Lectin-induced renal local complement activation is involved in tubular interstitial injury in diabetic nephropathy

Accepted Manuscript Lectin-induced renal local complement activation is involved in tubular interstitial injury in diabetic nephropathy

Jing-Min Zheng, Xian-Guo Ren, Zuan-Hong Jiang, De-Jun Chen, Wen-Jin Zhao, Li-Juan Li PII: DOI: Reference:

S0009-8981(18)30149-9 doi:10.1016/j.cca.2018.03.033 CCA 15124

To appear in:

Clinica Chimica Acta

Received date: Revised date: Accepted date:

20 October 2017 26 March 2018 26 March 2018

Please cite this article as: Jing-Min Zheng, Xian-Guo Ren, Zuan-Hong Jiang, De-Jun Chen, Wen-Jin Zhao, Li-Juan Li , Lectin-induced renal local complement activation is involved in tubular interstitial injury in diabetic nephropathy. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Cca(2018), doi:10.1016/j.cca.2018.03.033

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Lectin-induced renal local complement activation is involved in tubular interstitial injury in diabetic nephropathy

Jing-Min Zhengab* , Xian-Guo Renc# , Zuan-Hong Jianga, De-Jun Chena, Wen-Jin

a

PT

Zhaob, Li-Juan Lib

Department of Nephrology, Taizhou Hospital, Wenzhou Medical University, Linhai

RI

317000, Zhejiang Province, China b

National Clinical Research Center of Kidney Diseases, Jingling Hospital, Nanjing

Department of Pediatrics, Jingling Hospital, Nanjing University School of Medicine,

Nanjing 210002, Jiangsu Province, China

MA

#Co-first author

NU

c

SC

University School of Medicine, Nanjing 210002, Jiangsu Province, China

*Corresponding author: Professor Jing-Min ZHENG

ED

National Clinical Research Center of Kidney Diseases，Jingling Hospital, Nanjing University School of Medicine，305 Zhongshan Road, Nanjing 21002, Jiangsu

EP T

Province, China Tel.: +86 80 860426

AC C

Email: [email protected]

ACCEPTED MANUSCRIPT Abstract BACKGROUND-Complement has been suggested to be involved in diabetic nephropathy (DN), but the exact significance and underlying mechanisms remain unclear. Data about renal local complement activation in DN patients is scarce. The purpose of the study was to clarify the significance and mechanism of renal

PT

local complement activation in DN. METHODS-Sixty-two biopsy-proven DN patients were recruited. Renal

RI

expression of C1Q, factor B, C5b-9, MBL and MBL-associated serine protease 1 (MASP1) were detected and associated with the kidney damage.

SC

RESULTS-C5b-9, MBL and MASP1 was found to increase with the progression of DN. Especially, the level of C5b-9, MBL and MASP1 in tubular interstitium

NU

was closely associated with the damage degree of tubular interstitium. In addition, MBL and MASP1 co-localized and their levels in tubular interstitium correlated

MA

with the levels of C5b-9 in tubules and tubular interstitium. CONCLUSION-Increased renal local complement activation was present in DN

ED

patients and might contribute to the kidney damage, especially tubular interstitial damage. MBL pathway might play an important role in renal tubular interstitial

EP T

complement activation. Methods against complement activation or MBL pathway might be effective in reducing renal tubular interstitial damage in DN patients. Keywords

AC C

Complement; C5b-9; diabetic nephropathy; MBL

ACCEPTED MANUSCRIPT

1. Introduction As a part of immune system, complement is evolved to protect the hosts from invasion of pathogens, but it can also cause self-damage following inappropriate activation [1-3]. Involvement of complement in immune-associated kidney diseases

PT

has been well demonstrated [4-10]. Interestingly, recent studies suggested that complement might also be involved in diabetic nephropathy (DN), a traditionally

RI

recognized non- immune kidney disease [11-18]. However, the exact clinical significance and the underlying mechanisms are unclear. Of note, previous clinical

SC

observations focused mainly on the changes of circulating complement components

NU

[13-18]. Data about renal local complement activation in DN patients is still scarce.

Complement can be activated through one of the three pathways: the classical, the

MA

alternative and the mannose-binding lectin (MBL) pathway. In patients with diabetes mellitus, increased plasma levels of MBL pathway components have been

ED

reported and thought to be associated with the development of DN [13-18]. However, no study has reported the expression of MBL pathway components in the

EP T

renal tissue of DN patients.

To clarify the clinical significance of renal local complement activation in DN

AC C

patients and determine the mechanism of complement activation, the present study examined the deposition of complement activation products C5b-9 and the expression of C1Q, factor B, MBL and MBL-associated serine protease 1 (MASP1) in the renal biopsy specimens of DN patients. Information derived would contribute to a better understanding of the role of complement and the underlying mechanism in the disease. 2. Materials and methods 2.1. Study population. Sixty-two diabetic patients with biopsy-proven DN, who were hospitalized at the clinical unit of the nephrology centre of Jingling Hospital from 2013 to 2014, were

ACCEPTED MANUSCRIPT recruited. All these patients met the World Health Organization criteria for type 2 diabetes mellitus, and renal biopsy consistent with the diagnosis of DN and exclusion for other concomitant renal diseases. According to their clinical characteristics, patients were classified into microalbuminuria stage group (MG, urinary albumin excretion rate [UAE] >30 mg/24 h and <300 mg/24 h), proteinuria stage group (PG, UAE ≥300 mg/24 h and estimated glomerular filtration rate

PT

[eGFR, according to the equation of Modification of Diet in Renal Disease] ≥60 ml

RI

min−1 1.73 m−2 ) and renal insufficiency stage group (RIG, eGFR <60 ml min−1 1.73 m−2 ). Eleven renal samples (5 men and 6 women with an average age of 55±9)

SC

from the un-affected part of the kidneys which were removed because of renal carcinoma were served as normal controls when analyzing renal complement

NU

components. This study was approved by Ethics Committee of Jingling Hospital (approval number: 2013GJJ-100). All research work with human participants was

MA

in accordance with the ethical standards of the responsible committee on human experimentation and with the Declaration of Helsinki. Informed consent was

ED

obtained from every participant. 2.2. Histology.

formalin,

EP T

For light microscopy, renal biopsy specimens were fixed in 10% neutral buffered embedded

in

paraffin

and

sectioned

at

2

μm

thickness.

