 ’ cells ( 12). Immunologically, the disease is characterized by lymphopenia, decreased T4 ’ cells. rcversed T4/T8 ratios, skin test anergy, depressed T-cell mitogenic responses. and hypergammaglobulinemia (13). To date there are conflicting data regarding the abnormalities in NK cell activity in AIDS patients. Significantly (14, 15) to marginally (16) decreased NK cell activity in AIDS and PGL have been reported in some studies, whereas others have found NK cell activity to be normal (17). The numbers of subjects tested in these studies were small. An apparently contradictory finding is an increased number of lymphocytes bearing the Leu 7 marker in these patients (18, 19). We present the first large systematic study of natural cytotoxicity in AIDS: NK cytolytic activity and number of Leu I la-positive lymphocytes have been determined in heterosexual males, in asymptomatic homosexual men, and in PGL and AIDS patients. In addition, we studied the influence of plasma factors on NK cytolytic activity, which has not been previously reported. We found that in a number of study subjects NK cell activity is inhibited by autologous plasma. The role of NK cells in the pathogenesis of AIDS is discussed. MATERIAL AND METHODS Subjects. Subjects were participants in an epidemiologic investigation which included asymptomatic homosexual men, as well as PGL and AIDS patients (20). Age ranges and medians were similar among these groups. PGL was defined as lymphadenopathy of greater than 1 cm at two or more extrainguinal sites of at least 3 months’ duration. AIDS was diagnosed if culture- or biopsy-proven opportunistic infections, Kaposi’s sarcoma, or lymphoma were present. Both diagnoses are in accordance with the standards of the Center of Disease Control, Atlanta, Georgia (21). On each occasion we tested homosexual controls and PGL and AIDS patients simultaneously. Normal asymptomatic heterosexual males consisted of hospital personnel between 20 and 50 years of age. Lymphocyte separation. Peripheral blood was drawn in preservative-free heparinized syringes and lymphocytes were isolated by Ficoll-Paque (Pharmacia, Uppsala, Sweden) density gradient centrifugation. Cells were washed two times in phosphate-buffered saline, and resuspended in RPM1 1640 medium (GIBCO Laboratories, Grand Island, N.Y.) supplemented with 20 mM Hepes (4-(2-hydroxyethyl-1-piperazine ethanesulfonic acid) (GIBCO), and 2.5 x 10-j M 2-mercaptoethanol (Sigma, St. Louis, MO.) (complete medium). Lymphocytes were counted in a hemacytometer by diluting the cell preparation 1:20 in 2% acetic acid containing 300 ~l/lOO ml Sedi-Stain (Clay Adams, Beckton-Dickinson, Parsippany, N.J.); monocytes and granulocytes were excluded from the lymphocyte count. NK cell activity assay. Suspensions of the K562 cell line were cultured in complete medium supplemented with 1% L-glutamine (GIBCO) and 10% heatinactivated fetal calf serum (FCS) (GIBCO). Target cells were incubated for 60 min with 150 pCi Na, 51Cr0, (Frosst Laboratories, Montreal. Quebec) in 1 ml

NK

CELLS

IN ACQUIRED

IMMUNODEFICIENCY

SYNDROME

143

medium at 37°C in 5% CO,. After three washings the cells were resuspended in complete medium containing 10% FCS and incubated for another 60 min at 37°C in 5% CO,. After washing, the NK cell assay was performed in complete medium with 10% FCS and 1% glutamine. Ceils (1 x lo4 in 100 ~1 medium) were plated in wells of microtiter plates (Nunc, Roskilde, Denmark) and a varying number of effector cells in 100 ~1 medium were added. After 16 hr of incubation at 37°C and 5% CO,. 100 p,l of supernatant was collected and counted in a gamma counter (1270 rackgamma II, LKB, Turku, Finland). Maximum release of “Cr was determined by adding 100 )LI 2 M HCl to target cells, which were subsequently freeze-thawed three times. Spontaneous “‘Cr release was determined by incubation of target cells in culture medium only. Specific lysis was calculated by Eq. [l]: experimental release spontaneous release % specific lysis = maximum release spontaneous release

x 100.

