Professional Documents
Culture Documents
1 s2.0 S2468024919300890 Main
1 s2.0 S2468024919300890 Main
1 s2.0 S2468024919300890 Main
Background: Experimental studies support a role of complement activation in diabetic nephropathy (DN),
yet few clinical correlates exist. We evaluated urinary levels of sC5b-9 membrane attack complex (MAC) in
patients with overt DN, and examined its association with the glomerular filtration rate (GFR) decline,
proteinuria, and inflammatory biomarkers. We explored different complement pathways and compared
our findings to autoimmune glomerulonephritis.
Methods: We prospectively followed 83 patients with DN and obtained repeated measurements of pro-
teinuria, complement fragments (sC5b-9, C4a, C1q, mannose-binding lectin–associated serine protease
[MASP]-1, and factor Bb), monocyte chemoattractant protein-1 (MCP-1), and transforming growth factor
(TGF)-b1. We assessed independence and interactions using general linear models and repeated mea-
sures analyses and compared levels with subjects with active focal and segmental glomerulosclerosis,
ANCA-associated vasculitis, and membranous and IgA nephropathies (n ¼ 63).
Results: The diabetic cohort had an initial GFR of 25 9 ml/min per 1.73 m2 and a renal function decline of
2.9 3.0 ml/min per 1.73 m2 per year. All complement biomarkers were strongly intercorrelated and
associated with biomarker inflammation and fibrosis, proteinuria, and the rate of renal function decline.
There was a significant interaction (P ¼ 0.03) between the level of proteinuria and urinary sC5b-9: in in-
dividuals with higher levels of urinary MAC, the relationship between proteinuria and the rate of renal
function decline was more pronounced than in those with low urinary MAC. Finally, patients with DN had
levels of urinary sC5b-9 comparable to autoimmune glomerulonephritis, when stratified by the level of
proteinuria.
Conclusion: Urinary MAC is present in patients with overt DN at levels comparable to autoimmune
glomerulonephritis and correlates with the GFR decline, supporting that complement activation and its
measurement are clinically relevant in DN.
Kidney Int Rep (2019) 4, 797–805; https://doi.org/10.1016/j.ekir.2019.03.004
KEYWORDS: complement activation; diabetic nephropathy; fibrosis; inflammation; proteinuria; glomerulonephritis
ª 2019 International Society of Nephrology. Published by Elsevier Inc. This is an open access article under the CC BY-
NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Human Cytokine/Chemokine plex kit (Millipore, St. Table 1. Demographics and clinical features of the DN cohort
Charles, MO) on a multiplex platform (Eve Technolo- (n ¼ 83)
gies Corp, Calgary, Alberta, Canada). The lower At enrollment
threshold sensitivities of the assays were 2 and 10 ng/l Female (%) 20
for MCP-1 and TGF- b1. We report the values of uri- Ethnicity (%): white, Asian, African, Middle Eastern 87, 4, 4, 5
69 10
nary complement biomarkers in mg/mmol of creatinine
Age (yr)
GFR (ml/min per 1.73 m2) 25 9
and MCP-1 and TGF- b1 in ng/mmol of creatinine. Proteinuria (g/mmol creatinine) 0.10 (0.05–0.32)
Urinary MCP-1 (ng/mmol creatinine) 35 (22–67)
Urinary TGF-b1 (ng/mmol creatinine) 1.4 (0.0–3.0)
Statistical Analyses Urinary sC5b-9 (mg/mmol creatinine) 1.2 (0.3–6.8)
During follow-up
Normally distributed variables are presented as means
Systolic blood pressure (mm Hg) 142 18
with SDs. Because the distributions of all urinary bio- Diastolic blood pressure (mm Hg) 69 12
markers were nonparametric, they are presented as Mean arterial blood pressure (mm Hg) 94 12
medians with interquartile range, and correlation co- Antihypertensive medication (n) 4 (3–5)
efficients were calculated using Spearman’s rho. We RASB (%) 73
also categorized urinary sC5b-9 and proteinuria into HbA1C (%) 7.3 (6.5–7.9)
tertiles to illustrate our findings and used the Pearson Length of follow-up (yr) 2.1 (1.6–2.8)
Proteinuria (g/mmol creatinine) 0.13 (0.04–0.32)
correlation to perform trend tests with the rate of renal Urinary MCP-1 (ng/mmol creatinine) 40 (21–80)
function decline. Urinary TGF-b1 (ng/mmol creatinine) 1.7 (0.7–3.4)
To further assess the association between the GFR Urinary sC5b-9 (mg/mmol creatinine) 1.9 (0.5–10.4)
slope (dependent variable) and urinary sC5b-9 sepa- Rate of renal function decline (ml/min per 1.73 m2 per year) 2.9 3.0
rately from the overall proteinuria, we performed HbA1C, glycosylated hemoglobin; MCP, monocyte chemoattractant protein; RASB, renin
angiotensin system blockade; TGF, transforming growth factor.
