Journal
of General Virology (2000), 81, 351–357. Printed in Great Britain
...................................................................................................................................................................................................................................................................................
BAG-1, a novel Bcl-2-interacting protein, activates expression
of human JC virus
Laxminarayana R. Devireddy, Kotlo U. Kumar,‡ Mary M. Pater† and Alan Pater
Division of Basic Medical Sciences, Faculty of Medicine, Memorial University of Newfoundland, St John’s, Newfoundland, Canada
Transcription of the human polyomavirus JC virus (JCV) genome is regulated by cellular proteins
and the large tumour (T) antigen. Earlier studies led to the identification of nuclear factor-1 (NF1)-binding sites in the JCV enhancer by DNase I protection assays of extracts from retinoic acid
(RA)-differentiated P19 embryonal carcinoma (EC) cells. In this study, a cDNA clone that encodes
a protein capable of binding to the JCV NF-1 sites was isolated from an RA-differentiated EC cell
cDNA library. Sequence analysis revealed that the cDNA isolated was identical to the previously
described Bcl-2-interacting protein BAG-1 (Bcl-2-associated athano gene-1). Results from RNA
studies indicated that BAG-1 is expressed in several cell types. Co-transfection of a recombinant
BAG-1 expression plasmid with JCV promoters indicated that BAG-1 stimulates transcription of the
JCVE promoter and to a lesser extent the JCVL promoter. Mutations in the NF-1 sites in the JCVE
promoter eliminated the activation by BAG-1. Thus, BAG-1 is a novel transcription factor that may
play a role in JCV expression.
Introduction
The human polyomavirus JC virus (JCV) is the aetiological
agent of progressive multifocal leukoencephalopathy (PML)
(reviewed in Major et al., 1992 ; Weber & Major, 1997). This
virus is closely related to two other polyomaviruses, the simian
virus 40 (SV40) and BK viruses, which share more than 70 %
identity in their protein-coding regions (Frisque et al., 1984).
However, compared with BK virus and SV40, JCV has a
narrower host range and tissue tropism. The greatest divergence between JCV and the other polyomaviruses is in the
virus control region, which contains promoters for early and
late gene transcription and is organized as two 98 bp direct
repeats. Several lines of evidence gained in vitro and in vivo,
including in vitro transcription from the virus promoters in
Author for correspondence : Laxminarayana R. Devireddy. Present
address : Howard Hughes Medical Institute, University of Massachusetts
Medical Center, 373 Plantation Street, Suite 309, Worcester,
MA 01605, USA. Fax j1 508 856 5473. e-mail
Laxminarayana.Devireddy!umassmed.edu
† This work is dedicated to the fond memory of Dr Mary M. Pater, who
passed away on 2 November 1994.
‡ Present address : Dept of Molecular Genetics, Univ. of
Illinois–Chicago, 900 S. Ashland Avenue, Chicago, IL 60607, USA.
0001-6618 # 2000 SGM
neuronal and non-neuronal cell extracts (Ahmed et al., 1990 ;
Kumar et al., 1993) and transient transfection of glial and nonglial cells with CAT reporter plasmids (reviewed in Raj &
Khalili, 1995), indicated that the JCV early and late promoters
are more active in cells derived from the central nervous
system (reviewed in Raj & Khalili, 1995). Thus, these data
suggest that the tissue tropism of JCV, at least to a large extent,
rests on tissue-specific expression of the virus promoters.
Previous studies have indicated that the cell-specific activation
of the JCV promoter requires its association with multiple
cellular transcription factors present in neuronal cells (reviewed
in Raj & Khalili, 1995).
Embryonal carcinoma (EC) cells are pluripotent in nature
and can be differentiated with chemicals into several different
cell lineages. For example, retinoic acid (RA) and DMSO
induce neuron- and skeletal muscle cell-like morphologies,
respectively. Thus, the EC cell system serves as a model system
for studying neuron-specific gene expression. Previous studies
from our laboratory have shown that the JCV promoter is
active only in RA-differentiated and not in undifferentiated or
DMSO-differentiated EC cells (Nakshatri et al., 1990 ; Kumar et
al., 1993). Nuclear extracts prepared from RA-differentiated
cells protect nuclear factor-1 (NF-1)-like sequences in the JCV
promoter (Nakshatri et al., 1990 ; Kumar et al., 1993) and
disruption of these sites reduces the activity of the JCV
promoter (Kumar et al., 1993). Furthermore, a novel factor
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:15:20
DFB
L. R. Devireddy and others
present in RA-differentiated EC cells seems to interact with
these sites, as judged by the results of band-shift and UV crosslinking assays (Kumar, 1994). However, the identity and nature
of this protein remain unknown. Cloning and characterization
of such factors may shed more light on neuron-specific
expression of human JCV.
