In diabetic nephropathy (DN) proinflammatory chemokines and leukocyte infiltration correlate with tubulointerstitial injury and declining renal function. The atypical chemokine receptor ACKR2 is a chemokine scavenger receptor which binds and sequesters many inflammatory CC chemokines but does not transduce typical G-protein mediated signaling events. ACKR2 is known to regulate diverse inflammatory diseases but its role in DN has not been tested. In this study, we utilized ACKR2−/− mice to test whether ACKR2 elimination alters progression of diabetic kidney disease. Elimination of ACKR2 greatly reduced DN in OVE26 mice, an established DN model. Albuminuria was significantly lower at 2, 4, and 6 months of age. ACKR2 deletion did not affect diabetic blood glucose levels but significantly decreased parameters of renal inflammation including leukocyte infiltration and fibrosis. Activation of pathways that increase inflammatory gene expression was attenuated. Human biopsies stained with ACKR2 antibody revealed increased staining in diabetic kidney, especially in some tubule and interstitial cells. The results demonstrate a significant interaction between diabetes and ACKR2 protein in the kidney. Unexpectedly, ACKR2 deletion reduced renal inflammation in diabetes and the ultimate response was a high degree of protection from diabetic nephropathy.
Although hyperglycemia is the initiating and essential cause for all diabetic complications there is accumulating evidence that inflammatory processes activated by chronic elevated glucose are integral to the development of diabetic complications [
ACKR2 is a chemokine decoy receptor [
All animal procedures followed the NIH Guide for the Care and Use of Laboratory Animals and were approved by the University of Louisville Institutional Animal Care and Use Committee. ACKR2 mice on the C57BL/6 background originally from Charles River Italia (Calco, Italy) [
Glucose was assayed in serum samples obtained from nonfasted mice at 6 months of age by the Glucose (HK) Assay Kit (Sigma-Aldrich). At 2 months urine glucose was evaluated with Clinistix (Bayer). Albumin was measured from spot urine samples with a mouse albumin ELISA kit (Bethyl Laboratories, Montgomery, TX) within the linear range of the assay. Urine creatinine was measured with a creatinine assay kit (DICT-500, BioAssay Systems). Urine albumin was expressed as the ratio of albumin to creatinine (
Kidneys were fixed overnight in 10% neutral buffered formalin and embedded in paraffin. Sagittal tissue sections from the center of the kidney were stained with Masson’s trichrome using standard protocols. Stained slides were imaged with a 20x objective. Fibrosis was semiquantitatively scored by a blinded observer for the number of blue stained fibrotic areas per section. Renal inflammatory cell infiltration was evaluated by staining sections with rat anti-mouse CD45 antibody (Angio-Proteomie, Boston, MA). Positive staining was detected with HRP conjugated second antibody and diaminobenzidine (DAB). CD45 positive cell infiltration was evaluated by quantitating the DAB stained pixel area in 8 random, nonoverlapping 200x image fields from the cortical region per mouse with 3 mice per group. Digital images were taken by an observer blind to the identity of the section and the number of positive pixels was quantified by another observer blind to section identity. Pixel number was determined using the ability of Adobe Photoshop to select areas of matching color intensity.
RNA extraction was done with the RNeasy Mini Kit (Qiagen, Santa Clarita, CA, USA) from frozen kidneys. Extracted RNA was checked for quality on Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, USA). The RNA samples having RNA integrity number (RIN) above 8.8 (average 9.1) were used for probe preparation. A 100 ng aliquot of RNA from each mouse was used for probe preparation with an Ambion WT Expression kit. The kit generates sense-strand cDNA from total RNA for fragmentation and labeling was done with an Affymetrix GeneChip WT Terminal Labeling Kit (PN90067). Probes from 3 six-month-old female mice in each group were hybridized to Affymetrix mouse gene 1.0 ST exon arrays and scanned with a GCS 3000 7G scanner and signals were analyzed with Command Console software (Affymetrix, Santa Clara, CA). Gene expression profiles were uploaded to Ingenuity software (Ingenuity Systems,
Total RNA was extracted from whole kidney using TRIzol reagent (Invitrogen, Carlsbad, CA). The cDNA was synthesized with high-capacity cDNA archive kit (p/n 4322171, Applied Biosystems, Foster City, CA) and PCR was performed on an Applied Biosystems 7300 thermocycler with commercially available Taqman reagents (Assay on Demand, Applied Biosystems) for ccbp2 (ACKR2) (Mm00445551_m1), ccl2 (Mm00441242_m1), ccl5 (Mm01302428_m1), ccr2 (Mm04207877_m1), and ccr5 (Mm01216171_m1). Amplification was performed in duplicate using 40 cycles of denaturation at 95°C for 15 sec and primer annealing/extension at 60°C for 1 min. Expression data were normalized to 18s ribosomal RNA (Hs99999901-sl) or GAPDH RNA measured on the same samples. Relative expression ratio was calculated according to the
Immunohistochemistry with anti-human ACKR2 antibody was used for detection of ACKR2 expression in human kidneys: renal tissue biopsies (
Data are expressed as means ± SE. Comparisons between two groups were performed by
Enzymatic assays, necessary for accurate measurement of blood glucose in OVE diabetic mice [
Blood glucose and ACKR2 RNA in diabetic and normal mice, with and without deletion of the ACKR2 gene. (a) ACKR2 knockout did not affect blood glucose levels in free fed normal or diabetic mice.
