 |
||||
| 2008 JASN IMPACT FACTOR 7.505 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2008 JASN IMPACT FACTOR 7.505 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
*
Institute of Internal Medicine, Division of Nephrology, University of
Padova, Padova, Italy
§
Department of Biochemistry, University of Padova, Padova, Italy
Department of Internal Medicine, Division of Endocrinology, Metabolism,
and Pathobiochemistry, University of Tübingen,
Tübingen, Germany
Institute of Pathology, University of Munich, Munich, Germany.
Correspondence to Dr. Erwin D. Schleicher, Department of Internal Medicine, Division of Endocrinology, Metabolism, and Pathobiochemistry, Eberhard-Karls-Universität, Otfried-Müller-Straße 10, D-72076, Tübingen, Germany. Phone: 49-7071-29-87599; Fax: 49-7071-29-5974; E-mail: enschlei{at}med.uni-tuebingen.de
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(PKC-) and -ß1 and activation of PKC-, as well as
high glucose-induced activation of PKC-. These results suggest that GAG
inhibit TGF-ß1 overexpression at the transcriptional level, possibly via
inhibition of high glucose-activated PKC. The findings indicate the potential
of GAG therapy for the prevention of diabetic glomerulosclerosis by the
inhibition of chronic disease-induced TGF-ß1 mRNA overexpression. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recent reports have indicated the involvement of cytokines and growth factors in the development of diabetic nephropathy and other nephropathies (4,5,6,7). In particular, transforming growth factor ß (TGF-ß) has been implicated in their pathogenesis (5,6,7,8). Latent TGF-ß is the proform of the multifunctional cytokine TGF-ß. Cells secrete latent TGF-ß in a small or large form. The latter, containing a latent TGF-ß-binding protein (LTBP), is secreted by glomerular cells (9). Among many diverse effects, TGF-ß promotes the accumulation of extracellular matrix by increasing the synthesis of extracellular matrix components and reducing matrix degradation. Therefore, chronically enhanced induction of TGF-ß, in response to injury or disease, is suggested to be the major mediator of progressive fibrosis in experimental animal and human diseases associated with sclerosis, such as diabetic nephropathy (10). The beneficial effects of inhibiting TGF-ß action by injecting neutralizing antibodies have been demonstrated in various renal and non-renal disease models and injuries, including diabetic nephropathy (11,12). Recent data convincingly demonstrated that inhibition of TGF-ß activity by decorin, a proteoglycan that binds and neutralizes TGF-ß via its core protein, resulted in the suppression of matrix accumulation in nephritic glomeruli (14,13,14). Similarly, administration of recombinant latency-associated peptide, an endogenous TGF-ß-binding protein, reversed hepatic TGF-ß1-induced effects (15). Furthermore, the use of antisense oligonucleotides against TGF-ß1 mRNA blocked fibrosis in a mesangioproliferative nephritis model (16,17) and in cultured mesangial cells (18). Together, these data indicate that inhibition of TGF-ß1 activity at the protein level or blockage at the translational level attenuates the fibrosis-promoting effects of TGF-ß1.
We previously reported that chronic therapy with glycosaminoglycans (GAG) may prevent or cure experimental diabetic nephropathy (19,20). In this experimental model, daily treatment with low-molecular weight heparin or dermatan sulfate (DS) prevented structural changes in the glomerular basement membrane and changes in the albumin excretion rate, without affecting other parameters, e.g., glycemia or GFR (19,20). Although a number of mechanisms have been proposed to explain the renoprotective effects of GAG, we wondered, in light of the hypothesized key role of TGF-ß in diabetic nephropathy, whether these agents might be renoprotective because of their inhibition of the TGF-ß cascade.
In this report, we demonstrate that chronic treatment with a modified heparin (mH) GAG preparation with low anticoagulant activity prevents the clinical and histologic signs of diabetic nephropathy in a well characterized rat model of diabetic glomerulosclerosis, mimicking most characteristics of human diabetic glomerulosclerosis, and we provide evidence that the disease-induced overexpression of TGF-ß1 is suppressed by GAG/mH therapy. By studying the molecular mechanisms in an established cell culture system, we found that GAG/mH treatment of mesangial cells prevented the high glucose-mediated induction of TGF-ß1 mRNA expression and the subsequent increase in TGF-ß protein levels and bioactivity. Our data suggest the potential value of this treatment for diabetic patients with early signs of nephropathy and reveal that the renoprotective activity of heparin and GAG may be related to an anti-TGF-ß effect.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Study Design and Effect of GAG/mH Therapy on Albuminuria in Diabetic
Rats
This section of the study involved the use of new data, particularly on
TGF-ß, from previous reports
(19,20),
to which readers are referred for more details on the protocol, metabolic
control, renal function, and mortality rate. The study design is schematically
depicted in Figure 1A. Diabetes
mellitus was induced in 6-wk-old male Sprague-Dawley rats with streptozotocin.
One-half of the diabetic rats and one-half of the control rats were treated
subcutaneously with 15 mg/kg body wt per d GAG/mH, and the remaining one-half
of each group were treated subcutaneously with 2 ml/kg per d saline solution,
for 12 mo. None of the animals received insulin during the experiments.
|
Immunohistochemical Staining and Morphometric Analysis
Formaldehyde-fixed and paraffin-embedded tissue sections were
deparaffinized and treated with phosphate-buffered saline containing 0.5%
H2O2. Immunolocalization of collagen III was performed
as described (22).
