Signaling through Up-Regulated C3a Receptor Is Key to the Development of Experimental Lupus Nephritis¹

System lupus erythematosus (SLE)³ is an autoimmune disease caused by loss of tolerance, production of autoantibodies, and deposition of complement-fixing immune complexes (ICs) in tissues. In SLE patients, complement is systemically consumed, which is reflected by decreased circulating complement levels and deposition of complement fragments in damaged tissues, suggesting that complement activation is pathogenic (1).

The MRL/Mp-Tnfrsf6^lpr/lpr strain (commonly abbreviated as MRL/lpr) is an accurate mouse model of human SLE. This strain is on the autoimmune MRL/Mp background and bears the lymphoproliferation (lpr) gene, the result of a retroviral insertion in Tnfrsf6, leading to nearly complete absence of the proapoptotic Fas protein (2, 3). MRL/Mp^+/+ (MRL/+) mice are congenic to MRL/lpr but have normal Fas protein (3) and develop autoimmunity only later in life. MRL/lpr mice share many features of human SLE, including the production of autoantibodies, leading to the presence of complement activating ICs in the circulation and deposited in tissues. As with human SLE, many organs are involved in the disease in MRL/lpr mice. However, renal involvement is a key manifestation, which is felt to be the proximate cause of death in these animals (4, 5). The earliest changes in the kidney occur by 12 wk of age and include accumulation of ICs and proliferation within the mesangial area, at which time mild proteinuria develops (6). Later in the course of the disease, ICs localize in the peripheral capillary loops, and there is proliferation of both endothelial and mesangial cells and accumulation of monocytes and neutrophils. Ultimately, crescent formation and glomerulosclerosis occur in the terminal phase of disease (5).

Complement is part of the innate immune system and can be activated through one of three pathways—the classical, alternative, or mannose-binding lectin pathways. Central to each of these pathways is the cleavage of C3, resulting in production of C3a and C3b. Upon its generation, C3b attaches covalently to ICs and has binding affinity for a variety of circulating and cell-bound proteins (7, 8). The anaphylatoxin C3a is a 78-aa peptide derived from the N terminus of the α-chain of C3 and binds the G protein-coupled C3aR present on a variety of cells, including those of hematopoietic origin such as neutrophils and monocytes (9). Some nonhematopoietic cells also bear C3aR, including cultured human renal proximal tubular epithelial cells (10).

C3a contributes to inflammatory responses such as leukocyte accumulation and enhancement of vascular permeability occurring in various infectious and noninfectious states, including IC diseases (11). However, the role of C3a in the pathogenesis of the prototypical IC disease, SLE, has not been defined. From our previous studies, we found that inhibition of complement activation at the level of C3 convertases significantly reduced renal disease in MRL/lpr mice (12, 13, 14). Given that inhibition of C3 convertases prevents generation of C3a (as well as C3b, C5a, and C5b-9), it is conceivable the therapeutic effects we observed were secondary to its effect to limit generation of C3a. To clarify the role of C3a and C3aR in SLE and lupus nephritis, in the present study, we show that C3aR is up-regulated in MRL/lpr kidneys before the onset of renal disease and rises further as disease progresses, thereby supporting its relevance. Conclusive evidence for the role of C3a signaling through C3aR was then shown through the use of a specific small molecule antagonist during the evolution of disease in MRL/lpr mice.

To assess the renal expression of C3aR mRNA in animals before and well after the development of lupus nephritis and compare these to nondiseased controls, five MRL/lpr, five MRL/+, and five BALB/c mice at 6 and 24 wk of age were sacrificed for kidney harvest. Total RNA from renal cortex was extracted and cDNA produced as described previously (14). qRT-PCR was performed with the QuantiTect SYBR Green RT-PCR kit (Qiagen) using 5′-taaccagatgagcaccacca-3′ and 5′-tgtgaatgttgtgtgcatgg-3′ as mouse C3aR primers. Expression data were normalized to 18S RNA measured contemporaneously from the same samples.

