Introduction

Liver is one of the most important organs in material metabolism (Michalopoulos and DeFrances 2005). After partial hepatectomy (PH) (Higgins and Anderson 1931), the remnant hepatocytes are induced from quiescent state into rapid growth state for compensatory proliferation, which is called liver regeneration (LR) (Fausto et al. 2006; Xu et al. 2004). LR involves many processes such as cellular activation, cell proliferation and its regulation, structure-function reorganization and cell metabolism, etc. (Pahlavan et al. 2006). Presumably that organic acid metabolism is also necessary in the above process. Studies show that metabolism including organic acids metabolism comprises a set of sophisticated metabolic pathways, which are series of consecutive enzymatic reactions that produce specific products (Chou et al. 2006). Enzymes, namely proteins, play a key role in the above enzymatic reactions, and proteins, as the embodiment of life, involved many biochemical events, such as different inherent rule of codon usage in proteins may represent different biological implications (Zhang and Chou 1993, 1994, 1996); versatile architectures have made proteins possess varieties of functions (Chou 2005). Previous studies have showed that citrate synthase, isocitrate dehydrogenase 1, succinate-CoA ligase (GDP-forming) alpha subunit, fumarate hydratase and malate dehydrogenase promote biosynthesis of succinate, malate, citrate, isocitrate, fumarate and pyruvate in the citric acid cycle, respectively. Catalysis of guanidinoacetate methyltransferase results in synthesis of creatine. Quinolinate was synthesized under the catalytic effects of kynurenine 3-m onooxygenase, kynureninase (Alberati-Giani et al. 1996) and 3-hydroxyanthranilate 3, 4-dioxygenase. Butyrobetaine (gamma), 2-oxoglutarate dioxygenase 1 promotes carnitine biosynthesis. N-acetylneuraminate was catalyzed by CMP-N-acetylneuraminic acid synthetase. Synthesis of propionate was catalyzed by acetyl-CoA carboxylase alpha and acetyl-CoA dehydrogenase, medium chain, etc. Catalytic synthesis of Pyruvate was supported by pyruvate kinase (liver and RBC), lactate dehydrogenase A-like 6B (Tsuji et al. 1994), etc. Many diseases, such as Glutaric aciduria, Glutaryl-CoA dehydrogenase deficiency, are attributable to the disturbance of organic acid metabolism (Kolker et al. 2004; Varadkar and Surtees 2004). To investigate these biochemical events in rat LR at transcriptional level (Xu et al. 2005; Dransfeld et al. 2005), we employed the Rat Genome 230 2.0 array (Collins 2006) embracing the 106 genes related to organic acid metabolism to check genes expression profiles in regenerating liver after 2/3 hepatectomy based on querying organic acid metabolism-associated genes using the bioinformatics tools (Chou 2004; Cai and Chou 2005; Xu and Chang 2007; Shen and Chou 2007), and found that 55 genes is LR-associated. Their expression changes and correlation with LR as well as function were primarily analyzed (Yue et al. 2001).

Materials and methods

Regenerating liver preparation

Healthy SD rats (200–250 g) were obtained from the Animal Center of Henan Normal University. The 264 rats were separated into groups randomly and one group included six rats. PH was performed according to Higgins and Anderson (1931), the left and middle lobes of liver were removed. Rats were killed by cervical vertebra dislocation at 0.5, 1, 2, 4, 6, 8, 12, 16, 18, 24, 30, 36, 42, 48, 54, 60, 66, 72, 96, 120, 144 and 168 h after PH and the regenerating livers were observed at corresponding time point. The livers were wished three times in PBS at 4°C, then 100–200 mg livers from middle parts of right lobe (total mass: 0.1–0.2 g × 6 = 0.6–1.2 g) were gathered and stored at −80°C until use. The sham-operation (SO) groups were the same with PH ones but without liver removal. The laws of animal protection of China were enforced strictly.

