SEARCH

SEARCH BY CITATION

Keywords:

  • cyclooxygenase-2;
  • gastric cancer;
  • microRNA;
  • miR-101;
  • tumor growth

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Cyclooxygenase-2 (COX-2) plays an important role in the carcinogenesis and progression of gastric cancer. It has been demonstrated that COX-2 overexpression depends on different cellular pathways, involving both transcriptional and post-transcriptional regulation. MicroRNAs (miRNAs) are small, noncoding RNAs that function as post-transcriptional regulators. Here, we characterize miR-101 expression and its role in the regulation of COX-2 expression, which in turn, will provide us with additional insights into the potential therapeutic benefits of exogenous miR-101 for treatment of gastric cancer. Our results showed that miR-101 levels in gastric cancer tissues were significantly lower than those in the matched normal tissue (P < 0.01). Furthermore, lower levels of miR-101 were associated with increased tumor invasion and lymph node metastasis (P < 0.05). We also found an inverse correlation between miR-101 and COX-2 expression in both gastric cancer specimens and cell lines. Significant decreases in COX-2 mRNA and COX-2 levels were observed in the pre-miR-101-infected gastric cancer cells. One possible mechanism of interaction is that miR-101 inhibited COX-2 expression by directly binding to the 3′-UTR of COX-2 mRNA. Overexpression of miR-101 in gastric cancer cell lines also inhibited cell proliferation and induced apoptosis in vitro, as well as inhibiting tumor growth in vivo. These results collectively indicate that miR-101 may function as a tumor suppressor in gastric cancer, with COX-2 as a direct target.


Abbreviations
COX-2

cyclooxygenase-2

GAPDH

glyceraldehyde-3-phosphate dehydrogenase

miRNA

microRNA

PCNA

proliferating cell nuclear antigen

PE

phycoerythrin

SD

standard deviation

SEM

standard error of the mean

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Gastric cancer remains the second leading cause of cancer-related death worldwide, and is a major cause of cancer-related mortality in China [1]. Although overall survival in patients with gastric cancer has improved in recent years, owing to increased detection of early cancers and wider implementation of radical surgery [2, 3], the prognosis of advanced cancer remains poor, as safe and effective adjuvant therapy options are limited.

MicroRNAs (miRNAs) are small, noncoding RNAs that bind to partially complementary sites in the 3′-UTRs of target genes, leading to mRNA destabilization and translational suppression [4-6]. Growing evidence indicates that miRNAs are involved in crucial biological processes, including development, differentiation, apoptosis, and proliferation [7-9]. Furthermore, an increasing number of miRNAs have been shown to have important roles in cancer development and progression [10-12]. From a therapeutic point of view, miRNAs are very promising, as several studies have demonstrated the utility of miRNAs as biomarkers. Preliminary preclinical studies have also established that miRNAs may be therapeutically targeted in vivo [13, 14].

Cyclooxygenase catalyzes the formation of prostaglandin and other eicosanoids from arachidonic acid. Cyclooxygenase-2 (COX-2), the mitogen-inducible isoform, is constitutively expressed in gastric cancer [15, 16]. However, the molecular mechanisms underlying the aberrant expression of COX-2 in gastric cancer remain unclear. Regulation of COX-2 expression depends on different cellular pathways, involving both transcriptional and post-transcriptional mechanisms. Previous studies have shown that COX-2 inhibition by selective COX-2 inhibitors or small interfering RNA suppresses cell proliferation and induces apoptosis in human gastric cancer cells [17, 18]. Recently, others have found that miR-101 levels are decreased in patients with gastric cancer, colon cancer, and endometrial serous adenocarcinoma, all of which are characterized by COX-2 upregulation [19-21]. However, the exact mechanism by which the inverse correlation between miR-101 and COX-2 levels contributes to the pathophysiology underlying gastric cancer remains unknown. Therefore, the primary goals of this study were to examine the expression of miR-101 and COX-2 in gastric cancer specimens and to investigate the potential of exogenous miR-101 as a novel form of adjuvant treatment for patients with gastric cancer.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Expression of miR-101 and COX-2 in gastric cancer