The same primers used for qRT-PCR were used to generate a riboprobe to detect mouse C3aR by in situ hybridization. The 178-bp PCR product was cloned into a pCRII-TOPO vector with the TOPO TA cloning kit (Invitrogen Life Technologies), according to the manufacturer’s instructions. The plasmid was then sequenced to determine the orientation of the PCR product. Sense and antisense probes were synthesized and labeled using the DIG RNA Labeling Kit (Roche), according to the manufacturer’s instructions. RNA probes were then purified with phenol and chloroform extraction (Invitrogen Life Technologies). Four-micrometer sections from formalin-fixed, paraffin-embedded kidneys were baked at 50°C for 1 h and deparaffinated with xylenes. After rehydration, sections were digested with 10 μg/ml proteinase K (Roche) for 10 min at 37°C, followed by 2 h of prehybridization at 37°C in hybridization buffer (50% formamide, 5× SSC buffer (Roche), 10% dextran sulfate, and 5× Denhardt’s solution (Sigma-Aldrich)). Hybridization was conducted at 37°C overnight in hybridization buffer containing 200 ng/ml sense or antisense RNA probe. After hybridization, sections were digested with 20 μg/ml RNase A for 30 min at 37°C to remove the unbound probe. After a series of washings, the hybridization signal was detected using DIG Nucleic Acid Detection Kit (Roche), followed by counterstaining with methyl green. As control, sense probes were always used at the same time and conditions as antisense probes on consecutive sections from the same sample.

To observe the effects of antagonizing C3aR in lupus nephritis, we treated MRL/lpr mice with a selective nonpeptide antagonist of C3aR, N²-[(2,2-diphenylethoxy)acetyl]-l-arginine (SB 290157, to be denoted as antagonist of C3aR (C3aRa); Calbiochem), which has an IC₅₀ of 200 nM for mouse C3aR (15). Because of a measured t_1/2of 1.5 h in mice, we opted to administer C3aRa continuously using osmotic pumps (Alzet model 2001; Durect) chosen for weekly delivery based on the solubility of C3aRa from pilot studies (data not shown) and its stability. Twenty-seven male MRL/lpr mice (The Jackson Laboratory) were randomly divided into two groups to receive C3aRa (n = 13) or vehicle alone (n = 14). Starting at 13 wk of age, mice were implanted with osmotic pumps containing 0.42 g/kg C3aRa in 0.2 ml of 50% DMSO to deliver 60 mg/kg/day C3aRa, which was the most effective dose in a chronic rodent arthritis model (15) and one that completely blocked mouse neutrophil C3aR in an ex vivo binding assay (data not shown). Osmotic pumps containing C3aRa were replaced weekly. Control animals were treated identically, except the 50% DMSO solution did not include C3aRa in the osmotic pump. Serum and urine samples were collected biweekly. At 19 wk of age, all surviving animals were sacrificed for tissue harvest. These studies were approved by the University of Chicago Animal Care and Use Committee.

BUN and urinary albumin excretion were measured biweekly from 13 to 19 wk of age. BUN and urinary creatinine concentrations were detected with a Beckman Autoanalyzer (Beckman Coulter). Urinary albumin concentrations were measured with a mouse albumin ELISA kit (Bethyl Laboratories) (16). Urinary albumin is presented as the ratio of urinary albumin to creatinine concentrations (mg/mg).

Serum anti-dsDNA Abs were measured with an ELISA as described previously (12). Briefly, 96-well plates were coated with methylated BSA (Sigma-Aldrich), followed by calf thymus dsDNA (Sigma-Aldrich). Serial dilutions of sera were incubated for 2 h at room temperature, followed by HRP-conjugated goat anti-mouse IgG (Kierkegaard and Perry Laboratories) and o-phenylenediamine peroxidase substrate (Sigma-Aldrich). The A₄₅₀ was then measured. Pooled sera from several 24-wk-old MRL/lpr mice were used as controls in each ELISA plate. The anti-dsDNA Abs are presented as relative units by plotting against the standard curve. Sera from 24-wk-old MRL/+ and BALB/c mice were used as negative controls.

To measure circulating IC levels, an ELISA method was used as described previously (13). In brief, 96-well plates were coated with human C1q (Quidel). Serum samples were serially diluted and incubated for 2 h at room temperature, followed by HRP-goat anti-mouse IgG (Jackson ImmunoResearch Laboratories) and o-phenylenediamine peroxidase substrate (Sigma-Aldrich). Pooled sera from several 24-wk-old MRL/lpr mice were used as controls in each ELISA plate. Circulating IC levels were quantified by plotting against the standard curve and presented as relative unit.

For immunofluorescence microscopy, kidney sections were snap frozen in 2-methylbutane cooled on dry ice and kept at −80°C until use. For in situ hybridization and detection of apoptotic cells, kidney sections were fixed in 10% buffered formalin and embedded in paraffin and stored at room temperature.