[II

The tests were performed in quadruplicate; background ‘tCr release varied from 15 to 25%. To determine inhibition by plasma factors, cells (effector-totarget cell ratio of 5O:l) were incubated with 10% heat inactivated autologous plasma instead of FCS. Pooled normal human plasma (NHP) derived from asymptomatic heterosexual males served as control. Percentage plasma inhibition (%pi) was calculated by Eq. [2]: cpm in NHP %pi = 100 PI cpm in autologous plasma ’ Statistical analysis. When two sets of data were compared, significance of differences was determined by the Students t test. When more than two sets of data were compared, significance of differences was determined by one-way analysis of variance with Scheffe multiple comparisons. The relationship between variables was investigated using Spearman rank correlations (22). Experimental conditions. Incubation times of 4 and 16 hr are commonly used in NK cell assays. Less variation among normal subjects is found after a 16-hr incubation period (23). We tested 12 heterosexual men using 4- and 16-hr incubation periods and the results, at effector-to-target cell ratios of lOO:l, 50:1, and 25:1, were proportionate. Since the cytolytic response after a 16-hr incubation period was increased by approximately ‘/I to l/3, we choose this incubation period. The best discrimination between asymptomatic heterosexuals (9) and study subjects (23) was obtained at a 5O:l effector-to-target cell ratio. To assess the interassay variation, including the variation in antigenic expression of the target cells, the K562 cell line was cultured in two separate flasks for at least 1 week prior to testing. The two target cell populations were then labeled separately. Blood from 20 donors was divided and lymphocytes were separately isolated and counted. Each sample was assayed on either target cell population. The standard deviation (SD) between the two sets was 6.63%. Antigenic expres-

144

CREEMERS,

STARK,

AND

BOYKO

sion of the target cells did not appear to change since the cytolytic activity of lymphocytes from asymptomatic homosexuals on the different assay dates showed no significant difference. Membrane immunofuoresc~ence. Ficoll-Paque-purified cells were stained with flourescein isothiocyanate (FIT0conjugated monoclonal antibodies OKT,. OKT, (Ortho Diagnostic Systems, Raritan, N.J.) and Leu lla (Beckton-Dickinson Monoclonal Center, Inc. Mountain View, Calif.), as described previously (24). The number of fluorescent cells was determined on the FACS IV (BecktonDickinson) fluorescence-activated cell sorter. The absolute number of fluorescent lymphocytes was calculated from total white blood cell count determined on a Coulter Counter S plus V (Coulter Electronics, Inc. Hialeah, Fla.) and a 200-cell manual differential blood count. immunoglobulins and immune complexes. Serum concentrations of immunoglobulin classes were determined by nephelometry (Beckman Instruments, Inc., Clinical Instruments Division, Brea, Calif.). Soluble immune complex levels were established by the ‘zSI-labeled Clq assay (25). Lymphocyte transformation. Concanavalin A (Con A)- and phytohemagglutinin (PHA)-induced lymphocyte transformation was performed as described (26). Percentage plasma inhibition (% pi) was calculated by Eq. [3]: %pi=

loO-

cpm in autologous plasma cpm in NHP ’

[31

RESULTS NK cell activity. Significant differences in T4+ cells are found only between the AIDS group and other populations (P < 0.001) while the T4/T8 ratio of all groups differed significantly from each other (P < 0.050) (Fig. 1). The NK ceil responses of these populations are depicted on Fig. 2. There is no significant difference in reactivity between asymptomatic homosexuals (Z = 66.2, SD = 20.3) and heterosexuals (Z = 65.3, SD = 17.7). The asymptomatic homosexual population differs significantly from PGL patients (Z = 52.1, SD = 19.4, P < 0.005) with a difference in average target cell kill of 14%. PGL patients do not differ in cytolytic activity from AIDS patients (Z = 54.3, SD = 23.0). There was no clear relation between NK cell activity and a decrease in T4+ cells and T4/ T8 ratio (Y = 0.3815, P < 0.005 and Y = 0.3498, NS, respectively). Leu Zla-positive cells. Since the NK cell assay was performed with unseparated lymphocytes, it is possible that a decrease in activity is masked by a higher percentage of NK effector lymphocytes. We therefore determined the percentage and absolute number of Leu lla-positive lymphocytes in some subjects of each test group. The results, summarized in Table 1, show no significant increase or decrease in percentage of Leu 1la-positive cells. Percentage of Leu 1 la-positive cells was not correlated with NK cell activity (Y = 0.244, NS). To establish the role of NK cells in the pathogenesis of AIDS, the absolute number of NK cells in blood is of interest. The number of Leu 1 la-positive cells per milliliter of blood was significantly decreased in AIDS compared to the number in PGL patients (P

NK CELLS IN ACQUIRED

IMMUNODEFICIENCY

SYNDROME

145

1800

1400

m 5 " ; a 5 I3 at

1000

600

200

3.20 2.80

2 '0. z i 2.00 L ': :;

1.60

-0 :

@- 1.20 40 .40

A

B

C

D

FIG. 1. Absolute number of T helper (T4’) cells and ratio of T helper/suppressor (T4/%3) cells of the study populations. (A) Heterosexual controls; (B) asymptomatic homosexuals; (C) FGL patients; and (D) AIDS patients. Horizontal bars indicate the mean.

146

CREEMERS,

STARK,

100 I * : .I ... 1 u 80b *..*. -... 6 ...-*. z 60I; ----~ :+ ;” .: 1 i,. a 40 sp .. . I 2ot

AND BOYKO

:

:.: ... :...:‘$’ ye-.::.. :.‘”

. .. . .-. A .*

;;

it

. . & :.

:* .

.* .. :.

. .

. L--

~-

~~-

--.-

A

8

C

0

FIG. 2. NK cell activity in the study populations. (A) Heterosexual males; (B) asymptomatic homosexuals; (C) PGL patients: and (D) AIDS patients. Horizontal bars indicate the mean.

< O.OOS), but the difference between PGL patients and asymptomatic homosexual controls was not significant. Regression analysis suggests a correlation between number of Leu 1 la-positive cells and the number of T4+ and TS+ cells and T41 T8 ratio (values were r = 0.7792, P < 0.001; r = 0.6539, P < 0.001; and r = 0.4498, P < 0.001, respectively). This indicates that the decreased number of the Leu 1la-positive subpopulation decreases with other subpopulations. Znjluence of autologous plasma. In vivo NK cell activity may also be influenced by serum factors. To investigate this, cells from study subjects were tested in the presence of NHP or autologous plasma. The results, summarized in Fig. 3, show a decrease in reactivity in autologous plasma compared to NHP in asymptomatic homosexuals. (: = 55.5, SD = 23.5 vs X = 69.4, SD = 20.0; P < 0.025) and PGL (5 = 34.5, SD = 21.9 vs X 50.1, SD = 20.3, P < 0.025). The difference was not significant in AIDS patients (2 = 40.0, SD = 34.0 vs X = 54.0, SD = 30.7) or in heterosexual controls (Y = 67.2, SD = 15.9 vs X = 67.7, SD = 14.6). Activities of NK cells in NHP and FCS were identical (tested on 10 occasions). Cytotoxic activity in autologous plasma differed significantly between heterosexual males and PGL patients (P < 0.001) and between asymptomatic homosexuals and PGL patients (P < 0.005). The difference between the cytotoxic responses in autologous plasma of heterosexual and homosexual controls was not significant. NK cell activity measured in autologous plasma did not correlate with T4+ cells (r = 0.1298, NS) or T4/T8 ratio (r = 0.2578, NS). Only some subjects in each group showed a significant (P < 0.001) decrease of NK cell activity in TABLE THE~NCIDENCEOF

