Effect of high glucose on gene expression in mesangial cells: upregulation of the thiol pathway is an adaptational response
Jolean Morrison1,
Kristen Knoll1,
Martin J. Hessner2,3 and
Mingyu Liang1
1 Department of Physiology, Milwaukee, Wisconsin 53226
2 Max McGee National Research Center for Juvenile Diabetes, Department of Pediatrics, Medical College of Wisconsin, Milwaukee, Wisconsin 53226
3 Childrens Research Institute of the Childrens Hospital of Wisconsin, Milwaukee, Wisconsin 53226
 |
ABSTRACT
|
---|
Pathological alterations in glomerular mesangial cells play a critical role in the development of diabetic nephropathy, the leading cause of end-stage renal disease. Molecular mechanisms mediating such alterations, however, remain to be fully understood. The present study first examined the effect of high glucose on the mRNA expression profile in rat mesangial cells using cDNA microarray. Based on variation-weighted criteria and with a false discovery rate of 4.3%, 459 of 17,664 cDNA elements examined were found to be upregulated and 151 downregulated by exposure to 25 mM D-glucose for 5 days. A large number of differentially expressed genes belonged to several functional categories, indicating high glucose had a profound effect on mesangial cell proliferation, protein synthesis, energy metabolism, and, somewhat unexpectedly, protein sorting and the cytoskeleton. Interestingly, several thiol antioxidative genes (glutathione peroxidase 1, peroxiredoxin 6, and thioredoxin 2) were found by microarray and confirmed by real-time PCR to be upregulated by high glucose. These changes suggested that the oxidative stress known to be induced in mesangial cells by high glucose might be buffered by upregulation of the thiol antioxidative pathway. Upregulation of thiol antioxidative genes also occurred in high-glucose-treated human mesangial cells and in glomeruli isolated from rats after 1 wk of streptozotocin-induced diabetes, but not in human proximal tubule cells. High glucose slightly increased lipid peroxidation and decreased the amount of reduced thiols in rat and human mesangial cells. Disruption of the thiol antioxidative pathway by two different thiol-oxidizing agents resulted in a three- to fivefold increase in high-glucose-induced lipid peroxidation. In summary, the present study provided a global view of the short-term effect of high glucose on mesangial cells at the level of mRNA expression and identified the upregulation of the thiol antioxidative pathway as an adaptational response of mesangial cells to high glucose.
diabetes; diabetic nephropathy; microarray; oxidative stress; adaptation
 |
INTRODUCTION
|
---|
DIABETIC NEPHROPATHY, a common complication of diabetes, is the leading cause of end-stage renal disease in the United States (41). Diabetic nephropathy involves both glomerular and tubulointerstitial compartments. Abnormal extracellular matrix deposition, cellular growth, proliferation, and the impairment of function in glomerular mesangial cells contribute importantly to the progression of glomerular injury that eventually leads to glomerulosclerosis and renal failure in diabetic nephropathy (1, 33, 42).
While other abnormalities occurring in diabetes may participate in causing the pathological alterations of mesangial cells, hyperglycemia is believed to be the most important factor. High concentrations of glucose could damage mesangial cells through the stimulation of several interrelated pathways involving transforming growth factor-ß, the polyol pathway, protein kinase C, the hexoamine pathway, or oxygen free radicals (3, 11). It is conceivable, however, that additional pathways or currently unrecognized aspects of the above pathways may also participate in mediating the effect of high glucose on mesangial cells. Identifying those new molecular elements and integrating fragmental molecular events into a cohesive global view are important for obtaining deeper insights into the mechanism of diabetic nephropathy and for discovering potential new therapeutic targets.
High-throughput gene expression profiling (27) is a powerful approach for obtaining global views of biological events and for generating novel hypotheses. A number of previous studies have attempted to assess the effect of high glucose on mesangial cell gene expression on a large scale using techniques such as mRNA differential display (17), suppression subtractive hybridization (5, 35), and Affymetrix GeneChip (22). Interesting changes of gene expression were identified in these studies, demonstrating the discovery power of these high-throughput approaches. The present study extended this line of work by first producing an extensive gene expression profile of rat mesangial cells exposed to normal or high glucose. Taking advantage of recent advances in expression profiling and analytical techniques (27), the present study utilized a cDNA microarray containing
18,000 elements, an experimental design incorporating several replicates, and a variation-weighted statistical method for identifying differential expression. Results of the expression profiling indicated a number of prominent global characteristics of the effect of high glucose on mesangial cells that had not been emphasized previously. Furthermore, a hypothesis that high-glucose-induced oxidative injury in mesangial cells was buffered by upregulation of the thiol antioxidative pathway was generated by this expression profiling and subsequently tested.
 |
MATERIALS AND METHODS
|
---|
Materials.
Chemicals and molecular grade solutions were obtained from Sigma (St. Louis, MO) and Invitrogen (Carlsbad, CA), respectively, unless otherwise indicated. Plastic supplies for cell culture were from Corning (Corning, NY) or Becton-Dickinson (Franklin Lakes, NJ).
Cell culture.
