In 1980, Brechot detected integrated HBV DNA for the first time in human liver cancer tissues and PLC/PRF/5 cell line by Southern Blot (2), which directed a new way to study the mechanism by which HBV induced HCC at the molecular level. In 1982, Duck Hepatitis B Virus (DHBV), a virus similar to HBV, was found to replicate itself through reverse transcription (22). Therefore, researchers speculated that the integration of HBV DNA might be the direct cause for HBV-induced HCC since it had been recognized that the integration of HBV into host genome would lead to canceration.

Southern blot is the most popular method to detect integrated HBV DNA and can be used to determine whether HBV DNA is free or integrated in liver cells. At the beginning, a commonly used method was to construct a genomic library of HBV-induced HCC. Different fragments of HBV DNA were used as probes for hybridization with the genomic library and the positive clones were examined by DNA sequen-cing (11). Tsuei et al constructed a genomic library of the liver tissue of a nine-year-old HCC patient, using phage L47.1 as a vector (5). Then the positive clones were picked out through hybridization and sequencing. Results showed that there were deletion and rearrangement in the integrated HBV DNA. The inte-grated sequences included X gene and S gene, which had been rearranged. The integrated X gene still kept its function of trans-activating heterologous promoter.

A following method was in situ hybridization by which 3H-HBV DNA probe was used for in situ hybridization in paraffin sections of HCC and matched surrounding liver tissue. Results showed that HBV DNA was mainly distributed in cytoplasm with little in nucleus. Most of HBV DNA in matched surrounding tissue was free while that in cancer cells was integrated. Blum et al hold that the HBV DNA copies and replication activity could be estimated according to the quantity of the positive particles of HBV DNA in cells. Therefore, HBV DNA in situ hybridization provided a reliable method to investi-gate the relationship between HBV infection and HCC.

The invention and application of polymerase chain reaction (PCR) is also of significance for the studies on HBV DNA integration. The key issue of PCR is designing primers, which directly affect the efficiency of amplification (16, 29). Minami et al designed primers according to known sequences and human Alu repetitive sequences, and invented a PCR technique to clone integrated HBV DNA and consecutive se-quences (8). In order to avoid amplification of the DNA fragments between Alu repetitive sequences, dUTPs were introduced into the primers. Then the products were treated with uracil DNA glycosylase after ten cycles of amplification. By using this technique, the consecutive sequences to integrated HBV DNA has been successfully cloned in three cases of HCC tissues, which proves that it is a com-paratively ideal method to study the targets for HBV DNA integration.

Another important method is reverse PCR, which is based on PCR amplification (7). Tsuei et al found that HBV DNA could be detected in 80% of the HBV-induced HCC tissues, but the mechanism by which HBV DNA induce HCC was not clear (27). They cloned integrated HBV DNA and consecutive sequences by using reverse PCR, in which primers were designed according to direct repeat1 (DR1) and direct repeat2 (DR2) (26). Results showed that the amplified HBV DNA resided near DR1 and DR2 in nine out of ten cases of HCC, and all of them were Integronâ…. The results of Southern blot showed that in two out of five cases there were sequences specific to men, homologous to human long interspersed DNA element (LINE-1). In one integrant, it was found that HBV DNA was integrated into RNA-binding motif on Y chromosome (RBMY).

In addition, in situ PCR is also widely used in studies on HBV DNA integration. Matsui et al used fluorescence in situ polymerase chain reaction (FISPCR) to detect integrated HBV DNA in HCC tissues (15). Strong fluorescence was found in Alexander cells and could be analyzed quantitatively.

Integration is not necessary for HBV replication, so it is not a routine in the natural history of HBV infection, which is quite different from that of retroviruses (18). Most researchers used to think that HBV DNA integration occurs at the chronic stage of HBV infection. With the advance of research, HBV DNA integration was found at various stages of infection. Murakami et al used Alu-PCR to detect the liver samples of 19 acute and 12 chronic hepatitis B patients. Results showed that HBV DNA integration was found in 3 acute and all the chronic hepatitis patients (17). In the liver sample of one acute patient, HBV DNA was found to be integrated into the intron sequences of TNF inducible protein.

