Identification and characterization of a novel p42.3 gene as tumor-specific and mitosis phase-dependent expression in gastric cancer

Article metrics


Multiple genetic alterations are attributed to gastric cancer (GC); however, only a few critical genes have been identified so far. In this study, we isolated and characterized a novel gene p42.3, represented as tumor-specific and mitosis phase-dependent expression protein in GC cell line BGC823. Our data showed that the expression of p42.3 was cell cycle-dependent in GC cell lines. Moreover, p42.3 was specifically expressed in primary GC tissues but not in the matched normal mucosa of stomach, and this gene was expressed in diverse embryonic tissues. Furthermore, significant suppression of cell proliferation and tumorigenicity were detected and G2/M phase arrest was observed in cell line BGC823 depleted of p42.3 expression by RNAi technique, and we confirmed the expression changes of cyclin B1 and Chk2 following the silence of p42.3. Taken together, we cloned and characterized p42.3 gene that was specifically expressed in GC tumors but not in normal gastric mucosa, and the gene was associated with M-phase regulation. Moreover, p42.3 might be involved in cell proliferation and tumorigenesis; therefore, this gene might have potential applications in the diagnosis or treatment of GC.


Gastric cancer (GC) is the second most common cancer worldwide, and occurs at very high frequency in certain parts of China (You et al., 2005). Epidemiology investigation data showed that dietary factors, smoking and Helicobacter pylori infection played a role in high incidence and mortality rates of GC (Pinto-Santini and Salama, 2005). Importantly, molecular genetic studies have addressed the accumulations of multiple genes alterations involved in GC development, such as activation of oncogenes erbB-2, c-met, c-myc and k-ras, and inactivation of the tumor suppressor gene p53, E-cadherin, APC and RUNX3 (Li et al., 2002; Stock and Otto, 2005) in GC. Epigenetic alterations including DNA methylation (Toyota et al., 1999) and chromatin remodeling by histone modifications are associated with the development and progression of GC (Stock and Otto, 2005); however, these similar approaches have only been found to be associated with a limited number of GC cases, and additional molecules, which have key roles in gastric carcinogenesis, need to be identified.

The cell cycle is a collection of highly ordered processes that result in duplication of a cell (Nasmyth, 1996). There is a category of genes encoding components of cell cycle checkpoint, and their abnormalities increase gene instability and accelerate cellular evolution (Hartwell and Kastan, 1994). In other words, loss of cell cycle regulation is one of the most radical characteristics of neoplastic cells (Hanahan and Weinberg, 2000). At least four checkpoints exist in cell cycle, G1-S, S phase, G2-M checkpoint and spindle checkpoint to ensure the genome stability (Elledge, 1996). Genes involved in G1-S checkpoint, such as p53, p21, MDM2, p27 and Rb, are known for the critical functions in tumorigenesis (Hartwell and Kastan, 1994). Studies have demonstrated loss of function of other genes composing of the S phase, G2-M, and spindle checkpoint in cancers, such as recQ (Cobb and Bjergbaek, 2006), ATM, BRCA1, Cdc25, Chk1 and Chk2 (Deng, 2006). Although the foundational principles of cell cycle regulation have been clarified, we are still ignorant of its detailed mechanism (Murray, 2004) and specific gene involved in GC.

To understand further cell cycle regulation in GC and clarify the molecular mechanisms of gastric carcinogenesis, we performed mRNA differential display (mRNA DD) to screen differential genes between G1 and S phase in synchronized BGC823 cells. A full-length cDNA (3877 bp) was cloned and designated as p42.3 (GenBank, DQ150361).

In the study, we cloned full-length cDNA of p42.3 gene, and found that this gene consistently expressed in GC tissues but not in normal gastric mucosa, and its expression was cell cycle-dependent in GC cell lines. Furthermore, we have investigated correlation between p42.3 expression and biological characteristics in GC cell line BGC823 by depleting its expression.