general linear models using log-transformation (to Values are expressed as percent, mean SD, or median (interquartile range).
respect the assumptions of linear regression) of these 2
biomarkers as covariates with and without an interac-
tion term. RESULTS
We also verified these analyses using the estimated Cohort
GFRs, instead of the slopes, using repeated-measures The clinical and biochemical characteristics of the
analysis of variance. Because GFR estimations were study participants are given in Table 1. A total of 83
not done at predefined times, we established GFRs at patients were included in the cohort. All but 2 sub-
each 6-month interval based on the available data. We jects had type 2 diabetes, 80% were male and 87%
ran mixed repeated-measures analysis of variance white with a mean age of 69 10 years. The initial
adding to the “within-subject variable” (GFRs at GFR was 25 9 ml/min per 1.73 m2 and mean arterial
different time) a “between-subject variable”: one model pressure was 94 12 mm Hg. Overall, we obtained a
using log-transformed proteinuria, another using log- median 12 serum creatinine, 6 blood pressure, 3
transformed sC5b-9 and finally one using both bio- HbA1C, and 3 urine measurements per patient. Pa-
markers. We present the interactions with the time tients were followed for 2.1 (1.6–2.8) years. During the
effect, that is, to test whether the change in GFR over study, patients received on average 4 antihypertensive
time was influenced by that variable and report the F drugs and renin angiotensin system blockade in 73%
and partial eta (hp2) scores. Because the Mauchly’s tests of cases. Median HbA1C was 7.3%, 46% of partici-
of sphericity P values were always <0.001, we used the pants had an HbA1C #7%, and 12% a mean HbA1C
Greenhouse-Geisser correction for the degree of >8.5%. The median protein-to-creatinine ratio was
freedom. 0.13 (0.05–0.32) g/mmol. The mean decline in renal
In the subanalyses presented, we also separated function was 2.9 3.0 ml/min per 1.73 m2 per year.
individuals with low and high sC5b-9, C4a, and factor Only 2 individuals had a renal biopsy, and both
Bb by the point on a receiver operating characteristic demonstrated DN.
curve maximizing the sensitivity and specificity
product. To perform this analysis, the receiver Urinary Complement Biomarker Associations
operating characteristic curve used the GFR slope With the Proteinuria, Blood Pressure, Inflam-
divided into 2 groups using the median value as the matory, and Fibrosis Biomarkers and the Rate of
outcome. Two-tailed P values less than 0.05 were Renal Function Decline
considered statistically significant. Analyses were Urinary measurements of sC5b-9, factor Bb, C4a, and
performed using SPSS software (version 24; SPSS Inc., MASP-1 had a skewed distribution, more so compared
Chicago, IL). with proteinuria (Figure 1). The median average
Kidney International Reports (2019) 4, 797–805 799
CLINICAL RESEARCH K Pelletier et al.: Complement Activation in Diabetic Nephropathy
Figure 1. Distribution of urinary sC5b-9, C1q, factor Bb, C4a, mannose-binding lectin–associated serine protease (MASP)-1, and proteinuria
based on the average of all measurements per patients.
