Over-expression of the ribosomal protein L36a gene is associated with cellular proliferation in hepatocellular carcinoma

Authors

  • Jong-Hyun Kim,

    1. Department of Internal Medicine, Division of Gastroenterology and Hepatology, Institute for Medical Science, Chonbuk National University Medical School and Hospital, Chonju, Chonbuk, South Korea
    Search for more papers by this author
  • Kyung-Ran You,

    1. Department of Internal Medicine, Division of Gastroenterology and Hepatology, Institute for Medical Science, Chonbuk National University Medical School and Hospital, Chonju, Chonbuk, South Korea
    Search for more papers by this author
  • In Hee Kim,

    1. Department of Internal Medicine, Division of Gastroenterology and Hepatology, Institute for Medical Science, Chonbuk National University Medical School and Hospital, Chonju, Chonbuk, South Korea
    Search for more papers by this author
  • Baik-Hwan Cho,

    1. Department of Surgery, Division of Gastroenterology and Hepatology, Institute for Medical Science, Chonbuk National University Medical School and Hospital, Chonju, Chonbuk, South Korea
    Search for more papers by this author
  • Chan-Young Kim,

    1. Department of Surgery, Division of Gastroenterology and Hepatology, Institute for Medical Science, Chonbuk National University Medical School and Hospital, Chonju, Chonbuk, South Korea
    Search for more papers by this author
  • Dae-Ghon Kim

    Corresponding author
    1. Department of Internal Medicine, Division of Gastroenterology and Hepatology, Institute for Medical Science, Chonbuk National University Medical School and Hospital, Chonju, Chonbuk, South Korea
    • Division of Gastroenterology and Hepatology, Department of Internal Medicine, The Research Institute for Medical Science, Chonbuk National University Medical School and Hospital, 634-18 Keumam-dong, Dukjin-ku, Chonju, Chonbuk 561-172, South Korea
    Search for more papers by this author
    • fax: +82-63-254-1609


Abstract

Using messenger RNA (mRNA) differential display, we identified a single complementary DNA (cDNA) fragment (HG23T1) that was over-expressed in a hepatocellular carcinoma (HCC) specimen. We cloned the full-length HG23T1 gene by the rapid amplification of cDNA end (RACE) polymerase chain reaction (PCR) method. It perfectly matched the gene encoding human ribosomal protein L36a (RPL36A also referred to as RPL44). RPL36A mRNA was preferentially over-expressed in 34 of 40 HCC cases (85%, P < .001) and in all of 8 HCC cell lines. Ectopically over-expressed L36a ribosomal protein localized in the nucleoli of cells, and this localization seemed to be controlled by the N-terminal or the internal tetrapeptide consensus with its adjacent N-terminal domain. Over-expression of L36a led to enhanced colony formation and cell proliferation, which may have resulted from rapid cell cycling, and an antisense cDNA effectively reversed these alterations. In conclusion, RPL36A plays a role in tumor cell proliferation and may be a potential target for anticancer therapy of HCC. (HEPATOLOGY 2004;39:129–138.)

Hepatocellular carcinoma (HCC) is a major malignancy with an increasing incidence worldwide.1 The major risk factors for development of HCC are cirrhosis, chronic viral hepatitis B and C, exposure to aflatoxin, alcohol, and/or iron overload. HCCs are highly resistant to chemotherapeutic agents and radiotherapy.2 Despite a variety of treatment options, including surgical resection, chemoembolization, percutaneous injection of ethanol, radiofrequency thermal ablation, and liver transplantation, the prognosis for HCC is poor.3–5 Thus, the need to discover effective therapeutic molecules to suppress the growth of HCC is urgent. One approach is to find target genes and their molecular products and understanding how genetic and molecular changes can lead to cancer development through multistep carcinogenesis.

Altered gene expression caused either by mutations or by changes in the regulatory characteristics of HCC compared with corresponding nontumor tissues have been reported.6–8 Such alterations can occur in various cell cycle-related oncogenes, such as c-myc, K-ras, and H-ras, and/or in tumor suppressor genes, such as retinoblastoma and p53.9–12 In addition, glutamine synthetase, alfa-fetoprotein (AFP), the insulin receptor substrate-1-like gene, cyclin D1, MAGE-1, HIP/PAP, and CD24 have been reported to be over-expressed during hepatocarcinogenesis.13–19 However, these genetic changes do not precisely reflect the biologic nature or clinical characteristics of all HCCs. Previously, we identified and reported genes that were differentially expressed in HCC cells using the differential-display polymerase chain reaction (PCR) method.20 Among them, we focused preferentially on an up-regulated gene in human HCC in comparison with the nontumor surrounding tissues. This gene later proved to encode the ribosomal protein L36a (RPL36A) protein, which is localized in the nucleolus.