Immortalized rat mesangial cells obtained from American Type Culture Collection (Manassas, VA) were cultured in Dulbeccos modified Eagles medium containing 5 mM D-glucose that was supplemented with 0.4 mg/ml G418 and 15% fetal bovine serum. Cells were propagated for at least five passages in this medium in the investigators laboratory to verify stable growth characteristics and morphological features prior to treatment. Human mesangial cells and proximal tubule epithelial cells were obtained from Cambrex Bio Science (Walkersville, MD) and cultured in cell type-specific media supplied by the vendor of the cells. These media contained 57 mM D-glucose. Culture medium with additional D-glucose added to a final concentration of 25 mM was used for high-glucose treatment. Culture medium with 20 mM D-mannitol or L-glucose added was used as normal glucose control. No apparent differences in the measurements performed in this study were observed between cells with D-mannitol or L-glucose supplementation. Culture medium was changed every 2 days during treatment. The D-glucose concentration in the high-glucose culture medium after 2 days of incubation was above 18 mM as measured with a glucometer (Roche Diagnostics, Basel, Switzerland).
Construction of cDNA microarrays.
Rat cDNA microarrays were essentially constructed as described previously (12, 13). There were 18,432 elements printed, including 768 controls (GAPDH, ß-actin, Arabidopsis clones, printing buffer, etc.) and 17,664 rat cDNA clones obtained from University of Iowa. cDNA clones were amplified, purified, quantified, and reconstituted at 150 ng/µl in 3% DMSO/1.5 M betaine for printing using a GeneMachines OmniGrid printer (San Carlos, CA). Printed arrays were postprocessed using a nonaqueous method (9).
Microarray hybridization.
RNA extraction, cDNA labeling, microarray hybridization and scanning, and raw data extraction were performed as described previously (30, 31). Briefly, 30 µg of total RNA extracted from mesangial cells was reverse-transcribed to cDNA using oligo-(dT)1218 as the primer and labeled with fluorescent dyes Cy3 or Cy5. cDNAs derived from cells treated with normal and high glucose were labeled with different dyes, then pooled, purified, and hybridized together to a microarray. Hybridized microarrays were scanned and digitized using a confocal scanner. Raw values of fluorescent intensity in the signal and background areas of each spot were extracted using the software ImaGene 4.01 (BioDiscovery, Los Angeles, CA). The entire raw dataset has been deposited into the Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo, series GSE1027).
Four separate cell preparations were examined for each treatment condition. cDNAs from cells treated with normal glucose were labeled with Cy3 and those from high-glucose-treated cells were labeled with Cy5 in two microarrays, whereas the dyes were reversed in the other two microarrays. This design incorporated both biological replication and dye switching and had been shown by a previous analysis to be robust yet cost-effective (26).
Microarray data analysis and annotation.
Microarray data were processed as described previously (30, 31). Briefly, fluorescent intensity data were filtered and adjusted, and the ratio between high glucose and normal glucose was calculated for each cDNA element on the array that passed the detectability and quality filters. Ratios were ln-transformed and normalized. Of 17,664 cDNA elements, 7,499 passed the detectability and quality filters consistently in all four microarrays.
A statistical method for identifying differentially expressed genes incorporating permutation and considerations of variance among biological samples has been described previously (40). The original version of the above method was designed for analyzing intensity data. The two-color hybridization method used in the present study allows analysis to be performed using ratios within array. To preserve this advantage, the statistical method mentioned above was slightly modified to analyze ln(ratio) data in the present study. Mean and SD of ln(ratio) were calculated for each of 7,499 cDNA elements that had passed the filters and yielded ln(ratio)s in all four arrays. A t value was calculated for each element as the absolute value of mean divided by SD. The ln(ratio) data were rearranged to create balanced permutations that had an equal number of normal-glucose and high-glucose samples as well as an equal number of Cy3 and Cy5 in the numerators and the denominators. A tp value was similarly calculated for each element in each permutation. An arbitrary value of t was set as the threshold of differential expression. cDNA elements that had t values (tp in the case of permutated data sets) greater than the threshold t value were identified. False discovery rate was calculated as the average number of elements identified in permutated data sets divided by the number of elements identified in the actual experimental data set. The threshold t value was then adjusted until the desired false discovery rate was achieved.
The original document for cDNA elements printed on the array contained the description, GenBank accession number, clone ID, UniGene cluster ID, and other information for each element. Additional annotation was performed for cDNA elements identified as differentially expressed. This included searching homologies for EST elements, identifying gene functions, and establishing pathway affiliation. Numerous publications and publicly available databases were utilized, examples of which included the NCBI databases (http://www.ncbi.nih.gov), the rat genome database (http://rgd.mcw.edu), and Swiss-Prot (http://us.expasy.org/sprot).
Real-time PCR.
Real-time quantitative PCR was carried out using the ABI Prism 7900HT sequence detection system and SYBR Green reagents from Applied Biosystems (Foster City, CA). Results of characterization studies confirmed the wide dynamic range and the reproducibility of this assay. Primers were designed using Primer Express ver. 2.0 (Applied Biosystems) and are listed in Table 1. The RT-PCR reaction mixture contained 1x SYBR Green PCR master mix, 0.25 U/µl MultiScribe reverse transcriptase, 0.4 U/µl RNase inhibitor, 400 nM forward and reverse primers, and 10 ng total RNA, in a volume of 10 µl. Each reaction was performed in triplicate in clear 384-well plates at 48°C, 30 min; 95°C, 10 min; then 95°C, 15 s, and 60°C, 1 min, for 40 cycles. This was followed by the construction of a dissociation curve through increasing the temperature from 60°C to 95°C at a ramp rate of 2%. A single predominant peak was observed in the dissociation curve of each gene, supporting the specificity of the RT-PCR product. Ct numbers (the number of cycles at which fluorescent signals reached a detection threshold that was set within the exponential phase of PCR) were used to calculate the expression levels of genes of interest normalized to endogenous cellular 18S rRNA. The level of 18S rRNA was measured in parallel using primers (50 nM) from a ribosomal RNA control kit from Applied Biosystems.
