SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Objective

To examine the ability of injection of C-reactive protein (CRP) to treat systemic lupus erythematosus (SLE) in the (NZB × NZW)F1 (NZB/NZW) mouse and to use a nephrotoxic nephritis (NTN) model to further examine the mechanism of this activity.

Methods

NZB/NZW mice were given a single injection of 200 μg of CRP prior to disease onset or after the onset of high-grade proteinuria. Mice were monitored weekly for proteinuria and monthly for anti–double-stranded DNA (anti-dsDNA) antibodies. NTN was induced by immunization with rabbit IgG followed by rabbit anti-mouse glomerular basement membrane. Proteinuria was measured daily, and renal pathology was scored. CRP was injected at the time of disease induction or 9 days later.

Results

Treatment of NZB/NZW mice with CRP prior to disease onset delayed the onset of high-grade proteinuria by 16 weeks (P < 0.0001) and prolonged survival by 13 weeks (P < 0.002). CRP treatment of NZB/NZW mice during acute disease rapidly decreased proteinuria, and the treated mice remained aproteinuric for at least 10 weeks. Control and CRP-treated mice developed similar levels of anti-dsDNA. In C57BL/6 mice, injection of CRP either before or after induction of NTN suppressed proteinuria and glomerular pathology. CRP was completely ineffective in treating NTN in interleukin-10 (IL-10)–deficient mice.

Conclusion

CRP injection suppresses inflammation in the kidney in SLE and NTN. The requirement for IL-10 in this protection suggests that CRP must rapidly initiate an IL-10–dependent antiinflammatory process. These findings suggest that a major function of CRP during the acute-phase response is to limit tissue damage and modulate acute inflammation.

Systemic lupus erythematosus (SLE) is a systemic immune complex disease of humans that affects multiple organ systems. Perhaps the most severely affected organ is the kidney, and glomerulonephritis is the major cause of morbidity and mortality in patients with SLE (1). The current standard treatment for severe lupus nephritis is the alkylating agent cyclophosphamide (2). Although treatment is generally effective, this type of nonspecific immunosuppression is associated with a variety of serious side effects, including malignancy, sterility, and cystitis.

Recently, a variety of biologic agents have been used to treat SLE (2, 3). These agents act in several ways, including interfering with collaborations between B and T lymphocytes, directly eliminating effector cells, and blocking individual cytokines. Biologic agents have had various levels of success in treating animal models of SLE. Most agents require repeated treatment with high concentrations of monoclonal antibody or protein antagonists.

The most commonly studied animal model of human SLE is the female (NZB × NZW)F1 (NZB/NZW) mouse model. The NZB/NZW mouse model shares many features with the human disease, including severe proliferative glomerulonephritis, which is the major cause of death. NZB/NZW mice have high levels of circulating immune complexes, which interact with Fcγ receptors (FcγR) in the kidney to induce nephritis (4).

The innate immune system plays an important role in autoimmunity (5). One way in which innate immune system molecules may affect autoimmunity is through the recognition and clearance of autoantigens released from apoptotic or necrotic cells (6). Other possible mechanisms for protecting against autoimmune-mediated inflammation are by altering the cytokine response to inflammatory stimuli and by redirecting the adaptive immune system (7).

C-reactive protein (CRP) is the prototypical acute-phase reactant in humans and a component of the innate immune system (8, 9). CRP binds to nuclear antigens that are the target of the autoantibodies of patients with SLE (10, 11), as well as to damaged membranes and microbial antigens. CRP activates the classical complement pathway (12) and interacts with phagocytic cells through FcγR (13–15). CRP is protective against various inflammatory states, including endotoxin shock and inflammatory alveolitis (16–18). We recently showed that CRP protection against endotoxin shock requires FcγR and is associated with FcγR-dependent induction of interleukin-10 (IL-10) synthesis by macrophages (17).

We previously reported that CRP was protective against the accelerated disease in NZB/NZW mice injected with chromatin (19). More recently, Szalai et al (20) demonstrated that NZB/NZW mice transgenic for human CRP had a delayed onset of proteinuria and enhanced survival. The ability of CRP to prolong survival in NZB/NZW mice has been attributed to increased binding and clearance of autoantigens or immune complexes. However, the ability of CRP to regulate acute inflammation suggests an alternative mechanism for its beneficial effects in SLE.

