Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

The sulfonamides are commonly used for the therapy of infections in transplantation and for AIDS-related complications (Weinshilboum, 1989; Gordin et al., 1990). The use of these agents is frequently associated with adverse drug reactions. The most severe complications are designated hypersensitivity reactions. In these adverse events, the involvement of pharmacogenetically determined pathways of drug activation and detoxication has provoked interest in the pathogenesis of these rare but potentially life-threatening reactions (Sullivan, 1989). Oxidative metabolism of sulfonamides has been identified as an important initial component of the adverse reactions (Cribb et al., 1995). In sensitive patients, it appears that a fraction of a sulfonamide dose undergoes oxidative metabolism, yielding reactive metabolites, the first of which is an hydroxylamine (SMX-HA; Shear and Spielberg, 1986; Rieder et al., 1988; Cribb and Spielberg, 1992; Cribb et al., 1995). In comparison with control volunteers there appears to be enhanced toxicity to PBML in patients with symptoms of adverse reactions to sulfonamides, a result that implicates the hydroxylamine (Rieder et al., 1989; Shear et al., 1986). The initial production of reactive metabolites is followed by propagation of the reaction, which appears to be mediated on an immunologic basis (Leeder et al., 1991; Rieder et al., 1992). The apparent involvement of the immune system in the propagation of these reactions has prompted interest in how these reactive metabolites are involved in sulfonamide-mediated immunomodulatory effects in vitro.

The effects of reactive sulfonamide metabolites on cellular and humoral immunity are not fully understood. At low to moderate doses, SMX-HA treatment does not affect immune cell viability; at high doses, however, SMX-HA decreases the viability of PBML cultures (Shear and Spielberg, 1986; Rieder et al., 1988). Leeder et al. have demonstrated that SMX-HA inhibits NK cell activity without having any significant affect on cell viability (Leeder et al., 1991). Recently we showed that in sublethal concentrations, SMX-HA profoundly inhibits mitogen-induced cellular proliferation (Rieder et al., 1992) via a mechanism(s) that remains unknown.

The microbial products CsA, FK506 and rapamycin are potent immunosuppressive agents (Bierer et al., 1991). The clinical introduction of CsA has significantly improved the outcome of organ and bone marrow transplantation, and FK506 has been used with encouraging results in clinical transplantation trials (Bierer, 1993). Much research, performed over the past two decades, has uncovered the molecular mechanisms of immunosuppression for these agents and has advanced our understanding of signal transmission pathways in lymphocyte activation (for review see Bierer et al., 1993; Schreiber and Crabtree, 1992; Metcalfe and Richards, 1990; Sigal and Dumont, 1992).

CsA and FK506 inhibit signal transduction pathways that are characterized by an initial rise in intracellular calcium via ligation and activation of the TCR (Bierer et al., 1993). CsA and FK506 exert their antiproliferative effects by interfering with early T-cell signal transduction pathways, inhibiting the expression of several T-cell-derived cytokines and cytokine receptors at the transcriptional level. Addition of either agent prevents the transcription of mRNA encoding IL-2, IL-3, IL-4, IFN-, TNF- and others without inhibiting lymphokine receptor activation (Kronke et al., 1984; Tocci et al., 1989; Emmel et al., 1989). Exogenous IL-2 is able to reverse the inhibiting effects of either agent, which suggests that the block in T-cell function is proximal to this step (Dumont et al., 1990). Both CsA and FK506 act by binding to their cytoplasmic immunophilin (cyclophilin for CsA and FKBP for FK506) (Schrieber and Crabtree, 1992). This in turn binds calcineurin (Fruman et al., 1992; Liu et al., 1991), resulting in inhibition of the nuclear translocation and binding of several nuclear factors to the promoter region(s) of cytokine genes (O'Keef et al., 1992; Brunvand et al., 1988; Hoyos et al., 1989).

The structural analog of FK506, rapamycin, was originally used for its antifungal and antibacterial properties, and its immunosuppressive activity was more recently discovered (Morris, 1992). Rapamycin does not inhibit the early events in lymphocyte signal transduction; rather, rapamycin inhibits proliferation via inhibition of the lymphokine-dependent signaling required for cell division (Bierer et al., 1991; Dumont et al., 1990; Kay et al., 1991). Rapamycin inhibits cell cycle progression in the late G₁ stage, whereas FK506 and CsA inhibit the G₀ to G₁ transition (Bierer et al., 1993). Rapamycin binds the same immunophilin as FK506, FKBP; however, rapamycin is unable to inhibit cytokine gene transcription (Henderson et al., 1991). Rapamycin/FKBP complex appears to exert its immunosuppressive effects by interfering with signaling through cytokine receptors targeting p70 S6 kinase (and perhaps other kinases) in inhibiting T-lymphocyte proliferation (Bierer et al., 1993; Terada et al., 1994).

The use of glucocorticoids, CsA, FK506 and rapamycin in controlling allograft rejection and autoimmune diseases is associated with numerous side effects (Messer and Rietman, 1983; Starzl et al., 1989). Combination (double, triple, quadruple) immunosuppressive therapy has proved especially effective in maximizing immunosuppression (Kimball et al., 1991; Kahan et al., 1991) while minimizing the side effects of each agent, because the doses used per agent are far less than those required to achieve comparable immunosuppression when used alone (Fung et al., 1991). Here we investigate the interaction between SMX-HA and CsA, FK506 or rapamycin in order to determine potential sites for the immunomodulatory effects of sulfonamide reactive metabolites. Partly because of the immunosuppressive action of reactive metabolites of sulfonamides and because of the antibacterial effects of the parent sulfonamide, it is also possible that derivatives of sulfonamide metabolites may have a place in future combination immunosuppression.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Reagents. SMX-HA was synthesized using the method of Rieder et al. (1988), dissolved in DMSO and diluted to the indicated concentration with HEPES-balanced salt solution, pH 7.4. CsA (Sandoz Ltd., Basel, Switzerland), FK506 (Fujisawa Pharmaceutical Co., Deerfield, IL) and rapamycin (Syntex, Palo Alto, CA) were prepared as 10²M stock solution in 70% ethanol and were diluted in RPMI-1640 medium (GIBCO-BRL, Grand Island, NY) containing 10% (v/v) normal human type AB serum (Whittaker, Bethesda, MD). PHA and PMA were purchased from Sigma Chemical Co. (St. Louis, MO). ³H-thymidine (79 Ci/mmol) was obtained from New England Nuclear (Boston, MA).

PBML preparation. Venous blood from healthy volunteers was diluted 1:2 in HEPES buffer, layered on histopaque density gradient (SG 1.077; Sigma) and centrifuged for 20 min at 2000 × g. The interphase containing PBML was washed three times in HEPES buffer and resuspended in culture medium at 10⁶ cells/ml. Cell viability was assessed by the trypan blue exclusion principle.

Assessment of SMX-HA effect. PBML were incubated with culture medium (positive control), DMSO (vehicle control) or SMX-HA at concentrations ranging from 0 to 100 µM for 2 hr at 37°C. The cells were then washed at least twice with HEPES buffer and were resuspended in complete medium (RPMI-1640 + 10% pooled human type AB serum). Viable cell concentration was adjusted to 5 × 10⁵ cells/ml.

PBML proliferation assays. PBML (5 × 10⁴ cells) were cultured in triplicate in 96-well flat-bottom microtiter plates (Costar Plastics, Boston, MA) for 3 days at 37°C in a 5% CO₂ humidified atmosphere. The cells were then treated with test drug or controls, stimulated with PHA (5 µg/ml) and PMA (5 ng/ml) and incubated for 72 hr at 37°C. ³H-thymidine (1 µCi/well) was added to the cells 4 hr before culture termination, and mitogen-induced proliferation (in counts per minute) was assessed by measuring the cellular uptake of tritiated thymidine by liquid scintillation. The stimulation index was calculated as follows: The median-effect analysis defined by Kahan et al. (1991), describing synergy or antagonism for drug combinations, was used to assess immunosuppressive drug interactions and was expressed as

(1)

where f_a represents the fraction of the system that is affected (suppression index/100) by the drug at dose D, and f_u represents the fraction of the system unaffected (1  f_a) by the drug at dose D. D_m is the EC₅₀. The linear regression coefficient, r, describes the sigmoidicity of the concentration-response curve. Logarithmic conversion of this equation linearizes the relationship:

(2)

where m is the slope of the plot and D_m is the x-intercept when f_a/f_u = 1 or log(f_a/f_u) = 0. The linear regression coefficient, r, describes the data fit to the median-effect analysis. In order to determine the single (D_x) or combination (D1,D2)_x drug concentrations required to achieve x% inhibition, the median-effect relationship was used; it was expressed as

(3)

The interaction of drugs in combination was characterized by the CRI_x for inhibition level x, which was calculated as

(4)

where EC_x alone is the concentration of the agent alone producing inhibition level x, and EC_x mixture is the concentration when the drug is combined with the test agent.

Treatment-induced toxicity. PBML preincubated with 25 µM SMX-HA were plated and incubated (72 hr) with serial dilutions of CsA, rapamycin or FK506 (10⁵ to 10¹² M) to test for combination-induced PBML toxicity. Incubation conditions and materials were identical to those in the proliferation experiments described previously. Viable cells after the various combination drug treatments were quantified by 2,7-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein (BCECF) labeling and measurement on a Baxter Fluorescence Concentration Analyzer. The percent cell death was calculated by the formula

Quantification of IL-2 mRNA transcription. Human PBML were treated with SMX-HA (0-50 µM) and appropriate controls and incubated with PMA/PHA for 18 hr as described above. Steady-state IL-2 mRNA expression was assessed by reverse transcription polymerase chain reaction (RT-PCR). Total cellular RNA was extracted from PBML under strict RNase-free conditions (Blumberg, 1987) by the guanidium isothiocyanate phenol/chloroform method as described (Chomczynski and Sacchi, 1987) using TRIzol reagent (Gibco-BRL, Grand Island, NY). RNA content and purity were verified by spectrophotometry and by electrophoresis on a 1% agarose 2.2 M formaldehyde gel stained with ethidium bromide (0.5 mg/ml). cDNA was incubated with Superscript II RNase H-reverse transcriptase, RNase inhibitor, oligo dT_12-18 primer, dNTP mixture and 2.5 mM MgCl₂ (Gibco BRL, Grand Island, NY) according to the manufacturer's specifications. cDNA was amplified for 30 cycles under optimized (2.5 mM MgCl₂, Taq Polymerase, dNTPs all from Gibco BRL, Grand Island, NY) PCR conditions (denaturation for 45 sec at 92°C, annealing at 60°C for 30 sec and extension for 90 sec at 70°C) in a Perkin Elmer thermal cycler (Toronto, Ontario, Canada). PCR primers sequences for IL-2 (Clark et al., 1984) and -actin (Ng et al., 1985) were based on cloned cDNA sequences (obtained from Genebank) and analyzed by Geneworks software on a power Macintosh 7200 computer (Apple Macintosh computers, Cupertino, CA). The primer sequences for IL-2 and -actin follow.

IL-2

-actin

Quantification of IL-2 protein production. IL-2 cytokine production was quantified by photometric enzyme immunoassay using the streptavidin-coated microtiter plates by ELISA (Boehringer Manheim, Missassagua, Ontario, Canada). Cytokine production was initiated by PMA/PHA stimulation of SMX-HA-treated (0-100 µM) human PBML plated in 24-well microtiter plates at 10⁶ cells/well, and supernatant was aliquoted after 36 hr of incubation. Data were analyzed using a Molecular Devices plate reader at 590 nm and Soft Max software for windows (Molecular Devices Corp., Menlo Park, CA).

Endothelial cell proliferation. The transformed endothelial cell culture line S5C4 provided by Dr. R. R. Shivers (University of Western Ontario, London, Ontario, Canada) was plated in triplicate on 96-well microtiter plates with SMX-HA (0-100 µM) and -MEM media supplemented with 10% FCS and penicillin/streptomycin at a concentration of 1000 cells/well. Cell proliferation was quantified by colormetric conversion of tetrazoleum salts (MTT) (Mosmann, 1983) as described for the ELISA assay.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Effect of SMX-HA on T-lymphocyte proliferation. To assess the suppressive capacity of SMX-HA, PBML were preincubated for 2 hr with SMX-HA at 0 to 100 µM, washed and stimulated with PHA-PMA. Data from figure 1A demonstrate that SMX-HA inhibited, in a concentration-dependent manner, T-cell proliferation induced by PHA-PMA, with a marked suppression observed in the low-µM range. Figure 1B illustrates the logarithmic conversion (equation 2; see "Materials and Methods") of the median-effect equation (equation 1) for SMX-HA. The x-intercept of the line at y = 1 defines the term D_m, or EC₅₀. The calculated D_m or EC₅₀ value for SMX-HA was 26.87 µM.

View larger version (9K): [in this window] [in a new window]   Fig. 1.   a) SMX-HA-induced suppression of T-lymphocytes stimulated by PMA and PHA. Open squares represent the response of PBML preincubated with SMX-HA for 2 hr. Data represent the mean of experiments performed in triplicate using PBML of 15 control volunteers. b) Linearization of the dose-effect relationship and EC50 values for the standard curve for SMX-HA by logarithmic conversion of the median-effect equation.

Interaction of SMX-HA with CsA, FK506 and rapamycin. To assess the interaction between SMX-HA and other immunosuppressive agents, we added CsA, FK506 and rapamycin, at 10¹² M to 10⁵ M, to PBML cultures that had been pretreated for 2 hr with 25 µM SMX-HA. The cultures were subsequently stimulated for 72 hr with PHA-PMA. As shown in figures 2A, 3A, and 4A, CsA and FK506 significantly augmented SMX-HA suppressive activity, as compared with rapamycin. Furthermore, SMX-HA pretreatment induced, in a concentration-dependent fashion, an upward shift in the concentration-response curves for the test drugs. The synergy between SMX-HA and CsA or FK506 was observed at 25 µM and higher concentrations of SMX-HA (P < .05). In contrast, SMX-HA at 25 µM did not alter the antiproliferative effects of rapamycin. The linearization of the data by logarithmic conversion (equation 2) of the median-effect equation (equation 1) for the concentration-response curves and the calculated EC₅₀ values for the individual drugs are shown in figures 2B, 3B and 4B, respectively. All drugs induced a concentration-dependent inhibition of PHA-PMA-induced PBML proliferation that obeyed the median-effect principle, with the linear regression coefficient r > 0.75. 

View larger version (14K): [in this window] [in a new window]   Fig. 2.   a) The effect of SMX-HA preincubation on T-lymphocyte suppression in combination with CsA. Closed squares represent the standard curve for CsA; open diamonds represent 12.5 µM SMX-HA preincubation; open circles represent 25 µM SMX-HA preincubation; open triangles represents 50 µM SMX-HA preincubation; open squares represent 100 µM SMX-HA preincubation. Data represent the mean ± S.E. of experiments performed in triplicate using the PBML of five control volunteers. b) Linearization of the dose-effect relationship and EC50 calculation for the standard curve for CSA by logarithmic conversion of the median-effect equation.

View larger version (0K): [in this window] [in a new window]   Fig. 3.   a) The effect of SMX-HA preincubation on T-lymphocyte suppression in combination with FK-506. Closed squares represent the standard curve for FK-506; open diamonds represent 12.5 µM SMX-HA preincubation; open circles represent 25 µM SMX-HA preincubation; open triangles represents 50 µM SMX-HA preincubation; open squares represent 100 µM SMX-HA preincubation. Data represent the mean ± S.E. of experiments performed in triplicate using the PBML of five control volunteers. b) Linearization of the dose-effect relationship and EC50 calculation for the standard curve for FK506 by logarithmic conversion of the median-effect equation.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   a) The effect of SMX-HA preincubation on T-lymphocyte suppression in combination with rapamycin. Closed squares represent the standard curve for rapamycin; open diamonds represent 12.5 µM SMX-HA preincubation; open circles represent 25 µM SMX-HA preincubation; open triangles represent 50 µM SMX-HA preincubation; open squares represent 100 µM SMX-HA preincubation. Data represent the mean ± S.E. of experiments performed in triplicate using the PBML of five control volunteers. b) Linearization of the dose-effect relationship and EC50 calculation for the standard curve for rapamycin by logarithmic conversion of the median-effect equation.

CRI₅₀ calculations. EC₅₀ was calculated as described above for all the curves derived from combination studies. CRI₅₀ was calculated from the EC₅₀ values by using equation 4. The EC₅₀ and CRI₅₀ values are shown in table 1. Combinations of CsA, FK506 or rapamycin with 50 µM and 100 µM SMX-HA were not included, because these concentrations were associated with significant cytotoxicity, whereas at 12.5 µM and 25 µM, SMX-HA did not cause any reduction in cell viability (Rieder et al. 1992). At the CRI₅₀ level, CsA and FK506 showed greatly enhanced dose reduction in the presence of both 12.5 and 25 µM SMX-HA concentrations. CsA with 25 µM SMX-HA showed a 46-fold reduction in CsA concentration in order to achieve 50% suppression when compared with CsA alone. Similarly, 25 µM SMX-HA displayed a 64-fold reduction in FK506 concentration. These values were obviously greater than the CRI_x level of 2 that implies an additive relation for a two-drug combination. However, concentration reduction was not observed when SMX-HA was combined with rapamycin.

Treatment-induced cell death. Figure 5 illustrates the effects of test drug concentrations on PBML viability when the drug was combined with 25 µM SMX-HA. Significant cell death was not observed for any sample at this level of SMX-HA in combination with any concentration of test drug employed in the proliferative assays.

View larger version (13K): [in this window] [in a new window]   Fig. 5.   The effects of 25 µM SMX-HA preincubation followed by treatment with test drug (CsA, rapamycin or FK506) on in vitro PBML toxicity. Open squares represent 25 µM SMX-HA in combination with CsA; open diamonds represent 25 µM SMX-HA in combination with FK506; open circles represent 25 µM SMX-HA in combination with rapamycin. Data represent the mean ± S.E. of experiments performed in triplicate using the PBML of five control volunteers.

Effect of SMX-HA on -actin and IL-2 mRNA expression. The effect of SMX-HA on IL-2 steady-state mRNA expression in PMA/PHA-stimulated PBML was assessed by RT-PCR using -actin as a control for cDNA sample integrity and RT-PCR amplification variability. RNA content and purity were confirmed by UV absorbance spectrophotometry and Northern blot analysis (data not shown). SMX-HA incubation concentrations were at sublethal concentrations (< 50 µM SMX-HA for PBML) (fig. 6) to keep cell death to a minimum. Cell viability at the time of mRNA extraction was greater than 95% for all unstimulated/SMX-HA-treated samples (10, 25, and 50 µM SMX-HA) compared with unstimulated/nontreated samples assessed by trypan blue viability (data not shown). SMX-HA treatment of PBML had no effect on the levels of -actin mRNA detected by RT-PCR. PMA/PHA stimulation also showed no effect on the expression of the housekeeping gene -actin (fig. 6). PMA/PHA activation of human PBML stimulated the accumulation of IL-2 protein or mRNA, as illustrated in figure 6. DMSO controls, equivalent to the level of DMSO incubated with up to 50 µM SMX-HA, had no effect on the levels of -actin and IL-2 mRNAs detected in PMA/PHA-stimulated samples (fig. 6). SMX-HA at immunosuppressive concentrations in PBML did not inhibit the steady-state mRNA expression of IL-2 mRNA in vitro (fig. 6).

View larger version (73K): [in this window] [in a new window]   Fig. 6.   The effect of SMX-HA on PMA/PHA-induced IL-2-transcription in vitro. Background samples (BKG) represent untreated, unstimulated PBML; positive control samples (PC) represent untreated, PMA/PHA-stimulated PBML; dimethylsulfoxide samples (DMSO) represent stimulated PBML pre-treated with drug carrier; SMX-HA samples (10, 25, 50) represent stimulated PBML pretreated with SMX-HA (10, 25, 50 µM). -actin and IL-2 mRNA production was visualized by RT-PCR.

Effect of SMX-HA on IL-2 protein production. IL-2 protein production by PMA/PHA-stimulated PBML was quantified by ELISA analysis; data are presented in figure 7. Sublethal, immunosuppressive concentrations of SMX-HA (25 µM) did not interfere with PBML accumulation of IL-2 protein in vitro compared with nontreated samples (fig. 7). IL-2 protein was detected at approximately 0 to 150 pm in unstimulated samples and at approximately 750 to 4800 pm in PMA/PHA-stimulated untreated controls. Optical density at 590/650 nm was converted to % production of IL-2 protein with untreated PMA/PHA samples adjusted to 100%.

View larger version (10K): [in this window] [in a new window]   Fig. 7.   The effect of SMX-HA on the production of PMA/PHA-stimulated IL-2 protein in vitro. Supernatant IL-2 protein levels for SMX-HA-pretreated cells (0-100 µM) was quantified on the PBML isolated from four volunteers.

Effect of SMX-HA on endothelial cell proliferation. SMX-HA displayed a dose-dependent inhibition of proliferation on endothelial cell line S5C4 by MTT assay (fig. 8). Suppression of proliferation was not achieved by the parent drug SMX (fig. 8).

View larger version (27K): [in this window] [in a new window]   Fig. 8.   The effect of SMX (100 and 400 µM), DMSO and SMX-HA (0-400 µM) on S5L4 endothelial cell proliferation in vitro. Open squares represent day 0; open triangles represent day 1; open circles represent day 2; asterisks represent day 3; closed squares represent day 4; closed triangles represent day 5; open circles represent day 6. Data represent the error of experiments performed in triplicate.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

We previously reported the suppressive and cytotoxic capacity of the hydroxylamine-reactive metabolite of sulfamethoxazole, SMX-HA, to inhibit mitogen-induced cellular proliferation (Rieder et al., 1992). Here we extend earlier findings by assessing the interaction between SMX-HA and CsA, FK506 or rapamycin. A 2-hr preincubation with SMX-HA was used throughout the experiments to reduce the possibility of metabolite-drug interactions. In addition, a 2-hr incubation in HEPES-buffered saline was used to reduce metabolite binding to proteins present in FCS-supplemented culture media. Experimental conditions were designed to maximize metabolite-cell contact, with excess SMX-HA washed from the system and bound or intracellular SMX-HA remaining. Previous cytotoxicity data from our laboratory suggest a reduced cytotoxic effect when SMX-HA remains throughout the culture (unpublished data). Here we have demonstrated that SMX-HA in the low-micromolar range produces synergistic immunosuppression when incubated with CsA and FK506. This synergy is not due to enhanced cytotoxicity, because the cell viability in cultures treated with SMX-HA and CsA or FK506 was not significantly different from that in control cultures (fig. 5). However, the suppression index was greater than that of individual agents used (table 1).

One of the objectives of this study was to speculate on the mechanism of action of SMX-HA as determined by its interaction with CsA, FK506 and rapamycin. In this case, the immunosuppressives were used as probes to provide information about where in the process of lymphocyte activation the site(s) of action were likely to be. The use of these agents as probe compounds in elucidating possible mechanism(s) of immunosuppressive effects has been described previously (Bierer et al., 1993).

The synergy observed between SMX-HA and either CsA or FK506 indicates different sites of action of the combined agents. CsA and FK506 affect early stages of T-cell activation through inhibition of calcineurin activity (Sigal and Dumont, 1992; Kronke et al., 1984) and, consequently, blockade of cytokine and cytokine receptor expression (Tocci et al., 1989; Emmel et al., 1989). The fact that SMX-HA produces synergestic immunosuppression with these agents suggests that the reactive metabolite affects late stages of T-cell activation. The site(s) at which this effect occurs remain unknown. By contrast, SMX-HA failed to modulate the anti-proliferative effects of rapamycin. This suggests that both agents act at a similar stage of lymphocyte activation, which would suggest an effect in the late stages of T-cell activation.

Both CsA and FK-506 effectively block the production of IL-2, a major excitatory cytokine responsible for the cellular proliferative response (Mosmann, 1989, Bierer et al., 1993). SMX-HA was unable to alter IL-2 mRNA production (fig. 6) and IL-2 protein production at sublethal concentrations (25 µM). These observations demonstrate that SMX-HA must act through an immunosuppressive mechanism(s) different from CsA and FK506 proximal to IL-2 production.

