 |
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
*Department of Pathology, Division of Nephrology, University of Washington, Seattle Washington; and Department of Vascular Biology, The Hope Heart Institute, Seattle, Washington.
Correspondence to Dr. Charles E. Alpers, University of Washington Medical Center, Department of Pathology, Box 356100, 1959 NE Pacific St., Seattle, WA 98195. Phone: 206-543-5616; Fax: 206-543-3644;
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 (TGF-1) and type I collagen in mesangial cells. This study is an investigation of the role of SPARC in streptozotocin (STZ)–induced diabetic nephropathy (DN) of 6-mo duration in wild type (WT) and SPARC-null mice. SPARC expression was evaluated by immunohistochemistry (IHC) and by in situ hybridization (ISH). Deposition of type I and IV collagen and laminin was evaluated by IHC, and TGF- 1 mRNA was assessed by ISH. Renal function studies revealed no significant difference in BUN between diabetic SPARC-null mice and diabetic WT mice, whereas a significant increase in albumin excretion was detected in diabetic WT relative to diabetic SPARC-null mice. Diabetic WT animals exhibited increased levels of SPARC mRNA and protein in glomerular epithelial cells and in interstitial cells, in comparison with nondiabetic WT mice. Neither SPARC mRNA nor protein was detected in SPARC-null mice. Morphometry revealed a significant increase in the percentage of the glomerular tufts occupied by ECM in diabetic WT compared with nondiabetic WT mice, although there was no difference in the mean glomerular tuft area among groups. In contrast, diabetic SPARC-null mice did not show a significant difference in the percentage of the glomerular tufts occupied by ECM relative to nondiabetic null mice. Tubulointerstitial fibrosis was ameliorated in diabetic SPARC-null mice compared with diabetic WT animals. Further characterization of diabetic SPARC-null mice revealed diminished glomerular deposition of type IV collagen and laminin, and diminished interstitial deposition of type I and type IV collagen correlated with decreases in TGF- 1 mRNA compared with WT diabetic mice. These observations suggest that SPARC contributes to glomerulosclerosis and tubulointerstitial damage in response to hyperglycemia through increasing TGF- 1 expression in this model of chronic DN. E-mail: calp@u.washington.edu |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 (TGF-1) in renal cells (5,6 ). In vivo, SPARC is expressed during development (7) and is detected at sites of wound repair (8) and tissue remodeling (9). SPARC has also been implicated in the progression of renal injuries. In vivo, increased production of SPARC mRNA is observed in visceral glomerular epithelial cells and in interstitial cells in the course of passive Heymann nephritis (10) and in mesangial cells in the course of the anti-Thy-1 mesangial proliferative glomerulonephritis (11). In vitro, it has been demonstrated that SPARC may be involved in glomerular diseases by inhibiting mesangial cell proliferation and also by promoting production of ECM through regulation of TGF-1 expression and type I collagen production by mesangial cells (5,11 ).
Diabetic nephropathy (DN) is one of the major complications and causes of death in diabetes (12). Increased thickness of glomerular basement membrane and augmentation of ECM are recognized as pathologic hallmarks of diabetes. The latter particularly contributes to the reduction of filtration surface area, which becomes important in advanced DN. Increased synthesis of ECM components is associated with the accumulation of mesangial matrix in DN, which is, at least partially, a consequence of the activity of certain growth factors. Numerous growth factors have been implicated in the development of DN such as TGF-, a potent stimulus for the production of ECM through its regulation of the synthesis of ECM components, e.g., collagen type I (13,14 ).
A study in humans demonstrated that significantly elevated serum concentrations of SPARC could be detected in diabetic patients with fibrosing renal injury as revealed by renal biopsy (15). In an experimental model of rat DN induced by streptozotocin (STZ), a reduction of SPARC is observed in the early phase of injury, which coincides with the onset of diabetes-related kidney growth (16). The role of SPARC in DN, especially in the chronic phase of the disease, has not been extensively investigated and remains obscure. Mice with a homozygous deletion of the Sparc gene allow us to examine the role of SPARC in the pathogenesis of various diseases (17). Therefore, we conducted this study in SPARC-null mice to test the hypothesis, suggested by the study in humans cited above (15), that the absence of SPARC would ameliorate chronic DN.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Disease Model
Seven to ten-week-old male mice weighing 19 to 35 g at the onset of the experimental protocol were randomly divided into two groups: (1) a group in which mice were treated with STZ (Sigma Chemical Co., St. Louis, MO) and (2) a group in which mice were treated with citrate buffer adjusted to pH 4.5 (Sterile Water for Irrigation; Baxter Healthcare Corp., Deerfield, IL). STZ was dissolved in sterile citrate buffer and injected intraperitoneally into mice (120 mg/kg, up to 0. 4 ml) within 10 min of preparation. STZ or citrate buffer was administered at three time points occurring at 48-h intervals during the first week of this study.
