Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

IL-2 exerts multiple biological functions by binding to high-affinity receptors composed of , , and common  chain subunits (Rusterholz et al., 1999). Although undetectable on resting T cells, the  chain expression is triggered by antigen, a stimulus that can be mimicked by concanavalin A (Con A) or anti-T cell receptor (TCR) antibodies. Ligation of interleukin-2 receptor (IL-2R) with IL-2 triggers Ras-mitogen-activated protein kinase pathways, which are important for T cell proliferation (Reif et al., 1997) and Janus kinase-signal transducer and activator of transcription (Jak-STAT) pathway, which is involved in gene expression (Beadling et al., 1996). IL-2/IL-2R signaling is also important for T cell differentiation into effector T cells. IL-2 and IL-12 can stimulate mitogen-activated type 1 helper T (Th1) cells to produce IFN- and cytotoxic T lymphocytes (CTLs) to kill tumor cells, which may depend on an interaction between the p38 mitogen-activated protein kinase and Jak-STAT signaling pathways (Gollob et al., 1999). IL-2 is also necessary for IL-4-mediated differentiation of Th2 cells. Although Jak-STAT pathway emanating from the IL-2R is suppressed by preculture in IL-4, proliferative responses to IL-2 are augmented by IL-4-cultured cells (Castro et al., 1999). To date, effective clinical immune management has been molecularly localized to the inhibition of IL-2/IL-2R signaling. The blockade of IL-2 transcription by cyclosporin A (CsA) and IL-2-driven signaling by rapamycin have been demonstrated to clearly dampen immunological responsiveness (Woerly et al., 1996; Wang et al., 1999). Recent intensive studies on the biochemical process of IL-2/IL-2R signal transduction make this pathway a potential target for pharmacological intervention that can alter the progression of a broad range of T cell-mediated diseases.

During the past decade, the prodigiosin family has been suggested to be reference compounds for a growing family of drugs with potential therapeutic benefit. Although originally studied as potential antibiotic and cytotoxic compounds, some members of this class, particularly undecylprodigiosin (UP or prodigiosin 25-C), have been found to selectively suppresses T cell proliferation and to act primarily on CTLs, but not Th cells, macrophages, and B cells (Magae et al., 1996; Lee et al., 1998, 2000). It has been reported that UP blocks the cell cycle at the mid-late G₁ phase, before entry into the S phase, by inhibiting retinoblastoma protein expression (Songia et al., 1997; Mortellaro et al., 1999). PNU156804, a chemically synthesized derivative of UP, shows the same immunosuppressive activities as UP, and blocks IL-2-dependent proliferation and NF-B and AP-1 activation. Recently, we isolated prodigiosin (PDG) from a marine microorganism, Serratia marcescens (Han et al., 1998a). We previously reported that PDG was another member of prodigiosin family with immunosuppressive properties similar to that of UP, because PDG preferentially suppresses T cell, but not B cell, proliferation. In addition, our preliminary results suggested that PDG differed from UP, in that PDG inhibits the functions of Th cells, as indicated by the inhibition of T-dependent antibody response. It is also of note that PDG may be less toxic than UP, which was known to be toxic at 7 to 10 mg/kg (Tsuji et al., 1992; Songia et al., 1997). PDG does not induce body weight loss at therapeutic doses of 10 to 30 mg/kg (Han et al., 1998a). Furthermore, we observed during the present study that PDG does not induce body weight loss in nonobese diabetic (NOD) and DBA/1 mice with i.p. injection on alternate days for 16 weeks.

The primary objective of this study was to investigate the mode of action of PDG in vitro and to determine the biological effect of PDG in vivo. We show that PDG blocks T cell activation by inhibiting primarily IL-2/IL-2R signaling pathway and delays the progression of autoimmune diabetes and collagen-induced arthritis. Another intention of this study was directed at comparing the mode of action of PDG with that of CsA, upon which most current immunosuppressive protocols are based (Jain et al., 1999). Currently, two or more immunosuppressive drugs are used in a low-dose combination to achieve the maximum therapeutic effect and minimize the toxicity. A prerequisite for the combined use of different drugs is that they should have different modes of action and unrelated toxicities. Thus, many new immunosuppressants are being developed for use in combination therapy, in the expectation that they exert complementary and synergistic effects. In this study, we show that PDG has a mechanism that is distinct from that of CsA in terms of T cell activation, particularly with respect to the IL-2/IL-2R signaling pathway.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. PDG (molecular weight 323) was prepared from the culture broth of Serratia marcescens (Han et al., 1998a) and used at concentrations ranging from 3 to 30 ng/ml. Con A, lipopolysaccharide (LPS), and CsA were purchased from Sigma (St. Louis, MO), IL-2 from Chemicon International (Temecula, CA), and neutralizing antibodies from R & D Systems (Minneapolis, MN).

Cell Culture. Spleen cells were obtained from a specific pathogen-free BDF1 mouse (female, 6-7 weeks) and were freed of red blood cells by using a lysis buffer treatment (Han et al., 1998a). T cells were prepared from spleen cells. To deplete adherent macrophages, spleen cells were incubated for 1 h in a tissue culture Petri dish, and B cells were then depleted with Dynabeads coated with mouse pan-B antibody (anti-B220) according to the manufacturer's instructions (Dynal Biotech, Oslo, Norway). In the T cell-enriched population, CD3-positive T cells represented 85 to 90%, and contaminated CD19-positive B cells less than 10%, as determined by flow cytometric analysis. B cells were negatively prepared using Dynabeads coated with mouse CD3 antibody and peritoneal macrophages were isolated from the abdominal cavity. Cells were cultured in RPMI 1640 medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal calf serum (Hyclone, Logan, UT), glutamine, and 2-mercaptoethanol (Sigma).

Analysis of Cytokine Gene Expression. Splenic T cells were stimulated with 5 µg/ml Con A for 24 h and total RNA was extracted using an Ultraspec II RNA isolation kit (Biotech Laboratories Inc., Houston, TX). Reverse transcription-polymerase chain reaction (RT-PCR) and quantitative RT-PCR were performed to determine cytokine gene expression changes, as described previously (Han et al., 1998b). The primer sequences were as follows: IL-2, sense 5'-CTT GCC CAA GCA GGC CAC AG-3', antisense 5'-GAG CCT TAT GTG TTG TAA GC-3'; IFN-, sense 5'-AGC GGC TGA CTG AAC TCA GAT TGT AG-3', antisense 5'-GTC ACA GTT TTC AGC TGT ATA GGG-3'; and IL-4, sense 5'-GAA TGT ACC AGG AGC CAT ATC-3', antisense 5'-CTC AGT ACT ACG AGT AAT CCA-3'. We used 1.0 × 10⁶ molecules as internal standards for the quantification of IL-2, IFN-, and IL-4. After analyzing the band areas with an image analysis system (Multi-Analyst; Bio-Rad, Hercules, CA), the copy number of cytokine mRNA was calculated from the relative ratios of the relevant PCR product pairs. Cytokine protein levels were determined by enzyme-linked immunosorbant assay (ELISA), according to the manufacturer's instructions (R & D Systems).

Analysis of IL-2R Expression. Splenic T cells were stimulated with 5 µg/ml Con A. IL-2R expression levels were determined by RT-PCR, by using primers: sense, 5'-AAC AAC TGC AAT GAC GGT GA-3' and antisense, 5'-GCC CTC TCT CCC ATT AAA GC-3'. Anti-mouse CD25 antibodies conjugated with phycoerythrin (PE) was used for immunofluorescence analysis of cell surface IL-2R expression (Rusterholz et al., 1999). Double staining was performed using fluorescein isothiocyanate (FITC)-conjugated monoclonal antibodies to mouse surface markers, such as CD4 and CD8 (BD PharMingen, San Diego, CA).

Proliferation Assay. T and B cells were stimulated with 5 µg/ml Con A and LPS for 72 h, respectively. Cells were pulsed with 1 µCi/well [³H]thymidine (113 Ci/nmol; PerkinElmer Life Science Products, Boston, MA) for the last 18 h and harvested with an automated cell harvester (Inotech, Dottikon, Switzerland). The amount of [³H]thymidine incorporated into the cells was measured using a Wallac Microbeta scintillation counter (Wallac, Turku, Finland) (Han et al., 1998a).

Nitrite Quantification. Peritoneal macrophages were stimulated with 200 ng/ml LPS for 24 h. NO₂ accumulation was used as an indicator of nitric oxide production in the medium, as previously described (Jeon et al., 2000). The isolated supernatants were mixed with an equal volume of Griess reagent (1% sulfanilamide, 0.1% naphthylethylenediamine dihydrochloride, and 2% phosphoric acid) and incubated at room temperature for 10 min. Nitrite production was determined by measuring optical density at 540 nm by using NaNO₂ to generate a standard curve.

Analysis of Killing Activity and Perforin Expression of CTLs. Female C57BL/6 (H-2^b) mice were immunized i.p. with P815 mastocytoma cells (H-2^d, 2 × 10⁷ cells/mouse). The killing activity of CTLs was determined by a ⁵¹Cr release assay, as described previously (Lee et al., 2000). Perforin gene expression level was determined by RT-PCR. The primer sequences for perforin were as follows: sense, 5'-CAG CTC TTC CAC CTG CAG-3' and antisense, 5'-TTA AAG CTT ATA CAA GCC.

Electrophoretic Mobility Shift Assay (EMSA). Splenic T cells were stimulated with 5 µg/ml Con A for 6 h. EMSA was performed as described previously (Jeon et al., 2000). The oligonucleotide sequences were as follows: NF-B/Rel, 5'-GAT CTC AGA GGG GAC TTT CCG AGA GA-3'; NF-AT, 5'-CTG TAT GAA ACA AAT TTT CCT CTT TGG GC-3'; and AP-1, 5'-GAT CTG CAT GAG TCA GAC ACA-3'. Oligonucleotides for STATs (4, 5, and 5/6) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA).

Treatment of NOD Mice with PDG. Female NOD mice were obtained from both Taconic Farms (Germantown, NY) and Jackson Laboratories (Bar Harbor, ME). Animals were maintained under specific pathogen-free conditions and provided with sterile food and water ad libitum. NOD mice (n = 10) were treated i.p. with 10 mg/kg PDG on alternate days from 8 to 24 and 14 to 20 weeks of age to examined the effect of PDG on diabetes onset and progression, respectively. The accumulated incidence of diabetes was determined by measuring urine glucose levels once a week by using a Glucoseurea detection kit (Uropaper; Eiken Chemical Co., Ltd., Tokyo, Japan). On the last day, mice were killed for blood glucose analysis and histological examination (Yoon et al., 1999).

Treatment of Collagen-Induced Arthritic Mice with PDG. Male DBA/1 mice were obtained from Charles River Japan Inc. (Yokohama, Japan). Bovine type II collagen was diluted in 0.05 M acetic acid to a concentration of 2 mg/ml and emulsified in equal volumes of complete Freund's adjuvant (2 mg/ml of Mycobacterium tuberculosis strain H37RA; Difco, Detroit, MI). The mice (n = 13) were immunized intradermally at the base of the tail with 100 µl of emulsion. On day 21, the animals were given i.p. booster injections of 100 µg of type II collagen dissolved in phosphate-buffered saline. On day 28, the onset of arthritis was accelerated by a single i.p. injection of 40 µg of LPS. Mice were treated on alternate days with 10 mg/kg PDG from the following day of LPS injection. Mice were examined visually for the appearance of arthritis in the joints and severity scores (macroscopic score) were given as previously described (Joosten et al., 1997). The clinical severity of arthritis was graded on a scale of 0 to 2 for each paw, according to changes in the redness and swelling, where 0 indicates no changes; 0.5, significant swelling and redness; 1.0, moderate; 1.5, marked; and 2.0, maximal swelling and redness, and ankylosis. The macroscopic score (mean ± standard deviation) was expressed as a cumulative value for all paws. On the last day, the mice were killed for histological examination (Joosten et al., 1997).

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

PDG Selectively Inhibits T Cell Proliferation. We investigated the cell-type specificity of PDG by using Con A as a T cell-specific activator and LPS as an activator of B cells and macrophages. As shown in Fig. 1, PDG at 30 ng/ml completely inhibited the [³H]thymidine incorporation into Con A-stimulated T cells, whereas LPS-stimulated B cells and macrophages were unaffected by PDG. T cell specificity was shown from a concentration of 3 ng/ml PDG. We verified that PDG at fully active concentration did not induce cell death as determined by propidium iodide uptake experiment (data not shown).

View larger version (19K): [in this window] [in a new window]   Fig. 1.   PDG selectively inhibits T cell proliferation. Purified splenic T and B cells were activated with 5 µg/ml Con A and 5 µg/ml LPS, respectively. After incubation for 72 h, cellular proliferation was measured. The cpm values of control were 41,751 ± 1,348 (Con A) and 25,493 ± 608 (LPS). Peritoneal macrophages were activated with 200 ng/ml LPS for 24 h. The supernatants were subsequently isolated and analyzed for nitrite. The amount of nitrite produced by normal macrophages was 64 ± 5 nmol/106 cells. PDG was dissolved in dimethyl sulfoxide and added at final concentrations of 3 to 30 ng/ml.

Con A Increases Cytokine Transcriptions along with Activation of NF-B, NF-AT, AP-1, and STATs. Along with monoclonal antibodies to cell surface antigens, Con A was generally used to polyclonally activate T cells. As shown in Fig. 2A, the major function of Con A is the selective mitogenic stimulation of T cells. We also showed that mRNA expressions of IL-2, IL-4, and IFN- were strongly induced by Con A at the time points of 4 to 24 h and down-regulated thereafter (Fig. 2B). To study the transcription factor activation, T cells were stimulated with Con A for 6 h and nuclear extracts were prepared and analyzed by EMSA. As shown in Fig. 2C, Con A markedly induced DNA binding of NF-B, NF-AT, AP-1, STAT4, STAT5, and STAT5/6. Because Con A is not previously known as a STAT activator, we investigate the possible activation of STAT via endogenously produced cytokines. Cycloheximide, which was added to prevent de novo cytokine synthesis, strongly inhibited the activation of STATs (4, 5, and 5/6). In another experiment, antibody mixtures of anti-IL-2 (5 µg/ml), anti-IL-4 (0.3 µg/ml), anti-IFN- (0.1 µg/ml), and anti-IL-12 (0.1 µg/ml) were used to neutralize endogenously produced cytokines. These mixtures markedly reduced the STAT activation in Con A-activated T cells, suggesting that endogenously produced cytokines led to STAT activation.

View larger version (76K): [in this window] [in a new window]   Fig. 2.   Con A activates NF-B, NF-AT, AP-1, and STATs along with increase of gene expression of IL-2, IL-4, and IFN-. Fractionated T and B cells were treated with Con A at concentrations from 1 to 10 µg/ml. After incubation for 72 h, the degree of lymphocyte proliferation was measured (A). Splenic T cells were activated with 5 µg/ml Con A up to 72 h. IL-2, IL-4, and IFN- mRNA levels were analyzed by RT-PCR (B). Nuclear extracts were prepared 6 h after incubation and used in EMSA with the indicated probes, namely, NF-B, NF-AT, and AP-1 (C). In another experiment, cycloheximide (CHX, 10 µg/ml) or antibody mixtures of anti-IL-2 (5 µg/ml), anti-IL-4 (0.3 µg/ml), anti-IFN- (0.1 µg/ml), and anti-IL-12 (0.1 µg/ml) were added to the culture of T cells at the same time as the addition of Con A. Nuclear extracts were prepared 6 h after incubation and used in EMSA with the indicated probes, namely, STAT4, STAT5, and STAT5/6 (C).

PDG Inhibits IL-2R Expression, but not IL-2 Production. We determined the effect of PDG on IL-2 transcription by cytokine-specific quantitative RT-PCR (Han et al., 1998b). The number of IL-2 mRNA molecules was increased to 480,000 molecules/100 ng of total RNA by Con A, but PDG did not inhibit this response (Fig. 3A). The effects of PDG on IL-2 production were further confirmed at the protein level by ELISA. The protein level of IL-2 was increased to 1963 pg/ml by 5 µg/ml Con A, and this was not significantly suppressed by PDG (Fig. 3C). Next, we investigated the effect of PDG on mRNA and the surface levels of IL-2R on activated T cells. Splenic T cells activated with 5 µg/ml Con A for 24 h induced the synthesis of IL-2R mRNA, and this was strongly inhibited by PDG in a dose-dependent manner, as determined by RT-PCR analysis (Fig. 3B). The effects of PDG on the cell surface IL-2R expression of T cells were further examined by flow cytometry with anti-CD25 (IL-2R) antibody conjugated with PE. CD25 expression was increased by 70.9% after the activation of T cells with 5 µg/ml Con A and PDG strongly inhibited it in a dose-dependent manner (Fig. 3D). We further determined the effects of PDG on the IL-2R expression of CD4+ Th cells and CD8+ CTLs by using anti-CD25-PE/anti-CD4-FITC and anti-CD25-PE/anti-CD8-FITC antibodies, respectively. The ratios of CD25+CD4+ and CD25+CD8+ T cells were increased by 30.2 and 22.5% by 24 h of Con A treatment, respectively, and this was efficiently inhibited by PDG in a dose-dependent manner (Fig. 4).

View larger version (38K): [in this window] [in a new window]   Fig. 3.   PDG inhibits IL-2R formation, but not IL-2 production. Splenic T cells were activated with 5 µg/ml Con A for 24 h. PDG or CsA was dissolved in dimethyl sulfoxide and added at final concentrations of 3 to 30 ng/ml. The copy number of IL-2 mRNA was measured by quantitative RT-PCR. After analyzing the band areas by using an image analysis system (Multi-Analyst; Bio-Rad), the copy number of cytokine mRNA was calculated from the relative ratios of the relevant PCR product pairs (A). IL-2 protein level was determined by ELISA (C). IL-2R mRNA levels were analyzed by RT-PCR. After calculating the band areas using an image analysis system, the relative ratios to -actin were presented. Cell surface IL-2R expression was measured by flow cytometry using anti-CD25 (IL-2R) antibodies (5000 cells). UN, chemically untreated control cells; NA, Con A-treated naive cells; VH, vehicle (0.1% dimethyl sulfoxide)-treated cells.

View larger version (49K): [in this window] [in a new window]   Fig. 4.   PDG inhibits IL-2R expression of both Th cells and CTLs. Splenic T cells were activated with Con A (5 µg/ml) for 24 h. The levels of IL-2R expression in CD4+ Th cells and CD8+ CTLs (5000 cells) were determined with a double staining method by using anti-CD25-PE/anti-CD4-FITC or anti-CD25-PE/anti-CD8-FITC antibodies.

PDG Blocks IL-2R Expression of T Cells Regardless of IL-2. To assess the contribution of excessively added IL-2 on immunosuppressive activity of PDG and CsA, we supplemented the cultures with 10 units/ml IL-2. CsA strongly blocked the induction of IL-2R expression in T cells and the addition of IL-2 markedly recovered it, even though not completely (Fig. 5A). However, PDG inhibited IL-2R expression in T cells regardless of the presence of excessive amount of exogenous IL-2 (Fig. 5B), clearly indicating that PDG and CsA had different modes of action.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Exogenously added IL-2 recovers suppressive activity of CsA, but not that of PDG. Splenic T cells were activated with Con A alone or in combination with IL-2 (10 units/ml) for 24 h. CsA (A) or PDG (B) was dissolved in dimethyl sulfoxide and added at final concentrations of 3 to 30 ng/ml. Cell surface IL-2R expression was measured by flow cytometry with anti-CD25 (IL-2R) antibodies (5000 cells).

PDG Inhibits IFN- and IL-4 Production of Th Cells and CTL Activity. Splenic T cells were activated with 5 µg/ml Con A for 24 h, and cytokine-specific quantitative RT-PCR was performed. The number of mRNA molecules of IFN- and IL-4 increased to 6,558,000 and 640,000 molecules/100 ng of total RNA, respectively, and these were inhibited by PDG in a dose-dependent manner (Fig. 6, A and B). The effect of PDG on IFN- and IL-4 was further confirmed at the protein level by ELISA. As occurred for gene expression, IFN- and IL-4 protein expressions were strongly suppressed by PDG (Fig. 6, C and D). We next investigated the effect of PDG on CTLs. C57BL/6 mice were immunized with allogeneic antigen P815 cells, which induced active CTLs in the spleen 10 days after immunization. As shown in Fig. 7A, the killing activity of CTLs in P-815 cells is blocked by PDG at dosages ranging from 1 to 10 mg/kg. We investigated the effect of PDG on perforin, which is responsible for forming pores in the target cell membrane and plays a central role in the killing process. Perforin gene expression in spleen cells was significantly increased after immunization with alloantigen, and strongly inhibited by PDG (Fig. 7B).

View larger version (37K): [in this window] [in a new window]   Fig. 6.   PDG inhibits the IFN- and IL-4 production of Th cells. Splenic T cells were activated with 5 µg/ml Con A for 24 h. The mRNA copy number of IFN- (A) and IL-4 (B) was measured by quantitative RT-PCR. After analyzing the band areas with an image analysis system, the copy number of cytokine mRNA was calculated from the relative ratios of the relevant PCR product pairs. Protein levels of IFN- (C) and IL-4 (D) were determined by ELISA. UN, chemically untreated control cells; NA, Con A-treated naive cells; VH, vehicle (0.1% dimethyl sulfoxide)-treated cells.

View larger version (26K): [in this window] [in a new window]   Fig. 7.   PDG inhibits the cytotoxicity of CTLs. C57BL/6 mice (H-2b) were immunized with P815 (H-2d, 2 × 107 cells/ml) on day 0. PDG was injected on alternate days at doses of 1 to 10 mg/kg. Mice were sacrificed on day 10 and cytotoxic activity measured by 51Cr release assay (A). Effector cells were incubated with P-815 target cells at 25 to 100:1 of effector/target cells. Perforin gene expression levels were analyzed by RT-PCR. After calculating the band areas with an image analysis system, the relative ratios to -actin were presented (B). UN, chemically untreated normal mice; P-815, P-815-injected mice.

PDG Indirectly Inhibits Activation of STATs, but not That of NF-B, NF-AT, and AP-1. Con A differentially regulated the activation of transcription factors in T cells, as shown in Fig. 2C. Although PDG did not inhibit DNA binding of NF-B (Fig. 8A), NF-AT (Fig. 8B), and AP-1 (Fig. 8C) in Con A-activated T cells, this compound strongly blocked the activation of STAT4 (Fig. 9A), STAT5 (Fig. 9B), and STAT5/6 (Fig. 9C) in Con A-activated T cells. To determine the direct effect of PDG on STAT activation, we investigated the effect of PDG on cytokine-induced STAT activation in preactivated T cells. We induced DNA binding of STAT4 with IL-12 (Fig. 9D), STAT5 with IL-2 (Fig. 9E), and STAT6 with IL-4 (Fig. 9F), but it was not inhibited by PDG, suggesting that PDG inhibits cytokine signalings, leading to STAT activation, in Con A-treated T cells.

View larger version (58K): [in this window] [in a new window]   Fig. 8.   PDG does not inhibit DNA binding of NF-B, NF-AT, and AP-1. Splenic T cells were activated with Con A (5 µg/ml). PDG or CsA was added at 3 to 30 ng/ml. Nuclear extracts were prepared 6 h after incubation and used in EMSA with the indicated probes, namely, NF-B (A), NF-AT (B), and AP-1 (C). For competition, the nuclear extracts were incubated with competitor (Co), a 100-fold excess of unlabeled oligonucleotides, before adding the labeled probe.

View larger version (87K): [in this window] [in a new window]   Fig. 9.   PDG indirectly blocks DNA binding of STATs by inhibiting cytokine signalings. Splenic T cells were activated with Con A (5 µg/ml). PDG or CsA was added at 30 ng/ml. Nuclear extracts were prepared 6 h after incubation and used in EMSA with the indicated probes, namely, STAT4 (A), STAT5 (B), and STAT5/6 (C). For competition, the nuclear extracts were incubated with competitor (Co), a 100-fold excess of unlabeled oligonucleotides, before adding the labeled probe. Splenic T cells were preincubated with 5 µg/ml Con A for 48 h. After complete washing, these cells were rested for 24 h, further cultured with PDG (30 ng/ml) for 3.5 h, and subsequently treated with IL-12 (2 ng/ml), IL-2 (10 unit/ml), and IL-4 (20 ng/ml), respectively, for 21 min. Nuclear extracts were prepared and used in EMSA with the indicated probes, namely, STAT4 (D), STAT5 (E), and STAT5/6 (F).

PDG Delays Progression of Autoimmune Diabetes. We investigated the effects of PDG on autoimmune diabetes by using NOD mice. Accumulated diabetic incidence was measured in female NOD mice (n = 10; Taconic Farms) treated with PDG at 10 mg/kg on alternate days from 8 to 24 weeks of age. As shown in Fig. 10A, none of the PDG-treated NOD mice developed diabetes. In contrast, 80% (8/10) of the control NOD mice developed diabetes at 24 weeks of age. To assess whether PDG was effective after the onset of insulitis, we administered the drug into NOD mice (n = 10; Jackson Laboratories) from 14 to 20 weeks of age. This treatment delayed diabetes progression (Fig. 10B). These results demonstrate that PDG delays the onset and progression of autoimmune diabetes. Blood glucose levels of the NOD mice, measured on the last day of PDG administration, demonstrated that the blood glucose levels of PDG-treated mice were significantly lower than those of the control NOD mice (Table 1). Histological examination of the pancreatic islets at 24 weeks revealed that most of the islets (93%) from the PDG-treated mice were intact, whereas 86% of the islets examined in the control NOD mice exhibited insulitis (Table 1; Fig. 11).

View larger version (22K): [in this window] [in a new window]   Fig. 10.   PDG delays the progression of autoimmune diabetes in NOD mice. NOD mice (n = 10) obtained from Taconic Farms were treated on alternate days with 10 mg/kg PDG () from 8 (arrow) to 24 weeks of age (A). NOD mice (n = 10) obtained from Jackson Laboratories were treated on alternate days with 10 mg/kg PDG () from 14 (arrow) to 20 weeks of age (B). Control NOD mice () were treated with 0.4% Tween 80, which was used to dissolve PDG. Urine glucose levels were measured weekly using a Glucoseurea detection kit (Uropaper; Eiken Chemical Co., Ltd.). Diabetic NOD mice showed 250 to 500 mg/dl urine glucose level. Accumulated diabetes incidence (%, n = 10) was described.

View larger version (146K): [in this window] [in a new window]   Fig. 11.   Representative photographs of the pancreatic islets of NOD mice. NOD mice obtained from Taconic Farms were treated on alternate days with 10 mg/kg PDG (A and B) from 8 to 24 weeks of age. Control NOD mice were treated with 0.4% Tween 80, which was used to dissolve PDG (C and D). On the last day of drug administration, pancreatic islets were histologically examined with hematoxylin and eosin.

PDG Delays Progression of Collagen-Induced Arthritis. We investigated the effect of PDG on the onset of collagen-induced arthritis. Mice (n = 13) were treated with 10 mg/kg PDG on alternative days. As shown in Fig. 12, PDG markedly delays the onset of arthritis on the macroscopic arthritis score. In addition to visual scoring, we analyzed the histological features of the joints on the last day of the experiment. PDG markedly decreased the infiltration of inflammatory cells, resulting in a delay of disease progression (Fig. 13).

View larger version (27K): [in this window] [in a new window]   Fig. 12.   PDG delays the progression of collagen-induced arthritis. Male DBA/1 mice (n = 13) were immunized with bovine type II collagen/complete Freund's adjuvant emulsion, boosted with bovine type II collagen, and treated with LPS. Mice were treated on alternate days with 10 mg/kg PDG from the next day of LPS injection. Control DBA/1 mice were treated with 0.4% Tween 80, which was used to dissolve PDG. Mice were examined visually for the appearance of arthritis in the joints. The clinical severity of arthritis was graded on a scale of 0 to 2 for each paw, according to changes in the redness and swelling, where 0 indicates no changes; 0.5, significant swelling and redness; 1.0, moderate; 1.5, marked; and 2.0, maximal swelling and redness, and ankylosis. The macroscopic score (mean ± standard deviation) was expressed as a cumulative value for all paws.

View larger version (138K): [in this window] [in a new window]   Fig. 13.   Representative photographs of the joints of DBA/1 mice. On the last day of drug administration, the rear paws were removed and processed for histology (hematoxylin and eosin staining). Nonarthritic-interphalangeal (A) and -knee (C) joints of PDG-treated DBA/1 mice, and arthritic-interphalangeal (B) and -knee (D) joints of vehicle-treated DBA/1 mice are shown. Enhanced filtrate and severe cartilage surface disruption of arthritic mice were observed.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

In the present study, we have demonstrated that PDG selectively inhibits T cell activation. Although PDG does not block IL-2 production, it efficiently inhibits the IL-2R expression, and this results in a disruption of the IL-2/IL-2R signaling pathway in Con A-activated T cells. The mitogenic lectin Con A is polyclonal activator of T cells, leading to T cell proliferation presumably through the TCR. Although it is very difficult to separate the outcome of TCR signals from IL-2/IL-2R signals in Con A-induced T cell activation, it is evident that IL-2 plays a pivotal role in T cell growth. Past studies have shown that Con A- or anti-CD3-induced proliferation by splenic T cells from IL-2-, IL-2R-, and IL-2R-deficient mice was markedly reduced, but not completely, compared with control littermate T cells (Suzuki et al., 1995; Willerford et al., 1995; Zheng et al., 1998). The simplest interpretation of these results is that T cell proliferation is largely IL-2 driven, although a portion is independent of IL-2. Recently, it is shown that IL-2-independent response is largely driven by engagement of the TCR and either CD28 or CD40L (Razi-Wolf et al., 1996; Boulougouris et al., 1999). However, it should be stressed that the growth of T cells, which are proliferating in the absence of IL-2, is limited to two to three cell divisions. Furthermore, these T cells do not differentiate into effector T cells, as assessed by their minimal CTL activity and lack of IFN- secretion (Malek et al., 2001). PDG efficiently blocks T cell activation by inhibiting IL-2R formation, followed by the disruption of IL-2/IL-2R signaling. The lack of IL-2/IL-2R signaling in Con A-activated T cells is able, in part, to explain the down-regulation of IL-2-dependent gene expression. Perforin is an inducible component of the lytic machinery of CTLs, and its expression is known to be crucially regulated by IL-2/IL-2R signaling during CTL differentiation (Zhang et al., 1999). With respect to cytokine-secreting effector Th cells, IL-2 plays a critical role in priming naive CD4+ Th cells to become IL-4 and IFN- producers (Seder et al., 1994). However, we cannot rule out that PDG directly inhibits gene expression of perforin, IFN-, and IL-4. Although they are highly IL-2/IL-2R signaling-dependent, several redundant pathways may operate for the induction of both CTL activity and IFN- production. For example, T cells are activated and differentiated to CTLs and cytokine-secreting Th cells with IL-4 and anti-CD3 in the absence of IL-2/IL-2R signaling, even though its capacity is suboptimal (Malek et al., 2000). Thus, it remains to be determined whether PDG also inhibits such redundant signaling pathways.

Consistent with the case of TCR activation induced by anti-CD3/anti-CD28 or superantigen (Edmead et al., 1996), Con A-induced TCR activation also results in the induction of NF-AT, AP-1, and NF-B. In addition, we demonstrate here that Con A induces DNA binding of STATs (4, 5, and 6) in T cells. STAT activation in these T cells is markedly reduced by inhibiting de novo cytokine synthesis with cycloheximide and by neutralizing endogenously produced cytokines, including IL-2, IL-4, IL-12, and IFN-, with their respective antibodies. This result suggests that cytokine signaling plays a critical role in STAT activation in Con A-activated T cells, whereas Con A cannot be characterized as a direct STAT activator. Based on this finding, we can presume that PDG indirectly blocks STAT activation by inhibiting cytokine signaling in Con A-activated T cells.