Streptozotocin-induced diabetes in rats.
Male Sprague-Dawley rats with a 250300 g body wt were treated with an intraperitoneal injection of streptozotocin at 60 mg/kg body wt to induce type I diabetes. Streptozotocin was dissolved in 50 mM citrate buffer (pH 5.0) at 40 mg/ml. Rats were considered diabetic if their blood glucose (measured with a glucometer) was higher than 250 mg/dl both 48 h after the injection of streptozotocin and at the time of glomeruli isolation, i.e., 1 wk after injection. The ratio between right kidney weight and body weight at the time of glomeruli isolation was 0.0047 ± 0.0002 (n = 6) in streptozotocin-treated diabetic rats and 0.0037 ± 0.0001 (n = 7) in control rats (P < 0.05).
Isolation of glomeruli.
A commonly used sieving method (19), with some modifications, was used to isolate glomeruli from control and treated rats 1 wk after the streptozotocin injection. Rats were anesthetized, and the left kidney was flushed with ice-cold HBSS via the abdominal aorta. The renal cortex was cut into small pieces and transferred to a 50-ml beaker on ice. The tissue was ground up with a glass pestle and transferred to a stainless steel sieve with a pore size of 140-µm. The sieve was rinsed with
250 ml of ice-cold HBSS, and the flow-through was collected in a beaker on ice. The grinding and rinsing steps were repeated if large pieces of tissue remained on the sieve. The collected flow-through was poured onto another sieve with a pore size of 74 µm. The sieve was rinsed extensively with
500 ml of ice-cold HBSS. Glomeruli retained on the sieve were collected and inspected under a microscope equipped with a cold dissecting station. Glomeruli with >99% purity were pelleted by centrifugation and immediately resuspended in 1 ml TRIzol reagent (Invitrogen) for RNA extraction.
Lipid peroxidation assay.
Thiobarbituric acid reactive substance (TBARS) was measured as an index of lipid peroxidation. TBARS was measured as described previously (28, 29) with some modifications. Briefly, 100 µl of cell homogenate was combined with 300 µl of a 1:1:1 mixture of 15% TCA, 0.375% thiobarbituric acid, and 0.01% butylhydroxytoluene. After incubation at 95°C for 30 min, the reaction mixture was extracted with 1-butanol, and the absorbance was read at 535 nm. We used 1,1,3,3-methylmalonaldehyde as a standard.
Cellular reduced thiols assay.
The amount of cellular reduced thiols was measured using Ellman reagent. We combined 20 µl of cell homogenate with 75 µl of reaction buffer (30 mM Tris·HCl, pH 8.0, 3 mM EDTA), 25 µl of Ellman reagent (1.19 mg/ml in methanol), and 400 µl of methanol. The absorbance of the 3,000 g supernatant was read at 412 nm. N-acetyl-L-cysteine was used as a standard, and the amount of reduced thiols was normalized by protein content.
Statistics.
Microarray data were analyzed as described above. For other data, Students t-test or one-way ANOVA followed by Student-Newman-Keuls test was used when appropriate. P < 0.05 was considered significant. Results are shown as means ± SE.
 |
RESULTS
|
---|
Expression profiles.
Five-day treatment of rat mesangial cells with high glucose (25 mM D-glucose) resulted in the upregulation of 459 cDNA elements and the downregulation of 151 compared with normal glucose controls, representing 3.5% of 17,664 cDNA elements examined. The description, GenBank accession number, and mean and SD of the ln(ratio) of each of the 610 differentially expressed cDNA elements are shown in Supplemental Table S1 (available as Supplemental Material, from the Physiological Genomics web site).1
The threshold t value was 2.0, and the false discovery rate was 4.3% (see MATERIALS AND METHODS). A slightly smaller number of elements (568 elements) would be considered differentially expressed, while the false discovery rate would be reduced by more than one-half to 1.9%, if an additional criterion of absolute ln(ratio) > 0.18 (i.e., a change of more than 20%) was applied. This indicated that the majority of false positives were among the elements with smaller fold changes. We chose to report all identified elements without applying the additional ln(ratio) criterion because of the potential functional importance of some expression changes that were small in magnitude.
A selected set of 145 differentially expressed cDNA elements is shown in Tables 25 along with their GenBank accession numbers, brief notes of additional functional information, Swiss-Prot IDs, and mean and SD of their ln(ratio) values. These elements were emphasized because they represented genes that formed large clusters corresponding to several functional categories as shown in Tables 25. Specifically, 21 elements were found to be involved in cell cycle and proliferation, and 12 were found for the cytoskeleton (see Table 2); 15 elements were found for mitochondrial and energy metabolism, and 5 were found for oxidative stress (Table 3); 54 elements were found for protein synthesis, sorting, and degradation (Table 4); and 15 elements were found for signal transduction, and 24 were found for general transcriptional regulation (Table 5). This categorization was the result of bioinformatic analysis of all of the 610 identified elements (see MATERIALS AND METHODS). An example of the analysis was the homology search for 263 EST elements that had not previously been found to be homologous with known genes. Homology with known genes (mostly rat or mouse genes) with a BLAST score of more than 200 was found for 65 of these 263 elements, expanding the list of elements that could be linked to some biological functions.