Immunostaining for LTBP was performed similarly to the procedure used for
1 or 2 (IV) collagen staining
(21), except that tissue
sections were incubated with the rabbit LTBP-specific antiserum (1:80
dilution) at 4°C overnight; the same procedure was used for
immunolocalization of TGF-ß in tissue sections that had been pretreated
with 1 mg/ml bovine hyaluronidase in 0.1 M sodium acetate (pH 5.5) containing
0.85% NaCl and incubated at 37°C for 30 min. Mesangial cells were
identified by Thy-1.1 staining, as described
(21). The avidin-biotin
complexes were observed using 3,3'-diaminobenzidine, and the sections
were counterstained with hematoxylin and evaluated by light microscopy. The
sections from all experimental groups were incubated and developed
simultaneously. The IgG fraction of rabbit serum was used as a negative
control. Periodic acid-Schiff (PAS) and immunohistochemical staining was
evaluated by morphometric analysis, as described
(21). The images of >20
systematically sampled glomeruli in each animal were digitized from the light
microscope, using a video camera and a computer-based image analysis system
(VIDAS System; Kontron, Eching, Germany). Glomerular cross-sections were
displayed on the computer screen at a final magnification of x400, and
the area of the outlined polygon was measured with general-purpose image
analysis software (VIDAS System; Kontron), using an interactive procedure. The
positively stained mean glomerular areas were calculated for each animal.
Preparation of Riboprobes
cDNA fragments for TGF-ß1 and GAPDH were obtained from mouse spleen
and porcine mesangial cell RNA by reverse transcription (RT)-PCR
amplification, according to the previously described protocol
(18). The cDNA fragments were
then cloned into the transcription vector pGEM3Z (Promega). The TGF-ß1
probe (528 bp) is located near the 5'-terminus of the coding sequence
(corresponding to amino acids 32 to 207). After linearization of the plasmid,
single-stranded RNA probes complementary (antisense) or anticomplementary
(sense, negative control) to cellular mRNA were obtained by run-off
transcription using T7 or SP6 polymerase. For in situ hybridization,
the probes were labeled with digoxigenin, using the digoxigenin RNA labeling
kit (Boehringer); for the RNase protection assay, [-32P]UTP
(800 Ci/mmol; Hartmann, Braunschweig, Germany) and the Maxiscript kit (Ambion,
Heidelberg, Germany) were used.
In Situ Hybridization of TGF-ß1 in Kidney Sections and
Quantitative Analysis
Nonradioactive in situ hybridization and detection of the bound
riboprobes by gold-silver enhancement was performed with digoxigenin-labeled
sense and antisense TGF-ß1 riboprobes, as described
(21). Control experiments were
performed with sense TGF-ß1 and with hybridization buffer without probe.
All in situ hybridization experiments, including the silver
enhancement, were performed in parallel, to ensure identical reaction
conditions. All animals were investigated at two per group in every assay
session. The interassay variability was <10%. For quantitative analysis of
the in situ hybridization data, sections from each animal were
randomly selected, and grains were counted in glomerular and proximal tubular
cells. Only cells for which the nucleus and cytoplasm were clearly
attributable to that defined cell were evaluated. As many as 40
cells/glomerulus and 22 glomeruli/group were counted separately by two
investigators. Both investigators were unaware of the treatments received by
the different animals.
Mesangial Cell Culture
Porcine mesangial cells were isolated and characterized as described
previously (23). For
experiments, cells were grown in RPMI 1640 medium (10 to 30 mM glucose) or in
medium with phorbol myristate acetate (PMA) (0.1 µM), without or with
varying amounts of GAG, for the periods indicated. For experiments performed
with elevated glucose concentrations, mannitol was added to the control
medium. For TGF-ß1 bioassays, the conditioned media were centrifuged for
5 min at 4°C. The pellet was discarded, and the supernatant was collected,
divided into aliquots, and stored frozen at -20°C until the bioassay.
TGF-ß1 Bioassay (Mink Lung Cell Proliferation Assay)
Mink lung epithelial cells (CCL; American Type Culture Collection,
Rockville, MD) were maintained in Dulbecco's modified Eagle's medium (DMEM)
(Seromed, Berlin, Germany) with 10% fetal calf serum (FCS) and were used for
the TGF-ß growth inhibition assay essentially as described by Itoh et
al. (24). The cells were
trypsinized, washed with DMEM, suspended in standard RPMI 1640 medium with 10%
FCS, and seeded at 7 x 103 cells/0.2 ml in each well of
96-well dishes. After 3 h, mesangial cell supernatants were added. These
supernatants had been extensively dialyzed against serum-free RPMI 1640
medium, for equilibration with the different glucose concentrations, and were
supplemented with 1% FCS and 4 mM glutamine before being added to the mink
lung cells. For determination of total (latent plus active) TGF-ß levels,
supernatants were heat-activated at 85°C for 5 min. After 22 h of
incubation, the cells were pulsed for 2 h with 1.0 µCi
[3H]thymidine/well, washed twice with phosphate-buffered saline,
trypsinized, collected using a microculture harvesting device, and counted.