Both the DNA double-strand and DNA fragments of HBV can be integrated into liver cells in the form of monomer or oligomer. This kind of integration was also found in the liver cells of chimpanzees and woodchucks whose were infected with woodchuck hepatitis B virus. This indicates that the integration might be a direct nucleic acid-nucleic acid process. Integration may occur at different positions with more than one position in the same liver cell. It is known that HBV genome is complicated in construction and structural genes are overlapped with regulatory genes or other structural genes. Therefore, the integration of a fragment of DNA might involve several coding genes with different functions.

No complete HBV DNA integration has been found in the samples of spontaneous HBV infection in researches so far, and no integrated HBV DNA sequence has been found to be identical with another. Therefore, most researchers believe that HBV DNA integration is random. However, in 30% of different cancer tissues, the integration was found to occur at the adhesive terminus of HBV DNA, and the region of single chain was also a frequent position.

HBV DNA integration usually occurs at DR1 and DR2, two direct repeats at the end of HBV DNA chain. Since X gene is near direct repeats, it is the most frequent integrated fragments, and the next is S gene, which is upstream X gene. In sum, integrated DNA fragments focus on the coding gene domains of two transactivating proteins, which are HBsAg and X protein (6, 10, 12).

The sites of HBV DNA integration were analyzed to be random on the whole, although some hot spots had been discovered. In order to find preponderant integration targets of HBV DNA, Murakami et al used improved Alu-PCR to detect the targets in 60 samples of HCC (13). Results indicated that integration could occur in all chromosomes except chromosome 13, X and Y. Through this analysis, 68 sequences of HBV DNA integration targets were identified, including 7 repetitive sequences, 42 cell genes and 19 sequences encoding unknown proteins, of which 15 sequences were closely related to the occurrence of cancer. Some preponderant integration targets of HBV DNA were also found in this research, including hTERT (3/60) and some signal transduction factors.

In addition, Zhang et al found that the sites of HBV DNA integration were near oncogene erb B in HCC tissues (31). Analysis of the sequences revealed that the integrated HBV DNA was near epidermal growth factor (EGF) and other sites of tyrosine protein kinase. Berger et al did not find rearrangement of cell DNA in AL-14 and AL-26, two integrated clones of HBV DNA in PLC/PRF/5 cells. They only found deletion of small DNA fragments. A fragment of 2kb was deleted in AL-14 and 17bp was deleted in AL-26. There was a sequence of 182bp on the right lateral of AL-26, which was thought to be a part of the cell gene. But further study proved that it was an unusual fragment of HBV DNA. Surprisingly, this fragment was highly homologous to WHBV DNA. There were minisatel-lite-like repetitive sequences in AL-14 and AL-26, so they were thought to be the possible hot spots of HBV DNA integration (5).

DNA double-strand breaks (DSB) are potentially lethal lesions, which can be caused by ionizing radiation, chemicals, biological factors and so on. DSBs can be repaired by homologous recombination (HR) or non-homologous end jointing (NHEJ). In the past decade, there have been many studies on the relationship between HBV DNA integration and DNA damage and repair. All these studies indicated that DNA damage, especially DSBs could promote HBV DNA integration into host genome.

Petersen found that H₂O₂ could induce DHBV to integrate into LMH-D21-6, and the inhibitor of poly (ADP-ribose) polymerase (PARP) could increase DHBV integration (19). This indicated that the oxidative stress and DNA damage caused by H₂O₂ could promote HBV DNA integration. On the other hand, PARP was a ribozyme in eukaryotes, catalyzing poly ADP to be ribosylated. It played an important role in repairing DNA damage and maintaining stability of genome. In sum, oxidative DNA damage and inhibition of DNA repair could increase HBV DNA integration, which provided new ideas to study its molecular mechanism. After that, Zhang Lunli made further research on the effects of oxygen-derived free radicals on DHBV DNA integration in infected cells and on the possibility of transmission of integrated DHBV DNA from parent cells to offspring ones (30). In this study, he treated hepatoma cell line LMH-D21-6 with 10 mol/L of H₂O₂. Results showed that the occurrence of integration was 50% at this concentration, while that of integration in the cell without treatment with H₂O₂ was only 8.3%. He also found that the integrated DHBV DNA could be passed on to the DNA of daughter cells. In 2002, Dandri proved again that DSB could promote HBV DNA integration in in vitro infectious models (3). Recently, studies by Bill indicated that DNA DSBs might be a potential target for HBV DNA integration and it was possible that this integration can be done through NHEJ (1). Although the effects of DSBs and DNA repair on HBV DNA integration need further verifi-cation, these results shed new light on the mechanism of HBV DNA integration undoubtedly.