A total of 30 patients with histologically confirmed gastric adenocarcinoma were classified into four stages according to the American Joint Committee on Cancer cancer staging system [22]. The clinicopathological characteristics of these patients are shown in Table 1, along with the corresponding miR-101 and COX-2 mRNA levels. As shown in Fig. 1A, miR-101 expression was significantly downregulated in gastric tumor tissues relative to matched normal tissues, whereas COX-2 mRNA expression was markedly upregulated (P < 0.01). In contrast, a higher level of miR-101 was associated with a lower level of COX-2 mRNA in the normal tissues. Additional statistical analysis demonstrated an inverse relationship between miR-101 and COX-2 mRNA levels in gastric tumor (r = −0.77, P < 0.01) and matched normal tissues (r = −0.72, P < 0.01; Fig. 1B). There was no significant correlation between miR-101 expression and most clinicopathological characteristics, including age, gender, tumor site, histological type, and the degree of tumor cell differentiation. However, statistically significant differences in miR-101 levels were noted with respect to stage and lymph node metastasis (Table 1). Specifically, patients with stage III–IV disease or lymph node metastasis had significantly lower miR-101 levels than patients with stage I–II disease or those without lymph node metastasis. On the other hand, COX-2 mRNA levels were significantly higher in patients with stage III–IV disease or lymph node metastasis than in patients with stage I–II disease or those without such metastasis. These results are consistent with previous reports that overexpression of COX-2 is related to tumor invasion and lymph node metastasis in gastric cancer [16]. Furthermore, western blot analysis showed increased levels of COX-2 in tumor tissues as compared with matched normal tissues (Fig. 1C–F). An inverse correlation was also found between miR-101 and COX-2 expression in gastric tumor tissues (r = −0.765, P < 0.05; Fig. 1D) but not in matched normal tissues (r = 0.679, P > 0.05; Fig. 1E).

image

Figure 1. Expression of miR-101 and COX-2 in human gastric cancer specimens. (A) Quantitative real-time PCR was used for the detection of miR-101 and COX-2 mRNA expression. In each sample, miR-101 expression was normalized against U6 RNA, and COX-2 mRNA expression was normalized against GAPDH mRNA. (B) miR-101 expression correlated negatively with COX-2 mRNA levels in cancer tissues (r = −0.77, P < 0.01) and normal tissues (r = −0.72, P < 0.01). (C) Western blot analysis of COX-2 protein expression in gastric cancer and matched normal tissues obtained from seven randomly selected patients. N, normal tissues; T, cancer tissues). (D–F) The histogram represents the relative expression levels of miR-101 and COX-2. Each data point represents the mean ± standard error of the mean (n = 30). **P < 0.01 versus normal tissues. COX-2 expression was normalized to β-actin.

Download figure to PowerPoint

Table 1. The correlation between miR-101/COX-2 expression and patient characteristics. miR-101 expression was normalized against U6 RNA levels, and COX-2 mRNA levels were normalized against GAPDH mRNA
Characteristic n miR-101COX-2
Mean ± SEMP-valueMean ± SEMP-value
Age (years)
< 60181.40 ± 0.410.686.84 ± 1.400.98
≥ 60121.21 ± 0.26 6.81 ± 1.00 
Gender
Male231.18 ± 0.230.436.89 ± 0.880.89
Female71.61 ± 0.29 6.62 ± 1.98 
Tumor site
Cardia30.57 ± 0.120.447.38 ± 0.490.86
Corpus71.06 ± 0.34 5.50 ± 1.73 
Antrum201.47 ± 0.31 6.50 ± 0.80 
Stage
I + II102.13 ± 0.500.0054.24 ± 0.910.02
III + IV200.86 ± 0.16 8.12 ± 1.01 
Histological type
Tubular81.43 ± 0.570.916.93 ± 1.680.96
Papillary101.33 ± 0.41 6.50 ± 1.34 
Mucinous81.02 ± 0.27 7.33 ± 1.84 
Signet ring cell41.41 ± 0.66 6.17 ± 1.93 
Histological grading
Well and moderately differentiated171.37 ± 0.340.326.75 ± 1.030.69
Poorly differentiated131.17 ± 0.28 6.92 ± 1.30 
Lymph node metastasis
Absent102.16 ± 0.490.0033.59 ± 0.950.001
Present200.84 ± 0.16 8.44 ± 0.92 