Albuminuria was assessed by measuring albumin/creatinine ratio (ACR expressed as
Diabetic albuminuria was reduced by knockout of the ACKR2 gene at 2, 4, and 6 months of age. Urine albumin and creatinine were determined as described in Methods.
We evaluated the glomerular and tubular damage in OVE and OVE-ACKR2 mice at the age of 6 months as previously described [
Renal fibrosis is reduced by knockout of the ACKR2 gene in diabetic OVE-ACKR2 mice. (a) Representative images of renal fibrosis illustrated by trichrome staining in a kidney section for each genotype. Original magnification 200x. (b) Scoring of renal fibrosis by blind counting of blue stained fibrotic regions in trichrome stained kidney sections.
Infiltration of leukocytes in kidney was determined by staining with anti-CD45 antibody (Figure
Knockout of the ACKR2 gene reduces leukocyte infiltration in diabetic mice. (a) Representative images of CD45 staining, original magnification 200x. (b) Quantitative analysis of leukocyte infiltration scored as CD45 positive pixel area per visual field. Twenty-four random fields from 3 mice per group were measured.
Inflammatory chemokines CCL2 and CCL5 (ligands for ACKR2) are elevated in DN [
Kidney RNA levels of ACKR2 ligands CCL2 and CCL5 and their receptors. Values were determined by RT-PCR with Taqman probes using 18S as standard. Columns are mean + SE.
The global changes in gene expression profiles were evaluated by microarray. To confirm the reliability of the microarray results correlation coefficients were calculated between RT-PCR and microarray results for CCL2, CCL5, CCR2, and CCR5 based on the 12 samples used in both assays. For all but CCR5 the correlation was at least 0.96 (
Only 18 of 30,000 genes differed at the 0.05 level between the nondiabetic groups, FVB and ACKR2. Therefore, RNA expression of the OVE and OVE-ACKR2−/− diabetic groups was compared to one nondiabetic group, FVB. Using a minimal criterion of 1.5-fold change in expression and a
Ingenuity pathways in kidney affected by OVE diabetes and/or OVE-ACKR2 diabetes.
Ingenuity canonical pathway | OVE versus FVB | OVE-ACKR2 versus FVB | ||
---|---|---|---|---|
|
Ratio |
|
Ratio |
|
|
||||
Hepatic fibrosis |
|
26/147 | 0.003 | 6/147 |
Atherosclerosis signaling |
|
23/129 | NS | 2/129 |
Altered T cell and B cell signaling in rheumatoid arthritis |
|
17/92 | NS | 1/92 |
Graft-versus-host disease signaling |
|
10/50 | NS | 1/50 |
Glioma invasiveness signaling |
|
12/60 | NS | 1/60 |
|
||||
Communication between innate and adaptive immune cells |
|
18/109 | NS | 1/109 |
Dendritic cell maturation |
|
23/185 | NS | 0 |
Altered T cell and B cell signaling in rheumatoid arthritis |
|
17/92 | NS | 1/92 |
Pattern recognition receptors of bacteria and viruses |
|
19/106 | NS | 2/106 |
Leukocyte extravasation signaling |
|
27/199 | NS | 2/199 |
|
||||
Complement system |
|
13/35 | NS | 1/35 |
B cell development |
|
8/36 | NS | 0 |
NF- |
|
19/175 | NS | 0 |
p38 MAPK signaling |
|
14/106 | NS | 0 |
Antigen presentation pathway |
|
7/40 | NS | 0 |
|
||||
p38 MAPK signaling |
|
14/106 | NS | 0 |
Role of NFAT in regulation of the immune response |
|
16/198 | NS | 0 |
Nitrogen metabolism |
|
6/120 | NS | 1/120 |
Histidine metabolism |
|
7/112 | 0.00012 | 5/112 |
Arginine and proline metabolism |
|
8/176 | 0.00676 | 4/176 |
|
||||
Intrinsic prothrombin activation pathway |
|
8/32 | NS | 1/32 |
Coagulation system |
|
9/38 | NS | 0 |
Extrinsic prothrombin activation pathway |
|
6/20 | NS | 0 |
p38 MAPK signaling |
|
14/106 | NS | 0 |
HMGB1 signaling |
|
11/100 | NS | 0 |
|
||||
Dendritic cell maturation |
|
23/185 | NS | 0 |
Acute phase response signaling |
|
25/177 | 0.00813 | 6/177 |
TREM1 signaling |
|
10/66 | NS | 0 |
IL-8 signaling |
|
20/193 | NS | 2/193 |
NF- |
|
19/175 | NS | 0 |
|
||||
Dendritic cell maturation |
|
23/185 | NS | 0 |
Pattern recognition receptors of bacteria and viruses |
|
19/106 | NS | 2/106 |
Virus entry via endocytic pathways | 0.00014 | 13/100 | NS | 2/100 |
Clathrin-mediated endocytosis signaling | 0.00019 | 20/195 | NS | 2/195 |
Caveolar-mediated endocytosis signaling | 0.0015 | 10/85 | NS | 1/85 |
|
||||
LXR/RXR activation |
|
28/136 | 0.0002 | 7/136 |
TR/RXR activation | 0.0017 | 11/96 | NS | 2/96 |
Aryl hydrocarbon receptor signaling | 0.0028 | 14/159 | NS | 6/159 |
Nitrogen metabolism | 0.0037 | 6/120 | NS | 1/120 |
LPS/IL-1 mediated inhibition of RXR function | 0.0039 | 18/235 | NS | 9/235 |
In OVE kidney, many protective pathways such as immune response and cytokine signaling were activated, as indicated by the high number of RNAs with significantly altered expression. The same pathways in OVE-ACKR2 contained only a few RNAs with altered expression. With few exceptions, most of the biological pathways in Table
The effect of diabetes on kidney ACKR2 protein expression was evaluated in human DN and nondiabetic samples using a rat anti-human ACKR2 monoclonal antibody, previously evaluated on human samples [
Increased ACKR2 protein in diabetic human kidney sections stained with rat monoclonal antibody to human ACKR2. (a) Positive ACKR2 staining in diabetic kidney. Strongest staining in tubules (arrows) especially in a collapsed (arrow) tubule. (b) Minimal staining is seen on a serial section without primary antibody. The arrows indicate the same 2 tubules in images (a) and (b). (c) Sparse ACKR2 staining in a nondiabetic section. (d) At higher magnification granule-like deposits of ACKR2 can be seen in cytoplasm of proximal tubular epithelial cells in diabetic kidney. In the interstitial space ACKR2 staining is also visible in diabetic kidney monocytes (e) and endothelial cells (f). (g) Semiquantitative scoring of ACKR2 staining by a scorer blind to sample identity. Scores for proximal tubule and interstitial cells are higher in diabetic than nondiabetic samples.
This study demonstrates that the ACKR2 chemokine scavenger receptor has an unexpected important role in the development of diabetic kidney disease. Deletion of the ACKR2 gene in OVE diabetic mice produced a great reduction in albuminuria, accompanied by reduced severity of renal fibrosis, leucocyte infiltration, and inflammatory chemokine gene expression. In addition, ACKR2 protein content was elevated in several cell types in kidneys of DN patients.
Chemokines and cytokines regulate the inflammatory processes and contribute to progressive kidney damage in diabetes [
To determine if the ACKR2 RNA results indicate that diabetes alters ACKR2 protein immunohistochemistry studies were performed on human tissue since only an anti-human ACKR2 antibody has been validated [
The primary finding of this study was that ACKR2 deletion dramatically reduced DN. The reduction of albuminuria in OVE-ACKR2 mice was significant at the earliest age tested, two months. As OVE mice aged DN progressed and the protection by ACKR2 KO became more striking. At 6 months ACKR2 deletion produced a greater reduction in diabetic albuminuria. In addition several markers demonstrated reduced inflammation in OVE-ACKR2 kidneys compared to OVE kidneys. Histologically this was indicated by decreased leukocyte infiltration and less fibrosis. Gene expression data demonstrated that absence of ACKR2 prevented activation of multiple molecular pathways involved in immune or inflammatory processes in kidneys of diabetic mice. The finding of such potent renal protection from diabetes by deletion of ACKR2 was contradictory to our expectation, which was that deletion of ACKR2 would exacerbate DN by increasing renal inflammation. This expectation was based on the damage inflammation produces in DN and the anti-inflammatory potency of ACKR2 as a scavenger of proinflammatory chemokines. In several studies manipulation of ACKR2 levels modified tissue inflammation in a manner that would be predicted based on anti-inflammatory potency of ACKR2 as a chemokine scavenger: this was shown in experimental models of colitis and psoriasis, where deletion of ACKR2 increased colon [
Mechanisms to explain protection from DN by deletion of ACKR2 are not obvious. Protection was not due to reduced OVE diabetes since hyperglycemia was equivalent in OVE and OVE-ACKR2 mice (Figure
In
The authors declare that there is no conflict of interests regarding the publication of this paper.
The authors thank Yun Huang for mice mating and genotyping, Patricia Kralik for albuminuria assay, and Sabina J Waigel for gene array data discussion. This work was supported by Juvenile Diabetes Research Foundation Grants 1-INO-2014-116-A-N and 1-2011-588.