For each assay, a standard curve was established with 0.01 to 1 ng/ml human
recombinant TGF-ß1 (Life Technologies, Karlsruhe, Germany), with or
without GAG/mH. Preliminary experiments were performed to exclude any direct
inhibitory effects of GAG/mH on mink lung cell proliferation. Under the
conditions used, GAG/mH only marginally inhibited proliferation
(Table 1). To neutralize
TGF-ß activity, a rabbit anti-TGF-ß antibody (R&D Systems,
Minneapolis, MN) was added at a concentration of 10 µg/ml.
|
Determination of Mesangial Cell TGF-ß1 mRNA Levels by
RT-PCR
Mesangial cells were grown in 28-cm2 dishes and washed, and
total RNA was isolated as described
(23). Two hundred nanograms of
total RNA were used to synthesize cDNA with random priming, according to the
instructions provided by the manufacturer for the RNA-PCR kit (Perkin-Elmer).
Four microliters of the RT reaction mixture were used to amplify, in different
tubes, TGF-ß1 (161 bp) and the housekeeping gene GAPDH (983 bp). The PCR
amplification was performed in a final volume of 50 µl, containing 1.5 mM
MgCl2, 0.2 mM dNTP, 2 U of Taq DNA polymerase from a
freshly prepared 28:1 mixture of Taq antibody and Taq
polymerase, 0.4 µM primers, 50 mM KCl, and 10 mM Tris-HCl (pH 8). Kinetic
analysis was performed using 24 to 32 amplification cycles (94°C for 45 s,
60°C for 45 s, and 72°C for 2 min), to obtain quantitative data. PCR
fragments were separated by polyacrylamide gel electrophoresis and observed by
silver staining. Quantification was performed by direct densitometric analysis
of silver-stained bands (IBAS 2000; Kontron). Ratios of the OD values for the
TGF-ß1 (30 cycles) and GAPDH (24 cycles) PCR products were determined and
compared.
Determination of TGF-ß1 mRNA Levels by Northern Blotting and by
RNase Protection Assays
For Northern analysis, RNA was prepared from cultured mesangial cells with
the RNeasy kit (Qiagen, Hilden, Germany). Total RNA (20 µg) was separated
on a formaldehyde-containing agarose gel and transferred to a nylon membrane.
Digoxigenin-labeled RNA probes have been described. Hybridization was
performed overnight at 68°C in 5x SSC, 50% formamide, 0.1%
N-lauroylsarcosine, 0.02% sodium dodecyl sulfate (SDS), 2% blocking
reagent (Boehringer), and the filters were then washed with 2x SSC/0.1%
SDS at room temperature and with 0.5x SSC/0.1% SDS at 68°C. For
detection, a digoxigenin luminescence detection kit (Boehringer) was used.
TGF-ß1 mRNA levels were also assessed by RNase protection assays, as
described previously (23).
Determination of Protein Kinase C (PKC) Isoform Translocation and
PKC- Activity in Mesangial Cells
Mesangial cells were stimulated with PMA for 0.5 h, in the presence or
absence of 10 µg/ml GAG as indicated. Cellular fractions were prepared and
Western blotting was performed as recently described
(25). Isoform-specific
antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). The
specificity was evaluated as described
(25). For determination of
PKC- activity, mesangial cells were grown to subconfluency in plastic
dishes in RPMI 1640 medium with 15% FCS, transferred to DMEM (without sodium
pyruvate) with 0.5% FCS for synchronization, and cultured for 48 h, without or
with 10 µg/ml GAG/mH, in normal- or high-glucose medium or in medium with
PMA. Cells were then removed from culture dishes by scraping in test buffer
[20 mM Tris-HCl, pH 7.4, 250 mM saccharose, 2 mM ethylenediaminetetraacetate,
8 mM ethylene glycol bis(ß-aminoethyl
ether)-N,N,N',N'-tetraacetate, 20 mM mercaptoethanol, with
protease inhibitor cocktail (Boehringer)], centrifuged at 2000 rpm for 3 min,
and washed once with the same buffer. Packed cells were resuspended in 200
µl of test buffer, sonicated, and microfuged at 14,000 rpm for 20 min at
4°C. The supernatant was incubated overnight at 4°C with
anti-PKC- antibody (C-20 antiserum; Santa Cruz Biotechnology) bound to
protein A-Sepharose. The immunocomplexes were washed three times with 50 mM
Tris-HCl buffer (pH 7.5) containing the protease inhibitor cocktail. The
anti-PKC--immunoprecipitated activity assay was performed for 10 min at
30°C in a 30-µl incubation mixture containing 50 mM Tris-HCl (pH 7.5),
30 mM MgCl2, 20 µM [
-32P]ATP (specific
activity, 2000 cpm/pmol; Amersham Pharmacia Biotech, Milan, Italy), 1 mM
CaCl2, 50 µg/ml phosphatidylserine, and 0.1 mg/ml myelin basic
protein. Reactions were stopped by the addition of 1% SDS and 1%
ß-mercaptoethanol (final concentrations), followed by 5-min treatment at
100°C. Solubilized proteins were subjected to 0.1% SDS-15% polyacrylamide
gel electrophoresis, dried, and counted for radioactivity in a Packard Instant
Imager. The experiment was repeated three times.
Statistical Analyses
One-way ANOVA and ANOVA for repeated measures were used for morphometric
analysis of in situ hybridization and immunohistochemical data and
for the in vitro studies. Bonferroni's test for multiple comparisons
was applied.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Effects of GAG/mH Therapy on Structural Changes and TGF-ß
Expression in the Kidneys of Diabetic Rats
Diabetic rats exhibited an accumulation of PAS-stained mesangial matrix
(Figure 2A, b), which was not
observed in the control animals (Figure 2A,
a) or in the GAG/mH-treated animals
(Figure 2A, c). Similarly, the
tubular basement membranes stained more intensely in diabetic animals, whereas
these structures appeared unchanged in GAG/mH-treated diabetic animals,
compared with control animals. Morphometric quantification of the amount of
PAS-positive glomerular matrix confirmed the aforementioned observations
(Figure 2A, d). Because
previous reports indicated that collagen III is virtually absent in normal
glomeruli but is present at increased levels in diabetic glomeruli
(22), we studied the effect of
GAG/mH treatment on collagen III protein content. The immunohistochemical
analysis of collagen III revealed positive staining in the glomerular
mesangial matrix of untreated diabetic rats
(Figure 2B, b); staining was
significantly less extensive in normal glomeruli
(Figure 2B, a) and in
GAG/mH-treated diabetic rats (Figure 2B,
c). In glomeruli from diabetic rats, collagen III staining was
increased more than twofold, compared with both normal control animals
(P < 0.005) and GAG/mH-treated diabetic rats (P <
0.001) (Figure 2B, d). A
corresponding increase in staining in diabetes mellitus and its amelioration
by GAG/mH treatment were also obvious in the peritubular matrix
(Figure 2B). In earlier studies
with the same animals, we demonstrated that glomerular and tubular 1
(IV) collagen expression and deposition are prevented by chronic GAG/mH
treatment
(19,21).
|
Although we observed clearly increased TGF-ß1 protein staining in proximal tubular cells of diabetic animals (see below), insignificant TGF-ß1 staining could be observed in the glomerular matrix or in glomerular cells in all animal groups studied. Therefore, glomerular LTBP expression was analyzed by immunohistochemical analysis. We observed a fourfold increase in LTBP-labeled glomerular cells in diabetic animals (Figure 2C, b and d), whereas there was no significant difference between control and GAG/mH-treated rats (Figure 2C, a, c, and d). In addition to cellular LTBP staining, we occasionally observed extracellular deposition, which was enhanced in untreated diabetic animals. PAS, collagen III, and LTBP staining in GAG/mH-treated normal rats was unchanged, compared with control animals (data not shown).
In contrast to glomerular TGF-ß1 protein deposition, longterm diabetes mellitus had an enormous stimulatory effect on TGF-ß1 protein deposition in tubular epithelial cells (Figure 3). Because it appeared that whole cortical tubules, rather than single cells, were affected, we quantified the effect by counting TGF-ß1-positive tubules. As shown in Figure 3, the large increase in the number of TGF-ß1-positive tubules in animals with untreated diabetes mellitus (Figure 3B) was significantly reduced by GAG/mH treatment (Figure 3C). Furthermore, our data clearly demonstrated that no significant glomerular TGF-ß1 staining could be observed for untreated animals, whereas intense TGF-ß1 staining was observed in tubules in the same sections, confirming the activity of the TGF-ß1-specific antibody used for staining. Despite several attempts to reveal TGF-ß1 staining by treating the sections with urea, acid, or hyaluronidase or using antibodies from different sources, we could not detect glomerular TGF-ß1 staining.
|
GAG/mH Prevention of Renal Overexpression of TGF-ß1 mRNA in
Long-Term Diabetic Rats
In situ hybridization of renal sections for determination of
glomerular and tubular TGF-ß1 expression is shown in
Figure 4. Quantitative
evaluation of glomerular TGF-ß1 mRNA levels is presented in
Figure 5. Control sections
revealed a weak intracellular signal for TGF-ß1 mRNA
(Figure 4A), which was clearly
greater than the sense control (Figures
4D and
5). In diabetic animals, a
marked increase in TGF-ß1 mRNA levels was observed in all glomerular
cells (Figures 4B and
5), particularly mesangial
cells (identified by Thy-1.1 staining) (F = 5.89, P =
0.002). In GAG/mH-treated diabetic rats, TGF-ß1 mRNA expression appeared
unchanged, compared with control animals (Figures
4C and
5) and GAG/mH-treated normal
rats (Figure 5). Similarly, an
increase in TGF-ß1 mRNA expression in tubular epithelial cells of
diabetic animals was observed (Figure
4F); this increase was prevented by GAG/mH therapy
(Figure 4G). Quantification of
tubular TGF-ß1 mRNA expression confirmed TGF-ß1 mRNA overexpression
and its inhibition by GAG treatment (data not shown).