In the current study, we examined the effect of CRP on nephritis in NZB/NZW mice and found that a single injection of 200 μg of CRP not only markedly delayed the development of proteinuria, but also reversed active ongoing nephritis in this model. Both pretreatment and treatment during the acute disease significantly prolonged survival. In addition, CRP both prevented and reversed accelerated nephrotoxic nephritis (NTN), a rapid-onset immune complex disease that is induced in nonautoimmune mice by injection of antibody to glomerular basement membrane (anti-GBM). CRP protection in NTN was associated with a decrease in inflammatory and pathologic changes in glomeruli. This activity of CRP was dependent on IL-10, since CRP was ineffective in treating NTN in IL-10–deficient mice.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Animals.

Female NZB/NZW mice were obtained from The Jackson Laboratories (Bar Harbor, ME). C57BL/6 mice were obtained from the National Cancer Institute (Frederick, MD). IL-10–deficient mice on a C57BL/6 background were obtained from The Jackson Laboratories and bred at the Albuquerque Veterans Administration Animal Facility. All procedures involving animals were approved by the Institutional Review Board of the Department of Veterans Affairs Medical Center in Albuquerque.

Reagents.

Human CRP was purified from human pleural effusion fluid as described previously (10). All preparations were examined on overloaded sodium dodecyl sulfate–polyacrylamide gel electrophoresis gels to ensure purity. No bands other than the major band at ∼25 kd were seen. In addition, the preparations were examined for endotoxin by a quantitative chromogenic Limulus amebocyte lysate assay (BioWhittaker, Walkersville, MD). All preparations contained less than 0.3 ng of endotoxin/mg of protein.

To prepare anti-GBM, NZW rabbits were immunized with purified mouse glomeruli. The purification of mouse GBM and immunization of rabbits were performed essentially as described previously (21).

CRP treatment of NZB/NZW mice.

Eighteen-week-old NZB/NZW mice were divided into 2 groups (n = 10). One group was injected with saline (untreated control) and the other group was injected subcutaneously (SC) with 200 μg of CRP (early CRP treatment). The disease course was followed by measuring urinary protein levels using Chemstrips (Roche, Nutley, NJ). Grades of proteinuria were expressed as follows: 0 = none, 1+ = trace, 2+ = 30 mg/dl, 3+ = 100 mg/dl, and 4+ ≥500 mg/dl. Mice were bled monthly and serum was collected to measure autoantibody levels. When the control NZB/NZW mice had developed significant proteinuria at 30 weeks of age, 5 mice from this group with 4+ proteinuria were injected SC with 200 μg of CRP (late CRP treatment). Proteinuria was followed daily for 1 week and then weekly for the remaining course of the disease. Mice were euthanized for humanitarian reasons if they developed 4+ proteinuria accompanied by weight loss of >20%. Euthanized mice are included as deaths in the survival curves.

Autoantibody assays.

Antinuclear antibodies (ANAs) were detected by immunofluorescence on HEp-2 cell slides (Inova Diagnostics, San Diego, CA), using serial dilutions of mouse sera and Alexa Fluor 488–conjugated goat anti-mouse antibody (Molecular Probes, Eugene OR). Titers were designated as the last dilution at which staining was visible. A titer below 1:160 was considered negative. Autoantibodies to double-stranded DNA (anti-dsDNA) were determined using the Farr assay (Diagnostic Products, Los Angeles, CA), according to manufacturer's instructions, except that the mouse sera were diluted 1:5. The amount of anti-dsDNA antibodies was expressed as international units per milliliter, as determined by a standard curve supplied with the kit.

Enzyme-linked immunosorbent assay (ELISA) for anti-dsDNA antibody.

IgG antibodies to dsDNA were measured as previously described (22) using a 1:400 dilution of serum. An anti-dsDNA monoclonal antibody was used as a plate standard (23).

CRP treatment of NTN mice.

