Important Roles of Cellular MicroRNA miR-155 in Leukemogenesis by Human T-Cell Leukemia Virus Type 1 Infection

Human T-cell leukemia virus type 1 (HTLV-1) is the pathogen that causes the aggressive and lethal malignancy of CD4+ T-lymphocytes called adult T-cell leukemia/lymphoma (ATLL). MicroRNAs (miRNAs), a class of short, noncoding RNAs, regulate gene expression by targeting mRNAs for translational repression or cleavage. miRNAs are involved in many aspects of cell biology linked with formation of several cancer phenotypes. However, the relation between miRNAs and pathologic implication in ATLL is not well elucidated. Here, we evaluated the roles of cellular miRNAs in ATLL caused by HTLV-1. We found that the expression of miR-155 was increased in HTLV-1-positive T-cell lines. miR-155 expression was enhanced by Tax and binding of transcription factors, NF-κB and AP-1, on the transcription binding sites of miR-155 gene promoter region is important to increase the expression of miR-155 by Tax. Transfection of anti-miR-155 inhibitor, which inhibits the function of miR-155, inhibited the growth of HTLV-1-positive T-cell lines. On the other hand, the growth of HTLV-1-negative T-cell lines was not changed by transfection of anti-miR-155. Forced expression of miR-155 enhanced the growth of HTLV-1-positive T-cell lines. These findings indicate that targeting the functions of miRNAs is a novel approach to the prevention or treatment of ATLL.


Introduction
Human T-cell leukemia virus type 1 (HTLV-1) is the pathogen that causes adult T-cell leukemia/lymphoma (ATLL), which is a unique malignancy of CD4 + T cells [1]. HTLV-1 transforms the infected CD4 + T lymphocytes and causes ATLL in 2-4% of infected individuals 50-60 years after infection [1,2]. There are four subtypes of ATLL: acute, lymphoma, chronic, and smoldering types. Although the median survival time is more than 2 years in the relatively indolent chronic and smoldering types, the prognosis of ATLL in aggressive type, acute and lymphoma types, is very poor with a median survival time of less than 1 year [3]. However, at present, no curative therapy for ATLL has been established. Therefore, identification of good therapeutic targets and development of new therapeutic strategies for ATLL are urgent issues for public health in an endemic area.
The molecular mechanism of HTLV-1-mediated transformation and carcinogenesis is still unknown. Tax oncoprotein which is encoded by HTLV-1 is known as the major viral protein that confers prosurvival and proproliferative properties to infected cells at initial transformation [1]. Previous studies have shown that Tax can immortalize primary human T lymphocytes [4] and induce ATLL like phenotype in transgenic mice [5]. Tax is a 40 kDa transactivator protein and stimulates not only expression of viral genes but also those of cellular genes by interacting with numerous signaling pathways including nuclear factor κB (NF-κB), Janus kinase/signal transducer and activator of transcription, activator protein-1 (AP-1), phosphoinositide 3-kinase/Akt, serum-responsive factor, and the cAMP responsive elementbinding protein (CREB) pathways [6,7].
MicroRNAs (miRNAs) contain a broad class of small 20-25 nucleotide long endogenous RNAs. It has been shown that 2 ISRN Microbiology miRNAs play significant regulatory roles in gene expression by targeting mRNAs for translational repression or cleavage [8]. By regulation of gene expression, miRNAs are involved in many cellular phenotypes, including cell proliferation, differentiation, cell-cycle regulation, and immune surveillance. Alteration of miRNAs expression is not rare in human cancers and plays critical roles in carcinogenesis or cancer progression by changing the mRNA translation patterns [9,10]. It has been demonstrated that there are particular miRNAs that associate with the clinical futures of human cancers [11][12][13]. Therefore, these miRNAs might be useful for diagnostic markers and therapeutic targets of human cancers.
Oncogenic viruses such as Kaposi's sarcoma-associated herpes virus (KSHV), Epstein-Barr virus (EBV), human cytomegalovirus, and human immunodeficiency virus generate viral miRNAs [14,15]. Some of the viral miRNAs disturb both viral and cellular target gene expression, resulting in the inhibition of apoptosis and in the stimulation of cell growth [14,15]. However, miRNAs which are encoded by HTLV-1 have not been identified. Previously, it has been argued the relation between HTLV-1 infection and regulation of host miRNA expression, suggesting important roles of cellular miRNAs in survival and growth of HTLV-1-positive T cells [16][17][18]. Hypothetically, dysregulation of cellular miRNAs expression patterns by infection of HTLV-1 may impact the progression of ATLL.
Recently, we have performed miRNA expression profiling between HTLV-1-positive and -negative T-cell lines by microarray analysis to identify specific cellular miRNAs that play biological roles in the pathogenesis of ATLL (unpublished data). Through microarray data analysis, miRNAs whose expression levels are changed between HTLV-1-positive and -negative T-cell lines were selected. Among selected miRNAs, we focused on miR-155 because miR-155 has been implicated in normal hematopoiesis, immune response, and also in carcinogenesis of many human cancers [19,20]. miR-155 is encoded in the BIC gene and BIC RNA is processed into miR-155, that is, the functional miRNA [21]. It has been reported that miR-155 plays important roles in maturation and differentiation of lymphocytes [22]. Indeed, miR-155 expression is induced in activation of lymphocytes [19,23]. miR-155 knockout mice have impaired immune functions [22]. Transgenic overexpression of miR-155 in mouse studies has shown that these transgenic mice result in increased frequency of tumor formation [24]. However, the pathological roles of miR-155 in carcinogenesis by HTLV-1 infection are not well understood. Here, we investigated how miR-155 expression levels are regulated by HTLV-1 and possible roles of miR-155 in HTLV-1-positive T-cell growth.