The biogenesis of eukaryotic ribosomes occurs within a specific subnuclear compartment, the nucleolus, and requires the coordinated assembly of 4 different rRNAs and approximately 80 ribosomal proteins.21, 22 The 5.8S, 18S, and 28S rRNAs are synthesized by RNA polymerase I in the nucleolus, while the 5S rRNA is transcribed by RNA polymerase III in the nucleoplasm. In contrast, ribosomal proteins encoded by mRNAs are synthesized by RNA polymerase II and their proteins are imported from the cytoplasm into the nucleolus for assembly of 40S and 60S ribosomes after translation. Enhanced expression of ribosomal protein mRNAs, including acidic ribosomal phosphoproteins P0, S8, S19, L12, L23a, L27, L30, and L9, have been reported in HCC.23–25 However, over-expression of RPL36A and its implications have not yet been characterized. Thus, in this study, we report that RPL36A is over-expressed in HCC and that its functional role may be related to tumor cell proliferation.

Abbreviations:

HCC, hepatocellular carcinoma; AFP, alfa-fetoprotein; PCR, polymerase chain reaction; cDNA, complementary DNA; RACE, rapid amplification of cDNA end; PBS, phosphate-buffered saline; GFP, green fluorescent protein; NLS, nuclear localization signals.

Materials and Methods

Cloning of the L36A Ribosomal Protein Gene by 5′ RACE

Based on the 3′ untranslated region sequence of a single complementary (cDNA) fragment (HG23T1),20 an antisense primer (5′-GCAGTTGGGCTCAACGCACTCAAGCC-3′) was synthesized. Rapid amplification of the 5′ cDNA end (5′-RACE) was performed using the antisense primer following the protocol supplied with the Marathon cDNA Amplification kit (Clontech Laboratories, Palo Alto, CA). Based on the sequence of the 5′-RACE fragment, a sense primer (5′-CGCGTCGACGCGCTCACGCAAGCATG-3′) was synthesized and 3′-RACE was performed to obtain the full-length cDNA. The RACE product was cloned in the TA vector (Promega, Madison, WI), and the entire cDNA was sequenced. Nucleotide and predicted amino acid sequence searches were performed using BLAST.

Serologic Assays and Tissue Acquisition

Forty patients were selected and tested for hepatitis B virus or hepatitis C virus markers in their sera using a Cobra Core enzyme immunoassay kit (F. Hoffmann-La Roche, Basel, Switzerland). The level of AFP was quantitatively determined by a Microparticle Enzyme Immunoassay (Abbott Laboratories, North Chicago, IL). Primary HCCs and surrounding nontumor liver tissues were obtained from surgical resections of primary HCC. Written informed consent was obtained from each patient. HCC and nontumor tissues were histologically confirmed by pathologists. After resection, the tissues were rinsed in sterile phosphate-buffered saline (PBS) and were immediately stored in liquid nitrogen until use for RNA isolation. Total cellular RNA was extracted using the Tri-zole kit (MRC, Cincinnati, OH). This protocol conformed to the ethical guidelines of the Institutional Review Board (IRB).

Northern Blot Analysis

Samples containing 20 μg of total nontumor or tumor RNAs were fractionated on 1% agarose gels containing 2.2% formaldehyde and 50 mM 3-[N-morpholino]propanesulfonic acid and transferred to a nylon membrane. The membrane was cross-linked using an ultraviolet cross-linker (Stratagene, La Jolla, CA) and was individually hybridized with each cDNA probe generated from a digested cDNA insert labeled with α-32P dCTP (3,000 Ci/mmol; NEN Life Science Products, Boston, MA) by random priming. Hybridization and washing conditions were the same as described previously,20 and the blots were then exposed to x-ray film at −70°C. To confirm the amounts of mRNA loaded in each lane, the blots were hybridized afterward with an 18S ribosomal protein gene cDNA probe. Autoradiographs of Northern hybridizations were scanned using a Personal Densitometer SI (Molecular Dynamics, Sunnyvale, CA). Expression levels of each gene were normalized against the level of 18S RNA and the hybridized signals were then calculated as fold induction from nontumor or HCC tissues, or vice versa.

Southern Blot Analysis

Genomic DNA was extracted from each HCC sample by proteinase K digestion. Ten μg of each DNA sample was digested with EcoRI overnight at 37°C. The restriction fragments were separated by electrophoresis on a 0.7% agarose gel and then transferred onto a nylon membrane in 0.4 N NaOH. The membranes were prehybridized for 2 h and then hybridized with 32P-labeled RPL36A cDNA in hybridization buffer at 65°C overnight. Hybridization and washing conditions were the same as described previously.20 The blots were then exposed to X-ray film at −70°C.

Cellular Localization of the Functional RPL36A Protein

We constructed a RPL36A eukaryotic expression vector producing a fusion protein of ribosomal protein L36a with an enhanced version of green fluorescent protein (GFP) of Aequorea victoria at the amino terminal of the ribosomal protein L36a. For the GFP-RPL36A fusion expression vector, the human RPL36A gene was PCR amplified with a forward primer containing an EcoRI restriction enzyme site (5′-CGGAATTCCGCTCACGCAAGCATGGT-3′ and a reverse primer containing a KpnI restriction enzyme site (5′-CGGGGTACCCACTTAGAACTGGATCA-3′). The deletion constructs L36a(1–96), L36a(1–85), L36a(1–43), L36a(1–30), L36a(1–26), L36a(31–57), L36a(31–61), L36a(31–85), L36a(43–85), L36a(86–106), L36a(43–106), and L36a(31–106) were PCR amplified using a primer set containing an EcoRI site at the 5′ end and a KpnI site at the 3′ end. PCR was used to amplify the human RPL36A cDNA and the deletion constructs were ligated to pEGFP-C1 (Clontech) at EcoRI/KpnI sites in frame. We transfected Hep 3B, Chang liver, or NIH 3T3 cells with the expression vector (pEGFP-RPL36A). The cells were stained 48 h later with Hoechst 33258 and the transfected cells were examined using a laser-scanning microscope LCM 510 (Carl Zeiss, Jena, Germany). The fluorescent GFP-RPL36A fusion protein was localized in the nucleus, and accumulated mostly in the nucleoli. Cells with nucleolar localization of fluorescence were predominant in number at 48 h after transfection.

Cell Culture and Transfection

Mouse NIH/3T3, MCF10A, Chang liver, and human HCC cell lines, including Hep 3B, PLC/PRF/5, Hep G2, SK-HEP-1, and Huh-7, were all obtained from the American Type Culture Collection (ATCC, Manassas, VA). Primary HCC cell lines, including HKK-1, HLK-1, and SH-J1 cells,26 were isolated and their tumorigenicity was confirmed by xenotransplanting to nude mice. The cells were maintained in Dulbecco's modified Eagle medium supplemented with 10% fetal bovine serum at 37°C in a humidified atmosphere of 5% CO2 in air. For DNA transfection, a 384-base pair (bp) PCR-amplified fragment of RPL36A cDNA was ligated into the pEGFP-C1 (Clontech) mammalian expression vector. Lipofectin (Gibco BRL, Grand Island, NY) was used to transfect Chang liver cells. Several clones each of cells transfected with RPL36A and control cells were established in the presence of 600 μg/ml geneticin (G418) and were further characterized. Doubling times were determined from the growth curves.

Colony Generation Assay

We constructed the RPL36A eukaryotic expression vector as follows. The human RPL36A gene was PCR amplified with a forward primer containing an EcoRI restriction enzyme site (5′-CGGAATTCCGCTCACGCAAGCATGG-3′) and a reverse primer containing an XbaI restriction enzyme site (5′-CGAGATCTCGTGACATTTAGAACTGGAT-3′) and the PCR amplified human RPL36A DNA was ligated to pcDNA3.1/HisC (Invitrogen, Carlsbad, CA) at the EcoRI/XbaI site in frame in the sense orientation. For the expression vector in the antisense orientation, the human RPL36A gene was PCR amplified with a forward primer containing an EcoRI restriction enzyme site (5′-CGGAATTCGCTCACGCAAGCATGG-3′) and a reverse primer containing a KpnI restriction enzyme site (5′-CGGGGTACCCACTTAGAACTGGATCA-3′) and the PCR amplified human RPL36A DNA was ligated to pcDNA3 (Invitrogen) at the EcoRI/KpnI site. We transfected Hep 3B and Chang liver cells with those expression vectors. One day after transfection, cells were seeded in 10-cm Petri dishes and incubated in the presence of G418 for 3 wk. The colonies were visualized by washing the plates twice with PBS and incubating them for 1 h in fixative (10% [v/v] methanol, 10% [v/v] acetic acid). After staining for 30 min with fixative containing 0.5% (w/v) crystal violet, the plates were washed 3 times with fixation buffer, 2 times with water, and then dried in air.