C57BL/6 mice or C57BL/6 IL-10–/– mice were injected intraperitoneally at 6–8 weeks of age (day –7) with 0.25 mg of rabbit IgG in Freund's complete adjuvant (Sigma, St. Louis, MO). NTN was induced 7 days later (day 0) by intravenous injection of rabbit anti-GBM serum. C57BL/6 mice received single injections of 20 μl or 100 μl of anti-GBM on day 0 or 3 injections of 100 μl of anti-GBM antibody on days 0, 1, and 2. IL-10–deficient mice received a single injection of 100 μl of rabbit anti-GBM on day 0. A single SC injection of 200 μg of human CRP was given for treatment. Control mice were treated with an equal volume of saline. CRP treatment was administered either at the same time as the first anti-GBM injection, or 9–10 days later, after the mice had developed 5+ proteinuria. Proteinuria was followed daily using Albustix (Bayer, Elkhart, IN). Grades of proteinuria were expressed as follows: 0 = none, 1+ = trace, 2+ = 30 mg/dl, 3+ = 100 mg/dl, 4+ = 300 mg/dl, and 5+ = >2,000 mg/dl.

Histologic studies.

Kidneys were removed and fixed for 2 hours in Bouin's solution and then transferred to 70% ethanol. They were then embedded in paraffin, and 2μ sections were cut and stained with hematoxylin and eosin (H&E) or with periodic acid–Schiff (PAS) reagent. The sections were examined by one of us (JH) in a blinded manner and scored for glomerular and other renal changes. Glomerular lesions were scored on a 4-point scale according to the percentage of glomeruli involved and the severity of the lesions, where 1 = <10% and minimal; 2 = 10–25% and mild; 3 = 25–50% and moderate; and 4 = >50% and marked. Total lesion scores were also determined, which included tubular degeneration, proteinuria. and interstitial and perivascular inflammation. Thirty glomeruli in each kidney were examined.

Statistical analysis.

Survival curves were plotted according to the method of Kaplan and Meier and compared by the log-rank test (Mantel-Haenszel test). This analysis takes into account the time of death as well as the absolute numbers of mice surviving. Proteinuria scores are expressed as the mean ± SEM. Histopathology scores are expressed as the mean ± SD. Student's 2-tailed t-test was used to compare histopathology scores. Graphic and statistical analyses were performed using GraphPad Prism software version 4.0 (GraphPad Software, San Diego, CA).

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Delay in the onset of proteinuria and reversal of ongoing proteinuria by CRP treatment of NZB/NZW mice.

The onset of proteinuria in NZB/NZW mice occurs at ∼20 weeks of age. Mice undergo a gradual, progressive increase in proteinuria, which is associated with renal failure, severe glomerulonephritis, and mortality in 50% of the animals by ∼34 weeks of age (24). In this experiment, mice received a single injection of saline (untreated control) or 200 μg of CRP at 18 weeks of age (early CRP treatment). The mice were monitored weekly for proteinuria by dipstick analysis.

At 30 weeks of age, the control mice had a markedly higher degree of proteinuria (Figure 1A). Six of these 10 mice had developed high-grade, 4+ proteinuria. Five of these 6 control mice with 4+ proteinuria were injected with 200 μg of CRP (late CRP treatment) to determine whether CRP would affect ongoing renal disease. Mice that were injected with CRP at 18 weeks had a marked delay (16 weeks) in the onset of high-grade proteinuria (>3+) as compared with the control mice (P < 0.0001).

thumbnail image

Figure 1. Effect of C-reactive protein (CRP) treatment of (NZB × NZW)F1 (NZB/NZW) mice on the development of proteinuria. Eighteen-week-old NZB/NZW mice were divided into 2 groups (n = 10). One group was injected subcutaneously (SC) with saline (untreated control), and the other group was injected SC with 200 μg of CRP (early CRP treatment). A, The disease course was followed by daily measurement of urinary protein using Chemstrips. Proteinuria (graded as 0 = none, 1+ = trace, 2+ = 30 mg/dl, 3+ = 100 mg/dl, and 4+ = ≥500 mg/dl) was followed daily for 1 week and then weekly for the remainder of the disease course. Values are the mean ± SEM. B, Survival of NZB/NZW mice treated with CRP before or after the onset of nephritis. Mice were euthanized for humanitarian reasons if they developed 4+ proteinuria accompanied by weight loss of >20%. Euthanized mice are included as deaths in the survival curves. The 50% survival time for untreated control mice was 36 weeks of age. Mice that were pretreated with CRP at 18 weeks of age had a 50% survival time of 49 weeks (P = 0.0019 versus controls). Mice treated with CRP at 30 weeks of age had a median survival of 46 weeks (P = 0.0014 versus controls).