Leu

lla-POSITIVE Leu

1

LYMPHOCYTES Ila'

cells

(o/o of total lymphocytes) Subjects Asymptomatic homosexuals PGL AIDS

IN PGLAND

AIDS PATIENTS Leu I la+ lymphocytes (per ml blood)

N

X 2 SD

Range

.? + SD

Range

9 14 II

27.2 2 8.5 23.2 -+ 8.2 25.2 t 1.2

18-47 IO-40 13-31

691 r 356 534 2 311 257 2 146

322- 1054 135-1160 70-496

NK CELLS

IN ACQUIRED

IMMUNODEFICIENCY

SYNDROME

.

0 L--_ A

a

C

-

147

00 0 ~~~ D

FIG. 3. Influence of autologous plasma on NK cell activity of study populations. Filled symbols, NK cell activity in pooled normal human plasma; open symbols, NK cell activity in autologous plasma. (A) Heterosexual males; (B) asymptomatic homosexuals; (C) PGL patients; and (D) AIDS patients. Horizontal bars indicate the mean.

autologous plasma compared to in NHP: 13 out of 27 asymptomatic homosexuals, 15 out of 24 PGL patients, and 5 out of 8 AIDS patients showed a significant decrease. Percentage inhibition by autologous plasma in the individual assays in each study group is shown in Fig. 4. Inhibition in both asymptomatic homosexuals and PGL patients differed significantly from heterosexual controls (P < 0.001 for both). Percentage inhibition tends to increase in PGL and AIDS but differences were not significant. There was no correlation between percentage serum inhibition and T4+ cells (Y = 0.1222, NS) or T4/T8 ratio’s (r = 0.0450, NS). Plasma inhibition and other parameters of immunity. In 17 homosexuals the inhibition by autologous plasma was measured in both the NK cell assay, and in PHA- and Con A-induced blastogenesis. Inhibition by plasma in Con A- and PHAinduced blastogenesis correlated well (r = 0.8940, P < 0.001); correlation was moderate between percentage inhibition in the NK cell assay and in Con A- and PHA-induced blastogenesis (r = 0.5760 and r = 0.5717, respectively; P < 0.005 for both). Inhibition of cellular immune reactions at the effector cell level by immune complexes or excess antigen is a common phenomenon in virally induced diseases. In such cases, the response in normal serum or plasma can be increased by extensive washing of the effector cells (27). However, in three PGL patients with reduced activity in autologous plasma NK cell activity was not increased after six additional washes. Furthermore, no correlation was found between percentage inhibition and Clq values or concentrations of total Ig or Ig subclasses. DISCUSSION

We confirm that NK cell activity in normal human subjects varies widely (23, 28). Although some authors report a decrease in NK cell reactivity in PGL and AIDS patients (14, 15), our findings indicate that this is present in only a minority of study subjects. Monocytes and granulocytes may comprise up to 50% of the total Ficoll-Paque-separated white cell preparation in AIDS and PGL patients, versus 7- 12% in control subjects. We excluded monocytes and granulocytes from the lymphocyte count; failing to do so will cause falsely low results in NK cell activity.

148

CREEMERS,

60

. % -y ir A

STARK,

*.. .* .. ?zL is.& .. ..

a

AND BOYKO

.. . . . : 2. .

.

i. : . . :

C

D

FIG. 4. Percentage plasma inhibition in individual assays for the study groups. (A) Heterosexual males; (B) asymptomatic homosexuals: (C) PGL patients; and (D) AIDS patients. Horizontal bars indicate the mean.