3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyl Tetrazolium Bromide Test

Viable cells were adjusted with medium to a concentration of 1 × 104 cells/ml and were plated in 24-well plates. Cells were incubated for 5 d, and cell proliferation was examined by measuring cell viability using the tetrazolium salt test described previously.27 Briefly, 50 μl 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (5 mg in 0.9% sodium chloride; Sigma, St. Louis, MO) was added and incubated for 4 h at 37°C. After dissolving the precipitated dye in 200 μl dimethyl sulfoxide (Sigma) and 50 μl glycine buffer (pH 10.5), the absorbance was read at 540 nm in an enzyme-linked immunosorbent assay reader (Molecular Devices SPECTRAmax 340).

Synchronization and Cell Cycle Analysis

Subconfluent cultures were trypsinized and plated at 2 × 105 cells per 60-mm culture dish. A thymidine-aphidicolin double-block was used to synchronize the cells at the G1/S interface. Briefly, 2 mM thymidine (Sigma) was added to the exponentially growing cells. The thymidine was washed out 12 h later, and media containing 25 μM thymidine and deoxycytidine (Sigma) was added. After 14 hr, 2 μg/ml aphidicolin (Sigma) was added and 12 hr later the cells were released from the aphidicolin block, as described previously.28 Cell cycle distribution was performed at the indicated intervals following the release. PI (40 μg/100 μl PBS) was added to 1 × 106 cells suspended in 800 μl PBS together with 100 μl RNase A (1 μg/ml) and incubated at 37°C for 30 min before flow cytometric analysis of 2 × 104 cells. Using the Becton-Dickenson FACScan flow cytometer (San Jose, CA), red fluorescence due to propidium iodide bound to DNA was measured. Data were analyzed using the CellFIT cell cycle analysis program (Version 2.01.2).

Data Analysis

Statistical analysis was performed by the Chi-Square test or by a test of nomial proportion(s) using a statistical package (Minitab, State College, PA). A P-value less than .05 is considered significant.

Results

Identification of the RP36A Gene and Tissue Distribution of Its Expression

Previously, we isolated a 250-bp cDNA of the 3′ untranslated region of a single cDNA fragment (HG23T1) that was preferentially up-regulated in HCC tissues, as identified by the differential display method.20 In the present study, we isolated and sequenced a 402-bp clone using the RACE PCR method. The full length of this cDNA had 100% identity with human RPL36A (or referred to RPL44) mRNA (accession no. NM_021029) (Fig. 1A). Radiolabeled human RPL36A cDNA was hybridized at high stringency to human multiple tissue Northern blots bearing 2 μg poly(A)+ RNA. Pancreas, prostate, small intestine, peripheral blood leukocytes, and ovary expressed high levels of RPL36A mRNA. Human liver tissue and colon also expressed it (Fig. 1B).

Figure 1.

Identification of RPL36A gene expression and its tissue distribution. (A) A full-length fragment of RPL36A cDNA obtained by RACE PCR. Underlined italic letters indicate the antisense primer used during the initial 5′ RACE PCR. Underlined bold letters represent the sense primer used during the second 3′ RACE PCR. (B) Tissue distribution of RPL36A. Human adult tissue poly(A)+ Northern blots were probed with 32P-labeled RPL36A cDNA. The blots were stripped and sequentially hybridized with a probe for GAPDH cDNA as a loading control.

Expression of RPL36A mRNA and RPL36A Gene Amplification in HCC and HCC Cell Lines

RPL36A mRNA was analyzed by Northern blot to examine its expression profile in 40 HCC samples that were grouped into hepatitis B virus serology positive or negative, and which were further grouped into high serum AFP level (more than 400 ng/ml) or low serum AFP level. A 0.5-kb mRNA band was detected by Northern hybridization in HCC specimens (Fig. 2A) and in HCC cell lines, including Hep 3B, PLC/PRF/5, Hep G2, SK-HEP-1, Huh-7, HKK-1, HLK-1, and SH-J1 (Fig. 2B). Among 40 HCC specimens, 34 cases (85%) had increased levels of RPL36A mRNA, while 6 cases were negative or had trace amounts. The over-expression of RPL36A mRNA observed in HCC was not statistically related to either hepatitis B virus serology or to serum level of AFP (Table 1). All 8 HCC lines showed a higher expression of RPL36A mRNA than did the immortalized non-tumor cells MCF10A, NIH3T3, or Chang liver cells. Next, in order to determine if DNA amplification is the cause of overexpression of RPL36A in HCC, Southern blot analysis was performed. In 7 representative cases of HCC, any detectable gene amplification was not observed, compared to that of the loading control gene (Fig. 2C). Thus, these results suggest that the enhanced expression of RPL36A mRNA results from the transcriptional regulation, not from the gene amplification in HCC.

Figure 2.

Expression of RPL36A mRNA in HCC samples and in HCC cell lines. (A) Northern blot analysis of RPL36A mRNA in 13 representative HCC tissues. RPL36A is over-expressed in all of 13 HCC (T), compared to paired nontumor tissues (N). Twenty μg of nontumor or tumor total RNAs were fractionated on 1% agarose gels containing 2.2% formaldehyde and 50 mM 3-[N-morpholino]propanesulfonic acid and transferred to nylon membranes. The blots were hybridized with an RPL36A cDNA probe, and then hybridized afterward with an 18S cDNA probe as a loading control. Autoradiographs of Northern hybridizations were scanned using a personal densitometer. The expression level of each gene was normalized against the level of 18S mRNA. (B) Northern blot analysis of RPL36A mRNA in 8 HCC cell lines, compared to nontransformed cells. Twenty μg total RNAs from HCC cell lines or from non-transformed cells were fractionated on 1% agarose gels containing 2.2% formaldehyde and 50 mM 3-[N-morpholino]propanesulfonic acid and were then transferred to nylon membranes. The blots were hybridized with an RPL36A cDNA probe and afterward with an 18S cDNA probe as a loading control. (C) Southern blot analysis of genomic DNAs (10 μg) digested with EcoRI. DNAs were isolated from 7 representative HCC samples and the paired nontumor tissues. The membranes were prehybridized and then hybridized with 32P-labeled RPL36A cDNA. The blot was stripped and reprobed using β-actin cDNA as a loading control.

Table 1. Expression Profile of RPL36A mRNA in 40 HCC Tumor and Paired Nontumor Tissues
Association With HBVAFP LevelNumber (%)
  • NOTE. HBV (+), positive serology of HBV; HBV (−), negative serology of HBV. High AFP, more than 400 ng/ml AFP; Low AFP, less than 400 ng/ml AFP.

  • *

    P < .075; statistical analysis was performed by Chi-Square test or by a test of binomial proportion(s) using a statistical package.

HBV (+)High AFP4/7 (57.1)
 Low AFP11/12 (91.7)*
  15/19 (78.9)
HBV (−)High AFP3/4 (75)
 Low AFP16/17 (94.1)
  19/21 (90.5)
Total 34/40 (85)

Nucleolar Localization of Ribosomal Protein L36a

In order to determine the subcellular localization of ribosomal protein L36a, we fused the full length RPL36A sequence to the C-terminus of GFP. Transient expression of this fusion protein in NIH3T3 and in Chang liver cells showed that the ribosomal protein L36a was able to transport the reporter GFP protein to the nucleus, where it accumulated in nucleoli (Fig. 3A). In a control experiment, we observed that GFP expressed from pEGFP-C1 was not exclusively cytoplasmic or nuclear, but the staining was distributed throughout the nucleus and the cytoplasm. The GFP-RPL36A fusion protein was completely localized in the nucleolus. Similar results were observed in human hepatoma Hep 3B cells transiently transfected with the expression plasmid of the RPL36A/GFP fusion protein (data not shown). To identify region(s) of the protein responsible for its nuclear and nucleolar localization, we constructed and expressed several deletion mutants fused in frame to GFP (Fig. 3B). A tetra-peptide consensus (KR/KXR/K) detected in a significant number of nuclear localization signals (NLS)29 is also found near the N-terminus (27KKGK30), the C-terminus (97KKRK100), and an internal region (58KKGK61) of the ribosomal protein L36a. The minimal NLS and a domain N-terminal to the NLS has been reported to control nuclear and nucleolar accumulation of ribosomal proteins.30 Thus, we examined the nuclear or nucleolar targeting of the tetrapeptide and the adjacent domain N-terminal to the NLS. The deletion construct of amino acids 1–30, which contains the N-terminal tetrapeptide consensus and its adjacent N-terminal domain, showed nucleolar accumulation, whereas the deletion construct containing amino acids 1–26, which lacks the N-terminal tetrapeptide consensus, showed cytoplasmic, nuclear, and nucleolar signals. This pattern is different from the cytoplasmic and nuclear signals (with little staining in nucleoli) due to expression of GFP alone. The deletion construct of amino acids 31–61, which contains the internal tetrapeptide consensus and its adjacent N-terminal domain, showed nucleolar localization, whereas the deletion construct containing amino acids 31–57, which lacks the internal tetrapeptide consensus, showed cytoplasmic and nuclear signals as seen in the expression of GFP alone. However, the deletion construct of amino acids 86–106, which contain the C-terminal tetrapeptide consensus and its adjacent domain, showed cytoplasmic, nuclear, and nucleolar signals (Fig. 3C). Therefore, the N-terminal or the internal tetrapeptide consensus with its adjacent N-terminal domain may control the nuclear and nucleolar targeting of ribosomal protein L36a.