Real-Time RT-PCR.
Preparation of the samples for realtime RT-PCR was performed as described previously [28]. MirVana miRNA Isolation Kit (Ambion, Austin, TX, USA) was used for total RNA isolation from the cells. First-strand cDNA was synthesized by TaqMan-MicroRNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA). Real-time RT-PCR for miRNAs (TaqMan microRNA Assays, Applied Biosystems) was done on MX3000p realtime PCR system (Stratagene, La Jolla, CA, USA) as described previously [29]. PCR reactions were performed in triplicates, and the values relative to RNU48 were analyzed by the 2-ΔΔCt method [30].

Western
Blotting. Western blotting analysis was performed as described previously [31]. Cells were lysed in a buffer containing 62.5 mM Tris-HCl, pH 6.8, 2% sodium dodecyl sulfate (SDS), 10% glycerol, 6% 2-mercaptoethanol, and 0.01% bromophenol blue. Equal amounts of samples were exposed to electrophoresis on SDS-polyacrylamide gels followed by transfer to a polyvinylidene difluoride membrane (Millipore, Billerica, MA, USA). The protein expression was analyzed by immunoreactivity with the appropriate antibodies. The bands were visualized by enhanced chemiluminescence Plus reagent (GE Healthcare).
2.7. Reporter Assay. As previously described [28], appropriate reporter and effecter plasmids were transfected into the cells by MicroPorator MP-100 (Digital Bio Technology, Seoul, Korea). The reference plasmid phRL-TK (Promega, Madison, WI, USA), which contains the Renilla luciferase gene under the control of the herpes simplex virus thymidine kinase promoter, was cotransfected to correct for transfection efficiency. The cells were harvested by centrifugation ISRN Microbiology 3 48 h after transfection, washed with PBS, and lysed in reporter lysis buffer (Promega). Then luciferase activity was analyzed by using the Dual-Luciferase Reporter Assay System (Promega).

Electrophoretic Mobility Shift Assay (EMSA).
As previously described [36], nuclear proteins (5 μg) were incubated in 20 μL total reaction volume, containing 10 mM Tris-HCl, pH 7.5, 50 mM NaCl, 1 mM EDTA, 1 mM DTT, 5% glycerol, and 1 μg poly-deoxy-inosinic-deoxy-cytidylic acid (GE Healthcare, Waukesha, WI, USA). Then radiolabeled oligonucleotide probes were added and mixtures were incubated for 15 min at room temperature. The probes were prepared by annealing the sense and antisense synthetic oligonucleotides and filling in the overhanging ends in the presence of radiolabeled deoxyadenosine triphosphate and deoxycytidine triphosphate using the Klenow DNA polymerase. The oligonucleotides for the probes are the following: NF-κB binding site of the miR-155 gene promoter 5 -gactTACTATGGGATTTCCAGCTC-3 containing a putative NF-κB binding site (underlined), mutant NF-κB binding site 5 -gatcTACTATGCTCGAGCCAGCTC-3 containing a mutated NF-κB binding site (underlined), AP-1 binding site of the miR-155 gene promoter 5 -gactcggttatgagtcacaagtga-3 containing a putative AP-1 binding site (underlined), and mutant AP-1 binding site 5 -gat-cCGGTTATCTCGAGCAAGTGA-3 containing a mutated AP-1 binding site (underlined). The specificities of the DNA-protein complex were determined by adding antibodies specific for NF-κB or AP-1 family proteins to elicit supershifted bands. Antibodies were preincubated with the nuclear proteins for 45 min at room temperature and then incubated with radiolabeled probes. Samples were subjected to electrophoresis on 4% polyacrylamide gels with 0.25 × TBE buffer (22.3 mM Tris, 22.2 mM boric acid, and 0.5 mM EDTA), and the DNA-protein complexes were visualized by autoradiography.