Serum glucose levels and body weight were examined every 2 wk during the first 2 mo after injections and subsequently every 4 wk until sacrifice. Blood was collected from the orbital sinus.
Urinary glucose level was monitored semiquantitatively with Uristix reagent strips (Bayer Corporation, Elkhart, IN) every week during the first 2 mo according to the manufacturer’s instructions. Grading for urinary glucose was as follows; 0 (negative), 1+ (>100 mg/dl), 2+ (>250 mg/dl), 3+ (>500 mg/dl), 4+ (>1000 mg/dl), or 5+ (>2000 mg/dl). To render animals hyperglycemic without becoming ketoacidotic, when the urinary glucose level of diabetic mice was more than 5+, we administered long-acting insulin (Humulin U, Ultralente, human insulin HI-610; Eli Lilly and Co., Indianapolis, IN) subcutaneously at a dose of 1 unit three times every 2 d. Serum glucose levels after insulin treatment of the mice remained above 200 mg/dl.
The genotype of each animal was confirmed by PCR analysis on purified spleen DNA that was obtained at sacrifice.
Experimental Design
Twenty mice treated with STZ (SPARC-null mice, n = 13; WT mice, n = 7) and 13 mice treated with citrate buffer alone (SPARC-null mice, n = 7; WT mice, n = 6) were sacrificed 6 mo after these injections. Six-hour urine samples were collected in each mouse one day before sacrifice. A blood sample was obtained via cardiac puncture at sacrifice. Kidneys were removed and rapidly processed for light microscopy, immunohistochemistry, and in situ hybridization using conventional fixatives and techniques. Spleens were taken for genotyping by PCR analysis and snap-frozen in isopentane.
Renal tissues from these mice were evaluated by immunohistochemistry for SPARC, 
-smooth muscle actin (SMA, a marker of activated mesangial cells or myofibroblastic interstitial cells), Ki-67 (a marker of proliferating cells), Mac-2 (a marker of monocytes/macrophages), type I collagen (component of ECM) and type IV collagen, and laminin (components of the basement membrane). Expression of SPARC mRNA and TGF- 1 mRNA was evaluated by in situ hybridization.
Preparation of DNA and Genotyping
The Dneasy Tissue Kit (Qiagen Inc.) was used for DNA isolation from the spleens (up to 10 mg) according to the protocol of the manufacturer.
Mice were genotyped using three PCR primers as described previously (Life Technologies BRL, Rockville, MD) (18): 5'GATGAGGGTGGTCTGGCCCAGCCCTAGATGCCCCTCAC3', from the 5' end of the coding region of the SPARC gene; 5'CACCCACACAGCTGGGGGTGATCCAGATAAGCCAAG3', from the 3' end of the coding region, and 5'GTTGTGCCCAGTCATAGCCGAATAGCCTCTCCACCCAAG3', which was annealed with a sequence within the neomycin insertion region. A DNA product of 300 bp was demonstrated in WT animals, and one of 550 bp was demonstrated in SPARC-null mice. The DNA product isolated from heterozygotes was characterized by the presence of both bands. The PCR reaction contained DNA, 10 x PCR buffer, 50 mM MgCl2, 12.5 µM of each primer, 10 mM dNTP (Life Technologies BRL), and 2.5 units of Taq DNA polymerase (MBI Fermentas, Hanover, MD) in a 50-µl PCR reaction. Cycling conditions were: 96°C for 1 min, followed by 40 cycles at 96°C for 30 s, 60°C for 45 s, 68°C for 4 min, and 74°C for 10 min.
