SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Survival of patients with hepatocellular carcinoma (HCC) remains poor, which is largely attributed to active angiogenesis. However, the mechanisms underlying angiogenesis in HCC remain to be discovered. In this study, we found that long noncoding RNA associated with microvascular invasion in HCC (lncRNA MVIH) (lncRNA associated with microvascular invasion in HCC) was generally overexpressed in HCC. In a cohort of 215 HCC patients, the overexpression of MVIH was associated with frequent microvascular invasion (P = 0.016) and a higher tumor node metastasis stage (P = 0.009) as well as decreased recurrence-free survival (RFS) (P < 0.001) and overall survival (P = 0.007). Moreover, the up-regulation of MVIH served as an independent risk factor to predict poor RFS. We also found that MVIH could promote tumor growth and intrahepatic metastasis by activating angiogenesis in mouse models. Subsequent investigations indicated that MVIH could activate tumor-inducing angiogenesis by inhibiting the secretion of phosphoglycerate kinase 1 (PGK1). Additionally, in 65 HCC samples, MVIH expression was inversely correlated with the serum level of PGK1 and positively correlated with the microvessel density. Conclusion: Deregulation of lncRNA MVIH is a predictor for poor RFS of HCC patients after hepatectomy and could be utilized as a potential target for new adjuvant therapies against active angiogenesis. (HEPATOLOGY 2012;56:2142–2153)

Hepatocellular carcinoma (HCC) is currently the fifth-most common solid tumor worldwide and the second leading cause of cancer-related deaths in China.1, 2 Although remarkable progress has been made in recent decades, the details of the molecular mechanisms underlying HCC carcinogenesis remain to be elucidated.2, 3 Survival of patients with HCC has been improved with advancements in surgical techniques, but the median survival rate remains at approximately 50% (range, 17-69) after 5 years.4 This unfavorable prognosis is mainly because HCC is a highly vascularized type of tumor with frequent intra- or extrahepatic metastases. Blood vessels within tumors produced by angiogenesis are responsible for the poor survival of HCC patients.3, 5 Cancer classification using biomarkers may effectively define risk of recurrence, which allows for the use of appropriate treatments to acquire a better prognosis.6 But, to date, few measurable biomarkers for predicting HCC recurrence have been identified.

Long noncoding RNAs (lncRNAs) represent a subgroup of noncoding RNAs that are longer than 200 nucleotides. Recently, studies have shown that more genomic sequence is transcribed into lncRNAs than protein-coding RNAs.7, 8 Additionally, multiple studies have indicated that significant numbers of lncRNAs are involved and might play central roles in a variety of biological processes through complicated mechanisms.9-13 Notably, the deregulation of lncRNAs has also been shown to result in aberrant gene expression that contributes to the progression of a variety of human tumors, including HCC.14-17 However, compared with protein-coding genes, the clinical significance and functions of most deregulated lncRNAs in the progression and aggressiveness of HCC remain unknown.

In the present study, we report an lncRNA termed MVIH (lnc RNAs associated with microvascular invasion in HCC) (NCBI no.: AK094613) that is up-regulated in tumor tissues, compared to the corresponding noncancerous tissues, and correlates with the microvascular invasion of HCC. Moreover, MVIH serves as an independent risk factor for HCC patients' poor recurrence-free survival (RFS) after hepatectomy. Our results indicate that MVIH could promote tumor growth and intrahepatic metastasis in vivo. Furthermore, we demonstrate that the inhibition of phosphoglycerate kinase 1 (PGK1) secretion by its association with MVIH contributes to active angiogenesis both in vitro and in vivo.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Patients and Clinical Samples.

Forty fresh HCC tissues and paired adjuvant noncancerous tissue samples were randomly selected from patients undergoing hepatectomy at the Eastern Hepatobiliary Surgery Hospital (Shanghai, China) between May 1 and June 30, 2009. These tissues were used for quantitative real-time polymerase chain reaction (qRT-PCR) analysis. Another 215 fresh HCC tissues were collected from the same bank1 and used for survival analysis and further validation. Fresh tissue and serum samples were collected in the operating room and processed immediately within 15 minutes. Each sample was frozen and stored at −80°C. Paired noncancerous tissues were isolated from at least 2 cm away from the tumor border and were shown to lack tumor cells by microscopy. All patients in this study met the following inclusion criteria: Resected nodules were identified as HCC by pathological examination; no anticancer treatments were given before surgery; complete resection of all tumor nodules was verified by the cut surface being free of cancer by pathological examination; and complete clinical-pathologic and follow-up data were available. Patients that died of nonliver diseases or accidents were excluded. Clinical characteristics of all patients are listed in Supporting Table 1. Tumor differentiation was defined according to Edmondson's grading system.2 Micrometastases were defined as tumors adjacent to the border of the main tumor that were only observed under the microscope. Tumor staging was defined according to the sixth edition of the tumor node metastasis (TNM) classification system published by the International Union Against Cancer and the Barcelona Clinic Liver Cancer (BCLC) staging system. The study was approved by the institutional review board of Eastern Hepatobiliary Surgery Hospital. All patients gave their written informed consent to participate in the study. The data do not contain any information that could identify patients.

Detailed description of Patients and Methods can be found in the online Supporting Materials.

Results

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

The Novel lncRNA MVIH Is Up-regulated in HCC.

We previously identified systemic variations in the expression of lncRNAs between hepatitis B virus (HBV)-related HCC and paired nontumor samples using a microarray analysis.14 From that study, we noted that lncRNA MVIH was remarkably up-regulated (fold change, 3.654; P = 0.00205) in HCC according to the microarray data. (The microarray data discussed in that article have been deposited in NCBI Gene Expression Omnibus and are accessible through GEO Series accession number GSE27462;,http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE 27462.) To further validate this result, we analyzed MVIH expression using qRT-PCR in 40 pairs of predominately HBV-related HCC and the corresponding peritumoral tissues (Supporting Table 1, cohort 1) and found that MVIH was significantly up-regulated (P = 0.007) in HCC (Fig. 1A). We also noted that MVIH was located within the intron of the ribosomal protein S24 (RPS24) gene and overlapped with part of the 3′ end of the RPS24 transcript (Fig. 1B). To explore the potential relationship of the MVIH and RPS24 transcripts, we first examined the expression levels of RPS24 and MVIH in 40 HBV/HCC tissues (cohort 1). The results showed that no correlation (r = 0.060; P = 0.695) existed between the transcriptional levels of RPS24 and MVIH (Fig. 1C). Furthermore, lncRNA MVIH was statistically unchanged in SMMC-7721 and HCCLM3 cells transfected with two different short interfering (siRNAs) (designated as si-1 and si-2) against RPS24, despite significant reduction in RPS24 messenger RNA expression (Fig. 1D). Next, we performed a rapid amplification of complementary DNA ends (RACE) analysis to identify the 5′ and 3′ ends of the MVIH transcript. To amplify MVIH in a specific manner, we used intronic primers that would not recognize RPS24 (Fig. 1B). The full sequence of lncRNA MVIH is presented in Supporting Fig. 1. To determine whether the transcript of MVIH might encode proteins, we first predicted open reading frames (ORFs) of MVIH, and using an ORF finder from the National Center for Biotechnology Information (Bethesda, MD), several ORFs were predicted from lncRNA MVIH, and only one showed homologous protein with 56% identity in human. Second, we used a codon substitution frequency (CSF) analysis using PhyloCSF. We found that lncRNA MVIH had a very low CSF score (Supporting Fig. 2), which indicated that the transcript of lncRNA MVIH had no protein-coding potential. In conclusion, the overexpressed lncRN MVIH in HCC is a noncoding RNA and is transcribed independently of the RPS24 gene.

thumbnail image

Figure 1. lncRNA MVIH is up-regulated in HCC and transcribed independently of RPS24. (A) lncRNA MVIH is significantly up-regulated in 40 human HCC tissues, compared to the corresponding noncancerous tissues. Statistical differences were analyzed using the paired t test. Horizontal lines in box plots represent the median, boxes represent the interquartile range, and whiskers represent the 2.5th and 97.5th percentiles. (B) Schematic representation of the location of lncRNA MVIH within the RPS24 gene is shown. Exons and intron of the RPS24 gene are indicated by black boxes and black lines, and lncRNA MVIH is depicted as the gray box. The location of the forward (F) and reverse (R) primers used for RT-PCR and nested PCR are shown. Orientation of arrows indicates the direction of the transcription or amplification reaction. (C) Linear correlation between the expression of the MVIH and RPS24 was not observed. ΔCt values were used to measure gene expression, which was normalized by β-actin expression levels. (D) Expression of RPS24 was decreased, and MVIH was not significantly changed in SMMC-7721 and HCCLM3 cells with two different siRNAs against RPS24. Statistical differences were analyzed relative to siRNA control. ★P < 0.05.

Download figure to PowerPoint

The Association of lncRNA MVIH With Clinicopathological Characteristics and Prognosis After Hepatectomy.

We next sought to determine whether lncRNA MVIH expression level in HCC was associated with specific clinicopathological characteristics. We measured MVIH expression level in tumor tissues from another 215 HCC patients independent from 40 HCC patients of cohort 1 (Supporeing Table 1, cohort 2) by qRT-PCR. Median expression level was used as the cutoff. Low MVIH expression in 107 patients was classified as values below the 50th percentile (with an average ΔCt expression value of 6.863, ranking from 5.915 to 8.318, when compared with β-actin). High MVIH expression in 108 patients was classified as values above the 50th percentile (with an average ΔCt expression value of 4.149, ranking from 2.249 to 5.526, when compared with β-actin). We found that a higher MVIH expression level was significantly more frequent in tissues with increased microvessel invasion (χ2 = 6.267; P = 0.016) and advanced TNM tumor node metastasis T stage (χ2 = 9.332; P = 0.009; Table 1).

Table 1. Clinical Characteristics of 215 HCC Patients According to lncRNA MVIH Expression Level
VariablelncRNA MVIH Subgroup*P Value
LowHigh
  • Data are expressed as ratios.

  • Abbreviations: HBsAg, hepatitis B surface antigen; HBeAg, hepatitis B e antigen; ALT, aalanine aminotransferase.

  • *

    Median expression level was used as the cutoff. Low expression of lncRNA MVIH in 107 patients was classified as values below the 50th percentile. High lncRNA MVIH expression in 108 patients was classified as values at or above the 50th percentile.

  • P < 0.05 by the chi-squared test.

All cases107108 
Age, years, >55:≤5548:5935:730.069
Gender, male/female94:1385:230.099
HBsAg, positive/negative97:10100:80.632
HBeAg, positive,negative31:7637:710.464
Liver cirrhosis, with/witout64:4376:320.117
Serum albumin (g/L), >40:≤4077:3082:260.537
Serum bilirubin, μmol/L, >17:≤1723:8434:740.122
ALT, U/L, >40:≤4045:6254:540.275
AFP, ug/L, >20:≤2073:3485:230.091
Tumor size, cm, >5:≤530:7744:640.062
No. tumor, solitary/multiple83:2490:180.307
Edmondson's grade, I+II:III35:7234:740.884
Microvascular invasion, present/absent23:8440:680.016
Macrovascular invasion, present/absent8:9910:980.806
Micrometastases, present/absent43:6444:641.000
Encapsulation, no/complete51:5643:650.273
TNM stage, I:II:III78:16:1357:29:220.009*
BCLC stage, A:B:C83:16:881:17:100.873

We further examined whether MVIH expression level correlated with outcome of HCC patients after hepatectomy. Kaplan-Meier's survival curves were used to compare the low (n = 107) and high (n = 108) lncRNA MVIH subgroups, and the results are presented in Fig. 2A,B. Remarkably, patients with higher MVIH expression level had poorer RFS (P < 0.001) and overall survival (OS; P = 0.007).

thumbnail image

Figure 2. lncRNA MVIH could be an independent prognostic factor to predict RFS and OS. (A and B) Kaplan-Meier's analyses of correlations between lncRNA MVIH expression level and (A) RFS or (B) OS of 215 HCC patients is shown. (C) Multivariate analysis of HRs for RFS and OS is shown. HRs are presented as the means (95% CI). (D and E) Kaplan-Meier's analyses of correlations between MVIH expression level and (D) RFS or (E) OS of 164 early-HCC patients. Survival curves were compared using a long-rank test, and variables included in the multivariate analysis were selected using a univariate analysis.

Download figure to PowerPoint

A univariate analysis revealed that the gender, alpha-fetoprotein (AFP), tumor size, Edmondson's grade, microvascular invasion, macrovascular invasion, pathological satellite, encapsulation, TNM stage, BCLC stage, and MVIH expression level were significantly correlated with RFS or OS (Supporting Table 3). All 11 clinicopathological characteristics were further applied for multiple analyses. Cox's multivariate proportional hazards model indicated that tumor size (hazard ratio [HR]:2.541, 95% confidence interval [CI]: 1.519-4.254; P < 0.001), pathological satellite (HR: 2.365, 95% CI: 1.476-3.789; P < 0.001), MVIH expression level (HR: 1.987, 95% CI: 1.314-3.004; P = 0.001), and AFP (HR: 1.837, 95% CI: 1.067-3.162; P = 0.028) were independent risk factors for RFS. Intriguingly, only MVIH expression level (HR: 1.854, 95% CI: 1.050-3.272; P = 0.033) and tumor size (HR: 3.607, 95% CI: 1.632-7.971; P = 0.001). Pathological satellite was the same meaning as and supposed to be revised as micrometastase labeled in the figure 2C were significantly independent factors that affected the OS of HCC patients after hepatectomy (Fig. 2C).

Patients categorized with an early stage of HCC measured by the BCLC staging system have been shown to benefit from radical therapies, including hepatectomy.19 Therefore, we wanted to investigate whether MVIH expression level is associated with the prognosis of early-stage patients. Patients (n = 164) were classified as stage A according to the BCLC classification system. Interestingly, a higher MVIH expression level was consistently indicative of a shorter RFS (Fig, 2D; P = 0.003); however, the correlation between a higher MVIH expression level and OS were not statistically significant (Fig. 2E; P = 0.061). Taken together, these data indicate that MVIH expression level is an independent risk factor for HCC patients' recurrence of HCC patients.

lncRNA MVIH Promotes Tumor Growth and Intrahepatic Metastasis In Vivo.