|
|
GAG/mH Inhibition of High Glucose-Induced Overexpression of
TGF-ß mRNA and Protein and TGF-ß Bioactivity in Cultured Mesangial
Cells
To elucidate the molecular mechanisms of the effects of GAG/mH treatment on
diabetes mellitus-associated TGF-ß1 mRNA induction, we studied the
effects of GAG/mH on mesangial cells, the main target cells of diabetes
mellitus in glomeruli. Preliminary studies using in situ
hybridization indicated a dose-dependent, glucose-induced increase in
TGF-ß1 mRNA levels in cultured mesangial cells (data not shown). A
dose-response analysis demonstrated that significant effects were obtained
with 10 µg/ml GAG/mH. Therefore, this concentration was used in the in
vitro experiments. To determine whether there was structural specificity
of this GAG effect, GAG/DS was also studied. Quantitative analysis of the
grains demonstrated a more than threefold increase (10 versus 30 mM
glucose; P < 0.005), which was prevented by the addition of 10
µg/ml GAG/mH. In these preliminary experiments, we observed that GAG/DS,
although less potent, was also active. Because data obtained by in
situ hybridization are semiquantitative, we analyzed the effect by
Northern blotting. As shown in Figure
6, both GAG compounds prevented the high glucose-induced
TGF-ß1 mRNA increase, with only little effect on basal levels.
|
To evaluate whether the effect of GAG/mH on TGF-ß1 mRNA is translated to TGF-ß1 protein, we determined cellular TGF-ß1 mRNA and protein levels and TGF-ß1 bioactivity in the supernatants of GAG/mH-treated mesangial cells stimulated with increased glucose concentrations. Determination of the TGF-ß1 mRNA levels after RT-PCR amplification revealed a more than twofold, dose-dependent increase, which was prevented by 10 µg/ml GAG/mH (F = 55.4, P = 0.000) (Figure 7A). Furthermore, GAG/mH treatment attenuated high glucose-induced mesangial cell overproduction of TGF-ß protein (Figure 7B) and the formation of bioactive TGF-ß, without affecting basal levels (Figure 7C). To exclude a direct effect of GAG on TGF-ß1 protein levels and bioactivity, 10 µg/ml GAG (i.e., >1000-fold molar excess) was added to active recombinant TGF-ß1 (1 ng/ml) and incubated with mink lung cells. TGF-ß1 bioactivity was determined using the conventional mink lung cell proliferation assay. The results are presented in Table 1. Although the addition of 1 ng/ml TGF-ß1 reduced DNA synthesis by 86 and 65% compared with controls under the experimental conditions used, the presence of either GAG preparation used in the animal studies had only marginal effects on TGF-ß1 bioactivity. It is noteworthy that the addition of GAG had marginal effects on the proliferation of both control and TGF-ß1-treated cells, and no TGF-ß1-blocking effect, i.e., abolishment of the antiproliferative effect of TGF-ß1, was observed.
|
GAG/mH and GAG/DS Prevention of PMA-Induced Overexpression of
TGF-ß1 mRNA
To confirm the potency of GAG in preventing TGF-ß1 mRNA
overexpression, we used PMA, a known strong inducer of TGF-ß1 expression.
The addition of PMA increased mesangial cell TGF-ß1 mRNA levels two- to
three-fold after 9 to 12 h (Figure
8A). Coincubation with GAG/mH prevented PMA-induced TGF-ß1
overexpression, without affecting basal expression. To demonstrate that the
heparin structure and possible associated activities (e.g.,
anticoagulation and growth factor binding) are not necessary for the
inhibitory effect, a structurally unrelated GAG, i.e., GAG/DS (see
Materials and Methods), was used. The presence of increasing amounts of GAG/DS
dose-dependently prevented the PMA-induced TGF-ß1 mRNA increase, although
GAG/DS was less efficient than GAG/mH
(Figure 8B). Again, basal
TGF-ß1 mRNA levels were unchanged after the addition of GAG/DS.
|
Effects of GAG on Mesangial Cell PKC Translocation and PKC
Activity
The inhibitory effect of GAG on PMA-induced TGF-ß1 mRNA expression
indicated the involvement of PKC. Therefore, the effects of GAG/mH on
PMA-induced translocation of PKC-, -ß1, -
, and -
were
studied by Western blotting. We observed that GAG/mH inhibited PKC- and
-ß1 but not PKC- and - translocation to the mesangial cell
plasma membrane. Data for PKC- are presented in
Figure 9A. To evaluate whether
the effect of GAG on PKC translocation reflects an effect on PKC activity, we
studied the effect of GAG/mH on PKC- activity. This particular isoform
was studied more extensively because it has been reported that PKC-
expression is necessary for the inhibitory effect of heparin on cell
proliferation (26). As shown
in Figure 9B, GAG/mH inhibited
PMA-induced PKC activity. Stimulation by a high glucose concentration was also
studied, because this condition more closely resembles the pathophysiologic
situation in vivo. The addition of GAG/mH completely prevented high
glucose-induced PKC- activation, whereas GAG/mH had no effect on basal
PKC activity (Figure 9B).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 (IV) accumulation and synthesis and basement membrane thickening
(20,21).
Together, these data demonstrate that the treatment ameliorates the functional
and structural renal changes associated with diabetes mellitus. To investigate the molecular mechanisms of the inhibitory effects of GAG, renal TGF-ß1 expression was studied. In accordance with a previous study of short-term diabetic rats (29), increased glomerular and tubular expression of TGF-ß1 mRNA was observed for our long-term diabetic animals. TGF-ß1 protein levels were also increased in many cortical tubules of diabetic animals. However, we observed no significant glomerular TGF-ß1 protein deposition in our normal and long-term diabetic rats, although some studies reported significant increases in glomerular TGF-ß1 protein expression (5,29). It may well be that glomerular TGF-ß1 protein expression is low after 12 mo of diabetes mellitus and is therefore indistinguishable from that of control animals. To circumvent this problem, we studied the accumulation of LTBP, a peptide whose strict functional and molecular association with TGF-ß is known (9), and we observed significant increases in the glomeruli of diabetic animals, which were normalized by GAG/mH treatment (Figure 2C). The increased levels of LTBP-1 in diabetic rats could explain the increased TGF-ß activation (30), which, notwithstanding low, immunohistochemically undetectable, glomerular levels of TGF-ß protein, might produce increased local TGF-ß bioactivity. It is noteworthy that the plasma half-life of active TGF-ß1 is short; therefore, active TGF-ß1 may be rapidly cleared from glomeruli, providing an explanation for our results with long-term diabetic rats. By using long exposure times, however, we could observe glomerular TGF-ß1 mRNA expression, its increase in diabetes mellitus, and the inhibition by GAG/mH. Our finding that GAG/mH treatment prevented the diabetes mellitus-associated induction of glomerular and tubular TGF-ß1 expression suggested that inhibition of TGF-ß1 overexpression might be a primary target of GAG/mH action. To investigate the molecular mechanisms of GAG/mH action in more detail, a well established in vitro model for the investigation of mesangial cell expansion, i.e., mesangial cells cultured under high glucose conditions, was chosen. In these cells, we observed that GAG/mH, like TGF-ß1 antisense oligonucleotides (18,23), suppressed high glucose-induced TGF-ß1 mRNA and protein levels and bioactivity. Because the unequivocal demonstration of GAG inhibition of TGF-ß1 mRNA overexpression is a crucial point for elucidation of the mode of action, we used four different methods to verify the inhibitory GAG effect on TGF-ß1 mRNA levels. First, using in situ hybridization as a screening method, we could clearly demonstrate the inhibitory effect of GAG treatment on TGF-ß1 mRNA levels in single cells. Second, the use of RT-PCR to assess the dose dependence of the glucose effect demonstrated the inhibition of increased TGF-ß1 mRNA levels to basal levels. Although these two methods are not absolutely quantitative procedures, we recently demonstrated the usefulness of RT-PCR for the estimation of TGF-ß1 mRNA levels in mesangial cells and, in particular, their inhibition by GAG (31). Third, the widely used Northern blotting procedure confirmed the inhibitory effects of GAG/mH and GAG/DS on glucose-induced TGF-ß1 overexpression. Fourth, the dose dependence of the inhibitory effects of both GAG compounds was demonstrated using the RNase protection assay. Together, these data confirm that GAG inhibit TGF-ß function by preventing TGF-ß1 mRNA overexpression, arguing against the fact that GAG inhibit TGF-ß activation, e.g., by inhibition of tissue plasminogen activator formation and subsequent inhibition of latent TGF-ß activation to mature TGF-ß (32).
However, the data could also be explained by GAG/mH inhibition of any TGF-ß1 activity, resulting in interruption of the well known positive feedback mechanism and leading to reduced TGF-ß1 mRNA levels (12). To exclude the possibility that GAG bind directly to TGF-ß1 protein, thereby neutralizing TGF-ß1 activity, we incubated active TGF-ß1 with both GAG preparations. Evaluation of the bioactivity, using the conventional mink lung cell proliferation assay, after this coincubation demonstrated no inhibition of cellular TGF-ß1 bioactivity, indicating that neither GAG interfered with TGF-ß1 receptor binding or with intra- or postreceptor signaling, either by binding to active TGF-ß1 and thus blocking binding to TGF-ß receptors I and/or II or by interfering with intracellular processes, e.g., with type II receptor serine-threonine kinase. Our results are supported by the recent observation that heparin had no inhibitory effect on TGF-ß1 activity (33). In that report, the presence of 3 to 30 µg/ml heparin did not block the mesangial cell synthesis of fibronectin and collagen IV induced by the addition of 6 ng/ml bioactive TGF-ß1 protein. Taken together, our results suggest that GAG exert their inhibitory activity on increased de novo TGF-ß1 formation, without affecting basal TGF-ß1 expression. These observations are in line with the finding that GAG/mH treatment has little effect on renal TGF-ß1 mRNA levels in nondiabetic animals.
The effects of high glucose concentrations on mesangial matrix production
have been attributed to PKC activation
(8). Mesangial cell PKC
activation by the potent activator PMA resulted in more than two- to fourfold
increases in mesangial cell TGF-ß1 mRNA levels, which were completely
inhibited by treatment with both renoprotective GAG preparations (GAG/mH and
GAG/DS) used in this cell culture study and in previous experimental animal
studies
(19,20).