We found a small but significant difference in NK cell activity between asymptomatic homosexuals and PGL patients. Average target cell kill was similar in PGL and AIDS patients, and no correlation was found between NK cell activity and T4+ cells or T4/T8 ratio. Some decrease in activity can be explained by the fact that NK cell function is known to be regulated by interleukin 2 (IL-2) (29) via the production of immune (gamma) interferon, which is decreased in AIDS (30, 31). Also, a greater percentage of monocytes in the cell preparation might account for increased production of prostaglandin E2 (PGE,) and subsequent inhibition of NK cell activity (32). Our findings indicate that the ability of NK cells to mediate killing is not directly affected. NK cell activity was not directly related to percentage of Leu 1la-positive cells in the lymphocyte isolate. This is not surprising, since the number of NK effector cells is only one of many variables which determine NK cell activity in an individual. In addition the there have been subsets described within the Leu llapositive cell population which differ considerably in cytolytic activity (9). It has been postulated that NK cells mediate immunosurveillance, and prevent the establishment of malignancy (33). To determine whether NK cells contribute to the pathogenesis of AIDS, an evaluation of the in viva conditions is important. For this reason we investigated some of the factors which may determine the eventual NK cell activity in persons with PGL and AIDS. The absolute number of Leu Ila-positive lymphocytes appears to decrease with disease progression. The decrease showed a moderate correlation with both the T4+ and TSf subset. Lymphopenia with decrease in T4+ lymphocyte subset has been previously recognized in AIDS; however, the mechanism is unknown (34). We demonstrated that NK cell activity can be inhibited by a plasma factor. This factor was significantly present in asymptomatic homosexuals and in PGL and AIDS patients, but could not be demonstrated in heterosexuals. It’s presence reduces the average NK cell activity in PGL patients to approximately one half of the values found in heterosexuals. However, in a substantial number of PGL and AIDS patients inhibition by autologous plasma was absent. NK cell reactivity as measured in autologous plasma did not correlate with T4+ or T4/T8 ratio, nor

NK CELLS

IN ACQUIRED

IMMUNODEFICIENCY

SYNDROME

149

did the degree of inhibition. This factor may be due to concurrent viral infections (35) rather than being casually related to AIDS. Although the nature of the inhibitory plasma factor(s) remains unclear, our results indicate that inhibition is probably not caused by binding of nonspecific antibodies or immune complexes to the effector cell. The factor(s) may be similar to those causing inhibition in lymphocyte transformation, or it may represent absence of a sustaining plasma factor necessary for both lymphocyte transformation and NK cell activity. Antilymphocyte or anti-HLA antibodies (36, 37) or a factor described in certain AIDS patients which inhibits T-cell responses (38, 39) may also affect the NK cell response. While it is unlikely that factors responsible for in vivo impairment of NK cell activity are a direct result of infection by HTLV-III, they may contribute to a decreased effectiveness of NK cells in the majority of AIDS and PGL patients. The occasional exception (lack of inhibitory factor combined with high NK cell reactivity) in PGL and even AIDS patients, however, indicates that the view that NK cells are solely responsible for immunosurveillance is an oversimplification. ACKNOWLEDGMENTS The authors gratefully acknowledge the contribution of the primary care physicians of the Vancouver Lymphadenopathy-AIDS Study Group (P. Constance, B. Douglas, S. Fay. R. Nitz. M. Maynard, and B. Willloughby).

REFERENCES 1. 2. 3. 4. 5.