Cloning and identification of full-length cDNA of p42.3

mRNA DD analysis revealed the high-level expression of a cDNA fragment consisting of 200 nucleotides designated as A54-3 in cells synchronized in G1 phase compared with those in S phase (Figure 1a). Northern blot data showed that a transcript of approximately 4.0 kb was detected in human tumor cells (Figure 1b); in addition rapid amplification of cDNA ends (RACE) technique was performed to obtain the full-length cDNA (3877 bp). We confirmed 5′-end according to G-caping structure and 3′-end according to polyadenylation signal AATAAA followed by poly(A). The Upstream of start codon of putative open reading frame (ORF) had Kozak sequence. The ORF encodes a 389 aa protein estimated to have a molecular mass of 42.3 kDa. Therefore, full-length mRNA accorded with the features of a typical gene; we named this gene as p42.3 (GenBank: DQ150361). There are proline-rich regions in the amino-acid sequence, and the protein contains numerous potential phosphorylation sites for PKC, CK2, cAMP and cGMP. Additionally, there is an EF-hand motif at N terminal and a coiled-coil (CC) domain at C-terminal ends (Figure 2).

Figure 1

mRNA DD assay and cloning of full-length cDNA of p42.3. (a) Differential cDNA fragment A54-3 was screened out by mRNA DD. (b) Northern blot was performed in cancer cell lines to predict the transcript size. The loading levels are shown in the lower panel.

Figure 2

Analysis of nt and aa sequences of p42.3 cDNA. RACE technique was performed to amplify a full-length cDNA (3877 bp), which contained an ORF of 1170 bp (from 144 to 1313bp). The initiating methionine conformed to Kozak sequence criterion (shadowed sequence). The start codon (ATG) and stop codon (TAG) are shown in boldface. The canonical 3′-polyadenylation signal sequence (AATAAA) is shown in boldface. Double underlining indicated the proline-rich region (8–14aa, 97–169aa). The putative phosphorylation sites are shown in a box and the corresponding kinases are indicated in italic ( Single underlining indicated the CC domain (331–379aa, There is an EF-hand motif at N terminus (aa7–87).

Differential expression of p42.3 in the cell cycle phases

Twelve GC cell lines were examined by reverse transcription RT–PCR; the results showed that p42.3 mRNA was expressed in nine out of 12 (75%) GC cell lines (Figure 3a). We further examined its mRNA expression in BGC823 synchronized in G1, S, G2 and M phase by RT–PCR and demonstrated that p42.3 expression in G1 and M phases was higher than in S and G2 phases (Figure 3b). Moreover, immunocytochemistry analysis was performed in cancer cell lines including 7901, BGC823, MGC803 and AGS; p42.3 protein was detected both in nucleus and in cytoplasm. More importantly, we found that the level of p42.3 protein was the highest in M phase compared with other three phases. The morphology features of cells in M phase included the cells were orbicular and the chromosomes were congregated on metaphase plate or the cells were dividing. Staining with the pre-immune serum served as negative control (Figure 3c).

Figure 3

Detection of p42.3 in cancer cell lines. (a) RT–PCR analysis identified that p42.3 was detectable in 9/12 GC cell lines. (b) p42.3 expression was differential in BGC823 synchronized in G1, S, G2 and M phases. β-Actin served as control. (c) Immunocytochemistry analysis showed that the highest level of p42.3 protein was in the M phase of tumor cells (original magnification, × 200 and × 400). Staining by the preimmune serum served as the negative control (original magnification, × 200).

Consistent expression of p42.3 gene in primary GC tissues

p42.3 gene was frequently expressed in tumor cells. Our interest then focused on p42.3 expression profile in GC tissues and normal mucosa of stomach. First, RT–PCR was performed in 15 pairs of primary gastric carcinoma and the matched adjacent normal mucosa. We found that p42.3 gene was expressed in gastric tumor tissues but not in normal tissues among 8 out of 15 pairs of primary GC (53.3%), and in other 7 cases, it was totally not expressed (Figure 4a). To further confirm this observation, mRNA in situ hybridization (ISH) was performed in tissue microarray including 100 samples of primary GC and 20 samples of normal gastric mucosa. p42.3 gene was expressed in gastric tumor tissues (53 out of 100, 53%), while there was no expression in 20 specimens of normal tissues (P<0.01) (Figure 4b and Table 1). Then immunohistochemistry (IHC) was subsequently used to detect its protein expression in 44.3% (35 out of 79) gastric tissues; however, only two examples were detectable among 31 normal gastric tissues (P<0.01, Figure 4c and Table 2).