sC5b9-to-creatinine ratio in our cohort was 1.9 (0.5– urinary sC5b-9. To illustrate this, we dichotomized
10.4) mg/mmol. All complement biomarkers showed a urinary sC5b-9 based into “low” and “high” levels
strong correlation to each other (all P values # 0.001), based on the point on the receiver operating charac-
as well as to proteinuria. Urinary sC5b-9 also showed a teristic curve maximizing the sensitivity and specificity
correlation with the mean arterial pressure (correlation product (2 mg/mmol). We can see in Figure 3, that the
coefficient 0.29; P ¼ 0.009, Figure 2b), with the num- relationship between proteinuria and the rate of renal
ber of antihypertensive medications (correlation coef- function was pronounced in the presence of “high”
ficient 0.42; P < 0.001, Figure 2c), and with MCP-1 and urinary MAC (P < 0.001), whereas it was modest and
TGF-b1 (Figure 2d and e). sC5b-9 levels did not not statistically significant with “low” urinary MAC
correlate with the initial GFR. Finally, all complement (P ¼ 0.13).
biomarkers were strongly associated with the rate of We repeated these analyses using a repeated-
renal function decline by univariate analysis (Table 2 measures analysis of variance instead of the slope
and Figure 2f). and found similar results. Individually, proteinuria
(F2.7,222.0 ¼ 6.4, P ¼ 0.001, hp2 ¼ 0.07) and urinary
Interaction Between Urinary MAC and Protein- sC5b-9 (F2.7,221.2 ¼ 3.8, P ¼ 0.01, hp2 ¼ 0.05) were
uria With GFR Decline associated with the degree of change in GFR over
Because all the biomarkers of the complement studied time. When both were entered in the same model, the
are proteins, we sought to distinguish urinary sC5b-9 interaction between them was statistically significant
from the effects of proteinuria and performed a gen- (P ¼ 0.048). We therefore separated into low and
eral linear model addressing independence and in- high urinary MAC, as we did above when using the
teractions. Although a sC5b-9 effect independent of slope, and found that proteinuria was not associated
proteinuria was not found, an interaction existed be- with the change in GFR in the low MAC group
tween urinary MAC and proteinuria (P ¼ 0.03) indi- (F2.7,120.0 ¼ 1.1, P ¼ 0.35, hp2 ¼ 0.02), whereas it was
cating that the relationship of proteinuria with the rate in the high MAC group (F2.4,85.6 ¼ 6.7, P ¼ 0.001,
of renal function decline was modified by the level of hp2 ¼ 0.16).
800 Kidney International Reports (2019) 4, 797–805
K Pelletier et al.: Complement Activation in Diabetic Nephropathy CLINICAL RESEARCH
Figure 2. Associations between tertiles of urinary sC5b-9 (C1–C3) and proteinuria (a), mean arterial pressure (b), number of blood pressure
medications (c), urinary MCP-1 (d), TGF-b1 (e), and the rate of renal function decline (f). Values are expressed either as mean SD (full line) or
median with 25th and 75th percentile (dashed lines). All trend tests between tertiles of sC5b-9/creatinine and each outcome were statistically
significant (all P < 0.01).
Comparison of Different Pathways of Activation Bb,” and “low C4a-high factor Bb” were not statisti-
To further investigate how the activation of the clas- cally different.
sical and/or lectin pathways differed clinically from
that of the alternative pathway, we separated in- Comparison With Autoimmune GN
dividuals into categories of C4a and factor Bb levels We compared our findings with recently diagnosed
based on cutoffs determined by receiver operating autoimmune GN and measured these biomarkers in
characteristic curves (1.15 and 2.2 mg/mmol, for C4a urine samples from 15 subjects with focal and
and factor Bb, respectively). The slopes were 1.8 segmental glomerulosclerosis, 17 with IgA nephrop-
2.4, 1.4 1.6, 2.7 2.9, and 5.1 2.9 ml/min athies, 14 with membranous nephropathies, and 18
per 1.73 m2 per year with “both biomarkers low” (n ¼ with ANCA-associated vasculitis with a median
35), “low C4a-high factor Bb” (n ¼ 7), “high C4a-low proteinuria of 0.43 (0.19–0.74), 0.15 (0.11–0.43), 0.79
factor Bb” (n ¼ 16), and “both biomarkers high” (0.20–1.32), and 0.14 (0.06–0.65) g/mmol of creati-
(n ¼ 25), respectively (1-way analysis of variance, P < nine, respectively. The urinary excretion of sC5b-9
0.001 with “both high” faster than each of the 3 other and its corresponding proteinuria varied according
groups). The groups “both low,” “high C4a-low factor to the underlying glomerular disease. Figure 4 illus-
trates that for equal proteinurias, urinary sC5b-9 in
Table 2. Associations between urinary biomarkers and the rate of DN are comparable to autoimmune conditions in
renal function decline which complement activation is implicated in the
Variable Correlation coefficient P pathogenesis.
Proteinuria 0.50 <0.001
MCP-1 0.55 <0.001
DISCUSSION
TGF-b1 0.39 <0.001
sC5b-9 0.38 <0.001 The current study explored renal complement activa-
C4a 0.36 0.001 tion in DN in the clinical setting. We found that
MASP-1 0.34 0.002 complement components sC5b-9, C4a, C1q, MASP-1,
C1q 0.34 0.002 and factor Bb are present in the urine and strongly
Factor Bb 0.33 0.003
correlate with proteinuria, urinary MCP-1 and TGF-b1,
MASP, mannose-binding lectin–associated serine protease; MCP, monocyte chemo- blood pressure control, and the rate of renal function
attractant protein; TGF, transforming growth factor.
The distributions of biomarkers were nonparametric and tested using Spearman’s Rho. decline. In individuals with higher levels of urinary
Kidney International Reports (2019) 4, 797–805 801
CLINICAL RESEARCH K Pelletier et al.: Complement Activation in Diabetic Nephropathy
Figure 3. Interaction between proteinuria and the rate of renal function decline according to urinary sC5b-9 levels. Values are expressed as
mean SD.
MAC, the relationship between proteinuria and the rate the lectin, classical, and alternative pathways. Each of
of renal function decline was pronounced, unlike when these has its own triggers, but their activation leads to
urinary MAC was low, demonstrating a clinically the production of complement C3 and C5 convertases,
observable effect. In addition to high levels of urinary ultimately generating the C5b-9 complex, responsible
MAC, individuals with both elevated C4a and factor Bb for cell and organ damage. Glomerular proteinuria can
had a faster rate of renal function decline. Finally, at activate the complement in the tubules and the kidney
comparable levels of proteinuria, the patients with DN has the ability to synthesize most of the components of
had similar or higher levels of urinary sC5b-9 compared complement cascade.20,21 Evidence in DN points to-
with other active autoimmune glomerulonephritis. ward complement activation via the lectin and classical
Taken together, these findings support a meaningful complement pathways. Recent studies have demon-
clinical role of complement activation in agreement strated a relationship between high levels of circulating
with experimental findings. mannose-binding lectin, which initiate the complement
The complement system is an important component lectin pathway, and the development of persistent
of the innate immune system with 3 main pathways: proteinuria and a more rapid decline in renal function
Figure 4. Relation between urinary sC5b-9 and proteinuria in diabetic nephropathy (DN) and autoimmune glomerulonephritis (GN). The
purpose of this figure was to simultaneously compare urinary sC5b-9 levels stratified by proteinuria in different GN and demonstrate that the
magnitude of sC5b-9 in DN is comparable to autoimmune disorders in which complement activation is implicated in the pathogenesis. AAV,
ANCA-associated vasculitis; FSGS, focal segmental glomerulosclerosis; MN, membranous nephropathy.
in patients with DN.4–7 Glomerular C4d and sC5b-9 fragments in patients with DN who show evidence of
deposits correlate with the severity of DN and micro- complement activation. We found the greatest rate of
vascular and interstitial lesions.22 Furthermore, an progression when both factor Bb and C4a were present,
experimental model of nephrotic syndrome has shown supporting this hypothesis.