Download figure to PowerPoint

The control mice that received no rescue dose of CRP at 30 weeks all developed high-grade, 4+ proteinuria by 34 weeks despite the fact that 5 mice with the highest grade of proteinuria were removed from this group at 30 weeks for short-term, high-dose CRP treatment. In the group of mice rescued by CRP at 30 weeks, there was a rapid decrease in proteinuria within 2 days of treatment. The mice became completely free of measurable proteinuria, and significant proteinuria did not return until 12 weeks after late CRP treatment.

Prolongation of survival after CRP treatment of NZB/NZW mice.

Survival of the mice was also examined in the 3 groups. In general, proteinuria scores were consistent with survival levels (Figure 1B). The control mice died between 35 and 38 weeks of age, and the median survival was 36 weeks. Mice pretreated with CRP at 18 weeks of age lived much longer, with a median survival of 49 weeks (P = 0.0019 versus the saline-treated group). These mice developed proteinuria more slowly than did the control mice. The control mice developed 3+ proteinuria at a median age of 26.5 weeks, as compared with 42.5 weeks for the CRP-treated mice (P < 0.0001). Interestingly, the mice that were treated with CRP when they had developed high-grade proteinuria (at 30 weeks of age) again developed proteinuria at a similar age as the mice pretreated with CRP and had a similar median survival age of 46 weeks (P = 0.0014 versus the saline-treated group).

Autoantibody levels in NZB/NZW mice treated with CRP.

The primary mechanism for the development of renal pathology in NZB/NZW mice is the deposition of immune complexes in the kidneys. These immune complexes are made up of, in part, dsDNA and autoantibodies to dsDNA, and the development of anti-dsDNA antibodies correlates in time with the development of glomerulonephritis. We examined levels of anti-dsDNA antibodies by ELISA at monthly intervals in NZB/NZW mice treated with CRP at 18 weeks of age. CRP pretreatment did not significantly affect levels of IgG anti-dsDNA antibody, which were measured monthly up to 34 weeks of age (data not shown). There was also no change in the levels of IgG anti-dsDNA antibody in mice with ongoing disease that were treated at 30 weeks of age with CRP (data not shown).

The Farr assay detects a subset of anti-dsDNA antibodies that are associated with nephritis. We therefore also determined anti-dsDNA antibodies by the Farr assay in NZB/NZW mice at 30 weeks of age. No significant differences between CRP-treated (mean ± SEM 75 ± 28 IU/ml, median 48) and control, saline-treated (mean ± SEM 222 ± 122 IU/ml, median 46) mice were seen (P = 0.495).

NZB/NZW mice may also develop other autoantibodies that contribute to immune complex nephritis. We therefore determined ANA titers in CRP-treated and control mice at 30 weeks of age. Neither the ANA titers nor the staining patterns were significantly different between the 2 groups, with a mean ± SD ANA titer of 1:384 ± 110 (median 1:320) in saline-treated mice and 1:626 ± 160 (median 1:640) in CRP-treated mice (P = 0.58).

Prevention of proteinuria by CRP treatment in the NTN model.

Although NZB/NZW mice are considered to be the classic model of lupus nephritis, other animal models of immune complex–mediated nephritis have been developed. NTN is a well-described model of nephritis that has been extensively used to explore mechanisms of immune complex–mediated kidney disease. The disease is induced by injecting animals with rabbit IgG in Freund's complete adjuvant initially and then with rabbit anti-GBM about 1 week later. Mice rapidly develop high-grade proteinuria, azotemia, and severe glomerulonephritis. We sought to determine whether CRP would also protect from immune complex disease in this model in which nuclear antigens and autoantibodies do not have an obvious role.

NTN was induced in C57BL/6 mice at 6–8 weeks of age, and proteinuria was measured daily (Figure 2A). Mice were injected SC with either saline (NTN control) or 200 μg of CRP at the same time as the injection of 20 μl of anti-GBM serum (day 0; early CRP treatment). NTN control mice underwent a rapid increase in proteinuria that remained elevated from 1 day to at least 16 days after induction. In the group of mice injected with CRP, there was a complete absence of significant proteinuria (0 or 1+). On day 10 after anti-GBM treatment, half of the NTN control mice were injected with CRP (late CRP treatment). Mice that were treated with CRP showed complete reversal of proteinuria in 1 day and remained free of significant proteinuria. The mice from this experiment were euthanized on day 16, and renal histopathology was assessed. Minimal pathologic changes were observed in the kidneys of these mice. Nearly identical results were obtained in mice injected once with 100 μl of anti-GBM (data not shown).

thumbnail image

Figure 2. Effect of C-reactive protein (CRP) treatment of C57BL/6 mice with nephrotoxic nephritis (NTN) on the development of proteinuria. C57BL/6 mice were immunized with 0.25 mg of rabbit IgG in Freund's complete adjuvant on day –7. A, NTN was induced on day 0 by intravenous injection of 20 μl of rabbit anti–glomerular basement membrane (anti-GBM) serum. A single subcutaneous (SC) injection of 200 μg of human CRP was given for treatment, which was administered either at the same time as the anti-GBM injection (early CRP treatment) or 9–10 days later, after the mice had developed 5+ proteinuria (late CRP treatment). The disease course was followed by daily measurement of urinary protein using Chemstrips. Proteinuria was graded as 0 = none, 1+ = trace, 2+ = 30 mg/dl, 3+ = 100 mg/dl, and 4+ = ≥500 mg/dl. Values are the mean ± SEM. B, NTN was induced by 3 injections of 100 μl of rabbit anti-GBM antibody on days 0, 1, and 2. A single SC injection of 200 μg of human CRP was given for treatment, which was administered either at the same time as the first anti-GBM injection (early CRP treatment) or 9 days later, after the mice had developed 5+ proteinuria (late CRP treatment). Proteinuria was measured daily using Albustix and was graded as 0 = none, 1+ = trace, 2+ = 30 mg/dl, 3+ = 100 mg/dl, 4+ = 300 mg/dl, and 5+ = >2,000 mg/dl. Values are the mean ± SEM.

Download figure to PowerPoint

To determine the effects of CRP on renal pathology in NTN, an experiment was performed using an increased dose of anti-GBM (3 injections of 100 μl each on days 0, 1, and 2) (Figure 2B). Mice were injected SC with either saline (NTN control) or 200 μg of CRP (early CRP treatment) at the same time as the first injection of anti-GBM serum. In the group of mice injected with CRP, there was a complete absence of significant proteinuria (0 or 1+) in 4 of the 6 mice for the duration of the examination period. The other 2 CRP-treated mice developed 5+ proteinuria on days 8 and 10. On day 10 after anti-GBM treatment, half of the NTN control mice were injected with CRP (4 of the 12 saline-treated mice died within the 10-day time period). The mice that were treated with CRP during acute disease showed a rapid and almost complete reversal of proteinuria.

Decrease in the level of glomerular damage after CRP treatment in the NTN model.

The mice from the experiment shown in Figure 2B were euthanized on day 11, and their kidneys were fixed and examined microscopically (Figure 3). Significant changes in the NTN control mice were observed primarily in the glomeruli on sections stained with H&E. These changes included neutrophilic infiltration, fibrin thrombi in glomerular capillaries, with karyorrhectic debris, glomerular hypercellularity, hypertrophy of glomerular mononuclear cells, protein-rich fluid exudation into the glomerulus, and bridge formation between podocytes and parietal epithelium (early crescent formation). PAS staining (results not shown) revealed a slight increase in PAS-positive material in glomerular tufts and a prominent disruption of the organization of the basement membranes in the untreated NTN mice. In contrast, glomeruli from both groups of CRP-treated mice appeared normal on PAS staining.