Figure 4

Detection of p42.3 in gastric and embryonic tissues. (a) RT–PCR analysis showed that p42.3 was expressed in primary GC but not in the matched adjacent normal gastric mucosa (M, marker, N, normal gastric mucosa, T, gastric tumor). (b) ISH using an antisense and sense probe further demonstrated that p42.3 was expressed in GC but not in normal tissues. (c) Immunohistochemistry was performed to detect the protein of p42.3 in GC and normal tissues (N, normal gastric tissue, T, gastric tumor tissues). (d) p42.3 mRNA was expressed in diverse embryonic tissues.

Table 1 Frequent expression of p42.3 gene detected in GC tissues by mRNA ISH analysis
Table 2 Specific expression of p42.3 protein detected in GC tissues by IHC analysis

p42.3 expression in diverse embryonic tissues but not in adult tissues

Many oncogenes were expressed at fetal stage, p42.3 gene was expressed in GC cell lines and in GC tissues but not in normal gastric mucosa In addition, we wanted to know whether its expression was present in human embryonic tissues. RT–PCR was performed in 5-month embryonic tissues including stomach, intestine, colon, liver, brain, lung, heart, spleen and kidney. Our results showed that except in heart and lung, strong expression of p42.3 is seen in seven other tissues have strong (Figure 4d). In addition, we failed to detect p42.3 expression in noncancerous tissues from adult organs, such as stomach, colon, liver, breast brain, lung and cervix using IHC, RT–PCR and ISH, respectively (Supplementary Table 1 and Figures 1, 2 and 3).

Stable silencing p42.3 in GC cell line BGC823

To reveal the function of p42.3, we blocked its endogenous expression in BGC823 by RNAi technique. We isolated G418-resistant clones, which were stably transfected by vector-inserted different short-hairpin nucleotides targeting three sites of p42.3, and examined the p42.3 expression at both mRNA and protein level by semi-quantitative RT–PCR and western blot, respectively. We confirmed p42.3 expression was dramatically depleted in psiRNA-3C clone, which was targeted at position 2545–2565 bp of mRNA. In addition, p42.3 mRNA was undetectable, while the protein expression was partially blocked in another clone psiRNA-2D that was targeted at position 3406–3424 bp of mRNA. The parental BGC823 cell transfected with psiRNA-vector served as positive control (Figure 5a).

Figure 5

The effect on morphology and cell cycle progression by silencing of p42.3 in BGC823. (a) RT–PCR and western blot were performed to detect silence of p42.3 in single colony of BGC823 (1, psiRNA-vector; 2, psiRNA-2D and 3, psiRNA-3C), β-actin served as internal control. (b) Cyclin B1 and Chk2 were examined by western blot (1, BGC823; 2, psiRNA-vector and 3, psiRNA-3C), and β-actin served as control. (c) The morphologic was significantly altered after downregulation of p42.3. 4′,6-Diamidino-2-phenylindole (DAPI) staining further showed the alterations of nucleus in psiRNA-2D and psiRNA-3C (shown by arrows). Cells transfected by empty vector served as control. (d) FACS analysis confirmed psiRNA-2D and psiRNA-3C had G2/M phase arrest, and a small peak following G2/M peak was found in these two clones, respectively.

