Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

A novel form of immune intervention with potential for the treatment of autoimmune diseases has recently emerged. This new approach to immunotherapy is mediated by the G₁ blocker butyric acid. Butyric acid is a short-chain fatty acid derived from fiber polysaccharides that are metabolized by anaerobic bacteria in the gut. n-Butyrate has been primarily studied as a antineoplastic agent with the ability to inhibit proliferation and induce differentiation in a variety of transformed cell types in vivo (for a review, see Kruh et al., 1992). In addition to its effects on tumor cells, n-butyrate has been shown to induce antigen-specific unresponsiveness in CD4⁺ T cells in vivo (Gilbert and Weigle, 1993). The n-butyrate-induced T cell unresponsiveness required antigen stimulation and was not induced in T cells exposed to n-butyrate alone. The antigen specificity of the n-butyrate-induced T cell tolerance makes butyrate a good candidate for an immunotherapeutic agent.

Unfortunately, the immunotherapeutic potential for n-butyrate is limited by its short half-life in vivo (3-6 min) (Daniel et al., 1989). Even when it was administered by i.v. infusion, n-butyrate was found to be clinically ineffective as an anticancer agent (Novogrodsky et al., 1983; Miller et al., 1987).

To improve the therapeutic potential of butyrate, we synthesized the hydrochloride salts of three derivatives of butyric acid: 2-(4-morpholinyl)ethyl ester (MEB), 2-(4-morpholinyl)ethyl amide (MEBA), and the bisbutyryl ester/amide derivative of 1-piperidineethanol (BEB) (Fig. 1). The ester and amide functional groups of these compounds should undergo hydrolysis in vivo to release butyrate. This should lead to a more sustained release of butyric acid and significantly prolong the duration of action. Another advantage of these butyrate derivatives is that each contains an ionizable amino group. This not only allows the compound to be converted to water-soluble salts (e.g., hydrochloride) but also avoids the necessity of using the sodium salt of butyric acid, which could lead to sodium overload. In this study, we tested whether MEB, MEBA, and BEB could inhibit T cell activity in vivo, and we characterized the effects of MEB on T cell activity in vitro.

View larger version (9K): [in this window] [in a new window]   Fig. 1.   Molecular structure of butyric acid derivatives.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

Male C57BL/10, DBA/2, and C3H/HeJ mice at 6 to 8 weeks of age were purchased from Harlan Sprague-Dawley, Inc. (Indianapolis, IN).

Reagents and Antibodies

Imject keyhole limpet hemocyanin (KLH) was purchased from Pierce Chemical Co. (Rockford, IL). n-Butyrate was purchased from Sigma Chemical Co. (St. Louis, MO). Butyryl chloride, 4-(2-hydroxyethyl)morpholine, 4-(2-aminoethyl)morpholine, and 1-(2-hydroxyethyl)piperazine were purchased from Aldrich Chemical Co. (Milwaukee, WI).

Instrumentation

Proton NMR spectra were recorded at 500 MHz on a Bruker AM500 spectrometer and chemical shifts are reported in ppm (). Mass spectra were recorded on a Finnegan TSQ 700 spectrometer (direct exposure probe) at 70 eV electron ionization.

Butyric Derivatives

The following derivatives were used (Fig. 1).

