Predictive Significance Of Kidney Myeloid-RelatedProtein 8 Expression in Patients With Obesity- Or Type 2Diabetes-Associated Kidney Diseases
Mar 16, 2022
Contact: Audrey Hu Whatsapp/hp: 0086 13880143964 Email: audrey.hu@wecistanche.com
Takashige Kuwabara1, Kiyoshi Mori1,2*, Masato Kasahara1,3, Hideki Yokoi1, Hirotaka Imamaki1,Akira Ishii1, Kenichi Koga1, Akira Sugawara4, Shinji Yasuno3, Kenji Ueshima3, Takashi Morikawa5,Yoshio Konishi5, Masahito Imanishi5, Akira Nishiyama6, Kazuwa Nakao1,2, Masashi Mukoyama1
1 Department of Medicine and Clinical Science, Kyoto University Graduate School of Medicine, Kyoto, Japan,
2 Medical Innovation Center, Kyoto University GraduateSchool of Medicine, Kyoto, Japan,
3 Department of EBM Research, Institute for Advancement of Clinical and Translational Science, Kyoto University Hospital, Kyoto, Japan,
4 Department of Nephrology, Osaka Red Cross Hospital, Osaka, Japan,
5 Division of Nephrology and Hypertension, Osaka City General Hospital, Osaka, Japan,
6 Department of Pharmacology, Kagawa University Medical School, Kagawa, Japan
Abstract
Background and Objective: We have reported that toll-like receptor 4 (TLR4) and one of its endogenous ligands, myeloid-related protein 8 (MRP8 or S100A8), play an important role in the progression of diabetic nephropathy in mice. The aim of this study was to evaluate the significance of kidney MRP8 expression in patients with obesity- or type 2 diabetes-associated kidney diseases. Methods: In diabetic, obese, or control subjects, MRP8 mRNA and protein expression levels in renal biopsy samples were determined by real-time RT-PCR and immunohistochemistry (n = 28 and 65, respectively), and their associations with baseline and prognostic parameters were analyzed. Effects of MRP8 upon pro-inflammatory gene expressions were examined using macrophages. Results: Kidney MRP8 gene and protein expression levels were elevated in obese or diabetic groups compared to the control group. Among all subjects, by univariate linear regression analysis, glomerular MRP8-positive cell count and tubulointerstitial MRP8-positive area at baseline were both, respectively, correlated not only with various known risk factors for diabetic nephropathy (such as systolic blood pressure, proteinuria, and serum creatinine) but also with the extent of glomerulosclerosis and tubulointerstitial fibrosis. Independent factors predicting urinary protein levels a year later were examined by multivariate analysis, and they included glomerular MRP8-positive cell count (b = 0.59, P,0.001), proteinuria(b = 0.37, P = 0.002) and systolic blood pressure (b = 0.21, P = 0.04) at baseline, after adjustment for known risk factors. MRP8protein expression was observed in CD68-positive macrophages and atrophic tubules. In cultured mouse macrophages, MRP8 protein-induced proinflammatory cytokine expressions also triggered auto-induction of MRP8 in a TLR4-dependent manner.
Conclusions: Glomerular MRP8 expression appears to be associated with the progression of proteinuria in obese or types 2 diabetic patients, possibly by inducing inflammatory changes in macrophages through TLR4 signaling.
Effects of cistanche: treating kidney diseases
Introduction
Chronic inflammation plays an important role in the pathogenesis of diabetes or obesity and its cardiovascular complications [1]. The involvement of innate immune receptors and the endogenous ligands in the process of chronic inflammation has been implicated. Myeloid-related protein 8 (MRP8, also known asS100A8 or calgranulin A) was originally identified as a cytoplasmic calcium-binding protein in neutrophils and monocytes [2] and has become widely recognized as a potent endogenous ligand for toll-like receptor 4 (TLR4) in various diseases including septic shock, vascular and autoimmune disorders [3,4,5]. We have recently proposed that MRP8/TLR4 signaling plays an important role in the hyperlipidemia-induced progression of diabetic nephropathy [6]. Glomerular macrophages and collecting duct cells are major sources of MRP8 in mouse models of diabetic nephropathy[6] and renal fibrosis [7], respectively. Plasma levels of MRP8, which usually forms a heterodimeric complex with a binding partner MRP14 in the bloodstream, are increased in obese subjects [8,9]. However, there have been no reports investigating the renal expression of MRP8 in patients with obesity or type 2diabetes and its association with renal prognosis.
The aim of this study was to determine mRNA and protein expression levels of MRP8 in the kidney of Japanese patients with diabetic nephropathy (DN), obesity-related glomerulopathy(ORG), minimal change nephrotic syndrome (MCNS) or minor glomerular abnormality (MGA), which were all diagnosed by renal biopsy, and to evaluate whether renal MRP8 expression can predict renal outcomes.
Cistanche can improve kidney function
Materials and Methods
Ethics statement
The human study was conducted according to the principles expressed in the Declaration of Helsinki and was approved by the ethical Committees on Human Research of Kyoto UniversityGraduate School of Medicine and Osaka City General Hospital,respectively. All participants gave written informed consent. The animal study protocol was approved by the Animal ResearchCommittee of Kyoto University Graduate School of Medicine(Permit Number: Med Kyo 13318). All animal surgery was performed under sodium pentobarbital anesthesia, and all efforts were made to minimize suffering.