To evaluate the biological functions of lncRNA MVIH, we first examined the expression of MVIH in a variety of cell lines, including SMMC-7721, LO2, Huh7, HepG2, Hep3B, HCCLM3, and HCC97H, by qRT-PCR (Supporting Fig. 3A). To examine the biological significance of MVIH on tumor growth, we subcutaneously (SC) implanted firefly luciferase-labeled HCCLM3 cells transfected with lv-lncRNA-MVIH or lv-NC (Supporting Fig. 3B). We found that overexpression of lncRNA MVIH significantly promoted the growth of HCCLM3 cells (Fig. 3A,B). We then used SC tumor tissues from six tumors in each group (Fig. 3A, arrows) formed by HCCLM3 cells to establish orthotopic implanted models (12 mice in each group). Compared to the Lv-NC group, MVIH overexpression resulted in significantly frequent intrahepatic metastastic focis (5.17 ± 1.14 versus 2.58 ± 0.72; P = 0.005) and increased maximal tumor diameters (6.08 ± 1.56 versus 3.48 ± 1.02; P < 0.001) (Fig. 3C). Furthermore, HCCLM3 cells were injected through the splenic hilum to establish a liver metastasis tumor model. Nine of twelve mice injected with HCCLM3 cells with up-regulated MVIH and 4 of 12 mice in the control group showed luciferase signals 5 weeks after cell injections. Additionally, liver injected with HCCLM3 cells with up-regulated MVIH showed increased luciferase strength before mice were sacrificed (Supporting Fig. 3C). Then, livers were applied for hematoxylin and eosin (H&E) staining. We found that livers injected with HCC cells with lv-lncRNA-MVIH showed significant frequent intrahepatic metastasis (P = 0.037; Fig. 3D,E). No pulmonary and brain foci were detected in orthotopic implanted models and intrahepatic metastastic models before mice were sacrificed. Together, these results indicate that the overexpression of MVIH might enhance tumor growth and intrahepatic metastasis, which might potentially lead to a poor survival after surgery.

thumbnail image

Figure 3. Promotion of tumor growth and intrahepatic metastasis in vivo. (A) Photographs of tumors that developed in SC mouse models from HCCLM3/lncRNA/MVIH and HCCLM3/NC are presented. Tumors of lv-NC indicated by red arrows were not applied for orthotopic models, and the two smallest tumors of Lv-lncRNA-MVIH were indicated by black arrows. (B) Effects of lncRNA MVIH expression in SC mouse models are shown. Data are shown as the means ± SD. The statistical difference was analyzed by the two-sample t test. (C) Representative livers from orthotopic implanted mice of the lv-lncRNA-MVIH group and lv-NC group are shown in the right panel. Tumor number and maximal tumor diameters are compared between the lv-lncRNA-MVIH group and the lv-NC group are shown in the left panel. Data are shown as means ± SD. Statistical differences were analyzed by the two-sample t test. (D) Representative tumors by H&E staining for HCCLM3 control and MVIH overexpressing transplants in intrahepatic metastastic models. (E) Livers in lv-lncRNA-MVIH displayed frequent intrahepatic metastastic foci, compared to the control group. Horizontal lines in box plots represent the median, boxes represent the interquartile range, and whiskers represent the 25th and 75th percentiles. Statistical difference was analyzed by Mann-Whitney's U test. ★P < 0.05. SD, standard deviation.

Download figure to PowerPoint

lncRNA MVIH Associates With PGK1.

Recent studies have reported that lncRNAs usually perform their molecular functions through binding to specific proteins.12, 14 Therefore, we wanted to determine whether lncRNA MVIH played a key role in promoting tumor growth and metastasis through this mechanism. An RNA pull-down experiment was performed to identify proteins that associated with MVIH with its antisense strand as the negative control. We resolved the RNA-associated proteins on a sodium dodecyl sulfate/polyacrylamide gel electrophoresis (SDS-PAGE) gel, cut out the band specific to lncRNA-MVIH, and subjected them to mass spectrometry (MS). MS analysis of the band specifically bound to lncRNA MVIH revealed that PGK1 was specifically associated with MVIH (Fig. 4A; Supporting Table 4). The associated proteins from two independent RNA pull-down assays were used to further verify the association between PGK1 and MVIH. Among all five proteins identified by MS, only PGK1 was detected by western blotting with proteins from two independent RNA pull-down assays. The EZH2 protein, which has been reported to physically associate with lncRNAs and has been validated to have no association with lncRNA MVIH by RNA immunoprecipitation analysis (Supporting Fig. 4A), was included as a negative control in the western blotting assay (Fig. 4B).

thumbnail image

Figure 4. lncRNA MVIH physically associates with PGK1. (A) SDS-PAGE gel shows proteins bound to lncRNA MVIH and antisense RNA. The band highlighted by the arrow was submitted for MS analysis and was identified to be the PGK1 protein. (B) Western blotting analysis of two independent RNA pull-down assays from SMMC-7721 cellular extracts or an HCC tissue was performed. An Ab against the EZH2 protein was used as the negative control. (C and D) RIP experiments were performed using an Ab against PGK1 on extracts from SMMC-7721 cells or two fresh HCC tissues. Purified RNA was used for qRT-PCR using the SYBY-green method, and enrichment of lncRNA MVIH was normalized to input. Data are shown as means ± standard deviation.