View this table:
[in this window]
[in a new window]
|
Table 2. Large functional clusters of genes that exhibited differential expression in rat mesangial cells in response to high glucose: Cell cycle and proliferation and Cytoskeleton
|
|
View this table:
[in this window]
[in a new window]
|
Table 5. Large functional clusters of genes that exhibited differential expression in rat mesangial cells in response to high glucose: Signal transduction and Transcription
|
|
View this table:
[in this window]
[in a new window]
|
Table 3. Large functional clusters of genes that exhibited differential expression in rat mesangial cells in response to high glucose: Mitochondrial/energy metabolism and Oxidative stress
|
|
View this table:
[in this window]
[in a new window]
|
Table 4. Large functional clusters of genes that exhibited differential expression in rat mesangial cells in response to high glucose: Protein synthesis, sorting, and degradation
|
|
The expression profile of the 145 elements in Tables 25 revealed several global characteristics of the effect of a 5-day exposure to high glucose on mesangial cells. First, high glucose appeared to induce an overall elevation of cellular activity and metabolism. This was supported by upregulation of a large number of genes associated with cell cycle progression (e.g., cyclin B1, proliferating cell nuclear antigen, replication factor C), protein synthesis (e.g., eukaryotic initiation factor 4B, aspartyl-tRNA synthetase, eukaryotic initiation factor 5), and transcriptional activation (e.g., retinoic acid receptor RXR-
, general transcription factor II-1, cAMP-dependent transcription factor), and downregulation of genes related to protein degradation (e.g., protein-L-isoaspartate O-methyltransferase, cytosol aminopeptidase). The upregulation of genes related to protein synthesis and the downregulation of those for protein degradation were consistent with the known hypertrophic effect of high glucose on mesangial cells. Second, a smaller number of genes in several categories with apparently counteracting functions were also activated. Examples included genes involved in cell cycle arrest, such as cyclin G1 and prohibitin, and transcription repression, such as histone methyltransferase. Third, a somewhat surprisingly large number of genes related to protein sorting and trafficking (e.g., RAB2, RAB7, vacuolar protein sorting 29, archain 1, ARL-6 interacting protein-1) were found to be differentially expressed.
We found that 116 differentially expressed elements in Supplemental Table S1 also had some known functions. These are not included in Tables 25 because they did not seem to form large clusters of genes with similar roles. Some of these genes were known to be important for the effect of high glucose on mesangial cells. Notable examples included two elements (GenBank accession nos. AA964281 and AA964892), both encoding collagen type IV
1, that were both highly upregulated. This is consistent with the known effect of high glucose to induce increases in extracellular matrix deposition. Furthermore, several genes identified and validated to be differentially expressed in previous screening studies (5, 17, 22, 35) were confirmed by the present study. These included thrombospondin 1 (5, 17), a gene related to extracellular matrix, and vitamin D3-upregulated protein, a gene emphasized by Kobayashi et al. (22). However, many genes identified in previous screening studies were not confirmed by the present study, suggesting differences in species (human vs. rat) and treatment conditions (e.g., 21 days vs. 5 days) and possibly technique-related variations. Another well-known mediator in diabetic nephropathy, transforming growth factor-ß (38, 48), was not represented in this array. When a gene was represented by more than one cDNA element on the array, changes in the same direction were usually observed in those elements. The magnitude of change, however, varied in some cases, which could result from subtle differences in the amount or sequence of the cDNA probes.
Thiol antioxidative pathway.
Three cDNA elements representing glutathione peroxidase 1, peroxiredoxin 6, and thioredoxin 2 were found to be upregulated by high glucose (see Table 3). Glutathione peroxidase 1 is one of the major enzymes catalyzing the reduction of peroxides (such as hydrogen peroxide and lipid peroxide) using glutathione as the hydrogen donor (8). Peroxiredoxin 6 is a member of the peroxiredoxin family that has peroxide reducing activities similar to glutathione peroxidase 1, except that glutathione peroxidase 1 is selenium dependent, whereas peroxiredoxins are not (16, 43). Thioredoxin 2 is the mitochondria-specific isoform of thioredoxin that plays an important role in the reduction of cellular peroxides and protein disulfide bonds (2, 14, 18).
The upregulation of these three genes was of immediate interest to us because of the authors long-time interests in oxidative stress (28, 34, 46). As shown in Fig. 1, upregulation of these genes would indicate an increase in the activity of the thiol antioxidative pathway. Since it was widely known that high glucose induces substantial increases in the level of reactive oxygen species in several cell types (3) including mesangial cells (20, 21), it appeared that the upregulation of the thiol antioxidative pathway might be an adaptational response of mesangial cells that could buffer high-glucose-induced oxidative stress. It is worth noting that other major antioxidative genes, such as Cu-Zn superoxide dismutase, Mn superoxide dismutase, thioredoxin, thioredoxin reductase, and catalase were not differentially expressed based on the microarray data.

View larger version (18K):
[in this window]
[in a new window]
|
Fig. 1. Alterations of the thiol antioxidative pathway by high glucose. Open-box font (GPX1, PRDX6, and Trx2) indicates genes upregulated by high glucose (5 days) in rat and human mesangial cells. Solid bold font indicates genes represented in the microarray but not altered by high glucose. Gray font indicates genes not represented in the microarray. Genes encoding enzymes are shown in italic font. ROOH, peroxides; GSH, reduced glutathione; Trx, thioredoxin; Trx2, thioredoxin 2; GPX1, glutathione peroxidase 1; PRDX6, peroxiredoxin 6; GSR, glutathione reductase; TXNR, thioredoxin reductase; TXNR2, thioredoxin reductase 2; GSSG, oxidized glutathione.
|
|
To test this hypothesis, we first reexamined the mRNA levels of glutathione peroxidase 1, peroxiredoxin 6, and thioredoxin 2 using real-time PCR. As shown in Fig. 2A, results of the real-time PCR analysis in the original as well as additional cell preparations validated the upregulation of all three genes in rat mesangial cells in response to high glucose. The absence of differential expression of thioredoxin and catalase was also validated by real-time PCR. High glucose similarly upregulated the mRNA expression of glutathione peroxidase 1, peroxiredoxin 6, and thioredoxin 2 in human mesangial cells (Fig. 2B), suggesting that this upregulation was not likely the result of artificial culture conditions of one particular cell line. The expression of these genes in human proximal tubule epithelial cells, however, was not affected by high glucose, suggesting that this effect might be specific for certain cell types such as mesangial cells, rather than a nonspecific cellular response.