In unstimulated cells, addition of either GAG had no effect on basal
TGF-ß1 mRNA levels. Because these results indicated that the GAG may
exert their inhibitory effects on TGF-ß1 overexpression through
inhibition of PKC activation, the effects of GAG on PKC isoform translocation
were studied. We observed that PMA-stimulated translocation of PKC- and
-ß1 was prevented by both GAG preparations used. We never detected
PKC-ß2 in mesangial cells, in line with previous results demonstrating
that PKC-ß2 is absent in glomerular cells
(34). Although it is generally
accepted that translocation of PKC is accompanied by activation, we verified
our results for PKC- by showing that GAG/mH prevented high
glucose-induced activation of PKC-. Our data are intriguing because
they indicate that (1) GAG may inhibit PKC activation and
(2) the inhibitory effects of GAG on PKC isoforms demonstrate some
specificity (translocation of PKC- and - is not affected). In
agreement with our results, it was recently demonstrated, in peritoneal
fibroblasts, that heparin inhibited the glucose-stimulated PKC activity in the
membrane fraction (35). It is
noteworthy that, as observed for TGF-ß1 expression in animals and
mesangial cells, GAG treatment had no effect on basal PKC translocation or
activity.
Our observation that GAG treatment inhibits only the activated state (induced by PMA or high glucose concentrations) may be explained by the fact that TGF-ß1 gene expression is regulated by at least two different transcription factors; one controls basal transcription, whereas the second, i.e., activator protein-1 (AP-1), responds to PMA (36). Our findings are supported by reports on heparin inhibition of AP-1-mediated transactivation induced by PMA (37). Recent data from our laboratory demonstrated that AP-1-stimulated TGF-ß1 gene activation may be prevented by GAG treatment (38).
A number of studies have confirmed the renoprotective action of heparins and GAG in experimental diabetic nephropathy (19,20,21,27,28,39), although one recent report, which confirmed some of the glomerular morphologic effects, revealed a possible harmful effect on renal function, i.e., increased albuminuria (40). The reasons for these different results are probably the high heparin dosage and the type of commercially available heparin used by those authors, as discussed in detail (41). The fact that the structural heterogeneity of heparin and GAG suggests heterogeneous functions is confirmed by our studies showing that different GAG structures affect high glucose-induced mesangial cell TGF-ß1 mRNA expression differently (42).
The strategies thus far proposed to inhibit the TGF-ß loop are designed to act primarily on TGF-ß protein (11,12,13,14,15) or mRNA ready to be translated (16,17). All strategies were effective in preventing fibrosis in kidney and other organs in different models, and no side effects were observed for the treated animals, indicating that short-term blockage of TGF-ß is tolerable. However, total nonselective blockage of TGF-ß1 synthesis may be deleterious, because TGF-ß1 knockout mice die soon after birth as the result of an autoimmune-like disease (43), and loss of responsiveness to TGF-ß because of receptor mutations may lead to malignant cell transformation (44). Because these therapies were tested only with short-term protocols, the long-term consequences of these approaches are unknown. Our long-term in vivo studies demonstrate that GAG treatment prevents diabetes mellitus-induced overexpression of TGF-ß1, with no autoimmune-like disease, excess mortality rate, or occurrence of cancer (19,20). The results may be explained by the fact that GAG inhibit TGF-ß1 overexpression without affecting basal TGF-ß1 expression.
Heparin can have important side effects, including excessive anticoagulation, although we observed no hemorrhagic death among the animals treated with the modified GAG we used (19,20). Because it has been demonstrated that the renoprotection afforded by heparin-related drugs is not dependent on their anticoagulant activity or degree of sulfation (20,39,45), it should be possible to design GAG molecules or analogs to minimize the possible risks while maintaining the renoprotective activity.
Increased TGF-ß1 expression has also been observed for human patients with diabetic nephropathy (5), suggesting that the results obtained in animals are relevant to human disease. Treatment with low-molecular weight heparin reduced albuminuria in both micro- and macroalbuminuric patients with insulin-dependent diabetes mellitus (IDDM) (46,47). Danaparoid, which is a mixture of sulfated GAG consisting mainly of heparan sulfate, decreased proteinuria in a small, double-blind, crossover study of patients with IDDM with albumin excretion rates greater than 300 mg/24 h (48). Sulodexide, which is a formulation composed of the two GAG (80% fast-moving heparin and 20% DS) that were active in preventing diabetic nephropathy in the experimental model described here, was reported to reduce albuminuria in patients with IDDM or non-insulin-dependent diabetes mellitus (49,50), and this effect lasted for several weeks after withdrawal (50). Interestingly, for diabetic patients, the decrease in albuminuria was also documented with oral administration of the GAG sulodexide (51,52), a finding that confirms the possibility of safely and easily administering these agents via the oral route, as previously demonstrated (53).