Renal Morphology
Renal tissue was divided and fixed in either 10% neutral-buffered formalin or in methyl Carnoy solution, as described previously (19). Fixed tissues were processed, embedded in paraffin according to standard protocols, and sectioned at 4 µm. Slides were stained with periodic acid methenamine silver (silver), with periodic acid-Schiff (PAS), and with hematoxylin and eosin (H&E) using standard histologic procedures. Semiquantitative scoring of glomerular sclerosis in silver-stained tissue utilized five grades as described previously (20): 0 = normal glomerulus, 1+ = sclerosis involving less than 25% of glomerular surface area, 2+ = sclerosis involving 25 to 50%, 3+ = 50 to 75%, and 4+ = sclerosis involving 75% to 100% of glomerular surface area. Tubulointerstitial damage was also graded in the areas where 100 glomeruli were counted as described previously (21): 0 = no lesions showing cell infiltration and fibrosis, 1+ = minimal injury (single focus of lesion), 2+ = mild injury (more than two isolated foci), 3+ = moderate injury (more than five isolated foci), and 4+ = severe injury (more than ten isolated foci or diffuse infiltration and fibrosis).
Immunohistochemistry
Four-micrometer sections of formalin- and methyl Carnoy-fixed, paraffin-embedded tissue were processed by an indirect immunoperoxidase technique as described previously (19,22 ). The antibodies used in this study are listed in Table 1. Rabbit anti-mouse SPARC IgG from antiserum 5944 was affinity-purified as described previously (23). This polyclonal preparation immunoprecipitated SPARC from the conditioned media of murine glomerular mesangial cells, but did not react with any protein from the conditioned media of mesangial cells from SPARC-null mice (18). The specificity of the other antibodies has been established as referenced. For all samples, negative controls for the immunohistochemistry included substitution of the primary antibody with an irrelevant IgG from the same species or with phosphate-buffered saline (PBS).
|
SMA, types I and IV collagen, and laminin was performed using five grades as follows: 0 = no staining; 1+ = mesangial staining involving less than 25% of the area examined; 2+ = segmental mesangial staining involving 25 to 50% of mesangial areas present; 3+ = mesangial staining involving 50 to 75% of the areas examined; 4+ = diffuse mesangial staining involving more than 75% of areas examined, as described previously (19). The number of Mac-2-expressing cells and of Ki-67-positive cells in the glomerular tufts was identified by immunostaining and counted at a magnification of x400. Both the number of the intraglomerular positive cells and the number of glomeruli were counted, and the data are presented as stained cells per glomerular cross-section (gcs). Cells outside the glomerular tufts were excluded from these counts. At least 20 consecutive cross-sections of glomeruli (counting only those glomeruli containing more than 20 discrete capillary segments) were examined with each of the above antibodies. The degree of tubulointerstitial changes was graded on a scale from 0 to 4 as follows: 0 = normal kidney; 1+ = minimal abnormalities affecting less than 25% of the renal parenchyma; 2+ = mild, 25 to 50% involvement; and 3+ = moderate, 50 to 75% involvement, 4+ = severe, more than 75% of the kidney manifesting abnormalities.
Urine and Blood Samples
Serum glucose levels and semiquantitative urinary glucose levels at sacrifice were examined as mentioned above using the Genuine One Touch and One Touch Basic (Lifescan Inc., Milpitas, CA) and with Uristix reagent stripes (Bayer Corporation), respectively. Blood urea nitrogen (BUN) was determined by a commercial colormetric assay (Sigma). Urine samples were evaluated for proteinuria using the albumin/creatinine ratio. Albuminuria was evaluated using the Albuwell (Exocell, Inc., Philadelphia, PA) mouse albumin ELISA, and creatinine was measured using The Creatinine Companion (Exocell) according to the protocols of the manufacturer.
Molecular Probes
The molecular probes used in this study have previously been described (24,25 ):
1: a 974-kb (nt 421 to 1395) fragment of the mouse cDNA was inserted into the SmaI site of pGem7Zf+ (Promega), and was a gift from H.L. Moses, Dept. Cell Biology, Vanderbilt University, Nashville, TN (26). Both antisense and sense riboprobes were transcribed using reagents from Promega, except [35S]-UTP, which was obtained from New England Nuclear (Boston, MA). The details of this procedure have been published (27).
In Situ Hybridization
The in situ hybridization protocols have been described in detail (27). Negative controls included hybridization performed on replicate tissue sections using the sense riboprobe.
