In the present study, 25 patients were investigated with renal biopsy-proven anti-GBM disease. They all had linear IgG deposition accompanied by C3 along GBM on the kidney tissue. This provides evidence that the complement system was activated in the glomeruli. In the current study, we also demonstrated the staining of C4d along GBM in all patients. This indicates the activation of classical pathway. However, on renal biopsy, C1q deposition is seldom shown in patients . This makes the role of classical complement activation controversial in the kidney injury of anti-GBM disease.
In nephrotoxic nephritis (NTN) animal models, the role of complement activation via classical pathway has been studied using the C1q or C3 deficient mice [5–7]. In C1q knockout mice, C3 deposition caused by the induction of heterologous anti-GBM antibodies is attenuated, in comparison with the wild type mice [5, 6]. This indicates a pathogenic role in the mechanism of glomerular damage. However, in other studies, a protective role for C1q against renal inflammation is revealed . One explanation for these disparate findings is the two phases, the heterologous phase and the autologous phase, in the development of kidney injure in this model. The classical pathway of complement activation may exhibit different roles in different phases . Another explanation may be the difference between the immune complex mediated glomerulonephritis in animal models and the antibody dependent cell cytotoxicity mechanism in human patients. Thus the mouse models are not reflecting human anti-GBM disease exactly and are of limited value.
In the current study, we analyzed the clinical and pathological data and outcomes of the patients with anti-GBM disease and compared the patients with and without glomerular C1q deposition in kidney. No association was identified between glomerular C1q deposition and the severity of renal injury or the outcomes of patients. Neither the percentage of crescent in glomeruli nor the level of serum creatinine was correlated with glomerular C1q deposition. These findings indicate that the classical complement activation may not play a major pathogenic role in the mechanism of glomerular damage in human anti-GBM disease. It is consistent with our previous study, in which we detected the level of circulating and urinary C1q and found little correlation with the severity of kidney injury . The present study is limited due to the small sample size. The identification from multiple centers is needed in the future.
Both the classical and lectin pathways go through the activation of C4. C4d deposition was demonstrated on GBM in the glomeruli of our patients. The level of circulating and urinary mannose-binding lectin (MBL) was detected in our previous study, which had little correlation with the severity of kidney injury . Recently, we examined the MBL in the kidneys of our patients and found that MBL deposited diffusively on the GBM and mesangial area and did not co-localize with C5b-9. However, the C1q deposit was stained in a linear pattern on GBM and co-localized well with C5b-9. These findings indicate that the complement cascade is activated through the classical pathway, whereas it does not show a direct pathogenic role to the kidney injury in human anti-GBM disease.
Animal studies using C4 knockout mice, which lack a functional classical and lectin pathway, provided strong evidence for the involvement of alternative pathway in the complement activation and development of anti-GBM disease [5, 6]. After the induction of heterogeneous anti-GBM antibodies, C4-/- mice developed albuminuria similar to that observed in wild type mice. We speculated that both the classical pathway and alternative pathway participate in the complement activation in human anti-GBM disease. The alternative pathway may play more pathogenic role in the kidney injury. The complement might be activated firstly via the classical pathway and generate the inflammatory molecules, such as C3a, C5a and C5b-9, which may be further amplified by the alternative pathway and be crucial in the effector phase of kidney injury [18–20]. These need further investigations.
The reasons for the absence of C1q deposition in routine direct immunofluorescence on renal biopsy are unclear. Anti-C1q antibodies have been demonstrated to facilitate the deposition of C1q in target organ and on cell surface [14, 15]. In patients with lupus nephritis, the serum anti-C1q antibody levels are much higher in patients with C1q deposition in the kidney than those without C1q deposit . Coremans et al. have also detected positive anti-C1q antibodies in less than 50% of patients with anti-GBM disease . In the present study, we found 60% of patients having anti-C1q antibodies, whereas the deposition of C1q in glomeruli was not more frequently shown in the kidney. There may be two reasons for it. Firstly, the circulating anti-C1q antibodies were mostly in a lower level, which makes them less effective in facilitating the deposit of C1q. In SLE, the patients with lupus nephritis present much higher titers of anti-C1q antibodies than those without kidney injury. The higher titer of anti-C1q antibodies is also an important predictor for the renal flares [22, 23]. Although we did detect the presentation of anti-C1q antibodies in the patients with anti-GBM disease, they were all in a lower level as they were in other autoimmune disease . The lower titers may prevent the role of anti-C1q antibodies. Secondly, anti-C1q antibodies may help the autologous C1q deposit in healthy mice, but induce overt renal damage only in the context of glomerular immune complex disease [14–16]. As an organ-specific autoimmune disease, circulating immune complex does not play an important role in the pathogenesis of anti-GBM disease. The target organs are much prone to be damaged by the humoral and/or cellular mechanisms locally. There might be other explanation for the absence of glomerular C1q deposition. Unlike C3d and C4d, C1q does not bind covalently to its ligands, which results in its short half-life time in vivo and easy to be cleared by macrophages .