The antiviral drug amantadine has a direct inhibitory effect on T-lymphocytes

Immunopharmacology, 18 (1989) 195-204

Elsevier

195

IMO 00473

The antiviral drug amantadine has a direct inhibitory effect on T-lymphocytes Connie Clark 1'2, Mildred M. W o o d s o n 1, Vern B. Winge 1 and Herbert T. N a g a s a w a 1'3 1Medical Research Division, Veterans Administration Medical Center, Minneapolis, MN, and 2Department of Laboratory Medicine and Pathology and 3Department of Medicinal Chemistry, University of Minnesota, Minneapolis, MN, U.S.A. (Received 12 April 1989; accepted 8 August 1989)

Abstract: We investigated the effect of the antiviral drug amantadine (AmTd) on polyclonal activation of thymic-dependent (T) and thymic-independent (B) lymphocytes from normal mice. In the present studies, T-lymphocytes are defined by their response to concanavalin A (Con A) and B-lymphocytes by their response to lipopotysaccharide (LPS). Polyclonal activator-induced lymphocyte proliferation was assessed by quantifying cellular incorporation of tritiated thymidine. The results show that, in a dose-dependent manner, AmTd exhibits at least 2-fold greater inhibitory activity against Con A-responding T-cells than against LPS-responding B-cells. Further, several findings demonstrate that AmTd has a direct inhibitory effect on T-lymphocytes. First, AmTd pulse treatment of isolated T-cells, but not accessory cells, abolished the T-cell response to Con A. Second, AmTd pulse treatment of the cytotoxic T-lymphocyte line, CTLL-2, markedly reduced their ability to undergo IL-2-induced proliferation. Third, proliferation of T-cells which had already undergone activation by ConA was inhibited by AmTd. Further, the finding that addition of IL-1, IL-2 or both to cultures failed to reverse inhibition of the response to ConA argues that AmTd did not interfere with endogenous production of these lymphokines. Possible implications of these findings are discussed. Key words:

Amantadine; Inhibition; Proliferation; T-lymphocyte

Introduction The drug 1-aminoadamantane (amantadine, AmTd) is commonly used for tremor control in Parkinson's disease (Calne, 1982; Muenter, 1982), and to protect against infection with type A influenza virus (Aoki and Sitar, 1985; Mostow, 1987). In acCorrespondence: Connie Clark, Ph.D., Medical Research Division, Bldg. 31, Rm 308, Veterans Administration Medical Center, One Veterans Drive, Minneapolis, MN 55417, U.S.A. Abbreviations: AmTd, amantadine, 1-aminoadamantane; Con A, concanavalin A; cpm, counts per minute; CM, culture medium; CTLL, cytotoxic T-lymphocyte line; IL-1, interleukin-1; I1-2, interleukin-2; IL-2R, interleukin-2 receptor; LPS, lipopolysaccharide; RatCSF, rat cell culture supernatant fluid; SEM, standard error of the mean; T, thymic-dependent; B, thymicindependent; [3H]TdR, tritiated thymidine.

cord with the well-documented efficacy of AmTd prophylaxis, a recent study demonstrated that AmTd provided 78% protection of normal adults against experimental challenge with an influenza A wild-type virus (Sears and Clements, 1987). It is generally held that weak bases, such as AmTd, exert their antiviral effect by entering the cell in neutral form whereupon they undergo protonation and, in the process, elevate the lysosomal/endosomal pH (Goldstein et al., 1985). An elevated pH inhibits lysosomal enzyme activity as well as association/fusion of viral components with lysosomal/endosomal membranes (Marsh, 1984). Of interest was a recent report demonstrating that in vitro T-lymphocyte killing by an anti-CD5 ricin A-chain conjugate was markedly augmented in the presence of AmTd (Siena et al., 1987), sug-

0162-3109/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)

196 gesting that AmTd may have increased translocation of the T-lymphocyte immunotoxin into the cell. This finding, as well as AmTd's well-documented protective effect against infection with influenza A virus, suggested the possibility that, in addition to AmTd's inhibitory lysosomotropic effect on antigen-processing/presenting cells, AmTd may also have a direct effect on lymphocytes. Since a review of the literature revealed a paucity of information regarding possible AmTd-mediated modulation of lymphocyte activity, studies were undertaken to examine this question. In the present in vitro studies, thymic-dependent (T)-lymphocytes are defined by their response to concanavalin A (Con A), and thymic-independent (B)-lymphocytes by their response to lipopolysaccharide (LPS). Proliferation induced by Con A and by LPS was assessed by quantifying cellular uptake of tritiated thymidine. We report here that, in a dose-dependent manner, AmTd exhibits at least 2-fold greater inhibitory activity against Con A-responding T-cells than against LPS-responding B-cells. Moreover, evidence is presented demonstrating that AmTd does indeed have a direct inhibitory effect on T-cells. First, AmTd pulse treatment of T-cells, but not accessory cells, abolished the T-cell response to Con A. Second, AmTd pulse treatment of the accessory cell-independent cytotoxic T-lymphocyte line, CTLL-2, markedly reduced their ability to undergo IL-2-induced proliferation. Third, proliferation of T-cells which had already undergone activation by Con A was inhibited by AmTd. Further, AmTd did not interfere with endogenous production of IL-1 or IL-2. This was evidenced by the finding that addition of IL-1, IL-2 or both to cultures failed to reverse AmTd-induced inhibition of the response to Con A.