The extent of intraglomerular hybridization signal was graded semiquantitatively by an observer who was blinded to the origin of the histologic specimen: 0 = absent, 1 = mild (a few positive cells), 2 = moderate (several positive cells in a segmental distribution within the glomerulus), 3 = severe (many positive cells in a diffuse distribution).
Morphometric Analyses
Morphometric analysis was performed on H&E-stained and silver-stained histologic sections as described previously (22). In a blinded fashion, 15 consecutive glomerular cross-sections were photographed with a digital camera (Olympus DP11, Olympus America Inc., Melville, NY) and imported into Image-Pro Plus (Media Cybernetics, Silver Spring, MD). The following parameters were measured for each glomerular cross-section: (1) The number of glomerular nuclei and the glomerular tuft area (independent of the urinary space) on H&E-stained tissue and (2) the area of mesangial matrix and the glomerular tuft area on silver-stained tissue.
Statistical Analyses
All values are expressed as the mean ± SEM. Using the SPSS program, version 10.0 for Windows (SPSS Inc. Chicago, IL), we determined the survival rate by Kaplan-Meier test. Urinary glucose was compared by a nonparametric Mann-Whitney U test. Other data were examined with one-way ANOVA following Tukey post-hoc test. A P < 0.05 was considered to be statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Metabolic Characteristics and Renal Function
Metabolic parameters, body weight, and serum glucose level are presented in Table 2. The body weight of diabetic groups reached a nadir at 2 wk and gradually increased thereafter. Nondiabetic groups had a significant body weight gain compared with diabetic groups after 12 wk and at the end of the study period.
|
Renal function as measured by BUN revealed that diabetic groups of both genotypes showed a significant increase in BUN compared to nondiabetic groups (Table 2). However, no significant difference in BUN was detected between diabetic groups. Diabetic groups of both genotypes exhibited increases in albumin excretion, with an increased albumin/creatinine ratio compared with nondiabetic groups (Table 2). The albumin excretion of diabetic WT mice was significantly higher than that of diabetic SPARC-null mice. Diabetic groups of both genotypes had significant increases in both urinary glucose and urine volume, but there were no significant differences between the genotypes for these parameters (Table 2).
SPARC-Null Mice Show Reduced Accumulation of ECM
No segmental or global glomerulosclerosis was observed in renal tissue specimens obtained from nondiabetic WT or SPARC-null mice (Figure 1, B and E). Diabetic WT mice manifested extensive mesangial matrix expansion (Figure 1H), whereas no apparent increase of mesangial matrix was observed in diabetic SPARC-null mice (Figure 1K). No significant increase in glomerular cell number was noted in diabetic mice compared with nondiabetic mice (Figure 1, A, D, G, and J). The results of the semiquantitative analysis are summarized in Table 3. There was no significant difference in the scoring of glomerulosclerosis in silver-stained tissue between diabetic and nondiabetic SPARC-null mice by semiquantitative grading methods (see Materials and Methods). In contrast, diabetic WT mice showed a significant increase in the scoring of glomerulosclerosis compared with nondiabetic WT mice (diabetic 1.12 ± 0. 12 versus nondiabetic 0.26 ± 0. 02; P < 0.01), which was also significantly elevated compared with diabetic SPARC-null mice (WT 1.12 ± 0. 12 versus SPARC-null 0. 53 ± 0. 05; P < 0.01). In addition to the semiquantitative analysis, a quantitative analysis using morphometry was performed for the glomerular tuft area, glomerular matrix area, and glomerular cellularity (Figure 2). By this approach, there was no difference in the size of glomerular tuft among the four groups (Figure 2A). There were also no significant differences in mean cell number per glomerular tuft area among the four groups (Figure 2, B and C). In silver-stained histologic sections, diabetic WT mice demonstrated a significant increase in the area occupied by matrix per glomerulus, in comparison with nondiabetic WT mice (Figure 2D). In contrast, diabetic SPARC-null mice did not show a significant difference compared with nondiabetic null mice (Figure 2D). The same was true for the percentage of matrix per tuft area (Figure 2E).