Anti-miRNA Oligonucleotides and miRNA Precursors.
Anti-miR-155 oligonucleotide (anti-miR-155 inhibitor) used for miRNA knockdown and the miR-155 precursor (pre-miR-155 precursor) were purchased from Ambion. The cells were transfected with anti-miR-155 inhibitor or pre-miR-155 precursor at final concentration of 50 nM by MicroPorator MP-100. Anti-miR miRNA Inhibitors-Negative Control number 1 and Pre-miR miRNA Precursor Molecules-Negative Control number 1 (Ambion) were used for negative controls. As described previously [28], transfected cells were cultured for 12 h and then plated on 24-well culture dishes at 1 × 10 5 viable cells/mL. Then, the cells were cultured for additional 24, 48, and 72 h, and the effects of anti-miR-155 inhibitor or pre-miR-155 precursor on the cell growth were determined by counting the number of viable cells in triplicate. The number of the viable cells was counted by the Trypan Blue dye exclusion assay.
2.11. Statistical Analysis. All values were the mean ± SD from three separate experiments. Differences between samples were analyzed by the Student's t-test. A P value less than 0.05 was considered the presence of a statistically significant difference.

miR-155 Is Highly Expressed in HLTV-1-Positive T-Cell
Lines. MiRNA expression profiling in HTLV-1-positive Tcell line MT-2 and HLTV-1-negative T-cell line Jurkat were analyzed by microarray (unpublished data). Differentially expressed miRNAs between HTLV-1-positive and -negative T-cells exceeding 2-fold were selected. In this study, we focused on miR-155 among chosen miRNAs because miR-155 has been implicated in lymphocyte activation and in carcinogenesis of many human cancers [19,20]. Then, we confirmed the high expression of miR-155 in HTLV-1positive T-cell lines by TaqMan real-time RT-PCR ( Figure 1).

miR-155 Expression Is Enhanced by Tax.
Next, we determined the role of Tax in miR-155 expression by using Taxinducible cell line [25]. As we have shown previously [28], we confirmed that JPX-9 cells expressed Tax around 10 h after incubation with CdCl 2 by western blotting. Tax expression persisted at least until 120 h after induction (Figure 2(a)). Then, we analyzed the expression of miR-155 in the cells by TaqMan real-time RT-PCR. MiR-155 expression in JPX-9 cells treated with CdCl 2 started increasing from 48 h and constantly increased until 120 h after incubation with CdCl 2 (Figure 2(b)). MiR-155 expression was not altered by CdCl 2 in the parental Jurkat cells (data not shown). These findings indicate that the expression of miR-155 is enhanced by Tax in T cells.

