Inhibition of indoleamine 2, 3-dioxygenase-mediated tryptophan catabolism accelerates crescentic glomerulonephritis

Abstract

Immunomodulatory enzyme indoleamine 2, 3-dioxygenase (IDO) is one of the initial and rate-limiting enzymes involved in the catabolism of the essential amino acid tryptophan. Via catalysing tryptophan degradation, IDO suppresses adaptive T cell-mediated immunity and plays an important role in various forms of immune tolerance. Its role in T helper type 1 (Th1)-directed, cell-mediated crescentic glomerulonephritis (GN) is still unclear. Therefore, we investigated the activity and role of IDO in crescentic GN using a model of nephrotoxic serum nephritis (NTN), and IDO activity was inhibited by 1-methyl-tryptophan (1-MT) in vivo. Our results showed that activity of IDO, as determined by high performance liquid chromatography analysis of the kynurenine/tryptophan ratio, was increased markedly in the serum and renal tissue of NTN mice, and immunohistochemistry revealed that expression of IDO was up-regulated significantly in glomeruli and renal tubular epithelial cells during NTN. Treatment with 1-MT resulted in significantly exacerbated kidney disease with increased glomerular crescent formation, accumulation of CD4⁺T cells and macrophages in renal tissue, and augmented renal injury compared with phosphate-buffered saline-treated NTN mice, which was associated with enhanced Th1 responses and intrarenal cellular proliferation. These findings suggest that the development of NTN was regulated negatively by increased IDO activity, and IDO might play an important role in the pathogenesis of crescentic GN.

Introduction

Glomerulonephritis (GN) is an inflammatory disease of the renal glomeruli and is a major cause of end-stage renal failure. The most severe and rapidly progressive type of GN results from infiltration of inflammatory cells into the glomerular tuft and accumulation of cells and fibrin in Bowman's space; these forms are known as crescentic GN. The most widely used model of crescentic GN is nephrotoxic serum nephritis (NTN), in which heterologous anti-glomerular basement membrane (GBM) antibodies act as a stimulating antigen in the glomeruli. Increased evidence has implied that the CD4⁺ T helper 1 (Th1) cell-mediated response is the culprit for glomerular injury in crescentic GN [1,2]. The glomerular influx of CD4⁺ T cells and macrophages results in local cell proliferation, formation of glomerular crescents and severe glomerular injury [3–5]. This is independent of autologous antibodies [6] and can be attenuated by Th2 cytokines [7–9].

Indoleamine 2, 3-dioxygenase (IDO; EC 1·13.11·42) is a haem-containing dioxygenase which catalyses the first and rate-limiting step in the major pathway of tryptophan (trp) catabolism in mammals [10]. IDO is expressed widely on various immune cells, including macrophages, defined dendritic cells and monocytes [11], and also expressed on other tissue type cells such as mesenchymal stem cells, microglia cells, fibroblasts cells and renal tubular epithelial cells (TEC) [12–15], while Th1 cytokine interferon (IFN)-γ is the strongest evocator for IDO expression in those cells. IDO catalyses the cleavage of the trp pyrrol ring to N-formyl kynurenine (kyn), and catabolizes further to 3-hydroxykynurenine acid, 3-hydroxy-anthranilic acid and quinolinic acid, and these catabolites are referred to collectively as kynurenines. The resulting depletion of trp inhibits the proliferation of intracellular parasites such as Chlamydia trachomatis[16] and Toxoplasma gondii[17]. In addition to its anti-microbial effect, recent studies have shown that up-regulation of IDO activity can produce a local immunosuppressive environment. Either tryptophan depletion or increased levels of tryptophan catabolites can inhibit T lymphocyte replication, so IDO plays an immunosuppressant role in many respects including allergy, tumour immunology, maternal–fetus tolerance and human immunodeficiency virus infection [18–21]. In addition, enhanced IDO activity plays an important role in inducing immunological tolerance in many Th1-mediated autoimmunity diseases, such as experimental autoimmune encephalomyelitis (EAE), autoimmune diabetes in non-obese diabetic mice and collagen-induced arthritis [22–24].

