Induced experimental diabetes in animals was performed using a single injection of 50 mg/kg B.W. of STZ. This dose causes partial destruction of pancreatic β-cells responsible for producing and secreting insulin into the body, therefore, rats showed type 1 diabetes. The lack of insulin causes a decrease in the incorporation of glucose into peripheral tissues such as adipose and muscle, thereby increasing blood glucose [14]. In this experiment, STZ-injected rats showed a significant increase in blood glucose with a decrease in insulin levels, producing a diabetic condition (Table 1).

Regarding physiological parameters after induction of diabetes, significant differences in consumption of food and water, growth retardation and urine volume were observed in diabetic animals in comparison to healthy animals. Also, a decreased creatinine clearance (Ccr) was observed in diabetic animals compared with nondiabetic rats (P < 0.05). On the other hand, markers of renal damage in urine such as urea, albumin and protein excretion showed an increase in diabetic rats compared with those values obtained in nondiabetic animals (Table 1). Additionally, blood urea nitrogen and serum creatinine levels in diabetic rats indicate progressive renal damage, which is considered to be an indication of altered glomerular filtration rate in DN [15]. Our results are similar to those reported by Akbarzadeh et al. [16]. Likewise in this experiment, diabetic rats showed an increase in kidney weight compared to healthy rats (P < 0.05). In addition, these results are consistent to those described by Yamabe et al. [17], who observed an increase in kidney weight in STZ-induced diabetic male Wistar rats compared to healthy rats. This could be related to kidney hyperplasia caused by inflammation. Overall, these results suggest the presence of abnormalities in the renal function of diabetic animals.

Besides the biochemical changes in diabetic rats, morphological changes in kidney were carried out to confirm the development of DN. One of the morphological changes associated to DN is mesangial expansion due to increased mesangial matrix deposition, a mild increase in mesangial cellularity, and hypertrophy of mesangial cells [18]. Sections of rat kidney cortices obtained from the two experimental groups were stained with Hematoxylin-Eosin (H-E) and the results are shown in Figure 1.

The kidneys of healthy rats showed normal morphologies of glomeruli and tubules (Figure la). However, kidneys of the diabetic group showed glomerular damage and severe destruction of tubules (Figure lb). These results are similar to those reported by Omotayo et al. [19]. It has been observed that structural abnormalities of DN are similar in type 1 and type 2 diabetes and all renal compartments (glomeruli, tubulointerstitium, and renal vasculature) can be affected [20].

In recent years, gene expression profiling related with diabetic nephropathy has been studied in different animal models, such as STZ-induced diabetes in mice and rats, as well as spontaneous rodent models of diabetes. These studies have identified genes involved in the development and progression of this disease, but their precise molecular mechanisms have remained unclear and only a low number of those genes are considered biomarkers [3]. Therefore, it is important to identify new genes involved in this disease that can serve as molecular targets to broaden the development of novel therapeutic strategies.

By using SSH technique, 400 EST’s were obtained (200 down- and 200 up-regulated) from kidneys of diabetic rats. Northern blot analysis was performed to confirm the differential expression of the EST’s obtained from the cDNA libraries; however only 200 EST’s were transferred to nylon membranes (100 to each library) and hybridized with RNA probes from either diabetic or healthy rats. It was observed that 88 and 72 clones from the up- and down-regulated libraries, respectively, showed a marked differential expression in the presence of DN in STZ-induced rats. Typical expression pattern of cDNA arrangements of 47 clones from the subtractive library obtained is shown in Figure 2.

EST’s, based on their differential expression pattern, were selected and sequenced. The sequences of these clones were compared to those of the database from NCBI by means of the BLAST algorithm, and the results from the found genes are shown in Tables 2 and 3. Based on this, possible functions for the sequenced EST’s were grouped in: (1) Metabolism protein; (2) transporters and channels; (3) hypothetical proteins; (4) transcription factors; and (5) unknown function.