MEB. Butyryl chloride (6.07 g, 0.06 mol) was added with stirring to a cooled solution of 5 g (0.04 mol) of 4-(2-hydroxyethyl)morpholine in 20 ml of chloroform over 45 min and cooling was maintained for 6 h. The mixture was diluted with chloroform (15 ml) and washed three times with 20 ml of 5% sodium carbonate. The aqueous layer was washed with 15 ml of chloroform, and the combined organic layers were dried over anhydrous sodium sulfate, filtered, and concentrated in vacuo to afford 5.37 g (70%) of the ester as an orange liquid. Column chromatography of 500 mg on silica gel 60 (0.063-0.200 mm, 10-inch, 30-mm inside diameter) was performed using a gradient of ethyl acetate/hexane 1:3 to ethyl acetate. The recovered sample afforded the following: ¹H NMR (CDCl₃)  0.93 (t, 3 H, J = 7.4, CH₃), 1.63 (m, 2 H, J = 7.4, CH₃CH₂-), 2.28 (t, 2 H, J = 7.4, COCH₂-), 2.58 (ap bs, 4 H, ring N-CH₂-), 2.69 (t, 2 H, J = 5.7, CO₂CH₂-), 3.74 (t, 4 H, J = 4.5, ring -OCH₂-), 4.25 (t, 2 H, J = 5.7, NCH₂-); MS m/z 201 (M⁺), 130, 113, 100 (base peak). For 1-(4-morpholinyl)ethyl butyrate hydrochloride, a solution of 2.5 g of the crude free base in 60 ml of anhydrous ether was cooled in an ice bath with stirring while 14 ml of a cold 1.0 M solution of hydrogen chloride in anhydrous ether was added dropwise. The resulting white solid was filtered and recrystallized from 60 ml of tetrahydrofuran to yield 1.69 g of white crystals: m.p. 108.0-108.3°C.

MEBA. The amide was synthesized in 68% yield in a manner analogous to that described for MEB by treatment of 4-(2-aminoethyl)morpholine with butyryl chloride. ¹H NMR (CDCl₃)  0.91 (t, 3 H, J = 7.4, CH₃), 1.62 (m, 2 H, CH₃CH₂-), 2.13 (t, 2 H, J = 7.4, COCH₂-), 2.40 to 2.45 (m, 6 H, N-CH₂- and ring O-CH₂-), 3.30 to 3.33 (m, 2 H, N-CH₂-), 3.66 (t, 4 H, ring N-CH₂-), 5.96 (bs, 1 H, NH); MS m/z 200 (M⁺), 182, 157, 113, 100 (base peak). For 2-(4-morpholinyl)ethyl butanamide hydrochloride, the hydrochloride salt was prepared as described for MEB and recrystallized from tetrahydrofuran to afford hygroscopic crystals: m.p. 150.6-151.1°C.

BEB. Butyryl chloride (11.83 g, 0.11 mol) was added dropwise to a cooled solution of 4.84 g (0.04 mol) of 1-(2-hydroxyethyl)-piperazine in 20 ml of chloroform. Cooling was maintained for 2 h, and a white precipitate formed. Chloroform (15 ml) was added, and the mixture was stirred overnight, washed with 210 ml of cold 0.6 N sodium hydroxide solution, and then washed twice with 50 ml of cold water. The organic fraction was dried over anhydrous sodium sulfate, filtered, and concentrated in vacuo to yield 9.39 g (94%) of a clear yellow liquid. This was distilled: b.p. 144-146°C, (0.5 mm Hg) to give 6.08 g (65%) of MEB. ¹H NMR (CDCl₃)  0.897 (t, 3 H, J = 7.4, CH₃), 0.901 (t, 3 H, J = 7.4, CH₃), 1.59 (m, 2 H, CH₃CH₂), 1.61 (m, 2 H, CH₃CH₂), 2.24 (t, 2 H, J = 7.4, CH₂CO), 2.25 (t, 2 H, J = 7.4, CH₂CO), 2.42-2.43 (app t, 2 H, J = 5.2, axial CH ester end of ring), 2.43 to 2.44 (app t, 2 H, J = 5.2, equatorial CH ester end of ring), 2.59 (t, 2 H, J = 5.8, N-CH₂-CH₂-O), 3.41 (app t, 2 H, axial CH amide end of ring), 3.57 (app t, 2 H, equatorial CH amide end of ring), 4.16 (t, 2 H, N-CH₂-CH₂-O); MS m/z 270 (M⁺), 255, 242, 227, 199, 182, and 169 (base peak). For 2-(4-butanoylpiperazinyl)ethyl butanoate hydrochloride, the hydrochloride salt was prepared as described for MEB and recrystallized twice from tetrahydrofuran to afford white crystals: m.p. 130.1-130.7°C.