Downregulation of p42.3 affected the cell cycle progression

Morphologic changes were observed in the psiRNA-2D and the psiRNA-3C cells; especially, the cells of psiRNA-3C became flat and nonrefractile. Moreover, many giant cells, which were multinucleated and had pseudopodium, were found. 4′,6-diamidino-2-phenylindole (DAPI) staining further confirmed that the macronucleus composed of undivided poly-nucleus in the silenced cells (Figure 5c). Then fluorescence-activated cell sorter (FACS) analysis was performed to examine whether the downregulation of p42.3 alters the cell cycle progression. The results confirmed that cells of psiRNA-2D and psiRNA-3C were arrested at G2/M phase compared with the control cells. Moreover, there is a small peak after G2/M peak in both of cell clones; however, none of the peaks were seen in the control, indicating that the DNA content of these cells was higher than that of the G2/M phase cells (Figure 5d). We further examined the expression of cyclin B1 and Chk2 by western blot, which were two key genes involved in M-phase regulation; our results showed that cyclin B1 significantly decreased in psiRNA-3C; in contrast, Chk2 was increased in psiRNA-3C compared with parental BGC823 and psiRNA-vector (Figure 5b).

Suppression of proliferation and tumorigenicity in cell line BGC823 depleted p42.3

Our data showed that cell growth was significantly suppressed in psiRNA-3C compared with psiRNA-2D, parental BGC823 and psiRNA-vector (Figure 6a). Colony formation assay was used to evaluate the ability for anchorage-independent growth of cells in soft agar medium. Our data showed a significantly reduced colony formation of psiRNA-3C in both number and size, as compared with those cells transfected with vector alone (P<0.01). Colony formation of psiRNA-2D was not significantly decreased (Figure 6b and c).

Figure 6

Inhibition of cell proliferation and tumorigenesis after silencing of p42.3 in BGC823. (a) MTT assay showed that the cell growth of psiRNA-3C was strikingly inhibited; data are shown as the mean±s.d. of three independent experiments (*P<0.01). (b) Counting of the number of colonies showed that colony-forming activities of psiRNA-3C decreased significantly on soft agar. The data represent the mean±s.d. of three of independent experiments (*P<0.01). (c) The colonies of psiRNA-3C, psiRNA-2D and control formed on soft agar. (d) Inhibition of tumorigenesis of psiRNA-3C has statistical significance compared with control (*P<0.01). (e) The tumor induced by psiRNA-2D was smaller than control, while three mice injected with psiRNA-3C did not bear tumors, and other three mice developed much smaller tumors than control. (f) HE staining was performed in tissues of xenograft tumors.

To examine whether downregulation of p42.3 expression could inhibit the tumorigenicity of BGC823 in vivo, each of the cells, psiRNA-3C, psiRNA-2D and psiRNA-vector, was injected subcutaneously into nude mice. Tumors appeared slowly in athymic nude mice injected with psiRNA-3C compared with the control. Moreover, at the end of fourth week, four mice failed to bear tumors, and seven mice developed into small tumors in three independent experiments of psiRNA-3C. The results showed that both size and weight of xenograft tumor developed by psiRNA-3C were smaller than the control (P<0.05); the tumorigenicity of psiRNA-2D was not significantly inhibited (Figure 6d and e). The tumor cells of psiRNA-3C were well differentiated compared with psiRNA-2D and psiRNA-vector (Figure 6f).


The isolation and characterization of differentially expressed genes during carcinogenesis provide fundamental knowledge to understand the mechanisms responsible for malignant transformation (Herness and Naz, 2003; Lu et al., 2005). Gastric cell lines have been widely used as the model to study GC. In this study, GC cell line BGC823, which is an adherent, poorly differentiated, human gastric adenocarcinoma cell line (Ji et al., 2002), was used as the model to investigate genes expression in different cell cycle phases and further to understand the molecular mechanisms of carcinogenesis of stomach. Multiple genetic alterations drive the progressive transformation of normal cells into cancer cells (Hanahan and Weinberg, 2000). The advances in our understanding of the cell cycle reveal how fidelity is normally achieved by the coordinated activity of cyclin-dependent kinases, checkpoint controls and repair pathways (Hartwell and Kastan, 1994). Many genes involved in cell cycle regulation, especially in cell cycle checkpoint, were abnormal in carcinoma; however, additional molecules need to be identified to clarify relationship between cell cycle control and tumorigenesis.