Materials and Methods

Animals C57BL/6N female mice were obtained through a Veterans Administration-National Cancer Institute contract. Mice were fed certified Purina rodent chow and water ad libitum, and were about 2

months of age at the time of use. The Minneapolis Veterans Administration Medical Center Animal Research Facility is fully accredited by the American Association for Accreditation of Laboratory Animal Care.

Preparation of lymphocytes Single cell suspensions from spleens were prepared as previously described (Clark et al., 1987). Cell numbers were determined by hemocytometer counts and viability by dye exclusion. Enrichment for T-lymphocytes was accomplished using a mouse T-cell recovery column (Sci-Can Diagnostic, Novato, CA). The nonadherent T-cell population eluted from the column contained about 2% surface immunoglobulin-positive cells as assessed by FACS analysis of cells treated with fluorochrome-conjugated goat F(ab)2 anti-mouse IgG antibody (heavy and light chain specific, Cappel Laboratories, Cochranville, PA). Responsiveness to LPS was also used to monitor the presence of B-cells in the nonadherent T-cell population.

Mitogens Concanavalin A (Con A, Pharmacia, Piscataway, N J) and lipopolysaccharide (LPS, W from E. coli 0111:B4, Difco Labs., Detroit, M I ) w e r e dissolved in phosphate-buffered saline, pH 7.3, or RPMI-1640 (Grand Island Biological Co., Grand Island, NY). The stock solution of each mitogen was sterilized by membrane filtration, aliquoted, and stored at - 2 0 ° C or - 7 0 ° C .

Culture conditions for spleen cells Culture medium (CM) consisted of RPMI-1640 containing 2 mM L-glutamine, 5 ,uM 2-mercaptoethanol (Eastman Kodak Co., Rochester, NY) and 20 #g/ml of gentamicin sulfate (Sigma Chemical Co., St. Louis, MO). The medium was supplemented to contain an appropriate concentration(s) of Con A or LPS, and AmTd was added as needed. The cells were then added to provide a final concentration of 4 x 1 0 6 viable cells per ml of medium. For each condition, triplicate cultures (0.2 ml each) were plated and incubated in 96-well, flat-bottom culture trays (Linbro, Flow Labs., McLean, VA) at

197 37°C in a humidified atmosphere of 5% CO2 and 95% air. Proliferation was assessed by quantifying cellular incorporation of tritiated thymidine ([3H]TdR); 24 h before harvest, 1.4 #Ci of [3H]TdR (20 Ci/mmol, New England Nuclear, Boston, MA) was added to each culture. Except where otherwise noted, cultures were harvested on day 2 (day 0 is defined as the day cultures were initiated) using a multiple automated sample harvester (MASH II, Microbiological Associates, Bethesda, MD) and water washes. After drying, the filter discs (Whatman Inc., Clifton, N J) were immersed in scintillation fluid in filmware tubes (Nalge Co., Rochester, NY) and counted in a liquid scintillation spectrometer.

Preparation of lymphokine-containing culture supernatant fluid Lymphokine-containing rat cell culture supernatant fluid (RatCSF) was prepared by culturing spleen cells from inbred Lew/N rats in the presence of 1.0 #g of Con A per ml of medium for 48 h (Gillis et al., 1978). The supernatant fluid was collected, sterilized by membrane filtration, aliquoted, and stored at - 20°C. Human recombinant (r) IL-1 and mouse rIL-2 were obtained from Genzyme (Boston, MA). CTLL cells The IL-2-dependent murine cytotoxic T-lymphocyte line, CTLL-2, was obtained from the American Type Culture Collection (Rockville, MD). C T L L cells were maintained in medium consisting of RPMI-1640 supplemented to contain 15 mM HEPES, 2 mM L-glutamine, 2 mM sodium pyruvate, 10% fetal bovine serum (Hyclone Labs, Logan, Utah) and 40% RatCSF as a source of IL-2.

IL-2 assay Log-phase C T L L cells were plated in triplicate at 4 × 103 cells per culture (0.2 ml each), in medium containing successive 2-fold dilutions of rIL-2 or RatCSF as a source of IL-2. Cells were cultured for 24 h in a humid atmosphere of 5% COz and 95% air. Proliferation was assessed by uptake of [3H]TdR; 1.4 #Ci of [3H]TdR was added to each culture 4 h before the cultures were harvested.

Treatment of cells with amantadine Amantadine-HC1 (AmTd-HC1, Sigma Chemical Co., St. Louis, MO) was dissolved in phosphatebuffered saline, pH 7.3, sterilized by membrane filtration, aliquoted, and stored at - 20°C. For AmTd pulse treatment, cells were washed and incubated at 37°C for 4 h in medium alone and in medium containing an appropriate concentration(s) of AmTd. After washing, C T L L cells were cultured in medium containing successive 2-fold dilutions of RatCSF as a source of IL-2, while spleen T-cells were cultured in the absence and presence of Con A or LPS in medium without IL-2. Culture treatment of cells with AmTd consisted of initiating cultures of cells in medium which contained AmTd. Viability of spleen or C T L L cells following 4 h pulse treatment with 2.0 mM AmTd ranged from 83 to 96% of controls. Thus, AmTd did not exert a significant immediate cytotoxic effect on the cells. Analysis of data Based on cellular incorporation of [3H]TdR, proliferative responses were calculated as arithmetic mean counts per minute (cpm) of replicate cultures + the standard error of the mean (SEM). All experiments were done at least twice. Data transformations and statistical analysis were done using Statistix II (NH Analytical Software, Roseville, MN), Net cpm reflect the difference between mean cpm values in mitogen-stimulated and unstimulated cultures. Significance of the response differences was determined using the two-tailed Student's t-test. A p value of ~<0.05 was considered significant.

Results

Dose-dependent AmTd-induced inhibition of mitogen responses Spleen cells were cultured in medium alone, in medium containing from 0.01 to 1.0 mM AmTd, and in the absence and presence of Con A or LPS. The results depicted in Fig. 1 show that over the concentration range 0.1 to 1.0 mM AmTd, the drug exhibited a 2-fold greater inhibitory effect on Con Athan on LPS-responding cells.

198 I00 o=

400 [

~ , ~ L P S

o3°° _ t

o

2 °o.

Con

×

<->,~ o

m Control [ ] + 0.2 mM AmTd

I:i::

i []+°'4mMAmtd

200

03 +t

t00

*

u

.01

0.I

1.0

m M Amantadine

Fig. 1. Dose-dependent amantadine-induced inhibition of the response to Con A and to LPS. Spleen cells from normal C57BL/6N mice were cultured in the absence and presence of Con A (0.6/*g/ml) or LPS (80 #g/ml), in medium alone and in medium containing various concentrations of amantadine. Cultures were harvested on day 2. The data depict the combined results of five separate experiments. In the concentration range 0.1 to 1.0 m M A m T d , linear regression analysis indicated that the slope for Con A was 1.724, while the slope for LPS was 0.799. An asterisk indicates that responses of amantadine and control cultures differed with a p value of ~<0.05. Mean background cpm 4- SEM - 21060 =1- 6727. The mean net cpm ± SEM of untreated cells to Con A = 225269 ± 34443 and to LPS = 282007 ± 7895.

Effect of ArnTd on mitogen dose-responses and kinetics The dose-dependent AmTd-mediated inhibition of mitogen responses we observed could result if A m T d modified interactions between cells and mitogen in a manner which altered the concentration of mitogen required to evoke a m a x i m u m response. Alternatively, A m T d could induce changes which merely delayed cellular activation, thereby producing a shift in the kinetics of the response to Con A or LPS. The results show, however, that neither condition accounted for the inhibitory effect of A m T d on mitogen responses. It can be seen that m a x i m u m Con A-induced (Fig. 2, top panel) and LPS-induced (Fig. 2, b o t t o m panel) proliferation in the absence as well as in the presence of A m T d occurred at essentially the same concentration of Con A and at the same concentration of LPS. It can also be seen that A m T d did not cause a shift in the kinetics of the response to Con A (Fig. 3, top panel) or to LPS (Fig. 3, bottom panel).

7

o

0.3

.~

0.6

t.2

2.4

4.8

ConA (,ug/mlof medium)

E 400 o

~4

= 300 ~:~ 200 IO0 O

h

40

h 80

h

160

h 320

LPS (,ug/ml of medium)

Fig. 2. Amantadine does not alter the dose-response to Con A or LPS. Spleen cells from normal C57BL/6N mice were cultured in medium containing various concentrations of Con A or LPS, and in the absence and presence of 0.2 or 0.4 m M amantadine. Cultures were harvested on day 2. An asterisk indicates that responses of amantadine and control cultures differed with a p value of ~<0.05. Mean background cpm + SEM were: no amantadine = 7198 + 367; 0.2 m M amantadine - 1583 ~- 127: and 0.4 m M amantadine = 493 ± 33.

Exogenous lymphokines fail to reverse AmTd-mediated inhibition of the response to Con A Experiments were then done to examine whether A m T d might act by inhibiting cellular elaboration of endogenous IL-1 or IL-2; together, these lymphokines are generally held to provide signals essential to drive the proliferative response to Con A. Accordingly, cells were cultured in medium alone, in medium containing AmTd, and in the absence and presence of Con A. Two units of rIL- 1 or rIL-2 or 2 units each o f r I L - 1 and rIL-2 were added at the time cultures were initiated (0 h) or 24 h after cultures were initiated. The results show (Fig. 4) that addition of IL-1 or IL-2 or both to cultures failed to reverse AmTd-induced inhibition of the response to Con A.

199 Lymphokine Added

300 250

x

200

.=_o ~, 5o

ConA~¢=~'~

~E

uJ (z) +1

ICM

EZITL-I

q

o

•

I00

/

E

2oo

IL-2

I~IL I+2

~m

+02rnMAmTd

I-.-I +1

c) Z

8 0

e ~ . , , ~ ,

0

I

,

2

,

3

•

4

300 o

R I-I

2°°

LPsyC\

CM + AmTd 0 hrs

CM

CM +AmTd 24"hrs

Time of Lyrnphokine Addition

I--

Z

CM

oo

0

I

2 3 Doys in Culture

4

Fig. 3. Amantadine does not change the kinetics of the response to Con A or LPS. Spleen cells from normal C57BL/6N mice were cultured in the absence and presence of Con A (0.6 ~g/ml) or LPS (80/~g/ml), in medium alone and in medium containing 0.2 mM amantadine. Cultures were harvested on days 1, 2, 3 and 4. An asterisk indicates that responses of amantadine and control cultures differed with a p value of ~<0.05. Mean background cpm 4- SEM were: in the absence of amantadine, day 1 = 8127 4216, day 2 = 10070 + 585, day 3 = 5093 + 356, and day 4 = 1787 + 95; and in the presence of 0.2 mM amantadine, day 1 = 4539 ± 186, day 2 = 2601 i 98, day 3 = 642 ± 79, and d a y 4 = 207 ± 30.

Effect of A m T d on isolated T-cells and accessory cells Based on the preceding findings, it seemed likely that the inhibitory action of AmTd resulted from a direct effect of the drug on T-lymphocytes. Although stimulation of T-cells by Con A may not require mitogen processing, it is thought to require accessory cell elaboration of a co-stimulator activity such as IL-1 (Chestnut and Grey, 1986). To examine whether AmTd did in fact directly affect Tcells, cell mixing experiments were done. Isolated T-cells or irradiation-inactivated (3500 rads) accessory cells (spleen cells) were incubated in medium alone and in medium containing 2.0 mM AmTd for 4 h (pulse treatment). The cells were washed twice to remove the drug. T-cells and accessory cells were then cultured together in the absence and presence

Fig. 4. Neither IL-I nor IL-2 reverses amantadine-induced inhibition of Con A-responding cells. Spleen cells from normal C57BL/6N mice were cultured in the absence and presence of Con A (0.6/lg/ml), in medium alone and in medium containing 0.2 mM amantadine. Two units of IL-1 or IL-2 or two units each of IL-1 and IL-2 were added at the time cultures were initiated (0 h) or 24 h after cultures were initiated. Corresponding control cultures received an equivalent volume of medium. Cultures were harvested on day 2. An asterisk indicates that responses of control and lymphokine-supplemented cultures differed with a p value of ~<0.05. Mean background cpm were as follows. Zero (0) h lymphokine addition in the absence/presence of AmTd: culture medium (CM) alone = 7989/2595; CM + IL-1 = 13028/2021; CM + IL-2 = 9967/3059; CM + IL-1 + 2 = 11813/2941; and 24 h lymphokine addition in the absence/presence of AmTd: CM + 8462/2154; CM + IL-I = 10232/1561; CM + IL-2 = 9231/ 3025; CM + IL-1 + 2 = 10849/4124.

of Con A or LPS. As can be seen (Table I), treatment of T-cells with AmTd completely abolished their ability to respond to Con A. On the other hand, the response of untreated T-cells to Con A was essentially the same whether the accessory cells with which they were cultured had been treated with irradiation alone or treated with irradiation and also pulsed with AmTd. Thus, the results show that AmTd inhibited the response to Con A by virtue of a direct drug-mediated effect on T-lymphocytes. This result was not attributable to an immediate, drug-induced cytotoxic effect, since spleen and C T L L cell viability following 4 h pulse treatment with 2.0 mM AmTd ranged from 83 to 96% of controls. This accords with AmTd's previously reported lack of significant cytotoxicity for rat 3Y1 cells treated with 2.0 mM AmTd for 24 h (Shimura et al., 1987), for mouse 3T3 cells treated with 5.0 mM

200 TABLE I Alnantadine has a direct inhibitory effect on Con A-responding T-cells Cell treatment before coculture

Mitogen present in the culture medium (Data a r e c p m ± SEM x 10 3)

T-cells

Accessory cells

None

Untreated Amantadine Untreated Amantadine

3500 3500 3500 3500

0.4 0.2 0.5 0.2

rads rads rads AmTd fads AmTd

4± ± ±

Con A 0.0 0.0 0.0 0.0

37.3 0.7 40.5 0.4

± ± + ±

LPS 0.8 0.2 6.8 0.1

2.8 0.5 2.9 0.5

± m I ±

0.2 0.2 0.3 0.0

Isolated T-lymphocytes as well as irradiation-inactivated (3500 rads) spleen cells (as a source of accessory cells) were pulsed with culture medium alone or with culture medium containing 2.0 mM amantadine for 4 h at 37°C. After washing, T-cells (6 x 10S/culture) were comixed in culture with accessory cells (2 x 10S/culture) in medium alone and in medium containing Con A (0.6 llg/ml) or LPS (80 /~g/ml). Cultures were harvested on day 2. Cpm x 103 for irradiation-inactivated (3500 fads) accessory cells cultured alone (8 × 105/culture) were: culture medium (CM) = 0.3; CM + Con A = 0.2; and CM + LPS = 1.5; while those for irradiation-inactivated (3500 rads) and amantadine-pulsed (2.0 mM) accessory cells cultured alone (8 x i0S/culture) were: CM = 0.2: CM + ConA - 0.1; and CM + LPS - 0.2.

A m T d for 1 h (Schlegel et al., 1982), for a human hepatoma-derived cell line treated with 5.0 mM A m T d for 20 rain (Hortin and Strauss, 1986), and for the T 4 + T-cell line JM treated with 20.0 m M A m T d for up to 6 h (Maddon et al., 1986).

A m T d ' s effect on the cells or by an interaction with IL-2. To examine this, log-phase C T L L cells were washed, and then incubated in medium alone and in medium containing various concentrations of A m T d for 4 h (pulse treatment). After washing to

Effect of AmTd on IL-2-dependent proliferation o/" the cytotoxic T lymphoo'te line, CTLL-2 Since the foregoing experiments indicated that A m T d inhibited Con A-induced proliferation by virtue of its effect on T-lymphocytes, we examined whether A m T d would have a similar effect on proliferation of accessory cell-independent C T L L cells. C T L L cells were employed because an exogenous source of IL-2 is known to be both necessary and sufficient to support their proliferation (Gillis et al., 1978). In initial experiments, C T L L cells were cultured in IL-2-supplemented medium alone and in IL-2-supplemented medium containing 0.01 or 0.1 m M AmTd. The results show (Fig. 5) that proliferation of C T L L cells was marginally decreased in the presence of 0.01 m M AmTd, but was significantly decreased in the presence of 0.1 m M AmTd. It could not be discerned using this approach, however, whether inhibition resulted because of

CTLL: Culture Treatment "Control 50 I~1+O.OImM AmTd [ ] + O.I mM AmTd

Ohh h

gb o

x

ow I-4

"~ F-

50

4-1

o.

~: E 21

0

2 4 8 16 52 Rat CSF (reciprocalof the dilution inCM)

Fig. 5. Amantadine inhibits proliferation of IL-2-dependent CTLL cells. CTLL cells maintained in log-phase were washed and cultured (4 x 103/culture) for 24 h in the presence of various dilutions of RatCSF as a source of IL-2, and in the absence and presence of 0.01 or 0.1 mM amantadine. [3H]TdR was added 4 h before cultures were harvested. An asterisk indicates that responses of amantadine and control cultures differed with a p value of ~<0.05. Mean background cpm :k SEM in the absence of lL-2 = 341 4= 83.

201 8.0

CTLL: Pulse Treatment • Contol

~-4 ÷J

rv E "~'~ ,~,z

2.0

o.o

0

2

4

8

16

32

64

Rat CSF (reciprocal of the dilution in CM )

Fig. 6. Amantadine has a direct inhibitory effect on 1L-2-dependent CTLL cells. CTLL cells maintained in log-phase were washed and incubated at 37°C for 4 h in medium alone and in medium containing 0.4 or 1.0 mM amantadine. After washing, the cells were cultured (4 x 103/culture) for 24 h in the presence of various dilutions of RatCSF as a source of I1-2. [3H]TdR was added 4 h before cultures were harvested. An asterisk indicates that responses of amantadine-pulsed and medium-pulsed cells differed with a p value of ~<0.05. Mean background cpm + SEM in the absence of RatCSF were: medium-pulsed cells = 348 4120; 0.4 mM amantadine-pulsed cells = 87 4- 21; and 1.0 mM amantadine-pulsed cells = 64 4- 8.

remove the drug, CTLL cells were cultured in the absence and presence of various dilutions of RatCSF as a source of IL-2. It can be seen (Fig. 6) that pulse treatment of CTLL cells with AmTd (in the absence of IL-2) clearly decreased their ability to undergo IL-2-induced proliferation. Thus, as was the case for Con A-induced proliferation of normal T-lymphocytes, inhibition of IL-2-induced proliferation of CTLL cells was shown to result from a direct, AmTd-mediated effect on the cells.