|
|
|
Reduced Glomerular Accumulation of ECM Is Characterized by Type IV Collagen and Laminin Deposition in SPARC-Null Mice
To assess the specific composition of the ECM, we performed immunohistochemical staining for type I collagen, type IV collagen, and laminin. Nondiabetic mice of both genotypes demonstrated immunostaining for type IV collagen in the glomerular and tubular basement membranes. Glomerular staining for type IV collagen in nondiabetic mice was focal in distribution (Figure 3, B and E). A similar staining pattern for laminin was observed in nondiabetic kidneys of both genotypes (Figure 3, A and D). Diabetic WT mice demonstrated a significant increase in immunostaining for type IV collagen predominantly localized to the mesangium (Figure 3H; Table 4). Diabetic WT mice showed increased accumulation of laminin in both mesangium and glomerular capillary walls, in comparison with nondiabetic WT mice (Figure 3G; Table 4). In contrast, immunostaining for type IV collagen and laminin in diabetic SPARC-null mice compared with nondiabetic null mice was less (Figure 3, J and K; Table 4). In diabetic WT mice, deposition of type IV collagen and, to a lesser extent, of laminin was increased at sites of tubulointerstitial fibrosis compared with diabetic null mice (Figure 3, G, H, J, and K). Glomerular staining for type I collagen was consistently negative in both SPARC-null and WT mice (Figure 3, C, F, I, and L), but increased interstitial staining of this collagen was noted in both diabetic mice and was significantly enhanced in diabetic WT compared with diabetic null animals (Figure 3, I and L; Table 4).
|
|
|
Slight Elevation of SMA Is Apparent in Diabetic WT and SPARC-Null Mice
We asked whether a phenotypic change of mesangial cells or interstitial cells occurs in this model of experimental diabetes. Accordingly, tissue was stained for the mesangial activation marker SMA. All renal specimens obtained from SPARC-null and WT mice demonstrated intense staining for SMA by the smooth muscle cells of muscular arteries. A few glomeruli in nondiabetic mice exhibited mesangial SMA, usually in a segmental distribution within the glomerular tuft. No difference in SMA expression was noted between diabetic and nondiabetic SPARC-null mice, whereas diabetic WT mice showed a slight elevation of SMA in comparison with nondiabetic WT mice and diabetic SPARC-null mice (Table 4). Slight increases in SMA staining were also observed in the interstitial cells obtained from the diabetic groups, but there were no apparent differences between genotypes (Table 4). SMA was not expressed by tubular epithelium.
The mean number of proliferating cells in glomeruli as determined by Ki-67, and the mean number of monocytes/macrophages in glomeruli labeled with Mac-2, showed no significant differences between groups (Table 4).
Decreased Expression of TGF- 1 mRNA in SPARC-Null Mice
Weak expression of TGF-1 mRNA was detected in a pattern consistent with mesangial cell expression in nondiabetic mice of both genotypes, without clear differences (Figure 5, A and B). Diabetic WT mice showed a significant increase in glomerular expression of TGF- 1 mRNA, relative to nondiabetic WT mice (nondiabetic 1.3 ± 0.3 versus diabetic 2.5 ± 0.2; P < 0.01; Figure 5, A and C). There was no significant difference in glomerular TGF-1 mRNA between diabetic SPARC-null mice and nondiabetic null mice, although there was a trend toward increased production in the former groups (nondiabetic 1.4 ± 0.3 versus diabetic 1.9 ± 0.2; P = 0.09; Figure 5, B and D). The intensity of TGF-1 mRNA was substantially decreased in diabetic SPARC-null compared with diabetic WT mice, but the difference between these two groups did not reach significance (WT 2.5 ± 0.2 versus null 1.9 ± 0.2; P = 0.054).