HBV DNA integration into cells may damage the genome stability, cause DNA rearrangement and deletion and even induce canceration. The abnormal translocation of chromosomes was firstly observed in studies on these effects. Simon et al found integrated HBV DNA in HCC tissues and Hep40 cells (20). There was an abnormal chromosome 1p36 in HCC tissues, which did not exist in normal liver cells. Livezey et al found that 2.2.15 cells growed more slowly than Hep G2 cells (13). The former could grow into tumor in nude mice gradually while the latter could not. Besides, several chromosome rearran-gements and loss of heterozygosity (LOH) at the site of p53 were detected in 2.2.15 cells. Su et al cloned complete HBV DNA from Hep3B cells (23). Analysis of its sequence indicated that the open reading frame (ORF) of HBsAg was kept intact, but pre-S1 gene and C gene were rearranged. X protein was cut short at the carboxyl terminal, but its transactivation for SV 40 enhancer/promoter was reserved. Analysis with S1 nuclease revealed that there were HBV RNAs of 4.0, 2.9 and 2.2 kb in Hep3B cells, and their transcription was carried out under the guidance of pre-S2/S promoter. Somatic-cell hybrid mapping demonstrated that the flanking sequences belonged to chromosome 13 and chromosome 4, respectively, which indicated that HBV DNA integration caused rearrangement of DNA and the transactivation of X protein also played an important role in the formation of HCC.

Tokino et al found deletion of large fragments of genomic DNA in liver tissue integrated by HBV DNA and the fragments were 25kb, 12kb and 11kb, respectively (25). Every integron contained small fragments of viral DNA. After analyzing sequence of integrated the random clones, Tokino et al found that chromosome 11 and chromosome 17 were the most frequent ones to be integrated, but no specific integration sites were discovered (24). Slagle et al hold that there were possibly many reasons for HCC occurrence, and one of them was that the integrated HBV DNA activated the expression of important genes (21). Henderson et al found that the sequence at chromosome 18q11.1-q11.2 was located on the left of HBV DNA and the sequence at the end of chro-mosome 4 was located on its right, which indicated that HBV-induced HCC could be associated with the translocation of chromosomes (9).

HBV DNA might induce and promote the formation of HCC. Recently, Dejean found that in the cells of HCC, the ORF of chromosomes which contained HBV DNA had encoding constructions homologous to those of V-erb-A and steroid receptor genes (4). Further more, Zhang found fragments similar to oncogene erb-B and rearrangement of base sequence in the cells of liver cirrhosis and HCC (31). In sum, the integration of HBV DNA into liver cell causes chromosome aberration and DNA rearrangement, which may initiate the regulatory sequence of adjacent proto-oncogenes with the help of some carcinogenic agents. If the enhancer of HBV was inserted into a specific site, the transcription of oncogene may be enhanced and canceration induced. As Brechot speculated, the metabolism of liver cells was impaired by HBV infection to lead to the accumulation of carcinogene which induced canceration. Then the cancerous cells reproduced themselves quickly due to HBV DNA integration and gene rearrangement. This process was accelerated by the regeneration of liver cells and abnormality of immunity at the later stage of hepatitis.

In sum, since HBV DNA integration was disco-vered in 1980s, a great deal of research has been done and great progresses have been achieved. Through these studies, we obtained various methods to detect HBV DNA integration, the status of integration at different stages of infection, putative integration modes and the sequence of integrated HBV DNA as well as that of cellular DNA at integrated sites. And the roles of HBV DNA integration in the formation of HCC were also confirmed. However, there still exist many problems. Firstly, the complexity and diversity of HBV DNA integration need to be explained systematically and theoretically. Secondly, the mech-anism by which DSBs and their repair affect HBV DNA integration needs further study. Thirdly, HCC cells and Hep G2 cells are the most frequently used cell models in which HBV DNA has already been integrated into these cells stably, so the natural process of HBV infection, integration and canceration can not be observed. In a nutshell, there is a lot of work to be done in order to make clear the molecular mech-anism of HBV DNA integration.