miR-155 Promoter Activity Is Stimulated by Tax through
Activation of NF-κB and AP-1 Elements. To understand transcriptional mechanisms for Tax-induced activation of miR-155 expression, we analyzed the effects of Tax on the transactivation of miR-155 promoter by luciferase reporter assay. miR-155 gene promoter region has some high-probability transcription factor binding sites including NF-κB and AP-1 sties [34]. Each of these transcription factor binding sites was mutated in the reporter plasmid ( Figure 3(a)). Jurkat cells were transfected with the wildtype or the mutant luciferase reporter plasmid together with or without Tax expression plasmid. Forced expression of Tax enhanced the wild-type reporter activity. In contrast, mutation of the AP-1 site considerably suppressed the Taxstimulated induction of the miR-155 gene promoter activity, and mutation of NF-κB site also inhibited the effect of Tax on miR-155 gene promoter activity (Figure 3(a)). However, reduction levels of reporter activity by mutation in NF-κB site was modest than that by mutation in AP-1 site. Tax can activate the transcription of viral and cellular genes through two different enhancers: a cAMP responsive element (CRE)-like sequence and a κB element. Two wellknown Tax mutants, Tax M22 and Tax 703, differentially activate these pathways. Tax M22 effectively activated the CRE element but not the NF-κB element. Tax 703 activates the NF-κB element but does not affect CRE [32,33].
To confirm the transactivation-relevant signaling pathways, these Tax mutants were cotransfected along with the miR-155 promoter construct. In the present study, Tax M22 or 703 slightly activated the miR-155 promoter (Figure 3(b)).
On the other hand, transfection with two mutants together could activate miR-155 promoter (Figure 3(b)), indicating that Tax activates the miR-155 promoter in NF-κB-and CRE-dependent manners. However, the miR-155 promoter sequence does not have sequences suggestive of sites for CRE, suggesting that activation of miR-155 promoter by Tax is not directly mediated by the CRE pathways. Previously, it has been shown that Tax mutants M22 and 703 cannot activate AP-1 in a T-cell line and the coexpression of Tax M22 and Tax 703 can activate the AP-1 site [37]. It might be possible that both NF-κB and AP-1 signaling are necessary to achieve full activation of miR-155 promoter by Tax in T-cells.

NF-κB Highly Bound to miR-155 Promoter in HTLV-1-Positive T-Cell Lines.
We next examined NF-κB-miR-155 promoter binding in HTLV-1-positive and -negative T-cell lines by EMSA. The oligonucleotide probe with an NF-κB site of miR-155 gene promoter was used. Protein-DNA-binding activity was highly detected in HTLV-1-positive T-cell lines.
In contrast, those in HTLV-1-negative T-cell lines were weak (Figure 4(a)). Observed protein-DNA complexes in HTLV-1-positive T-cell lines were revealed to be specific for NF-κB by competition assays using cold probes (Figure 4(b)). The complex was supershifted by antibodies specific for NF-κB p50, p65, and c-Rel, indicating that the NF-κBbinding protein complexes contained these NF-κB subunits (Figure 4(b)). These results indicate that NF-κB proteins activate the expression of miR-155 by binding to miR-155 gene promoter in HTLV-1-positive T-cell lines.

AP-1 Highly Bound to miR-155 Gene Promoter in HTLV-1-Positive T-Cell Lines.
Then, we analyzed the effects of AP-1 on the expression of miR-155 gene by EMSA. Like NF-κB, highly AP-1 binding activity was detected in HTLV-1positive T-cell lines. On the other hand, AP-1 binding activity in HTLV-1-negative T-cell lines was weak ( Figure 5(a)). Observed protein-DNA complexes in HTLV-1-positive T-cell lines were revealed to be specific for AP-1 by competition assays using cold probes ( Figure 5(b)). The AP-1-binding protein complexes from HTLV-1-positive T-cells included Fra2, JunB, or JunD ( Figure 5(b)). These results support the idea that AP-1 proteins activate miR-155 by stimulation of transcriptional activity of miR-155 gene in HTLV-1-positive T-cell lines.

Discussion
In this study, we found the stimulatory roles of cellular miR-155 in proliferation of HTLV-1-positive T cells. It has been well known that HTLV-1 Tax activates transcription factors, NF-κB and AP-1 signaling pathways [6,7]. We showed that these transcription factors bind to the miR-155 gene promoter, resulting in stimulating the expression of BIC gene whose transcripts are processed into miR-155. Indeed, HTLV-1-positive T-cell lines expressed higher levels of miR-155 that HTLV-1-negative cell lines (Figure 1). Overexpression of miR-155 has been shown to correlate with many human cancers [19,20]. Oncogenic viruses, such as KSHV and Marek's disease virus-1, a herpes virus that causes a lymphoproliferative disorder in chickens, encode their own miR-155 orthologues [38,39]. For example, miR-K12-11, KSHV viral miRNA, has significant homology with cellular miR-155. Previous studies have shown a correlation between EBV and expression of miR-155 [40]. Although EBV does not encode any miR-155 orthologues, EBV type III latency genes enable high cellular miR-155 expression levels in the EBV-infected cells [34]. Like these oncogenic viruses, our current study linked HTLV-1 infection to high expression of miR-155. In addition to our study, other studies have also demonstrated that cellular miR-155 expression was enhanced in HTLV-1-positive T-cell lines or peripheral blood mononuclear cells from ATLL patients [16][17][18]. These findings implicate that cellular miR-155 plays crucial role in the biology and pathogenesis of HTLV-1. Cellular miR-155 expression seems to be regulated by multiple signaling pathways. It has been shown that miR-155 can be induced by B-cell receptor activation through a transcription factor AP-1 site on its promoter region [34]. On the other hand, two other studies have shown that EBV latent protein, LMP1 upregulates miR-155 expression in B cells through an NF-κB-mediated mechanism [41,42]. In our study, we found that both AP-1 and NF-κB influence expression of miR-155 gene (Figure 3). Although luciferase assay with reporter plasmids containing mutations on binding sites of each transcription factor revealed that both AP-1 and NF-κB sites are necessary for full activation of miR-155 promoter, effect of mutation on AP-1 site more clearly decreased promoter activity than that of mutation on NF-κB site (Figure 3(a)). These results suggested that AP-1 is more important than NF-κB for activation of miR-155 gene promoter. Further investigations are needed to conclusively verify which transcription factor is more important for activation of miR-155 promoter activity.
In this study, we found that overexpressed miR-155 enhances the growth of HTLV-1-positive T cells ( Figure 6). Moreover, inhibition of miR-155 functions by anti-miR-155 reduced the growth of these cells. In contrast, this effect of directly targets is the transcript encoding the transcriptional repressor, BACH1, which binds to consensus AP-1 promoter elements and suppresses AP-1 activity [43]. In this study, we found that AP-1 activates miR-155 expression by binding AP-1 site on the promoter of miR-155 gene (Figures 3 and  5). Considering the known activation of AP-1 by HTLV-1 Tax as well as the constitutive activation of AP-1 in primary ATLL cells [44], it is possible that the suppression of AP-1 inhibitors such as BTB and CNC homology 1 (BACH1) by miR-155 may derepress AP-1 promoter activity [45]. In addition, miR-155 can target other transcriptional repressor gene, human immunodeficiency virus type 1 enhancer binding protein 2 (HIVEP2) [34]. HIVEP2 protein inhibits the expression of oncogene c-myc through an NF-κB binding sequence [46]. It has been shown that HTLV-1 Tax stimulates transactivation of c-myc promoter through NF-κB activation [47]. It Recently, we showed that miR-146a plays a crucial role in the growth of HTLV-1positive T cells [28]. Similar to miR-155, we demonstrated that miR-146a expression was induced by Tax through NF-κB activation and anti-miR-146a inhibited specifically the proliferation of HTLV-1-positive T-cell lines but not that of -negative T-cell lines [28]. Although Tax plays an important role in overexpression of both miR-155 and miR-146a, other viral proteins and RNAs such as HTLV-1 basic leucine zipper factor might have some functions to modulate cellular miRNA expression in T cell. A recent study has shown that loss of miR-31 is responsible for oncogenic NF-κB activity and malignant phenotypes in ATLL [48]. Not only single miRNA but also combination of several cellular miRNAs which are abnormally expressed in HTLV-1-positive T cells may affect carcinogenesis by HTLV-1 infection. Therefore, the contribution of other cellular miRNAs in carcinogenesis related with HTLV-1 infection must be examined. From this point of view, further investigations might detect new functions of these miRNAs in carcinogenesis by HTLV-1 infection.

Conclusion
Our study demonstrated that miR-155 is highly expressed in HTLV-1-positive T-cell lines. Tax induced miR-155 expression by activation of NF-κB and AP-1 signaling pathways.
Inhibition of miR-155 by anti-miR-155 inhibitor suppressed the growth of HTLV-1-positive T-cell lines. On the other hand, the growth of HTLV-1-negative T-cell lines was not changed by transfection of anti-miR-155. Forced expression of miR-155 increased the growth of HTLV-1-positive T-cell line. Our findings suggest that highly expressed miR-155 inhibit translation of its target mRNAs that enhance the growth of HTLV-1-infected T cells, resulting in progression of ATLL. We propose that, miR-155 is one of the key molecules that are involved in the progression of ATLL.