Th1 Cell Clones

The KLH-specific Th1 cell clones were developed in C57BL/10 mice and characterized as Th1 cell clones based on their ability to secrete interleukin (IL)-2 but not IL-4. The Th1 clones were passed every 7 to 14 days using KLH, irradiated syngeneic spleen cells as antigen-presenting cells (APCs), and IL-2-containing conditioned medium from rat spleen cells stimulated with concanavalin A (ConA CM) using a previously described protocol (Gilbert et al., 1990).

Measuring Effects of Butyrate Derivatives on Antibody Production In Vivo

C57BL/10 mice (five mice/group) were injected i.p. with 100 µg of ovalbumin in conjunction with complete Freund's adjuvant on day 0. In one experiment, the mice also received one i.p. injection/day of the butyrate derivatives (0.091 mmol) on days 1 to 3. Serum samples were obtained 10 days after the initial injection with ovalbumin and tested for the presence of anti-ovalbumin antibodies using an enzyme-linked immunosorbent assay (ELISA). To perform the ELISA, 96-well plates (Costar 3595) were first incubated with ovalbumin (100 µl/well of 100 µg/ml in PBS) overnight at 4°C. The plates were then washed four times with PBS and 0.5% Tween 20, blocked with 1% fetal calf serum for 30 min at 37°C, and washed again. Individual serum samples were added (diluted 1:100 or 1:1000 in PBS), and the plates were incubated for 2 h at 20°C. The plates were next washed seven times with PBS/Tween, and alkaline phosphate (AP)-labeled goat anti-mouse IgG, IgA, and IgM (heavy + light) (Zymed Laboratories, South San Francisco, CA) was added (1:1000) for 1 h at 20°C. The plates were again washed seven times with PBS/Tween, and AP substrate (1 mg/ml) was added. After 10 min, immunoglobulin levels were quantified by an ELISA reader (absorbance, 405 nm). The concentration of anti-ovalbumin was determined by comparison with a standard curve obtained using mouse anti-ovalbumin antibody (Sigma Chemical). To measure isotype-specific anti-ovalbumin antibodies, serum samples (diluted at 1:300 or 1:1000) were incubated on the ovalbumin-coated plates as described earlier. After washing, biotinylated detecting antibodies directed against mouse IgG_2a (rat IgG₁, clone R19-15), IgG_2b (rat IgG_2a, clone R12-3), IgG₁ (rat IgG₁, clone A85-1), or IgM (rat IgG_2a, clone R6-60.2) (all purchased from PharMingen, La Jolla, CA) were added at 2.5 µg/ml for 1 h at 20°C followed by AP-labeled ExtrAvidin (Sigma) for 1 h at 20°C and AP substrate. Immunoglobulin levels were presented as absorbance measurements.

In a second set of experiments, male C57BL/10 mice (five mice/group) were injected i.p. with 100 µg of ovalbumin in conjunction with complete Freund's adjuvant on day 0, followed by a single i.p. injection of saline or MEB (0.15 mmol) on day 2 or 3. On day 10, the mice received 100 µg of ovalbumin in conjunction with incomplete Freund's adjuvant s.c. at the base of the tail. After an additional 7 days, cells from the periaortic and mesenteric lymph nodes were enriched for CD4⁺ T cells by negative selection as described by Griffin et al. (2000). The T cells were then incubated at 1 × 10⁵ in half-area (100 µl/well) 96-well plates (Costar 3696) along with 2 × 10⁵ irradiated (2000R) spleen cells from untreated C57BL/10 mice as APCs and various concentrations of ovalbumin. Proliferation was measured on day 4 by assessing incorporation of [³H]thymidine deoxyribose ([³H]TdR) after a 12-h pulse.

Examining Effects of Butyrate Derivatives on T Cells In Vitro

Th1 cells (5 × 10⁴ cells/well in 96-well plates; Costar) were stimulated with 10% IL-2-containing ConA CM in the presence of various concentrations of different butyrate derivatives. In some cultures, Th1 cell proliferation was measured after 2 days. In other cultures, the Th1 cells were washed after 24 h, and fresh IL-2-containing medium was added. Proliferation was measured in the washed, IL-2-restimulated cultures after an additional 2 days.