BGC823 cells were synchronized in G1 and S phase, and cDNA fragment A54-3 was initially identified as a candidate gene that is highly expressed in G1 phase by mRNA DD. A54-3 was highly homologous to c9orf140 (99%) located on Homo sapiens 9q34.3 chromosomal locus (Strauberg et al., 2002). We cloned full-length cDNA of c9orf140 and renamed c9orf140 as p42.3 (DQ150361). The homology of the human p42.3 gene to its mouse counterpart is RIKEN cDNA 2010317E24 (ang, 81% in nucleotide and 78% in amino-acid sequence), which was located on Mus musculus chromosome 2A3 and expressed in early neuroectoderm of embryo (Murata et al., 2004). The high conservation of p42.3 in mammals also suggested that this gene may have important roles in cell biology; however, its function is totally unknown.

One key finding of our study showed that p42.3 might be involved in cell cycle regulation. We confirmed differential expression of p42.3 in the phases of cell cycle; more importantly, the peak of p42.3 protein expression was in M phase, and its level gradually reduced after cell division. Downregulation of p42.3 showed the morphology of the cells was significantly altered. A morphologically enlarged and polynuclear cells were further identified, which indicated that they could not normally progress division. Moreover, the percentage of cells in G2/M phase was evidently increased compared with the control, which indicated that there was G2/M phase-arrest following the downregulation of p42.3. A specific peak observed in psiRNA-2D and psiRNA-3C by FACS assay suggested that this portion of cells was multiploid, which correlated well with the morphologic changes mentioned above.

Moreover, we confirmed expression status of cyclin B1 and Chk2, which are the two key genes involved in M-phase regulation. In response to DNA damage at G2 phase, Chk2 can phosphorylate Cdc25, creating a binding site for protein of the 14-3-3 family (Chaturvedi et al., 1999) that sequester Cdc25 in cytoplasm where it cannot dephosphorylate cyclin B1/cdc2, and thereby, lead to inhibition of mitosis (Kumagai and Dunphy, 1999); our data showed that upregulation of Chk2 and downregulation of cyclin B1 were found after silence of p42.3, which could result in G2/M arrest in the BGC823 cells-deleted p42.3 protein. Therefore, p42.3 gene might be involved in regulation of cell division, and additional studies are required to fully clarify the cause of retarding effect on cell-cycle progression by silencing of the p42.3 gene.

Another critical finding of our study was that we confirmed p42.3 was specifically expressed in GC tissues but not normal gastric mucosa, which suggested this gene might have a potential role in GC development. Then we identified that p42.3 was expressed in human fetal tissues but not in noncancerous tissues of adult organs (data not shown). It is generally believed that a fetal protein, highly expressed at fetal stages and quickly shut down after birth, could reoccur at tumor stages. Such molecule is likely to be a tumor marker and might play a role as oncogene involved in carcinogenesis (Zeng et al., 2002).

Our results further supported our hypothesis that p42.3 might stimulate cellular proliferation and malignant transformation. Silencing of this gene in BCG823 by RNAi resulted in significant inhibition of cancer cell proliferation and colony formation in vitro, and significantly reduced tumorigenicity in nude mice. Moreover, well-differentiated phenotype was observed in BGC823 cells-blocked p42.3 gene expression. Thus, these findings provided evidence that p42.3 indeed played important role in tumorigenesis, the molecular mechanism responsible for stimulating cellular proliferation and malignant transformation that needs to be further investigated.

Our data confirmed that the ORF of p42.3 cDNA can be translated into a 42.3 kDa protein in prokaryotic cells. A BLAST search for sequence homology in the GenBank database revealed that p42.3 has no significant homology to any genes with a known function. Interestingly, the p42.3 protein is proline-rich, which could bind with SH3 domains and were widely known to play a role in mediating protein–protein interactions involved in cellular signal (Kay et al., 2000). Another motif is EF-hand at the N-terminal end of p42.3 protein; some EF-hand proteins such as the S100 proteins have been reported to be involved in cancer (Santamaria-Kisiel et al., 2006). In addition, a CC domain involved in protein interactions existed in p42.3 protein (Zhang et al., 2001). Further analysis showed that p42.3 protein might be regulated by phosphorylation. These analyses indicated that p42.3 protein had functional domain or motifs to interact with other proteins.

In summary, we cloned full-length cDNA of p42.3 and demonstrated that p42.3 was highly expressed in M phase and specifically expressed in gastric tumors compared with normal tissues. Our results indicated that p42.3 gene had a critical role in cellular proliferation and tumor formation, and our present data suggested that p42.3 gene might be involved in the regulation of mitosis. Therefore, p42.3 might serve as a molecular marker or predictive target for GC.

Materials and methods


This study encompassed 194 patients (66 pairs of GC and matched adjacent normal tissues, 128 GC samples) surgically resected in School of Oncology, Peking University and Peking University Second Hospital. Histopathologic analysis was performed independently by two pathologists.

Cell lines and cell culture

GC cell lines BGC823, MGC803, SGC7901, PAMC82, MKN45, SNU1, SNU5, SNU16, RF1, RF48, AGS and N87, and prostate tumor cell line PC3, were cultured in Dulbecco's modified Eagle's medium (DMEM, Gibco BRL, Grand Island, NY, USA), supplemented with 5% fetal bovine serum (FBS); all cell lines were maintained at 37°C in 5% CO2.

Cell synchronization

BGC823 cells were cultured in DMEM media supplemented with 5% FBS at 37°C in 5% CO2. High-pressure N2O plus TdR (thymine deoxyriboside) (Sigma, St Louis, MO, USA) double-block method was used to synchronize cells in G1, S, G2 and M phases, and all synchronized cells were analysed by FACS method.

RNA extraction and mRNA DD

mRNA DD was performed according to Handbook for HIEROGLYPH mRNA Profile Kit (Genomyx Corporation, Foster City, CA, USA). Briefly, 10 μg of total RNA extracted from G1- and S-phase cells was converted to cDNA. PCR was performed using anchored primers. PCR products were run and then subjected to autoradiography. Differential bands were cut out from the gel, and cDNA fragments were eluted and reamplified according to kit recommendation.

Northern blot

Fifteen micrograms of total RNA was separated by formaldehyde-agarose gel electrophoresis and transferred onto nitrocellulose membrane, p42.3. cDNA was used as probe after labeling with [32P]dCTP by random priming according to manufacturer's instruction (Promega, Madison, WI, USA). The quality of the RNAs was confirmed by 18s and 28s rRNA.

Cloning full-length cDNA of p42.3

Full-length cDNA of the gene was cloned using the SMART RACE cDNA amplification kit (Clontech, Palo Alto, CA, USA) according to the manufacturer’s instruction. After extension of 5′ and 3′ ends, full-length 3877 bp cDNA of c9orf140 was obtained. We named this gene as p42.3 and submitted it to Genbank (DQ150361).


Total RNA of cell lines and tissues was converted to cDNA using oligo-dT primer and Moloney murine leukemia virus (M-MLV) reverse transcriptase according to the manufacture’s manual (Promega). The PCR products were separated on 1.5% agarose gels and stained with ethidium bromide. β-Actin served as an internal positive control.

mRNA in situ hybridization

To detect p42.3 expression in the human GC specimens, mRNA ISH on tissue array section was performed according to the manufacturer’s instruction (Roche, Mannheim, Germany) using (DIG)-labeled antisense- or sense-strand RNA probes consisting of p42.3 nucleotides 2987–3371 bp.

Immunohistochemistry and Immunocytochemistry

Tissue micro-arrays were prepared as described previously (Tang et al., 2004) and immunohistochemically incubated with p42.3 antibody using ABC kit (Pierce, Rockford, IL, USA). Tumor cell lines were fixed on slides; staining was subsequently done as described above; preimmune serum staining was used as negative control.

Western blot analysis

Fifty micrograms cell line protein was separated by 12% Tris-glycine polyacrylamide gel and transferred to polyvinylidene fluoride (PVDF) membrane (Bio-Rad, Hercules, CA, USA). The blot was probed with antibodies. Antibody binding was detected using enhanced chemiluminescence (ECL) detection system (Amersham, Piscataway, NJ, USA) according to the manufacturer’s protocol.