Hematoxylin–eosin, Periodic acid–Schiff’s reagent (PAS), Masson’s trichrome and

AC C

Periodic acid Methenamine silver staining were performed. Glomerular, tubular interstitial and vascular lesions in biopsies were recorded, classified and scored according to criteria described by Tervaert et al [19]. Neutrophils, eosinophils, plasma and mononuclear cells were identified according to the morphological characteristics of these cells on hematoxylin and eosin-stained sections, and the number of these cells in the tubular interstitium was counted by two pathologists. For each section, at least ten randomly selected cortical regions were examined, and the area of the selected regions in each section was measured using NIS Element BR3.4 software (Nikon, Shinagaw-ku Tokyo, Japan). The number of each kind of inflammatory cells was presented as the number of the cells

ACCEPTED MANUSCRIPT per area. The relative interstitial volume (RIV) was measured according to methods described by Okon et al [20]. For each section, ten randomly selected cortical regions were measured and the average value was taken as the RIV of the section. 2.3. Immunohistochemistry and quantitative analysis.

PT

As the antibody for C5b-9，MBL and MASP1 was not work well in paraffin section, immunohistochemical staining was performed on 4-m thick frozen sections using

RI

routine methods. Briefly, the unfixed renal tissue was embedded in OCT compound (Sakura Tissue-Tek; Bayer, Leverkusen, Germany), snap-frozen in liquid nitrogen,

SC

and stored at -80°C. Subsequently, 4-m sections were prepared and stored at -20°C until use. Endogenous peroxidase was blocked with hydrogen peroxide. After

NU

blocked for 30 min in 10% new-born calf serum, the samples were rinsed in PBS and incubated for 2 hours at room temperature with different first antibodies, such

MA

as mouse anti- human C5b-9 monoclonal antibody (Catalogue number ab66768, Abcam, Cambridge, UK, 1:500), mouse anti-human C1q antibody (Catalogue

ED

number ab71089, Cambridge, UK, 1:100), rabbit anti- human complement factor B antibody (Catalogue number NBP1-89985, Novus, Colorado, USA, 1:100), rabbit

EP T

anti-human MASP1/MASP3 polyclonal antibody (Catalogue number NBP1-85460, Novus, Colorado, USA, 1:500) and rabbit anti-human MBL polyclonal antibody (Catalogue number NBP1-85518, Novus, Colorado, USA, 1:25). After washing

AC C

with PBS, the tissues were incubated with HRP polymer (The second antibody, Quanhui Company, Zhuhai, Guangdong, China) for 45 minutes, then, the sections were washed and

developed

with diaminobenzidine.

Each section was

counterstained with haematoxylin. Normal homologous serum was used to replace the first antibody as a negative control. Image-pro Plus 6.0 software (Media Cybernetics, Bethesda, MD, USA) was used to measure the level of C5b-9, C1q, factor B, MBL, MASP1/MASP3 quantitatively in the sections. For evaluating the expression level of the molecules in tubules and tubular interstitium, at least 10 high power fields in the cortex region of each section were measured. For evaluating the glomerular expression level of these molecules,

ACCEPTED MANUSCRIPT at least 5 glomerulus from each section were measured (sections with less than 5 glomerulus were not included in the assay). To determine whether C1q, MBL and MASP1/3 was increasingly expressed in a glomerulus, here, we defined the situation of increased expression of C1q, MBL and MASP1/3 in a glomerulus as the expression level higher than the average + 1.96 S (standard deviation) of that in the

PT

normal control group. 2.4. Immunofluorescence staining.

RI

Immunofluorescence staining for MASP1 light chain was performed on frozen sections using routine methods. Briefly, the sections were blocked for 30 min in

SC

10% new-born calf serum, rinsed in PBS and incubated with goat anti-human MASP1 light chain antibody (Catalogue number SC-50843, Santa Cruz

NU

Biotechnology, CA, USA, 1:100) for 2 hours. After washed with PBS, the tissues were incubated with Cy3- labeled donkey anti- goat IgG antibody (Catalogue number

washed,

mounted

MA

A21206, EarthOx, CA, USA, 1:300) for 30 minutes. Then, the sections were and observed

under

fluorescence

microscope. Normal

ED

homologous serum was used to replace the first antibody as a negative control. The immuno-staining intensity for MASP1 light chain in the tubular interstitium was

EP T

evaluated by two blinded observers and scored as follows: 0=negative staining, 1 = minimal staining, 2 = moderate staining, 3 = strong staining. The average value of the scores made by the two observers was used to represent the leve l of MASP1 in

AC C

each sample.

2.5. Double immunofluorescence staining for MBL and MASP1. Double immunofluorescence staining for MBL and MASP1 was performed on frozen sections using routine methods. Briefly, after blocked in 10% new-born calf serum for 30 min, washed in PBS and incubated with goat anti- human MASP1 light chain antibody (1:100) for 2 hours, the sections were incubated with the Cy3 labeled donkey anti- goat IgG antibody (1:100) for 30 minutes. Then, the sections were washed and incubated with the rabbit anti- human MBL antibody (1:25) for another 2 hours. Then, the sections were washed and incubated with alexa fluor488- labeled donkey anti-rabbit IgG antibody (Invitrogen, CA, USA, 1:1000)

ACCEPTED MANUSCRIPT for 30 minutes. After washing, the sections were mounted and observed under laser scanning confocal microscope. 2.6. Measurement of plasma MBL, plasma C5b-9, urine N-acetyl-β- glucosaminidase (NAG), urine retinol-binding protein (RBP) and urine neutrophil gelatinase associated lipocalin (NGAL) level.

PT

An ELISA kit from BD Biosciences (Catalogue number 558315, NY, USA) was used to measure plasma C5b-9 level. An ELISA kit from Elabscience (Catalogue number

RI

E-EL-H1305c, Wuhan, Hubei, China) was used to measure plasma MBL level. Kits from Biostec (Catalogue number CR2337, C109200T and C110200T respectively,

SC

Chongqing, China) were used to measure urinary creatinine, NAG and RBP level. The urinary NGAL level was analyzed using “Neutrophil gelatinase-associated lipocalin

NU

(NGAL) Assay kit” from Vazyme (Nanjing, Jiangsu, China). The urine NAG, RBP and NGAL level were corrected by the level of urinary creatinine and expressed as

MA

nanogram per milligram creatinine for NGAL, microgram per milligram creatinine for RBP, and Unit per gram creatinine for NAG to minimize the variations from urine

2.7. Statistical Analyses.

ED

collection.

EP T

All data were analyzed using SPSS version 19.0 (SPSS, Chicago, IL, USA). Data were presented as mean ± SD for continuous variables with normal distribution and One way ANAVA was used to analyzed these data. Data were presented as median

AC C

(interquartile range) for continuous variables without normal distributio n and absolute

value

and

percentage

for

frequency

of categorical

variables.

Mann–Whitney test was used for continuous variables without normal distribution. Correlation analysis was carried out using Spearman coefficients. All statistical tests were two-tailed, and P values <0.05 were considered significant. 3. Results 3.1. Clinical and pathological parameters of DN patients. Among the 62 DN patients, 11 were allocated to the MG, 17 to the PG and 34 to the RIG. There were 7 men in the MG, 11 men in the PG and 23 men in the RIG. The mean ages of the patients in the three groups were 52.8±13.1, 47.8±10.5 and

ACCEPTED MANUSCRIPT 49.8±9.8, respectively. The median eGFR was 107.0 (91.4-120.6), 85.3 (67.0-109.4) and 37.9 (23.5-49.9) mlminute-1 1.73 m-2 in the MG, PG and RIG, respectively. The number of patients with hypertension was 8, 14 and 33 in the MG, PG and RIG, respectively. Glomerular, tubulointerstitial and renal blood vessel injury increased

PT

from the MG to the PG and RIG (Table 1).

3.2. Immuno-staining for C5b-9.

RI

Immuno-staining for C5b-9 could be observed in all the normal renal specimens (Fig.1A). They were distributed in Bowman's capsular membrane, glomerular

SC

mesangial area, tubules (mainly in tubular basement membrane, TBM), tubular interstitium and the wall of blood vessels. With the progression of DN, the levels of

NU

C5b-9 in the glomerulus, tubules and tubular interstitium increased. The levels of C5b-9 in glomerulus seemed to be associated with the degrees of mesangial

MA

expansion and culminated in the sclerotic glomerulus. The levels of C5b-9 in tubules seemed to be associated with the damage degrees of the tubules. Higher

ED

immuno-staining for C5b-9 was found in the atrophy tubules. Also, deposition of C5b-9 increased in the wall of blood vessels in DN patients.

EP T

Results of bivariate correlation analysis showed that the level of C5b-9 in tubular and interstitial area correlated significantly with indices of tubular interstitial

AC C

damage (Table 2).

3.3. Plasma C5b-9 level in DN patients. To determine whether the increased renal deposition of C5b-9 in DN patients originated from systemic complement activation, we measured the plasma C5b-9 at the same time. As shown in Fig.2, no significant change in the level of plasma C5b-9 in DN patients was found. Also, the level of plasma C5b-9 didn’t correlate with the renal C5b-9 level. 3.4. Immuno-staining for C1q, factor B, MBL and MASP1 heavy chain. As shown in Fig.3, weak immuno-staining for C1q and factor B could be observed

ACCEPTED MANUSCRIPT in the renal tissue of normal controls. The immuno-staining for C1q was found mainly in the renal tubular interstitium and in the glomerular mesangial area in the normal controls. No obvious change in the expression of C1q was observed except that in the sclerosis area of some glomerulus, where higher depositio n of C1q was observed (Fig.3A). Immuno-staining for factor B was found mainly in tubular

PT

interstitium and on the brush border side of the tubules in the normal controls. No significant change in factor B level in the tubular interstitium was observed in the

RI

DN patients when compared with the normal controls. Higher immuno-staining for factor B was observed in a fewer tubules of DN patients, which distributed mainly

SC

on the brush border side (Fig.3B g-h). This phenomenon was observed in 29 of the

NU

62 patients.

As shown in Fig. 4, only trace immuno-staining for MBL and MASP1 heavy chain,

MA

which is a common component of MASP1 and MASP3 (MASP1/3), was observed in the normal renal specimens (Fig.4A(b) and Fig.4B(h) respectively). To varied

ED

degrees, immuno-staining for MBL and MASP1/3 increased in all the renal tissues of DN patients. The increased MBL and MASP1/3 in DN patients were found

EP T

around the tubules, appearing to be expressed by tubular interstitial cells (Fig.4A(c) and Fig.4B(i), respectively). The levels of MBL and MASP1/3 in tubular interstitium increased with the progression of DN (The tubular interstitial MBL and

AC C

MASP1/3 level increased from MG to PG and RIG, Fig.4(D)) and seemed to be closely associated with the severity of tubular interstitial damage (F ig.4A(d)-(f) and Fig.4B(j)-(l), respectively). Analysis based on serial renal sections showed that MBL and MASP1/3 co-localized in the renal tissues of DN patients (Fig.4(C)). In addition, increased immuno-staining for MBL and MASP1/3 was also observed in the mesangial area of some injured glomerulus. Such glomerulus was observed in 28 of the 52 DN patients examined (10 patients with less than 5 glomerulus were not included in the assay). 3.5. Immuno-staining for MASP1 light chain.

ACCEPTED MANUSCRIPT As antibody against MASP1 heavy chain detected both MASP1 and MASP3, another antibody against MASP1 light chain was further used to detect MASP1 only

by

immunofluorescence

(the

antibody

can

only

be

used

in

immunofluorescence but not in immunohistochemistry). The results of MASP1 light chain staining was similar to that of MASP1 heavy chain staining:

PT

immuno-staining for MASP1 light chain was found mainly in tubular interstitium; the immuno-staining intensity of MASP1 in tubular interstitium increased with the

SC

RI

increase of tubular interstitial damage (Fig.5).

3.6. Double immuno-fluorescence staining for MBL and MASP1.

NU

Results of double immuno- fluorescence staining for MBL and MASP1 showed that

renal tubular interstitium (Fig.6).

MA

MBL and MASP1 co- localized in the renal tissue of DN patients, especially in the

3.7. Plasma MBL level in DN patients.

ED

As shown in Fig.7, the level of plasma MBL increased in DN patients compared with the normal control group. However, no significant correlation between the

EP T

plasma MBL level and renal MBL deposition was observed (p>0.05). 3.7. Correlation of the levels of renal MBL, MASP1, C1q and factor B with the levels

AC C

of C5b-9 and renal tubular and interstitial injury. As shown in Table 3, the expression levels of MBL and MASP1/3 in renal tubular interstitium were found to correlate significantly with the level of C5b-9 in tubular interstitial area and the indices reflecting tubular injury, including tubular and interstitial injury score, urine RBP, urine NAG and urine NGAL level. No significant correlation of the levels of MBL and MASP1/3 in glomerulus with the level of C5b-9 was observed. Also, no significant correlation of the renal levels of C1q and factor B with renal C5b-9 and indices of renal injury was found. 4. Discussion

ACCEPTED MANUSCRIPT For the first time, the present study evaluated renal deposition of complement activation product C5b-9 in type 2 DN patients. The level of C5b-9 increased with the development of DN and associated closely with kidney damage. In the mean while, no significant change in the level of systemic complement activation product C5b-9 was found. These results indicated that the increased renal C5b-9 was

PT

produced locally and might contribute to the progression of the disease. Previously, Falk et al examined C5b-9 deposition in 12 type 1 DN patients and observed a close Taken together,

RI

association between C5b-9 deposition and kidney damage [21].

these findings suggested that the renal local activation of complement system may

SC

play an important role in both type1 and type 2 DN. Thus complement s ystem may serve as a novel treatment target for reducing renal damage and retarding the

NU

progression of the disease.

To determine the mechanism underlying renal local complement activation, the

MA

present study further examined the expression of C1q (a key molecule of the classical pathway), complement factor B (a key component of the alternative

ED

pathway), MBL and MASP1 (two key molecules of the MBL pathway) in the renal specimens of DN patients. Both MBL and MASP1 were found to be increasingly

EP T

expressed in the renal tubular interstitium in DN patients. The levels of MBL and MASP1 in tubular interstitium were associated closely with the degree of tubular interstitial damage. Also, they were found to correlate with the level of complement

AC C

activation product C5b-9 in tubular interstitial area. Based on these observations, we speculated that MBL pathway- induced renal local complement activation might be involved in renal tubular interstitial damage in type 2 DN patients. In contrast to the expression of MBL and MASP1, no obvious change in the expression of C1q was observed except that in the sclerosis area of some glomerulus. As we could not determine that the deposition of C1q in the sclerosis area of the glomerulus is before or after the damage of the glomerulus here (Because the normal glomerular structure in this area is damaged severely, C1q from plasma is easy to enter into the area), the exact significance of C1q, which represents the classical pathway, in the glomerular damage needs further

ACCEPTED MANUSCRIPT investigation. However, data from our study does not support the role of the classical pathway in the renal tubular interstitial injury since no significant change in C1q level was observed in tubular interstitial area compared with the normal control group. In addition, in the present study, we observed increased complement factor B in

PT

some of the renal tubules, a phenomenon present in 29 of the 62 DN patients. The increased complement factor B was found to be distributed mainly on the brush

RI

border side of the tubules but not on the side adjacent to TBM (C5b-9 was observed distributed mainly in TBM and interstitium in the renal tubular interstitial area).

SC

This phenomenon might suggest that the alternative pathway is involved in urinary complement activation, which has been well reported in previous studies [22-23].

NU

However, as no significant association between factor B level and indices of tubular damage was observed, the exact significance of the increased factor B in renal

MA

tubular damage needs to be further investigated.

Apart from the vital function of host defense to infection, MBL pathway has been

ED

reported to be involved in a lot of human diseases [1,24-29]. But as a defense system against the invasion of pathogens, can MBL pathway be activated in renal

EP T

tubular interstitium of DN patients? The MBL pathway is usually triggered by the binding of its pattern recognition molecules, such as MBL, to their carbohydrate ligands which are usually on the surfaces of various pathogens. Once binding to its

AC C

ligands, MBL forms complexes with MBL-associated serine proteases (MASPs, including MASP1, MASP2 and MASP3). And then, MASPs are activated and induce the formation of the C3 convertase (C4b2a) by directly cleaving C4 and C2 [30-31], which finally initiates the cascade reaction of complement activation. However, evidences from recent studies demonstrated that MBL can also recognize apoptotic cells, cell debris and glycosylated molecules [30, 32]. As one of the most important mechanisms underlying diabetic kidney damage, increased glycosylation level has been proved in DN patients [33-36]. Also, the injury of tubules and interstitium, which may result in cell debris in tubular interstitium, is a common feature in patients with DN [37-41]. The increased glycosylated molecules and cell

ACCEPTED MANUSCRIPT debris in tubular interstitium might be recognized by MBL, which is necessary for the triggering of MBL pathway. In addition, in the present study, the increased MBL and MASP1 were found to co-localize in tubular interstitium, which means that after recognition of its targets, MBL can easily form complex with MASP1 and initiate complement activation cascade. So we speculated that MBL can recognize

PT

the possible ligands, like glycosylated molecules, apoptotic cells and cell debris seen in the DN, and then activate the complement system in tubular interstitium. In

RI

support of this idea, Falk et al reported that in the TBM of type 1 DN patients C5b-9 was distributed on circular membranous structures, which were thought to be

SC

fragmented cellular processes and cellular debris [21].

NU

In the present study, increased expression of MBL and MASP1 were also found in some glomerulus of about half of the DN patients examined. However, as we did

MA

not find significant correlation between the levels of MBL and MASP1 and the deposition of complement activation product C5b-9 in the glomerulus, further

ED

investigations are still needed to clarify the exact significance of MBL pathway in

EP T

glomerular damage of DN patients.

It should be pointed out that there are several limitations in the present study. First, it is a cross-section observation; further perspective study is still needed to confirm

AC C

our conclusions. Second, our observations were based on the specimens of renal biopsy. Deviation from renal biopsy might be unavoidable. Third, as only 62 patients were included, deviation from the relative small sample might also be present. Fourth, as the patients were transferred to our center (The clinical unit of the nephrology centre of Jingling Hospital) from all parts of the country and had been treated with various medicines (including some traditional Chinese medicines) at different dosage for different time before renal biopsy was carried out, which made it difficult to precisely group the patients and compare the difference between different therapies, the present study did not include the influence of different therapies in the complement activation.

ACCEPTED MANUSCRIPT 5. Conclusion In summary, our study indicates that increased renal local complement activation is present in patients with type 2 DN. Renal local complement activation might have important roles in the pathogenesis of the disease. Among the complement activation pathways, the role of MBL pathway in the renal local complement

PT

activation and kidney damage, especially in the renal tubular interstitial damage of DN patients, deserves more attention. Methods against complement activation

RI

or MBL pathway might be effective in retarding the progression of the disease. These are valuable for the better recognition of the disease and the development

AC C

EP T

ED

MA

NU

SC

of novel therapeutic methods.

ACCEPTED MANUSCRIPT Author contributions Jing-Min Zheng conceived and designed the study, participated in the performance of the experiments and the revision of the manuscript, and supervised the project. Xian-Guo Ren participated in designing of the study, performance of the experiments and drafting the manuscript. Zuan-Hong Jiang

PT

participated in interpretation of data, drafting and revision of the manuscript. De-Jun Chen participated in the conception of the study, interpretation of data

RI

and revising the article. Wen-Jin Zhao and Li-Juan Li participated in the performance of the experiments, the analysis and interpretation of data and

SC

revising the article. All authors read and approved the final version of the

NU

manuscript.

We are grateful to Dr. Ming-Cao Zhang, who is affiliated to the National Clinical

MA

Research Center of Kidney Diseases, Jingling Hospital, for his excellent technical assistance in section preparation. We acknowledge Dr. Ming-Lin Zhou, who is

ED

affiliated to the National Clinical Research Center of Kidney Diseases, Jingling

AC C

EP T

Hospital, for her help in data analysis.

ACCEPTED MANUSCRIPT Conflicts of Interest

AC C

EP T

ED

MA

NU

SC

RI

PT

The authors declare that there is no conflict of interest.

ACCEPTED MANUSCRIPT Acknowledgments This work was supported by a grant from the National Natural Science

AC C

EP T

ED

MA

NU

SC

RI

PT

Foundation of China (No. 81370828).

ACCEPTED MANUSCRIPT References [1]

D. Ricklin, E.S. Reis, J.D. Lambris, Complement in disease: a defence system turning offensive, Nat Rev Nephrol. 12 (2016) 383-401.

[2]

M. van Lookeren Campagne, E.C. Strauss, B.L. Yaspan, Age-related macular degeneration: Complement in action, Immunobiology. 221 (2016)

[3]

PT

733-9.

L.G. Mellbin, M. Bjerre, S. Thiel, T.K. Hansen, Complement activation and

RI

prognosis in patients with type 2 diabetes and myocardial infarction: a report

R.J. Popat, M.G. Robson, Complement and glomerular diseases, Nephron Clin Pract. 128 (2014) 238-42.

[5]

NU

[4]

SC

from the DIGAMI 2 trial, Diabetes Care. 35 (2012) 911-7.

N. Maillard, R.J. Wyatt, B.A. Julian, K. Kiryluk, A. Gharavi, V.

MA

Fremeaux-Bacchi, J. Novak, Current Understanding of the Role of Complement in IgA Nephropathy, J Am Soc Nephrol. 26 (2015) 1503-12. A. Mella, M. Messina, A. Lavacca, L. Biancone, Complement cascade and

ED

[6]

kidney transplantation: The rediscovery of an ancient enemy, World J

[7]

EP T

Transplant. 4 (2014) 168-75. D.J. Birmingham, L.A. Hebert, The Complement System in Lupus Nephritis, Semin Nephrol. 35 (2015) 444-54. R.A. Sinico, N. Mezzina, B. Trezzi, G.M. Ghiggeri, A. Radice, Immunology

AC C

[8]

of membranous nephropathy: from animal models to humans, Clin Exp Immunol. 183 (2016) 157-65.

[9]

S.J. Gou, J. Yuan, C. Wang, M.H. Zhao, M. Chen, Alternative complement pathway activation products in urine and kidneys of patients with ANCA-associated GN, Clin J Am Soc Nephrol. 8 (2013) 1884-91.

[10]

R. Ma, Z. Cui, Y.H. Liao, M.H. Zhao, Complement activation contributes to the injury and outcome of kidney in human anti-glomerular basement membrane disease, J Clin Immunol. 33 (2013) 172-8.

ACCEPTED MANUSCRIPT [11]

J.M. Forbes, M.E. Cooper, Mechanisms of diabetic complications, Physiol Rev. 93 (2013) 137-88.

[12]

J. Ostergaard, T.K. Hansen, S. Thiel, A. Flyvbjerg, Complement activation and diabetic vascular complications, Clin Chim Acta. 361 (2005) 10-9.

[13]

M. Saraheimo, C. Forsblom, T.K. Hansen, A.M. Teppo, J. Fagerudd, K.

PT

Pettersson-Fernholm, S. Thiel, L. Tarnow, P. Ebeling, A. Flyvbjerg, P.H. Groop, G. FinnDiane Study, Increased levels of mannan-binding lectin in

RI

type 1 diabetic patients with incipient and overt nephropathy, Diabetologia. 48 (2005) 198-202.

T.K. Hansen, M.A. Gall, L. Tarnow, S. Thiel, C.D. Stehouwer, C.G.

SC

[14]

Schalkwijk, H.H. Parving, A. Flyvbjerg, Mannose-binding lectin and

[15]

NU

mortality in type 2 diabetes, Arch Intern Med. 166 (2006) 2007-13. T.K. Hansen, C. Forsblom, M. Saraheimo, L. Thorn, J. Waden, P. Hoyem, J.

MA

Ostergaard, A. Flyvbjerg, P.H. Groop, G. FinnDiane Study, Association between mannose-binding lectin, high-sensitivity C-reactive protein and the

(2010) 1517-24.

L.Z. Guan, Q. Tong, J. Xu, Elevated serum levels of mannose-binding lectin

EP T

[16]

ED

progression of diabetic nephropathy in type 1 diabetes, Diabetologia. 53

and diabetic nephropathy in type 2 diabetes, PLoS One. 10 (2015) e0119699. [17]

L. Jenny, R. Ajjan, R. King, S. Thiel, V. Schroeder, Plasma levels of

AC C

mannan-binding lectin-associated serine proteases MASP-1 and MASP-2 are elevated in type 1 diabetes and correlate with glycaemic control, Clin Exp Immunol. 180 (2015) 227-32.

[18]

J.A. Ostergaard, S. Thiel, P. Hovind, C.B. Holt, H.H. Parving, A. Flyvbjerg, P. Rossing, T.K. Hansen, Association of the pattern recognition molecule H-ficolin with incident microalbuminuria in an inception cohort of newly diagnosed type 1 diabetic patients: an 18 year follow-up study, Diabetologia. 57 (2014) 2201-7.

[19]

T.W. Tervaert, A.L. Mooyaart, K. Amann, A.H. Cohen, H.T. Cook, C.B. Drachenberg, F. Ferrario, A.B. Fogo, M. Haas, E. de Heer, K. Joh, L.H. Noel,

ACCEPTED MANUSCRIPT J. Radhakrishnan, S.V. Seshan, I.M. Bajema, J.A. Bruijn, S. Renal Pathology, Pathologic classification of diabetic nephropathy, J Am Soc Nephrol. 21 (2010) 556-63. [20]

K. Okon, A. Szumera, M. Kuzniewski, Are CD34+ cells found in renal interstitial fibrosis?, Am J Nephrol. 23 (2003) 409-14. R.J. Falk, S.P. Sisson, A.P. Dalmasso, Y. Kim, A.F. Michael, R.L. Vernier,

PT

[21]

Ultrastructural localization of the membrane attack complex of complement

[22]

RI

in human renal tissues, Am J Kidney Dis. 9 (1987) 121-8.

Y. Morita, H. Ikeguchi, J. Nakamura, N. Hotta, Y .Yuzawa, S. Matsuo,

SC

Complement activation products in the urine from proteinuric patients, J Am Soc Nephrol. 11(2000): 700-7.

NU

[23] J.L. Ogrodowski, L.A. Hebert, D. Sedmak, F.G. Cosio, J. Tamerius, W. Kolb, Measurement of SC5b-9 in urine in patients with the nephrotic syndrome,

[24]

MA

Kidney Int. 40(1991):1141-7.

A. Fearn, N.S. Sheerin, Complement activation in progressive renal disease,

[25]

ED

World J Nephrol. 4 (2015) 31-40.

L. Bao, P.N. Cunningham, R.J. Quigg, Complement in Lupus Nephritis:

[26]

EP T

New Perspectives, Kidney Dis (Basel). 1 (2015) 91-9. B.E. Berger, The Alternative Pathway of Complement and the Evolving Clinical-Pathophysiological Spectrum of Atypical Hemolytic Uremic

[27]

AC C

Syndrome, Am J Med Sci. 352 (2016) 177-90. C.M. Nester, R.J. Smith, Complement inhibition in C3 glomerulopathy, Semin Immunol. 28 (2016) 241-9.

[28]

D. Ricklin, J.D. Lambris, New milestones ahead in complement-targeted therapy, Semin Immunol. 28 (2016) 208-22.

[29]

G. Bajic, S.E. Degn, S. Thiel, G.R. Andersen, Complement activation, regulation, and molecular basis for complement-related diseases, EMBO J. 34 (2015) 2735-57.

ACCEPTED MANUSCRIPT [30]

M.H. Beltrame, A.B. Boldt, S.J. Catarino, H.C. Mendes, S.E. Boschmann, I. Goeldner, I. Messias-Reason, MBL-associated serine proteases (MASPs) and infectious diseases, Mol Immunol. 67 (2015) 85-100.

[31]

H. Sekine, M. Takahashi, D. Iwaki, T. Fujita, The role of MASP-1/3 in complement activation, Adv Exp Med Biol. 735 (2013) 41-53. J. Fortpied, D. Vertommen, E. Van Schaftingen, Binding of

PT

[32]

mannose-binding lectin to fructosamines: a potential link between

RI

hyperglycaemia and complement activation in diabetes, Diabetes Metab Res Rev. 26 (2010) 254-60.

S.B. Peterson, G.W. Hart, New insights: A role for O-GlcNAcylation in

SC

[33]

diabetic complications, Crit Rev Biochem Mol Biol. 51 (2016) 150-61. Y.M. Sun, Y. Su, J. Li, L.F. Wang, Recent advances in understanding the

NU

[34]

biochemical and molecular mechanism of diabetic nephropathy, Biochem

[35]

MA

Biophys Res Commun. 433 (2013) 359-61.

C.M. Zheng, W.Y. Ma, C.C. Wu, K.C. Lu, Glycated albumin in diabetic

ED

patients with chronic kidney disease, Clin Chim Acta. 413 (2012) 1555-61. [36] F.P. Schena, L. Gesualdo, Pathogenetic mechanisms of diabetic nephropathy,

[37]

EP T

J Am Soc Nephrol. 16 Suppl 1 (2005) S30-3. C.J. Magri, S. Fava, The role of tubular injury in diabetic nephropathy, Eur J Intern Med. 20 (2009) 551-5.

AC C

[38] P. Bjornstad, M.A. Lanaspa, T. Ishimoto, T. Kosugi, S. Kume, D. Jalal, D.M. Maahs, J.K. Snell-Bergeon, R.J. Johnson, T. Nakagawa, Fructose and uric acid in diabetic nephropathy, Diabetologia. 58 (2015) 1993-2002.

[39]

G. Tramonti, Y.S. Kanwar, Review and discussion of tubular biomarkers in the diagnosis and management of diabetic nephropathy, Endocrine. 43 (2013) 494-503.

[40]

M.A. Lanaspa, T. Ishimoto, C. Cicerchi, Y. Tamura, C.A. Roncal-Jimenez, W. Chen, K. Tanabe, A. Andres-Hernando, D.J. Orlicky, E. Finol, S. Inaba, N. Li, C.J. Rivard, T. Kosugi, L.G. Sanchez-Lozada, J.M. Petrash, Y.Y. Sautin, A.A. Ejaz, W. Kitagawa, G.E. Garcia, D.T. Bonthron, A. Asipu, C.P.

ACCEPTED MANUSCRIPT Diggle, B. Rodriguez-Iturbe, T. Nakagawa, R.J. Johnson, Endogenous fructose production and fructokinase activation mediate renal injury in diabetic nephropathy, J Am Soc Nephrol. 25 (2014) 2526-38. M. Zhan, I.M. Usman, L. Sun, Y.S. Kanwar, Disruption of renal tubular mitochondrial quality control by Myo-inositol oxygenase in diabetic kidney

EP T

ED

MA

NU

SC

RI

PT

disease, J Am Soc Nephrol. 26 (2015) 1304-21.

AC C

[41]

ACCEPTED MANUSCRIPT Figure captions

Fig.1. Renal deposition of C5b-9 increased with the development of diabetic

PT

nephropathy (DN). (A): Representative pictures showing deposition of C5b-9 in the renal tissues of normal controls and DN patients ((a): Negative control of

patients).

RI

immunohistochemistry, (b): A picture from normal control; (c): A picture form DN (B): Results of semi-quantitative analysis, *p<0.05, **p<0.01. Solid black

SC

line in (d) and (e): median. Scale bar in (a-c): 50 m.

NU

Fig.2.

MA

No significant change in the level of plas ma C5b-9 in DN patients at different clinical stages of the disease was found when compare d with normal controls . NCG: Normal control group; MG: Microalbuminuria stage group; PG: Proteinuria

ED

stage group; RIG: renal insufficiency stage group. Sold black line: median.

EP T

Fig.3

Results of immuno-staining for C1q and comple ment factor B. (A): Results of

AC C

immuno-staining for C1q ((a): Negative control of immunohistochemistry; (b): A representative picture from normal control; (c): A representative picture from DN patients; (d): A representative picture showing increased C1q in the sclerosis area of the glomerulus of DN patients). (B): Results of immuno-staining for complement factor B ((e): Negative control of immunohistochemistry; (f): A representative picture from normal control; (g): A representative picture from DN patients; (h): Partial magnification of picture (g) showing increased complement factor B distributed on the brush border of the tubules (shown by black arrow)). Scale bar in (a)-(h): 50 m. Fig.4.

ACCEPTED MANUSCRIPT Renal immuno-staining for MBL and MASP1/3 increased with the development of diabetic nephropathy (DN). (A): Immuno-staining for MBL ((a): Negative control of immunohistochemistry; (b): A representative picture from normal control; (c): A representative picture from DN patients; (d)-(f): Representative pictures showing that the level of MBL in renal tubular interstitium increased with the increase of tubular

PT

interstitial damage in DN patients). (B): Immuno-staining for MASP1/3 ((g): Negative control of immunohistochemistry; (h): A representative picture from normal control;

RI

(i): A representative picture from DN patients; (j)-(l): Representative pictures showing that the level of MASP1/3 in renal tubular interstitium increased when tubular

SC

interstitial damage increased in DN patients). (C): Results of immuno-staining for MBL and MASP1/3 respectively on serial renal sections, showing that MBL (m) and

NU

MASP1/3 (n) co- localized in the renal tissues of DN patients. (D): Results of semi-quantitative analysis, * p<0.05; **p<0.01. Solid black line in D: median. Scale

MA

bar in A-C: 50 m.

ED

Fig.5.

Renal immunofluorescence-staining for MASP1 light chain increased with the of

diabetic

EP T

development

nephropathy

(DN).

(a): Negative

control of

immunofluorescence; (b): Immuno-staining for MASP1 in the renal tissues of normal

AC C

controls. (c)-(e): Immuno-staining for MASP1 in the renal tissues of DN patients showing that the level of MASP1 in the renal tubular interstitium increased with the increase of tubular interstitial damage. (f): Results of quantitative analysis, * p<0.05; **p<0.01. Scale bar in (a)-(e): 50 m. Fig.6. Results of double immuno-fluorescence staining for MBL and MASP1. MASP1 (a) and MBL (b) were found to be co- localized in the renal tissue of diabetic nephropathy patients ((c): Merged picture of MASP1 and MBL). (d)-(f): Negative control of immunofluorescence ((d): MASP1; (e):MBL; (f): Merged picture of (d) and

ACCEPTED MANUSCRIPT (e)) . Scale bar in (a)-(f): 50 m. Fig.7. Plas ma MBL level in patients with diabetic nephropathy (DN). NCG: Normal

AC C

EP T

ED

MA

NU

SC

RI

PT

control group; DNG: DN patient group. Sold black line: median; * p<0.05.

SC

RI

PT

ACCEPTED MANUSCRIPT

AC C

EP T

ED

MA

NU

Figure.1

PT

ACCEPTED MANUSCRIPT

AC C

EP T

ED

MA

NU

SC

RI

Figure.2

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC C

EP T

ED

Figure.3

AC C

EP T

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Figure.4

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC C

EP T

ED

MA

Figure.5

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC C

EP T

ED

MA

Figure.6

PT

ACCEPTED MANUSCRIPT

AC C

EP T

ED

MA

NU

SC

RI

Figure.7

ACCEPTED MANUSCRIPT Table1 Clinical and pathological variables of participating patients. MG

PG

RIG

11

17

34

52.8±13.1

47.8±10.5

49.8±9.8

84 (36-120)

108 (24-162)

120 (84-144)

7/4

11/6

23/11

26.5±3.3

N Age (year) Duration of diabetes mellitus

PT

Parameters

(months)

RI

25.8±2.1

7.1 (6.0-8.6)

7.2 (5.8-9.7)

7.0 (5.8-8.6)

11.0 (9.2-14.5)

11.1

11.9

(9.0-13.0)

(10.5-13.6)

45.0±2.1

39.0 ±7.5**

34.1±5.5**##

MA

Male/female

135.0±16.8

127.8±18.7

113.0±17.5**##

7.1(6.4-8.4)

8.5(6.5-9.1)

7.5(6.4-8.5)

54(46-68)

69(48-76)

57(46-69)

Serum cholesterol (mmol/L)

4.3 (4.1-5.4)

5.2 (3.8-5.9)

6.2 (4.8-7.7) **#

Serum low density cholesterol

2.7(1.9-3.6)

2.5(1.7-3.3)

3.2(2.4-4.3)

0.7(0.6-1.0)

0.8(0.7-0.9)

0.9(0.7-1.2)

1.8 (1.0-2.4)

1.8 (1.1-3.8)

1.8 (1.2-2.5)

70.7 (60.1-91.1)

80.4

149.0

(68.5-107.8)

(131.3-253.3)

Fasting blood glucose (mmol/L) blood

glucose

NU

Postprandial (mmol/L)

Serum albumin (g/L) Hemoglobin (g/L)

AC C

(mmol/L)

EP T

(mmol/mol)

ED

Glycated hemoglobin (%) Glycated hemoglobin

26.4±3.3

SC

Body mass index

Serum high density cholesterol (mmol/L)

Serum triglyceride (mmol/L) Serum creatinine (µmol/L)

**##

eGFR (ml/min/1.73m-2 )

Proteinuria (g/24h)

107.0

85.3

37.9

(91.4-120.6)

(67.0-109.4)

(23.5-49.9)

0.38 (0.35-0.41) 1.8 (0.7-4.3)** 5.2 (2.9-7.4)**##

ACCEPTED MANUSCRIPT systolic blood pressure (mm

132 (130-137)

140 (130-149)

131 (122-150)

80 (72-85)

80 (77-90)

78 (75-86)

8 (72.7%)

14 (82.4%)

33 (97.1%)

Class Ⅱa N(%)

7 (63.6%)

3 (17.6%)

3 (8.8%)

Class Ⅱb N(%)

4 (36.4%)

6 (35.3%)

4 (11.8%)

Class Ⅲ N(%)

0 (0%)

Class Ⅳ N(%)

0 (0%)

Hg) diastolic blood pressure (mm Hg) Hypertension N (%)

RI

PT

Glomerular lesions

21 (61.8%)

1 (5.9%)

6 (17.6%)

0 (0%)

0 (0%)

0 (0%)

7 (63.6%)

6 (35.3%)

2 (5.9%)

4 (36.4%)

7 (41.2%)

6 (17.6%)

0 (0%)

3 (17.6%)

12 (35.3%)

0 (0%)

1 (5.9%)

14 (41.2%)

Score=0 N(%)

0 (0%)

0 (0%)

0 (0%)

Score=1 N(%)

1 (9.1%)

0 (0%)

0 (0%)

Score=2 N(%)

5 (45.5%)

7 (41.2%)

7 (20.6%)

Score=3 N(%)

4 (36.4%)

2 (11.8%)

1 ( 2.9%)

Score=4 N(%)

1 (9.1%)

8 (47.1%)

26 (76.5%)

(Interstitial fibrosis and tubular

MA

atrophy score plus interstitial inflammation score)

Score=2 N(%) Score=3 N(%)

EP T

Score=4 N(%)

ED

Score=1 N(%)

Score=5 N(%)

SC

NU

Tubular and interstitial lesions

7 (41.2%)

AC C

Vascular lesions (score)

Data are presented as mean ± SD for continuous variables with normal distribution, median (interquartile range) for continuous variables without normal distribution and absolute value and percentage for frequency of categorical variables. eGFR: estimated glomerular filtration rate; MG: Microalbuminuria stage group; PG: Proteinuria stage

ACCEPTED MANUSCRIPT group; RIG: renal insufficiency stage group.**p<0.01 vs MG; #p<0.05 vs PG;

AC C

EP T

ED

MA

NU

SC

RI

PT

##p<0.01 vs PG.

ACCEPTED MANUSCRIPT

Table 2 Correlation between the level of C5b-9 and indices reflecting tubular interstitial damage (n=62). Indices reflecting tubular interstitial damage

C5b-9 in tubular interstitial area p

Tubular and interstitial injury score

0.519

<0.01

Relative interstitial volume

0.556

<0.01

Mononuclear cell number in tubular interstitium Plasma cell number in tubular interstitium

0.526 0.297

<0.01 <0.05

Neutrophil number in tubular interstitium

0.389

<0.01

Infiltrating cell number in tubular interstitium

0.525

<0.01

Urine RBP level

0.486

<0.01

0.363

<0.01

RI

PT

r

Urine NGAL level

AC C

EP T

ED

MA

NU

SC

RBP: Retinol-binding protein; NGAL: Neutrophil gelatinase associated lipocalin.

ACCEPTED MANUSCRIPT Table 3 Correlation of the levels of tubular interstitial MBL and MASP1 with the level of tubular interstitial C5b-9 and indices of renal tubular and interstitial injury (n=62). MASP1/3 level p

r

p

C5b-9 level

0.516

<0.01

0.324

<0.05

TIS

0.448

<0.01

Urine NGAL

0.364

<0.01

Urine RBP level

0.502

<0.01

0.465

<0.01

Urine NAG level

0.322

0.371

<0.01

PT

r

SC

MBL level

<0.01

0.373

<0.01

RI

0.446

<0.05

NU

TIS: Tubular and interstitial injury score; RBP: Retinol-binding protein; NAG:

AC C

EP T

ED

MA

N-acetyl-β-glucosaminidase; NGAL: Neutrophil gelatinase associated lipocalin.

ACCEPTED MANUSCRIPT Highlights

EP T

ED

MA

NU

SC

RI

PT

Renal local complement activation is found in type 2 diabetic nephropathy patients. MBL pathway is proposed to contribute to the renal local complement activation. MBL pathway is proposed to contribute to renal interstitial damage in the disease.

AC C

  