Four-micrometer cryostat sections were fixed in ether-ethanol and directly stained with FITC-conjugated Abs to mouse C3, IgG, IgA (Cappel), and IgM (Sigma-Aldrich). The staining intensity and distribution was semiquantitatively scored from 0 to 4 in a blinded manner as described previously (16).

To stain for neutrophils and monocyte/macrophages, slides were first fixed in ether/ethanol and then blocked with 10% normal goat serum. Slides were then sequentially incubated in rat anti-mouse neutrophil Ab (Serotec), followed by FITC-conjugated goat anti-rat IgG (absorbed with mouse Ig; Serotec) in 10% normal mouse serum for neutrophils or stained directly with FITC-conjugated rat F4/80 mAb (Serotec). To quantify cells in each mouse kidney section, neutrophils and monocyte/macrophages were counted in at least 20 low-power fields (lpf, 200×) in a blinded manner as described by others (17).

To detect apoptotic cells in tissue sections, the TACS · XL-Blue label in situ apoptosis detection kit (Trevigen) was used according to the manufacturer’s instruction. Briefly, paraffin-embedded sections were deparaffinated and rehydrated. The free 3′-OH ends of nuclear DNA fragments were labeled with BrdU using TdT. Sections were then incubated with biotinylated anti-BrdU Ab and streptavidin-HRP. A positive control treated with nuclease and unlabeled negative experimental controls were also included. Apoptotic cells, which stained intensively blue, were counted in a blinded manner from at least 20 lpf (×200)/animal.

qRT-PCR was performed on total renal cortical RNA using QuantiTect SYBR Green RT-PCR kit. As with C3aR expression analyses, data were normalized to 18S RNA. The primers used are provided in Table I.

Frozen renal cortical tissue protein was obtained and quantified as described previously (14). Equal amounts of samples were separated by SDS-PAGE and electrophoretically transferred to a polyvinylidene difluoride membrane (Millipore). Membranes were blocked with 5% nonfat milk for 1 h at room temperature and incubated with Abs to mouse C3aR (BMA Biomedicals), serine 380-phosphorylated phosphatase and tensin homologue deleted on chromosome 10 (PTEN), or serine 473-phosphorylated Akt (Cell Signaling Technology). Membranes were then incubated with peroxidase-conjugated anti-chicken IgY (Sigma-Aldrich) or anti-rabbit IgG Ab (Pierce). Chemiluminescent substrate (Pierce) was used to develop signals. Membranes were then stripped followed by probing with anti-actin (Sigma-Aldrich), anti-PTEN, or anti-Akt Abs (Cell Signaling Technology). Controls in which the primary Ab was omitted were negative.

Data are expressed as mean ± SEM and were analyzed using Minitab (version 12; Minitab) and Stata (version 8; Stata) software. Log-rank tests were used when comparing survival rates between the two groups over time. Repeated measures ANOVA was used when comparing measures in the two groups over time. For the comparison between two groups at one time point, t testing was used for parametric data, and Mann-Whitney testing was used for nonparametric data. Potential correlations among variables were determined by calculating Pearson product moment correlation coefficients and their p values.

Using in situ hybridization, the mRNA for C3aR was found primarily in tubular cells in the kidneys of normal BALB/c (data not shown) and MRL/+ mice (Fig. 1,A). In prediseased (6 wk; Fig. 1,B) and diseased (24 wk; Fig. 1,C) MRL/lpr mice, C3aR mRNA was markedly up-regulated in kidneys relative to these age-matched controls. In addition to increased tubular cell expression of C3aR mRNA, it was also expressed in glomeruli of MRL/lpr mice at both ages, which appeared to be localized to glomerular epithelial cells (Fig. 1,D, arrows). In the older MRL/lpr mice, cells infiltrating glomeruli and the interstitium also bore C3aR. In all cases, in situ hybridization with sense probe was negative in consecutive sections, confirming the specificity of the approach. Consistent with these histological data were qRT-PCR studies measuring C3aR mRNA in renal cortex, which showed a marked increase comparing both 6- and 24-wk-old MRL/lpr to MRL/+ mice (Fig. 1,E). C3aR mRNA was translated into protein as shown by Western blotting (Fig. 1 F) in which the increased expression comparing MRL/lpr to MRL/+ mice, as well as the rise over time in MRL/lpr mice, was apparent.