Overexpression of miR-101 inhibits COX-2 expression in gastric cancer cells

To better understand the regulation of COX-2 mRNA, we analyzed the 3′-UTR sequence of human COX-2 mRNA by use of an algorithm from the Sanger Institute (http://microrna.sanger.ac.uk/) to search for potential miRNA target sites [23]. Although several miRNA candidates (such as miR-101, miR-143, and miR-144) were identified as potential regulators of COX-2 expression, miR-101 showed a high binding score for its target site on COX-2 mRNA. The alignment of miR-101 with its putative target sequences on the human COX-2 3′-UTR is shown in Fig. 2A. Previous studies have also demonstrated, by luciferase activity assay, that miR-101 interacts directly with the 3′-UTR of COX-2 mRNA, leading to its post-transcriptional repression [19, 24]. Therefore, we constructed a recombinant lentiviral vector containing pre-miR-101 (Lenti-miR) to mediate miR-101 overexpression in gastric cancer cell lines (SGC-7901 and BGC-823). The infective efficiencies of lentivirus in SGC-7901 and BGC-823 cells all reached up to 90% without cytotoxicity at a multiplicity of infection of 1, which was used as the standard multiplicity of infection in the following experiments. As shown in Fig. 3A, no significant change in the expression of miR-101 was observed after infection with negative control (Lenti-neg) in either SGC-7901 or BGC-823 cells. In contrast, the levels of miR-101 in both SGC-7901 and BGC-823 cells infected with Lenti-miR increased by 10.3-fold and 12.9-fold, respectively, as compared with Lenti-neg-infected cells. To examine the effect of miR-101 on COX-2 expression, gastric cancer cell lines were infected with Lenti-miR, and quantitative real-time PCR analysis showed that COX-2 mRNA levels markedly decreased in cells infected with Lenti-miR (Fig. 3B). The downregulation of COX-2 level by miR-101 was also confirmed with western blotting analysis (Fig. 3C,D). Additional statistical analysis demonstrated that the levels of miR-101 in SGC-7901 and BGC-823 cells correlated inversely with the levels of COX-2 mRNA (r =−0.997 and r = −0.975, respectively; P < 0.001) and COX-2 (r = −0.85 and r = −0.817, respectively; P < 0.01). These data collectively suggest that miR-101 overexpression in gastric cancer cell lines inhibits COX-2 mRNA and COX-2 expression.

image

Figure 2. Schematic representation of miRNA expression vector and miR-101 with its binding site in the COX-2 3′-UTR. (A) Complementarity of the sequences between miR-101 and the COX-2 3′-UTR. (B) The double-stranded pre-miR-101 oligonucleotides were cloned into the linearized pcDNA 6.2-GW/EmGFP-miR vector. (C) The mature miR-101 sequence inserted was confirmed by DNA sequencing.

Download figure to PowerPoint

image

Figure 3. miR-101 inhibits COX-2 expression in gastric cancer cells. The expression levels of miR-101 (A) and COX-2 mRNA (B) were measured by quantitative real-time PCR analysis. (C) COX-2 expression was detected by western blot analysis. (D) The histogram represents the expression of COX-2 relative to that of β-actin. All data (the Lenti-neg group was set to 1) represent the mean ± SD of six independent experiments. **P < 0.01 versus the Lenti-neg or the noninfected group.