The ability of the butyrate derivative to induce Th1 cell anergy was examined using a protocol previously developed for n-butyrate-induced T cell tolerance (Gilbert and Weigle, 1993). Briefly, Th1 cells were incubated in primary cultures at 2.5 × 10⁵ cells/ml, along with MEB (1 mM), KLH (50 µg/ml), and 5 × 10⁶/ml irradiated syngeneic spleen cells as APC. Alternatively, the Th1 cells were incubated in primary cultures containing MEB and IL-2 (10% ConA CM). Control primary cultures received MEB and APCs but no antigen or IL-2. After incubation for 24 h at 37°C, the cells in the primary cultures were harvested, washed free of MEB, and reincubated at 2.5 × 10⁵ml in secondary cultures without MEB. The Th1 cells in the secondary cultures were stimulated with 10% IL-2-containing ConA CM, or with 5 × 10⁶/ml irradiated syngeneic spleen cells as APCs, and KLH. After 2 days in the secondary cultures, the Th1 cells were assessed for proliferation (pulsed with [³H]TdR for 12 h).

The ability of MEB to induce alloantigen-specific T cell unresponsiveness was tested by incubating spleen cells from DBA/2 mice (H-2^d) (2.5 × 10⁵ in 200 µl/wells) with stimulator cells [2.5 × 10⁵ irradiated (2000R) spleen cells from C57BL/10 mice (H-2^b)]. MEB (1 mM) was added to some wells of the mixed lymphocyte reaction (MLR) 24 h after the initiation of culture. After an additional 3 days, the MLR cultures were washed and rested for an additional 2 days. The T cells from the MLR were then isolated and reincubated at 2.5 × 10⁵/well with either the initial alloantigen (spleen cells from C57BL/10 mice) or with a third-party alloantigen [spleen cells from C3H/HeJ mice (H-2^k)]. Proliferation in both the primary and secondary MLR was measured on day 5 by assessing incorporation of [³H]TdR after a 12-h pulse.

DNA Analysis. To examine DNA content, Th1 cells were fixed in prechilled 70% ethanol overnight at 4°C. The fixed Th1 cells were next washed in PBS, resuspended in 1 ml of staining buffer containing RNase (1 mg/ml; Sigma Chemical Co.) and propidium iodide (50 µg/ml; Sigma Chemical Co.), incubated for 20 min in the dark at 20°C, and analyzed by flow cytometry using a FACScalibur (Becton Dickinson, Mountain View, CA). The data were analyzed using the ModFit DNA analysis program (Verity Software House).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Butyrate Derivatives Reversibly Inhibited IL-2-Induced Proliferation of Th1 Cells. To test the ability of the butyrate derivatives to suppress T cell activity in vivo, IL-2-stimulated Th1 cells were incubated in the presence of n-butyrate or the butyrate derivatives. Some cultures of treated Th1 cells were washed after 24 h and restimulated with IL-2 without butyrate derivatives. As shown in Fig. 2, the ester and ester/amide derivatives of n-butyrate, MEB and BEB, respectively, were comparable to n-butyrate in their ability suppress IL-2-induced proliferation of Th1 cells. In addition, similar to n-butyrate-treated T cells, T cells treated with butyrate derivatives regained their ability to proliferate to IL-2 once the compounds were washed out of the cultures. This latter observation means that the cell cycle-blocking effects of the MEB and BEB were not due to drug-induced toxicity.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Butyrate derivatives inhibit proliferation of Th1 cells. KLH-specific Th1 cells (clone D3) were stimulated with IL-2 (10% ConA CM) in the presence of different concentrations of n-butyrate () or butyrate derivatives (, MEB; , MEBA; , BEB). In some cultures, Th1 cell proliferation was measured after 2 days. In other cultures, the inhibitors were washed out after 24 h, and the Th1 cells were reincubated with fresh IL-2-containing medium. Proliferation of the Th1 cells restimulated with IL-2 in the absence of the inhibitors was measured after an additional 2 days. This experiment was repeated three times with very similar results.

