|
|
||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online August 7, 2002 as doi:10.1096/fj.02-0074fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||


* Laboratory of Human Carcinogenesis, National Cancer Institute, Bethesda, Maryland, USA;
Department of Radiation Oncology, Medical College of Virginia, Richmond, Virginia, USA; and
Division of Hematology, Center for Biologics Evaluation and Research, Food and Drug Administration, Bethesda, Maryland, USA
3Correspondence: Laboratory of Human Carcinogenesis, National Cancer Institute, National Institutes of Health, 37 Convent Dr., MSC 4255, Building 37, Room 2C25, Bethesda, MD 20892-4255, USA. E-mail: xin_wei_wang{at}nih.gov
SPECIFIC AIMS
Hepatitis B virus (HBV) is a major risk factor for hepatocellular carcinoma (HCC). HBV encodes an oncogenic HBx gene that functions as a transcriptional coactivator of multiple cellular genes. To understand the role(s) of HBx in the early genesis of HCC, we systematically analyzed gene expression profiles by serial analysis of gene expression (SAGE) in freshly isolated human primary hepatocytes infected with a replication-defective adenovirus containing HBx.
PRINCIPAL FINDINGS
1. SAGE libraries in human primary hepatocytes without or with HBx expression
To ensure high efficiency of HBx expression in freshly isolated human hepatocytes, we constructed an adenoviral vector encoding HBx. Ad-HBx infection at the multiplicity of infection (MOI) of 5 resulted in an efficient HBx expression in freshly isolated hepatocytes, as determined by Western blot with anti-HBx monoclonal antibodies and Northern blot analysis. No visible cytotoxic effect was observed in hepatocytes infected with Ad-HBx or Ad-CMV at this MOI (data not shown). SAGE libraries were constructed from Ad-HBx- and Ad-CMV-infected human primary hepatocytes. A total of 19,501 tags were generated, of which 9815 were from Ad-HBx-infected primary hepatocytes and 9686 were from control Ad-CMV. Sequence analysis identified a total of 1184 distinct genes from the Ad-HBx-infected primary hepatocytes and 1426 from the control. In total, 93% were known genes, 2.2% as ESTs, 3.6% as unknown transcripts as listed in the SAGE database, and 1.2% were unmatched in the SAGE database. The introduced HBx gene also contains a SAGE tag (GAGACCACCG). The MOI dose used in this experiment produced a moderate abundance of HBx transcripts, which represented 0.14% of total transcripts analyzed, similar to the expression of genes encoding profilin1, gelsolin, or apolipoproteins, etc. The abundance of HBx transcripts is less than that of many other cellular transcripts, such as ß-actin (0.2%), ferritin (0.25%), GAPDH (0.38%), antitrypsin (0.68%), albumin (0.78%), and M-phase phosphoprotein homologue (2.73%).
2. Gene expression pattern characteristic of normal hepatocytes
From nearly 10,000 SAGE tags analyzed in normal human hepatocytes, we have listed
70 genes expressed in greatest abundance, and these genes represented > 19% of the total transcripts analyzed. Examples of these transcripts include albumin, antitrypsin, serum amyloid A1, orosomucoid 1, and glucose phosphate isomerase, which represent most abundant liver-specific genes, as well as many common housekeeping genes such as actins, myosin, GAPDH, and many ribosomal genes.
3. The profile of the differentially expressed genes deregulated by HBx
In the transcripts analyzed, 81.6% were not altered by HBx expression, whereas 9.8% were induced and 8.6% were repressed > threefold (Fig. 1
B). A total of 57 genes were up-regulated with > 5-fold, which include two genes with a > 10-fold increase in HBx-expressing cells. Most of these known genes were categorized as enzymes, transcription factors, and ribosomal proteins. In contrast, 46 genes were down-regulated by at least 5-fold; many are either unknown genes or ESTs, including four genes with > a 10-fold decrease compared with the control.
|
4. Validation of SAGE data by Megarray
To validate the identity and the quantification of genes identified by SAGE, we selected 39 known genes according to the gene consortium databases (Research Genetics) that matched to our SAGE tags. These clones were chosen for their known identity and their wide ranges of expression profiles. Two unknown clones also were included in this analysis. The cDNA of these clones were quantitatively spotted in an array manner onto a 5 x 6 cm membrane, which we referred to as Megarray. Probes were made from the same RNA samples used for the SAGE library construction and derived from primary hepatocytes infected with either Ad-HBx or Ad-CMV by reverse transcription with 32P labeling. Probes were then hybridized to the Megarray in duplicate. The ratios of the dot intensity between Ad-HBx and Ad-CMV were plotted against the folds of SAGE tags. There is a significant correlation between the two methods (r=0.8, P<0.005), with an exception of only three clones. Of these three clones, the SAGE tag (ACTGGGGAAT), originally identified as Ran-, binding protein 1 was shown to have a wrong identity because of an error in the Unigene database. Thus, the Megarray analysis proved useful in validating our SAGE data and provides confidence for the identity of the HBx-targeted transcripts.