Figure 3.

Localization of GFP-RPL36A fusion protein in cells. (A) NIH3T3 and Chang liver cells grown on coverslips were transiently transfected with 2 μg of plasmid encoding the GFP-RPL36A fusion protein (pEGFP-RPL36A) or with a vector control (pEGFP). The cells were fixed with 4% paraformaldehyde and permeabilized in 0.4% Triton-X for 10 min at room temperature for the indicated times. The cells were stained 15 min with 1 μg/ml Hoechst 33258 to visualize nuclei, and were stored in 50% glycerol in PBS at 4°C. GFP fluorescence was then analyzed using a LSM 510 confocal microscope. Bar, 10 μm. (B) Structure and subcellular distribution of the RPL36A-GFP fusion protein. Gray box, L36a protein sequences; Black box, tetrapeptide consensus sequences; BNTS, putative bipartite nuclear targeting sequences (amino acids 27–43, 43–59, and 86–102); C, cytoplasmic; N, nuclear; NO, nucleolar localization; ≥, more predominant or equal signal. (C) Confocal analysis of GFP localization directed by L36a deletion mutants. Bar, 10 μm.

Increased Clonogenic Survival by Ectopic Over-Expression of Ribosomal Protein L36a

Clonogenic assays are a very sensitive way to assess the cell proliferative potential.31 Therefore, to determine the effects of RPL36A expression on cell survival, we performed colony generation assays with Hep 3B cells or with normal liver-derived Chang liver cells because those cells express relatively less RPL36A mRNA (Fig. 2). Both Hep 3B and Chang liver cells transfected with pcDNA3.1/HisC-RPL36A produced a 2-fold increase in colony number compared with vector control cells. Furthermore, cells transfected with an antisense RPL36A cDNA were about 50% the level of control cells in colony formation (Fig. 4). We observed similar results in cells transfected with GFP/RPL36A fusion expression plasmids (data not shown). However, inhibition of colony formation in this type of assay can be explained either by a block of cell cycling or by the induction of cell death. Thus, we needed to further analyze the functional role of ribosomal protein L36a.

Figure 4.

Effect of L36a expression on clonogenicity of Chang liver and Hep 3B cells. The cells were transfected with an expression vector (pcDNA3.1/HisC- RPL36A), an antisense expression vector (pcDNA3-RPL44AS), or an empty control vector (pcDNA3.1/HisC). After 3-wk of selection with G418, the colonies were visualized by staining for 30 min with fixative containing 0.5% (w/v) crystal violet. Vertical bars represents mean ±SE from 2 independent experiments in duplicate. *P < .01 versus control.

Ectopic Over-Expression of Ribosomal Protein L36a Enhances Cell Proliferation

To determine the role of ribosomal protein L36a over-expression, we introduced the GFP/RPL36A fusion expression system into Chang liver cells. Two types of Chang liver cells (GFP-S1 and GFP-S8), which stably expressed the fusion protein localized in nucleoli, were isolated (Fig. 5A). Cells stably transfected with an antisense RPL36A cDNA (AS-3 and AS-4) were isolated and showed low levels of RPL36A mRNA (Fig. 5B). We then measured the cell proliferative activity by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide assay. Compared with control cells transfected with the empty vector (C1 and C3), the GFP-S1 and the GFP-S8 cells were highly proliferative during culture for 5 d. In contrast, the AS-3 and AS-4 cells showed lower proliferative activity than vector control cells (pcDNA3-C1 and pcDNA3-C2) (Fig. 5C). To exclude the possibility that induction of cell death contributes to this difference in proliferative activity or colony formation, we examined the susceptibility of the cells to chemotherapeutic drug (doxorubicin)-induced apoptosis because we did not find any cellular death during the cell culture. Treatment with doxorubicin did not elicit significant changes in apoptotic cell death among cells expressing sense or antisense RPL36A, compared with vector control cells (data not shown). Thus, we assume that the increased proliferative activity of the cells resulted from enhanced cell cycling.

Figure 5.

Effect of L36a expression on cell proliferation. (A) Confocal microscopic findings of stable transfectants expressing the GFP-RPL36A fusion protein (GFP-S1 and GFP-S8) or the vector control (GFP-C1 and GFP-C3). The cells were stained for 15 min with 1 μg/ml Hoechst 33258 to visualize the nuclei, and were stored in 50% glycerol in PBS at 4°C. GFP fluorescence was then analyzed using an LSN 510 confocal microscope. Bar, 10 μm. (B) Expression of RPL36A mRNA in cell lines stably transfected with an expression vector of antisense RPL36A cDNA. Twenty μg total RNAs from the stable transfectants and from the vector control cells were fractionated on 1% agarose gels containing 2.2% formaldehyde and 50 mM 3-[N-morpholino]propanesulfonic acid and were transferred to nylon membranes. The blots were hybridized with an RPL36A cDNA probe, and then hybridized afterward with a GAPDH cDNA probe as a loading control. (C) Cell proliferation was determined by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide assay. Cells stably transfected with a sense RPL36A or with an antisense RPL36A cDNA were plated in quadruplicate in 24-well plates. At the indicated time intervals, the monolayer cells were processed for the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide assay as described in Materials and Methods, and the absorbance was read at 540 nm in an enzyme-linked immunosorbent assay reader.

Effect of Ectopic Ribosomal Protein L36a Over-Expression on the Cell Cycle

Chang liver cells that stably over-express GFP-RPL36A showed greater cell numbers than vector control cells during culture. Thus, we investigated whether ribosomal protein L36a over-expression is associated with accelerated cell cycling. For synchronization, the thymidine-aphidicolin double-block was used to synchronize the cells at the G1/S interface. Cells were then analyzed after release from the aphidicholin block (Fig. 6). The fraction of cells at G1 is minimal (19.1%) in the control vector cells (pcDNA3-C4) 9 h after release, whereas it occurred 12 h after release in cells transfected with the antisense RPL36A cDNA, implying that the antisense-transfected cells had delayed progression through the cell cycle. Cells stably expressing ribosomal protein L36a showed the minimal fraction of G1 3 h after release. Thus, cells stably over-expressing ribosomal protein L36a showed more rapid cell cycling than did the vector control cells.

Figure 6.

Flow-cytometric histogram of synchronized cells. The cell cycle distribution of cells was performed after the aphidicholin-thymidine double-block using a FACScan, as described in Materials and Methods. Cells transfected with a sense RPL36A (S1), an antisense RPL36A cDNA (AS4), or a vector control (pcDNA3-C4) were analyzed at 3-h intervals after release from the G1 block. Data were analyzed as single-parameter frequency histograms in an SFIT model. DNA contents are plotted on the linear abscissa. C, control.

Discussion

Previously, we observed 14 ribosomal protein genes, including P0, S20, L27a, L8, L31, L37a, S24, S27a, L21, L35a, S8, L37a, S3a, and L36A (L44), that were over-expressed in hepatitis B virus-associated HCC.20 Among those, we were most interested in the over-expression of RPL36A because it was preferentially over-expressed in HCC tissues (85%) and cell lines more than any other genes observed. The mechanism for the enhanced expression of ribosomal protein mRNAs in tumors has been unclear. However, NF-α1 (EF2) and NF-β1 transcription factors were previously reported to regulate expression of the human ribosomal protein gene RPS14.32 Furthermore, GA-binding protein (GABP) (also called nuclear respiratory factor 2 or E4 transcription factor 1) was identified as a regulator of several ribosomal proteins.33 GABP is a positive regulator of the RPL30 gene and can serve as a repressor of the RPS16 gene. The present study revealed that the expression of RPL36A mRNA could be also transcriptionally regulated. However, it remains unclear which transcription factor(s) play roles in the regulation of ribosomal protein L36a and other ribosomal proteins in HCC.

The nucleolus is the site of ribosomal biogenesis and is where the ribosomal proteins are associated with the precursor-rRNAs (pre-RNAs), which are concomitantly processed into mature rRNA molecules.34 The nuclear transport of proteins depends upon the presence of one or more NLS. These sequences have been found throughout the polypeptide chain35 and, in most cases, consist of either short basic amino acid sequences, such as the NLS of the SV40 large T-antigen (126PKKKRKV132),36, 37 or of longer bipartite sequences consisting of 2 stretches of basic amino acids separated by about 10 amino acids.38 NLS are both necessary and sufficient to target a cytoplasmic protein to the nucleus.37 The L36a protein contains 3 tetrapeptide consensus NLS sequences (KR/KXR/K).29 In addition, there are 3 putative bipartite nuclear targeting peptide sequences, 27KKGKDSLYAQGKRRYDR43, 43RKQSGYGGQTKPIFRKK59, and 86KRCKHFELGGDKKRKGQ102. Those tetrapeptide consensus sequences completely overlap the N-terminal and the C-terminal, partially with the internal putative bipartite nuclear targeting sequences. Using RPL36A fused with GFP, we have shown that the minimal NLS detected at the N-terminus or in internal putative bipartite nuclear targeting sequences plays a role sufficient for the nucleolar targeting of ribosomal protein L36a, as found in the minimal NLS of S7 (KRPR) and of S6 (KRRR and KKPR).30 Similarly, domain II of the human ribosomal L7a protein drives the nucleolar accumulation of a reporter protein with the cooperative action of a short basic amino acid sequence.39

Ribosomal protein mRNAs, including P0, S8, S19, L12, L23a, L27, L30, and L9, were over-expressed in HCC.23–25 Among them, L12, together with the P0, P1, and P2 complex, forms the moiety of the GTPase domain in eukaryotic ribosomes.40–42RPL12-deficient yeast cells are viable but grow very slowly.41 Furthermore, the expression of RPL23A mRNA and cell growth are reduced in melanoma cells after treatment with interferon.43 Enhanced expression of an antisense RPL23A sequence in human cervical carcinoma cells reduces colony formation and cell growth, indicating that a direct antiproliferative effect results from inhibiting the expression of this gene. This mounting evidence suggests that enhancement of these ribosomal subunits could play some causal role in the proliferation potential of cancer cells, particularly HCC cells. Ribosomal protein L27 is expressed at higher levels in human fetal tissues than in adult tissues.44 The phorbol ester-induced growth suppression, terminal differentiation, and apoptosis of leukemia cells is also associated with a coordinated suppression of ribosomal protein L27 expression,45 which implies that ribosomal protein L27 expression may be related to the dedifferentiation. In this study, ectopic over-expression of a sense RPL36A cDNA increased colony formation and cell growth. The forced expression of antisense RPL36A mRNA reversed these effects, which suggests that the expression of the RPL36A gene may be involved in cell survival or cell proliferation.

Constitutive expression of human ribosomal protein L7 after transfection of RPL7 cDNA into Jurkat T-lymphoma cells leads to cell cycle arrest and to the apoptotic death of most transfectants.46 The N-terminal region of human and rodent protein L7 carries a sequence motif that is similar to the basic-region-leucine zipper domain characteristic of some eukaryotic transcription factors. The basic-region-leucine zipper-like domain mediates the formation of L7 homodimers that interact with cognate sites on mRNA and can regulate the translation of a distinct set of mRNA.47, 48 Therefore, the expression of ribosomal proteins may also alter cell cycle or cell survival. In the present study, there was no difference in cell survival or apoptotic cell death between the RPL36A transfectants and vector control cells in the absence or presence of apoptotic insults. However, cells transfected with the sense cDNA of RPL36A showed accelerated cell cycling, whereas cells transfected with the antisense cDNA of RPL36A showed delayed progress of the cell cycle, implying that the tumor mediated-over-expression of RPL36A may be involved in the enhanced cell cycling that leads to tumor cell growth.

In summary, our data suggest that over-expression of RPL36A may be associated with hepatocarcinogenesis and that it functionally relates to tumor cell proliferation. Thus, RPL36A may be a potential target gene for anticancer therapy of HCC.

Acknowledgements

The authors thank the Korea Basic Science Institute for the laser scanning microscope.

Ancillary