RNA isolation and purification

Total RNA was isolated from frozen livers according to the manual of Trizol reagent (Invitrogen Corporation, Carlsbad, CA, USA) (Knepp et al. 2003) and then purified base on the guide of RNeasy mini kit (Qiagen Inc., Valencia, CA, USA) (Nuyts et al. 2001). Total RNA samples were checked to exhibit a 2:1 ratio of 28 S rRNA to 18 S rRNA intensities by agarose electrophoresis (180 V, 0.5 h). Total RNA concentration and purity were estimated by optical density measurements at 260/280 nm (Arkin et al. 1998).

cDNA, cRNA synthesis and purification

cDNA, cRNA purification were based on the way established by Affymetrix. 1–8 μg total RNA as template was used for cDNA synthesis. cRNA labeled with biotin was synthesized utilizing 12 μl cDNA as the template, and cDNA and cRNA were purified according to the purification procedure of GeneChip Analysis (Li et al. 2005). Measurement of cDNA, cRNA concentration and purity were the same as above (Arkin et al. 1998).

cRNA fragmentation and microarray detection

15 μl (1 μg/μl) cRNA incubated with 6 μl 5 × fragmentation buffer at 94°C for 35 min was digested into 35–200 bp fragments. The hybridization buffer prepared according to the way Affymetrix provided was added to the prehybridized Rat Genome 230 2.0 microarray produced by Affymetrix, then hybridization was carried out at 45°C for 16 h on a rotary mixer at 60 rpm. The microarray was washed and stained by GeneChip fluidics station 450 (Affymetrix Inc., Santa Clara, CA, USA). The chips were scanned by GeneChip Scan 3000 (Affymetrix Inc.), and the signal values of gene expression were observed (Collins 2006).

Microarray data analysis

The normalized signal values, signal detections (P, A, M) and experiment/control (Ri) were obtained by quantifying and normalizing the signal values using GCOS (GeneChip operating software) 1.2 (Collins 2006).

Normalization of the microarray data

To minimize the technical error from the microarray analysis, each sample was hybridized three times to the gene chips. The average value of three measurements was normalized, and statistics and cluster analyses were conducted on these values with GeneMath, GeneSpring (Silicon Genetics, San Carlos, CA, USA) and Microsoft Excel Software (Microsoft, Redmond, WA, USA) (Collins 2006; Eisen et al. 1998; Werner 2001).

Identification of genes associated with liver regeneration

The nomenclatures of organic acid biosynthesis or catabolism were adopted from the GENEONTOLOGY database (http://www.geneontology.org), and inputted into at NCBI (http://www.ncbi.nlm.nih.gov) and RGD (http://www.rgd.mcw.edu) to identify the rat, mouse and human genes associated with organic acid metabolism. In addition, according to maps of biological pathways embodied by GENMAPP (http://www.genmapp.org), KEGG (http://www.genome.jp/kegg/pathway.html) and BIOCARTA (http://www.biocarta.com/genes/index.asp), the genes associated with organic acid metabolism were collected, and reconfirmed through literature searches of the pertinent articles. Besides the rat genes, the genes which were now thought to only exist in mouse and/or human and which exhibited a greater than twofold change in the rat regenerating livers were referred to as rat homologous genes. The genes that displayed reproducible results with the three independent analyses and that showed a greater than twofold change in expression in at least one time point as a significant difference (P ≤ 0.05) or an extremely significant difference (P ≤ 0.01) between PH and SO, were considered being associated with LR.

Results

Expression changes of the genes associated with organic acid metabolism during liver regeneration

According to the data of databases at NCBI, GENMAPP, KEGG, BIOCARTA and RGD, organic acid metabolism involves 124 genes, and among them, 106 genes were contained in the Rat Genome 230 2.0 array. Fifty-five of them displayed meaningful changes in expression at least in one time point post-PHx, showed significant or extremely significant differences in expression when comparing PH with SO, and exhibited the reproducible result with three independent analysis with Rat Genome 230 2.0 array, suggesting that the genes were associated with LR. It was indicated that there were 23 up-regulated genes, 16 down-regulated genes and 16 up/down-regulated genes during LR. The range of up-regulation was from 2-fold to 30-fold compared with control, and that of down-regulation was 2-fold to 222-fold (Table 1).

Table 1 Expression abundance of 55 organic acid metabolism-associated genes during rat liver regeneration
Full size table

Initial expression and total expression profiles of the genes associated with organic acid metabolism during liver regeneration

At each time point of LR, the numbers of initially up-, down-regulated and totally up-, down-regulated genes were in sequence: both 7 and 3 at 0.5 h; 3, 2 and 9, 4 at 1 h; 2, 2 and 6, 5 at 2 h; 0, 2 and 6, 4 at 4 h; 0, 3 and 3, 7 at 6 h; 1, 4 and 4, 12 at 8 h; 0, 4 and 3, 13 at 12 h; 4, 6 and 5, 12 at 16 h; 2, 0 and 7, 7 at 18 h; 0, 0 and 9, 5 at 24 h; 1, 1 and 8, 5 at 30 h; 1, 0 and 7, 9 at 36 h; 2, 1 and 12, 5 at 42 h; 0, 1 and 8, 9 at 48 h; 0, 0 and 6, 1 at 54 h; 0, 0 and 6, 0 at 60 h; 2, 0 and 11, 0 at 66 h; 0, 0 and 6, 0 at 72 h; 1, 0 and 7, 4 at 96 h; 0, 0 and 6, 3 at 120 h; 0, 0 and 8, 2 at 144 h; 0, 0 and 7, 1 at 168 h (Fig. 1). There were totally 26 initially up-regulated and 29 initially down-regulated genes during LR, i.e. at the initiation phase of LR (0.5–4 h after PH), G0/G1 transition (4–6 h after PH), cell proliferation (6–66 h after PH), cell differentiation and structure-function reconstruction (72–168 h after PH), the number of initially up and initially down-regulated genes were 12 and 9, 0 and 5, 13 and 20, 1 and 0, respectively. The total number of the up- and down-regulated genes in LR was 151 and 114, and expression frequencies of total 55 genes in the above four phases of LR were 28 and 16, 9 and 11, 89 and 85, 34 and 10, respectively (Fig. 1).

Fig. 1
figure 1

The initial and total expression profiles of 55 genes associated with organic acid metabolism at each time point of liver regeneration. X-axis: Recovery time after partial hepatectomy, Y-axis: The number of the genes with the measurable expression changes of every time point. Grey bars: Up-regulated gene; White bars: Down-regulated. Blank bars: Initial expressing genes, Dotted bars: The total expressing genes. Gene expression changes span the whole liver regeneration. Initial up-regulated genes dominate at 0.5–2, 18, 30–42, 66 and 96 h, initial down-regulated genes predominate at 4–16 and 48 h, no genes display initial expression at other time points

Full size image

Expression patterns of the genes associated with organic acid metabolism during liver regeneration

According to the function and expression profiles of gene during LR, classification of the expression patterns of total 55 genes into 14 types (Fig. 2) are as follows. (1) acetate biosynthesis, involving one up-regulated gene two down-, and three up/down-; (2) propionate biosynthesis, three up- and two down-; (3) propionate catabolism, three up-regulated, three down-, and one up/down-; (4) pyruvate biosynthesis, seven up-regulated and two up/down-; (5) pyruvate catabolism, one up-regulated and four down-; (6) succinate biosynthesis, two up-regulated; (7) malate biosynthesis, three up-regulated; (8) citrate biosynthesis, two up-regulated and two down-; (9) isocitrate biosynthesis, one up-regulated, one down- and one up/down-; (10) fumarate biosynthesis, one down- and two up/down-; (11) creatine biosynthesis, one down-regulated and one up/down-; (12) carnitine biosynthesis, one up-regulated, two down- and five up/down-; (13) N-acetylneuraminate biosynthesis, four up-regulated and one down-; (14) quinolinate biosynthesis, two down-regulated and one up/down-regulated (Fig. 2).

Fig. 2
figure 2

Expression patterns of 55 genes associated with organic acid metabolism during LR Expression patterns were obtained by the analysis of detection data of Rat Genome 230 2.0 array with Microsoft Excel. 1 acetate biosynthesis; 2,3 propionate metabolism; 4,5 pyruvate metabolism; 6 succinate biosynthesis; 7 malate biosynthesis; 8 citrate biosynthesis; 9 isocitrate biosynthesis; 10 fumarate biosythesis; 11 creatine biosynthesis; 12 carnitine biosynthesis; 13 N-acetylneuraminate biosynthesis; 14 Quinolinate biosynthesis. Real lines: the up-regulated genes; broken lines: the down-regulated genes; real-broken lines: the up/down-regulated genes. X-axis indicates the recovery time (h) after PH; Y-axis shows logarithm ratio of the signal values of genes at each time point to control

Full size image

Discussion

Organic acids, the intermediate metabolites of many substances, have significant biological actions. Among acetate biosynthesis-promoting genes, aspa was up-regulated at 0.5–4, 48 h (Namboodiri et al. 2000); aldh1a4 at 0.5–4, 8, 24 and 48 h, aldh1b1 at 4, 8 and 18–24 h, cach at 8–18 and 36 h and acas2l at 48 h were down-regulated, while acyp1 was up/down-regulated. As for the genes contributing to propionate biosynthesis, the expression levels of three genes acaca (Zhang and Kim 1996), abat and acadm mRNA were increased, however, hadha showed downtrend at 0.5, 36 h. dhcr24 and slc27a2, the genes accelerating propionate degradation, displayed down-regulation and up/down-regulation in LR. It is inferred from expression changes of the above genes that there exhibit an increase in acetate biosynthesis in the forepart, propionate biosynthesis in the forepart, prophase and early metaphase, degradation of propionate were blocked middle metaphase.

Among the genes positively regulating pyruvate biosynthesis, expect for down-regulation of genes hagh at 8–12 h and idhb at 16 h, ten genes including pklr were up-regulated at the interval of 0.5–6 and 18–168 h (Wang et al. 2002). In addition, Multi-time down-regulation of five pyruvate catabolism-prompting genes such as pc during 8–48 h (Jitrapakdee et al. 1996); of three genes enhancing succinate biosynthesis, suclg1 being down-regulated at 16, 30, 42 and 96 h, sucla2 at 18 h and aldh5a1 at 0.5–6, 54–72 and 120–168 h (Mehta et al. 2006) being up-regulated; genes me1, me2 and mor1, whose expression magnified at 42 h, at 18–36 h and at 30, 42 and 96 h (Chow et al. 2005), respectively, belong to the malate biosynthesis-accelerating genes; two citrate biosynthesis-promoting genes cs and aco1 were in a sequence expressed significantly increased at 16 h and decreased at 12 h; isocitrate biosynthesis-enhancing genes aco1 12 h and idh1 at 4–12 h both exhibited the low levels in expression (down-regulated more than twofold); down-regulation of the fumarate biosynthesis-promoting gene fh at 16, 36, 48 and 120–144 h (Coustou et al. 2006), which speculates that pyruvate biosynthesis in the forepart, metaphase and anaphase; degradation of pyruvate was blocked at metaphase; succinate biosynthesis in the forepart and anaphase; malate biosynthesis in the metaphase. Whereas, isocitrate biosynthsis shows an decrease in the forepart and fumarate at anaphase. Noticeably, the increased expression of genes ldhal6b and mor1 spanned almost during the entire LR, and reached peaks with 29.9- and 8.9-fold of the control at 18 h and at 42 h, respectively, which indicate that the two genes play key roles separately in pyruvate biosynthesis and malate biosynthesis in corresponding time of LR.

Both gamt and ckb, the creatine biosynthesis-promoting genes, were down-regulated at 12–18 h (Braissant et al. 2005; Shen et al. 2002), eight genes including bbox1, the carnitine biosynthesis-enhancing genes, had multiple-time down-regulations within the period of 0.5–12 h, while trended up at 24–48 h and took on a tendency to change into zigzag form; genes cmas, nt5m, phpt1, mtmr6 and mtmr7 have a promotive effect on N-acetylneuraminate biosynthesis, expect for downtrend of cmas (Tanner et al. 2005) in expression change, the latter four genes showed down-regulation; three genes such as kmo, etc., playing the role in quinolinate biosynthesis, were decreased in expression level at 6–24 h (Ito et al. 2004), presuming creatine biosynthesis attenuated at early metaphase; carnitine biosynthsis shows an decrease in the forepart and prophase and enhancement at middle metaphase; N-acetylneuraminate biosynthesis swelled at 36, 66 and 96 h; quinolinate biosynthesis weaken at forepart and early metaphase. It is well known that carnitine is helpful for transportation of energy substance required for cell proliferation (Lai et al. 1998), its biosynthesis showed enhancement at middle metaphase, which might support energy for cell proliferation.

In summary, the treatment of experimental material in this study is marked with long times and multiple time points, and expression changes of genes related to organic acid metabolism after rat PH were detected with high-throughput gene expression analysis at transcriptional level. It was primarily proved that there exhibit an increase in acetate biosynthesis in the forepart, propionate biosynthesis in the forepart, prophase and early metaphase, pyruvate biosynthesis in the forepart, metaphase and anaphase, succinate biosynthesis in the forepart and anaphase; malate biosynthesis in the metaphase and N-acetylneuraminate biosynthesis at 36, 66 and 96 h. Whereas, carnitine biosynthsis shows an decrease in the forepart and prophase and enhancement at middle metaphase; at the same time, isocitrate in the forepart, quinolinate at forepart and early metaphase, and creatine at early metaphase perform the restrained biosynthesis, respectively; in addition, degradation of pyruvate and propionate were blocked at metaphase. However, DNA → mRNA → protein → function are influenced by many factors including protein interaction. Therefore, the above results need to be further analyzed with the techniques, such as Northern blotting, protein chip, RNA interference, protein-interaction, etc.