As all these studies indicate that Th1-like responses can be controlled by IDO, we hypothesized that IDO might contribute to the disease course of crescentic GN, which might also be mediated by CD4⁺ Th1 cells. To determine whether IDO plays a negative regulatory role in crescentic GN mice, we assessed IDO activity and expression in serum and renal tissue and the effect of 1-methyl-tryptophan (1-MT), a specific inhibitor of IDO, on the development of crescentic GN in a B6 mouse model of NTN.

Materials and methods

Animals

Eight-week-old male C57BL/6 mice were obtained from the Third Military Medical University Daping Animal Services (Chongqing, China) and housed in a specific pathogen-free mouse facility.

Preparation of rabbit anti-mouse GBM serum

Rabbit anti-mouse GBM serum was prepared as described by Xie et al.[25]. Briefly, glomeruli of B6 mice were isolated from the renal cortex through a series of sieves of decreasing pore size (250-, 150- and 75-µm mesh), and disrupted by sonication. The GBM fractions were then collected by centrifugation. For immunization, 1 mg GBM protein was emulsified with 1 ml Freund's complete adjuvant (Sigma Chemicals Co., St Louis, MO, USA) and was administered to the rabbit by subcutaneous (s.c.) injection. Titres of anti-mouse GBM serum were raised in New Zealand white rabbits by repeated immunization against glomerular sonicates.

Induction of crescentic GN and treatment with 1-methyl-DL-tryptophan

Mice were first sensitized by s.c. injection with rabbit immunoglobulin (Ig)G (250 µg/mouse) in complete Freund's adjuvant (Sigma; day 0). After 5 days, 0·25 ml rabbit anti-mouse GBM serum was given intravenously. NTN mice were injected intraperitoneally (i.p.) daily with 10 mg of 1-MT (Sigma Aldrich Co., St Louis, MO, USA; the endotoxin level is less than 1 EU/µg) dissolved in phosphate-buffered saline (PBS) from 24 h prior to the sensitization (day –1) to the end of the experiment as the 1-MT-treated NTN group (n = 15), and NTN mice were injected i.p. daily with PBS alone as the NTN control group (n = 15). Non-immunized mice (n = 15) served as the normal control group without GN. IDO activity, immune responses and renal injury were assessed on day 19.

Measurement of IDO activity in serum and renal tissue

To detect IDO activity, serum samples and renal tissue of the three groups of mice were collected at the end of the experiments, and serum samples were frozen in –20°C until high performance liquid chromatography analysis. Renal tissues were dissected and homogenized in 2 µl perchloric acid (3.6%; Roth, Karlsruhe, Germany) per mg of weight, and after centrifugation (at 4°C/15 000 g) the protein-free supernatant was frozen at –20°C. Trp and kyn concentrations were measured by reverse-phase high performance liquid chromatography, as described previously [26]. Trp was measured by fluorescence detector (Model Prostar 360, Varian Inc., Middelburg, the Netherlands) at an excitation wavelength of 286 nm and an emission wavelength of 366 nm, and kyn was measured by a fluorescence detector (Hewlett Packard G13144, Palo Alto, CA, USA) at 360 nm. To estimate IDO activity, the kyn/trp ratio was calculated.

Assessment of renal injury

Glomerular crescent formation was assessed on 3-µm paraffin sections stained with periodic acid-Schiff reagent. Glomeruli were considered to exhibit crescent formation when two or more layers of cells were observed in the Bowman's space. A minimum of 40 glomeruli for each animal were counted to determine the percentage of crescent formation. Proteinuria (mg/24 h) were measured by a modified Bradford method [27] on urine which was collected during the final 24 h of the experiments. Serum creatinine concentrations at the end of the experiments were measured by an enzymatic creatinase assay [28].

Assessment of renal macrophages and CD4⁺ T cell infiltration

Macrophages and CD4⁺ T cells were demonstrated by immunoperoxidase staining on the frozen 6-µm-thick kidney sections, as described previously [29]. Using FA/11 (rat anti-mouse CD68; ATCC, Manassas, VA, USA) for macrophages and GK1.5 monoclonal antibody (mAb) (rat anti-mouse CD4; ATCC) for CD4⁺ T cells. Isotype-matched control IgG (rat IgG2b; Pharmingen, San Diego, CA, USA) served as a negative control. The number of cells bearing CD4 or CD68 determinants was assessed in a minimum of 20 randomly selected high-power fields (hpf)/animal, and results were expressed as cells/glomerulus for glomeruli and cells/hpf for interstitiumi.

