Gain-of-function factor H–related 5 protein impairs glomerular complement regulation resulting in kidney damage

Genetic variation within the factor H–related (FHR) genes is associated with the complement-mediated kidney disease, C3 glomerulopathy (C3G). There is no definitive treatment for C3G, and a significant proportion of patients develop end-stage renal disease. The prototypical example is CFHR5 nephropathy, through which an internal duplication within a single CFHR5 gene generates a mutant FHR5 protein (FHR5mut) that leads to accumulation of complement C3 within glomeruli. To elucidate how abnormal FHR proteins cause C3G, we modeled CFHR5 nephropathy in mice. Animals lacking the murine factor H (FH) and FHR proteins, but coexpressing human FH and FHR5mut (hFH-FHR5mut), developed glomerular C3 deposition, whereas mice coexpressing human FH with the normal FHR5 protein (hFH-FHR5) did not. Like in patients, the FHR5mut had a dominant gain-of-function effect, and when administered in hFH-FHR5 mice, it triggered C3 deposition. Importantly, adeno-associated virus vector-delivered homodimeric mini-FH, a molecule with superior surface C3 binding compared to FH, reduced glomerular C3 deposition in the presence of the FHR5mut. Our data demonstrate that FHR5mut causes C3G by disrupting the homeostatic regulation of complement within the kidney and is directly pathogenic in C3G. These results support the use of FH-derived molecules with enhanced C3 binding for treating C3G associated with abnormal FHR proteins. They also suggest that targeting FHR5 represents a way to treat complement-mediated kidney injury.

The complement system is an important component of the immune response to pathogens, particularly meningococcal infection. Complement activation is tightly regulated to prevent its effectors from damaging host tissue. Impaired control of activation, termed complement dysregulation, is associated with tissue injury, including age-related macular degeneration and renal disease. Complement-mediated kidney damage is exemplified by thrombotic microangiopathy in atypical hemolytic uraemic syndrome and glomerular damage in C3 glomerulopathy (C3G) and IgA nephropathy (IgAN).Complement factor H (FH) is a plasma protein that down-regulates C3 activation through the complement alternative pathway. The essential role of FH is illustrated by homozygous FH-deficient patients who have acquired severe C3 deficiency due to uncontrolled C3 consumption (1). The FH protein family includes five factor H–related proteins (FHR1 through 5), and all are composed of subunits called short consensus repeat (SCR) domains. While FH contains both regulatory and binding SCR domains for the activated C3 fragment C3b, the FHR proteins contain only binding domains, suggesting different functions. The importance of FHR proteins in renal pathology is derived from the associations between susceptibility to C3G and abnormal FHR proteins (2–8). C3G is characterized by dominant glomerular C3 deposition and glomerular damage (9). The prototypic example of FHR-associated C3G is CFHR5 nephropathy (3, 10). Affected individuals have a heterozygous internal duplication within the CFHR5 gene. The normal FHR5 protein consists of 9 SCR domains, whereas the abnormal FHR5 protein (FHR5mut), due to duplicated exons encoding the first two SCR domains, consists of 11 SCR domains. There are now several examples of abnormal FHR proteins and C3G (2, 4–8).Both FHR1 (11–13) and FHR5 (14, 15) influence susceptibility to IgAN, a glomerular disorder characterized by galactose-deficient IgA1 immune deposits and C3 deposition. How FHR proteins influence glomerular C3 deposition in both C3G and IgAN remains poorly understood. FHR1 and FHR5 proteins can antagonize the ability of FH to down-regulate C3 activation in vitro (7, 16). However, FHR1, FHR4, and FHR5 can also promote C3 activation in vitro independently of FH (17–19). From these in vitro observations, it can be hypothesized that the degree of C3 deposition in response to a complement-activating trigger within the kidney (e.g., IgA1 immune deposits) depends on the relative interactions between local complement activation and either FH (inhibition of activation) or the FHR proteins (promotion of activation). The degree of complexity in this system is also governed by context-specific interactions between the FHR proteins and surface glycans (20). However, the lack of appropriate in vivo models (21) due to interspecies differences in the FHR proteins has prevented researchers from modeling FHR-associated renal pathology to identify mechanisms of injury and therapeutics, an approach that has been very successful for FH-associated renal pathologies (22–30).To overcome these limitations, we developed murine strains consisting of 1) mice lacking the entire 664 kb FH-FHR locus and therefore deficient in FH and all the FHR proteins (delFH-FHR), 2) mice lacking the 537 kb FHR locus and therefore expressing normal FH but not the FHR proteins (delFHR), and 3) mice expressing human FH (hFH) and FHR5/FHR5mut in the absence of the mouse proteins. Using these unique models, we show that delFH-FHR animals develop a spontaneous C3G, that the absence of the FHR proteins in the delFHR strain did not result in spontaneous renal disease, and that delFH-FHR animals transgenically expressing hFH with the FHR5mut associated with CFHR5 nephropathy, spontaneously develop C3G, recapitulating the key features of the human disease. Finally, we demonstrate that adeno-associated virus (AAV) vector delivery of a homodimeric mini-FH (HDM-FH) molecule, previously shown to be efficacious in reducing glomerular C3 in FH-deficient mice (30), can significantly reduce glomerular C3 in our CFHR5 nephropathy mouse model and, thus, could be a future therapeutic approach for treating C3G associated with abnormal FHR proteins.