Study subjects
Proteinuric patients with obesity or type 2 diabetes who underwent renal biopsy were enrolled in this study. Patients with infectious disease, cancer, liver disease or collagen disease were excluded. Proteinuria was defined as urinary protein greater than0.5 g/g creatinine or urinary albumin greater than 300 mg/g creatinine in at least two consecutive measurements. Obesity was defined as a body mass index (BMI) greater than 25.0 (kg/m2). Type2 diabetes was diagnosed in accordance with the criteria of theWorld Health Organization. Biochemical measurements on admission for renal biopsy were used as baseline characteristics for cross-sectional analysis. Estimated glomerular filtration rate(eGFR) was calculated using a simplified prediction equation proposed by the Japanese Society of Nephrology: eGFR (ml/min/1.73 m2) = 1946 [age (years)]20.287 6 [serum creatinine (mg perdl)]21.094 60.739 (for females), which is a validated local modification of MDRD [10]. The serum concentrations of creatinine were measured using an enzymatic method.
For immunohistochemistry, 65 Japanese patients who underwent renal biopsy at the Department of Medicine and ClinicalScience, Kyoto University Hospital between 2000 and 2011 were analyzed. Biopsy-proven diagnoses of all patients during this period are listed in Table S1 in File S1. The subjects examined in this work included DN (n = 19), ORG (n = 10) and non-obese,non-diabetic control subjects who were diagnosed as MGA(n = 19) or MCNS (n = 17). Some cases in these categories were excluded because residual samples available contained less than 10glomeruli. Definition of DN consisted of (1) more than 5 years duration after the onset of diabetes, (2) existence of micro- or macro-albuminuria, (3) compatible histopathological changes with DN such as glomerular basement membrane thickening, mesangial expansion, nodular sclerosis (Kimmelstiel-Wilson nodules)and/or arteriolar hyalinosis, and (4) exclusion of other causes for renal disorders [11]. ORG was defined morphologically as focal segmental glomerulosclerosis and/or glomerulomegaly in subjects having both obesity and proteinuria, whose definitions were described above [12,13].
For mRNA expression analysis, low-quality samples, in which18S ribosomal RNA (rRNA) levels were lower than the detection sensitivity limit by real-time RT-PCR, were excluded. Subjects enrolled consisted of 22 type 2 diabetic patients who underwent renal biopsy at Osaka City General Hospital between 2000 and2010, and 6 non-diabetic control subjects, who had biopsy-provenMGA.
Tables 1 and 2 summarize the baseline clinical characteristics of the patients who were examined by immunohistochemical or gene expression analysis, respectively. For light microscopy, the tissue specimens were processed according to standard procedures. Sections were stained with hematoxylin-eosin, periodic acid-Schiff, periodic-acid methenamine silver or Masson trichrome(Fig. S1). The ratios of the number of glomeruli with global sclerosis among that of total glomeruli and the relative areas of tubulointerstitial fibrosis were evaluated independently by two pathologists unaware of diagnosis and clinical data.
Effects of cistanche: treating kidney diseases
Definition of renal outcomes
The following two prognostic indicators were examined by linear regression and logistic regression analyses, respectively: (1)the extent of proteinuria measured at one year post-biopsy, and (2)renal event defined as the annual increase in serum creatinine by .50% from baseline or initiation of chronic dialysis.
Immunohistochemistry
Immunohistochemistry of MRP8 and CD68 was carried out using kidney sections (thickness 4 mm) fixed with 4% buffered paraformaldehyde. After antigen retrieval by citrate buffer, kidney sections were incubated with 10% goat serum, followed by mouse anti-human MRP8 (1:100; BMA biomedicals, Augst, Switzerland)[14] or mouse anti-human CD68 antibodies (1:50; DAKO, Ely, UK), respectively. Primary antibodies were visualized with horseradish peroxidase-conjugated secondary antibody and 3,3-diaminobenzidine tetrahydrochloride (Dako USA, Carpinteria, CA). Nuclei were counterstained with hematoxylin. MRP8-positive cells were counted in more than 10 glomeruli, and MRP8-positive area in tubulointerstitium was measured quantitatively to obtain an average for each subject using MetaMorph 7.5software (Molecular Devices, Downingtown, PA, USA). Colocalization of CD68- and MRP8-positive cells was evaluated with serial sections. There was neither MRP8 nor CD68 signal in negative controls stained without first antibody (Fig. S2). Bypreincubation of anti-MRP8 antibody with 20 molar excess of recombinant human MRP8 protein (Life Technologies, Carlsbad, CA, USA) at 4uC overnight, staining was markedly reduced, if not completely, further supporting the specificity of the antibody (Fig.S3).
Cistanche to treat kidney
Evaluation of mRNA expression
Frozen kidney sections were separated into glomeruli and nonglomerulus tissues by laser capture micro-dissection (LM200; Olympus, Tokyo, Japan) as previously described [15]. Total RNA was extracted with RNeasy mini kit (Qiagen, Tokyo, Japan).mRNA expression levels were determined by TaqMan real-time PCR (Applied Biosystems, Foster City, CA, USA) [16,17]. Expression levels of all genes were normalized by 18S rRNA(internal control) levels. See Table S2 in File S1 for primer and probe sequences. Eukaryotic 18S rRNA was detected with PreDeveloped TaqMan Assay Reagents (Applied Biosystems).
Table 1. Baseline clinical characteristics of patients at renal biopsy who were analyzed for MRP8 protein expression by immunohistochemistry.
MGA: minor glomerular abnormality, MCNS: minimal change nephrotic syndrome, ORG: obesity-related glomerulopathy, DN: diabetic nephropathy, BMI: body mass index, BUN: blood urea nitrogen, CRP: C-reactive protein. Data are means 6 SD. *Overall differences between MGA, MCNS, ORG and DN groups were compared byANOVA.doi:10.1371/journal.pone.0088942.t001
MRP8 treatment of macrophages
Bone marrow-derived macrophages were generated from wild-type or TLR4 knockout (KO) mice [18] on C57BL/6J genetic background (Oriental BioService, Kyoto, Japan) as described previously [6]. Briefly, following lysis of red blood cells, bone marrow cells were resuspended in a medium containing 20% fetal calf serum and 50 ng/ml recombinant human macrophage-colony-stimulating factor (Peprotech, Rocky Hill, NJ, USA), and cultured at 37uC in 5% CO2 atmosphere. On day 7, macrophages were incubated with recombinant mouse MRP8 (Abnova, Taipei, Taiwan) or vehicle for 4 hours. Polymyxin B (25 mg/ml, NacalaiTesque, Kyoto, Japan) was added to each well to minimize contamination of endotoxin as described previously [3,19]. Noendotoxin was detected at any concentration of MRP8 tested hereafter incubation with 25 mg/ml of polymyxin B by ToxinSensorChromogenic LAL Endotoxin Assay Kit (GenScript, Piscataway, NJ, USA). Total RNA from cells was extracted with RNeasy MiniKit, and mRNA expression levels of interleukin-1 beta (IL-1b), tumor necrosis factor-alpha (TNFa), and MRP8 were determined by TaqMan real-time RT-PCR. Expression levels of all genes were normalized by rodent GAPDH levels (internal control, Pre-Developed TaqMan Assay Reagents). Primer and Probe sequences for real-time PCR are listed in Table S2 in File S1.
Statistical analysis
Data are expressed as means 6 SD, or means 695%confidence interval (CI) when appropriate. For the comparison among four groups, one-way or two-way ANOVA with Bonferroni’s post-hoc analysis was used, and categorical variables were compared using x2 test. The student’s unpaired t-test was applied for comparison between two groups as appropriate. Spearman correlation coefficients were estimated to determine associations between two variables. To examine the effects of baseline covariates determining the extent of glomerular or tubulointerstitial MRP8 expression or urinary protein levels one year after biopsy, univariate and multivariate linear regression analyses were performed. Logistic regression analysis was used to analyze explanatory variables predicting the occurrence of renal events. All data were analyzed using StatView 5.0 software (SAS InstituteInc., Cary, NC, USA). P values,0.05 were considered statistically significant.
Effects of cistanche: treating kidney diseases
Results
We compared MRP8 protein expression levels in the kidney among DN, ORG, and non-obese, non-diabetic control (MGA and CNS) groups. Immunohistochemical analysis revealed that both glomerular MRP8-positive cell count (Fig. 1A) and tubulointerstitial MRP8-positive area (Fig. 1B) in DN were significantly larger than those in other groups including MGA, MCNS, and ORG (P<0.01). ORG subjects also showed a tendency of elevated MRP8expression compared to MGA and MCNS (Fig. 1A, 1B). Furthermore, glomerular MRP8 mRNA expression levels in DN subjects were significantly higher compared to non-DM control subjects (P,0.01, Fig. 1C). In non-glomerulus tissues, MRP8mRNA expression levels were much lower than those in glomeruli, both in non-DM and DM groups. Abundant MRP8 protein expression in the tubulointerstitium of DN cases was not clearly reflected into increased mRNA expression, which may be partly caused by deposition of blood-derived proteins in the tubulointerstitium as discussed in the next section. As shown in representative photos (Fig. 1D–G, see Fig. S4 in detail), renal biopsy samples from MGA and MCNS subjects showed few MRP8-positive cells in glomeruli (Fig. 1D, 1E and Fig. S4). In ORG subjects, someMRP8-positive cells appeared in glomeruli and tubulointerstitium(Fig. 1F and Fig. S4). In DN subjects, a marked increase of MRP8-expressing cells in glomeruli and significant expansion of MRP8-positive areas in the tubulointerstitium were observed in a focal manner (Fig. 1G and Fig. S4). Of note, MRP8-positive cells were absent in nodular sclerosing lesions of diabetic glomeruli (Fig. S4:DN case 2, 3) as described previously for sclerotic lesions in ANCA-associated glomerulonephritis [20]. Paired immunohistochemistry for CD68 and MRP8 in serial sections suggested thatMRP8 signals were, at least in part, observed in macrophages expressing CD68 (Fig. 2), as we reported in a mouse model of diabetic nephropathy [6]. Besides, focally injured atrophic tubular epithelial cells also strongly expressed MRP8, which were surrounded by MRP8(+)-, CD68(+)-positive macrophages (Fig. 2, Fig S4: DN case 3-5). In the cases with nephrotic range proteinuria, MRP8 staining was also observed along brush borders of proximal tubules both in MCNS and DN cases (Fig. S4). Since the sample number of mRNA expression was too small for multivariate analysis, the following analyses were performed using data from patients studied by immunohistochemistry.
Table 2. Baseline clinical characteristics of patients at renal biopsy who were analyzed for MRP8 mRNA expression by real-time RT-PCR.
The associations between kidney MRP8 signals and baseline clinical parameters at the time of renal biopsy were analyzed cross-sectionally (Table 3). By univariate analysis, glomerular and/or tubulointerstitial MRP8 protein expression was significantly correlated to age, systolic and diastolic blood pressures, urinary protein, serum levels of creatinine, BUN and HDL cholesterol,eGFR, and extent of global glomerulosclerosis and tubulointerstitial fibrosis. These parameters were further examined by multivariate analysis after excluding diastolic blood pressure,eGFR and BUN because of collinearity. Percentage of tubulointerstitial fibrosis was independently correlated with glomerularMRP8 signals (b = 0.62, adjusted P = 0.02) and tubulointerstitialMRP8 signals (b = 0.85, adjusted P,0.001), respectively. Additionally, tubulointerstitial MRP8 signals were also independently correlated with baseline proteinuria (b = 0.20, adjusted P = 0.01).Scattered plot analyses between MRP8 signals in glomeruli or tubulointerstitium and clinical parameters indicated that MCNSgroup had a distinct distribution pattern from other groups, especially as to urinary protein and serum LDL cholesterol levels(Fig. S5A–D, S6A–D). Exclusion of MCNS group improved correlation between MRP8 signals and urinary protein or serum LDL-cholesterol levels (Fig. S5E–F, S6E–F). Therefore, we carried out a sub-analysis excluding MCNS patients and found that urinary protein was an independent factor correlated with glomerularMRP8 signals by multivariate analysis (Table 4; b = 0.36, adjusted P = 0.03).
Next, we performed linear regression or logistic regression analyses to identify explanatory factors predicting renal outcomes which were the extent of proteinuria a year later and renal event within a year. Since there was a good association between glomerular and tubulointerstitial MRP8 signals (Fig. S7; R = 0.67,P,0.001), these parameters were alternatively enrolled in further analyses. We evaluated the association between baseline parameters and urinary protein at 1 year after renal biopsy by multiple regression analysis. As shown in Table 5, glomerular MRP8 signal(b = 0.59, adjusted P,0.001) was a predictive factor for the extent of proteinuria a year later, as well as baseline systolic blood pressure (b = 0.21, adjusted P = 0.04) and baseline proteinuria(b = 0.37, adjusted P = 0.002) were. These parameters were independent of other known diabetic nephropathy risk factors including renal dysfunction (serum creatinine) and extent of global sclerosis and tubulointerstitial fibrosis [11,21–24]. On the other hand, tubulointerstitial MRP8 signal (b = 0.34, adjusted P = 0.09)was not an independent predictive factor for urinary protein levels a year later. Renal events occurred in 7 patients (6 in DN and 1 in ORG cases) within a year after renal biopsy. By univariate analysis, not only the extent of glomerulosclerosis and tubulointerstitial fibrosis, and glomerular and tubulointerstitial MRP8 signals, but also blood pressures, renal dysfunction, and urinary protein levels at baseline were significant predictive factors for the occurrence of renal events. However, by multivariate analysis, their covariates were canceled out by each other (Table S3 in File S1), likely due to high correlations among these parameters.
Finally, we examined the potency of MRP8 as an endogenous ligand for TLR4 using cultured macrophages. In bone marrow-derived macrophages from wild-type mice, MRP8 protein-induced upregulation of proinflammatory cytokine genes such as IL-1b andTNFa and also triggered auto-induction of MRP8, in a dose-dependent manner between 10–1000 ng/ml. These effects ofMRP8 were suppressed approximately by two-thirds in macrophages obtained from TLR4 KO mice (P,0.01) (Fig. 3).
Figure 1. Immunohistochemical and mRNA analyses for MRP8 in kidney biopsy samples. Quantification of glomerular MRP8-positive cell count (A) and tubulointerstitial MRP8-positive area (B). mRNA expression of MRP8 in glomerular and non-glomerular fractions (C). Open bars: non-obese, non-diabetic controls which are MGA or MCNS, closed bars: ORG or DN (A–C). Representative pictures of MGA, MCNS, ORG, and DN groups (D–G). MGA: minor glomerular abnormality, MCNS: minimal change nephrotic syndrome, ORG: obesity-related glomerulopathy, DN: diabetic nephropathy. *P,0.01.doi:10.1371/journal.pone.0088942.g001
Discussion
The present study has demonstrated that MRP8 is abundantly expressed in glomeruli and tubulointerstitium of patients with DNas compared to ORG and non-obese, non-diabetic control (MGAand MCNS). Furthermore, in ORG subjects, MRP8 expression levels tended to be higher than MGA or MCNS subjects. In baseline cross-sectional investigation including all subjects, by univariate linear regression analysis, glomerular MRP8-positive cell count and tubulointerstitial MRP8-positive area were both, respectively, correlated not only with various known risk factors for diabetic nephropathy (such as systolic blood pressure, proteinuria, and serum creatinine) but also with the extent of glomerulosclerosis and tubulointerstitial fibrosis. By multivariate analysis, tubulointerstitial MRP8-positive area was significantly correlated with proteinuria and tubulointerstitial fibrosis. Glomerular MRP8-positive cell count was significantly correlated with tubulointerstitial fibrosis in the primary analysis, and with proteinuria in a sub-analysis excluding MCNS group. Immunohistochemistry indicated that MRP8 was expressed, at least partly, by CD68(+)-expressing macrophages and atrophic tubules. These findings raise the possibility that kidney MRP8 signals in glomeruli or tubulointerstitium may serve as novel markers of diabetic nephropathy.
Figure 2. Localization of CD68 and MRP8 protein expression in serial sections of diabetic nephropathy cases. Expression of CD68(A, B) and MRP8 expression (C, D) in paired renal specimens (A and C, orB and D). Arrows indicate colocalization of CD68 and MRP8 signals.doi:10.1371/journal.pone.0088942.g002
In the prognostic study, multivariate analysis revealed that urinary protein levels at a year after renal biopsy were independently associated with glomerular MRP8-positive cell count, urinary protein and systolic blood pressure at baseline. Of note, glomerular MRP8 expression showed the strongest correlation with urinary protein a year later (b= 0.87), even stronger than baseline urinary protein (b = 0.78), by univariate analysis. It is partly because glomerular MRP8 expression is not largely elevated in ‘benign’ forms of proteinuria such as ones observed in MCNSpatients, whose levels of proteinuria are extremely high at diagnosis by renal biopsy but are usually resolved within a year after initiation of immunosuppressive therapy. These findings suggest that glomerular MRP8 expression may possess a unique predictive nature as a disease marker, which cannot be substituted by baseline proteinuria or routine pathological analysis evaluating global glomerulosclerosis and tubulointerstitial fibrosis. Moreover, we speculate that glomerular MRP8 expression is not a simple marker or bystander but an active player in glomerular injury as discussed below.
The coefficient of determination (R2) calculated with explanatory parameters enrolled in multiple regression analysis was 0.52* and 0.74#, respectively. y, years; BP, blood pressure; gCr, g creatinine; T-chol, total cholesterol; HDL-Chol, HDL cholesterol; LDL-chol, LDL cholesterol; GS, glomerulosclerosis; TI, tubulointerstitial.doi:10.1371/journal.pone.0088942.t003
Table 4. Sub-analysis of the relationship between baseline clinical parameters and MRP8 signals, after exclusion of MCNS group.
Table 5. Multiple regression analysis for identification of factors predicting urinary protein levels 1 year after renal biopsy
Figure 3. Effects of MRP8 upon bone marrow-derived macrophages. Bone marrow-derived macrophages (BMDM) were stimulated with recombinant mouse MRP8 for 4 hours. Error bars indicate 95% CI and statistical analyses were performed with log-transformed values. Two-way ANOVA revealed significant effects of genotypes, MRP8 concentrations and their interactions for expression of all 3 genes (P,0.001 for all comparisons). n = 4. WT, wild-type; KO, knockout; IL-1b, interleukin 1 beta; TNFa, tumor necrosis factor-alpha. *P,0.01 among different concentrations, #P,0.01 among genotypes.doi:10.1371/journal.pone.0088942.g003
In another attempt of a longitudinal study, logistic regression analysis failed to find out any independent predictors for the occurrence of a renal event within a year. We speculate that this is partly because tubulointerstitial MRP8 expression and tubulointerstitial fibrosis were two potent predictors for renal event in univariate analysis but their significances were canceled by each other in multivariate analysis. These two parameters showed a strong correlation (R = 0.68, P,0.001) (Fig. S8), suggesting that these two parameters might be equivalent to predicting renal events. Indeed, interstitial MRP8 expression showed quite a similar pattern to interstitial fibrosis evaluated by Masson trichrome staining. The quantity of interstitial MRP8 largely depends on the positive signals in atrophic tubules rather than those in macrophages, whose feature differs from that of glomerular MRP8 in a punctate distribution. Furthermore, a small sample size, short observation period, and few subjects who developed renal events might have reduced the detection power. Since MRP8 expression in tubular epithelial cells plays a causative role in the progression of tubulointerstitial inflammation in a mouse model of renal fibrosis[7], further analysis will be needed to clarify the role of tubulointerstitial MRP8 in DN.
In accordance with our previous study [6], MRP8 mRNA was upregulated predominantly in the glomerular fraction of human DN subjects as compared to control subjects with MGA. On the other hand, MRP8 protein expression was observed not only in the glomerulus but also in the tubulointerstitium. In this regard, it should be noted that there were two distinct patterns of MRP8staining in the tubulointerstitium of DN. One was intense and focal staining in severely atrophic tubules. The other was mild staining distributed along the brush border of proximal tubules, which was also found in ORG and MCNS. The latter signals likely represent MRP8 protein derived from the blood and reabsorbed by proximal tubules, which should not be accompanied by increasedMRP8 mRNA expression. Concerning proteins other than MRP8, we and others have recently reported similar phenomena of immunoreactive protein detection in the proximal tubules caused by reabsorption but not by renal synthesis [25,26]. On the other hand, since a little MRP8 staining remained in antibody absorption test, especially in glomerular exudative lesions and severely-scarred, fibrotic lesions around atrophic tubules, the presence of non-specific signals cannot be completely negated (Fig. S3).
Glomerular MRP8 signals mainly showed punctuate patterns in DN subjects (Fig. 1, Fig. S4). Since both of CD68 and MRP8 were detected by mouse monoclonal antibodies, the localization of these molecules was evaluated by serial sections, not by double immunostaining. Staining patterns of MRP8 were compatible with those in other inflammatory renal disorders including IgAnephritis [27], membranoproliferative glomerulonephritis [14], and ANCA-related glomerulonephritis [20], in which macrophages were suggested as a major source of MRP8, as we reported in a rodent model [6]. In addition, neutrophils could be considered as another source of MRP8 affecting vascular complications [28]. Currently, we are investigating the molecular mechanism whyMRP8 is predominantly upregulated in myeloid-lineage cells infiltrating glomeruli. In vitro study revealed that MRP8 induced inflammatory cytokine expression and also potentiated expression of MRP8 itself in macrophages in a TLR4-dependent manner. Additionally, MRP8-positive cells were absent in nodular sclerosing lesions, suggesting that glomerular MRP8 might reflect ongoing glomerular damage [20]. Importantly, a largest-scale human study reported that MRP8 gene expression in blood mononuclear cells of type 1 diabetic patients is significantly elevated in subjects with diabetic complications including nephropathy [29].
Since inhibition of the renin-angiotensin system (RAS) is an important determinant of renal outcomes, we examined the effects of RASblockade on kidney MRP8 expression. We found no significant difference in kidney MRP8 mRNA expression between DN patients treated with or without RAS blockade (Fig. S9), probably because cases treated with RAS blockade tended to have more severe hypertension and proteinuria than cases without RAS blockade
In obese humans and mice, increased plasma MRP8/14complex may reflect a degree of obesity and originate from adipocytes as well as leukocytes [8,9]. Our ORG cases had mild staining of MRP8 in proximal tubules, suggesting increased plasma levels of MRP8. In contrast, there was no significant correlation between tubulointerstitial MRP8 expression and body mass index (Fig. S6G). Therefore, local MRP8 expression in the kidney may serve better as a marker for renal injury rather than for obesity [8,9].
Cistanche for improving kidney dysfunction
Our study has several limitations. The sample number of each group studied was small. Non-identical subjects were enrolled in the mRNA and immunohistochemical analyses. Since we only analyzed patients who underwent renal biopsy, the composition of patients investigated here may not reflect those in general type 2diabetic patients or in general chronic kidney disease subjects. Although age was not retained as an independent factor associated with MRP8 signals in our data (Table 3), it is known that aging associated with chronic inflammation [30]. The effects of age cannot be completely neglected. Although most MRP8 signals were lost in the antibody absorption test, there were some positive signals remaining, which might be caused by the non-specific binding of the first antibody. As discussed above, investigation of renal MRP8expression by renal biopsy helps us understand the pathophysiology and prognosis of chronic kidney diseases, especially associated with obesity and diabetes, but has a disadvantage for routine and repeated use in out-patient clinics
In summary, the present study suggests that the expression ofMRP8 in the kidney reflects the current pathological status and also predicts renal outcomes in patients with obesity or type 2diabetes. Further investigations studying urinary MRP8 levels among obese or diabetic patients on large scale may be warranted
Supporting Information
Figure S1 Representative photos showing renal biopsy sections of a DN patient stained with (A) periodic acid Schiff, (B) periodic-acid methenamine silver or (C)Masson trichrome. The ratio of the number of glomeruli with global sclerosis (arrows) among that of total glomeruli and the relative area of tubulointerstitial fibrosis was 33% and 65%, respectively, in this patient. (TIFF)
Figure S2 Immunohistochemistry for MRP8 and CD68proteins in DN patients. Photos in the right column show negative control experiments without 1st antibody. (TIF)
Figure S3 Antibody absorption test for MRP8 staining.PBS: phosphate-buffered saline, rhMRP8: recombinant humanMRP8.(TIF)
Figure S4 Representative photos of MRP8 expression in MGA, MCNS, ORG, and DN groups.(TIF)
Figure S5 Correlation between glomerular MRP8-positive cell count and clinical parameters. The log-transformed values of MRP8 signals were used. The correlations were analyzed using all of 4 groups (A–D, G) or 3 groups excluding the MCNS group (E, F). Open circles: minor glomerular abnormality(MGA), closed circles: minimal change nephrotic syndrome(MCNS), open triangles: obesity-related glomerulopathy (ORG), closed triangles: diabetic nephropathy (DN).(TIF)
Figure S6 Correlation between tubulointerstitial MRP8-positive area and clinical parameters. The log-transformed values of MRP8 signals were used. These correlations were analyzed using all of the 4 groups (A–D, G) or 3 groups excluding the MCNS group (E, F).(TIF)
Figure S7 Correlation between glomerular and tubulointerstitial MRP8 expression. The log-transformed values ofMRP8 signals were used.(TIF)
Figure S8 Correlation between tubulointerstitial MRP8-positive area and tubulointerstitial fibrosis. The log-transformed values of MRP8 signals were used.(TIF)
Figure S9 Renal mRNA expression of MRP8 in DN patients with or without renin-angiotensin blockade.N.S.: not significant. n = 15 (Yes), 6 (No). Among 22 DN cases, information about medication was not available in one patient.(TIF)
File S1 Supporting Tables.
Table S1, Pathological diagnoses of all cases who underwent renal biopsy at Department of Medicine and Clinical Science, Kyoto University Hospital between 2000 and 2011. Table S2, Primer and probe sequences for TaqMan real-time RT-PCR. Table S3, Logistic regression analysis for the occurrence of a renal event within a year. (DOC)AcknowledgmentsWe gratefully acknowledges S. Tanaka for statistical advice, Y. Ogawa, N.Igarashi, and C. Kimura for technical assistance, and A. Yamamoto and SOgino for secretary assistance. Author ContributionsConceived and designed the experiments: TK KM MK HY MI AN KNMM. Performed the experiments: TK HI AI KK TM YK MI AN. Analyzed the data: TK KM MK HY AS SY KU KN MM. Contributed reagents/materials/analysis tools: TK KM. Wrote the paper: TK KMMM.
Effects of cistanche: treating kidney diseases
References
1. Donath MY, Shoelson SE (2011) Type 2 diabetes as an inflammatory disease.Nat Rev Immunol 11: 98–107.
2. Odink K, Cerletti N, Bruggen J, Clerc RG, Tarcsay L, et al. (1987) Twocalcium-binding proteins in infiltrating macrophages of rheumatoid arthritis.Nature 330: 80–82.
3. Vogl T, Tenbrock K, Ludwig S, Leukert N, Ehrhardt C, et al. (2007) Mrp8 andMrp14 are endogenous activators of Toll-like receptor 4, promoting lethal,endotoxin-induced shock. Nat Med 13: 1042–1049.
4. Croce K, Gao H, Wang Y, Mooroka T, Sakuma M, et al. (2009) Myeloidrelated protein-8/14 is critical for the biological response to vascular injury.Circulation 120: 427–436.
5. Loser K, Vogl T, Voskort M, Lueken A, Kupas V, et al. (2010) The Toll-like receptor 4 ligands Mrp8 and Mrp14 are crucial in the development of autoreactive CD8+ T cells. Nat Med 16: 713–717.
6. Kuwabara T, Mori K, Mukoyama M, Kasahara M, Yokoi H, et al. (2012)Exacerbation of diabetic nephropathy by hyperlipidemia is mediated by Toll-like receptor 4 in mice. Diabetologia 55: 2256–2266.
7. Fujiu K, Manabe I, Nagai R (2011) Renal collecting duct epithelial cells regulate inflammation in tubulointerstitial damage in mice. J Clin Invest 121: 3425–3441.
8. Sekimoto R, Kishida K, Nakatsuji H, Nakagawa T, Funahashi T, et al. (2012)High circulating levels of S100A8/A9 complex (calprotectin) in male Japanese with abdominal adiposity and dysregulated expression of S100A8 and S100A9 in adipose tissues of obese mice. Biochem Biophys Res Commun 419: 782–789.
9. Mortensen OH, Nielsen AR, Erikstrup C, Plomgaard P, Fischer CP, et al. (2009)Calprotectin—a novel marker of obesity. PLoS One 4: e7419.
10. Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, et al. (2009) Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis53: 982–992.
11. Gross JL, de Azevedo MJ, Silveiro SP, Canani LH, Caramori ML, et al. (2005)Diabetic nephropathy: diagnosis, prevention, and treatment. Diabetes Care 28:164–176.
12. Kambham N, Markowitz GS, Valeri AM, Lin J, D’Agati VD (2001) Obesity-related glomerulopathy: an emerging epidemic. Kidney Int 59: 1498–1509.
13. Praga M, Morales E (2006) Obesity, proteinuria, and progression of renal failure.Curr Opin Nephrol Hypertens 15: 481–486.
14. Kawasaki Y, Hosoya M, Takahashi A, Isome M, Tanji M, et al. (2005) Myeloidrelated protein 8 expressions on macrophages is a useful prognostic marker for renal dysfunction in children with MPGN type 1. Am J Kidney Dis 45: 510–518.
15. Nishiyama A, Konishi Y, Ohashi N, Morikawa T, Urushihara M, et al. (2011)Urinary angiotensinogen reflects the activity of intrarenal renin-angiotensin system in patients with IgA nephropathy. Nephrol Dial Transplant 26: 170–177.
16. Ogawa Y, Mukoyama M, Yokoi H, Kasahara M, Mori K, et al. (2012)Natriuretic peptide receptor guanylyl cyclase-A protects podocytes from aldosterone-induced glomerular injury. J Am Soc Nephrol 23: 1198–1209.
17. Yokoi H, Mukoyama M, Mori K, Kasahara M, Suganami T, et al. (2008)Overexpression of connective tissue growth factor in podocytes worsens diabetic nephropathy in mice. Kidney Int 73: 446–455.
18. Hoshino K, Takeuchi O, Kawai T, Sanjo H, Ogawa T, et al. (1999) Cutting edge: Toll-like receptor 4 (TLR4)-deficient mice are hyporesponsive to lipopolysaccharide: evidence for TLR4 as the Lps gene product. J Immunol162: 3749–3752.
19. Suganami T, Tanimoto-Koyama K, Nishida J, Itoh M, Yuan X, et al. (2007)Role of the Toll-like receptor 4/NF-kappaB pathway in saturated fatty acid-induced inflammatory changes in the interaction between adipocytes and macrophages. Arterioscler Thromb Vasc Biol 27: 84–91.
20. Pepper RJ, Hamour S, Chavele KM, Todd SK, Rasmussen N, et al. (2013)Leukocyte and serum S100A8/S100A9 expression reflects disease activity in ANCA-associated vasculitis and glomerulonephritis. Kidney Int 83: 1150–1158.
21. Ravid M, Brosh D, Ravid-Safran D, Levy Z, Rachmani R (1998) Main risk factors for nephropathy in type 2 diabetes mellitus are plasma cholesterol levels, mean blood pressure, and hyperglycemia. Arch Intern Med 158: 998–1004.
22. Adler AI, Stratton IM, Neil HA, Yudkin JS, Matthews DR, et al. (2000)Association of systolic blood pressure with macrovascular and microvascular complications of type 2 diabetes (UKPDS 36): a prospective observational study.BMJ 321: 412–419.
23. Ruggenenti P, Remuzzi G (1998) Nephropathy of type-2 diabetes mellitus. J Assoc Nephrol 9: 2157–2169.
24. Taft JL, Nolan CJ, Yeung SP, Hewitson TD, Martin FI (1994) Clinical and histological correlations of decline in renal function in diabetic patients with proteinuria. Diabetes 43: 1046–1051.
25. Matsusaka T, Niimura F, Shimizu A, Pastan I, Saito A, et al. (2012) Liverangiotensinogen is the primary source of renal angiotensin II. J Am Soc Nephrol23: 1181–1189.
26. Kuwabara T, Mori K, Mukoyama M, Kasahara M, Yokoi H, et al. (2009)Urinary neutrophil gelatinase-associated lipocalin levels reflect damage to glomeruli, proximal tubules, and distal nephrons. Kidney Int 75: 285–294.
27. Kawasaki Y, Suyama K, Go H, Imamura T, Ushijima Y, et al. (2009)Accumulation of macrophages expressing myeloid-related protein 8 associated with the progression of sclerotic changes in children with IgA nephropathy.Tohoku J Exp Med 218: 49–55.
28. Nagareddy PR, Murphy AJ, Stirzaker RA, Hu Y, Yu S, et al. (2013)Hyperglycemia promotes myelopoiesis and impairs the resolution of atherosclerosis. Cell Metab 17: 695–708.
29. Jin Y, Sharma A, Carey C, Hopkins D, Wang X, et al. (2013) The Expression of inflammatory Genes Is Upregulated in Peripheral Blood of Patients with Type 1Diabetes. Diabetes Care.
30. Longo VD, Finch CE (2003) Evolutionary medicine: from dwarf model systems to healthy centenarians? Science 299: 1342–1346.
Citation: Kuwabara T, Mori K, Kasahara M, Yokoi H, Imamaki H, et al. (2014) Predictive Significance of Kidney Myeloid-Related Protein 8 Expression in Patients with Obesity- or Type 2 Diabetes-Associated Kidney Diseases. PLoS ONE 9(2): e88942. doi:10.1371/journal.pone.0088942
Editor: Utpal Sen, University of Louisville, United States of America
Received September 11, 2013; Accepted January 14, 2014; Published February 18, 2014
Copyright: 2014 Kuwabara et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported in part by Grant-in-Aid for Diabetic Nephropathy and Nephrosclerosis Research from the Ministry of Health, Labour and Welfare of Japan (to KM), research grants from the Japanese Ministry of Education, Culture, Sports, Science, and Technology (to TK, KM, and MM), from the Japan Foundation for Applied Enzymology (to TK) and from the Smoking Research Foundation (to MM). The funders had no role in study design, data collection, and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: keyem@kuhp.kyoto-u.ac.jp