BAG-1 (Bcl-2-associated athano gene-1) is a novel Bcl-2interacting protein that was first identified as a multifunctional
molecule that binds to the anti-apoptotic protein Bcl-2 and
promotes cell survival (Takayama et al., 1995). Since its
discovery, BAG-1 has been shown to form complexes with
several other proteins, including tyrosine kinase growth-factor
receptors such as hepatocyte- and platelet-derived growth
factors (Bardelli et al., 1996), the serine\threonine protein
kinase Raf-1 (Wang et al., 1996 ; reviewed in Wang & Reed,
1998) and the retinoic acid receptor (Liu et al., 1998).
Alternative translation of BAG-1 generates a series of proteins
with variable lengths (Packham et al., 1997). Longer forms of
BAG-1, with unique amino-terminal domains, have also been
reported to form complexes with steroid-hormone receptors
(Packham et al., 1997 ; Froesch et al., 1998). Binding of BAG-1
to these molecules modulates their activity. The mechanism by
which BAG-1 influences the activities of such diverse proteins
can perhaps be attributed to its ability to bind heat-shock
proteins directly, which in turn interact with multiple target
proteins in cells. BAG-1 contains an amino-terminal region
with similarity to ubiquitin and a central region that binds Bcl2 (Takayama et al., 1995). Its carboxy terminus is required
for formation of complexes with Raf-1 and growth factor
receptors.
It is well established that the expression of JCV early and
late genes, encoding the tumour (T) antigen and capsid
proteins, respectively, determines the narrow host range and
specificity of this virus. Studies from our lab and other
laboratories have demonstrated that the JC promoter\enhancer
functions more efficiently in neuronal cells than in nonneuronal cells. The studies presented here attempt to identify
the proteins that bind the JCV promoter\enhancer region.
Here, we report the cloning of a novel protein by Southwestern
screening of a P19 RA-differentiated cDNA library with a JCV
NF-1 probe. Interestingly, this protein is identical to a Bcl-2interacting protein, BAG-1. Furthermore, this protein is able to
bind the NF-1 sequence and transactivate JCV promoters.
Methods
Cells, plasmids and transfection procedures. P19 EC, U87
MG human glioblastoma and HeLa cells were maintained in Dulbecco’s
modified Eagle’s medium. P19 cells were differentiated with 1 % DMSO
into skeletal or cardiac muscle cells or differentiated with 300 nM all
trans-retinoic acid (Sigma) into neuron-like cells as described previously
(Nakshatri et al., 1990). Plasmids pCMV, pBluescript KS(j) and pUC19
were purchased from Invitrogen, Stratagene and New England Biolabs,
respectively. pJCVE, JCVL, RIIE, II (bearing mutations in two NF-1 sites
"!
present within the two 98 bp repeats) and DM (bearing mutations in all
"!
DFC
three NF-1 sites), CAT reporter plasmids and pRSV-β-Gal were described
by Nakshatri et al. (1990) and Kumar et al. (1993). BKE CAT (BK virus
early promoter linked to the CAT gene) was described by Nakshatri et al.
(1991). Cells were transfected with the amounts of plasmids indicated by
the calcium phosphate method as described by Kumar et al. (1993).
Oligonucleotides. The sequences of wild-type (WT) and mutant
(mt) JCV NF-1 oligonucleotides used are : WT NF-1, 5h TGGCTGCCAGCCAA 3h ; and mt NF-1, 5h gtaCTGCCAGaCcAGCA 3h (see
Kumar et al., 1993). Lower-case letters indicate mutated bases.
Construction and screening of the cDNA library. A λgt22A
cDNA library from mRNA of P19 RA-differentiated cells was made by
using a kit from GIBCO-BRL according to the manufacturer’s instructions.
Screening of the cDNA expression library was performed by the
Southwestern blot method of Vinson et al. (1988). Freshly prepared E. coli
Y1090R− bacteria were infected with recombinant phage from the P19
RA λgt22A library at " 5i10% p.f.u. per 150 mm plate. The infection
was performed by incubating at 37 mC for 30 min. Next, 7n5 ml 0n7 % top
agarose was mixed with the bacteria and the mixture was overlaid onto
1n5 % T-Tyn (tryptone, yeast extract and NaCl) agar plates. After
incubation at 42 mC for 4 h, IPTG-impregnated nitrocellulose filters were
overlaid onto the plates and the plates were incubated for additional 6 h
at 37 mC. The filters were then removed from the plates, air-dried and
immersed in binding buffer (6 M guanidine–HCl ; 25 mM HEPES, pH 7n9 ;
3 mM MgCl ; 4 mM KCl and 1 mM DTT) for 5 min at 4 mC with gentle
#
agitation. The binding buffer was then diluted with an equal volume of
the same buffer but containing variable amounts of guanidine–HCl and
the filters were incubated in these diluted buffers for 5 min each. The
filters were then washed with binding buffer without guanidine–HCl. The
filters were then blocked by incubating in binding buffer containing 5 %
non-fat dried milk powder for 30 min at 4 mC. After rinsing in binding
buffer containing 0n25 % milk powder, the filters were probed with 10'
c.p.m.\ml nick-translated JCV NF-1 oligonucleotide concatemer in
binding buffer containing 0n25 % milk powder and 10 µg\ml sonicated
salmon sperm DNA. After 3 h hybridization at 4 mC, the filters were
washed three times for 15 min each with binding buffer containing 0n25 %
milk powder. Finally, the filters were wrapped in Saran Wrap and exposed
to X-ray film overnight at k70 mC. Positive plaques were identified and
re-screened until they were homogeneously positive.
Sequence analysis. The cDNA, which was cloned into pBluescript
KS(j) (Stratagene), was sequenced by using a kit from United States
Biochemical according to the manufacturer’s instructions. The sequence
was compared with known sequences in GenBank by BLAST search.
Northern blot analysis. Total RNA from undifferentiated (UD),
RA- and DMSO-differentiated EC cells, U87 MG and HeLa cells was
extracted by using a kit from Promega, according to the manufacturer’s
instructions. Twenty µg total RNA was electrophoresed on a 1 %
formaldehyde–agarose gel and transferred to nylon membrane (Gilman
Sciences). The filter was blocked with Denhardt’s solution and probed
with 10' c.p.m.\ml nick-translated kNF-1 cDNA overnight at 60 mC.
After hybridization, the filter was washed and exposed to X-ray film. The
hybridized probe was stripped with 0n1iSSC and 0n1 % SDS and
reprobed with an actin probe as described above.
Gel-shift analysis. The sequences of JCV NF-1 oligonucleotides
were described by Kumar et al. (1993). The probes were prepared by endlabelling the oligonucleotides in the presence of [α-$#P]dCTP. The
binding reactions were done in a volume of 35 µl in buffer containing
10 mM Tris–HCl, pH 7n8 ; 1 mM EDTA ; 5 mM DTT ; 150 mM KCl ;
12 % glycerol ; 10 µg poly(dI.dC), 10 ng labelled probe and 4 µl in vitrotranslated BAG-1. The reaction mixture was incubated at room
temperature for 30 min. The reaction products were resolved on a 4 %
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:15:20
Regulation of JCV transcription by BAG-1
non-denaturing PAGE gel. Gels were then dried and subjected to
autoradiography.
Southwestern blot analysis. Nuclear extracts from UD, RA- and
DMSO-differentiated EC cells, HeLa and U87 MG cells were resolved on
a 10 % SDS–PAGE gel and transferred to PVDF membrane in a semi-dry
transfer apparatus (Bio-Rad) according to the manufacturer’s instructions.
The filters were then blocked overnight at 4 mC with 5 % milk powder
in a binding buffer containing 25 mM HEPES–NaOH, pH 7n9 ; 5 mM
MgCl ; 0n5 mM DTT and 25 mM NaCl. After blocking, the filters were
#
probed with 10' c.p.m.\ml nick-translated JCV NF-1 oligonucleotide in
binding buffer. The filters were then washed with binding buffer without
milk powder, dried and subjected to autoradiography.
CAT assays. CAT assays were performed as described previously
(Nakshatri et al., 1990). The percentage of acetylation was quantified by
liquid scintillation and normalized on the basis of β-galactosidase activity.
Results
Isolation of a cDNA encoding a JCV NF-1 regionbinding protein from P19 RA-differentiated cells
Previous studies from our laboratory indicated that NF-1
sites present in the JCV enhancer play an important role in the
neuron-specific expression of JCV (Nakshatri et al., 1990 ;
Kumar et al., 1993). Furthermore, a novel protein seems to
interact with the NF-1 motif (Kumar, 1994). This protein is
(a)
UD
RA
DMSO
U87
HeLa
(a)
2.37
kNF-1
1.35
(b)
2.37
Actin
1.35
Fig. 2. Northern blot analysis of kNF-1 RNAs from various cell lines. Total
RNAs were extracted from the cell lines as described in Methods and
20 µg RNA was used in Northern blot analysis. kNF-1 cDNA was used as
a probe (a). The same blot was then stripped and reprobed with an actin
probe (b). Numbers on the right indicate sizes of molecular mass markers
(kb).
present in both neuronal and non-neuronal cells and is able to
interact specifically and directly with the JCV NF-1 motif. To
examine the identity of this protein, an NF-1 oligonucleotide
probe was used to screen a λgt22A cDNA expression library
from RA-differentiated EC cells. After screening more than
2i10& plaques, two positive plaques were identified. The
(b)
Fig. 1. Sequence analysis of kNF-1 cDNA. (a) Nucleotide sequence of kNF-1 and the deduced amino acid sequence. Numbers
on the left indicate the positions of nucleotides. Amino acid positions are indicated on the right. The first three start methionines
are in bold. The polyadenylation signal is in bold and doubly underlined. (b) Amino acid sequence comparison of kNF-1 and
BAG-1. The sequences were aligned by using the GAP program of GCG. Vertical lines indicate identity.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:15:20
DFD
L. R. Devireddy and others
(a)
(b)
Fig. 3. Analysis of DNA-binding properties of the kNF-1-encoded protein. (a) Gel-shift analysis of in vitro-translated kNF-1.
Lanes : 1, free probe ; 2–4, binding reactions containing 4 µl in vitro-translated kNF-1 protein with no competitor (lane 2), a
200-fold excess of WT JCV NF-1 oligonucleotide (lane 3) or a 200-fold excess of mutant JCV NF-1 oligonucleotide (lane 4) ;
5–7, binding reactions containing 4 µl rabbit reticulocyte lysate with no competitor (lane 5), a 200-fold excess of WT JCV NF-1
oligonucleotide (lane 6) or a 200-fold excess of mutant JCV NF-1 oligonucleotide (lane 7) ; 8–10, binding reactions containing
4 µl in vitro-translated luciferase protein with no competitor (lane 8), a 200-fold excess of WT JCV NF-1 oligonucleotide (lane
9) or a 200-fold excess of mutant JCV NF-1 oligonucleotide (lane 10). The arrow indicates the position of the free probe.
(b) Southwestern blot analysis of nuclear extracts from indicated cell types. Nuclear extracts (50 µg) or BSA (50 µg) were
resolved on a 10 % SDS–PAGE gel, transferred to PVDF membrane and probed with WT JCV NF-1 oligonucleotide. The arrow
indicates the position of the 30 kDa protein. The faster-migrating band in P19 RA-differentiated cells may be a proteolytically
processed form. The numbers on the right indicate the positions of molecular mass markers (kDa).
specificity of interaction of these plaques with the NF-1 probe
was further confirmed. Replica filter papers containing plaques
of the two positive recombinant phages were hybridized with :
(i) WT NF-1 ; (ii) mt NF-1 (Kumar et al., 1993) and (iii) a nonspecific oligonucleotide. Only the WT NF-1 probe (data not
shown), and not the mutant or non-specific probes, interacted
with the positive recombinant plaques (data not shown).
Analysis of DNA extracted from the two positive plaques
yielded the same sequence, and only one plaque was
characterized further. The positive recombinant phage encoding the JCV NF-1 region-binding protein was termed kNF1 (Kumar, 1994). Analysis of kNF-1 revealed a " 1 kb DNA
fragment (Fig. 1 a) with perfect sequence identity to BAG-1
(Fig. 1 b ; Takayama et al., 1995). The kNF-1 cDNA contains an
open reading frame (ORF) of 229 amino acids with the
potential to encode a 30 kDa protein. The ORF corresponds to
sequences spanning the entire BAG-1 sequence. BAG-1 is a
DFE
Bcl-2-interacting protein isolated from a mouse embryonic
cDNA library in the yeast two-hybrid system (Takayama et al.,
1995). The sequence identity of kNF-1 to BAG-1 is likely to
reflect the isolation of the same gene.
Northern blot analysis
JCV expresses its genes and replicates efficiently in brain
cells (Kenney et al., 1984) and the NF-1 sites in the JCV
regulatory region were shown to be critical for neuron-specific
expression (Kumar et al., 1993). To determine the cell-specific
expression of kNF-1, Northern blot analysis was performed
with RNAs derived from UD, RA- and DMSO-differentiated
EC cells, U87 MG and HeLa cells. Probing and washing were
performed under high-stringency conditions to eliminate crosshybridization. As shown in Fig. 2 (a), a major " 1n5 kb RNA
species was detected in UD, RA- and DMSO-differentiated EC
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:15:20
Regulation of JCV transcription by BAG-1
(a)
(b)
Fig. 4. Transactivation of JCV promoters by kNF-1. (a) Plasmids containing the CAT reporter gene under the control of the JCV
early (JCVE) or late promoter (JCVL), the BKV early promoter (BKVE) and one 98 bp repeat (RIIE) were transfected (2 µg each)
into HeLa cells alone or with 10 µg kNF-1 expression plasmid (pCMV-kNF-1). The graph below shows the adjusted percentage
acetylation values. (b) Integrity of NF-1 sites is required for kNF-1-mediated transactivation. A JCV enhancer bearing mutations
either in the two NF-1 sites present within the 98 bp repeats (II10 CAT) (lanes 1 and 2) or in all the three NF-1 sites (DM10
CAT) (lanes 3 and 4) (Kumar et al., 1993) was co-transfected with (lanes 2 and 4) or without (lanes 1 and 3) 10 µg kNF-1.
Cell extracts were prepared 48 h post-transfection and CAT activity was measured. The graph below shows the adjusted
percentage acetylation values.
cells, whereas a " 1n7 kb RNA species was detected in HeLa
cells. From the intensity of the bands, it appears that kNF-1
RNA was present at approximately similar levels in all cell
types with the exception of U87 MG cells, which did not
express kNF-1 RNA. Stripping and reprobing the same blot
with an actin probe showed that equal amounts of RNA were
loaded on the gel (Fig. 2 b). These observations indicate that a
major 1n5 kb kNF-1 RNA species is produced in neuron-like
cells.
Nucleoprotein complex formation
The binding activities of crude nuclear extracts and in vitrotranslated kNF-1 were tested by Southwestern and gel-shift
analyses, respectively. To examine the ability of in vitrotranslated kNF-1 protein to bind the JCV NF-1 oligonucleotide,
a gel-shift assay was performed. As shown in Fig. 3 (a), in vitrotranslated kNF-1 protein reduced the mobility of the JCV NF1 probe (lane 2). This retardation was eliminated by the
addition of a 200-fold excess of unlabelled WT JCV NF-1
oligonucleotide (lane 3), but not by the addition of the mutant
JCV NF-1 oligonucleotide (lane 4), confirming the specificity of
binding of kNF-1 protein to the JCV NF-1 oligonucleotide. No
retardation was observed with a reticulocyte lysate (lanes 5–7)
or with in vitro-translated luciferase protein (lanes 8–10).
The DNA-binding activity of kNF-1 was further confirmed
in Southwestern (DNA–protein) blot assays. Nuclear extracts
from the cell types indicated were resolved on an SDS–PAGE
gel, transferred to PVDF membrane and probed with WT JCV
NF-1 oligonucleotide. A 30 kDa protein was detected in P19
RA, P19 DMSO and HeLa cell extracts (Fig. 3 b). The fastermigrating protein in P19 RA cells may be a proteolytically
degraded kNF-1 (Devireddy, 1995). However, a 30 kDa
protein was not detected in extracts from U87 MG glioblastoma cells. Furthermore, these cells are negative for kNF-1
mRNA expression (Fig. 2). A 30 kDa protein was also detected
in extracts from P19 RA, P19 DMSO and HeLa cells in
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:15:20
DFF
L. R. Devireddy and others
Western blot assays using a BAG-1-specific antibody (data not
shown). The identity and nature of the 97 kDa protein present
in all cell types are not known. However, a similar-sized
protein present in glial and non-glial cells was also shown to
interact with the JCV B domain (which harbours the NF-1 site)
(Khalili et al., 1988). In summary, these results suggest that
BAG-1 binds the JCV NF-1 oligonucleotide.
Transcriptional activity of kNF-1 cDNA in HeLa cells
To examine the transcriptional activity of the kNF-1
protein, the kNF-1 cDNA was cloned into a eukaryotic
expression vector, pRc\CMV. Non-neuronal HeLa cells that
do not support JCV transcription were employed (Tada et al.,
1989). Reporter plasmids with the CAT gene under the control
of the JCV early (JCVE) or late (JCVL) promoters were
transfected into HeLa cells alone or with pCMV-kNF-1. A
basal level of activity from the JCVE and JCVL promoters was
detected in HeLa cells (Fig. 4 a). Co-transfection of kNF1 cDNA resulted in 6-fold and 5-fold activation of the JCVE
and JCVL promoters, respectively (Fig. 4 a). Co-transfection of
pRc\CMV alone had no effect (data not shown). The RIIE
expression plasmid, containing only one JCV 98 bp repeat,
was also transactivated, by 2n5-fold (Fig. 4 a). The BK virus
early promoter was not transactivated by kNF-1 cDNA (Fig.
4 a). A JCV enhancer\promoter containing mutations in the
NF-1 sites was not transactivated by kNF-1, suggesting that
the integrity of these sites is necessary for kNF-1-mediated
transactivation (Fig. 4 b). In summary, these results suggest that
kNF-1 specifically transactivated JCV promoters.
Discussion
In this study, we isolated a cDNA (kNF-1) encoding a JCV
NF-1 motif-binding protein from P19 RA-differentiated cells
that is identical to a Bcl-2-interacting protein, BAG-1 (Fig. 1).
Furthermore, the kNF-1 protein bound specifically to the JCV
NF-1 region and up-regulated transcription of the viral
promoters in non-glial cells (Figs 3 and 4). Northern blot
analysis for detection of kNF-1 mRNA revealed the presence
of BAG-1 mRNA of varying sizes (Fig. 2).
JCV infection occurs during childhood in the majority of
the population, but only during immunosuppression does this
virus become pathogenic and cause PML. The virus replicates
only in neuronal cells, but the mechanism that determines
neuron-specific expression of virus genes is not completely
understood. It has been postulated that a regulatory pathway
that includes participation of neuronal and non-neuronal
transcription factors plays a role in the neuron-specific
regulation of JCV (reviewed in Raj & Khalili, 1995). The
functional redundancy of promoter elements and the requirement for multiple cellular transcription factors further
support the hypothesis of multiple cis and trans determinants of
JCV neurotropism.
DFG
The data presented here led to the identification of a novel
transcription factor, BAG-1, that bound specifically to the JCV
NF-1 sequences and up-regulated transcription of the JCV
early and late promoters. Recent studies indicate that BAG-1
and its longer isoforms BAG-1M and BAG-1L influence a wide
variety of cellular phenotypes through their interaction with
Hsc70\Hsp70, including resistance to apoptosis, promotion of
cell proliferation, enhancement of tumour cell migration and
alteration of transcriptional activity of steroid-hormone receptors (Takayama et al., 1995 ; Wang et al., 1996 ; Bardelli et al.,
1996 ; Liu et al., 1998 ; Froesch et al., 1998). The ability of the
BAG-1 family of proteins to modulate the chaperone activity
of Hsc70\Hsp70 may have different consequences for the
various proteins whose functions are controlled by conformational change. For example, BAG-1 proteins have been
reported to bind Bcl-2, Raf-1, androgen receptor and PDGF
receptor and enhance their function. Conversely, interaction of
BAG-1 with retinoic acid receptor, glucocorticoid receptor and
Siah-1 seems to inhibit their activity (Takayama et al., 1995 ;
Wang et al., 1996 ; Bardelli et al., 1996 ; Liu et al., 1998 ;
Kullmann et al., 1998 ; Matsuzawa et al., 1998 ; Froesch et al.,
1998). The Bcl-2 protein, which has an anti-apoptotic function,
recruits cytosolic proteins to internal membranes, including
Raf-1 and BAG-1 (Wang et al., 1996 ; Takayama et al., 1998).
Bcl-2 also negatively regulates p53 transcriptional activity
(Froesch et al., 1999), possibly by sequestering transcription
factors necessary for p53 transcriptional activity. Analogously,
BAG-1 may modulate JCV expression by interacting with
transcription factors.
It is intriguing that a ubiquitously expressed, anti-apoptotic
and chaperone regulator BAG-1 activates the expression of
JCV. Many distinct promoter elements, including NF-1,
contribute to the tropism of the JCV promoter (Nakshatri et al.,
1990 ; Kumar et al., 1993). The restriction in expression of
individual transcription factors to glial cells, including Tst-1,
NF-1 and Sp1, may confer neuron-specific expression to JCV
(Nakshatri et al., 1990 ; Henson et al., 1992 ; Wegner et al.,
1993). Therefore, the relative abundance of transcription
factors, rather than a single determining factor, may account
for the neurotropism of JCV.
This work was part of the MSc and PhD theses of L. R. D. and K. U. K.,
respectively, submitted to the Memorial University of Newfoundland.
We thank Doreen Bailey and Jose Teodoro for help in preparation of the
manuscript. This work was supported by grants from the National Cancer
Institute of Canada (with funds from the Canadian Cancer Society) and
the Medical Research Council of Canada. L. R. D. and K. U. K. contributed
equally to this work.
References
Ahmed, S., Chowdhury, M. & Khalili, K. (1990). Regulation of a human
neurotropic virus promoter, JCVE : identification of a novel activator
domain located upstream from the 98 bp enhancer promoter region.
Nucleic Acids Research 18, 7417–7423.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:15:20
Regulation of JCV transcription by BAG-1
Bardelli, A., Longati, P., Albero, D., Goruppi, S., Schneider, C.,
Ponzetto, C. & Comoglio, P. M. (1996). HGF receptor associates with
Matsuzawa, S., Takayama, S., Froesch, B. A., Zapata, J. M. & Reed,
J. C. (1998). p53-inducible human homologue of Drosophila seven in
the anti-apoptotic protein BAG-1 and prevents cell death. EMBO Journal
15, 6205–6212.
Devireddy, L. R. (1995). Characterization of a cDNA encoding a novel
transcription factor regulating expression of human neurotropic JC virus. MSc
thesis, Memorial University of Newfoundland, St John’s, Newfoundland,
Canada.
Frisque, R. J., Bream, G. L. & Cannella, M. T. (1984). Human
polyomavirus JC virus genome. Journal of Virology 51, 458–469.
Froesch, B. A., Takayama, S. & Reed, J. C. (1998). BAG-1L protein
enhances androgen receptor function. Journal of Biological Chemistry 273,
11660–11666.
absentia (Siah) inhibits cell growth : suppression by BAG-1. EMBO Journal
17, 2736–2747.
Nakshatri, H., Pater, A. & Pater, M. M. (1990). Activity and enhancer
binding factors for JC virus regulatory elements in differentiating
embryonal carcinoma cells. Virology 177, 784–789.
Nakshatri, H., Pater, M. M. & Pater, A. (1991). Activity and enhancer
binding factors for BK virus regulatory elements in differentiating
embryonal carcinoma cells. Virology 183, 374–380.
Packham, G., Brimmell, M. & Cleveland, J. L. (1997). Mammalian cells
express two differently localized Bag-1 isoforms generated by alternative
translation initiation. Biochemical Journal 328, 807–813.
Raj, G. V. & Khalili, K. (1995). Transcriptional regulation : lessons from
the human neurotropic polyomavirus, JCV. Virology 213, 283–291.
Tada, H., Lashgari, M., Rappaport, J. & Khalili, K. (1989). Cell typespecific expression of JC virus early promoter is determined by positive
and negative regulation. Journal of Virology 63, 463–466.
Froesch, B. A., Aime-Sempe, C., Leber, B., Andrews, D. & Reed, J. C.
(1999). Inhibition of p53 transcriptional activity by Bcl-2 requires its
membrane-anchoring domain. Journal of Biological Chemistry 274,
6469–6475.
Henson, J. W., Saffer, J. & Furneaux, H. (1992). The transcription factor
Sp1 binds to the JC virus promoter and is selectively expressed in glial
cells in human brain. Annals of Neurology 32, 72–77.
Kenney, S., Natarajan, V., Strike, D., Khoury, G. & Salzman, N. P.
(1984). JC virus enhancer-promoter active in human brain cells. Science
Takayama, S., Sato, T., Krajewski, S., Kochel, K., Irie, S., Millan, J. A. &
Reed, J. C. (1995). Cloning and functional analysis of BAG-1 : a novel
Bcl-2-binding protein with anti-cell death activity. Cell 80, 279–284.
Khalili, K., Rappaport, J. & Khoury, G. (1988). Nuclear factors in human
Takayama, S., Krajewski, S., Krajewska, M., Kitada, S., Zapata, J. M.,
Kochel, K., Knee, D., Scudiero, D., Tudor, G., Miller, G. J., Miyashita, T.,
Yamada, M. & Reed, J. C. (1998). Expression and location of
brain cells bind specifically to the JCV regulatory region. EMBO Journal
7, 1205–1210.
Hsp70\Hsc-binding anti-apoptotic protein BAG-1 and its variants in
normal tissues and tumor cell lines. Cancer Research 58, 3116–3131.
Kullmann, M., Schneikert, J., Moll, J., Heck, S., Zeiner, M., Gehring, U.
& Cato, A. C. (1998). RAP46 is a negative regulator of glucocorticoid
Vinson, C. R., LaMarco, K. L., Johnson, P. F., Landschulz, W. H. &
McKnight, S. L. (1988). In situ detection of sequence-specific DNA
receptor action and hormone-induced apoptosis. Journal of Biological
Chemistry 273, 14620–14625.
Kumar, K. U. (1994). Transcriptional regulation of human JC virus. PhD
thesis, Memorial University of Newfoundland, St John’s, Newfoundland,
Canada.
Kumar, K. U., Pater, A. & Pater, M. M. (1993). Human JC virus perfect
palindromic nuclear factor 1-binding sequences important for glial cellspecific expression in differentiating embryonal carcinoma cells. Journal of
Virology 67, 572–576.
binding activity specified by a recombinant bacteriophage. Genes &
Development 2, 801–806.
Wang, H. G. & Reed, J. C. (1998). Bcl-2, Raf-1 and mitochondrial
regulation of apoptosis. Biofactors 8, 13–16.
Wang, H. G., Takayama, S., Rapp, U. R. & Reed, J. C. (1996). Bcl-2
interacting protein, BAG-1, binds to and activates the kinase Raf-1.
Proceedings of the National Academy of Sciences, USA 93, 7063–7068.
Weber, T. & Major, E. O. (1997). Progressive multifocal leukoencephalopathy : molecular biology, pathogenesis and clinical impact.
Intervirology 40, 98–111.
Wegner, M., Drolet, D. W. & Rosenfeld, M. G. (1993). Regulation of JC
virus by the POU-domain transcription factor Tst-1 : implications for
progressive multifocal leukoencephalopathy. Proceedings of the National
Academy of Sciences, USA 90, 4743–4747.
226, 1337–1339.
Liu, R., Takayama, S., Zheng, Y., Froesch, B., Chen, G. Q., Zhang, X.,
Reed, J. C. & Zhang, X. K. (1998). Interaction of BAG-1 with retinoic
acid receptor and its inhibition of retinoic acid-induced apoptosis in
cancer cells. Journal of Biological Chemistry 273, 16985–16992.
Major, E. O., Amemiya, K., Tornatore, C. S., Houff, S. A. & Berger,
J. R. (1992). Pathogenesis and molecular biology of progressive
multifocal leukoencephalopathy, the JC virus-induced demyelinating
disease of the human brain. Clinical Microbiology Reviews 5, 49–73.
Received 3 August 1999 ; Accepted 21 October 1999
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:15:20
DFH