Butyrate Derivatives Inhibited Antibody Production to Thymus-Dependent Antigen In Vivo. The butyrate derivatives were next tested for their ability to suppress lymphocyte activity in vivo. In this experiment C57BL/10 mice were injected i.p. on day 0 with thymus-dependent antigen ovalbumin followed by i.p. injections of either MEB, MEBA, or BEB on days 1 to 3. When anti-ovalbumin levels in the serum of the mice were tested 10 days after administration of the antigen, MEB was shown to significantly decrease by 65% the ability of the mice to generate a primary antibody response to a thymus-dependent antigen compared with control mice treated with saline (Fig. 3). Mice treated with a second butyrate derivative, BEB, also produced significantly less antigen-specific antibody than control mice. In contrast, the third butyrate derivative, MEBA, was unable to suppress antigen-specific antibody production in vivo. These results show that the ester and the ester/amide derivatives of butyrate suppressed lymphocyte function both in vitro and in vivo, whereas the amide analog of butyrate was ineffective both in vitro and in vivo.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   MEB inhibits primary antigen-specific antibody production in vivo. C57BL/10 mice were injected i.p. on day 0 with ovalbumin, followed by i.p. injections of saline () or one of three different butyrate derivatives (, MEBA; , BEB; , MEB) on days 1 to 3. Levels of total anti-ovalbumin antibody in the serum 10 days after antigen administration were measured. *, significantly different from response of mice treated with saline, P < .05.

Butyrate Derivatives Inhibited Antigen-Specific T Cell Responses In Vivo. If the butyrate derivatives suppressed antibody production to a thymus-dependent antigen by inactivating the antigen-specific CD4⁺ T cells required for B cell help, then theoretically the butyrate derivatives need be present only during an early stage, during which the CD4⁺ T cells would otherwise be activated by antigen. To test the possibility that short-term exposure to butyrate derivatives could alter the T cell response to antigen in vivo, a second experiment was conducted in which C57BL/10 mice were injected i.p. with ovalbumin on day 0, followed by a single i.p. injection of saline or MEB on day 2 or 3. The mice were reimmunized with ovalbumin s.c. on day 10. In this study, isotype-specific anti-ovalbumin antibody was measured to more precisely delineate the effect of MEB on antigen-specific antibody production. In addition, to look more directly on the effect of MEB on CD4⁺ T cells, lymph node CD4⁺ T cells isolated from mice 6 days after the second immunization with ovalbumin were examined for their ability to proliferate to ovalbumin in vitro. Treatment with a single dose of MEB significantly decreased the production of IgG_2a and IgG_2b anti-ovalbumin antibody during the primary antibody response (Fig. 4A). IgG₁ and IgM anti-ovalbumin antibody production were also decreased, albeit not dramatically, if MEB was administered on day 2, but not on day 3, after immunization. An evaluation of the antibody response generated by a second exposure to ovalbumin revealed that IgG_2a and IgG_2b anti-ovalbumin antibody remained dramatically low in mice treated with MEB on either day 2 or day 3 after their initial immunization with ovalbumin (Fig. 4B). IgG1 anti-ovalbumin production during the secondary antibody response was also significantly decreased, whereas the IgM anti-ovalbumin antibody production after reimmunization with ovalbumin was unaffected by the initial treatment with MEB. The MEB-induced decrease in IgG antigen-specific antibody production correlated with a significant loss of antigen-specific proliferation observed in the CD4⁺ T cells isolated from antigen-primed mice treated with MEB on day 2 or day 3 after immunization (Fig. 4C). Taken together, it appears that even a brief exposure to the butyrate derivative MEB in vivo can induce antigen-specific unresponsiveness in CD4⁺ T cells.

View larger version (28K): [in this window] [in a new window]   Fig. 4.   MEB induces antigen-specific T cell inactivation in vivo. C57BL/10 mice were injected i.p. with ovalbumin/CFA on day 0, followed by a single i.p. injection of saline or MEB (0.15 mmol) on day 2 or 3. The mice were reimmunized with ovalbumin s.c. on day 10. Isotype-specific anti-ovalbumin antibody generated during the primary immune response (A) and the secondary immune response (B) were measured in the serum at day 10 and day 16, respectively. C, lymph node CD4+ T cells isolated from the mice 6 days after the second immunization with ovalbumin were stimulated with ovalbumin in vitro and examined for proliferation. , saline; , MEB day 2; , MEB day 3. *, significantly different from response of CD4+ T cells isolated from mice treated with saline, P < .05. This experiment was repeated, and the proliferation data represent the mean values of the two experiments.

Butyrate Derivative Induced Antigen-Specific Inactivation in CD4⁺ T Cells In Vitro. Experiments were conducted in vitro to confirm the ability of MEB to induce antigen-specific unresponsiveness in CD4⁺ T cells. This confirmation included testing whether antigen was required for MEB-induced T cell anergy. Th1 cells were treated with MEB in the presence or absence of antigen or exogenous IL-2. The Th1 cells were then removed from the primary cultures, washed free of MEB, and restimulated with antigen or IL-2 in secondary cultures; tolerized Th1 cells are characterized by the fact that although they lose their ability to proliferate when restimulated with antigen, their continued expression of IL-2 receptors enables them to proliferate when stimulated with exogenous IL-2. The results of Fig. 2 had shown that Th1 cells treated with IL-2 and MEB, although blocked in primary cultures, retained their ability to proliferate in response to IL-2 once the MEB had been washed from the cultures. Here we show that Th1 cells pretreated with IL-2 and MEB also retained their ability to proliferate to antigen once MEB had been washed from the cultures (Fig. 5A). In contrast to Th1 cells pretreated with IL-2 and MEB, Th1 cells pretreated with antigen and MEB lost their ability to proliferate in antigen-stimulated secondary cultures (Fig. 5B). The fact that the Th1 cells pretreated with antigen and MEB, although unable to respond to antigen, could still proliferate in secondary cultures stimulated with exogenous IL-2 suggested that the lack of antigen responsiveness in these Th1 cells was not due to a loss of viability. Th1 cells incubated in primary cultures with MEB alone or in medium alone retained their ability to proliferate in response to antigen stimulation in secondary cultures. This result showed that antigen-activated, but not IL-2-activated, Th1 cells became unresponsive to a subsequent stimulation with antigen after exposure to MEB.

View larger version (13K): [in this window] [in a new window]   Fig. 5.   MEB induces antigen-specific T cell inactivation in vitro. KLH-specific Th1 cells (clone D3) were incubated in primary cultures with (A) MEB (1 mM) and/or IL-2 (10% ConA CM) or (B) MEB (1 mM) and/or KLH (50 µg/ml). After 2 days, the Th1 cells were isolated from the primary cultures, rested, and then reincubated in secondary cultures stimulated with antigen (KLH, 50 µg/ml) or IL-2 (10% ConA CM). Th1 cell proliferation in the secondary cultures was measured. *, significantly different from response of Th1 cells exposed to MEB alone, P < .01.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   MEB-induced T cell unresponsiveness is antigen-specific. A, spleen cells from DBA/2 mice (H-2d) were stimulated in a primary MLR with irradiated spleen cells from C57BL/10 mice (H-2b) in the presence () or absence () of MEB (1 mM). As a negative control, some wells were stimulated with spleen cells from DBA/2 mice (). Proliferation was measured on day 5. B, after 3 days, some wells in the primary MLR cultures were washed and rested for an additional 2 days. The spleen cells that had been stimulated in the primary MLR with spleen cells from C57BL/10 mice in the presence () or absence () of MEB were then isolated and reincubated with either the original alloantigen (spleen cells from C57BL/10 mice), a third-party alloantigen [spleen cells from C3H/HeJ mice (H-2k)], or IL-2 (ConA CM). Proliferation was measured on day 5 of the secondary MLR. *, significantly different from response of spleen cells not exposed in the primary MLR to MEB, P < .01.