RNAi plasmid constructs and transfection

We expressed short hairpin RNA molecules targeted to three sites at nucleotides 3170–3188, 3406–3424 and 2545–2565bp of p42.3 mRNA. The sequences were as follows: No. 1 (5′-IndexTermTCCCACTCCCAGTTTACAGCTTGATCAAGAGTCAAGCTGTAAACTGGGAGTTT-3′, 5′-IndexTermCAAAAAACTCCCAGTTTACAGCTTGACTCTTGATCAAGCTGTAAACTGGGAGT-3′), No. 2 (5′-IndexTermTCCCACTGTGATGACACCCGGAAATCAAGAGTTTCCGGGTGTCATCACAGTTT-3′, 5′-IndexTermCAAAAAACTGTGATGACACCCGGAAACTCTTGATTTCCGGGTGTCATCACAGT-3′), No. 3 (5′-IndexTermTCCCAGGGCACTTTGGTACACTGTCTCAAGAGGACAGTGTACCAAAGTGCCCTTT-3′, 5′-IndexTermCAAAAAAGGGCACTTTGGTACACTGTCCTCTTGAGACAGTGTACCAAAGTGCCCT-3′). Oligonucleotides were annealed and ligated to psiRNA-Hh1NEO plasmid (InvivoGen, San Diego, CA, USA). BGC823 was transfected, and stable transfectants were isolated as described previously (Deng et al., 2005), the efficacy of the knockdown was identified by RT–PCR and western blot.

MTT assay

BGC823 cells were plated onto 96-well culture plates, and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was added to the cells at 1–6 days. MTTwas removed after 4 h incubation, and then dimethyl sulfoxide (DMSO) was pipetted to solubilize the formazan product. The absorbancy at 570 nm was assayed by micro-plate reader (Bio-Rad).

Flow cytometric analysis

The cells were routinely harvested and washed with phosphate-buffered saline (PBS), and then fixed in ethanol. The fixed cells were stained with propidium iodide (PI) before analysis by FACsorter (Becton Dickinson, Flanklin Lakes, NJ, USA). Finally, the cell cycle profiles were interpreted by CellQuest software.

Soft agar colony formation assay

Soft agar assays were performed as described previously (Deng et al., 2005). Anchorage-independent growth was measured by counting the colony number and observation of the size of colony.

Tumorigenicity assay

BGC823 cells blocked by p42.3-RNAi were injected subcutaneously into the left dorsal flank of nude mice (BALB/c), and the right side injected with BGC823 infected by vector alone served as control. Tumor size was measured every week, at the end of the experiment the mice were killed, and the tumor specimens were weighed and fixed. Histological examination was performed with hematoxylin and eosin staining.

Accession codes




  1. Chaturvedi P, Eng WK, Zhu Y, Mattern MR, Mishra R, Hurle MR et al. (1999). Mammalian Chk2 is a downstream effector of the ATM-dependent DNA damage checkpoint pathway. Oncogene 18: 4047–4054.

  2. Cobb JA, Bjergbaek L . (2006). RecQ helicases: lessons from model organisms. Nucleic Acids Res 34: 4106–4114.

  3. Deng CX . (2006). BRCA1: cell cycle checkpoint, genetic instability, DNA damage response and cancer evolution. Nucleic Acids Res 34: 1416–1426.

  4. Deng H, Guo RF, Li WM, Zhao M, Lu YY . (2005). Matrix metalloproteinase 11 depletion inhibits cell proliferation in gastric cancer cells. Biochem Biophys Res Commun 326: 274–281.

  5. Elledge SJ . (1996). Cell cycle checkpoints: preventing an identity crisis. Science 274: 1664–1673.

  6. Hanahan D, Weinberg RA . (2000). The hallmarks of cancer. Cell 100: 57–70.

  7. Hartwell LH, Kastan MB . (1994). Cell cycle control and cancer. Science 266: 1821–1828.

  8. Herness EA, Naz RK . (2003). A novel human prostate-specific gene-1 (HPG-1): molecular cloning, sequencing, and its potential involvement in prostate Carcinogenesis. Cancer Res 63: 329–336.

  9. Ji JF, Chen X, Leung SY, Chi JT, Chu KM, Siu TY et al. (2002). Comprehensive analysis of the gene expression profiles in human gastric cancer cell lines. Oncogene 21: 6549–6556.

  10. Kay BK, Williamson MP, Sudol M . (2000). The importance of being proline: the interaction of proline-rich motifs in signaling proteins with their cognate domains. FASEB J 14: 231–241.

  11. Kumagai A, Dunphy WG . (1999). Binding of 14-3-3 proteins and nuclear export control the intracellular localization of the mitotic inducer Cdc25. Genes Dev 13: 1067–1072.

  12. Li QL, Ito K, Sakakura C, Fukamachi H, Inoue K, Chi XZ . (2002). Causal relationship between the loss of RUNX3 expression and gastric cancer. Cell 109: 113–124.

  13. Lu M, Zhang LP, Maul RS, Sartippour MR, Norris A, Whitelegge J et al. (2005). The novel gene EG-1 stimulates cellular proliferation. Cancer Res 65: 6159–6166.

  14. Murata T, Furushima K, Hirano M, Kiyonari H, Nakamura M, Suda Y . (2004). Ang is a novel gene expressed in early neuroectoderm, but its null mutant exhibits no obvious phenotype. Gene Expr Patterns 5: 171–178.

  15. Murray AW . (2004). Recycling the cell cycle: cyclins revisited. Cell 116: 221–234.

  16. Nasmyth K . (1996). Viewpoint: putting the cell cycle in order. Science 274: 1643–1645.

  17. Pinto-Santini D, Salama NR . (2005). The biology of helicobacter pylori infection, a major risk factor for gastric adenocarcinoma. Cancer Epidemiol Biomarkers Prev 14: 1853–1858.

  18. Santamaria-Kisiel L, Shaw GS, Shaw GS . (2006). Calcium-dependent and -independent interactions of the S100 protein family. Biochem J 396: 201–214.

  19. Stock M, Otto F . (2005). Gene deregulation in gastric cancer. Gene 360: 1–19.

  20. Strauberg RL, Feingold EA, Grouse LH, Derge JG, Klausner RD, Collins FS . (2002). Generation and initial analysis of more than 15 000 full-length human and mouse cDNA sequences. Proc Natl Acad Sci USA 99: 16899–16903.

  21. Tang ZB, Zhao M, Ji JF, Yang GB, Hu FL, He JS et al. (2004). Overexpression of gastrin and c-met protein involved in human gastric carcinomas and intestinal metaplasia. Oncol Rep 11: 333–339.

  22. Toyota M, Ahuja N, Suzuki H, Itoh F, Ohe-Toyota M, Imai K et al. (1999). Aberrant methylation in gastric cancer associated with the CpG island methylator phenotype. Cancer Res 59: 5438–5442.

  23. You WC, Li JY, Zhang L, Jin ML, Chang YS, Ma JL et al. (2005). Etiology and prevention of gastric cancer: a population study in a high risk area of China. Chin J Diq Dis 6: 149–154.

  24. Zeng JZ, Wang HY, Chen ZJ, Ulrich A, Wu MC et al. (2002). Molecular cloning and characterization of a novel gene which is highly expressed in hepatocellular carcinoma. Oncogene 21: 4932–4943.

  25. Zhang PN, McAlinden A, Li S, Schumacher T, Wang HL, Hu S et al. (2001). The amino-terminal heptad repeats of the coiled-coil neck domain of pulmonary surfactant protein D are necessary for the assembly of trimeric subunits and dodecamers. J Biol Chem 276: 19862–19870.

Download references


This work was supported by grants from the National Natural Science Foundation of China (39625016), State Key Basic Research Program of China (2004CB518708) and the National High Technology Research and Development Program of China (2002BA711A11).

Author information

Correspondence to Y Lu.

Additional information

Supplementary Information accompanies the paper on the Oncogene website (

Supplementary information

Rights and permissions

Reprints and Permissions

About this article


  • gastric cancer
  • p42.3
  • cell cycle
  • RNAi

Further reading