View larger version (22K):
[in this window]
[in a new window]
|
Fig. 2. The upregulation of thiol antioxidative genes was validated in rat and human mesangial cells as well as rat glomeruli by real-time PCR. The upregulation of all three genes by high glucose (5 days) was statistically significant in both rat (A) and human (B) mesangial cells. GPX1 and TXN2, but not PRDX6, were also significantly upregulated in glomeruli isolated from diabetic rats 1 wk after streptozotocin (STZ) injection (C). Gene expression levels in human mesangial cells and rat glomeruli were analyzed by real-time PCR. *P < 0.05 vs. NG-treated cells or control rats. HG, high glucose; NG, normal glucose; GPX1, glutathione peroxidase 1; PRDX6, peroxiredoxin 6; TXN2, thioredoxin 2.
|
|
mRNA expression levels of these genes were further examined in glomeruli of diabetic rats to determine whether the changes observed in cultured mesangial cells also occurred in vivo. Glutathione peroxidase 1 and thioredoxin 2 mRNAs were significantly upregulated in glomeruli isolated from rats with 1 wk of streptozotocin-induced diabetes compared with controls (Fig. 2C). Peroxiredoxin 6, however, was not differentially expressed in glomeruli. Thioredoxin and catalase were not differentially expressed, either, similar to mesangial cells in culture.
We then examined the effect of high glucose on TBARS, an index of lipid peroxidation, and the amount of cellular reduced thiols in cultured mesangial cells. As shown in Fig. 3, high glucose significantly increased TBARS in rat and human mesangial cells by 12% and 25%, respectively. Meanwhile, high glucose decreased the amount of cellular reduced thiols in rat and human mesangial cells by 13% and 21%, respectively. These changes support the presence of high-glucose-induced oxidative injury and increases in the consumption of reduced thiols. On the other hand, the small magnitude of these changes suggested that either the effect of high glucose on lipid peroxidation and thiol consumption was mild, or an otherwise larger effect was buffered by the activation of protective pathways. The latter possibility would be consistent with the upregulation of thiol antioxidative gene expression shown above.

View larger version (21K):
[in this window]
[in a new window]
|
Fig. 3. High glucose increased lipid peroxidation and decreased the amount of cellular reduced thiols slightly. Rat (A) and human (B) mesangial cells were treated with high (HG) or normal (NG) glucose for 5 days. TBARS, thiobarbituric acid reactive substance; n = 811. *P < 0.05 vs. normal glucose.
|
|
High-glucose-induced oxidative stress would be exacerbated by suppression of the thiol antioxidative pathway if the upregulation of this pathway had been buffering the oxidative damage. Indeed, thiol-oxidizing agents 5,5'-dithiobis(2-nitrobenzoate) (DTNB, 1 mM) or diamide (100 µM) substantially exacerbated high-glucose-induced increases of TBARS in rat mesangial cells by three- to fivefold (Fig. 4A). DTNB or diamide alone did not have significant effects on TBARS when cells were cultured in media containing normal glucose supplemented with D-mannitol or L-glucose. DTNB similarly exacerbated high-glucose-induced increase in TBARS in human mesangial cells (Fig. 4B). Diamide at this dose, however, was cytotoxic to human mesangial cells, causing cell death regardless of glucose concentrations.

View larger version (14K):
[in this window]
[in a new window]
|
Fig. 4. Thiol oxidizing agents exacerbated high-glucose-induced lipid peroxidation in mesangial cells. Rat (A) and human (B) mesangial cells were treated with high glucose (HG) in the absence or presence of 5,5'-dithiobis(2-nitrobenzoate) (DTNB, 1 mM) or diamide (100 µM) for 5 days. As parallel controls, cells were treated with normal glucose in the absence or presence of DTNB or diamide at the same concentrations. Results were expressed as percent increases relative to corresponding normal glucose control with or without DTNB or diamide. Diamide at this dose was cytotoxic to human mesangial cells. TBARS, thiobarbituric acid reactive substance; n = 611. *P < 0.05 vs. HG alone.
|
|
In summary, 5 days of high-glucose exposure and 1 wk of streptozotocin-induced diabetes upregulated the mRNA expression of thiol antioxidative genes in mesangial cells in culture and glomeruli in vivo, respectively. High glucose slightly decreased the amount of cellular reduced thiols in mesangial cells, and high-glucose-induced increases of lipid peroxidation were substantially exacerbated by thiol-oxidizing agents.
 |
DISCUSSION
|
---|
The present study provided the first comprehensive profile of mRNA expression in rat mesangial cells cultured in normal and high glucose for 5 days. A hypothesis derived from the microarray data was tested, and for the first time the results demonstrated an adaptational role of the thiol antioxidative pathway in high-glucose-induced mesangial cell abnormalities.