This report suggests possible molecular target sites of PDG action. We show that PDG efficiently blocks IL-2R gene expression, indicating that PDG affects certain signaling pathways for IL-2R transcription. On the other hand, PDG does not inhibit IL-2 transcription, suggesting that several signaling pathways for IL-2 production are not targets of PDG action. Our results show that PDG does not affect the activation of NF-AT, NF-B, and AP-1, which play a critical role in enhancing IL-2 transcription (Yasui et al., 1998). IL-2R gene expression is tightly regulated through changes in its rate of transcription. Mitogenic stimuli rapidly induce IL-2R gene expression in T cells, and it is regulated by a potent proximal enhancer between nucleotides 299 and 228 that contains NF-B and CArG motif and between 137 and 64 that binds Elf-1 and HMG-I(Y) (John et al., 1995; Sperisen et al., 1995). However, IL-2 induces IL-2R transcription via an IL-2-responsive enhancer, whose activity depends on the cooperative binding of IL-2-induced STAT5 to two sites and of constitutively active Elf-1 to a third site (Rusterholz et al., 1999). In cells induced to express IL-2R with Con A, none of the IL-2-responsive enhancer sites is occupied. Further investigation of the activation of transcription factors involved in IL-2R gene expression may provide valuable information in identifying the molecular target proteins of PDG; this will be undertaken in our next study.

Here we demonstrate that PDG has a different mechanism of action from CsA and UP, a well known member of the prodigiosin family. PDG blocks the formation of the IL-2/IL-2R complex by the selective inhibition of IL-2R expression. On the other hand, PNU156804, a chemically synthesized analog of UP, does not inhibit the formation of IL-2/IL-2R complex (Mortellaro et al., 1999), and CsA blocks this pathway by inhibiting both IL-2 and IL-2R expression. Another difference is that PDG does not directly inhibit the activation of transcription factors, such as NF-B, AP-1, STATs, and NF-AT, which are inhibited by either UP or CsA, respectively (Mortellaro et al., 1999). Although they are structurally related to each other, PDG and UP show different modes of action on Th cells and CTLs. Although UP did not inhibit Th cell functions, PDG inhibits Th cell functions by inhibiting the expression of IL-2R, IFN-, and IL-4. PDG also down-regulates perforin gene expression of CTLs, whereas UP does not affect the gene or protein expression of perforin, but only abrogates perforin activity, as assayed by immunoblotting (Togashi et al., 1997). These results suggest that PDG has a unique target molecule, which is different from UP and CsA. Thus, PDG could be potentially used as an immunosuppressant in combination with CsA in therapy of allograft rejection and several autoimmune diseases, because they have different mode of action, which is a prerequisite for the combined use of different drugs.

Preventive approaches to autoimmune diseases either act directly on the immune system or prevent target cells from expressing autoantigen, making them less vulnerable to the immune system. In this report, we describe the promising effect of PDG upon autoimmune diabetes and collagen-induced arthritis, and partially suggest the mode of action of PDG. Activated Th1 cells and CTLs are believed to regulate the onset and progression of autoimmune diseases. In general, Th1 cytokines, such as IL-2 and IFN-, appear to play a pathological role in the destruction of  cells in autoimmune diabetes. For example, monoclonal antibody to IFN- prevents the development of diabetes in NOD mice and BioBreeding (BioBreeders, Watertown, MA) rats (Debray-Sachs et al., 1991). IL-2 activates and differentiates pre-CTLs into -cell-specific CTLs, which are capable of performing cytotoxic functions through the use of perforin and granzyme. Perforin-deficient NOD mice showed a decreased incidence of diabetes and a delayed onset of the disease (Jun et al., 1999). Th1 cytokines are also proinflammatory, leading to the induction of tumor necrosis factor- and IL-1 from macrophages, which play major roles in cartilage destruction in the synovial membranes of the joint (Thornton et al., 2000). In murine collagen-induced arthritis, administration of IFN- can exacerbate disease and administration of anti-IL-2R antibodies can inhibit disease onset (Banerjee et al., 1988). Our results indicate that PDG efficiently blocks the functions of Th1 cells and CTLs through the inhibition of IL-2/IL-2R signaling, IFN- production, and perforin expression, and this may result in, in part at least, the delay of the progression of autoimmune diabetes and collagen-induced arthritis.

In summary, our present results demonstrate that PDG has a distinct mode of action in that it primarily inhibits IL-2R expression, which plays a pivotal role in T cell growth, and suggest the possibility that PDG will be a good candidate compound for autoimmune disease immunotherapy.