In the present study, the main objective was to identify some differentially expressed genes involved in DN. As aforementioned, based on hybridization signals, 160 EST’s were selected for sequencing, from which 72 were down- and 88 over-expressed genes (Tables 2 and 3, respectively). BLAST analysis revealed that at least 80% of the genes had been previously reported in the GeneLibrary database. However, 20% of these fragments did not match any gene published in the gen bank, thus raising the possibility of newly identified genes.

Animal experiments were conducted in accordance with ethical guidelines of the University of Queretaro. Male Wistar rats (weighing 250–300 g) were obtained from Rismart S.A., Mexico. The animals were housed at 20–25 °C with a 12-h light-dark cycle, through the experimental period. Animals were allowed free access to water and a base diet (Zeigler rat chow) containing: protein, 18%; fat, 4%; carbohydrates, 54%; fiber, 5%; and sufficient minerals and vitamins to maintain the health of the rats. Diabetes was induced by a single intraperitoneal injection of freshly prepared STZ (Sigma Chemical Company, USA, 50 mg/kg body weight; dissolved in a citrate buffer, 0.01 M, pH 4.5) after overnight food deprivation. Nondiabetic rats received the same volume of vehicle (citrate buffer). After 1 week, rats with blood glucose ≥250 mg/dL were considered diabetic for this experiment. Both nondiabetic (group 1) and diabetic (group 2) groups included eight rats each one and were fed with base diet. Blood samples were taken from tail vein and the glucose concentration was measured by a glucose oxidase assay (Glucose Accutrend, Roche, Germany) and measured by a reflective glucometer (Accutrend, GC). One day before sacrifice, animals were housed in metabolic cages to obtain urine for determination of biochemical parameters related to DN. At week four, animals were anaesthetized and blood samples were withdrawn by cardiac puncture and the serum was immediately separated from the blood samples by centrifugation and stored at −80 °C until analysis. Kidneys were immediately removed, thoroughly washed with cold-ice saline phosphate buffer, and weighed. A piece of kidney was stored in formaldehyde solution and another part was frozen in liquid nitrogen and stored at −80 °C until analysis.

DN was assessed biochemically by determining urine volume, urinary urea, albumin, protein and also blood urea. Likewise, blood and urine creatinine levels were used for quantification of creatinine clearance (Ccr). All these parameters were measured using commercial kits (Randox Laboratories Ltd., UK). Creatinine clearance (C_Cr) was calculated on the basis of urinary Cr, serum Cr, urine volume, and body weight using the following equation: C_Cr (mL/min) = [urinary Cr (mg/dL) × urine volume (mL)/serum Cr (mg/dL) × 1440 (min)]. Also, kidney/body weight ratio was calculated. Data were expressed as relative kidney weight (percentage of body weight). Serum insulin levels were determined under fasting conditions using a commercial kit (Millipore, USA) according to the manufacturer’s instructions.

Renal tissues were fixed in 4% paraformaldehyde solution and embedded in paraffin. Sections were cut at 7 μm with a microtome and deparaffined with xylene. The sections were stained with Hematoxylin-Eosin (H-E) and observed under a light microscope at magnifications of 200×.

Total RNA of diabetic and normal rats was obtained by RNEASY protocol (QIAGEN, Hilden, Germany). Integrity and size of RNA was analyzed by means of electrophoresis in agarose gel with formaldehyde. RNA quantification and assessment of purity were measured spectrophotometrically by the absorbance ratio (260/280 nm).

1 μg of total RNA of each condition (healthy and diabetic) was used as template for synthesizing the first cDNA chain with Superscript II reverse transcriptase (Life Technologies, Rockville, MD, USA) and the protocol of SMART™ cDNA Synthesis, following the instructions of the supplier (Clontech, Palo Alto, CA, USA). The cDNA was amplified by long distance PCR (LD-PCR) with 15, 18, 21, 24, and 27 parallel cycles to compare and guarantee an optimal and suitable amount of PCR products for the construction of the cDNA subtractive library [12]. CROMA-SPIN^TM columns (Clontech, Palo Alto, CA, USA) were used to guarantee highly pure cDNA.

Subtractive hybridization by suppression (SSH) was carried out using the PCR-Select™ cDNA subtraction kit (Clontech, Palo Alto, CA, USA). Tester and driver cDNAs were digested using RsaI to yield blunt ends. In order to obtain up-regulated genes in kidney of diabetic rats, the mRNA of these rats was employed as the “tester” and kidney mRNA of healthy rats as the “driver”. On the other hand, to obtain down-regulated genes from kidney of diabetic rats, kidney mRNA of diabetic rats was employed as the “driver” and kidney mRNA of healthy rats as the “tester”. The tester cDNA fragments were proportionally divided into two aliquots and each one ligated in separated reactions with adapter 1 and adapter 2, resulting in two tester cDNA populations. A small amount of each tester population was mixed with an excess of driver population (5 μg), subsequently heat-denatured, and allowed to anneal for 8 h at 68 °C. The two samples from the first hybridization were combined and annealed overnight at 68 °C with additional fresh-denatured driver cDNA. A primary PCR was conducted to amplify those cDNAs that represented differentially expressed genes. A second PCR was carried out using nested primers 1 and 2 R (Clontech, Palo Alto, CA, USA) to reduce the level of unspecific amplification and for the preparation of Northern blot probes. The products of the secondary PCR amplification were monitored by agarose gel electrophoresis, and the fragments larger than 250 bp were cut out of the agarose gel using a sterile scalpel and purified by using a QIAEXII protocol (QIAGEN, Hilden, Germany).

The differential PCR fragments were ligated into the PCR 4-TOPO cloning vector, according to the manufacturer’s instructions (Invitrogen, Carlsbad, CA, USA). 2 μL of these ligation reactions was used to transform chemically competent Escherichia coli TOP 10 cells (Invitrogen, Carlsbad, CA, USA). The transformed cells were planted on Luria-Bertani (LB) agar containing kanamycin, and the white colonies were selected. Individual E. coli colonies were inoculated into LB media containing kanamycin/ampicillin and shaken overnight at 37 °C. Inserts were screened by restriction analysis using EcoRI (Invitrogen, Carlsbad, CA, USA) and agarose gel electrophoresis. Each selected colony of the library was kept in cryogenic vials with 500 μL of each culture and 500 μL of 100% sterilized glycerol. The collection of colonies (library of differentially expressed genes) was kept at −80 °C. The products were sequenced and subject to homology search by using BLAST program at the web site of National Center of Biotechnology Information [41].

To confirm the expression status of the genes isolated by SSH, Northern blot analysis was performed as previously described [42]. Briefly, the mRNA was isolated from kidneys of diabetic and healthy rats as described above. Of the isolated mRNA, 2 μg were spotted into positive-charged nylon membranes (Hybond^TM-N⁺ Amersham Bioscience, UK Limited) using the slot manifold (Amersham Bioscience). Probes were created from the previously isolated SSH fragments by incorporating fluorescein-11-dUTP (Gene Images CDP-star random prime labeling module) according to the manufacturer’s instructions (Amersham Pharmacia Biotech Inc. Piscataway, NJ, USA). After ultraviolet (UV) cross-linking the RNA to the membranes (UV stratalinker 2400, stratagem Co., La Jolla, CA, USA), the blots were hybridized, washed, and detected whit the Gene Images CDP-star detection module (Amersham Pharmacia Biotech Inc. Piscataway, NJ, USA) using anti-fluorescein alkaline phosphatase conjugated and CDP-star detection reagent (Amersham Pharmacia Biotech Inc. Piscataway, NJ, USA) and followed by 3 h of exposure.

The T-student analysis was performed to evaluate the significant differences (P ≤ 0.05) in the expression levels of the diabetic kidney against the healthy kidney (JMP 5.0.1., software). The differential expression analysis consisted of the comparison of band intensity for each gene. The confirmation of the results was validated by Northern blot analysis. The normalization was based on the ratio between the gene of each diabetic control absolute values followed by the ratio of healthy control relative value expressed as fold of change in Tables 2 and 3. The cut off value to assign genes that increase/decrease expression was based on the relative value 1.0 produced by the normalization ratios.