The expression profiles established in the present study exhibit several global characteristics that are consistent with pathological alterations known to occur in high-glucose-treated mesangial cells or in diabetic glomeruli, such as accelerated proliferation (45), hypertrophy (42), and increased deposition of extracellular matrix (32). Perhaps more interestingly, a large number of genes in several categories whose importance in the effect of high glucose on mesangial cells had not been emphasized were found to be differentially expressed. One such category included at least 12 genes related to the cytoskeleton and cell-cell or cell-matrix interactions. The importance of the cytoskeleton in diabetic nephropathy has only been reported in a small number of studies (6, 47), including a previous screening using suppression subtractive hybridization (5). Another such category included a large number of genes related to protein sorting and trafficking (Table 4). The role these cellular components and activities play in the effect of high glucose on mesangial cells and in diabetic nephropathy deserves further study.
Another important global characteristic observed in the present study is that simultaneous activation frequently occurs for genes or pathways with apparently opposing functions (cell cycle progression vs. arrest, protein synthesis vs. degradation, transcriptional activation vs. repression, etc.). This constant balancing between acting and counteracting cellular components could indeed be a general mechanism for achieving dynamic adaptation and precise control of cellular functions. Such a fundamental process is sometimes overlooked. One implication of this fundamental process is that obtaining a more reliable view of cellular functions may require evaluation of the balance among all redundant, complementary, and opposing pathways (27), rather than one or two isolated components or pathways.
Upregulation of the thiol antioxidative pathway is an example of such dynamic adaptation. Increased generation of superoxide anion, possibly from the mitochondrial respiratory chain and other sources, has been proposed as a common event that activates various mechanisms contributing to the development of diabetic complications (3). Large increases of reactive oxygen species have been observed in endothelial cells (10, 36) and mesangial cells (20, 21) treated with high glucose as well as in glomeruli isolated from diabetic rats (23). Antioxidants and antioxidative genes such as superoxide dismutase have generally been shown to be beneficial in the treatment of diabetes and its complications, including diabetic nephropathy (7, 24, 25). The effect of high glucose and diabetes on antioxidative pathways per se, however, is more controversial. Various responses (upregulation, downregulation, or no changes) have been reported for several antioxidative genes (4, 15, 23, 37), which might be partially related to genetic susceptibility to diabetic complications. Results of the present study support the notions that the thiol antioxidative pathway in mesangial cells can be upregulated by high glucose and diabetes and that such upregulation may be an adaptational response that could partially buffer the deleterious effect of high-glucose-induced oxidative stress.
Several important questions emerged from these data and could become interesting subjects for future studies. The in vitro and in vivo functional significance of high-glucose-induced alterations in the expression of several pathways identified in the present study (in addition to the thiol pathway) obviously deserves investigation. It would also be worthwhile to profile gene expression in mesangial cells exposed to high glucose for a shorter or longer duration or in combination with other treatments. Expression profiling in tissues obtained directly from human subjects or animal models of diabetic nephropathy would be highly valuable (39, 44). In regard to the thiol pathway, it remains to be determined whether upregulation of the thiol antioxidative pathway in vivo, as indicated by the data obtained from isolated glomeruli, is an intrinsic protective mechanism in diabetic nephropathy. If so, it would be important to determine whether this protective mechanism diminishes or becomes overwhelmed by injurious processes as the disease progresses and whether diabetic nephropathy can be ameliorated by maintaining or further enhancing this mechanism through gene delivery or other interventions.
 |
GRANTS
|
---|
This study was supported in part by National Heart, Lung, and Blood Institute Grants HL-29587 and HL-49219.
 |
ACKNOWLEDGMENTS
|
---|
We thank the technical staff for assistance in manufacturing microarrays and the Microarray Group in the Department of Physiology for helpful discussion.
 |
FOOTNOTES
|
---|
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: M. Liang, Dept. of Physiology, Medical College of Wisconsin, 8701 Watertown Plank Road, Milwaukee, WI 53226 (E-mail: mliang{at}mcw.edu).
10.1152/physiolgenomics.00031.2004.
1 The Supplementary Material (Table S1, a complete list of the 610 differentially expressed cDNA elements) for this article is available online at http://physiolgenomics.physiology.org/cgi/content/full/00031.2004/DC1. 
