To the Editor: In their article, Tamori et al. (1) describe a common integration of hepatitis B virus (HBV) DNA into the MLL2 gene in 3 of 15 hepatocellular carcinomas (HCC). In all of the three cases of HBV integration into the MLL2 gene, they observed MLL2 RNA expression by reverse transcription-PCR in the tumor tissues only and not in the surrounding noncancerous tissues. In addition, they found MLL2 gene amplification in one case.

Insertional mutagenesis caused by HBV integration has attracted interest for a long time. In many HCC cases, the studies were not able to identify any preferred site or gene for HBV integration; therefore, HBV integrations seemed to occur randomly.

With recent accumulating information on the human genome and progress in strategies to identify viral-host junctions, however, a common HBV integration into the human telomerase reverse transcriptase gene was reported by several groups independently. Ferber and colleagues (2) reported HBV-human telomerase reverse transcriptase sequences in one of eight HCCs and in one of five hepatoma-derived cell lines by restriction site PCR. Brechot and colleagues reported 2 cases with HBV integration into human telomerase reverse transcriptase among 50 cases of HCCs using HBV-Alu PCR and 1 case of integration among 10 previously published cases (3).

Given an extremely high frequency (3 of 15 HCC) of HBV integration into the MLL gene, we searched for reports of the MLL gene being affected by HBV integration in the literature and Genbank. During this search, we noticed that the sequence referred to as MLL2 by Tamori et al. is called MLL4 at present. MLL2 was originally reported at ch19 but was renamed as MLL4. Currently, in the present database, MLL2 is considered a different gene that is located at ch12.

HBV integration in the MLL4 gene at ch19 was observed in 0 of 50 cases by HBV-Alu PCR (3), 0 of 17 cases by a conventional phage cloning method (4), 1 of 10 cases in the literature (cited in ref. 3), and 0 of 17 cases in the Genbank. Thus, only 1 of 94 cases of HCCs and cell lines has shown MLL4 affected by HBV.

The reason for this discrepancy (P < 0.01 by χ2 test) in the frequency of MLL4 gene involvement between Tamori et al. and previous studies is unclear. However, we would like to point out that Tamori et al. did not clarify a clonal origin of the HBV integration. In Fig. 2, they showed MLL4 gene amplification in cancerous tissues using a genomic Southern blot hybridized with an MLL4 probe. But they did not show radiography of the same membrane hybridized with an HBV probe, which is essential for confirmation of clonal expansion of integrated HBV DNA. It was also curious that we could not see any band in the noncancerous tissues that was detected by the MLL4 probe.

Because they used a highly sensitive cassette ligation PCR, it is possible that they detected a small quantity of HBV integration that occurred after MLL4 genome amplification had been completed. We also noted that the MLL4 gene contains a cluster of PstI site (four sites between nt 17000 and 17600 of Genbank AD000671) that they used for digesting genomic DNA and cassette ligation before PCR. The possibility that genome amplification of MLL4 with a cluster of PstI site might result in artificial PCR products should carefully be eliminated.

Despite this paucity of data, Tamori et al. have revealed genome amplification of MLL4 genes and its exclusive RNA expression in relation to some HCCs through an investigation of HBV integration. The role of HBV integration in MLL4 gene expression and amplification is of great interest and further study should be conducted.

In response:We would like to thank Dr. Minami and colleagues for their interest in our study (1) about HBV DNA integration into the human genome of hepatocellular carcinoma. However, we disagree with several of their comments. First, we did not state in our article that MLL2 was a common target for HBV DNA integration in hepatocellular carcinoma. We suggested that MLL2 gene is one of the targets for HBV DNA integration. Microarray analysis showed that HBV DNA integration altered human gene expression during hepatocarcinogenesis.

We also appreciate the kind comment of Dr. Minami et al. that MLL2 is called MLL4 at present, and we apologize to the readers because use of the older term MLL2 may have caused confusion. MLL2 was named so only because of its close homology to MLL. The older term MLL2 gene was discovered in 1999 (2). Another homologous gene, ALR, was identified in 1997 (3). Currently, we refer to MLL2 as MLL4 or MLL2/4 and ALR as MLL2.

Next, amplification with the cassette ligation method depends on the target-genome sequence. Restriction enzymes most likely do not function without the restriction enzyme–sensitive sequence. Genome methylation is another important factor for the successful operation of this method. Alu-PCR has been shown to be more efficient than ligation-mediated PCR for amplification to an unknown flanking sequence (4). We agree with this. However, there is no evidence that the rate of artificial errors is higher with ligation-mediated PCR. We amplified the MLL2/4-HBV DNA fragments in hepatocellular carcinomas from different patients at different dates. Three different junctions were found between MLL2/4 and HBV DNA. We proposed that HBV DNA randomly integrated into the genomic site of MLL2/4. Up to the present, ∼60 genes have been reported as viral insertion sites (5). In the future, new genes will be found, and previously reported genes will be redetected. It is impossible to rule out the possibility of coincidental results on the basis of previous data. Consensus has been reached that HBV DNA is randomly integrated into human genome. However, recent studies have suggested that HBV DNA integration might accumulate at a specific gene(s) in human cancer, such as human telomerase reverse transcriptase (hTERT). Ferber et al. (6) found two junctions between HBV DNA and hTERT in hepatocellular carcinoma and three junctions between HPV DNA and hTERT in cervical cancer. We detected three junctions between HBV DNA and MLL2/4 in hepatocellular carcinoma. Our detected frequency of integration sites is consistent with previous findings.

Next, Dr. Minami et al. suspected that PstI clusters might induce artificial errors with our procedure. We cannot understand the relation between PstI clusters and amplification errors. Our three fragments of MLL2/4 successive to HBV DNA did not have a restriction site for PstI, excluding ligation sites. We could directly amplify the HBV-MLL2/4 fragment from raw genomic DNA extracted from three hepatocellular carcinomas with the use of two primers across the junction between HBV and MLL2/4. TAKARA Bio, Inc. (Otsu, Japan), the manufacturer of the cassette ligation kit, has commented that basic studies have shown no evidence of technical errors related to clustering of restriction sites, and no customer has filed claims about artificial errors with this kit. We conclude that HBV DNA was clearly integrated into human MLL2/4 genome in three hepatocellular carcinomas.

Finally, we hope for continued cooperation in studies of HBV DNA insertional mutagenesis with researchers interested in this field.

References

1
Tamori A, Yamanishi Y, Kawashima S, et al. Alteration of gene expression in human hepatocellular carcinoma with integrated hepatitis B virus DNA.
Clin Cancer Res
2005
;
11
:
5821
–6.
2
FitzGerald KT, Diaz MO. MLL2: a new mammalian member of the trx/MLL family of genes.
Genomics
1999
;
59
:
187
–92.
3
Prasad R, Zhadanov AB, Sedkov Y, et al. Structure and expression pattern of human ALR, a novel gene with strong homology to ALL-1 involved in acute leukemia and to Drosophila trithorax.
Oncogene
1997
;
15
:
549
–60.
4
Nelson DL, Ledbetter SA, Corbo L, et al. Alu polymerase chain reaction: a method for rapid isolation of human-specific sequences from complex DNA sources.
Proc Natl Acad Sci U S A
1989
;
86
:
6686
–90.
5
Murakami Y, Saigo K, Takashima H, et al. Large scaled analysis of hepatitis B virus (HBV) DNA integration in HBV related hepatocellular carcinomas.
Gut
2005
;
54
:
1162
–8.
6
Ferber MJ, Montoya DP, Yu C, et al. Integrations of the hepatitis B virus (HBV) and human papillomavirus (HPV) into the human telomerase reverse transcriptase (hTERT) gene in liver and cervical cancers.
Oncogene
2003
;
22
:
3813
–20.
1
Tamori A, Yamanishi Y, Kawashima S, et al. Alteration of gene expression in human hepatocellular carcinoma with integrated hepatitis B virus DNA.
Clin Cancer Res
2005
;
11
:
5821
–6.
2
Ferber MJ, Montoya DP, Yu C, et al. Integrations of the hepatitis B virus (HBV) and human papillomavirus (HPV) into the human telomerase reverse transcriptase (hTERT) gene in liver and cervical cancers.
Oncogene
2003
;
22
:
3813
–20.
3
Murakami Y, Saigo K, Takashima H, et al. Large scaled analysis of hepatitis B virus (HBV) DNA integration in HBV related hepatocellular carcinomas.
Gut
2005
;
54
:
1162
–8.
4
Wang Y, Lau SH, Sham JS, Wu MC, Wang T, Guan XY. Characterization of HBV integrants in 14 hepatocellular carcinomas: association of truncated X gene and hepatocellular carcinogenesis.
Oncogene
2004
;
23
:
142
–8.