|
1 mRNA was barely detectable in the interstitium of nondiabetic mice (Figure 5, A and B), whereas in diabetic mice, especially WT, TGF-1 mRNA was markedly augmented in areas of tubulointerstitial injury (Figure 5, C and D). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 by mesangial cells, leading to production of ECM components (5), its localization in myofibroblast-like interstitial cells in humans in settings of interstitial fibrosis (24), a report of increased circulating levels of SPARC in diabetic patients with fibrosing renal injury (15), and evidence that SPARC participates in early events of diabetic nephropathy as revealed in STZ-induced diabetes in the rat (16). This study used genetically modified mice deficient in SPARC as one approach to test the importance of SPARC in mediating a well-characterized model of experimental DN. Two principal findings emerged from this study: (1) in the absence of SPARC, the severity of DN is markedly diminished; (2) the absence of SPARC was correlated with the suppression of the normally increased production of TGF-1 that occurs in this model, indicating one mechanism for the diminished accumulation of ECM that is observed in diabetic SPARC-null mice. In this study, increased production of SPARC mRNA and protein was detected in diabetic WT mice compared with nondiabetic WT mice. In both diabetic and nondiabetic WT mice, SPARC expression in glomeruli was detected only in podocytes. This pattern of expression by podocytes is consistent with previous reports in rats (10,28,29 ) and humans (24), although it has been reported under some disease conditions that mesangial cells produce SPARC as well (11,28 ). Mesangial production of SPARC was not detected in this study. Increased expression of SPARC in diabetic WT mice compared with nondiabetic WT mice was also detected in interstitial cells in areas of interstitial injury. This result is also consistent with data on renal interstitial fibrosis in rats (29) and humans (30).
The WT diabetic animals developed the characteristic features of STZ-induced DN as previously established: mesangial matrix expansion in the glomeruli, tubulointerstitial injury, and an absence of vascular pathology (31). The reduced severity of DN in SPARC-null compared with WT mice was defined by the following features: (1) reduction of mesangial matrix expansion with diminished accumulation of type IV collagen and laminin and (2) decreased interstitial injury with diminished deposition of type IV and type I collagen in the tubulointerstitium. A functional correlate to this morphologic improvement was a significant decrease in albumin excretion in the diabetic SPARC-null mice as compared with diabetic WT mice. These results are particularly relevant to human DN, which is characterized by increased expression of type IV collagen and laminin in the glomerular mesangial areas and in the peripheral capillary walls (32,33 ).
We speculate the beneficial effect of the SPARC deletion in diabetic nephropathy may be indirect, given that glomerular SPARC expression was identified in podocytes rather than mesangial cells, where the principal injury of experimental DN is localized. We speculate SPARC is active in part through physicochemical interactions with specific growth factors, such as that reported with PDGF B-chain (4), or may act to modify production of matrix regulatory molecules, such as TGF-, which may be produced and released by glomerular epithelial cells in certain disease settings.
It is well established that TGF-1 is a potent stimulator of ECM production in glomerular injury and may be the most important growth factor in determining the extent of renal fibrosis after injury. In the present study, we observed significant augmentation of TGF-1 mRNA production as well as marked accumulation of glomerular ECM in diabetic WT mice compared with nondiabetic WT mice, consistent with previous studies in animal model systems of diabetes (34–36 ) and in human DN (35). A key finding of this study was that diabetic SPARC-null mice demonstrated suppression of the typical increases in TGF-1 mRNA that occur in this model. It has been shown that TGF- mediates the production of type I and type IV collagens in mesangial cells under conditions of high glucose in vitro (37). Exposure of cultured glomerular epithelial cells to TGF-1 and high glucose concentrations also increased the production of the basement membrane components, laminin and type IV collagen (38). It was previously demonstrated that SPARC-null mesangial cells express significantly decreased steady-state levels of TGF-1 mRNA and correspondingly less secreted TGF-1 protein (5). We now demonstrate the relevance of this observation to the in vivo setting, where the absence of SPARC resulted in decreased synthesis of TGF- 1 mRNA by mesangial cells, which presumably led to the reduced accumulation of ECM in glomeruli and basement membrane. These findings indicate SPARC can be an important mediator of sclerosing glomerular injury and acts at least in part by its ability to enhance TGF- 1 expression.
In addition to the glomeruli, TGF-1 expression was substantially increased in diabetic WT mice in areas of tubulointerstitial injury, as compared with diabetic SPARC-null mice. We found that the absence of SPARC had a protective effect on tubulointerstitial injury with diminished production of TGF-1. It has been demonstrated that TGF- stimulates secretion of types I and IV collagen by cultured renal interstitial fibroblasts (39). SPARC has been co-localized to cells with a myofibroblastic-like phenotype (i.e., with co-expression of SMA) in the damaged interstitium in chronic human disease such as chronic transplant rejection (24,30 ). Given the central role of TGF-1 in mediating interstitial fibrosis as established in numerous studies of experimental injury (reviewed in reference 40), and the presence of SPARC in relevant sites of tubulointerstitial injury in disease, these findings indicate that the effect of SPARC on TGF-1 production might be an important pathogenic mechanism in the exacerbation or amelioration of fibrosing renal injury. We recognize that the deficiency of SPARC may exert a beneficial effect in DN by mechanisms not directly involving TGF-1, but pathways by which this might occur are not yet clearly defined.