Renal expression of IDO and proliferating cell nuclear antigen

The IDO and proliferating cell nuclear antigen (PCNA) were identified in paraffin-embedded sections (4–6 µm) by immunohistochemistry. The sections were treated first with 0·3% hydrogen peroxide to block non-specific staining and digested lightly with pepsin for antigen retrieval, and then incubated with rat anti-mouse IDO mAb (Upstate, SC, USA) 1:300 and rat anti-mouse PCNA mAb (Dako, Cambridge, UK) 1:1000 respectively. After washing, sections were further stained with horseradish peroxidase (HRP)-labelled goat anti-rat IgG (Sino American Biotechnology Co., Beijing, China) at a 1:5000 dilution. Reactivity was detected using 3′3′-diaminobenzidine reagent sets (Sino-American Biotechnology Co.), according to the manufacturer's instructions. The tissues were then counterstained lightly with haematoxylin; positive staining was detected as brown coloration of the tissues. The number of PCNA-positive cells were expressed as cells/glomerulus for glomeruli and cells/hpf for interstitiumi.

Splenocyte production of cytokines

To assess the cytokine production by splenocytes, spleen cells were collected at the end of the experiment; single-cell suspensions were prepared and cultured with protein G-purified rabbit globulin as described previously [9]. Briefly, splenocytes were isolated by sieving, and red blood cells were removed by lysis. Cells were suspended in Dulbecco's modified Eagle's medium (supplemented with 0·3 mg/ml L-glutamine, 100 U/ml penicillin, 0·1 mg/ml streptomycin and 10% fetal calf serum) and planted into 96-well plates at a concentration of 5 × 10⁶ cells/ml per well. Cells then were cultured with rabbit globulin at a concentration of 10 µg/ml in a humidified environment with 5% CO₂ at 37°C. The concentrations of IFN-γ and interleukin (IL)-4 secreted by splenocytes cultured with rabbit globulin for 72 h were measured by enzyme-linked immunosorbent assay (ELISA) kits for IFN-γ and IL-4, as described previously [8]. We took as controls the cytokines secreted by splenocytes from the two groups of NTN mice incubated without rabbit globulin for 72 h.

Measurement of circulating IgG subclasses

Antigen-specific IgG1 and IgG2a serum levels were measured by ELISA using serum collected at the end of the experiments, as described previously [9]. Microtitre plates were coated with purified rabbit IgG (Sigma-Aldrich) in PBS, washed and blocked with bovine serum albumin (20 mg/ml in PBS). Serially diluted mouse sera (1:100–1:1600) were added to these plates, and HRP-conjugated anti-mouse IgG1 and IgG2a (1:4000; Southern Biotechnology Association, Birmingham, AL, USA) were used to detect IgG subclasses. The plates were washed and incubated with 2,2′-azino-di-3-ethylbenzthiazoline sulphonate (0·1 M; Boehringer Mannheim, Mannheim, Germany), and absorbance was read at 405 nm to compare the concentration of each isotopy between two treatment groups. Sera from non-immunized C57BL/6 mice were tested to provide baseline antibody levels.

Statistical analysis

Results are expressed as the mean ± standard error of the mean. The unpaired t-test was used for statistical analysis. Values of P < 0·01 were considered significant.

Results

Determination of IDO activity during NTN

The IDO catabolizes tryptophan to its by-product, kynurenine. To confirm IDO activity, kyn and trp concentrations were measured and kyr/trp ratios (µmol/mmol) were calculated in the serum and renal tissues of NTN mice; control samples were derived from non-immunized normal animals. Concentrations of kyn and trp were detectable in serum and renal tissue and low kyn/trp ratios were evaluated in non-immunized normal animals, but increased kyn concentrations and decreased trp concentrations were detected in the serum and renal tissue of NTN mice (P < 0·01, Fig. 1a and b); kyn/trp ratios were markedly increased. As expected, treatment with 1-MT decreased the kyn/trp ratios significantly in serum and renal tissue compared with PBS-treated NTN mice. These data indicate that IDO enzyme activity was up-regulated markedly in serum and renal tissue during NTN, and were blockaded effectively by 1-MT.