that complement-deficient rats developed significantly A possible limitation of our study is that we did not
milder tubulointerstitial injuries than did complement- have measurements of plasma complement biomarkers,
sufficient animals, despite equivalent levels of pro- and it was thus not possible to address if urinary levels
teinuria.23 Complement activation also participates in simply represent overflow of plasma levels. However,
the deleterious effect of hypertension.24,25 Animal recent studies showed that there was no significant
models of angiotensin-2–induced hypertension show correlation between plasma levels of complement
reduced cardiomyocyte hypertrophy, cardiac inflam- components, proteinuria, and GFR in DN,22 as well as
mation, and perivascular fibrosis using a C5a receptor in patients with autoimmune glomerulonephritis.29,31
antagonist.26 Only a few experimental studies have Nevertheless, because we know that plasma levels of
assessed the therapeutic effect of blocking complement sC5b-9 are increased in patients with DN, urinary
signaling in DN. Diabetic mice with increased com- excretion due to change in glomerular permeability
plement activation had a more severe diabetic ne- cannot be excluded. sC5b-9 is a large molecule
phropathy27 and blockade of complement component (approximately 1000 kDa) that is not filtered by the
C5 diminished the severity of albuminuria.28 normal glomerulus29 and healthy individuals have
To our knowledge, this is the first prospective study undetectable urinary levels.30,31 To further address
with repeatedly measured urinary complement bio- this, we compared urinary levels of sC5b-9 of patients
markers in the diabetic population to find a correlation with DN with other immunologic glomerulonephritis,
with the rate of renal function decline. Previous studies in which a clinically meaningful role of complement
have addressed sC5b-9 in the clinical setting: Ogro- activation exists and can be the target of therapy.33 Our
dowski et al.29 reported preliminary data showing that results demonstrated that autoimmune GN excretes
urinary MAC excretion was increased not only in pa- sC5b-9 at levels comparable to DN. This supports that
tients with membranous nephropathy but also in those increased MAC excretion in DN is not solely caused by
with DN, which was correlated with the degree of a passive filtration of plasma sC5b-9 in the setting of
proteinuria. A small study in 15 patients with DN with higher glomerular permeability, but is locally
nephrotic proteinuria also reported urinary excretion expressed and implicated in the pathogenesis of DN, at
of sC5b-9 and CD59 (a MAC inhibitor).30 In this study, least in a subgroup of patients. It is also possible that
however, urinary levels of sC5b-9 correlated with certain individuals in the DN cohort unknowingly had
CD59, but not with proteinuria or cross-sectional GFR. an underlying autoimmune GN. Such misdiagnosis of
Morita et al.31 later found that urinary excretion of DN is inherent to most diabetic studies in which a renal
MAC was elevated in 99 patients (17 with DN) with biopsy is not mandatory.6,7,34,35
nephrotic syndrome, and that excretion was correlated Other limitations should be outlined. We only had 2
with degree of proteinuria. Interestingly, in this study, individuals with type 1 diabetes and 80% of the cohort
individuals with minimal change disease, which does were male, reducing the generalizability of the study.
not lead to chronic kidney disease, had very low levels We must also underline that most individuals in the
of urinary MAC despite a mean proteinuria of 4.89 g/g DN cohort corresponded to the Kidney Disease:
of creatinine. Improving Global Outcomes chronic kidney disease
It was not possible to highlight the activation of one categories G3b/G4-A3 and that the results presented
specific complement pathway in our cohort. MASP-1 here are not necessarily applicable to earlier stages of
and C1q were both correlated with proteinuria and the disease.36 Finally, the small sample could not allow
renal function decline. These 2 proteins could thus us to adjust for other important risk factors of pro-
have led to the formation of C4a fragment, and both the gression in DN (e.g., among others, blood pressure
lectin and classical pathways could be implicated. As- control and antihypertensive drugs, HbA1c, type of
sociations with urinary factor Bb, present in fewer diabetes treatment).
