Introduction

A single-nucleotide polymorphism (SNP) in the CFH gene (rs1061170), which encodes the complement factor H (FH) protein, increases AMD risk by ~7-fold in those homozygous for the variant [1]. The monomeric form of C-reactive protein (mCRP) and the membrane attack complex of complement (MAC) are both deposited in eyes with AMD and with high-risk CFH genotypes [2, 3]. Here we sought to assess whether the CRP and MAC deposition described in human choroid/Bruch’s membrane is due to systemic or local factors by directly comparing circulating and ocular levels of CRP and MAC in the same donors.

Methods

Eyes and post-mortem serum samples from the same donors were obtained from the Iowa Lions Eye Bank (Iowa City, IA). Serum and RPE/choroid tissue from 18 donors was employed for each of the ELISA experiments. All donors were genotyped for the Y402H SNP [3]. Equal amounts of total protein (tissue) or equal volumes of serum were evaluated using ELISA, according to the manufacturer’s instructions (CRP Kit, RayBiotech, Norcross, GA; C5b-9 MAC Kit, Quidel, San Diego, CA).

Results

When CRP levels in RPE/choroid were compared with serum levels from the same donors, a trend toward increased CRP levels for genetically high-risk eyes compared to low-risk eyes was observed, which was consistent with our and others’ previous immunohistochemical findings for monomeric CRP [3,4,5] (Fig. 1a), although this difference in total CRP did not reach statistical significance in the current study. Serum CRP levels did not differ significantly based on CFH genotype (Fig. 1b). When directly comparing tissue to serum CRP levels for each donor, the slope of the regression line was found to be positive and significantly non-zero, suggesting a modest correlation between tissue and serum CRP levels (p < 0.05, R2 = 0.3543; Fig. 1c). These data indicate a weak, but detectable, relationship between CRP levels in tissue and serum from the same individuals. This correlation suggests that the mCRP present in the aging eye is derived from circulating pCRP. This is also consistent with the absence of detectable CRP mRNA in retina, RPE, or choroid (unpublished observations).

Fig. 1
figure 1

CRP (a–c) and sC5b-9/MAC (d–f) levels in the RPE/choroid and serum of genotyped human donors. Box-and-whisker plots, showing median, interquartile range, and minimum and maximum values, for RPE/choroid tissue (a, d) and serum (b, e) for low-risk (YY) and high-risk (HH) CFH donors. Direct comparisons between serum and tissue protein levels reveal a regression line with a significantly non-zero slope for CRP (R2 = 0.3543; c) but not for MAC (R2 = 0.0377; f). *p < 0.05

Full size image

MAC levels were found to be significantly elevated (p < 0.05) in the RPE/choroid tissue for high-risk compared to low-risk CFH donors (Fig. 1d). Notably, this is the third independent human cohort showing this genotype–phenotype relationship. In contrast, we did not observe a difference in circulating sC5b-9 levels in the serum based on genotype (Fig. 1e). Furthermore, there is no apparent correlation between tissue and serum MAC/sC5b-9 levels when directly compared for each donor (Fig. 1f).

Conclusions

Accumulation of complement complexes in the choroid is a tissue-specific event and is not proportional to circulating complexes. These data suggest that serum levels of the MAC are not related to tissue levels and most likely cannot be used as a predictive measure for AMD or for tissue-associated complement injury in AMD. Taken together, these data lend further support for the notion that MAC accumulation in AMD is choriocapillaris-specific and that therapies addressing MAC injury to the choroid will be beneficial in AMD.

Summary

What was known before:

  • CRP and MAC accumulate in aging choriocapillaris.

What this study adds:

  • In the case of CRP, levels are related to circulating CRP in the serum. In the case of MAC, levels in choriocapillaris are independent of those in serum, indicating activation of complement in the local environment.