Download figure to PowerPoint

Overexpression of miR-101 inhibits the growth and induces the apoptosis of gastric cancer cells

SGC-7901 and BGC-823 cells infected with Lenti-miR exhibited morphological changes such as increasingly round shape, cell volume enlargement, and the presence of vacuoles of different sizes (Fig. 4A). To determine whether miR-101 affects gastric cancer cell proliferation, cell numbers were counted at 24, 48 and 72 h after Lenti-miR infection. As shown in Fig. 4B, miR-101-transfected cells had lower rates of cell proliferation than Lenti-neg-infected and noninfected cells. Furthermore, Lenti-miR infection resulted in downregulation of proliferating cell nuclear antigen (PCNA) expression in gastric cancer cells (Fig. 4C,D). The effects of miR-101 on the apoptosis of gastric cancer cells were assessed by flow cytometric analysis at 48 h after lentiviral infection. In Fig. 5A, the lower right panels correspond to apoptotic cells that stain with annexin V-R-phycoerythrin (PE) only. miR-101 infection resulted in 5.4% and 5.6% increases in SGC-7901 and BGC-823 apoptotic cells, respectively (Fig. 5B). In addition, Lenti-miR infection resulted in downregulation of Bcl-2 expression and upregulation of Bax expressions in gastric cancer cells (Fig. 5C,D). All of these data indicate that overexpression of miR-101 is involved in both cell proliferation and apoptosis of gastric cancer cell lines.

image

Figure 4. Ectopic expression of miR-101 inhibits gastric cancer cell proliferation. (A) Cell morphology was examined by fluorescence microscopy at 48 h after lentiviral infection (original magnification, ×100). Morphological changes were seen only in Lenti-miR-infected gastric cancer cells (left panels). Magnified images of Lenti-miR-infected SGC-7901 and BGC-823 cells clearly show the presence of different-sized vacuoles (right panels). (B) Cell numbers were determined by cell count at 24, 48 and 72 h after lentiviral infection. (C) The expression of PCNA in gastric cancer cells was measured by western blot analysis at 48 h after lentiviral infection. (D) The histogram represents the expression of PCNA relative to that of β-actin. Each data point represents the mean ± SD from six independent experiments. *P < 0.05, **P < 0.01 versus the Lenti-neg or noninfected group.

Download figure to PowerPoint

image

Figure 5. Ectopic expression of miR-101 induces apoptosis in gastric cancer cells. (A) Flow cytometry-based annexin V-R-PE labeling of apoptotic cells. (B) The histogram represents apoptosis rates. (C) The expression levels of Bcl-2 and Bax in gastric cancer cells were measured by western blot analysis at 48 h after lentiviral infection. (D) The histogram represents the expression levels of Bcl-2 and Bax relative to that of β-actin. Each data point represents the mean ± SD from six independent experiments. *P < 0.05, **P < 0.01 versus the Lenti-neg or noninfected group.

Download figure to PowerPoint

miR-101 inhibits tumor growth in vivo

To further explore whether miR-101 is involved in tumorigenesis, a gastric xenograft tumor model was established. Noninfected, Lenti-miR-infected or Lenti-neg-infected SGC-7901 and BGC-823 cells were injected separately into six groups of nude mice (n = 6). As shown in Fig. 6A,B, injection of Lenti-miR-infected cells induced a significant reduction in tumor growth as compared with the mice injected with Lenti-neg-infected or noninfected cells at later time points. The average tumor weight was also notably lower in the Lenti-miR-infected group than in the Lenti-neg-infected or noninfected group (Fig. 6C). Furthermore, the levels of miR-101 in the xenografts of Lenti-miR-infected cells were much higher than those in Lenti-neg-infected or noninfected cells (Fig. 6D). In contrast, the levels of COX-2 in the xenografts of Lenti-miR infected cells were much lower than those in Lenti-neg-infected or noninfected cells (Fig. 6E,F). These data collectively indicate that miR-101 can effectively reduce tumor growth by suppressing COX-2 expression in vivo.