Download figure to PowerPoint

To further validate the association between the lncRNA MVIH and PGK1, we used a PGK1 antibody (Ab) to perform a radioimmunoprecipitation (RIP) assay on SMMC-7721 cellular extracts and two fresh HCC tissues. Consistently, we detected a significantly higher enrichment level of lncRNA MVIH with the PGK1 Ab (Fig. 4C,D), compared to the immunoglobulin G control.

lncRNA MVIH Activates Angiogenesis by Inhibiting the Secretion of PGK1.

Next, we wanted to determine whether the association between lncRNA MVIH and PGK1 had functional significance. PGK1 can be secreted by tumor cells and inhibit angiogenesis,18 which is critical for tumor growth and metastasis. Therefore, we hypothesized that MVIH might regulate tumor-inducing angiogenesis by modifying the secretion of PGK1. To test this hypothesis, we examined the concentration of PGK1 by an enzyme-linked immunosorbent assay (ELISA) assay on the tumor-cell–conditioned medium (TCM) from the Huh7 and HCCLM3 cell lines transfected with lv-lncRNA-MVIH or lv-NC (Supporting Fig. 3B). We found that the PGK1 concentrations in the TCM from cells with up-regulated MVIH were significantly lower than the comparable negative control (Fig. 5A). To further study the functional significance of this finding, an in vitro capillary tube-formation assay was utilized. With the TCM from the tumor cells transfected with lv-lncRNA-MVIH, human umbilical vein endothelial cells (HUVECs) developed more capillary-like structures than the negative controls (Fig. 5B,C). Therefore, we speculated that active angiogenesis facilitated by lncRNA MVIH might be responsible for the increased tumor growth and intrahepatic metastasis in vivo. To further test this hypothesis, the tumor tissues from the SC mouse model were applied for an immunohistochemistry (IHC) analysis with an Ab against CD34. The results indicated that HCCLM3 cells overexpressing MVIH developed tumors with a higher microvessel density (MVD) than the negative controls (10.6 ± 3.0 versus 5.5 ± 2.5; P = 0.019; Fig. 5D,E).

thumbnail image

Figure 5. lncRNA MVIH regulates angiogenesis by inhibiting the secretion of PGK1. (A) Concentration of PGK1 in conditioned medium from Huh7 or HCCLM3 cells was measured by an ELISA assay. The HCC cell line without lentivirus incubation was designated as the mock treatment. Data are shown as means ± SD. Statistical differences were analyzed by the two-sample t test. (B and C) Up-regulated MVIH could activate HCC cells to promote the tube formation of HUVECs. Representative photographs of tube formation (B) and the number of branch points (C) are presented. SFM, serum-free medium. (D) Representative photographs of an IHC analysis of tumors from SC models are presented, and the MVD of tissues was calculated (E). Data are shown as means ± SD. Statistical differences were analyzed by the two-sample t test. All groups were compared to the lv-NC control group for statistical comparisons. ★P < 0.05.

Download figure to PowerPoint

The PGK1 protein is known to be an enzyme of glycolysis. Therefore, we wondered whether the association between MVIH and PGK1 would play an important role in this process. To test this idea, Huh7 and HCCLM3 HCC cell lines with up-regulated MVIH and negative controls were harvested to examine the transcriptional levels of glucose transporter 1, hexokinase 2, pyruvate kinase isozyme M2, phosphofructokinase, and aldolase B, fructose-biphosphate, which are key enzymes for glycolysis.20, 21 However, no significant aberrant expression of the tested genes was detected (Supporting Fig. 2B). Next, we examined the levels of lactate acid production in the Huh7 and HCCLM3 cells described above, and the results revealed that lactate acid levels were unchanged (Supporting Fig. 2C). Collectively, we concluded that lncRNA MVIH could weaken the function of PGK1 in angiogenesis inhibition by reducing the secretion of PGK1, but could not regulate the function of PGK1 in glycolysis.

lncRNA MVIH Expression Level Inversely Correlates With Serum Level of PGK1 and Positively Correlates With MVD in Clinical Samples.

We further measured MVD and serum levels of PGK1 in 65 HCC tissues with well-preserved serum samples from cohort 2 by IHC and ELISA assays, respectively (cohort 3; Supporting Table1). Additionally, we analyzed the associations among lncRNA MVIH expression level, angiogenesis, and serum level of PGK1. Intriguingly, patients with a higher lncRNA MVIH expression level in primary HCC tissues showed a substantial down-regulation of serum PGK1 (Fig. 6A; high lncRNA-MVIH versus low lncRNA-MVIH = 9.589 ± 2.351 versus 7.121 ± 3.462; P = 0.001). Furthermore, MVIH expression was linearly correlated with the serum level of PGK1 (Fig. 6B; r = −0.404; P < 0.001). Additionally, MVIH expression was inversely correlated with MVD in tissues (high lncRNA-MVIH versus low lncRNA-MVIH = 27.6 ± 11.9 versus 16.4 ± 9.1; P < 0.001) (Fig. 6C,D). Thus, the up-regulation of MVIH is associated with a high MVD in primary HCC tissues and a low PGK1 level in serum. These results further support the hypothesis that lncRNA MVIH could activate tumor-inducing angiogenesis by reducing the secretion of PGK1.

thumbnail image

Figure 6. Correlation of lncRNA MVIH expression with serum level of PGK1 and MVD in HCC patients. (A) Serum levels of PGK1 were significantly lower in HCC patients with higher MVIH expression levels. Data are shown as means ± SD. Statistical significance was calculated using the two-sample t test. (B) Relative expression of MVIH was plotted against the serum level of PGK1. (C) Representative photographs of the IHC analysis are shown, and lncRNA MVIH expression level and serum level of PGK1 are displayed in the bottom panel. Three representative patients with relative low, median, and high MVIH expression were designated as sample 1, sample 2, and sample 3. (D) MVD was compared between the low and high MVIH-expressing subgroups of the 65 patients. Data are shown as means ± SD. Statistical significance was calculated using the two-sample t test. SD, standard deviation.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Here, we identified a new lncRNA transcript (MVIH), which was significantly up-regulated in HCC. A majority of the MVIH sequence is located within the intron of the RPS24 gene, which encodes a protein belonging to the S24E family of ribosomal proteins.22 Despite the interrelated location of MVIH within the RPS24 gene, these two genes are independently transcribed. The independent regulation of the two transcripts suggests that MVIH might play a role in biological behaviors independent of RPS24.

With a cohort of 215 randomly selected HCC patients, we determined that the up-regulation of MVIH was significantly associated with numerous cancer phenotypes, including microvascular invasion (P = 0.016), frequent recurrence (P < 0.000), and cancer-related death (P = 0.007). A multivariate analysis revealed that MVIH expression level was an independent risk factor for RFS (HR, 1.987;, 95% CI: 1.314-3.004; P = 0.001) and OS (HR, 1.854;, 95% CI: 1.050-3.272; P = 0.033) after surgery. The molecular classifications of HCC using coding genes have identified an abundance of valuable biomarkers23, 24; however, a poor overlap exists between these biomarkers of HCC patients. Using a combination of different types of transcripts as prognostic indicators might be a resolution to establish more-accurate prognostic gene signatures.25 Recently, growing evidence has indicated that noncoding RNAs, predominantly microRNAs (miRNAs), could predict the outcomes of HCC patients.26, 27 Given the fact that lncRNAs are more abundant than miRNAs,13 we speculated that numerous lncRNAs with the potential to serve as prognostic indicators remain to be elucidated.

Furthermore, we found that MVIH overexpression could predict the frequent recurrence (P = 0.003) of early-stage HCC patients. Therefore, our data indicate that MVIH might be an attractive biomarker for risk prognostication and personal therapy screening of HCC patients after hepatectomy. Overexpression of MVIH might be a warning sign that postoperative patients should be closely monitored and receive appropriate adjuvant therapies. The failure to predict overall survival of the panel of early-stage HCC patients (low lncRNA-MVIH/high lncRNA-MVIH = 84.14%: 73.17%; P = 0.061) might be attributed to the short follow-up (range, 4-30 months). Further validation of this result by a prolonged follow-up of this patient cohort or another cohort of HCC patients should be undertaken.

We then explored that HCC cells with overexpression of MVIH might play a critical role in tumor growth and intrahepatic metastasis in vivo. Interestingly, we found that up-regulated MVIH could not affect the proliferation or migration of HCC cells in vitro (data not shown). Considering that MVIH expression was correlated with microvessel invasion in HCC samples, we designed subsequent experiments to determine the role of lncRNA MVIH in angiogenesis.

PGK1 is a glycolytic enzyme that catalyzes the conversion of 1.3-diphosphoglycerate to 3-phosphoglycerate. In addition to influencing glycolysis, PGK1 can be secreted by tumor cells and contributes to the inhibition of angiogenesis.18 The angiogenesis switch is activated surprisingly early during the multistage advancement of HCC tumors and is essential for HCC growth and metastasis.3 Therefore, we speculated that a negative regulation of PGK1 should exist and be activated to promote tumor progression. In this study, we revealed that the up-regulation of MVIH significantly diminished the secretion of PGK1 of Huh7 and HCCLM3 cells (Fig. 5A). The tumor cells transfected with lv-lncRNA-MVIH generated TCM with an elevated ability to promote HUVECs to form vessel branches. Tumors formed by HCCLM3 cells with up-regulated MVIH showed more microvessels in the nude mouse models, compared to controls. Consistently, HCC patients with up-regulated MVIH also showed a high MVD and serum level of PGK1 (Fig. 6A,B). These results suggest that up-regulated MVIH may account for the diminished secretion of PGK1 and, in turn, activates angiogenesis as well as the accelerated growth and metastasis of HCC. Therefore, down-regulating MVIH expression may serve as a new potential strategy for adjuvant therapy against angiogenesis. Although our results revealed that MVIH could reduce the serum level of PGK1, the mechanisms underlying this process are still unclear and will require further hypothesis-driven studies.

In conclusion, we revealed that MVIH plays a key role in activating angiogenesis. Our results therefore suggest that lncRNA MVIH might be a predictor for HCC patients' poor RFS after hepatectomy and a therapeutic target for the inhibition of angiogenesis in HCC.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

The authors acknowledge Jie Wang for his detailed bioinformatics analysis of MVIH and Xiao-Fei Ye and Ya-Lin Sun for their professional statistical analysis.

References

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information
  • 1
    Thomas MB, Jaffe D, Choti MM, Belghiti J, Curley S, Fong Y, et al. Hepatocellular carcinoma: consensus recommendations of the National Cancer Institute Clinical Trials Planning Meeting. J Clin Oncol 2010; 28: 3994-4005.
  • 2
    El-Serag HB, Rudolph KL. Hepatocellular carcinoma: epidemiology and molecular carcinogenesis. Gastroenterology 2007; 132: 2557-2576.
  • 3
    Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 2011; 144: 646-674.
  • 4
    Takayama T. Surgical treatment for hepatocellular carcinoma. Jpn J Clin Oncol 2011; 41: 447-454.
  • 5
    Fernandez M, Semela D, Bruix J, Colle I, Pinzani M, Bosch J. Angiogenesis in liver disease. J Hepatol 2009; 50: 604-620.
  • 6
    Mann CD, Neal CP, Garcea G, Manson MM, Dennison AR, Berry DP. Prognostic molecular markers in hepatocellular carcinoma: a systematic review. Eur J Cancer 2007; 43: 979-992.
  • 7
    Wilusz JE, Sunwoo H, Spector DL. Long noncoding RNAs: functional surprises from the RNA world. Genes Dev 2009; 23: 1494-1504.
  • 8
    Nagano T, Fraser P. No-nonsense functions for long noncoding RNAs. Cell 2011; 145: 178-181.
  • 9
    Tian D, Sun S, Lee JT. The long noncoding RNA, Jpx, is a molecular switch for X chromosome inactivation. Cell 2010; 143: 390-403.
  • 10
    Orom UA, Derrien T, Beringer M, Gumireddy K, Gardini A, Bussotti G, et al. Long noncoding RNAs with enhancer-like function in human cells. Cell 2010; 143: 46-58.
  • 11
    Hung T, Wang Y, Lin MF, Koegel AK, Kotake Y, Grant GD, et al. Extensive and coordinated transcription of noncoding RNAs within cell-cycle promoters. Nat Genet 2011; 43: 621-629.
  • 12
    Huarte M, Guttman M, Feldser D, Garber M, Koziol MJ, Kenzelmann-Broz D, et al. A large intergenic noncoding RNA induced by p53 mediates global gene repression in the p53 response. Cell 2010; 142: 409-419.
  • 13
    Guttman M, Amit I, Garber M, French C, Lin MF, Feldser D, et al. Chromatin signature reveals over a thousand highly conserved large non-coding RNAs in mammals. Nature 2009; 458: 223-227.
  • 14
    Yang F, Zhang L, Huo XS, Yuan JH, Xu D, Yuan SX, et al. Long noncoding RNA high expression in hepatocellular carcinoma facilitates tumor growth through enhancer of zeste homolog 2 in humans. HEPATOLOGY 2011; 54: 1679-1689.
  • 15
    Ting DT, Lipson D, Paul S, Brannigan BW, Akhavanfard S, Coffman EJ, et al. Aberrant overexpression of satellite repeats in pancreatic and other epithelial cancers. Science 2011; 331: 593-596.
  • 16
    Panzitt K, Tschernatsch MM, Guelly C, Moustafa T, Stradner M, Strohmaier HM, et al. Characterization of HULC, a novel gene with striking up-regulation in hepatocellular carcinoma, as noncoding RNA. Gastroenterology 2007; 132: 330-342.
  • 17
    Gupta RA, Shah N, Wang KC, Kim J, Horlings HM, Wong DJ, et al. Long non-coding RNA HOTAIR reprograms chromatin state to promote cancer metastasis. Nature 2010; 464: 1071-1076.
  • 18
    Lay AJ, Jiang XM, Kisker O, Flynn E, Underwood A, Condron R, Hogg PJ. Phosphoglycerate kinase acts in tumour angiogenesis as a disulphide reductase. Nature 2000; 408: 869-873.
  • 19
    Bruix J, Llovet JM. Prognostic prediction and treatment strategy in hepatocellular carcinoma. HEPATOLOGY 2002; 35: 519-524.
  • 20
    Kitamura K, Hatano E, Higashi T, Narita M, Seo S, Nakamoto Y, et al. Proliferative activity in hepatocellular carcinoma is closely correlated with glucose metabolism but not angiogenesis. J Hepatol 2011; 55: 846-857.
  • 21
    Moreno-Sanchez R, Rodriguez-Enriquez S, Marin-Hernandez A, Saavedra E. Energy metabolism in tumor cells. FEBS J 2007; 274: 1393-1418.
  • 22
    Jones AM, Marzella R, Rocchi M, Hewitt JE. Mapping of the human ribosomal small subunit protein gene RPS24 to the chromosome 10q22-q23 boundary. Genomics 1997; 39: 121-122.
  • 23
    Neo SY, Leow CK, Vega VB, Long PM, Islam AF, Lai PB, et al. Identification of discriminators of hepatoma by gene expression profiling using a minimal dataset approach. HEPATOLOGY 2004; 39: 944-953.
  • 24
    Iizuka N, Oka M, Yamada-Okabe H, Nishida M, Maeda Y, Mori N, et al. Oligonucleotide microarray for prediction of early intrahepatic recurrence of hepatocellular carcinoma after curative resection. Lancet 2003; 361: 923-929.
  • 25
    van Malenstein H, van Pelt J, Verslype C. Molecular classification of hepatocellular carcinoma anno 2011. Eur J Cancer 2011; 47: 1789-1797.
  • 26
    Zhang J, Yang Y, Yang T, Liu Y, Li A, Fu S, et al. microRNA-22, downregulated in hepatocellular carcinoma and correlated with prognosis, suppresses cell proliferation and tumourigenicity. Br J Cancer 2010; 103: 1215-1220.
  • 27
    Hou J, Lin L, Zhou W, Wang Z, Ding G, Dong Q, et al. Identification of miRNomes in human liver and hepatocellular carcinoma reveals miR-199a/b-3p as therapeutic target for hepatocellular carcinoma. Cancer Cell 2011; 19: 232-243.

Supporting Information

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
HEP_25895_sm_SuppFig1.tif1059KSupporting Information Figure 1. Full-length of human lncRNA MVIH gene cloning. (A) Left: Agarose gel electrophoresis of nested PCR products from the 5′-RACE procedure and the 3′-RACE procedure. Molecular weight markers (base pairs) are indicated on the right. The major PCR product is marked by an arrow on the left. Right: Sequencing of second-round PCR products revealed the boundary between the universal anchor primer and lncRNA MVIH sequences. The guanine marked by a vertical line indicates a putative transcriptional start site or a putative transcriptional end site. The arrows indicate the transcriptional directions. (C) Nucleotide sequence of the full-length human lncRNA MVIH gene. Sequence added based on reference sequence in NCBI database is indicated in red.
HEP_25895_sm_SuppFig2.tif8136KSupporting Information Figure 2. Codon substitution frequency scores (CSFs) of lncRNA-MVIH and RPS24. The whole transcription region is divided to some blocks by chromosome positions of exon, intron of RPS24 and MVIH. The average score of each block was taken as the CSFs of corresponding block. The blue bars represent exons of RPS24 and the green bar represents the transcript of lncRNA MVIH. The black lines represent regions un-transcribed. TSS, transcription start site. TES, transcription end site.
HEP_25895_sm_SuppFig3.tif12783KSupporting Information Figure 3. (A) Expression pattern of lncRNA MVIH in HCC cell lines and normal cell line (LO2). (B) The expression of the lncRNA MVIH in Huh7 and HCCLM3 cells transfected with a lentivirus encoding the lncRNA MVIH (Huh7-lncRNA-MVIH and HCCLM3-lncRNA-MVIH) or a puromycin control (Negative control, Huh7-NC and HCCLM3-NC) was quantified by RT-PCR. The data are shown as the means ± standard deviation of three independent biological replicates. The statistical difference was analyzed by Two-sample t-test. (C) Representative photographs of luciferase signals in 1 week and 5 weeks were taken by an IVIS system after injection with HCCLM3 cells through the splenic hilum.
HEP_25895_sm_SuppFig4.tif2993KSupporting Information Figure 4. (A) RIP experiments were performed using an antibody against EZH2 on extracts from SMMC-7721 cells and one fresh HCC tissue. The purified RNA was used for q RT-PCR using the SYBY-green method, and the enrichment of the lncRNA MVIH was normalized to the input. The lncRNA HEIH, which was previously revealed to associate with EZH2, was applied for a positive control. The data are shown as the means ± S.D. The statistical difference was analyzed by Two-sample t-test. (B) Expression of key genes in glycolysis was detected using q RT-PCR. Huh7 and HCCLM3 transfected with Lv-lncRNA-MVIH showed no significant alteration of glycoltic markers expression compared with control cells. (C) The production of lactate acid in Huh7 and HCCLM3 cells transfecting LV-lncRNA-MVIH or LV-NC was measured by lactate acid detection kit (JianCheng, NanJing, China) according to manufacturer's instructions.The data were showed as mean ± S.D. The statistical difference was analyzed by Two-sample t-test.
HEP_25895_sm_SuppInfo.doc150KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.