Low Levels of Factor H Family Proteins During Meningococcal Disease Indicate Systemic Processes Rather Than Specific Depletion by Neisseria meningitidis

Abstract

Neisseria meningitidis, the causative agent of meningococcal disease (MD), evades complement-mediated clearance upon infection by 'hijacking' the human complement regulator factor H (FH). The FH protein family also comprises the homologous FH-related (FHR) proteins, hypothesized to act as antagonists of FH, and FHR-3 has recently been implicated to play a major role in MD susceptibility. Here, we show that the circulating levels of all FH family proteins, not only FH and FHR-3, are equally decreased during the acute illness. We did neither observe specific consumption of FH or FHR-3 by N. meningitidis, nor of any of the other FH family proteins, suggesting that the globally reduced levels are due to systemic processes including dilution by fluid administration upon admission and vascular leakage. MD severity associated predominantly with a loss of FH rather than FHRs. Additionally, low FH levels associated with renal failure, suggesting insufficient protection of host tissue by the active protection by the FH protein family, which is reminiscent of reduced FH activity in hemolytic uremic syndrome. Retaining higher levels of FH may thus limit tissue injury during MD.

Keywords: FHR; Neisseria meningitidis; complement; factor H; meningococcal disease.

Figure 1

FH family protein levels are low at the acute stage of MD. (A, B) Differences in FH, FHR-1/1 homodimers, FHR-1/2 heterodimers, FHR-2/2 homodimers, FHR-3, FHR-4A and FHR-5 homodimers as assessed at the acute stage (samples obtained during the first or second day of hospitalization, n = 106) compared to levels at convalescence (n = 91). Two children appeared to carry a homozygous CFHR3/CFHR1 deletion, as evidenced by the lack of either protein in their convalescent sample. They were excluded from the analysis for FHR-1/1, FHR-1/2, FHR-2/2 and FHR-3 (n = 104 and n = 89 for acute stage and convalescence). Acute serum samples comprised 88 samples drawn at day 1 and 18 samples drawn at day 2 of hospitalization, for patients of whom no day 1 sample was available. Levels of FHR-2/2 were calculated based on FHR-1/1 and FHR-1/2 levels. Shaded area indicates 95% range in healthy patients, with dashed line indicating the median ****p < 0.0001. Scatter dot plots depict median and interquartile range (IQR) as red lines. Statistical significance was tested using a Mann-Whitney test. (C) Correlations (r) between the relative decreases of FH family proteins (ratios between acute and convalescent levels, by dividing acute levels over convalescent levels) were assessed using Pearson’s measure of association, followed by the Benjamini-Hochberg procedure to control for the false discovery rate (FDR, set to 0.05). Blue shades indicate, from light to dark: p < 0.05; p < 0.01; p < 0.001; and p < 0.0001. (D) Examples of correlations in (C), showing relative decrease in FH levels versus relative decrease in FHR levels. A/C ratio, acute/convalescent ratio; Conv., convalescent.

Figure 2

FH family protein level dynamics during the acute stage of MD. (A) FH family protein levels in paired samples (n = 17) as assessed during the first four days of infection (day 1 until day 4), compared with the concentration at convalescence (C). Shaded area indicates 95% range in healthy patients, with dashed line indicating the median. Friedman test followed by Dunn’s multiple comparisons test, with every acute stage dataset compared to the levels found at convalescence. ****p < 0.0001; ***p < 0.001; **p < 0.01; *p < 0.05; ns, not significant. (B) FH family protein levels as in (A), normalized to the levels found at convalescence. (C) Total IgG levels in unpaired samples (maximum n = 16) as assessed during the first four days of infection and at convalescence. The 95% range in healthy patients is not depicted, due to variability of total IgG levels during childhood. Lines depict median and IQR.

Figure 3

FH family proteins per clinical syndrome. Protein levels of FH, FHR-1/1, FHR-1/2, FHR-3, FHR-4A and FHR-5 at the acute stage (first sample obtained during hospitalization, n = 106), according to the diagnosed clinical syndrome: meningococcal meningitis (M, n = 14), meningococcal septicemia (S, n = 75), or both (M+S, n = 17). Statistical significance was tested using a Kruskal-Wallis test, followed by a Dunn’s multiple comparisons test. Lines depict median and IQR. **p < 0.01; *p < 0.05; ***p < 0.001; ns, not significant.

Figure 4

Associations of acute stage FH family protein levels with clinical and laboratory parameters. Pearson correlation coefficients (r), considering twelve severity markers and seven FH family proteins (including the different dimers). Correlations were assessed using Pearson’s measure of association, followed by the Benjamini-Hochberg procedure to control for the false discovery rate (FDR, set to 0.05). Blue shades indicate, from light to dark: p < 0.05; p < 0.01; p < 0.001; and p < 0.0001. WCC, white cell count; aPTT, activated partial thromboplastin time; INR, international normalized ratio; CRP, C-reactive protein; PIM, pediatric index of mortality; GMSPS, Glasgow Meningococcal Septicemia Prognostic Score.

Figure 5

FH family proteins are low in patients who receive renal support. Serum levels of FH, FHR-1/1, FHR-1/2, FHR-2/2, FHR-3, FHR-4A and FHR-5 at the acute stage of patients who did (n = 15) or did not (n = 90) receive renal support. Both surviving and non-surviving patients are included here. Statistical significance was tested using a Mann-Whitney test. Lines depict median and IQR. ****p < 0.0001; ***p < 0.001; **p < 0.01; ns, not significant.