image

Figure 6. miR-101 inhibits gastric tumor growth in vivo. Lenti-miR-infected, Lenti-neg-infected or noninfected SGC-7901 and BGC-823 cells were injected separately into six groups of nude mice (n = 6). (A) Tumor volume was measured, and tumor growth curves were made. (B) The mice were killed at the completion of the experiment, and each tumor lump was removed from the body. (C) The mean tumor weight of six groups at the end of the experiment. (D) The expression of miR-101 in the xenografts was measured by quantitative real-time PCR analysis and normalized against U6 RNA levels. (E) COX-2 expression in the xenografts was detected by western blot analysis. (F) The histogram represents the expression of COX-2 relative to that of β-actin. Data are shown as mean ± SD from six independent experiments. *P < 0.05, **P < 0.01 versus Lenti-neg or non-infected group.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

miRNAs are small, noncoding RNAs that are approximately 22 nucleotides in size and can regulate the expression of hundreds of genes by targeting their mRNAs post-transcriptionally [4-6]. It has been reported that miRNA expression profiles differ between normal and cancer tissues [20, 25]. Although low levels of miR-101 have been identified in many malignant tumors [10, 19-21], few studies have investigated miR-101 expression in association with clinicopathological characteristics in gastric cancer. The present study has demonstrated that miR-101 is significantly downregulated in gastric cancer tissues as compared with matched normal tissues. Furthermore, we have analyzed the relationship between miR-101 expression and the clinicopathologic characteristics of patients with gastric cancer. We have clearly shown that miR-101 downregulation closely correlates with tumor staging and lymph node metastasis. Specifically, the level of miR-101 is significantly lower in patients with stage III–IV disease or lymph node metastasis than in those with stage I–II disease or without lymph node metastasis. In view of the above data, we speculate that miR-101 plays an important role in gastric cancer progression. This study has also demonstrated that COX-2 expression is significantly higher in cancer tissue than in normal tissue, and that overexpression of COX-2 correlates with both tumor staging and lymph node metastasis of gastric cancer. Our results are consistent with previous reports that overexpression of COX-2 is associated with depth of invasion and lymph node metastasis [16, 26, 27]. In addition, the inverse correlation between COX-2 mRNA/COX-2 expression and miR-101 expression seen in gastric cancer tissues is also validated in gastric cancer cell lines. Collectively, our data indicate that aberrant miR-101 expression may be a novel mechanism by which COX-2 mRNA and COX-2 expression are upregulated in gastric cancer.

It is well known that tumorigenesis results from an imbalance between cell proliferation and apoptosis, which is maintained by different signal transduction pathways. In the current study, we have observed that miR-101 overexpression significantly decreases cell proliferation and induces apoptosis in gastric cancer cell lines (SGC-7901 and BGC-823). This is further supported by the finding that the overexpression of miR-101 inhibits tumor growth in nude mice. These results support previous findings that miR-101 suppresses cell proliferation and induces apoptosis in vitro, as well as inhibiting tumor growth of liver, stomach, colon and prostate cancers in vivo [10, 20, 28-30]. Furthermore, we have shown that exogenous miR-101 significantly reduces PCNA expression in SGC-7901 and BGC-823 cells. PCNA is a nuclear protein that is synthesized in late G1 and S phases of the cell cycle, and it is used to monitor changes in cellular growth status [31, 32]. The modulation of PCNA expression is an important indicator of early changes in cellular proliferation, and provides a potential mechanism by which miR-101 inhibits gastric cancer cell proliferation. Recently, we have also demonstrated that treatment with NS-398, which is a selective COX-2 inhibitor, significantly reduces PCNA expression in human pancreatic carcinoma cells [33]. Therefore, miR-101-mediated suppression of cell proliferation is most likely attributable to inhibition of COX-2-associated PCNA expression in gastric cancer cells. Collectively, our data suggest that miR-101 may be an antioncogenic miRNA that regulates COX-2 mRNA and COX-2 expression in gastric cancer cells. In addition to gastric cancer, exogenous miR-101 has also been shown to suppress the proliferation and growth of colon cancer, endometrial serous adenocarcinoma and prostate cancer cells by directly inhibiting COX-2 expression [19, 21, 29].

It has been previously shown that apoptosis is, in part, modulated by the Bcl-2 family, including apoptosis-inhibiting gene products (Bcl-2, Bcl-XL, Mcl-1, A1, and Bcl-w) and apoptosis-accelerating gene products (Bax, Bak, Bcl-XS, and Bim) [34]. In this study, western blot analysis revealed the simultaneous downregulation of Bcl-2 expression and upregulation of Bax expression in gastric cancer cells infected with Lenti-miR. Data from both in vivo and in vitro studies have shown that upregulation of COX-2 expression reduces the apoptosis rate by upregulating of Bcl-2 expression [35, 36]. Recently, we have demonstrated that NS-398 significantly decreases Bcl-2 levels but increases Bax levels in human gastric cancer cells [17]. These results indicate that miR-101-mediated apoptosis is most likely attributable to the downregulation of COX-2 expression in gastric cancer cells.

On the basis of the degree of complementarity with the 3′-UTRs of targets, miRNAs may directly cleave the mRNAs or inhibit protein synthesis. Perfect or nearly perfect base pairing induces target mRNA cleavage, whereas imperfect base pairing induces mainly translational silencing of the targets [37]. Our bioinformatics analysis has revealed that the 3′-UTR of human COX-2 mRNA has a complementary site for the seed region of miR-101. In addition, it has been well demonstrated by luciferase activity assay in previous studies that miR-101 directly interacts with COX-2 mRNA, thereby leading to its post-transcriptional repression [19, 24]. Therefore, we did not perform any luciferase activity assay in the current study to confirm whether COX-2 is a true target of miR-101. We used quantitative real-time PCR and western blot analysis to confirm that COX-2 is a target of miR-101 in both gastric cancer tissues and cell lines. Specifically, overexpression of miR-101 in gastric cancer cells significantly reduces endogenous COX-2 expression at both the mRNA and protein levels. These results highlight the fact that miR-101 interacts with COX-2 and downregulates its expression at the transcriptional and translational levels.

In conclusion, we have demonstrated that miR-101 directly downregulates COX-2 expression in gastric cancer cell lines, and that this occurs through transcriptional and post-transcriptional repression mechanisms. Moreover, ex vivo data suggest that a lower level of miR-101 expression correlates with a higher level of COX-2 expression in gastric cancer tissues. Consequently, miR-101 deregulation in gastric cancer cells represents one of the main mechanisms underlying COX-2 overexpression in gastric cancer, and the miR-101 level in tumor tissues can also serve as an important biomarker in gastric cancer prognosis. Overexpression of COX-2 has been shown to be involved in apoptosis resistance, angiogenesis, decreased host immunity, and enhanced invasion and metastasis. Therefore, better understanding of the molecular mechanism of COX-2 downregulation by miR-101 not only provides new insights into gastric carcinogenesis and malignant progression, but also offers a novel approach for treatment of gastric cancer.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Patient tissue samples

Tumor and matched normal tissue (at least 5 cm away from primary tumor) samples were obtained from 30 patients undergoing surgery for gastric cancer at the Department of Gastrointestinal Surgery, The First Affiliated Hospital of Nanjing Medical University (Nanjing, China). The samples were placed in RNAlater (Ambion, Austin, TX, USA) in the operating room, and subsequently frozen at −80 °C until RNA extraction. None of the patients received any nonsteroidal anti-inflammatory drugs, chemotherapy or irradiation prior to operation. All tumor samples were reviewed by a pathologist and histologically confirmed to be adenocarcinomas, which were classified on the basis of the modified World Health Organization classification system [38]. All patients provided informed consent to participate in the study, in accordance with the Helsinki Declaration and the ethical guidelines of Nanjing Medical University.

Cell lines and culture

Human gastric cancer cell lines SGC-7901/moderately differentiated and BGC-823/poorly differentiated (Cell Bank of Chinese Academy of Sciences, Shanghai, China) were grown in RPMI 1640 (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum and penicillin/streptomycin. Human embryonic kidney cells (HEK-293FT; Shanghai Institute of Cell Biology) were maintained in DMEM (Gibco) supplemented with 10% fetal bovine serum and penicillin/streptomycin/geneticin. All cells were incubated at 37 °C in a humidified incubator in an atmosphere of 5% CO2.