5. Up-regulation of genes involved in protein synthesis, gene transcription, and protein degradation
Close inspection of the liver cell transcripts up-regulated by HBx revealed an interesting gene expression profile. HBx expression in primary human hepatocytes appeared to be associated with an induction of genes belonging to three categories. From a total of 71 ribosomal protein transcripts identified, 35 are increased by twofold or greater in HBx-expressing hepatocytes. The other two categories included nine transcripts encoding transcription factors with a zinc finger motif and nine transcripts encoding proteasome subunits and ubiquitin or ubiquitin-associated proteins, respectively. Some ribosomal genes L4, L26, and L38 showed a significant change in response to HBx expression with an increase of six-, seven-, or eightfold, respectively. Some of the ribosomal transcripts can be detected only in the HBx group. Similarly, all nine transcription factors with zinc finger motifs and seven of nine proteasome-associated factors were undetectable in control hepatocytes. It is possible that the magnitudes of induction in these transcripts by HBx expression are underestimated.
6. Activation of ribosomal genes by HBx is independent of the presence of an NES motif
HBx is a nuclear cytoplasmic shuttling protein with its primary localization in the cytoplasm due to the presence of an NES sequence. We sought to determine whether the multiple gene deregulations by HBx are via a nuclear export pathway in a NES-dependent manner. We used a semiquantitative RT-PCR and analyzed the expression level of three representative ribosomal transcripts (RPL26, RPL14, and RPL4) as standardized by 18S ribosomal RNA in primary hepatocytes infected with an adenovirus containing wild-type HBx or an export-deficient HBx mutant with mutations at its NES motif. We found no significant difference between mutant and wild-type HBx expressions of all three ribosomal proteins. These data indicate that HBx-mediated deregulation of ribosomal genes is associated with mechanisms other than its effect on nuclear export pathway.
CONCLUSIONS AND SIGNIFICANCE
Using a SAGE technique, we have generated global gene expression profiles in normal primary human hepatocytes and HBx-expressing hepatocytes. These profiles may help us to understand potential function of HBx as an oncogene at the molecular level. The transcriptome of isolated primary human hepatocytes may also provide a useful database for future SAGE analysis of liver diseases.
Gene expression profiling is a useful tool to generate hypotheses. Our new findings were that genes of many ribosomal proteins were highly expressed in HBx-expressing hepatocytes. It is plausible that HBx may induce c-myc expression, thereby inducing ribosomal genes that may be necessary to enhance protein synthesis during HBx-stimulated cell proliferation and cell transformation. In addition, we observed that many transcription factors, including the zinc-finger protein family such as c-myc-associated zinc-finger protein and zfp-36, as well as activating transcription factor 4, were induced in HBx-expressing hepatocytes. We found that HBx can induce the expression of genes belonging to the protein degradation pathway such as ubiquitin, ubiquitin-specific protease 25, and proteasome alpha type 2, which is consistent with a recent study that HBx both structurally and functionally interacted with the proteasome complex resulting in the trans-activation of multiple genes. The up-regulation of proteins in proteasome complex could be a feedback response to eliminate overexpressed HBx in hepatocytes after infection. An increase in expression of these genes by HBx may explain the unusual ability of HBx to function as a transcription coactivator of multiple genes. These results indicate that HBx may function as a major regulator in a common cellular pathway, which in turn regulates protein synthesis, gene transcription, and protein degradation (Fig. 2
).
|
The Megarray method developed in this study has validated the SAGE data, which allows us to detect multiple genes simultaneously. This analysis demonstrates that the SAGE technique is feasible and that Megarray is a useful and efficient tool to validate expression profiles.
In summary, SAGE has helped to identify novel transcripts associated with HBx. Further characterization of these genes and the rest of the inventory of genes may shed light on some of the genetic alterations during early liver cancer development.
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.02-0074fje; to cite this article, use FASEB J. (August 7, 2002) 10.1096/fj.02-0074fje ![]()
2 Current address: Division of Hematology, Center for Biologics Evaluation and Research, Food and Drug Administration, Bethesda, Maryland, USA. ![]()
This article has been cited by other articles:
![]() |
A Budhu, Y Chen, J.W. Kim, M Forgues, K Valerie, C.C. Harris, and X.W. Wang Induction of a unique gene expression profile in primary human hepatocytes by hepatitis C virus core, NS3 and NS5A proteins Carcinogenesis, July 1, 2007; 28(7): 1552 - 1560. [Abstract] [Full Text] [PDF] |
||||
![]() |
Z. Hu, Z. Zhang, J. W. Kim, Y. Huang, and T. J. Liang Altered Proteolysis and Global Gene Expression in Hepatitis B Virus X Transgenic Mouse Liver J. Virol., February 1, 2006; 80(3): 1405 - 1413. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Forgues, M. J. Difilippantonio, S. P. Linke, T. Ried, K. Nagashima, J. Feden, K. Valerie, K. Fukasawa, and X. W. Wang Involvement of Crm1 in Hepatitis B Virus X Protein-Induced Aberrant Centriole Replication and Abnormal Mitotic Spindles Mol. Cell. Biol., August 1, 2003; 23(15): 5282 - 5292. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. W. Kim and X. W. Wang Gene expression profiling of preneoplastic liver disease and liver cancer: a new era for improved early detection and treatment of these deadly diseases? Carcinogenesis, March 1, 2003; 24(3): 363 - 369. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |