Lymphocyte Function

or antigen-induced transformation. Lymphocyte responses to the mitogens phytohemagglutinin and concanavalin A were not changed by exposure to amphotericin B, 5-fluorocytosine, or the combination ofboth drugs. T-lymphocyte receptors for sheep erythrocytes and B-lymphocyte surface immunoglobulin and receptors for complement were not changed by treatmnent with amphotericin B or 5fluorocytosine.

Amphotericin B (AmB) is a polyene antibiotic that binds to sterols in the fungal cell membrane, leading to leakage of cell contents and death of the cell (9). 5-Fluorocytosine (5-FC) is a fluorinated pyrimidine which is used almost exclusively in combination with AmB in infections caused by organisms such as Candida, Torulopsis, and Cryptococcus. Using in vitro assays for T-lymphocyte function and an in vivo murine model of T-cell fimction, several 273,1977). However, immunostimulatory effects of AmB also have been reported in murine systems (4,10,11,14,15,20). The effects of 5-FC and the combination of 5-FC and AmB on the immune system have not been studied.
The effects of these antifungal agents on the immune system may be of great significance in the treatment of patients with fungal diseases. Immunosuppression could be detrimental to successful treatment of fungal infections. On the other hand, immunostimulatory effects of antifungal antibiotics could prove to be a useful adjunct to host defenses against fungi. This study specifically evaluated the effect of AmB, 5-FC, and the combination of both drugs on Band T-lymphocyte function in vitro.
MATERIALS AND METHODS Antifungal drugs. AmB was obtained from E. R. Squibb & Sons, Inc., Princeton, N.J. To each vial containing 50 mg of sterile lyophilized AmB was added sterile distilled water so that the final concentration of AmB was 5,000 Mg/ml. This stock solution was stored in aliquots at -70°C until used. Further dilutions were made with phosphate-buffered saline (PBS), pH 7.2. 5-FC powder (kindly supplied by Hloffman-La Roche, Inc., Nutley, N.J.) was diluted in PBS to a concentration of 5,000 ,Lg/ml, filtered through a 0.45-,m membrane filter (Millipore Corp., Bedford, Mass.), and stored in aliquots at -70°C until used. Further dilutions were made with PBS.
T-cell rosette formation. Ficoll-Isopaque-separated mononuclear cells (5) from six healthy donors were washed three times in Hanks balanced salt solution (HBSS) and incubated in HBSS for 2 h at 370C with concentrations of 1 and 5 ug of AmB per ml and 10 and 100 ,ug of 5-FC per ml. Control cells were incubated with PBS. The cells were then washed three times with HBSS, and T-celi rosettes were prepared by the method of Jondal et al. (12). A 0.25-ml quantity of cell suspension (4 x 108 cells per ml), 0.25 ml of sheep erythrocytes (108 cells per ml), and 1 drop of latex particles were mixed together. The mixture was incubated for 10 min at 37°C, spun at 200 x g for 5 min, and placed at 40C overnight. Those lymphocytes with three or more sheep erythrocytes attached were scored as T rosettes.
B-cell complement receptors and surface immunoglobulin. Washed Ficoll-Isopaque-separated mononuclear cells from six healthy donors were incubated in HBSS for 2 h at 37°C with 1 and 5 Ag of AmB per ml and 10 and 100 jg of 5-FC per ml. Control cells were incubated with PBS. Zymosan (Sigma Chemical Co., St. Louis, Mo.) was boiled in 10 volumes of 0.9% NaCl, centrifuged at 5,000 x g, washed three times with HBSS, and suspended at a concentration of 1 mg/ml. This suspension was mixed with fresh human serum for 30 min at 37°C (allowing fixation of complement), washed three times with HBSS, diluted to a final concentration of 0.5 mg/ml, and frozen at -70°C in 0.4-ml aliquots (16).
After incubation with the drugs, the lymphocytes were washed with HBSS, and 0.2 ml of a suspension of 4 x 100 cells per ml was mixed with 0.4 ml of complement-coated zymosan particles. The suspension was centrifuged at 200 x g for 5 min and counted immediately. Lymphocytes with three or more zymosan particles attached were considered to have complement receptors.
For determination of surface immunoglobulin, 0.1 ml of cell suspension (2 x 107 cells per ml) was combined with 0.1 ml of fluorescein-conjugated goat antihuman immunoglobulin M (Meloy Laboratories, Springfield, Va.) following the 2-h,,incubation with drugs. This mixture was incubated at 4°C for 30 mi, the cells were washed three times with cold HBSS, and the percentage of fluorescent cells was determined with a Zeiss standard microscope fitted with fluorescent accessories. Lymphocyte transformation. Ficoll-Isopaqueseparated mononuclear cells from 12 healthy donors were washed twice with Eagle minimum essential medium (Grand Island Biological Co., Grand Island, N.Y.) and suspended at a concentration of 5 x 105 cells per ml in Eagle miniimal essential medium with 20% fetal calf serum (Grand Island Biological Co.), 1 mM L-glutamine, and 10 ,ug of streptomycin per ml added. Each glass culture tube (12 by 75 mm) received 1 ml of cell suspension. Quantities of 10 pl of varying concentrations of the following were added to the appropriate cultures: phytohemagglutinin-M (PHA; Difco Laboratories, Detroit, Mich.), undiluted to 1:30 dilution; concanavalin A (ConA; Pharmacia Fine Chemicals, Inc., Piscataway, N.J.), 2 to 20 ug/ml; and streptokinase-streptodornase (SK-SD; American Cyanamid Co., Lederle Laboratories Div., Pearl River, N.Y.), 4 to 40 U of SK. AmB was added to the appropriate cultures in concentrations varying from 0.2 to 10 ,ug/ml.   After the appropriate incubation period (3 days for mitogens, 6 days for SK-SD), the cultures were pulselabeled for 2 h with 1 ,uCi of [methyl-3H]thvmidine (specific activity, 6 Ci/mmol). The amount of radioactivity incorporated into the acid-precipitable fraction was assayed in a Beckman liquid scintillation counter (13). Results were expressed as mean disintegrations per minute (DPM) or as the stimulation index, which is the ratio of DPM in stimulated cultures to DPM in control cultures. For each volunteer the dose of antigen or mitogen giving the maximum response was used for all calculations. Statistical analysis. The data were evaluated with the Student t test for paired samples.

RESULTS
The expression of lymphocyte surface receptors and surface immunoglobulin was not altered by treatment with AmB or 5-FC. The percentage of T and B lymphocytes were the same in control and drug-treated mononuclear cell preparations (Table 1). Trypan blue dye exclusion showed >95% viability of all preparations after the 2-h incubation period with drugs.
Spontaneous lymphocyte transfornation was significantly depressed by concentrations of AmB varying from 0.2 to 5.0 ,ug/ml (P < 0.005) (Fig. 1). Further studies showed that concentrations as low as 0.05 ,ug of AmB per ml suppressed spontaneous transfornation by 40%. The mitogenic response, expressed as DPM, to both PHA and ConA was unchanged by the addition of AmB to the cultures. However, the response to the specific antigen SK-SD was significantly diminished (P < 0.005) by the addition of AmB.
In contrast to AmB, 5-FC, in concentrations from 10 to 200 ,g/ml, had no effect on spontaneous transformation, mitogenic response to PHA or ConA, or the response to the specific antigen SK-SD (Fig. 2j. The combination of both AmB and 5-FC appeared to act in a fashion similar to that of AmB alone (Fig. 3). Spontaneous transformation and FIG. 1. Effect of varying concentrati on spontaneous lymphocyte transformat on the response of lymphocytes to PH.A (A), and SK-SD (0), expressed as mec each concentration ofAmB. An asterisk P < 0.(K)5 (Student's t test for paired sam the response to SK-SD were both di the same extent by the combination a with AmB alone (P < 0.005). The I PHA and ConA was not significant after exposure to the combination o 5-FC.
If the data are expressed in terms tion index, it appears that AmB s combination with 5-FC enhances th4 response. However, this increase is to a drop in background spontaneot mation and does not reflect a true absolute DPM.

DISCUSSION
AmB is currently the mainstay of chemotherapy. Patients with dissem gal infections are frequently immuno as a result of their underlying diseas in some instances as a result of their c ing fungal infection (6,18). Any furth sion of the immune response by the agent could prove detrimental to tbh cure of the infection. Because of the r of factors influencing the immune patients with serious fungal infections, we sought to assess the effects of AmB, 5-FC, and the combination of both drugs in a well-defined in vitro system.
In vitro, AmB significantly diminished both spontaneous transformation and the response to the specific antigen SK-SD. These effects occurred at concentrations of drug achievable in humans (3). In contrast to the studies of Tarnvik and Ans6hn (19), we were unable to document an effect on the mitogenic response to PHA and ConA, even at concentrations as high as 10 ,ug/ml. This differential inhibitory effect on specific antigenic response but not on mitogenic response is similar to that reported after vaccination with live viral vaccines (8,17) and after treatment with corticosteroids (2, 7) and cytotoxic drugs (1). However, it is possible that this difference could be related to the fact that cultures stimulated with antigens were incubated longer (6 days) than those incubated with mi- epressed to 1-s as they were E response to z ly changed _ f AmB and A of stimulaalone or in e mitogenic o due wholy as transforincrease in F antifungal inated funisuppressed ies and also )verwhelmier suppresantifungal ie eventual multiplicity system in 17th ICAAC, Abstr. no 273). Others have shown AmB to be immunostimulatory to murine lymphocyte function both in vitro and in vivo (4,10,11). All studies showing AmrB to have immunostimulatory effects on lymphocytes have used murine cells; immunostimulation has not been documented using human cells. Perhaps this difference in test systems explains the differing results found between our study and those of others (4, 10, 11). 5-FC had no effect on in vitro lymphocyte responses, and the combination of AxnB and 5-FC showed no more suppression of lymphocyte transformation than that found with AmB alone.
Thus, in a well-defined and wholly in vitro system, AmB, but not 5-FC, diminished both spontaneous and antigen-induced lymphocyte transfornation. In this system, AmB selectively inhibited transformation to antigen and was not totally immunosuppressive. Whether this in vitro phenomenon has relevance to patients with serious fungal infections requiring AmB treatment is conjectural. Further in vivo studies in animal models of chronic antifungal therapy should be pursued to assess further the immunosuppressive potential of AmB.