Discussion

Weak bases such as AmTd, in free base form, readily permeate cell membranes (Ohkuma and Poole, 1981). Subsequent intralysosomal protonation of the base elevates the pH, which results in inactivation of various lysosomal acid hydrolases (Ohkuma and Poole, 1978), and also prevents virus-cell membrane fusion (Richman et al., 1986). It is thought that AmTd protects against influenza A virus infection by this mechanism and, in addition, inhibits

degradation of antigen by accessory cells, the latter being a requisite step in the process of accessory cell presentation of antigen to T-lymphocytes for initiation of an immune response. AmTd has, however, also been reported to inhibit protein synthesis (Shimura et al., 1987), to inhibit phospholipases and induce lipidosis (Hostetler and Richman, 1982), to inhibit functions of the secretory pathway (Hortin and Strauss, 1986), and to modify interaction of membrane components (Herrmann et al., 1985; Tverdislov et al., 1986). Chloroquine, which is thought to be a similarly-acting weak base, exhibits a similar range of biological effects (Patterson and Oxford, 1986). A recent report demonstrating that killing by a T-cell-specific immunotoxin was markedly augmented in the presence of AmTd (Siena et al., 1987) suggests that AmTd may modify the plasma membrane in a manner which increases translocation of the immunotoxin. These findings prompted us to investigate AmTd's effect on lymphocyte activity. In the present in vitro studies, mitogen-induced proliferation was used to assess AmTd's effect on lymphocytes. T-lymphocytes were defined by their proliferative response to Con A, and B-lymphocytes by their proliferative response to LPS. We report here that AmTd exhibits a dose-dependent inhibitory effect on Con A- and LPS-stimulated lymphocyte proliferation. Of particular interest was the finding that AmTd has at least a 2-fold greater inhibitory effect on Con A-responding T-cells than on LPS-responding B-cells. Moreover, several lines of evidence demonstrate that AmTd-induced inhibition of T-cell proliferation results from a direct drug-mediated effect on T-cells. First, AmTd pulse treatment of T-cells, but not accessory cells, abolished the T-cell response to Con A. Second, AmTd pulse treatment of the accessory cell-independent cytotoxic T-lymphocyte line, CTLL-2, markedly reduced their ability to undergo IL-2-induced proliferation. Third, proliferation of T-cells which had already undergone activation by Con A was inhibited by AmTd. Moreover the finding that addition of IL-1, IL-2, or IL-1 plus IL-2 to cultures failed to reverse AmTd-induced inhibition of the lymphocyte response to Con A (Fig. 4) shows

202 that AmTd did not interfere with endogenous production of these lymphokines. Although stimulation of unprimed T-cells by Con-A is not thought to require mitogen processing, it is thought to require accessory cell elaboration of a co-stimulator activity such as IL-1 (Chestnut and Grey, 1986). The prevailing view is that accessory cell-elaborated IL-1 activates T-cells to endogenous synthesis of IL-2 and expression of IL-2 receptors (IL-2R) (Sprent and Webb, 1987). Subsequent binding of IL-2 by IL-2Rs provides the final signal that triggers T-cells to proliferate. C T L L cells, on the other hand, depend solely on an exogenous source of IL-2 to undergo proliferation. The present experiments show that following ArnTd pulse treatment, T-cells did not recover their ability to respond to Con A (Table I), and C T L L cells did not recover their ability to respond to IL-2 (Fig. 6). If AmTd's effect was limited to deacidification of cellular lysosomes/endosomes, recovery of responsiveness to Con A by T-cells and to IL-2 by C T L L cells would be predicted, since the effects of lysosomotropic weak bases have been shown to be readily reversed when the drug is withdrawn (Jensen, 1988; Scala and Oppenheim, 1983). Significantly, studies by Richman et al. (1981) demonstrated that although AmTd concentrates in lysosomes, the antiviral effect is lost immediately upon removal of the drug from the medium. Thus, the anti-influenza A virus activity of AmTd is independent of the intralysosomal concentration of the drug. These findings, together with the well-documented membrane-perturbing effects of AmTd (Herrmann et al., 1985; Tverdislov et al., 1986), suggest that AmTd's inhibition of Con A-responding T-cells and C T L L cells may more likely reflect effects directed to the plasma membrane. AmTd was clearly less inhibitory to LPS-responding cells. Experiments which showed that the degree of inhibition of LPS responses by AmTd lessened as the concentration of LPS increased (Fig. 2, lower panel) suggest the possibility of an interaction between AmTd and LPS. The finding that LPS binds to various mammalian cells in a nonsaturable manner (Kirkland et al., 1988) supports the view that acquisition of LPS responsiveness may result

from insertion of the lipophilic LPS into the cell membrane (Jakobovitz et al., 1982). Presuming this to be the case, it might be speculated that increasing the concentration of LPS may interfere with AmTd's effect on the plasma membrane. This might reasonably explain our finding that, as the concentration of LPS was increased, AmTd's inhibitory effect on the response to LPS decreased (Fig. 2, lower panel). It has been shown that Con A binding to cells is rapid, and that a pulse as short as 3 h is sufficient to induce proliferation of lymphocytes (Hadden, 1988). Further, while detectable levels of IL-2R are expressed as early as 4 8 h after stimulation with Con A, optimal levels of IL-2Rs are seen about 24 h after culture initiation (Malek et al., 1985). Using this time-frame, which parallels the response kinetics of our system, we compared the effect of AmTd on resting and activated cells. Thus, AmTd (0.2 mM) was added to LPS- or Con A-containing cultures at the time cultures were initiated (resting cells) or 24 h after they were initiated (mitogen-activated cells). Compared to resting cells, cells cultured for 24 h showed reduced sensitivity to AmTd; the reduction in sensitivity was about 37% for LPS and about 8% for Con A (data not shown). This indicates that AmTd is markedly inhibitory to Con A-responding T-cells well beyond the time when early, direct T-cell interaction with accessory cells or stimulation by accessory cell-elaborated factors might occur. Again, this argues for a direct effect of AmTd on Con A-responding cells. Moreover, whether resting or activated, AmTd was significantly more inhibitory for Con A- than for LPS-responding cells. The inhibitory effect of AmTd on proliferation of Con A- and on LPS-stimulated lymphocytes was not due to drug-mediated cytotoxicity. This was evidenced by the fact that cell viability following 4 h pulse treatment with 2.0 mM AmTd did not differ significantly from that of cells incubated with medium alone. Lack of significant cytotoxicity has been reported for rat 3Y1 cells treated with 2.0 mM AmTd for 24 h (Shimura et al., 1987), for mouse 3T3 cells treated with 5.0 mM AmTd for 1 h (Schlegel et al., 1982), and for human hepatoma-derived

203 cells treated with 5.0 mM AmTd for 20 min (Hortin and Strauss, 1986). Although the present work clearly demonstrates that AmTd has a direct inhibitory effect on T-cells, additional studies are required to elucidate the mechanism(s) involved. Together with findings by others, however, several implications arising from the present studies deserve brief mention. Previously documented AmTd-mediated effects on the plasma membrane include (1) an increase in fluidity (Herrmann et al., 1985), (2) inhibition of ligand clustering in clathrin-coated pits (Schlegel et al., 1982), and (3) blocking of receptor-mediated endocytosis (Garcia and Sanchez, 1983). These findings, together with the well-documented efficacy of prophylactic AmTd-mediated protection against infection by influenza A virus, collectively argue for more persistent AmTd-mediated effects on the immune system than those which might accrue from the transitory, readily reversible deacidification of lysosomal/endosomal systems of cells. In addition, Shobukhov et al. (1983) demonstrated that shortterm in vitro treatment of influenza virus-infected murine leukemia Rauscher cells with AmTd interrupted the viral infectious cycle for up to three months. Moreover, Sears and Clements (1987) found that experimental challenge of healthy humans with influenza virus, after short-term pretreatment with AmTd, caused a 100-fold decrease in virus shedding compared to placebo controls. Clearly, the outcome in the studies cited above is also incompatible with the notion that AmTd's effect is restricted to a transitory, readily reversible deacidification of lysosomal/endosomal vesicles. Studies are under way using AmTd and 1-nitroadamantane to determine whether replacement of the amine group, which gives AmTd its lysosomotropic characteristics, with a nitro group alters this agent's inhibitory effect on T-lymphocytes.

Acknowledgements This work was supported by a grant from the Veterans Administration. We thank Dr. Hollis Krug for critically reviewing this manuscript.

References Aoki FY, Sitar DS. Amantadine kinetics in healthy elderly men: implications for influenza prevention. Clin Pharmacol Ther 1985;37:137. Calne DB. The role of various forms of treatment in the management of Parkinson's disease. Clin Neuropharmacol 1982;5:Suppl l:$38. Chestnut RW, Grey HM. Antigen presentation by B cells and its significance in T-B interactions. Adv Immunol 1986;39:51. Clark C, LaSota I, Borch RF. Low dose x-irradiation and 4hydroperoxycyclophosphamide distinguishes three Lyt-l+ lymph node T-cell subtypes that respond to specific antigen in vitro. J Leukocyte Biol 1987;41:330. Garcia GM, Sanchez CM. Dansylcadaverine and rimantadine inhibition of phagocytosis, PAF-acether release, and phosphatidylcholine synthesis in human polymorphonuclear leukocytes. Immunopharmacology 1983;6:317. Gillis S, Ferm MM, Ou W, Smith KA. T-cell growth factor: parameters of production and a quantitative microassay for activity. J Immunol 1978;120:2027. Goldstein JL, Brown MS, Anderson RGW, Russell DW, Schneider WJ. Receptor-mediated endocytosis: concepts emerging from the LDL receptor system. Annu Rev Cell Biol 1985;1:1. Hadden JW. Transmembrane signals in the activation of T lymphocytes by mitogenic antigens. Immunol Today 1988;9:235. Herrmann A, Lentzsch P, Lassmann G, Ladhoff AM, Donath E. Spectroscopic characterization of vesicle formation on heated human erythrocytes and the influence of the antiviral agent amantadine. Biochim Biophys Acta 1985;812:277. Hortin G, Strauss AW. Effects of acidotropic compounds on the secretory pathway inhibition of secretion and processing of the third and fourth components of complement. Biochem Biophys Res Commun 1986;136:603. Hostetler KY, Richman DD. Studies on the mechanism of phospholipid storage induced by amantadine and chloroquine in Madin Darby canine kidney cells. Biochem Pharmacol 1982;31:3795. Jakobovitz A, Sharon N, Zan-Bar I. Acquisition of mitogen responsiveness by non-responding lymphocytes upon insertion of appropriate membrane components. J Exp Med 1982;156:1274. Jensen PE. Protein synthesis in antigen processing. J Immunol 1988;141:2545. Kirkland TN, Ziegler E J, Tobias P, Ward DC, Michalek SM, McGheeJR, Macher I, Urayama K, Appelmelk BJ. Inhibition of lipopolysaccharide activation of 7OZ/3 cells by antilipopolysaccharide antibodies. J Immunol 1988;141:3208. Maddon PJ., Dalgleish AG, McDougal JS, Clapham PR, Weiss RA, Axel R. The T4 gene encodes the AIDS virus receptor and is expressed in the immune system and the brain. Cell 1986;47:333. Malek TR, Schmidt JA, Shevach EM. The murine IL 2 receptor. III. Cellular requirements for the induction of IL 2 receptor

204 expression on T cell subpopulations. J lmmunol 1985:134:2405. Marsh M. Review article: the entry of enveloped viruses into cells by endocytosis. Biochem J 1984;218:1. Mostow SR. Prevention, management, and control of influenza. Role of amantadine. Am J Med 1987:82:(6A):35. Muenter MD. Initial treatment of Parkinson's disease. Clin Neuropharmacol 1982:5:Suppl 1:$2. Ohkuma S, Poole B. Fluorescence probe measurement of the intralysosomal pH in living cells and the perturbation of pH by various agents. Proc Natl Acad Sci USA 1978:75:3327. Ohkuma S, Poole B. Cytoplasmic vacuolation of mouse peritoneal macrophages and the uptake into lysosomes of weakly basic substances. J Cell Biol [981;90:656. Patterson S, Oxford JS. Early interactions between animal viruses and the host cell: relevance to viral vaccines. Vaccine 1986;4:79. Richman DD, Hostetler KY, Yazaki PJ, Clark S. Fate of influenze A virion proteins after entry into subcellular fractions of LLC cells and the effect of amantadine. Virology 1986; l 51:200. Richman DD, Yazaki P, Hostetler KY. The intracellular distribution and antiviral activity of amantadine. Virology 1981;112:81. Scala G, Oppenheim JJ. Antigen presentation by human monocytes: evidence for stimulant processing and requirement for interleukin 1. J Immunol 1983:131:1160.

Schlegel R, Dickson RIF, Willingham MC, Pastan 1H. Amantadine and dansylcadaverinc inhibit vesicular stomatitis virus uptake and receptor-mediated endocytosis of alpha 2-macroglobulin. Proc Natl Acad Sci USA 1982:79:2291. Sears SD, Clements ML. Protective cfllcacy of low-dose amantadine in adults challenged with wild-type inlluenza A virus. Antimicrob Agents Chemother 198731: I470. Shimura H, Umeno Y, Kumura G. Effects of inhibitors of the cytoplasmic structures and functions on the early phase of infection of cultured cells with simian virus 40. Virology 1987:158:34. Shobukhov VM, Linitskaia GL, Galegov GA. Inhibiting action of adamantane derivatives on chronic influenzal infection in a tissue culture. Vopr Virusol 1983;3:325. Siena S, Villa S, Bregni M, Bonnadonna G, Gianni M. Amantadine potentiates T lymphocyte killing by an anti-pan-T cell (CD5) ricin A-chain immunotoxin. Blood I987;69:345. Sprent J, Webb SR. Function and specificity o f T cell subsets in the mouse. Adv Immunol 1987;41:39. Tverdislov VA, E1-Karadagi S, Kharilonenkov IG, Glaser R, Donath E, Herrmann A. Lentzsch P, Donath J. Interaction of the antivirus agents remantadine and amantadine with lipid membranes and the influence on the curvature of human red cells. Gen Physiol Biophys 1986;5:61.