Given the increased expression of C3aR in lupus mouse kidneys as early as 6 wk of age, which is before the development of renal disease, we reasoned that signaling through this G protein-coupled receptor would be relevant in the evolution of lupus nephritis. To test this directly, C3aRa was used to treat MRL/lpr mice starting at 13 wk of age, at which time clinical autoimmune disease had begun (as shown by the presence of albuminuria and anti-dsDNA Abs, see below) and continued until animals were 19 wk of age. Thus, C3aRa was administered during the full evolution of autoimmune disease in MRL/lpr mice. As shown in Fig. 2, 5 of the 14 (35.7%) vehicle-treated animals died by 19 wk, while mortality was completely prevented in all 13 C3aRa-treated animals during the 6 wk of treatment (p = 0.019).

To determine the effect of C3aRa treatment on renal function in these mice, BUN levels were measured biweekly. As shown in Fig. 3, BUN levels in control animals increased as the mice aged consistent with the progressive development of renal failure. In contrast, mice treated with C3aRa had significantly lower BUN levels over time (p = 0.021). By the end of the study, 9 of 14 control animals had developed renal failure (defined by BUN levels > 60 mg/dl) from which 5 had died, while only 4 of 13 C3aRa-treated animals developed renal failure, which was not fatal in any animal. Thus, C3aR blockade with C3aRa forestalled development of renal failure, thereby preventing mortality in MRL/lpr mice.

In addition to BUN levels, we measured albuminuria as another important indicator of renal disease. Control MRL/lpr mice developed heavy albuminuria over time, which was reduced by C3aRa (p = 0.040). At the beginning of the study, all MRL/lpr mice had abnormal urinary albumin excretion (more than twice the upper limit of normal); at the study conclusion, albuminuria in control and C3aRa-treated groups was 9.0 ± 3.6 and 5.8 ± 1.5 mg/mg creatinine, respectively. C3aR blockade delayed the time of onset and the actual number of MRL/lpr mice that developed such albuminuria (defined as urinary albumin/creatinine > 2.5; p = 0.018 by repeated measures ANOVA). For example, at 15 and 19 wk, the cumulative development of marked albuminuria in control and C3aRa-treated groups was 64.3% (9 of 14) vs 30.8% (4 of 13) and 100% (14 of 14) vs 69.2% (9 of 13), respectively. Notably, all MRL/lpr mice that developed renal failure also had this degree of albuminuria, including the five control animals that died during the study.

Given that C3aR may be relevant to development of an abnormal immune response (such as in promoting Th2 immunity in murine asthma (18, 19)), circulating anti-dsDNA Abs were measured as reflective of the underlying autoimmunity in SLE. As expected by starting the study when we did, there were detectable anti-dsDNA Abs in all animals at 13 wk of age, while sera from both BALB/c and MRL/+ mice were negative in this assay. During the 6 wk of treatment, the serum anti-dsDNA in both groups rose, and there were no significant differences when we compared the two groups over time (data not shown).

The complement system is important for the clearance of ICs from the circulation, which is focused on the deposition of C4b and C3b onto ICs (20, 21). Blockade of C3aR with C3aRa, as used in this study, should not have interfered with this activation. Consistent with this, circulating ICs were no different over time comparing C3aRa-treated and control MRL/lpr mice (data not shown).

MRL/lpr mice develop many characteristics of human lupus nephritis, including glomerular deposition of ICs and complement activation products of varying composition, as well as an inflammatory cell infiltration. In 19-wk-old MRL/lpr mouse kidneys from both groups, C3, IgG, IgA, and IgM strongly deposited in the glomerular mesangium and peripheral capillary loops, which was unaffected by treatment with C3aRa (data not shown). Given that both neutrophils and monocytes bear C3aR and are relevant to lupus nephritis, we quantified the extent of their infiltration in MRL/lpr mouse kidneys. C3aRa treatment significantly reduced both neutrophil and monocyte/macrophage infiltration into kidneys of lupus mice (Fig. 4).

Because C3aR blockade with C3aRa led to protection from renal disease as characterized by a reduction in neutrophil and monocyte infiltration, renal cortical expression of mRNAs for IL-1β, IL-2, IL-18, IFN-γ, TNF-α, TGF-β1, ICAM-1, CCR2/MCP-1R, CCL5/RANTES, CCL8/MCP-2, and CXCL2/MIP-2 were measured to provide insight into potential mechanism(s) behind these findings. These 11 mediators were chosen based on published literature as being relevant to SLE or C3a signaling (11, 22, 23, 24, 25, 26, 27) or were identified in our unpublished microarray data as being affected by C3 convertase inhibition (14). In initial experiments, we studied four animals from C3aRa-treated and control animals, which had been randomly assigned beforehand but whose phenotype was reflective of their respective group. As shown in Fig. 5,A, only IL-1β and CCL5/RANTES expression was significantly changed among this diverse group of mediators. In follow-up experiments, we examined the expression of IL-1β and CCL5/RANTES in all surviving animals by qRT-PCR. Inhibition of C3aR with C3aRa led to significantly reduced expression of IL-1β (32.5 ± 8.2 and 14.3 ± 2.4 U in control and C3aRa-treated groups, respectively, p < 0.0001; Fig. 5,B) and CCL5/RANTES (23.6 ± 3.2 and 9.0 ± 2.3 U in control and C3aRa-treated groups, respectively, p < 0.0001; Fig. 5 C), confirming that the overexpression of these two cytokines appears to be mediated, at least in part, through signals delivered through C3aR.

Of note is that C3aR mRNA expression was not significantly changed when its protein was blocked with C3aRa. At 19 wk, the relative renal cortical C3aR expression in control and C3aRa-treated groups was 21.6 ± 3.0 and 28.0 ± 3.5 U, respectively. When compared with data from younger and older MRL/lpr mice (cf, Fig. 1 E), it appears that C3aR mRNA expression rises continuously over time (i.e., C3aR expression was 4.7 ± 0.8, 21.6 ± 3.0, and 62.0 ± 17.1 U in 6-, 19-, and 24-wk-old control MRL/lpr mice, respectively).

In addition to inflammatory changes, we assessed apoptosis by using TUNEL to detect and quantify apoptotic cells. In control MRL/lpr mouse kidneys, a considerable number of apoptotic cells could be found in diseased glomeruli and in the interstitium (data not shown). In contrast, mice treated with C3aRa had fewer apoptotic cells in these areas, which was significantly less than controls (p = 0.045; Fig. 6 A). The relevance of apoptosis to lupus nephritis is supported in these studies by the significant positive correlation between the number of apoptotic cells and the extent of renal disease as measured by BUN values (r² = 0.74, p = 0.006).

Given the role of activated protein kinase B (PKB)/Akt to oppose apoptosis (28, 29), we evaluated its phosphorylation at serine 473 by immunoblotting. C3aRa-treated mice had greater quantities of renal cortical phospho-PKB/Akt (serine 473) compared with control MRL/lpr mice (Fig. 6 B). Thus, in kidneys of MRL/lpr mice, C3aR-related (but Fas-independent) signals lead to apoptosis, which may involve reduction of activated PKB/Akt.

Pilot studies were performed to identify phosphoproteins altered as a result of signaling through C3aR and which therefore might promote renal disease in MRL/lpr mice. From these, we focused on the tumor suppressor gene PTEN, which is translated into the ∼55-kDa PTEN protein. The expression of phospho-PTEN (serine 380) was increased in control MRL/lpr mice relative to C3aRa-treated mice, while the total mass of PTEN protein was unaffected (Fig. 7). Thus, signaling through C3aR in kidneys of lupus mice appears to lead to increased quantities of phospho-PTEN, which can be blocked by C3aRa.

Many aspects of disease occurring in MRL/lpr lupus mice are felt to accurately reflect that which occurs in human SLE, including its renal manifestations. There is a growing appreciation that signaling of the C3a anaphylatoxin through its G protein-coupled receptor, C3aR, has relevance to a variety of inflammatory conditions. In the present study, we focused on the role of C3aR in these MRL/lpr lupus mice and found that C3aR mRNA and protein were markedly up-regulated in the kidneys of MRL/lpr mice preceding and following development of active renal disease, which indicates that the up-regulated C3aR expression is not just an accompanying phenomenon and could contribute directly to the pathogenesis of the disease. Stimulated by these findings, we administered a specific antagonist of C3aR to mice during the time renal disease was evolving. Such C3aRa-treated mice had significantly less renal disease than vehicle-treated controls, which manifested as decreased albuminuria and BUN levels. Furthermore, no C3aRa-treated animal died from renal failure compared with 35.7% of controls. These data clearly show an important role for C3a signaling through C3aR in the development of fatal renal disease in SLE.

Although normal MRL/+ and BALB/c mice had expression of C3aR mRNA in renal tubular cells, there was significantly up-regulated expression of C3aR mRNA in the kidneys of MRL/lpr mice. As might be expected from its known localization on hematopoietic cells, as well as our findings regarding its renal distribution, in these MRL/lpr kidneys, C3aR mRNA was present in inflammatory cells infiltrating the kidney and in tubular cells. In addition, C3aR mRNA was expressed in glomerular visceral and parietal epithelial cells in MRL/lpr mice. There are a number of cytokines relevant to SLE that are known to increase expression of anaphylatoxin receptors, including IL-1β (30), IL-6 (31), and IFN-α and IFN-γ (32). Many cytokines, including these, are present early in the life of MRL/lpr mice and increase with age (33, 34, 35); thus, it is conceivable that one or a combination of these cytokines are responsible for the marked increase of C3aR mRNA in MRL/lpr kidneys relative to MRL/+ controls. Although the role for C3aR (and C5aR) on infiltrating cells, including neutrophils and monocytes, has been studied in detail, the role for anaphylatoxin receptors on other cells such as those of epithelial origin is much less well understood (36). There is evidence that active C3aR is present in cultured human proximal tubular cells, and this can also be up-regulated by various cytokines (10). The exact role for glomerular and tubular epithelial expression of C3aR is speculative but certainly could involve their important role in renal function and structural integrity and/or their immunological activity (37, 38). Irrespective of the exact role for C3aR, the elevated expression in lupus kidneys suggested to us it could be of functional significance to disease.

MRL/lpr mice develop significant renal disease, which is felt to cause death of most if not all animals. Consistent with this, the five control animals that died in this study did so with renal failure. Inhibition of C3aR with a small molecule antagonist prevented the early mortality of MRL/lpr mice until 19 wk of age when the study was terminated. This effect on survival was clearly due to prevention of renal failure and is supported by the significant reduction in BUN and albuminuria as functional measures of renal disease. In addition, there was a significant reduction in neutrophil and monocyte/macrophage infiltration in kidneys of C3aRa-treated MRL/lpr mice compared with the surviving controls. This may represent the importance of direct C3aR signaling on these hematopoietic cells and/or to local effects brought about by C3aR on intrinsic renal cells to induce their recruitment into the kidney. Admittedly, these studies were performed relatively early in the course of disease in these MRL/lpr mice; as such, we can only state with certainty that C3aRa delayed renal disease and mortality, as has been the case with other forms of therapy now in clinical use, such as corticosteroids and cyclophosphamide (39). Overall, C3a signaling through C3aR likely represents one factor out of many that are operative in the genesis of lupus nephritis.

Circulating IC levels and glomerular deposition of Igs and C3 were not changed in animals receiving C3aRa, indicating that the formation and clearance of ICs, their deposition in glomeruli, and the subsequent complement activation up to C3 were not affected by this C3aR blockade. Because activation of C1, C4, and C3 individually can result in a diversity of effects, some of which may be beneficial in addition to their well-known proinflammatory effects (7, 8), this could represent an advantage to the strategy of using a C3aR antagonist. For instance, complete absence of C3 did not alter renal disease in MRL/lpr mice, which could be attributed, at least in part, to altered IC handling and their excess deposition in glomeruli (40). In contrast, the C3 inhibitor, Crry, was protective in MRL/lpr mice despite a marked alteration in IC handling, but in this case, shifting ICs away from glomeruli (13). Adding to the complexity is that factor B and D deficiencies were beneficial in MRL/lpr mice, illustrating a role for the alternative pathway in this disease (41, 42). Even signaling of C3a through its C3aR can have complex, and at times unexpected, actions in the whole animal (43); in part, this may be because C3aR has been shown to be present on APCs and T lymphocytes (44) where it could affect immunity (18). At least as measured by the anti-dsDNA Ab response, C3aRa did not appear to affect the autoimmune response in MRL/lpr mice in these studies. Thus, these data suggest that the effect of C3aR blockade to attenuate lupus nephritis and prolong survival in MRL/lpr mice was largely independent of any effect on autoantibody development, IC formation in circulation and deposition in glomeruli, and the subsequent complement activation by these glomerular-bound ICs.

Many cytokines, chemokines, and adhesion molecules are involved in the pathogenesis of lupus nephritis. IL-1β expression is up-regulated in MRL/lpr mouse kidneys (34, 45) and has been found to be important in inducing endothelial ICAM-1 and VCAM-1 expression (46). The chemokine RANTES/CCL5 is not only expressed in macrophages, T lymphocytes, and fibroblasts (47, 48) but also in kidney mesangial cells and tubular cells (49, 50). The relevance of RANTES/CCL5 to experimental lupus nephritis is suggested by its overexpression both before and during renal disease in MRL/lpr mouse kidneys (51). Furthermore, artificial overexpression of RANTES via gene transfer leads to the local recruitment of macrophages and T lymphocytes in MRL/lpr mouse kidneys (51). Taken together, these data support that IL-1β and RANTES may contribute to the development of lupus nephritis. In our study, blockade of C3aR significantly reduced renal cortical IL-1β and RANTES mRNAs in MRL/lpr mice, suggesting the beneficial effect of C3aR blockade in lupus nephritis may involve reduced production of IL-1β and RANTES.

Apoptosis can occur through receptor-mediated and mitochondrial pathways. The importance of apoptosis to SLE has become clear, given that apoptotic cells represent a vast reservoir of autoantigens that appear to be processed incorrectly in this disease (52). MRL/lpr mice lack Fas (3), a member of the TNFR family, and consequently have impaired apoptosis with expansion of autoreactive lymphocytes, although apoptosis can occur in these mice through alternatively available pathways. The complement system has relevance to apoptosis, both in the clearance of apoptotic cells (8, 53) and in its potential to lead to apoptosis, such as via direct effects of C5a (54) or C5b-9 (55, 56, 57) or by recruiting inflammatory cells that can themselves induce apoptosis (58, 59). A clear role for apoptosis has been shown in a variety of acute renal diseases (53, 56, 60, 61, 62), as well as in the changes that evolve more chronically in lupus nephritis (63, 64). In models of lupus nephritis, inhibition of apoptosis with anti-Fas ligand Abs (65) or with a pan-caspase inhibitor (66) led to protection from renal disease. In our study, we observed over a 50% decrease in the number of apoptotic cells in the renal cortex of mice treated with C3aRa compared with control mice. The finding that there was a significant correlation between apoptotic cell counts and BUN, which is arguably one of the best overall measures of renal disease severity, lends additional support to the notion that apoptosis is relevant to lupus nephritis. There is an increasing appreciation that apoptosis and inflammation can be linked rather than being mutually exclusive (67, 68). Hence, the finding that C3aRa blocked both inflammatory cell infiltration and apoptosis is not inconsistent and could even be occurring through a common mechanism (69). In addition, the mass of PKB/Akt phosphorylated on serine 473, which is the first step toward its membrane association and full activation (70), was significantly increased by C3aR blockade, which may have contributed to the reduction in apoptosis observed in these studies. Furthermore, control lupus mice had significantly greater quantities of phospho-PTEN (serine 380) than C3aRa-treated mice. Given that PTEN is an important negative regulator of PI3K, with profound implications on cell growth and survival (71, 72) and immune cell activation (73), these results are likely to be relevant to the underlying pathophysiology of lupus nephritis and how signaling through C3aR contributes to disease.

In conclusion, our study demonstrates that blockade of mouse C3aR using a specific antagonist prevented spontaneous mortality in MRL/lpr mice through its effect to lessen lupus nephritis. Because C3aR is present on both intrinsic renal cells and extrinsic inflammatory cells, these observed beneficial effects may be related to C3aRa reducing recruitment of infiltrating cells directly by blocking their C3aR and/or indirectly by reducing renal cortical expression of IL-1β and RANTES. In addition, there was less apoptosis in C3aRa-treated kidneys, which both correlated with this lesser disease severity and also with activation of the PKB/Akt pathway. Finally, our data suggest that C3aR signaling leads to PTEN inactivation, which may be detrimental and can be reversed by receptor blockade. Overall, these results clearly show there is an important role for the C3a anaphylatoxin in lupus nephritis and that blockade of C3aR represents a potentially viable treatment for this important disease. Given that this very same inhibitor effectively antagonizes human and mouse C3a receptors in similar nanomolar concentrations (15), these results could support studies in human lupus with this compound.

We thank Dr. Yasushi Nakagawa (Section of Nephrology, University of Chicago, IL) for always being willing to help with our sample measurements and Kristen Kasza (Department of Health Studies, University of Chicago, IL) for performing statistical analyses.

The authors have no financial conflict of interest.