In conclusion, our results demonstrate that application of GAG/mH prevents (1) albuminuria, enhanced mesangial matrix expansion, and increased glomerular and tubular expression of TGF-ß1 mRNA in long-term diabetic rats, (2) hyperglycemia- and PMA-induced TGF-ß1 mRNA and protein overexpression in mesangial cells, and (3) PMA- and hyper-glycemia-induced mesangial cell PKC translocation and activation. Together with previous results, these data indicate the possibility that GAG therapy may represent a new therapeutic option for the treatment and/or prevention of diabetic nephropathy and possibly other chronic TGF-ß-related nephropathies and fibrotic processes.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 (IV) collagen expression and
deposition in long-term diabetic rats is prevented by chronic
glycosaminoglycan treatment. Lab Invest74
: 484-495,1996[Medline]
expression is required for heparin inhibition of rat smooth muscle
cell proliferation in vitro and in vivo. J Biol
Chem 271:25928
-25935, 1996This article has been cited by other articles:
 |
|
 |
|
  J. Borawski, M. Dubowski, K. Pawlak, and M. Mysliwiec Effect of Sulodexide on Plasma Transforming Growth Factor-{beta}1 in Healthy Volunteers Clinical and Applied Thrombosis/Hemostasis, February 1, 2010; 16(1): 60 - 65. [Abstract] [PDF]
|
|
|
|
 |
|
 M. Rossini, T. Naito, H. Yang, M. Freeman, E. Donnert, L.-J. Ma, S. R. Dunn, K. Sharma, and A. B. Fogo Sulodexide ameliorates early but not late kidney disease in models of radiation nephropathy and diabetic nephropathy Nephrol. Dial. Transplant., January 7, 2010; (2010) gfp724v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. L. Heerspink, T. Greene, J. B. Lewis, I. Raz, R. D. Rohde, L. G. Hunsicker, S. L. Schwartz, S. Aronoff, M. A. Katz, G. M. Eisner, et al. Effects of sulodexide in patients with type 2 diabetes and persistent albuminuria Nephrol. Dial. Transplant., June 1, 2008; 23(6): 1946 - 1954. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. J. Lewis and X. Xu Abnormal Glomerular Permeability Characteristics in Diabetic Nephropathy: Implications for the therapeutic use of low-molecular weight heparin Diabetes Care, February 1, 2008; 31(Supplement_2): S202 - S207. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.-M. Myint, Y. Yamamoto, T. Doi, I. Kato, A. Harashima, H. Yonekura, T. Watanabe, H. Shinohara, M. Takeuchi, K. Tsuneyama, et al. RAGE Control of Diabetic Nephropathy in a Mouse Model: Effects of RAGE Gene Disruption and Administration of Low-Molecular Weight Heparin Diabetes, September 1, 2006; 55(9): 2510 - 2522. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Lang Regulating Renal Drug Elimination? J. Am. Soc. Nephrol., June 1, 2005; 16(6): 1535 - 1536. [Full Text] [PDF]
|
|
|
|
|
|
K. Shibuya, K. Kanasaki, M. Isono, H. Sato, M. Omata, T. Sugimoto, S.-i. Araki, K. Isshiki, A. Kashiwagi, M. Haneda, et al. N-Acetyl-Seryl-Aspartyl-Lysyl-Proline Prevents Renal Insufficiency and Mesangial Matrix Expansion in Diabetic db/db Mice Diabetes, March 1, 2005; 54(3): 838 - 845. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Gross, M. J. de Azevedo, S. P. Silveiro, L. H. Canani, M. L. Caramori, and T. Zelmanovitz Diabetic Nephropathy: Diagnosis, Prevention, and Treatment Diabetes Care, January 1, 2005; 28(1): 164 - 176. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Menne, J.-K. Park, M. Boehne, M. Elger, C. Lindschau, T. Kirsch, M. Meier, F. Gueler, A. Fiebeler, F. H. Bahlmann, et al. Diminished Loss of Proteoglycans and Lack of Albuminuria in Protein Kinase C-{alpha}--Deficient Diabetic Mice Diabetes, August 1, 2004; 53(8): 2101 - 2109. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Yao, D.-Y. Huang, I. L. Pfaff, X. Nie, M. Leitges, and V. Vallon Evidence for a role of protein kinase C-{alpha} in urine concentration Am J Physiol Renal Physiol, August 1, 2004; 287(2): F299 - F304. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Weigert, K. Brodbeck, F. C. Brosius III, M. Huber, R. Lehmann, U. Friess, S. Facchin, S. Aulwurm, H. U. Haring, E. D. Schleicher, et al. Evidence for a Novel TGF-{beta}1-Independent Mechanism of Fibronectin Production in Mesangial Cells Overexpressing Glucose Transporters Diabetes, February 1, 2003; 52(2): 527 - 535. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Gambaro, I. Kinalska, A. Oksa, P. Pont'uch, M. Hertlova, J. Olsovsky, J. Manitius, D. Fedele, S. Czekalski, J. Perusicova, et al. Oral Sulodexide Reduces Albuminuria in Microalbuminuric and Macroalbuminuric Type 1 and Type 2 Diabetic Patients: The Di.N.A.S. Randomized Trial J. Am. Soc. Nephrol., June 1, 2002; 13(6): 1615 - 1625. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. ALEXAKIS, A. GUETTOUFI, P. MESTRIES, C. STRUP, D. MATHE, C. BARBAUD, D. BARRITAULT, J.-P. CARUELLE, and P. KERN Heparan mimetic regulates collagen expression and TGF-{beta}1 distribution in gamma-irradiated human intestinal smooth muscle cells FASEB J, July 1, 2001; 15(9): 1546 - 1554. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |
),
diabetic ([UNK]), and GAG/mH-treated diabetic ([UNK]) animals; 100 glomerular
sections/animal were evaluated. *P < 0.005.