Potential target sites for the immunosuppressive effects of SMX-HA include interference with cytokine-mediated signaling or blockade of the transition from the G₁ to the S phase of the cell cycle. In addition, it remains unknown whether sulfonamide metabolite-induced immunosuppression occurs as a result of effects on specific cellular subset(s). Research is currently underway to elucidate the cellular target(s) and precise molecular site(s) for the immunosuppressive effects of SMX-HA.

Inhibition of cell proliferation by SMX-HA pretreatment was also observed on human endothelial cells (fig. 8). Thus lymphoid cells are not the only cell type affected by the toxic or antiproliferative effects of SMX-HA. This suggests that the reactive metabolite may irreversibly bind to membrane-bound or even intracellular macromolecules, inducing cytotoxicity at high concentrations and interference of proliferation at sublethal concentrations. The antiproliferative effects of SMX-HA on human endothelial cells are diminished at reduced metabolite concentrations (fig. 8), which suggests that the recovery after a 2 hr treatment of the compound is dependent on the concentration of SMX-HA. At 100 µM SMX-HA, the cytotoxic effect of the metabolite is overwhelming; at lower concentrations, on the other hand, endothelial cell recovery is observed (fig. 8). For this reason, it is difficult to separate the cytotoxic effects of the molecule from possible antiproliferative effects. However, in human PBML it is easier to segregate cytotoxic events from antiproliferative effects due to suppression observed at sublethal SMX-HA concentrations (Rieder et al., 1992).

A desired goal of any immunosuppressive drug combination is to maximize immunosuppression and minimize toxicity. In our study, cell viability was not reduced when SMX-HA was added at low concentrations (12.5 µM and 25 µM) to any of the tested immunosuppressants. Because of the significant cytotoxicity associated with higher concentrations of SMX-HA (Rieder et al., 1992), we did not include combinations of SMX-HA at 50 µM and 100 µM with CsA and FK506 in the present study. The synergy between SMX-HA and FK506 or CsA was observed at low and nontoxic concentrations of test drugs. Given the volume of distribution of sulfamethoxazole, the concentrations of SMX-HA detected in patient urine samples and the fact that 3% to 5% of a given dose appears to be converted to reactive metabolites in vivo, it is possible that patients treated with sulfonamides achieve SMX-HA plasma concentrations that approach 25 µM (Rieder et al., 1988). Thus it is possible that the antiproliferative effects of SMX-HA that we have demonstrated in vitro could also occur in vitro during sulfonamide treatment, especially during high-dose therapy prescribed for the management of infectious complications due to immunosuppression in transplantation recipients. It remains to be demonstrated whether SMX-HA plays a significant role in enhancing the immunosuppression of CsA- and FK506-treated individuals. Our data suggest that concurrent therapy with sulfonamides and immunosuppressive agents such as CsA or FK506 may enhance the antiproliferative effects of these agents. Taken together, these results point to the potential of developing derivatives of reactive sulfonamide metabolites that, in conjunction with immunosuppressives such as CsA, may enhance immunosuppression, thus allowing for a substantial reduction in immunosuppressant dosesa reduction that, in turn, may prevent or lower the risk of nonspecific toxicity that is often observed with monotherapy.

Given the fact that the sulfonamides are analogs of many environmental chemicals, these results may also enhance our understanding of the potential impact of low-level environmental exposure to reactive metabolites on immunity. Reactive metabolites of a number of compounds have been demonstrated to produce unpredictable effect on immunity (Rieder, 1992); Shear and Spielberg, 1988).

Current immunosuppressives can produce effective immunosuppression, but they often produce undesired toxicity. Our observations on synergistic immunosuppression produced by SMX-HA suggest that detailed study of the mechanism(s) of immunosuppression produced by reactive drug metabolites may be useful in the development of novel immunosuppressives. In addition, understanding how reactive metabolites affect specific targets in the immune system may be useful in the development of these compounds as probes for determining synergistic and antagonistic effects of novel compounds on immunity.