Fig. 1.

High performance liquid chromatography analysis of indoleamine 2, 3-dioxygenase (IDO) activity. Kyurinine/tryptophan (kyn/trp) ratio in the serum and kidney of non-immunized normal mice and nephrotoxic serum nephritis (NTN) mice treated with 1-methyl-tryptophan (1-MT) or phosphate-buffered saline. Non-immunized mice (n = 15) had a low-level (kyn/trp) ratio (µmol/mmol), but IDO activity was increased significantly in the serum (a) and kidney (b) of NTN mice (n = 15) at the end of the experiments (day 19), and 1-MT blockade (n = 15) effectively decreased the IDO activity during NTN. *P < 0·01 compared with non-immunized normal mice; **P < 0·01 compared with 1-MT treated NTN mice.

Effect of 1-MT treatment on renal injury

Non-immunized mice exhibited normal renal histology (Fig. 2a), while mice immunized with rabbit anti-mouse GBM serum developed crescentic GN. Treatment with 1-MT exacerbated significantly the histological features of crescentic GN compared with PBS-treated NTN mice (Fig. 2b and c). The percentage of crescent formation was increased significantly after 1-MT treatment compared with PBS-treated NTN mice (1-MT treated: 38·9 ± 6·8%; PBS-treated: 23·2 ± 4·1%, P < 0·01; Fig. 3a). Proteinuria and serum creatinine values of 1-MT-treated NTN mice at the end of the experiments were both elevated significantly compared with PBS-treated NTN mice (P < 0·01; Fig. 3b and c).

Fig. 2.

Renal histology of non-immunized mice and nephrotoxic serum nephritis (NTN) mice treated with 1-methyl-tryptophan (1-MT) or phosphate-buffered saline (PBS) on day 19. Representative glomeruli are shown from the three groups of mice; the non-immunized mice did not develop crescentic glomerulonephritis (GN) (a), while the other mice groups did (b,c). Crescents were assessed by periodic acid-Schiff staining on paraffin-embedded kidney sections. PBS-treated NTN mice developed proliferative crescentic GN (b), which was exacerbated by 1-MT blockade (c) (magnification: ×400).

Fig. 3.

Renal injury, as assessed by percentage of crescent formation (a), proteinuria (b) and serum creatinine (c) in 1-methyl-tryptophan (1-MT) treated nephrotoxic serum nephritis (NTN) mice (n = 15) and phosphate-buffered saline (PBS)-treated NTN mice (n = 15). Percentage of crescent formation and serum creatinine on day 19 and urinary protein excretion during the last 24 h of experiments were both increased prominently in the 1-MT treated NTN mice (*P < 0·01, n = 15) compared with PBS-treated NTN mice (n = 15).

Renal accumulation of CD4⁺ T cells and macrophages

To determine whether 1-MT has an impact on renal infiltrated inflammatory cells during NTN, we assessed and compared the number of intrarenal CD4⁺ T cells and macrophages in the 1-MT- and PBS-treated NTN mice by immunostaining. We detected a significantly enhanced infiltration of CD4⁺ T cells both in glomeruli (P < 0·01; Fig. 4a) and tubulointerstitiumi (P < 0·01; Fig. 4c) after 1-MT treatment, and infiltration of macrophages was also increased notably in both glomeruli (P < 0·01; Fig. 4b) and tubulointerstitiumi (P < 0·01; Fig. 4d) in 1-MT-treated NTN mice compared with PBS-treated NTN mice.

Fig. 4.

Intrarenal accumulation of CD4⁺ T cells and macrophages 1-methyl-tryptophan-treated nephrotoxic serum nephritis (NTN) mice (n = 15) and phosphate-buffered saline-treated NTN mice (n = 15) on day 19.

The IDO expression and intrarenal cell proliferation

To understand the role of IDO in kidney injury, we measured IDO expression in renal tissue during NTN. Immunohistochemical staining showed that IDO expression was absent from kidneys of normal non-immunized mice (Fig. 5a), but IDO expression was up-regulated strongly within glomeruli and TEC of NTN mice (Fig. 5b); IDO-positive cells in glomeruli exhibited the morphology of infiltrated monocytes/macrophages. No staining was observed when isotype control antibody was used.

Fig. 5.

Immunohistochemical staining for indoleamine 2, 3-dioxygenase (IDO). Non-immunized normal mice showed no IDO expression (a). However, the expression of IDO was up-regulated strongly in glomeruli of nephrotoxic serum nephritis mice, and the morphology of IDO-positive cells was similar to the infiltrated macrophages (b); IDO expression was also found in tubular epithelial cells (magnification: ×400).

Cellular proliferation within glomeruli and interstitial cells, which were assayed by immunohistochemical staining using rat anti-mouse PCNA mAb, increased significantly in NTN mice compared with normal non-immunized mice, and IDO blockade with 1-MT enhanced glomerular and interstitial cell proliferation significantly compared with PBS-treated NTN mice (P < 0·01; Fig. 6a and b).

Fig. 6.

Intrarenal cells proliferation in 1-methyl-tryptophan (1-MT)-treated nephrotoxic serum nephritis (NTN) mice (n = 15) and phosphate-buffered saline-treated NTN mice (n = 15). Proliferating cells were detected in renal tissue by immunohistochemistry on paraffin-embedded kidney section using rat anti-mouse proliferating cell nuclear antigen monoclonal antibody. 1-MT treatment during NTN increased cellular proliferation significantly within diseased glomeruli (a, *P < 0·01, cells/glomerulus) and interstitial (b, *P < 0·01, cells/high-power field).

Splenocyte cytokine production

As immune deviation from Th2 to Th1 responses is a feature of NTN, we measured the level of IFN-γ, a signature cytokine of Th1 cells, and IL-4, an important Th2 cytokine, and evaluated the changes in Th1/Th2 balance after 1-MT treatment. Two control groups had low levels of both IFN-γ and IL-4, and there were no statistically significant differences between the data of the two control groups. However, those cytokines secreted by splenocytes were both increased significantly after antigen was restimulated (P < 0·01; Fig. 7a and b). In the presence of antigen, the IFN-γ level secreted by splenocytes from 1-MT-treated NTN mice was increased significantly compared with PBS-treated NTN mice (P < 0·01; Fig. 7a). In contrast, splenocyte IL-4 production was decreased significantly in 1-MT-treated NTN mice compared with PBS-treated NTN mice (P < 0·01; Fig. 7b).

Fig. 7.

Effect of 1-methyl-tryptophan (1-MT) blockade on splenocyte cytokine production in vitro. Cytokine levels were measured by enzyme-linked immunosorbent assay on supernatants that were collected from cultured splenocytes. Two control groups had low levels of interferon (IFN)-γ and interleukin (IL)-4, but levels of those cytokines were both increased significantly after antigen was restimulated (a,b, *P < 0·01 compared with control groups; n = 15). After antigen was restimulated, the production of IFN-γ (a) by splenocytes from 1-MT treated nephrotoxic serum nephritis (NTN) mice was increased prominently compared with phosphate-buffered saline (PBS)-treated NTN mice (**P < 0·01; n = 15). However, the production of IL-4 (b) by splenocytes from 1-MT-treated NTN mice was decreased markedly compared with PBS-treated NTN mice (**P < 0·01; n = 15).

Assessment of circulating antigen-specific antibody subclass levels

Serum levels of antigen-specific IgG1 and IgG2a subclasses in the normal control group were very low, but were both increased significantly in NTN mice. Serum levels of IgG1 isotypes decreased significantly in 1-MT-treated NTN mice compared with PBS-treated NTN mice (P < 0·01; Fig. 8a), while IgG2a isotype rose markedly in 1-MT-treated NTN mice compared with PBS-treated NTN mice (P < 0·01; Fig. 8b). These results show that IDO blockade had affected the IgG subclasses, and the changes of IgG subclasses were consistent with splenocyte cytokine production after 1-MT treatment.

Fig. 8.