thumbnail image

Figure 3. Decrease in renal pathology in mice with nephrotoxic nephritis (NTN) treated with C-reactive protein (CRP). Mice were treated as described in Figure 2B. Kidneys were removed, fixed for 2 hours in Bouin's solution, transferred to 70% ethanol, and then embedded in paraffin; 2μ sections were cut and stained with hematoxylin and eosin. A, Section from a control C57BL/6 mouse. B, Section from a C57BL/6 mouse with NTN (day 11 after anti–glomerular basement membrane [anti-GBM] treatment) showing increased glomerular cellularity, with hypertrophy of glomerular cells, neutrophilic glomerular infiltrates (small arrows), intraglomerular fibrin thrombi (large arrow), fluid leakage into Bowman's space, and early crescent formation. Adjacent tubules contain protein-rich fluid (arrowhead). C, Section from a C57BL/6 mouse with NTN (day 11 after anti-GBM) that received late treatment (day 9) with 200 μg of CRP. D, Section from a C57BL/6 mouse with NTN (day 11 after anti-GBM) that received early treatment (day 0) with 200 μg of CRP. Late or early treatment of mice with 200 μg of CRP markedly attenuated the glomerular pathology associated with NTN. E, Kidney from a C57BL/6 mouse with NTN (day 11 after anti-GBM), which did not respond to CRP treatment on day 0. Changes are similar to those in the untreated NTN mouse, with neutrophilic glomerular infiltrates (small arrows), intraglomerular fibrin thrombi (large arrow), and protein-rich fluid in adjacent tubules (arrowhead).

Download figure to PowerPoint

Table 1 shows the scores for the pathologic changes observed in glomeruli on H&E-stained sections from the 3 NTN groups and the control uninjected mice. Both groups of CRP-treated mice had significantly decreased glomerular lesion scores compared with the untreated NTN control mice. Overall lesion scores, including tubular changes, perivascular and interstitial inflammation, as well as glomerular changes, were also higher in the untreated NTN mice compared with both CRP treatment groups (data not shown). Scores in the 2 mice in the early CRP treatment group that developed high levels of proteinuria (nonresponders) were similar to those in the untreated NTN mice, and their scores are shown separately in Table 1. In contrast to the decreased pathologic changes in glomeruli of CRP-treated mice, examination of frozen sections by immunofluorescence showed similar levels of both rabbit and mouse IgG in glomeruli of CRP-treated and untreated NTN mice (data not shown).

Table 1. Effect of CRP treatment on renal pathology in mice with NTN*
TreatmentGlomerular lesion scoreP
  • *

    Glomerular lesions seen on hematoxylin and eosin–stained kidney sections obtained on day 11 of nephrotoxic nephritis (NTN) (see Figure 3) were scored for the percentage of glomeruli involved and the severity of the lesions, as follows: 1 = <10% and minimal; 2 = 10–25% and mild; 3 = 25–50% and moderate; and 4 = >50% and marked. Values are the mean ± SD of 3–4 mice (30 glomeruli/mouse).

  • C-reactive protein (CRP)–treated NTN mice versus untreated NTN mice.

  • Individual scores are shown for the 2 mice in the day 0 CRP group that developed proteinuria.

Control (uninjected C57BL/6 mice)0.29 ± 0 
Untreated NTN mice2.90 ± 1.16 
Late CRP treatment (day 10)0.86 ± 0.530.024
Early treatment with CRP
 Responders1.00 ± 0.450.028
 Nonresponders3.89 and 4.00 

Abrogation of CRP-mediated protection from NTN in IL-10–deficient mice.

We have previously observed that C57BL/6 mice challenged with lipopolysaccharide after prior injection of human CRP have a marked increase in serum levels of IL-10 (17). IL-10 is a potent antiinflammatory cytokine that has profound effects on the inflammatory response and is important for the generation of regulatory T cells (25, 26). We hypothesized that the protective effect of CRP against NTN might depend on the production of IL-10. Therefore, NTN was induced in IL-10–deficient mice in the same manner as described for the C57BL/6 mice, with a single injection of 100 μl of anti-GBM. CRP treatment was the same.

The IL-10–deficient mice had a disease course that was similar to that in the C57BL/6 mice, with a rapid onset of severe proteinuria that persisted throughout the course of the experiment (Figure 4). In contrast to C57BL/6 mice, CRP provided no significant protection from disease either when injected at the time of, or 10 days after, treatment with anti-GBM antibody.

thumbnail image

Figure 4. Lack of C-reactive protein (CRP)–mediated protection from nephrotoxic nephritis (NTN) in interleukin-10 (IL-10)–deficient mice. IL-10–deficient mice on a C57BL/6 background were immunized with 0.25 mg of rabbit IgG in Freund's complete adjuvant on day –7. On day 0, 6 mice were injected subcutaneously (SC) with 200 μg of CRP (early CRP treatment) or 12 mice were injected intravenously with saline and 100 μl of rabbit anti–glomerular basement membrane (anti-GBM). A group of 4 control NTN mice with 5+ proteinuria was injected SC with 200 μg of CRP on day 9 (late CRP treatment). Proteinuria was measured daily using Albustix and was graded as 0 = none, 1+ = trace, 2+ = 30 mg/dl, 3+ = 100 mg/dl, 4+ = 300 mg/dl, and 5+ = >2,000 mg/dl. Values are the mean ± SEM.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

The major new finding of this study is that a single injection of CRP can produce a long-lasting effect on the autoimmune process in NZB/NZW mice. Despite the short half-life of injected human CRP in the mouse (∼4 hours) (27), the down-regulation of the inflammatory process in the kidney persists for more than 2 months. The finding that CRP can rapidly reverse ongoing severe nephritis suggests that this is not an effect on immunization or immune complex clearance, but rather, is a direct regulatory effect on inflammation. This process is not limited to autoimmune nephritis, but was also seen in NTN, another immune complex–mediated model of nephritis. CRP was not protective against NTN in IL-10–deficient mice, which led to the hypothesis that CRP initiates an IL-10–dependent antiinflammatory process. Taken together, these findings suggest that a major function of CRP during the acute-phase response is to limit tissue damage and modulate acute inflammation.

The use of CRP to treat NZB/NZW mice has been reported previously by us and by other investigators (19, 20). However, this is the first study to use a single injection of CRP either before the onset of disease or during the active, acute phase of nephritis. The degree of protection observed was greater than that seen in our early study in which CRP was injected multiple times in association with chromatin-coated latex beads (19) or the recent study by Szalai et al in which human CRP–transgenic mice were used (20). Continuous exposure to low levels of CRP extended survival by 8 weeks in that study (20), whereas in the current study, a single dose of CRP extended survival by 13 weeks. Furthermore, to our knowledge, this is the first study to show that CRP can reverse ongoing active nephritis in any animal model.

The pentraxins have been reported to play an important role in the prevention of autoimmune disease. One popular model suggests that the pentraxins are important for the clearance and removal of autoantigens from the circulation and for clearance of apoptotic cells, which express nuclear antigens on cell surface blebs (6, 28–30). In fact, it has been reported that a knockout of the serum amyloid P component gene in the mouse (the major pentraxin in this species) leads to a lupus phenotype (31). The experiments presented here do not rule out this process, but it is very unlikely that this mechanism explains all of the current findings, for several reasons. First, the CRP treatments did not affect the levels of autoantibodies in the NZB/NZW mice, a corollary of the “waste disposal” hypothesis. Second, the effect of CRP on ongoing nephritis was extremely rapid (on the order of a few days). Autoantibody levels or immune complexes in the kidney are not expected to change substantially in this short period of time. Third, the protection provided by CRP in the NTN model, if the 2 mechanisms are similar, could not reflect clearance of known autoantigens, since they do not play a role in this model.

It was concluded in previous studies of transgenic expression of human CRP in NZB/NZW mice that CRP protects against SLE by increasing the clearance of immune complexes and preventing their deposition in the renal cortex (20). However, it appears unlikely that this mechanism is responsible for the observed effects on ongoing immune complex disease in NZB/NZW mice. In the current studies, CRP treatment was immediately effective in mice that had already accumulated immune complexes and developed nephritis.

To determine whether the protective effect of CRP is restricted to the NZB/NZW model of SLE, we tested the effect of CRP on the onset of proteinuria in MRL/lpr mice. In ongoing experiments, CRP treatment at 6 weeks of age completely prevented the onset of significant proteinuria for at least 14 weeks and delayed the development of lymphadenopathy as well (Rodriguez W, et al: unpublished observations). In previous studies by Szalai et al (32), CRP also delayed or prevented the development of organ-specific autoimmune disease in a model of experimental autoimmune encephalomyelitis. Thus, the effectiveness of CRP treatment may extend to multiple models of autoimmunity. The use of a normal human serum protein to treat autoimmune disease is novel and is especially significant for human therapy, since this molecule is not likely to have inherent toxicity or general immunosuppressive activity. CRP is normally found at low levels, but may be present at high levels for many years in various conditions, such as rheumatoid arthritis. It has been reported that CRP levels are inappropriately low as compared with other markers of inflammation, such as the erythrocyte sedimentation rate, and that serum IL-6 levels fail to correlate with serum CRP levels in SLE (33, 34). Perhaps this inadequate CRP response contributes to the susceptibility to SLE exacerbation. Indeed, recent genetic studies of CRP gene polymorphisms suggest that low baseline expression of CRP is associated with an increased risk of SLE (35).

Because the autoimmune mouse model requires long-term observation and the mouse strain and immunologic background cannot be readily manipulated, we turned to another model of immune complex–mediated nephritis. NTN is an immune complex–mediated disease that is dependent on FcγR for its expression (21, 36). Although NTN most closely resembles Goodpasture's syndrome in humans, the renal disease in NTN resembles SLE in many respects and has been used as a surrogate model for lupus nephritis. In mice with NTN, CRP both prevented the development of high-grade proteinuria and reversed ongoing proteinuria. CRP-treated mice showed rapid improvement in glomerular lesions as well as reversal of proteinuria. Examination of frozen sections of kidney by immunofluorescence showed similar levels of both rabbit and mouse IgG in glomeruli of CRP-treated and NTN control mice. These results are most consistent with suppression of the inflammatory response in the kidney by CRP. The finding that CRP had similar effects in NTN and SLE suggests that the effect of CRP on nephritis is not limited to autoimmune disease, but rather, immune complex–mediated disease in general.

Our results using IL-10–deficient mice indicate that IL-10 is required for CRP-mediated protection from NTN. IL-10 is an antiinflammatory cytokine with the ability to inhibit the production of inflammatory cytokines by a wide variety of cells (26). In several cases, it has been demonstrated that IL-10 may modulate the severity of immune complex–mediated diseases, including lupus nephritis (37) and NTN (38–40). In our previous study on endotoxin shock, CRP-mediated protection was associated with its ability to enhance IL-10 production by macrophages (17). The enhancement of IL-10 production by CRP both in mice and in bone marrow macrophage cultures required the FcR γ-chain. In other studies, we have presented evidence that CRP binds to the γ-chain associated receptor FcγRI (14). In a mouse model of experimental autoimmune encephalomyelitis, Szalai et al (32) showed that transgenic expression of human CRP decreased disease activity and increased IL-10 production in vitro. Immune complex–mediated inflammation can also be regulated by FcγRIIb-dependent mechanisms (41). CRP protection in endotoxin shock also requires this receptor (17). However, our preliminary results indicate that CRP can protect mice from NTN in the absence of FcγRIIb (Rodriguez W, et al: unpublished observations).

The protection provided by CRP was long-lasting when injected either before the onset of nephritis or during the height of nephritis. The mechanism by which CRP induces this long-lasting suppression is unknown. However, one must invoke a long-lived cell type, which could provide persistent suppression of the response to immune complexes. Macrophages may persist in tissues for long periods of time and can, under some circumstances, suppress inflammation. Another possibility is that exposure of T cells to antigen in the presence of IL-10 may lead to the development of a T regulatory cell as described by Groux et al (42). The search for these cells is ongoing.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES