Three Murine Leukemia Virus Integration Regions within 100 Kilobases Upstream of c-myb Are Proximal to the 5' Regulatory Region of the Gene through DNA Looping.
ABSTRACT Retroviruses integrated into genomic DNA participate in long-range gene activation from as far away as several hundred kilobases. Hypotheses have been put forth to account for these phenomena, but data have not been provided to support a physical mechanism that explains long-range activation. In murine leukemia virus-induced myeloid leukemia in mice, integrated proviruses have been found upstream of c-myb in three regions, named Mml1, Mml2, and Mml3 (25, 50, and 70 kb upstream, respectively). The transcription factor c-Myb is an oncogene whose dysregulation and/or mutation can lead to human leukemia. We hypothesized that the murine c-myb upstream region contains regulatory elements accessed by the retrovirus. To identify regulatory sites in the murine c-myb upstream region, we looked by chromatin immunoprecipitation with microarray technology (ChIP-on-chip) for histone modifications implicating gene activation in normal cells. H3K4me3, H3K4me1, and H3K9/14ac were enriched at Mml1 and/or Mml2 in the myeloblastic cell line M1, which expresses c-myb. The enrichment of all of these histone marks decreased with differentiation-induced downregulation of the gene in M1 cells but increased and spread in tumor cells containing integrated provirus. Importantly, using chromosome conformation capture (3C)-quantitative PCR assays, interactions between the 5' region, including the promoter and all Mml sites (Mml1, Mml2, and Mml3), were detected due to DNA looping in M1 cells and tumor cells with provirus in Mml1, Mml2, or Mml3. Therefore, our study provides a new mechanism of retrovirus insertional mutagenesis whereby spatial chromatin organization allows distally located provirus, with its own enhancer elements, to access the 5' regulatory region of the gene.
- SourceAvailable from: PubMed Central[Show abstract] [Hide abstract]
ABSTRACT: Retroviral vectors have been successfully used therapeutically to restore expression of genes in a range of single-gene diseases, including several primary immunodeficiency disorders. Although clinical trials have shown remarkable results, there have also been a number of severe adverse events involving malignant outgrowth of a transformed clonal population. This clonal expansion is influenced by the integration site profile of the viral integrase, the transgene expressed, and the effect of the viral promoters on the neighbouring host genome. Infection with the pathogenic human retrovirus HTLV-1 also causes clonal expansion of cells containing an integrated HTLV-1 provirus. Although the majority of HTLV-1-infected people remain asymptomatic, up to 5% develop an aggressive T cell malignancy. In this review we discuss recent findings on the role of the genomic integration site in determining the clonality and the potential for malignant transformation of cells carrying integrated HTLV-1 or gene therapy vectors, and how these results have contributed to the understanding of HTLV-1 pathogenesis and to improvements in gene therapy vector safety.Viruses. 01/2014; 6(11):4140-4164.
- [Show abstract] [Hide abstract]
ABSTRACT: Cellular commitment to differentiation requires a tightly synchronized, spatial-temporal interaction of regulatory proteins with the basic DNA and chromatin. A complex network of mechanisms, involving induction of lineage instructive transcription factors, installation or removal of histone modifications and changes in the DNA methylation pattern locally orchestrate the three dimensional chromatin structure and determine cell fate. Maturation of myeloid lineages from haematopoietic stem cells has emerged as a powerful model to study those principles of chromatin mechanisms in cellular differentiation and lineage fate selection. This review summarizes recent knowledge and puts forward novel ideas on how dynamics in the epigenetic landscape of myeloid cells shape the development, immune-activation and leukaemic transformation outcome.Journal of molecular biology. 08/2014;
- [Show abstract] [Hide abstract]
ABSTRACT: Insertional mutagenesis screens of retrovirus-induced mouse tumors have proven valuable in human cancer research and for understanding adverse effects of retroviral-based gene therapies. In previous studies, the assignment of mouse genes to individual retroviral integration sites has been based on close proximity and expression patterns of annotated genes at target positions in the genome. We here employed next-generation RNA sequencing to map retroviral-mouse chimeric junctions genome-wide, and to identify local patterns of transcription activation in T-lymphomas induced by the murine leukemia gamma-retrovirus SL3-3. Moreover, to determine epigenetic integration preferences underlying long-range gene activation by retroviruses, the colocalization propensity with common epigenetic enhancer markers (H3K4Me1 and H3K27Ac) of 6,117 integrations derived from end-stage tumors of more than 2,000 mice was examined.Retrovirology. 05/2014; 11(1):36.
Three Murine Leukemia Virus Integration Regions within 100
Kilobases Upstream of c-myb Are Proximal to the 5= Regulatory
Region of the Gene through DNA Looping
Junfang Zhang, Jan Markus,* Juraj Bies, Thomas Paul,* and Linda Wolff
Laboratory of Cellular Oncology, Center for Cancer Research, National Cancer Institute, Bethesda, Maryland, USA
controlling essential cellular processes, such as proliferation,
differentiation, and apoptosis (7, 21). A role for c-MYB in human
T-cell leukemia (T-ALL) has been reported recently, where the
18). Altered c-MYB expression also plays a role in human colon
and breast carcinoma (6, 13, 37). These reports followed years of
studies in avian and murine models which demonstrated that
overexpression or mutations in c-myb can release its oncogenic
potential, especially in myeloid cells (30, 40).
c-myb was a primary target of insertional mutagenesis when
venously into adult BALB/c mice following intraperitoneal injec-
tion of pristane to induce an inflammatory response (26, 33, 42).
In this animal model, 100% of the tumors were shown to have
undergone c-myb DNA rearrangements due to virus integration.
Promoter insertion combined with the formation of gag-myb
RNA fusions was the most common mechanism of activation.
Therefore, an important feature was the ability of the enhancer/
to activate transcription from the c-myb locus, bypassing the nor-
In a similar model where pristane-treated DBA/2 mice were
injected with amphotropic 4070 virus (41), two-thirds had inte-
grations directly into c-myb, and additional proviral integration
sites were found far upstream of c-myb. These upstream integra-
Mml3, located approximately 25, 50, and 70 kb upstream of the
c-myb promoter, respectively (9, 16). Interestingly, many of these
one of these regions.
The mechanism by which these proviral insertions in Mml1,
Mml2, or Mml3 contributes to leukemia development has been
within this 100-kb region, we hypothesized that the upstream re-
gion of c-myb contains regulatory elements that control expres-
by the provirus to somehow activate gene expression. We have
addressed this model by analyzing histone modifications within
gene regulation. Indeed, enrichments of histone methylation and
acetylation marks, which identify enhancers, were found near
proviruses and were associated with c-myb expression. Further
analysis of the spatial organization of the same 100-kb region,
using a quantitative chromosome conformation capture PCR
(3C-qPCR) assay, revealed looping structures that, in tumors, al-
low integrated proviral LTRs access to the 5= control region of
c-myb. This provides the first evidence for a long-range mecha-
nism of retrovirus gene activation through a 3-dimensional chro-
Received 1 May 2012 Accepted 9 July 2012
Published ahead of print 18 July 2012
Address correspondence to Linda Wolff, firstname.lastname@example.org.
*Present address: Jan Markus, Cancer Research Institute, Slovak Academy of
Sciences, Bratislava, Slovakia; Thomas Paul, Celgene, San Diego, California, USA.
Copyright © 2012, American Society for Microbiology. All Rights Reserved.
jvi.asm.org Journal of Virology p. 10524–10532 October 2012 Volume 86 Number 19
MATERIALS AND METHODS
1640 medium with 10% (vol/vol) heat-inactivated horse serum (Invitro-
gen). All tumor cell lines (16) established in vitro from granuloma and/or
ascites were cultured in Dulbecco’s modified Eagle medium with 10%
(vol/vol) fetal bovine serum. For interleukin-6 (IL-6) treatment, M1 and
tumor cells were seeded at a density of 1 ? 105cells/ml in medium con-
taining IL-6. IL-6 stocks were prepared as described previously (32).
Quantitative real-time PCR analysis. Total RNA was isolated using
using a cDNA reverse transcription kit (Applied Biosystems). Quantita-
tive real-time PCR was performed in triplicate with predesigned c-myb
gene expression assays (Mm 00501741-m1; Applied Biosystems). Data
were normalized to a mouse glyceraldehyde-3-phosphate dehydrogenase
(GAPDH) control (Applied Biosystems). Relative quantitation was car-
ysis was performed using GraphPad Prism 5 software. The Student t test
was used on measurements of c-myb expression from M1 and tumor
samples from 3 experimental replicates.
ChIP-on-chip analysis. Chromatin immunoprecipitation with mi-
croarray technology (ChIP-on-chip) was conducted as previously de-
scribed (28). Cells were fixed in 0.8% formaldehyde for 6 min at room
temperature. After lysis, samples were sonicated to a size range of 200 to
1,000 bp. Chromatin (150 to 200 ?g) was immunoprecipitated with an-
tibodies for H3K4me3 (ab8580; Abcam), H3K4me1 (ab8895; Abcam),
H3K9/14Ac (06-599; Upstate), CTCF (ab70303; Abcam), H3K9me3
(ab8898; Abcam), H3K27me3 (17-622; Upstate), or rabbit IgG (15006;
Sigma-Aldrich). A 10% aliquot was removed as an input fraction. ChIP
DNA and input DNA were amplified using a WGA2 kit (Sigma-Aldrich).
A total of 2.5 ?g of amplified DNA was labeled with Cy3 (input) or Cy5
(IP) dUTP (PerkinElmer Life and Analytical Sciences) using the CGH
labeling kit (Invitrogen).
Custom 8-by-15,000 tiling arrays (Agilent) contained probes span-
ning mouse chromosome 10 (chr.10) from bp 020600015 to 021199991
(genome browser-mm8; University of California–Santa Cruz). Probes
were designed using eArray (Agilent) and a covered 600-kb region sur-
rounding c-myb (?40-bp spacing). A total of 3 ?g of labeled ChIP and
input DNA was cohybridized to the chip for 40 h at 65°C, washed, and
Data were extracted with Feature Extraction 9.1 software and analyzed
can be accessed at the GEO database with accession number GSE34770.
3C assay. The 3C-qPCR protocol was performed as described previ-
ously (8), with minor modifications. Cross-linking was performed by in-
cubating 1 ? 107cells in 10 ml of fresh medium supplemented with 1%
formaldehyde for 10 min at room temperature. The reaction was
quenched by addition of glycine to a final concentration of 0.125 M.
Nuclei were harvested by lysis of the cells in ice-cold lysis buffer (10 mM
Tris-HCl, pH 7.5; 10 mM NaCl; 0.2% NP-40; 1? complete protease in-
buffer 2 (NEB) containing 0.3% SDS and incubated at 37°C for 1 h while
being shaken. Triton X-100 was added to 2%, followed by incubation for
1 h at 37°C to sequester the SDS. The cross-linked DNA was digested
overnight with 400 U HindIII. Digested DNA was diluted with ligation
buffer (50 mM Tris-HCl, 10 mM MgCl2, 10 mM dithiothreitol [DTT], 1
X-100 to 1%, 1 h of incubation was performed at 37°C. DNA was ligated
by using 100 U T4 DNA ligase in 7 ml 1? ligation buffer for 4 h at 16°C,
?g final) was added and DNA was incubated overnight at 65°C to de-
37°C with RNase (300 ?g final) and purified by phenol-chloroform ex-
traction and ethanol precipitation. The purity assessment and loading
adjustment were based on qPCR.
To prepare a qPCR control template in which all possible ligation
chromosome (BAC) clone (9) was completely digested with HindIII and
ligated at a high DNA concentration to reach a random ligation. The
the standard curve for each test primer set by qPCR.
Quantitative PCR analysis of 3C DNA. The TaqMan probe and bait
primers were designed close to the HindIII restriction site of the c-myb
promoter bait fragment. Test primers were designed close to restriction
sites of each candidate interacting fragment. (Sequences of the test and
3C DNA and Universal PCR Master Mix (Applied Biosystems) were used
for the TaqMan real-time PCR. Standard curves were performed for each
run using serial dilutions of the control template prepared from the K9
BAC clone. Relative interactions were determined by the values corre-
lated using the parameters of the standard curve (b, intercept; a, slope) as
10(CT ? b)/a. For normalization, values in different 3C samples were di-
vided by the value of the ERCC3 internal cross-linking control (35).
Microarray data accession number. Normalized and raw data files
have been submitted to the GEO database under accession number
TABLE 1 Sequence of qPCR primers used for 3C-qPCR analysis of the
mouse c-myb locusa
fragment no. Test primer sequence
sequencesarethefollowing(5=to3=):B1,ATTATGGAGGCGAGAGAGGTGT; B2, TCAT
TATGGAGGCGAGAGAGGTGT; B3, ATTATGGAGGCGAGAGAGGTGTCA; and B4,
TCATTCATTCATTATGGAGGCGAGAGAGG. For each ligation product, the bait
primer giving the best amplification efficiency was used as indicated in the table. The
ligation product-specific primers (so-called test primers) were designed downstream of
the 5= HindIII site of each restriction fragment (fragments 1 to 31). The sequence of the
TaqMan probe used is 5=- 6-carboxyfluorescein [FAM]- AATCTTTGCAGCTGCCTGC
CTGTCAGC-3=BGH. Internal interaction controls were performed using the following
ERCC-3 primers as described before (35): forward primer (5= to 3=), GCCCTCCCTGA
AAATAAGGA; reverse primer (5= to 3=), GACTTCTCACCTGGGCCTACA; ERCC-3
TaqMan probe, 5=- FAM-AAAGCTTGCACCCTGCTTTAGTGGCC-3=BGH.
DNA Looping Aids Retrovirus Insertional Mutagenesis
October 2012 Volume 86 Number 19 jvi.asm.org 10525
c-myb is expressed in tumor cell lines with provirus integrated
in c-myb upstream regions. c-Myb is an essential regulator of
hematopoiesis, and its expression is largely restricted to progeni-
tor cells and is downregulated as cells differentiate (7). As shown
in Fig. 1A, expression of c-myb RNA is significantly decreased in
is expressed in tumor cell lines with integrated provirus in the
Mml1, Mml2, or Mml3 region at levels similar to or higher than
those of undifferentiated M1 cells (Fig. 1A). These tumor cells
no response (Fig. 1B). This indicates that there is a positive corre-
lation between the presence of provirus upstream of c-myb and
Histone H3K4 trimethylation and histone acetylation at
both the c-myb gene promoter and Mml1 are associated with
in influencing gene expression and genome function. To provide
evidence for upstream transcriptional regulatory regions that
might be involved in positively influencing c-myb expression in
ments of H3K4 trimethylation (H3K4me3) and acetylation of
H3K9 and H3K14 (H3K9/14ac) in M1 cells. These histone modi-
fications have been found by others to be frequently present at
ChIP-on-chip analysis using a tiling microarray representing a
600-kb region surrounding the c-myb gene on mouse chr.10
cells, both H3K4me3 and H3K9/14ac were found at its transcrip-
tion start site. Interestingly, there was also strong enrichment of
these marks in the Mml1 region, with 3 peaks between ?25 and
?40 kb that indicate the presence of regulatory elements (Fig. 2A
and B). It should be noted that in previous studies, no transcripts
of sequences representing the three peaks, E1 (3.8k), E2 (1.8k),
and E3 (1.9k) fragments were cloned upstream of the c-myb pro-
Luciferase assays show that sequences within one of the regions
increased luciferase activity (Fig. 3C), indicating the presence of
at both c-myb and Mml1 sites in M1 cells significantly decreased
when c-myb was downregulated in conjunction with IL-6 treat-
ment (Fig. 2B), supporting their role in c-myb transcription.
Further analysis showed that, in tumor cells with integrations
increase in H3K4me3 at the c-myb promoter and at Mml1 com-
tumor cell line (30-2-7) with a provirus at Mml3, we observed a
similar broad distribution of the H3K4me3 mark at c-myb and
Mml1. We hypothesize that the presence of viral enhancers in
integrated proviruses is responsible for the increase in histone
H3K4me1 in the c-myb upstream region is increased and al-
the distribution of the H3K4me1 mark can predict regulatory el-
ements, such as enhancers (11). ChIP-on-chip data showed that
this modification was strikingly abundant at both Mml1 and
were differentiated with IL-6 treatment for 5 days (Fig. 2C). We
also observed that, in tumors with insertions in Mml1 and Mml3,
enrichment of H3K4me1 was present at all Mml regions. Com-
pared to M1 cells, the presence of provirus at either Mml1 or
Mml3 caused an expansion of this histone modification through-
out the upstream region of c-myb.
Changes in H3K27me3 and H3K9me3, which are reported to
be associated with the transcriptional repression states, were not
ing cells (data not shown).
All three retrovirus integration regions interact with the 5=
tone modification data above suggest the existence of regulatory
c-myb expression levels. Interestingly, the presence of provirus in
one region causes increases of the enhancer-associated histone
distal loci and the c-myb promoter. Therefore, we decided to look
FIG 1 Expression of c-myb RNA in tumor cell lines with integrated provirus. Expression levels were determined by quantitative reverse transcription PCR. (A)
cell lines and M1 cells after treatment with IL-6 for 0, 3, 6, and 24 h. Data are normalized to initial c-myb expression in individual cell lines. Error bars represent
SD (n ? 3).
Zhang et al.
jvi.asm.orgJournal of Virology
titative chromosome conformation capture assay (3C-qPCR) (5,
detect long-range chromatin interactions in vivo (5). A region of
Mml2, and Mml3 regions, was examined. Interaction frequency
was detected by quantitative real-time PCR with a TaqMan probe
designed in the bait. To relate the spatial conformation of the
c-myb locus to its transcriptional status, the 3C-qPCR assay was
FIG 2 Histonemodificationspresentatthec-mybgeneandMmlregionscorrelatewithc-mybexpression.(A)ChIP-on-chipdataofhistonemodificationsknownto
mouse chr.10. Four known genes on the locus are shown on top by solid rectangles with arrows that indicate their transcriptional orientation. Locations of viral
integrations sites Mml1, Mml2, and Mml3 are shown. The area within the box, the c-myb gene through Mml1, is expanded in panel B. (B) Detailed analysis of the
separately upstream of the c-myb promoter controlling a luciferase reporter gene (B) and transfected into NIH 3T3 and 293T cell lines. The green horizontal
arrow shows the transcription orientation of the gene. One of the regions (E1) increases the luciferase transcription by 2- to 4-fold compared to the promoter-
only control (C). Data are normalized to cotransfected Renilla gene expression. Error bars represent SD (n ? 3).
DNA Looping Aids Retrovirus Insertional Mutagenesis
October 2012 Volume 86 Number 19jvi.asm.org 10527
the bait fragment. The locations of HindIII fragments are indicated below the graph. Mml1, Mml2, and Mml3 regions are marked by red arrows. Data are
tumor cell lines. Virus-chromosomal DNA junction sequences were determined by a shotgun cloning method as described previously (34). Genomic DNA was
Zhang et al.
jvi.asm.orgJournal of Virology
lated c-myb (Fig. 4A). Prominent peaks of interactions were de-
tected around ?20, ?25, ?50, and ?73 kb. Surprisingly, all 4
peaks were stable, in that they did not disappear with downregu-
lation of c-myb in IL-6-treated M1 cells.
Remarkably, each of the interaction peaks was located proxi-
mal to a virus integration region, either Mml1, Mml2, or Mml3
(Fig. 4A), suggesting that all of the proviruses upstream of c-myb
come in close proximity to the c-myb promoter and affect c-myb
transcription. In the Mml1 region, two interaction peaks were
fragment 9 (?26 kb). We have determined the proviral/chromo-
somal DNA junction sequences for proviruses within these two
Mml1 peaks. These sequences, from tumors (30A2-2-6 and 30-3-
by the three arrows in the Mml1 region in Fig. 4A. Many other
ment 18) and ?73 kb (fragment 26). The interaction frequencies
of these sites are more than 4-fold higher than those of the Mml1
cells did not change upon a differentiation-induced decrease in
The 3C data in M1 cells suggest that provirus in the upstream
region comes close to the c-myb promoter through DNA loops
and thereby affects the oncogene transcription in retrovirus-in-
duced leukemia. To determine whether the provirus indeed
comes in close proximity to the c-myb promoter in tumors with
c-myb locus in tumor cells with proviruses in Mml1 (30C-18),
Mml2 (30-2-9), and Mml3 (30-2-7). The data show that the pro-
moter interaction regions overlap all three virus integration re-
the tumor cells, virus comes in close vicinity to the c-myb pro-
moter through integration into preformed, stable loops.
The spatial organization of the c-myb locus was also examined
in NIH 3T3 cells, because this represents another tissue type, one
which does not express c-myb (Fig. 4D). Interestingly, cross-link-
ing frequency data for the Mml1 region indicate the absence of
interaction between the Mml1 region and the promoter in NIH
3T3 cells. However, cross-linking frequencies in the vicinities of
Mml2 and Mml3 are similar to that in M1 cells. Interestingly, the
tein is a multivalent factor with widespread regulatory functions,
and it has been implicated in mediating intrachromosomal con-
binding in the mouse genomic c-myb locus to see if its binding fit
our model of looping at this locus as determined by 3C-qPCR.
ChIP-on-chip was carried out using antibody to CTCF and the
microarray described above (Fig. 5A). Three CTCF enrichment
?70 kb from the promoter (Fig. 5B). These three CTCF binding
sites overlap the provirus integration regions (indicated by red
vertical arrows). The ChIP data showed that all CTCF binding
sites overlap long-range promoter interaction regions (Fig. 5B),
indicating that CTCF plays a role in loop formation.
This study describes a new mechanism for gene activation by ret-
viruses within 100 kb upstream of the c-myb gene in murine my-
eloid leukemias are inserted specifically at DNA sites that interact
with the 5= region, including the promoter of the gene, through
looping (Fig. 6). Since a well-accepted mechanism of insertional
mutagenesis is enhancer insertion, the mechanism by which the
far-upstream proviruses activate c-myb may be a modification of
this mechanism (22, 27). Since it was reported that differential
expression of c-myb is regulated by a transcriptional arrest mech-
anism in the first intron (2, 31) and the bait used in the confor-
of intron 1, we cannot rule out the possibility that the integrated
proviruses in the Mml regions act through prevention of attenu-
ation at the elongation block.
Sequences in the LTR enhancer elements presumably are re-
previously found that proviruses integrated in the Mml1 region
between 0.4 and 0.9 kb, and we know that the U3 LTR sequences
of structural genes, only an LTR remained in these proviruses.
Although such a model could be predicted based upon recent
studies that show that transcriptional activation in higher eu-
karyotes frequently involves the long-range action of regulatory
elements (10, 24), this is the first time it has actually been shown
that the provirus is in a position to interact with the promoter
and/or control region in intron 1 through a 3-dimensional chro-
In this study, we applied a quantitative 3C assay that was re-
tion regions are close to the promoter by loop formation in M1
(Fig. 4C). The Mml1 region contains 2 loops, giving a total of 4
Interestingly, M1 cells that are induced to differentiate with IL-6
and do not express appreciable levels of c-myb still maintain the
same chromatin looping structure observed in cells with active
transcription. Therefore, this chromatin looping structure is in-
with an arrow depicts the c-myb gene and its transcriptional orientation. (C) Long-range interactions detected between the 5= c-myb region and Mml regions in
tumor cells. 3C-qPCR assays were performed at the same time in M1 cells and tumor cell lines containing a provirus in one of the Mml regions. The upstream
HindIII fragments examined in these experiments are indicated. Integration sites in tumor cell lines are marked by red arrows. The locations of integrated
panel B. The Mml3 location was previously determined by Southern blot analysis. Data are normalized to the ERCC3 internal cross-linking frequency control
(means and SEM; n ? 3). (D) 3C-qPCR assay of NIH 3T3 cells. Data are normalized to the ERCC3 control (means and SEM; n ? 3).
DNA Looping Aids Retrovirus Insertional Mutagenesis
October 2012 Volume 86 Number 19 jvi.asm.org 10529
sufficient by itself for c-myb expression and may require addition
chromatin conformation in a nonhematopoietic cell line, NIH
looping structure at the c-myb locus exist in different cell types.
Apparently, a novel interaction peak that is not found in the he-
matopoietic cells was present very close to the promoter at ap-
proximately ?11 kb. The fact that looping structures vary be-
erythroid cells have a different pattern of looping than we found
here in myeloid cells (36).
The finding that CTCF binds in the vicinity of the looping
structure is not surprising, in that it is known that CTCF contrib-
utes to looping associated with gene activation (12, 29). Although
we discovered 4 dominant promoter-interacting loops, only 3
FIG 5 CTCF is recruited to the promoter interaction regions. (A) ChIP-on-chip experiment using antibody specific for CTCF and the microarray described in
and the chromosome structure of the c-myb locus in M1 cells. 3C-qPCR data of the c-myb locus in M1 cells (details are given in the legend to Fig. 4). Potential
matrix attachment regions (MARs) were predicted within a 100-kb region upstream of c-myb by the online program MARWIZ (http://genomecluster.secs
.oakland.edu/marwiz). In silico data show that potential MARs are located at the boundaries of the loops identified in the 3C assay (alignments are shown by
Zhang et al.
jvi.asm.orgJournal of Virology
not associated with this DNA binding protein. Interestingly, oth-
by cohesin and mediator complexes in the absence of CTCF (14).
Matrix attachment regions (MARs) are evolutionarily conserved
reported to act as boundary elements for chromatin functional
domains (23). The MARWIZ online program was used to predict
cells. In silico data show that potential MARs are located at the
boundaries of the loops identified in the 3C assay (Fig. 5B).
Integration at a site in the genome can be a consequence of both
imagine that both play a role in the far-upstream region of c-myb.
have preferences not only for the vicinity of the 5= ends but also for
insertion within a kilobase of DNase hypersensitivity sites (4, 19).
Here, we mapped histone H3 modification sites in the c-myb
regions in myeloid cells contain active histone modifications, es-
suggested by others who reported that a transgene integrated ap-
proximately 77 kb upstream of the c-myb disrupts sequences that
eage-restricted progenitor cells (25). In addition, studies on the
human HBS1L-MYB intergenic interval associated with elevated
genic region contains regulatory sequences that could be impor-
tant in hematopoiesis by controlling MYB expression (38).
We did not observe a significant change in repression-associ-
ated histone modifications (H3K27me3 and H3K9me3) with
downregulation of c-myb (data not shown). Perhaps these modi-
fications are more generally involved in bivalent stem cell states
cells differentiate and begin to form specialized cells (39).
This study provides a new mechanism of retroviral insertional
distal to a gene access to the gene’s promoter and 5= control re-
gion. It is hypothesized that the enhancers of the provirus can act
at the promoter in a manner similar to that when the virus is
integrated directly in the 5= end of the gene and provide enhancer
We thank Richard Koller in our laboratory for assistance in tumor cell
and Gene Expression, for their much valued and excellent advice.
The work was supported by the Intramural Program at the National
Cancer Institute, Center for Cancer Research. J.M. was partially sup-
ported by grant 2/0135/09 from the Slovak Grant Agency VEGA.
the human genome. Cell 129:823–837.
2. Bender TP, Thompson CB, Kuehl WM. 1987. Differential expression of
c-myb mRNA in murine B lymphomas by a block to transcription elon-
gation. Science 237:1473–1476.
3. Clappier E, et al. 2007. The C-MYB locus is involved in chromosomal
translocation and genomic duplications in human T-cell acute leukemia
(T-ALL), the translocation defining a new T-ALL subtype in very young
children. Blood 110:1251–1261.
4. Daniel R, Smith JA. 2008. Integration site selection by retroviral vectors:
molecular mechanism and clinical consequences. Hum. Gene Ther. 19:
5. Dekker J, Rippe K, Dekker M, Kleckner N. 2002. Capturing chromo-
some conformation. Science 295:1306–1311.
6. Drabsch Y, et al. 2007. Mechanism of and requirement for estrogen-
regulated MYB expression in estrogen-receptor-positive breast cancer
cells. Proc. Natl. Acad. Sci. U. S. A. 104:13762–13767.
7. Greig KT, Carotta S, Nutt SL. 2008. Critical roles for c-Myb in hemato-
poietic progenitor cells. Semin. Immunol. 20:247–256.
8. Hagege H, et al. 2007. Quantitative analysis of chromosome conforma-
tion capture assays (3C-qPCR). Nat. Protoc. 2:1722–1733.
9. Haviernik P, et al. 2002. Linkage on chromosome 10 of several murine
retroviral integration loci associated with leukaemia. J. Gen. Virol. 83:
10. Heintzman ND, et al. 2009. Histone modifications at human enhancers
reflect global cell-type-specific gene expression. Nature 459:108–112.
11. Heintzman ND, et al. 2007. Distinct and predictive chromatin signatures
of transcriptional promoters and enhancers in the human genome. Nat.
12. Hou CH, Zhao H, Tanimoto K, Dean A. 2008. CTCF-dependent en-
hancer-blocking by alternative chromatin loop formation. Proc. Natl.
Acad. Sci. U. S. A. 105:20398–20403.
13. Hugo H, et al. 2006. Mutations in the MYB intron I regulatory sequence
increase transcription in colon cancers. Genes Chromosomes Cancer 45:
14. Kagey MH, et al. 2010. Mediator and cohesin connect gene expression
and chromatin architecture. Nature 467:430–435.
15. Koch CM, et al. 2007. The landscape of histone modifications across 1%
of the human genome in five human cell lines. Genome Res. 17:691–707.
16. Koller R, et al. 1996. Mml1, a new common integration site in murine
leukemia virus-induced promonocytic leukemias maps to mouse chro-
mosome 10. Virology 224:224–234.
17. Kouzarides T. 2007. Chromatin modifications and their function. Cell
18. Lahortiga I, et al. 2007. Duplication of the MYB oncogene in T cell acute
lymphoblastic leukemia. Nat. Genet. 39:593–595.
19. Lewinski MK, et al. 2006. Retroviral DNA integration: viral and cellular
determinants of target-site selection. PLoS Pathog. 2:611–622. doi:
20. Liebermann DA, Hoffman-Liebermann B. 1989. Proto-oncogene ex-
FIG 6 Model of the long-range interactions between retrovirus integration
regions and the 5= end of c-myb in tumor cells. Proviruses in Mml1 (red lines)
or Mml2 and Mml3 (blue lines) come in close proximity to the 5=-end regu-
latory region of the gene by DNA looping. CTCF binding near the interaction
regions is suggested to be involved in the looping formation. Potential MARs
were found at the boundaries of the looping structure.