Butyrate Derivative Blocked Activated Th1 Cells in G₁. Because MEB appeared to suppress antigen-specific T cell responses both in vivo and in vitro, further characterization of its mechanism of action was conducted. n-Butyrate-induced T cell tolerance has been linked to the ability of the compound to block cell cycle progression of T cells in G₁ (Gilbert and Weigle, 1993). Although it was shown that MEB, as well as other butyrate derivatives, inhibited the proliferation of activated Th1 cells, it was not known where in the cell cycle this inhibition occurred. An analysis of DNA content showed that similar to n-butyrate, essentially all the Th1 cells stimulated with IL-2 in the presence of MEB remained in G₀/G₁ (Fig. 7).

View larger version (12K): [in this window] [in a new window]   Fig. 7.   MEB blocks Th1 cell cycle progression in G0/G1. KLH-specific Th1 cells (clone C3) were left unstimulated (resting) or were stimulated with IL-2 (10% ConA CM) in the presence or absence of MEB (1 mM). The Th1 cells were collected after 48 h and assayed for DNA content.

View larger version (13K): [in this window] [in a new window]   Fig. 8.   MEB sequesters stimulated Th1 cells in G0/G1. KLH-specific Th1 cells (clone C3) were stimulated with IL-2 (10% ConA CM) at the initiation of culture and then again at 24, 48, and 96 h. Some cultures also received a single dose of MEB (1 mM) at times 0, 15, 24, or 39 h after the initiation of culture. The Th1 cells were collected at 0, 48, 72, 96, or 120 h after the initiation of culture and assayed for DNA content. , G0/G1; , S; , G2/M.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

n-Butyrate derivatives designed to possess G₁ blocker activity both in vitro and in vivo were synthesized. The ester (MEB) and ester/amide (BEB) derivatives of butyrate were found to suppress IL-2-stimulated proliferation of Th1 cells in vitro. Unlike MEB and BEB, the amide analog of butyrate, MEBA, did not suppress Th1 cell proliferation in vitro. The lack of activity of MEBA may be related to the slower metabolic hydrolysis of the amide bond in MEBA compared with the ester bond in MEB and BEB. When tested in vivo, both MEB and BEB, but not MEBA, were shown to significantly suppress a primary antibody response to a thymus-dependent antigen. Suppression of antibody production could reflect inhibition of T and/or B cell function. However, subsequent in vivo experiments to more specifically examine the effect of MEB on T cell activity revealed that MEB induced antigen-specific unresponsiveness in CD4⁺ T cells. The T cell unresponsiveness induced in mice immunized with ovalbumin and treated with MEB was manifested as an inability of lymph node CD4⁺ T cells to proliferate when stimulated with ovalbumin in vitro. Although this finding does not negate the possibility that MEB may also inactivate antigen-activated B cells, it clearly demonstrates that ester analogues of butyrate can induce unresponsiveness in antigen-specific CD4⁺ T cells.

MEB was able to inactivate CD4⁺ T cells in vivo even if administered in a single dose. If MEB works in vivo as it does in vitro (i.e., by inducing anergy in antigen-stimulated T cells), it would be necessary for MEB to be present only during the narrow window of time when T cell stimulation by antigen occurs in vivo. Because T cell activation in draining lymph nodes has been shown to occur 2 to 3 days after immunization of naïve mice (MacLennan et al., 1997; Garside et al., 1998), MEB was administered in a single dose on either day 2 or day 3 after administration of the antigen, in this case ovalbumin. Both primary and secondary anti-ovalbumin antibody production was inhibited in the MEB-treated mice. However, not all of the isotypes of anti-ovalbumin antibody were suppressed equally. IgM anti-ovalbumin was slightly inhibited during the primary antibody response and was totally unaffected during the secondary antibody response. Because the requirement for antigen-specific T cell help during the production of IgM is less stringent than that needed for the production of IgG subclasses of Ig (Steele et al., 1996), this finding would suggest that MEB is better at suppressing specific T cell responses than it is at inhibiting nonspecific T cell help or B cell activity.

Although a single dose of MEB had little effect on antigen-specific IgM production, this treatment regimen was shown to suppress primary IgG_2a and IgG_2b anti-ovalbumin antibody production and to block the generation of the memory T cells required for a secondary IgG_2a or IgG_2b anti-ovalbumin antibody response. MEB also decreased the generation of memory T cells required for a secondary IgG₁ antibody response. However, the effect of MEB treatment on IgG₁ antibody production was less profound than the effect of MEB on IgG_2a or IgG_2b. IgG₁ production has been shown to be dependent on IL-4, and thus largely driven by Th2 cells, whereas IgG_2a production is enhanced by interferon-, and thus driven by Th1 cells (Stevens et al., 1988). The relationship between IgG_2b and a particular CD4⁺ T cell subset is less well-defined, but because IL-4 has been shown to suppress IgG_2b (Kuhn et al., 1991), it is not unlikely that Th1 cells rather than Th2 cells promote IgG_2b production in vivo. Consequently, it is possible to interpret the differential effect of MEB on isotype-specific antibody production in vivo by postulating that Th1 cells are more susceptible than Th2 cells to MEB-induced unresponsiveness. Certainly the results showing that MEB induced antigen-specific unresponsiveness in Th1 cells in vitro underscore the likelihood that Th1 cells, both in vitro and in vivo, are susceptible to MEB-induced tolerance. The suggestion that Th2 cells are less susceptible than Th1 cells to MEB-induced unresponsiveness is in accordance with other methods of inducing T cell tolerance, which have similarly documented the relative resistance of Th2 cells to tolerance induction (Gilbert et al., 1990; Williams et al., 1990). The fact that the memory IgG₁ antibody response was suppressed to some degree in mice treated with MEB suggests that although Th2 cells may be somewhat resistant to MEB-induced unresponsiveness, somewhat longer exposure to MEB or perhaps higher doses of MEB may be expected to more completely suppress Th2-mediated IgG₁ production.

MEB-induced T cell unresponsiveness was not generalized but was reserved for T cells that are simultaneously stimulated with antigen. Unlike Th1 cells exposed to both antigen and MEB in vitro, Th1 cells exposed to MEB alone or to MEB and IL-2 did not lose their ability to respond to a subsequent antigen challenge. Along these same lines, splenic T cells stimulated in vitro with an alloantigen in the presence of MEB lost their ability to proliferate in response to a subsequent challenge with the initial alloantigen but retained their ability to proliferate when stimulated with a third-party alloantigen. Taken together, these results underscore the antigen specificity of MEB-induced T cell unresponsiveness.

No toxicity was observed in this study when mice were treated with MEB at a dosage that approximated 0.7 g/kg/day. Even if administered at high doses for extended periods of time, it seems unlikely that MEB would be toxic. Concerns about possible sodium overload and lack of efficacy have precluded studies examining the potential toxicity of high doses of n-butyrate. However, the arginine salt of butyrate was shown to be nontoxic in humans even when perfused at doses as high as 2 g/kg/day (Perrine et al., 1994). In any case, the efficacy of a single dose of MEB suggests that short-term use of the compound would be effective, thus eliminating any possible toxicity associated with long-term use.

MEB was shown to induce eventual G₁ sequestration of activated Th1 cells regardless of when it was added during the cell cycle. This finding suggests that MEB will be effective in treating an ongoing T cell response, a valuable characteristic of an immunotherapeutic agent. Many methods of inducing antigen-specific T cell unresponsiveness for the treatment of autoimmunity are very useful in preventing the initiation of the disease process but are much less effective in treating an already established autoimmune response (Meyer et al., 1996; Gaupp et al., 1997; Bai et al., 1998). In addition, the use of butyrate derivatives to treat autoimmune disease does not require identification of the specific autoantigens targeted by the self-reactive lymphocytes. Theoretically, the butyrate derivatives would inactivate any CD4⁺ T cell that was simultaneously being stimulated with antigen, thus encompassing all autoreactive CD4⁺ T cells activated in response to any self-antigens. Taken together, the results presented suggest that short-term use of butyrate derivatives can be used in vivo to induce antigen-specific inactivation of at least the Th1 cell-like subset of CD4⁺ T cells, thus providing the basis for a novel method of immune intervention with potential for the treatment of autoimmune disease. Such a treatment regimen would have definite advantages over most existing immunotherapies, which consist of the long-term use of drugs that induce generalized immune suppression and may produce significant clinical side effects.