 |
References
|
---|
- Adler S. Structure-function relationships associated with extracellular matrix alterations in diabetic glomerulopathy. J Am Soc Nephrol 5: 11651172, 1994.[Abstract]
- Arner ES and Holmgren A. Physiological functions of thioredoxin and thioredoxin reductase. Eur J Biochem 267: 61026109, 2000.[Abstract/Free Full Text]
- Brownlee M. Biochemistry and molecular cell biology of diabetic complications. Nature 414: 813818, 2001.[CrossRef][ISI][Medline]
- Ceriello A, Morocutti A, Mercuri F, Quagliaro L, Moro M, Damante G, and Viberti GC. Defective intracellular antioxidant enzyme production in type 1 diabetic patients with nephropathy. Diabetes 49: 21702177, 2000.[Abstract]
- Clarkson MR, Murphy M, Gupta S, Lambe T, Mackenzie HS, Godson C, Martin F, and Brady HR. High glucose-altered gene expression in mesangial cells. Actin-regulatory protein gene expression is triggered by oxidative stress and cytoskeletal disassembly. J Biol Chem 277: 97079712, 2002.[Abstract/Free Full Text]
- Cortes P, Mendez M, Riser BL, Guerin CJ, Rodriguez-Barbero A, Hassett C, and Yee J. F-actin fiber distribution in glomerular cells: structural and functional implications. Kidney Int 58: 24522461, 2000.[CrossRef][ISI][Medline]
- Craven PA, Melhem MF, Phillips SL, and DeRubertis FR. Overexpression of Cu2+/Zn2+ superoxide dismutase protects against early diabetic glomerular injury in transgenic mice. Diabetes 50: 21142125, 2001.[Abstract/Free Full Text]
- Dickinson DA and Forman HJ. Glutathione in defense and signaling: lessons from a small thiol. Ann NY Acad Sci 973: 488504, 2002.[Abstract/Free Full Text]
- Diehl F, Grahlmann S, Beier M, and Hoheisel JD. Manufacturing DNA microarrays of high spot homogeneity and reduced background signal. Nucleic Acids Res 29: e38, 2001.[Abstract/Free Full Text]
- Giardino I, Edelstein D, and Brownlee M. BCL-2 expression or antioxidants prevent hyperglycemia-induced formation of intracellular advanced glycation endproducts in bovine endothelial cells. J Clin Invest 97: 14221428, 1996.[Abstract/Free Full Text]
- Haneda M, Koya D, Isono M, and Kikkawa R. Overview of glucose signaling in mesangial cells in diabetic nephropathy. J Am Soc Nephrol 14: 13741382, 2003.[Free Full Text]
- Hessner MJ, Wang X, Hulse K, Meyer L, Wu Y, Nye S, Guo SW, and Ghosh S. Three color cDNA microarrays: quantitative assessment through the use of fluorescein-labeled probes. Nucleic Acids Res 31: e14, 2003.[Abstract/Free Full Text]
- Hessner MJ, Wang X, Khan S, Meyer L, Schlicht M, Tackes J, Datta MW, Jacob HJ, and Ghosh S. Use of a three-color cDNA microarray platform to measure and control support-bound probe for improved data quality and reproducibility. Nucleic Acids Res 31: e60, 2003.[Abstract/Free Full Text]
- Hirota K, Nakamura H, Masutani H, and Yodoi J. Thioredoxin superfamily and thioredoxin-inducing agents. Ann NY Acad Sci 957: 189199, 2002.[Abstract/Free Full Text]
- Hodgkinson AD, Bartlett T, Oates PJ, Millward BA, and Demaine AG. The response of antioxidant genes to hyperglycemia is abnormal in patients with type 1 diabetes and diabetic nephropathy. Diabetes 52: 846851, 2003.[Abstract/Free Full Text]
- Hofmann B, Hecht HJ, and Flohe L. Peroxiredoxins. Biol Chem 383: 347364, 2002.[ISI][Medline]
- Holmes DI, Abdel Wahab N, and Mason RM. Identification of glucose-regulated genes in human mesangial cells by mRNA differential display. Biochem Biophys Res Commun 238: 179184, 1997.[CrossRef][ISI][Medline]
- Holmgren A. Thioredoxin and glutaredoxin systems. J Biol Chem 264: 1396313966, 1989.[Free Full Text]
- Ito O, Alonso-Galicia M, Hopp KA, and Roman RJ. Localization of cytochrome P-450 4A isoforms along the rat nephron. Am J Physiol Renal Physiol 274: F395F404, 1998.[Abstract/Free Full Text]
- Kang BP, Frencher S, Reddy V, Kessler A, Malhotra A, and Meggs LG. High glucose promotes mesangial cell apoptosis by oxidant-dependent mechanism. Am J Physiol Renal Physiol 284: F455F466, 2003. First published November 5, 2002; 10.1152/ajprenal.00137.2002.[Abstract/Free Full Text]
- Kiritoshi S, Nishikawa T, Sonoda K, Kukidome D, Senokuchi T, Matsuo T, Matsumura T, Tokunaga H, Brownlee M, and Araki E. Reactive oxygen species from mitochondria induce cyclooxygenase-2 gene expression in human mesangial cells: potential role in diabetic nephropathy. Diabetes 52: 25702577, 2003.[Abstract/Free Full Text]
- Kobayashi T, Uehara S, Ikeda T, Itadani H, and Kotani H. Vitamin D3 up-regulated protein-1 regulates collagen expression in mesangial cells. Kidney Int 64: 16321642, 2003.[CrossRef][ISI][Medline]
- Koya D, Hayashi K, Kitada M, Kashiwagi A, Kikkawa R, and Haneda M. Effects of Antioxidants in Diabetes-Induced Oxidative Stress in the Glomeruli of Diabetic Rats. J Am Soc Nephrol 14: S250S253, 2003.[Abstract/Free Full Text]
- Kubisch HM, Wang J, Luche R, Carlson E, Bray TM, Epstein CJ, and Phillips JP. Transgenic copper/zinc superoxide dismutase modulates susceptibility to type I diabetes. Proc Natl Acad Sci USA 91: 99569959, 1994.[Abstract/Free Full Text]
- Kuroki T, Isshiki K, and King GL. Oxidative stress: the lead or supporting actor in the pathogenesis of diabetic complications. J Am Soc Nephrol 14: S216S220, 2003.[Free Full Text]