Hailer, H., Harrison, M., Kiessling, R., and Wigzell, H., Nature (London) 270, 609, 1977 Saksela, E., Timonen, T., Ranki, A., and Hayry, P, Immunoi. Rev. 44, 71, 1979. Hailer, O., Kiessling, R., Om, A., and Wigzell, H., J. Exp. Med. 145, (411, 1977. Herberman, R. B., Nunn, M. E., and Lavrin, D. H., Znt. J. Cancer 16, 216. 1975. Herberman, R. B., and Holden, H. T., In “Advances in Cancer Research” (G. Klein and S. Weinhouse, Eds.), pp. 305-377, Academic Press, New York, 1978. 6. Hurme,M., andSinvola,M., Cell. Immunol. 84, 276. 1984. 7. Morretta, A., Pantales, G., Mingari, M. C., Melioli. G., Morretta. L.. and Cerottini, J. C., Eur. J. Immunol. 14, 121, 1984. 8. Abu, T., and Balch, C. M., J. lmmunol. 127, 1024. 1981. 9. Lanier, L. L., Philips, J. H., Warner, N. L., and Babcock. G. F., J. Leuk. Rio/. 35, 11, 1984. 10. Lanier, L. L., Engleman, E. G., Gatenby. P., Babcock, G. F.. Warner, N. L., and Herzenberg, L. A., Immunol. Rev. 74, 143, 1983. 11. Popovic, M., Sarngadharan, M. G., Read, E.. and Gallo, R. C., Science ( Wa.&&ron, D.C.) 224, 497, 1984. 12. Klatzmann, D.. Barre-Sinoussi, F., Nugeyre, M. T., Dauguet, C., Vilmer, E., Griscelli, C., BrunVezinet, F., Rouziouxa, C., Gluckman, J. C., Chermann. J. C.. and Montagnier, L., Science, (Washingron, D.C.) 225, 59, 1984. 13. Siegal,E P.,Semin. Oncol. 11, 29, 1984. 14. Cunningham-Rundles, S., Safai, B.. Metroka, C., Krown. S. E., Rubin, B. Y.. and Stahl, W. M.. In “AIDS, the Epidemic of Kaposi’s Sarcoma and Opportunistic Infections” (A. E. Friedman-

Klien, andL. J. Laubenstein,Eds.)pp. 153-160,Masson,New York, 1984. 15. Lederman, M. M., Ratnoff, 0. D., Scillian, J. J., Jones, P. K.. and Schackter. B., N. Eq. J. Med. 308, 79, 1983. 16. Siegal, F. P., Lopez, C., Hammer. G. S., Brown, A. E.. and Komfeld. S. J., N. Eng. J. Med. 305, 1439, 1981. 17. Schroff. R. W., Gottlieb, M. S., Prince, H. E., Chai, L. L., and Fahey, J. L., C/in. Immune/. and Immunopathol. 27, 300, 1983.

BLOOD

BY T-CELL

SUBSET

L

3

h

‘h

E3

/

1800

DONOR SCREENING Ts

rh

TS

h

Th

7

157 : 1

h

.. :. :: .:S.

1200

600

600

.. :.

Low Th : 1s Dono

2 ;1 u” 1600

Healthy

E3

4 s

‘h

3

L

1‘h

T

1800.

h

High Risk

s

rh

Patients

C: 1‘S

T ‘h

1200,

1200

&-l-J-

.‘. f .t.

::.

600

8

I I.

:. .:: ;.

.::. ; .;.. .. .. .. ..

Lymphadenopalhy

600

I Symptomatic

Patients

.i:. . -i. . . . .. . :.. . i’ AIDS

PotIonto

FIG. 3. Absolute counts of Th and Ts cells in blood donors with abnormal Th:Ts results, with AIDS, patients with lymphadenopathy syndrome or symptoms, and healthy homosexual high risk for developing AIDS. Within each group. individuals are separated according to they had: (A) an isolated decrease in Th (less than 6OO/ml), (B) an isolated increase in Ts than 6OO/ml), or (C) a decrease in Th with a concommitant increase in Ts.

patients men at whether (greater

the first determination is a consequence of an increase in the number of Ts cells; the Th count is within the normal range. At the time of the second determination, 5 months later, the total T-cell count had decreased and the Th:Ts was persistently abnormal, with an interval decrease in the helper/inducer subset to below normal levels. The patient is being followed carefully. DISCUSSION

We have analyzed T-lymphocyte subsets in 8715 consecutive volunteer blood donors. Serial analysis of repeat donors showed that for donors with values in the normal range, Th:Ts is a stable laboratory parameter within a given individual over periods of several months. The mean value for Th:Ts in our donors is similar to those reported by others on the basis of smaller data bases, and in agreement with these studies, our results show a statistically significant trend for an increasing value for Th:Ts with increasing age (16). The functional and clinical significance of this relationship is at present unclear. We also noted an age-independent effect of serologic CMV status on Th:Ts, with CMV-positive donors