In conclusion, we have demonstrated that chronic STZ-induced DN of 24-wk duration was less severe in SPARC-null compared with WT mice. This was evidenced by the reduction in albuminuria, mesangial expansion, and tubulointerstitial injury with decreased production of ECM components, the latter presumably a consequence of reduced production of TGF-1. This study indicates that SPARC can be an important modulator of glomerulosclerosis and interstitial damage in response to hyperglycemia and might be an attractive target for therapeutic intervention in human diabetic nephropathy.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Kamijo-Ikemori, T. Sugaya, A. Sekizuka, K. Hirata, and K. Kimura Amelioration of diabetic tubulointerstitial damage in liver-type fatty acid binding protein transgenic mice Nephrol. Dial. Transplant., October 14, 2008; (2008) gfn573v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Zhang, Y. Zhang, G. Ning, D. K. Deb, J. Kong, and Y. C. Li Combination therapy with AT1 blocker and vitamin D analog markedly ameliorates diabetic nephropathy: Blockade of compensatory renin increase PNAS, October 14, 2008; 105(41): 15896 - 15901. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Daniel, K. Schaub, K. Amann, J. Lawler, and C. Hugo Thrombospondin-1 Is an Endogenous Activator of TGF-{beta} in Experimental Diabetic Nephropathy In Vivo Diabetes, December 1, 2007; 56(12): 2982 - 2989. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Socha, M. Manhiani, N. Said, J. D. Imig, and K. Motamed Secreted Protein Acidic and Rich in Cysteine Deficiency Ameliorates Renal Inflammation and Fibrosis in Angiotensin Hypertension Am. J. Pathol., October 1, 2007; 171(4): 1104 - 1112. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Rentz, F. Poobalarahi, P. Bornstein, E. H. Sage, and A. D. Bradshaw SPARC Regulates Processing of Procollagen I and Collagen Fibrillogenesis in Dermal Fibroblasts J. Biol. Chem., July 27, 2007; 282(30): 22062 - 22071. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. V. Durvasula and S. J. Shankland Mechanical strain increases SPARC levels in podocytes: implications for glomerulosclerosis Am J Physiol Renal Physiol, September 1, 2005; 289(3): F577 - F584. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.-L. Zhao, P. C.Y. Tong, F. M.M. Lai, B. Tomlinson, and J. C.N. Chan Association of Glomerulopathy With the 5'-End Polymorphism of the Aldose Reductase Gene and Renal Insufficiency in Type 2 Diabetic Patients Diabetes, November 1, 2004; 53(11): 2984 - 2991. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Sadlier, S. B. Connolly, N. E. Kieran, S. Roxburgh, D. P. Brazil, L. Kairaitis, Y. Wang, D. C. H. Harris, P. Doran, and H. R. Brady Sequential Extracellular Matrix-focused and Baited-global Cluster Analysis of Serial Transcriptomic Profiles Identifies Candidate Modulators of Renal Tubulointerstitial Fibrosis in Murine Adriamycin-induced Nephropathy J. Biol. Chem., July 9, 2004; 279(28): 29670 - 29680. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. L. Hudkins, D. G. Gilbertson, M. Carling, S. Taneda, S. D. Hughes, M. S. Holdren, T. E. Palmer, S. Topouzis, A. C. Haran, A. L. Feldhaus, et al. Exogenous PDGF-D Is a Potent Mesangial Cell Mitogen and Causes a Severe Mesangial Proliferative Glomerulopathy J. Am. Soc. Nephrol., February 1, 2004; 15(2): 286 - 298. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Taneda, K. L. Hudkins, S. Topouzis, D. G. Gilbertson, V. Ophascharoensuk, L. Truong, R. J. Johnson, and C. E. Alpers Obstructive Uropathy in Mice and Humans: Potential Role for PDGF-D in the Progression of Tubulointerstitial Injury J. Am. Soc. Nephrol., October 1, 2003; 14(10): 2544 - 2555. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||