patients, is more surprising, as the alternative com- Future studies are needed to validate that urinary
plement pathway has not been linked with DN in complement fragments are predictive of outcomes in
previous studies. However, deposition of C3b mediated the clinical setting, and not a mere reflection of the
by the classical or lectin pathways can activate the level of proteinuria. Of interest, sC5b-9 should be
alternative pathway and amplify the global comple- tested in earlier stages of DN with lower proteinuria
ment response by a feedback loop.32 It can thus be and preserved GFR to see whether its expression still
plausible to find some level of alternative pathway exists. If such findings are confirmed, we should then
Kidney International Reports (2019) 4, 797–805 803
CLINICAL RESEARCH K Pelletier et al.: Complement Activation in Diabetic Nephropathy
contemplate whether inhibition of the complement 10. Bus P, Chua JS, Klessens CQF, et al. Complement activation
pathways with newly developed agents should be in patients with diabetic nephropathy. Kidney Int Rep. 2018;3:
302–313.
investigated in DN, as done in animal models. Uri-
nary sC5b-9 measurements could then help us 11. Tervaert TW, Mooyaart AL, Amann K, et al. Pathologic clas-
sification of diabetic nephropathy. J Am Soc Nephrol.
determine which patients might benefit from this 2010;21:556–563.
approach.
12. Yiu WH, Li RX, Wong DWL, et al. Complement C5a inhibition
moderates lipid metabolism and reduces tubulointerstitial
DISCLOSURE fibrosis in diabetic nephropathy. Nephrol Dial Transplant.
2018;33:1323–1332.
All the authors declared no competing interests. A small
13. Gnudi L, Coward RJM, Long DA. Diabetic nephropathy:
part of the results pertaining to measurements of perspective on novel molecular mechanisms. Trends Endo-
monocyte chemoattractant protein-1 and transforming crinol Metab. 2016;27:820–830.
growth factor-b1 were published in Diabetes Research and 14. Nosadini R, Velussi M, Brocco E, et al. Course of renal func-
Clinical Practice in 2013.15 tion in type 2 diabetic patients with abnormalities of albumin
excretion rate. Diabetes. 2000;49:476–484.
28. Fujita T, Ohi H, Komatsu K, et al. Complement activation 33. Jayne DRW, Bruchfeld AN, Harper L, et al. Randomized trial
accelerates glomerular injury in diabetic rats. Nephron. of C5a receptor inhibitor avacopan in ANCA-associated
1999;81:208–214. vasculitis. J Am Soc Nephrol. 2017;28:2756–2767.
29. Ogrodowski JL, Hebert LA, Sedmak D, et al. Measurement of 34. Wanner C, Heerspink HJL, Zinman B, et al. Empagliflozin and
SC5b-9 in urine in patients with the nephrotic syndrome. kidney function decline in patients with type 2 diabetes: a
Kidney Int. 1991;40:1141–1147. slope analysis from the EMPA-REG OUTCOME trial. J Am Soc
30. Lehto T, Honkanen E, Teppo AM, et al. Urinary excretion of Nephrol. 2018;29:2755–2769.
protectin (CD59), complement SC5b-9 and cytokines in mem- 35. Saraheimo M, Forsblom C, Hansen TK, et al. Increased
branous glomerulonephritis. Kidney Int. 1995;47:1403–1411. levels of mannan-binding lectin in type 1 diabetic patients
31. Morita Y, Ikeguchi H, Nakamura J, et al. Complement acti- with incipient and overt nephropathy. Diabetologia. 2005;48:
vation products in the urine from proteinuric patients. J Am 198–202.
Soc Nephrol. 2000;11:700–707. 36. Levin A, Stevens PE. Summary of KDIGO. 2012 CKD Guide-
32. Noris M, Remuzzi G. Overview of complement activation and line:behind the scenes, need for guidance, and a framework
regulation. Semin Nephrol. 2013;33:479–492. for moving forward. Kidney Int. 2014;85:49–61.