Inbred Wistar rats were bred under specific pathogen-free conditions at the Institute for Animal Experimentation of the Kagawa University School of Medicine Kagawa, Japan. Eight male Wistar rats from the ages of post-natal day 3, and post-natal week 1, 2, 4, 8 and 36 were used in this study. Eight fetal livers were also obtained from pregnant rats at day 20. We performed hepatectomy under ether anesthesia. For immunohistochemistry, small sections of the liver tissue were preserved in 10% formalin. The remaining sections of the liver tissue were rapidly placed in liquid nitrogen, transferred individually to pre-weighed tubes containing RNAlater (Ambion, Tokyo, Japan) and stored at −80°C until analysis.

The frozen tissue samples were thawed on ice and dissolved in TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA, USA). Total RNA was extracted from the tissues using an miRNeasy Mini kit (Qiagen, Tokyo, Japan) according to the manufacturer’s instructions. Briefly, 700 μl of TRIzol reagent containing samples was mixed vigorously with 140 μl of chloroform, incubated at room temperature for 3 min, followed by centrifugation at 12,000 × g for 15 min at 4°C. The upper aqueous phase was transferred to another tube containing 525 μl of 100% ethanol. The mixture was loaded into RNeasy Mini columns, followed by serial washing with solutions provided with the kit. Finally, RNA was collected into RNase-free water for further experiments. All RNA samples used in this study showed A260/280 ratios between 2.0 and 2.1. The integrity of RNA was determined using a NanoDrop 2000 spectrophotometer (Thermo Scientific, Rockford, IL, USA). The quality of total RNA was determined using the RNA Nano 6000 chips on an Agilent 2100 Bioanalyzer (Agilent Technologies Inc., Osaka, Japan), and all RNA samples used for microarray analyses had RIN values >8.2. These RNA samples were stored at −80°C.

Total RNA was labeled with Hy3 dye using the array labeling kit (Exiqon, Vedbæk, Denmark). Total RNA (2 μg) was incubated with a spike for 30 min at 37°C and then at 95°C for 5 min. Hy3 dye and labeling enzyme were added to the samples. The enzyme was then heat-inactivated at 16°C for 1 h and 65°C for 15 min, protected from light and then received an addition of hybridization buffer. The sample was loaded onto the arrays by capillary force using 3D-Gene miRNA oligo chips (version 16; Toray Industries, Inc., Kanagawa, Japan). The chips enabled the examination of the expression of 679 miRNAs printed in duplicate spots. The arrays were incubated at 32°C for 16 h, then briefly washed in a 30°C wash buffer solution (0.5X SSC, 0.1% SDS), rinsed in wash buffer solution (0.2X SSC, 0.1% SDS) followed by a wash in another buffer solution (0.05X SSC), according to the manufacturer’s instructions (Toray Industries, Inc.). The arrays were spun for 1 min at 1,000 rpm for drying, followed by immediate scanning using a Toray 3D-Gene 3000 miRNA microarray scanner (Toray Industries, Inc.). The relative expression level of each miRNA was calculated by comparing the average signal intensities of the valid spots with their mean value throughout the microarray experiments following normalization to their adjusted median values.

To illustrate the alterations in the expression levels of the 13 upregulated or downregulated miRNAs, we created a heatmap in which each cell represents the expression level of each of the 13 miRNAs for each of 8 individual rats at different stages of development, namely the fetal period, at post-natal day 3 and post-natal week 1, 2, 4, 8 and 36. The heatmap was color-coded according to the log2-transformed expression level. The center level of the color code is set as the median value over all of the values used in the heatmap. Briefly, in the heatmap, white color represents mean values, red indicates gains and blue represents losses.

The avidin-biotin complex (ABC) immunohistochemical method was used. Following formalin fixation, the tissues were dehydrated in a graded series of ethanol and embedded in paraffin. The serial sections were mounted onto glass slides. Immunohistochemical staining for proliferating cell nuclear antigen (PCNA) was performed using the following procedure: the sections were deparaffinized, hydrated and quenched for endogenous peroxidase with 0.3% hydrogen peroxide in PBS at room temperature for 30 min. The sections were permeabilized in PBS with 0.3% polyoxyethylene (10) octyl phenyl ether (Triton X-100) (Wako, Osaka, Japan) for 1 h before quenching the endogenous peroxidase activity. The sections were then incubated overnight at room temperature with a mouse anti-PCNA antibody (1:100 dilution; Dako, Glostrup, Denmark). Antibody binding was detected using a Vectastain Elite ABC kit (Vector Laboratories, Burlingame, CA, USA) and 3,3′-diaminobenzidine as a chromogen. After staining, all sections were counterstained with Mayer’s hematoxylin. The specificity of immunostaining was examined using non-immune mouse IgG as a negative control for the primary antibody. Images were captured using an Olympus BX51 microscope and Olympus DP72 camera (Olympus, Tokyo, Japan). The nuclear labeling index for the PCNA-positive cells (positive nuclei/total counted) was determined by evaluating at least 500 hepatocytes at random in the microscopic field by 2 observers (T. Masaki and S. Mimura).

Frozen liver tissue samples were homogenized in a protein extraction solution (PRO-PERP™; Intron Biotechnology, Inc., Gyeonggi, Korea), and cell lysis was then induced by incubating the samples for 20 min on ice. The samples were prepared by centrifugation at 13,000 rpm for 5 min at 4°C. The supernatant was then collected. The protein concentration was measured by a dye-binding protein assay performed according to the Bradford method (12). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed according to the method of Laemmli (13), and western blot analysis was performed as previously described by Towbin et al (14) using anti-β-actin monoclonal antibody (Sigma-Aldrich; A5442, used at 1:3,000) and cyclin D1 (Thermo Fisher Scientific, Waltham, MA, USA; RB-9041, used at 1:1,000) as primary antibodies and horseradish peroxidase-inked anti-mouse and anti-rabbit IgG secondary antibodies (GE Healthcare UK, Buckinghamshire, UK; used at 1:2,000).

Total RNA (20 μl) from all liver tissues used in this study was loaded into each lane, and the bands of 18S and 28S ribosomal RNA in the gel were detected using the 2100 Bioanalyzer (Fig. 1). Based on these results, RNA samples extracted from various liver tissues were shown to be of adequate quality.

Using a custom microarray platform, we analyzed the expression levels of 679 rat miRNA probes. Four miRNAs were found to be gradually upregulated with age in the rat livers. The 4 upregulated miRNAs were miR-29a (spot no. 1), miR-29c (spot no. 2), miR-195 (spot no. 3) and miR-497 (spot no. 4) (Figs. 2 and 3). By contrast, 9 miRNAs were gradually downregulated. These downregulated miRNAs were miR-301a (spot no. 5), miR-148b-3p (spot no. 6), miR-7a (spot no. 7), miR-93 (spot no. 8), miR-106b (spot no. 9), miR-185 (spot no. 10), miR-450a (spot no. 11), miR-539 (spot no. 12) and miR-301b (spot no. 13). The heatmap (Fig. 4) clearly demonstrates increasing trends for 4 miRNAs and decreasing trends for 9 miRNAs in the expression levels among the age groups. It also showed that the trends in miRNA expression were fairly consistent among individual rats and were constant in both increasing and decreasing directions.

The typical immunohistochemical staining pattern of PCNA at various stages of development in the rat livers, including fetal liver (Fig. 5A), post-natal day 3 liver (Fig. 5B), post-natal week 1 liver (Fig. 5C), post-natal week 2 liver (Fig. 5D), postnatal week 4 liver (Fig. 5E), post-natal week 8 liver (Fig. 5F) and post-natal week 36 liver (Fig. 5G). The labeling index of PCNA in the hepatocytes in the developing livers decreased with age (Fig. 5H).

Western blot analysis was used to examine the cell cycle of the aging rat liver by using an antibody against cyclin D1. The cyclin D1 level was hardly detectable in the fetal liver (Fig. 6). However, cyclin D1 was detected in the liver at post-natal day 3. The highest level was detected in the post-natal week 1 liver and the level slightly decreased with age. The amount of β-actin (an internal control for protein loading) was almost the same in all age groups.