Lentiviral vector construction and cell infection

Lentiviral vectors for the delivery of miRNA were designed and made with reagents and protocols included in the BLOCK-iT Pol II miR RNAi Expression Vector Kit (Invitrogen, Carlsbad, CA, USA). The pre-miR-101 (5′-TGCTG TACAG TACTG TGATA ACTGA AGTTT TGGCC ACTGA CTGAC TTCAG TTAAC AGTAC TGTA-3′; the mature miR101 sequence is underlined, and the complementary sequence to the 3′-UTR of COX-2 mRNA is in bold) and its reverse complement (5′-CCTGT ACAGT ACTGT TAACT GAAGT CAGTC AGTGG CCAAAACTTC AGTTA TCACA GTACT GTAC-3′) were annealed and inserted into the pcDNA 6.2-GW/EmGFP-miR expression vector, which contains the full pre-miRNA 5′-flanking and 3′-flanking regions, as well as the cocistronic Emerald GFP (EmGFP) gene (Fig. 2B). The pDONR 221 vector was used as an intermediate to transfer the pre-miRNA expression cassette into the lentiviral expression plasmid (pLenti6/V5-DEST), with Gateway Technology (Invitrogen). The mature miR-101 sequence inserted was confirmed by DNA sequencing (Fig. 2C). pLenti6/v5-GW/EmGFP-miR101 vector (3 μg) and ViraPower Packaging mix (9 μg) were cotransfected into HEK-293FT cells with Lipofectamine 2000 (Invitrogen). At 24 h post-transfection, the culture media as virus stocks were harvested, centrifuged at 1800 g for 10 min, and filtered through a 0.45-μm poly(vinylidene difluoride) Durapore membrane (Millipore, Billerica, MA, USA) to remove any nonadherent packaging cells. SGC-7901 and BGC-823 cells were infected with the recombinant lentiviruses, and stable cell lines were selected with blasticidin (3 μg·mL−1; Invitrogen) for 2 weeks.

RNA isolation and quantitative real-time PCR

RNA was extracted from gastric cancer tissues and cell lines with Trizol reagent (Invitrogen), according to the manufacturer's instructions, and this was followed by reverse transcription with RevertAi First Strand cDNA Synthesis Kits (Invitrogen). miRNAs were isolated with the mirVana RNA isolation kit (Ambion), and reverse transcribed with the TaqMan miRNA Reverse Transcription kit (Applied Biosystems, Foster City, CA, USA), according to the manufacturer's instructions. Mature miR-101 expression was measured with a stem–loop RT-PCR technique, as previously described [39]. The expression levels of miR-101 and COX-2 mRNA were analyzed by RT-PCR with an ABI 7500 thermocycler (Applied Biosystems), and cycle threshold values were determined with the manufacturer's software. miR-101 levels were normalized against U6 RNA (Applied Biosystems), and COX-2 mRNA levels were normalized against glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA. The relative expression levels were calculated with the 2∆∆CT method [40]. All primers and parameters for reverse transcription and RT-PCR are summarized in Table 2.

Table 2. Sequences of pre-miR-101 oligonucleotide and primers in the study. A scrambled oligonucleotide that has been validated as not producing identifiable effects on known miRNA function was used as a negative control
OligonucleotidesSequence (5′- to 3′)
Pre-miR-101 forward TGCTGTACAGTACTGTGATAACTGAAGTTTTGGCCACTGACTGACTTCAGTTAACAGTACTGTA
Pre-miR-101 reverse CCTGTACAGTACTGTTAACTGAAGTCAGTCAGTGGCCAAAACTTCAGTTATCACAGTACTGTAC
Negative control forward TGCTGAAATGTACTGCGTGGAGACGTGTTTTGGCCACTGACTGACTTCAGT TAACAGTACTGTA
Negative control reverse CCTGTACAGTACTGTTAACTGAAGTCAGTCAGTGGCCAAAACACGTCTCCACGCAGTACATTTC
Quantitative real-time PCR primers
U6 forward CTCGCTTCGGCAGCACATA
U6 reverse AACGCTTCACGAATTTGCGT
GAPDH forward CACTGGCGTCTTCACCACCAT
GAPDH reverse GTGCAGGAGGCATTGCTGAT
COX-2 forward CTGGCGCTCAGCCATACAG
COX-2 reverse CACCTCGGTTTTGACATGGGT

Western blot analysis

Preparation of cell or tissue lysates and SDS/PAGE analysis were performed as previously described [41, 42]. The antibodies used includes the mAbs mouse anti-COX-2 Ig, anti-Bcl-2 Ig and anti-Bax Ig, and the polyclonal antibodies anti-PCNA Ig (1 : 1000 dilution; Santa Cruz Biotechnology, Santa Cruz, CA, USA) and mouse anti-β-actin Ig (1 : 500 dilution; Sigma Chemical Co., St Louis, MO, USA). Bands were quantified with densitometric image analysis software (quantity one; Bio-Rad, Hercules, CA, USA). The relative expression levels of COX-2, PCNA, Bcl-2 and Bax were normalized to that of β-actin.

Cell growth and apoptosis assays

SGC-7901 and BGC-823 cells were infected with Lenti-miR, and viable cell numbers were determined at 24, 48 and 72 h postinfection with a hemocytometer. Morphological changes of the infected cells were evaluated with a fluorescence microscope (Eclipse E-800; Nikon, Tokyo, Japan). For quantitative analysis of apoptosis, cells were harvested at 48 h after infection, washed with NaCl/Pi, and stained with annexin V-R-PE and 7-aminoactinomycin D with the ApoScreenAnnexinVApoptosis Detection Kit (Southern Biotech, Birmingham, AL, USA). The cells were then subjected to flow cytometry according to the manufacturer's instructions, and the stained cells were analyzed with a FACScan flow cytometer (Becton Dickinson, San José, CA, USA).

Tumorigenicity assays in nude mice

Four-week-old BALB/c athymic nude mice (Shanghai SLAC Laboratory Animal Co. Ltd, Shanghai, China) were divided into six groups of six mice each, and injected subcutaneously with 2 × 106 cells in 0.15 mL of NaCl/Pi into their right axillary fossa at a single site. The tumor size was measured with calipers at 2-day intervals from the 10th day to the 20th day (for SGC-7901 cells) or from the 14th day to the 24th day (for BGC-823 cells) after inoculation. The tumor volume (cm3) was calculated with the following formula: (shortest diameter)2 × (longest diameter) × 0.5. The mice were killed at the completion of the experiment, and the implanted tumors were removed and weighed. All experimental procedures were performed in accordance with the Guide for the Care and Use of Laboratory Animals (NIH publication no. 80-23, revised 1996) and the institutional ethical guidelines for animal experiments.

Statistical analysis

Statistical analyses were performed with spss version 11.0 (SPSS, Chicago, IL, USA). The relationships between the clinicopathological characteristics and the expression of miR-101 or COX-2 were examined with the χ2-test, the Wilcoxon matched pair test, or the Kruskal–Wallis nonparametric test. The correlation between miR-101 and COX-2 levels was determined with the Spearman rank correlation coefficient. The data are shown as mean ± standard deviation (SD) or mean ± standard error of the mean (SEM), and were analyzed by ANOVA with Dunnett's multiple comparison tests. A P-value of < 0.05 was considered to be statistically significant.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We thank C. Wang (The First Affiliated Hospital of Nanjing Medical University, Nanjing, Jiangsu Province, P. R. China) for insightful discussions on the area relevant to this study. The present study was supported by the Nature Science Foundation of China (NSFC), No. 81072030, and a grant from the scientific research foundation for outstanding medical talent, Health Bureau of Jiangsu Province, No. RC207046.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References