- Liang M, Briggs AG, Rute E, Greene AS, and Cowley AW Jr. Quantitative assessment of the importance of dye switching and biological replication in cDNA microarray studies. Physiol Genomics 14: 199207, 2003. First published June 10, 2003; 10.1152/physiolgenomics.00143.2002.[Abstract/Free Full Text]
- Liang M, Cowley AW Jr, and Greene AS. High throughput gene expression profiling: A molecular approach to integrative physiology. J Physiol 554: 2230, 2004.[Abstract/Free Full Text]
- Liang M, Croatt AJ, and Nath KA. Mechanisms underlying induction of heme oxygenase-1 by nitric oxide in renal tubular epithelial cells. Am J Physiol Renal Physiol 279: F728F735, 2000.[Abstract/Free Full Text]
- Liang M and Knox FG. Nitric oxide reduces the molecular activity of Na+,K+-ATPase in opossum kidney cells. Kidney Int 56: 627634, 1999.[CrossRef][ISI][Medline]
- Liang M, Yuan B, Rute E, Greene AS, Olivier M, and Cowley AW Jr. Insights into Dahl salt-sensitive hypertension revealed by temporal patterns of renal medullary gene expression. Physiol Genomics 12: 229237, 2003. First published December 10, 2002; 10.1152/physiolgenomics.00089.2002.[Abstract/Free Full Text]
- Liang M, Yuan B, Rute E, Greene AS, Zou AP, Soares P, McQuestion GD, Slocum GR, Jacob HJ, and Cowley AW Jr. Renal medullary genes in salt-sensitive hypertension: a chromosomal substitution and cDNA microarray study. Physiol Genomics 8: 139149, 2002. First published January 2, 2002; 10.1152/physiolgenomics.00083.2001.[Abstract/Free Full Text]
- Mason RM and Wahab NA. Extracellular matrix metabolism in diabetic nephropathy. J Am Soc Nephrol 14: 13581373, 2003.[Abstract/Free Full Text]
- Mauer SM, Lane P, Hattori M, Fioretto P, and Steffes MW. Renal structure and function in insulin-dependent diabetes mellitus and type I membranoproliferative glomerulonephritis in humans. J Am Soc Nephrol 2, Suppl: S181S184, 1992.
- Morrison J, Liang M, and Cowley AW Jr. Renal expression of NAD(P)H oxidase and nitric oxide synthases in consomic rats: Using real-time PCR as a targeted profiling tool (Abstract). J Am Soc Nephrol 14: 136, 2003.
- Murphy M, Godson C, Cannon S, Kato S, Mackenzie HS, Martin F, and Brady HR. Suppression subtractive hybridization identifies high glucose levels as a stimulus for expression of connective tissue growth factor and other genes in human mesangial cells. J Biol Chem 274: 58305834, 1999.[Abstract/Free Full Text]
- Nishikawa T, Edelstein D, Du XL, Yamagishi S, Matsumura T, Kaneda Y, Yorek MA, Beebe D, Oates PJ, Hammes HP, Giardino I, and Brownlee M. Normalizing mitochondrial superoxide production blocks three pathways of hyperglycaemic damage. Nature 404: 787790, 2000.[CrossRef][ISI][Medline]
- Sechi LA, Ceriello A, Griffin CA, Catena C, Amstad P, Schambelan M, and Bartoli E. Renal antioxidant enzyme mRNA levels are increased in rats with experimental diabetes mellitus. Diabetologia 40: 2329, 1997.[CrossRef][Medline]
- Sharma K and Ziyadeh FN. Hyperglycemia and diabetic kidney disease. The case for transforming growth factor-beta as a key mediator. Diabetes 44: 11391146, 1995.[Abstract]
- Susztak K, Sharma K, Schiffer M, McCue P, Ciccone E, and Bottinger EP. Genomic strategies for diabetic nephropathy. J Am Soc Nephrol 14, Suppl 3: S271S278, 2003.
- Tusher VG, Tibshirani R, and Chu G. Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci USA 98: 51165121, 2001.[Abstract/Free Full Text]
- United States Renal Data System. USRDS Annual Data Report. http://www.usrds.org/adr.htm. Accessed November 19, 2003.
- Wolf G and Ziyadeh FN. Molecular mechanisms of diabetic renal hypertrophy. Kidney Int 56: 393405, 1999.[CrossRef][ISI][Medline]
- Wood ZA, Schroder E, Robin Harris J, and Poole LB. Structure, mechanism and regulation of peroxiredoxins. Trends Biochem Sci 28: 3240, 2003.[CrossRef][ISI][Medline]
- Yechoor VK, Patti ME, Saccone R, and Kahn CR. Coordinated patterns of gene expression for substrate and energy metabolism in skeletal muscle of diabetic mice. Proc Natl Acad Sci USA 99: 1058710592, 2002.[Abstract/Free Full Text]
- Young BA, Johnson RJ, Alpers CE, Eng E, Gordon K, Floege J, Couser WG, and Seidel K. Cellular events in the evolution of experimental diabetic nephropathy. Kidney Int 47: 935944, 1995.[ISI][Medline]
- Yuan B, Liang M, Yang Z, Rute E, Taylor N, Olivier M, and Cowley AW Jr. Gene expression reveals vulnerability to oxidative stress and interstitial fibrosis of renal outer medulla to nonhypertensive elevations of ANG II. Am J Physiol Regul Integr Comp Physiol 284: R1219R1230, 2003. First published January 23, 2003; 10.1152/ajpregu.00257.2002.[Abstract/Free Full Text]
- Zhou X, Hurst RD, Templeton D, and Whiteside CI. High glucose alters actin assembly in glomerular mesangial, and epithelial cells. Lab Invest 73: 372383, 1995.[ISI][Medline]
- Ziyadeh FN, Sharma K, Ericksen M, and Wolf G. Stimulation of collagen gene expression, and protein synthesis in murine mesangial cells by high glucose is mediated by autocrine activation of transforming growth factor-beta. J Clin Invest 93: 536542, 1994.[ISI][Medline]