Circulating antigen-specific immunoglobulin (Ig)G1 and IgG2a levels in non-immunized normal mice (n = 15) and nephrotoxic serum nephritis (NTN) mice treated with 1-methyl-tryptophan (1-MT) (n = 15) or phosphate-buffered saline (PBS) (n = 15). Serum antigen-specific Ig subclass levels were measured by enzyme-linked immunosorbent assay and tested at five dilutions (1:100 to 1:1600) at the end of the experiments. Baseline represents Ig subclass levels in non-immunized normal C57BL/6 mice. (a)IgG1 levels were reduced significantly in 1-MT-treated NTN mice compared with PBS-treated NTN mice (P < 0·01 at all dilutions; while (b) IgG2a levels rose markedly in 1-MT treated NTN mice compared with PBS-treated NTN mice (P < 0·01 at all dilutions).

Discussion

Crescentic GN in mice resembles crescentic GN in man [30] and are characterized similarly by accumulation of Th1 effectors (CD4⁺ T cells and macrophages), which results in crescent formation and renal injury. Many researchers have shown that activation or tolerance of CD4⁺ T cell play an important role in the development of crescentic GN [2,5], but the underlying mechanisms terminating immune attacks have been poorly understood. IDO, a haem-containing dioxygenase, can catalyse tryptophan degradation and produce a microenvironment of tryptophan depletion and kynurenines accumulation, which can regulate T cell-mediated immune responses in inflammatory diseases, transplant rejection, pregnancy and autoimmunity diseases in vitro and in experimental mouse systems [10,11]. Therefore, IDO has emerged recently as a new paradigm in immunology. IDO is known to be induced effectively by IFN-γ, which is a strong activator for Th cell differentiation in the direction of Th1. Although IFN-γ is the key cytokine involved in the Th1 response, in several models of organ transplantation and organ-specific autoimmunity, for example, EAE and experimental autoimmune uveoretinitis, IFN-γ was found to have a protective effect [31–33]. Kitching et al. also reported that IFN-γ-deficient (IFN-γ^–/–) mice developed more severe autoimmune anti-GBM GN with increased crescent formation, intrarenal cellular proliferation and leucocyte infiltration, but the mechanisms behind the protective effect of IFN-γ in this model of crescentic GN remain complex and incompletely understood [28]. Until recently evidence suggested that the immune regulatory property of endogenous IFN-γ in many Th1-directed autoimmune diseases and solid organ transplantation were associated with its role in stimulating the up-regulation of IDO expression [34–35]. Based on these data we speculated that IDO might play an important role in the development of crescentic GN; consequently we investigated the activity and role of IDO in NTN mice.

High serum levels of kyn/trp have been reported in many T cell-mediated immune diseases [22,23,36–38]; furthermore, several studies have reported that significantly elevated IDO activity was found in serum and renal tissue of kidney allograft rejection, renal ischaemia/reperfusion injury and chronic renal failure [39–41]. By measuring the kyn/trp ratio, we also found that the activity of IDO was increased significantly in serum and renal tissue during NTN compared with non-immunized normal mice. In addition, we further tested the expression of IDO in renal tissue of NTN mice, and our results showed that IDO was expressed in glomeruli and renal TEC of NTN mice. Although the cellular source of IDO expression in glomeruli during NTN has not yet been defined precisely, the cellular morphology of IDO-positive cells is similar to monocytes/macrophages. Considering that macrophage is the main infiltrated inflammatory cell in crescentic GN [3,42], and up-regulation of IDO expression in peripheral infiltrated monocytes/macrophages has also been reported in other immune diseases [22,43], we speculate that those IDO-positive cells in glomeruli are probably infiltrated macrophages. Meanwhile, we found that IDO expression was up-regulated in TEC compared with non-immunized mice and, in fact, several studies have reported that IDO was expressed by TEC under some cytokines and inflammatory mediators released in pathological conditions such as kidney allograft rejection and renal ischaemia/reperfusion injury [15,39,40]. IDO has proved to be beneficial to tissues expressing the enzyme, as it is able to limit immune effector and prevent exaggerated immune activity by inhibiting the proliferation of inflammatory cells [44]. Hence, up-regulated IDO expression in renal tissue of NTN mice might play an important role in the development of crescentic GN.

To confirm the immunoregulatory role of IDO in NTN, we used 1-MT, a known competitive inhibitor of IDO, to blockade IDO activity in a B6 mouse model of NTN. In 1-MT-treated NTN mice, we demonstrated a significantly decreased kyn/trp ratio during NTN compared with PBS-treated NTN mice, suggesting that 1-MT could blockade effectively the activity of IDO. Blockade of IDO by systemic administration of 1-MT in NTN mice exacerbated crescentic GN significantly, with increased glomerular crescent formation, accumulation of CD4⁺ T cells and macrophages in renal tissue and aggravation of the renal injury. These data suggest that IDO plays an important role in the negative regulatory feedback of immunological mechanisms in crescentic GN.

It has been demonstrated that IDO exerts a critical role for controlling inflammatory cell proliferation [44,45], so we detected the expression of PCNA in renal tissue of 1-MT-treated NTN mice. Our data show that blockading IDO with 1-MT significantly enhanced cell proliferation in glomeruli and interstitium during NTN compared with PBS-treated NTN mice. These results suggest that increased intrarenal CD4⁺ T cells and macrophages after blockading with 1-MT were associated with augmented intrarenal cell proliferation. Up-regulation of IDO expression in glomeruli and TEC during NTN might produce a microenvironment of tryptophan depletion and kynurenine accumulation, which could attenuate CD4⁺T cell and macrophage accumulation locally within renal tissue by reducing cellular expansion. Thus, IDO might play an important role in inhibiting infiltrated leucocyte expansion in the renal tissue during NTN.

The systemic balance of Th1/Th2 plays an important role in affecting the development of this crescentic GN [2,7], while numerous studies show that IDO activity is associated with the Th1/Th2 paradigm [46,47], and inhibition of IDO with 1-MT can also alert the Th1/Th2 balance [23,48]. To confirm whether exacerbated kidney disease with 1-MT treatment is associated with Th1/Th2 bias, which is a systemic immune response and different from the mechanisms of local IDO blockade in renal tissue, we further investigated the balance of Th1/Th2 after 1-MT treatment by assessed the levels of cytokines secreted by splenocytes and circulating IgG1 and IgG2a subclasses. By measuring the concentrations of cytokines in the culture supernatant of splenocytes stimulated with rabbit globulin, we found that treatment with 1-MT could up-regulate IFN-γ production significantly, while down-regulating production of IL-4 compared with PBS-treated NTN mice. As IFN-γ/IL-4 are the representative cytokines of the Th1/Th2 response, these results demonstrate that inhibition of IDO would aggravate Th1 deviation in NTN mice. By measuring the levels of circulating IgG subclasses, we found that IgG1 isotypes decreased significantly, while the IgG2a isotype rose markedly in 1-MT-treated NTN mice compared with PBS-treated NTN mice; this change of IgG subclass was consistent with splenocyte cytokine production. Combined with the results of splenocyte production of cytokines, our study further proved that 1-MT treatment would augment Th1 responses in NTN mice.

The development of crescentic GN is exacerbated by Th1 cytokines and attenuated by Th2 cytokines. These results suggest another mechanism: that 1-MT exacerbates renal injury by enhancing the development of a damaging Th1 response during the nephritogenic immune response of NTN. Similarly, inhibition of IDO with 1-MT can also exacerbate Th1-driven collagen-induced arthritis, but attenuate the Th2-mediated anterior chamber-associated immune deviation by enhancing the ratio of Th1/Th2 [23,47].

In conclusion our studies demonstrate, for the first time, that up-regulation of IDO activity in serum and renal tissue of NTN mice plays a protective role in Th1-mediated crescentic GN. Inhibition of IDO with 1-MT leads to significantly exacerbated renal injury with increased glomerular crescent formation and accumulation of CD4⁺ T cells and macrophages in renal tissue. These results are attributed to systemic immune response changes in enhanced Th1 deviation and local IDO blockade effects of increased intrarenal cell proliferation, which are two distinct types of mechanism. Further studies into the exact nature of IDO in pathogenic mechanisms of autoimmune nephritis are needed. Those researches may lead to the development of novel therapies for autoimmune nephritis.