DNA Looping Aids Retrovirus Insertional Mutagenesis
October 2012 Volume 86 Number 19 jvi.asm.org 10531
pression and dissection of the myeloid growth to differentiation develop-
mental cascade. Oncogene 4:583–592.
21. Lieu YK, Reddy EP. 2009. Conditional c-myb knockout in adult hema-
topoietic stem cells leads to loss of self-renewal due to impaired prolifer-
ation and accelerated differentiation. Proc. Natl. Acad. Sci. U. S. A. 106:
22. Maeda N, Fan H, Yoshikai Y. 2008. Oncogenesis by retroviruses: old and
new paradigms. Rev. Med. Virol. 18:387–405.
23. Mirkovitch J, Mirault ME, Laemmli UK. 1984. Organization of the
higher-order chromatin loop: specific DNA attachment sites on nuclear
scaffold. Cell 39:223–232.
24. Morse RH. 2010. Epigenetic marks identify functional elements. Nat.
25. Mukai HY, et al. 2006. Transgene insertion in proximity to the c-myb
gene disrupts erythroid-megakaryocytic lineage bifurcation. Mol. Cell.
26. Nason-Burchenal K, Wolff L. 1993. Activation of c-myb is an early bone-
marrow event in a murine model for acute promonocytic leukemia. Proc.
Natl. Acad. Sci. U. S. A. 90:1619–1623.
27. Neel BG, Hayward WS, Robinson HL, Fang J, Astrin SM. 1981. Avian
leukosis virus-induced tumors have common proviral integration sites
and synthesize discrete new RNAs: oncogenesis by promoter insertion.
28. Paul TA, Bies J, Small D, Wolff L. 2010. Signatures of polycomb repres-
sion and reduced H3K4 trimethylation are associated with p15INK4b
DNA methylation in AML. Blood 115:3098–3108.
29. Phillips JE, Corces VG. 2009. CTCF: master weaver of the genome. Cell
30. Ramsay RG, Gonda TJ. 2008. MYB function in normal and cancer cells.
Nat. Rev. Cancer 8:523–534.
31. Reddy CD, Reddy EP. 1989. Differential binding of nuclear factors to the
intron 1 sequences containing the transcriptional pause site correlates
with c-myb expression. Proc. Natl. Acad. Sci. U. S. A. 86:7326–7330.
32. Schmidt M, Nazarov V, Stevens L, Watson R, Wolff L. 2000. Regulation
of the resident chromosomal copy of c-myc by c-Myb is involved in my-
eloid leukemogenesis. Mol. Cell. Biol. 20:1970–1981.
33. Shen-Ong GL, Wolff L. 1987. Moloney murine leukemia virus-induced
myeloid tumors in adult BALB/c mice: requirement of c-myb activation
but lack of v-abl involvement. J. Virol. 61:3721–3725.
34. Slape C, et al. 2007. Retroviral insertional mutagenesis identifies genes
that collaborate with NUP98-HOXD13 during leukemic transformation.
Cancer Res. 67:5148–5155.
local histone modification in the beta-globin locus. Genes Dev. 20:2349–
36. Stadhouders R, et al. 2011. Dynamic long-range chromatin interactions
control Myb proto-oncogene transcription during erythroid develop-
ment. EMBO J. 31:986–999.
37. Thompson MA, Flegg R, Westin EH, Ramsay RG. 1997. Microsatellite
deletions in the c-myb transcriptional attenuator region associated with
over-expression in colon tumour cell lines. Oncogene 14:1715–1723.
38. Wahlberg K, et al. 2009. The HBS1L-MYB intergenic interval associated
in erythroid cells. Blood 114:1254–1262.
distinct functions in active and inactive genes. Cell 138:1019–1031.
40. Wolff L. 1996. Myb-induced transformation. Crit. Rev. Oncogenesis
41. Wolff L, Koller R, Davidson W. 1991. Acute myeloid leukemia induction
by amphotropic murine retrovirus (4070A): clonal integrations involve
c-myb in some but not all leukemias. J. Virol. 65:3607–3616.
42. Wolff L, Mushinski JF, Shen-Ong GL, Morse HC, III. 1988. A chronic
inflammatory response. Its role in supporting the development of c-myb
and c-myc related promonocytic and monocytic tumors in BALB/c mice.
J. Immunol. 141:681–689.
Zhang et al.
jvi.asm.org Journal of Virology