| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
*
Molecular Biophysics and Biochemistry and
Genetics, Yale University School of Medicine, New Haven, CT 06520
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our laboratory has developed a transient transfection assay for the
presence of activities that support switch recombination
(5, 6, 7, 8, 9). This assay measures recombination between Sµ and
S
3 region sequences carried by a shuttle vector. In the first step,
recombination occurs during transfection of mammalian cells; in the
second step, recombinants are enumerated by genetic selection in
Escherichia coli. Similar substrates have been developed by
Lieber and collaborators (10, 11). Other laboratories have
developed integrating substrates that undergo recombination in
transfected cell lines, including substrates carried on both plasmid
and retroviral vectors (12, 13, 14, 15).
The extrachromosomal switch substrates have proved to be a powerful tool for assaying switch recombination activities in murine cells. The substrates effectively recapitulate chromosomal switch recombination, in that they recombine at high frequencies in primary LPS-activated murine B cells (5) and in pre-B and B cell lines, but not other cell types (6, 9, 10); the majority of recombination junctions map to sites within the S regions (5, 6, 7, 8, 9, 10, 11); junctions are heterogeneous in site and sequence (5, 6, 7, 8, 9, 10, 11); and mutations are evident within a few hundred nucleotides of one side of the junction (11). However, thus far the use of extrachromosomal switch substrates has been limited to murine cell lines, because the substrates developed by both our own laboratory (5, 6, 7, 8, 9) and Lieber’s laboratory (10, 11) are carried on a polyoma-based shuttle vector, and polyoma does not replicate in human cells. To study how EBV infection affects switch recombination activities, we have now modified the extrachromosomal substrates to allow analysis of recombination in human cells. Using recombination cassettes carried on an SV40-based shuttle vector, we have examined recombination activities in the EBV-negative lymphoma, BJAB, and its EBV-positive derivative, BJAB-BI, as well as in other cell types. We have found that recombination levels were very low in BJAB but that BJAB-B1 and group I and group III EBV-positive BL cells supported high levels of switch substrate recombination. EBV infection can therefore induce activities that support switch recombination. One B cell-specific activity implicated in switch recombination is LR1, a sequence-specific duplex DNA binding factor identified and characterized by our laboratory (16, 17, 18). LR1 is a transcriptional regulator of the c-myc protooncogene (19) and of the EBV Fp promotor (20), and LR1 binds specifically to the G-rich consensus sequence, GGNCNAG(G/C)CTG(G/A). We further show that recombination frequencies correlated with levels of LR1 duplex DNA binding activity.
Our data support a model in which EBV infection mimics Ag activation by inducing activities essential to switch recombination. These activities may be important in replication or recombination of the EBV genome, and they are also implicated in the translocations of c-myc to the Ig loci.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The extrachromosomal switch substrates were derived from pHL22 and pHL122 (5), which were designed to assay recombination in murine cells and carried the polyoma early region. To allow replication in human cells, we replaced the polyoma early region with the SV40 large T and replication origin to create the following constructs:
pML22. The recombination cassette was excised from pHL22 as a
5.9-kb NotI-EcoRV fragment, which includes 2.2 kb
of Sµ sequence, the leftward promoter of bacteriophage 
(PL), and 2.5 kb of S3 sequence. This
fragment was cloned into shuttle vector pGG51 (a gift from Dr. Michael
Lieber, University of Southern California, Los Angeles, CA), which
contains the SV40 origin of replication and large T Ag coding sequence,
the prokaryotic ColE1 replication origin, and an ampicillin resistance
gene; pGG51 had been digested with SalI and BglII
and adapted with NotI linkers, as had the vector (Fig. 1
).
pML23. A 962-bp HindIII fragment carrying a
kanamycin resistance gene was excised from pBS.Km2, filled with Klenow,
and inserted into the NarI site of pML22 just downstream of
S3 (Fig. 1).
|
PL, and
S3 sequences was excised from pHL122 and cloned into pGG51 at its
SalI and (filled) BglII sites.
The PL promoter is a conditionally lethal
marker. Plasmids carrying PL cannot produce
antibiotic-resistant transformants of nonlysogenic strains of E.
coli, because active transcription of PL
interferes with plasmid replication. These plasmids can replicate in
lysogens, in which the prophage produces the repressor protein
which prevents transcription of PL. Upon
recombination involving the S regions, PL is
deleted, and the resulting recombinants can transform a nonlysogenic
strain of E. coli.
Plasmids were propagated in E. coli DH10B() and purified
by Qiagen (Chatsworth, CA). Before transfection of eukaryotic cells,
all plasmid preparations were tested for the
PL marker by comparison of the number of
transformants of DH10B and the isogenic lysogen, DH10B(). The
frequency of inactivation of PL was in all
cases <10-4, which is in the range of the
mutation rate in E. coli (data not shown). Restriction
mapping of plasmid recovered from DH10B, in which
PL had been inactivated, showed that the
mutation events that had occurred during propagation in E.
coli were single base changes or deletions and insertions so small
that they did not affect the restriction map.
Cell lines, cell culture, and transfection
The following human cell lines were used in the switch substrate
recombination assay: Raji, a group III EBV-positive BL
(21); Akata, a group I EBV-positive BL (22);
BJAB, an EBV-negative BL (23); BJAB-B1, an EBV-infected
derivative of BJAB (24); BJE2, an Epstein-Barr nuclear Ag
(EBNA)-2 cDNA-transfected derivative of BJAB (25); and
HT1080, a fibrosarcoma cell line (26). B cell lines were
cultured in RPMI 1640 and HT1080 was cultured in MEM-
. All media
were supplemented with 10% FBS, 100 U/ml penicillin, 100 µg/ml
streptomycin, and 50 µM 2-ME.
Optimal transfection conditions were determined by transfection with a reporter construct in which a chloramphenicol acetyltransferase (CAT) gene was driven by the CMV IE1 promoter-enhancer (20). Akata, BJAB-B1, BJAB, and BJE2 cell lines were transfected using DEAE-dextran, as described previously (9). Briefly, 2 x 107 cells were pelleted, washed with RPMI 1640, and resuspended in 1 ml RPMI 1640 containing 500 µg/ml DEAE-dextran (Pharmacia, Piscataway, NJ) and 10 µg Qiagen-purified plasmid DNA. After 30 min incubation at 37°C, cells were washed with RPMI 1640 and resuspended in 30 ml culture medium. For transfection of Raji cells, 4 x 107 cells were washed with PBS (137 mM NaCl, 2.7 mM KCl, 8.1 mM Na2HPO4, 1.5 mM KH2PO4, pH 7.4), resuspended in 0.8 ml PBS, and mixed with 10 µg DNA. The reaction was loaded into a 0.4-cm gap electroporation cuvette (Bio-Rad, Foster City, CA) and electroporated at 350 V, 960 µF on a Bio-Rad gene pulser. Transfection of HT1080 cells was conducted with the Calcium Phosphate Transfection System (Life Technologies, Gaithersburg, MD) as recommended by the manufacturer.
Switch substrate recombination assay
The switch substrate recombination assay has been described in
detail previously (5, 6, 9). Briefly, cells were
transfected with substrate, and 42 h after transfection low m.w.
DNA was isolated by rapid alkaline lysis. A compatible plasmid
pACYC184, which contains a gene encoding chloramphenicol resistance,
was added to recovered DNA as an internal control for transformation,
and samples were digested with DpnI to destroy molecules
that had not replicated in the eukaryotic cells. Recombination
substrates were propagated on dam+ E.
coli strains, whereas pACYC184 was propagated on a
dam- strain of E. coli and was
therefore resistant to DpnI digestion. Equal aliquots of DNA
were transformed into E. coli DH10B and the isogenic
lysogen DH10B() by electroporation. The recombination frequency
(R) was calculated as the ratio of ampicillin-resistant
(AmpR) (or in some experiments
kanamycin-resistant (KanR)) transformants of the
nonlysogen compared with the lysogen, normalized for transformation
efficiency as measured by chloramphenicol-resistant
(CmR) transformants:
 |
Mapping of recombinants
Minipreps of recombinant DNA were produced by alkaline lysis, and deletion endpoints of recombinant plasmids were mapped by digestion with HindIII, NotI, and XhoI.
Gel mobility shift assay
Gel mobility shift assays were conducted as described (16, 19). Nuclear extracts from cultured cells were incubated with
32P-labeled synthetic duplex oligonucleotides
containing an LR1 site from the murine S1 switch region, or a mutant
site (G14A) differing at a single base pair. DNA-protein complexes were
resolved in a 7% polyacrylamide gel in 90 mM Tris-borate, 1 mM EDTA,
pH 8.3.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We first assayed recombination of pML22 in BJAB, its EBV-infected
derivative, BJAB-B1, and in the human fibroblast line, HT1080. As shown
in Table I, the recombination frequency
in BJAB was R = 1.1%. This low level was comparable
with that in the fibroblast line, HT1080. In contrast, the
recombination frequency in BJAB-B1 was 5-fold higher, R
= 5.5%. This is comparable with the level observed with murine
substrates recovered from transfection of B cell lines
(9).
|
100 genes, but cells transformed by EBV express only a
small fraction of them. BLs are classified as group III if they express
EBV genes including the nuclear proteins, EBNA-1, -2, -3A, -3B, -3C,
and -LP, several latent membrane proteins (LMPs), and two EBER RNAs;
and as group I if the only EBNA expressed is EBNA-1. We tested the
ability of representatives of these two classes of EBV-transformed BLs
to support switch substrate recombination. The recombination level in
Akata (group I) was R = 6.5%, comparable with that of
BJAB-B1. It was significantly higher in Raji, R =
11.7%. BJE2 cells, which stably express EBNA-2 but no other EBV genes,
supported only background levels of recombination. These results
show that genes encoded by EBV enhance switch substrate
recombination in these BL cell lines. Enhanced recombination activities in EBV+ lymphoma cells target switch region sequences
It was critical to map recombinants to determine whether
recombination was targeted to S region sequences. In the construct
pML22, the ampicillin resistance marker is directly upstream of Sµ,
and the bacterial replication origin is directly downstream of S3.
This biases recovery of molecules in which recombination breakpoints
are beyond the S regions. To map recombination junctions, we therefore
constructed another switch substrate, pML23, in which a 1-kb kanamycin
resistance gene was inserted just downstream of the S3 region. After
transfection of human cells with pML23, we selected either
ampicillin-resistant or kanamycin-resistant recombinants. The
ampicillin-resistant recombinants (AmpR) included
all those with acceptor endpoints downstream of the S3 region,
whereas the kanamycin-resistant recombinants
(KanR) included all those with donor endpoints
located upstream of the Sµ region. We compared recombination
frequencies of pML22 and pML23 to determine whether introduction of the
kanamycin cassette affected recombination levels. As shown in Fig. 2, recombination frequencies of the two
constructs were comparable, regardless of selection conditions.
|
PL region. Selection for ampicillin resistance
allowed all recombinants with acceptor junction endpoints in S3 and
in the Kan gene to be scored (Fig. 3A). Among
ampicillin-resistant recombinants recovered from transfection of BJAB,
50% of acceptor junctions mapped within S3. In contrast, among
ampicillin-resistant recombinants recovered from transfection
of BJAB-B1, 92% of acceptor junctions mapped within S3.
Similarly high levels of targeting of recombination within S3
were observed in recombinants recovered from transfection of both
EBV-transformed BLs, Akata and Raji. Selection for kanamycin resistance
allowed all recombinants with junction endpoints in Sµ and in the
Amp gene to be scored (Fig. 3B). Among
kanamycin-resistant recombinants recovered from transfection of
BJAB, 25% of donor recombination junctions were in Sµ; whereas in
recombinants recovered from BJAB-B1, 67% of donor junctions were
within Sµ. Targeting of both donor and acceptor endpoints in BJAB-B1
was similar to that observed for other EBV-positive lines, whereas
targeting of recombination in the fibroblast, HT1080, resembled that
observed in BJAB.
|
90% of
acceptor (or downstream) endpoints in the appropriate S regions.
Recombination within sequences flanking Sµ has also been reported in
the CH12F3 lymphoma, which switches from IgM to IgA production after
cytokine stimulation (28). Our mapping data show that
recombination of pML23 in BJAB-B1 and the EBV-positive BL lines, Akata
and Raji, produces junctions similar to those observed during
chromosomal switch recombination. Thus, the activities induced by EBV
appear to be those that carry out switch recombination in
vivo. Transcriptional activator enhances substrate recombination
Recombination of the extrachromosomal switch substrates in primary
murine B cells was shown to be dependent on the presence of
transcriptional regulatory regions (5). To ask whether a
transcriptional activator can enhance substrate recombination, we
assayed recombination of the substrate pML122, which carries the Ig
heavy chain intron enhancer and promoter sequences just upstream of
Sµ. Table II compares recombination of
pML122 and pML22 in transfections conducted in parallel. The
recombination frequency of pML122 was 2- to 3-fold higher than that of
pML22 in all lines examined (Table II). Moreover, recombination of
pML122 in Raji reached a level of R = 20%, comparable
with the very high level of substrate recombination evident in
transfection of primary B cells activated by LPS with the analogous
murine substrate, pHL122 (5).
|
3
region in recombinants recovered from transfections of BJAB, BJAB-B1,
and Raji (Fig. 4). The transcriptional
regulatory region thus enhanced targeting of recombination, as also
occurs in transfection of murine cells (6). Donor
junctions mapped both within EµPH and within
Sµ. A considerable fraction of chromosomal switch junctions produced
in vivo also lie upstream of the G-rich Sµ region (27, 28).
|
LR1 is a B cell-specific, sequence-specific DNA binding protein
complex which is induced in primary murine B cells activated for switch
recombination (16, 17, 18, 19). LR1 has multiple potential duplex
DNA binding sites in each of the G-rich switch regions. LR1 is a
heterodimer of two subunits, nucleolin and a specific isoform of
heterogeneous ribonucleoprotein D, and both subunits bind very tightly
to G-G paired structures which may form during transcription of the
G-rich S regions. This has led to a working model for switch
recombination in which LR1 may participate in synapsis of the two S
regions activated for recombination (reviewed in Ref. 29).
Because of the clear differences in the levels and targeting of
recombination in BJAB and its EBV-positive derivative, BJAB-B1, it was
of interest to determine whether LR1 DNA binding activity might
correlate with recombination. We therefore isolated nuclear extract
from Raji, Akata, BJAB-B1, BJAB, BJE2, and the fibroblast, HT1080, and
assayed LR1 DNA binding activity by gel mobility shift. These assays
compared binding to two different 32P-labeled
duplex synthetic oligonucleotides. One (S1) carried a wild-type LR1
binding site from the murine S1 switch region, and the other (G14A)
carried a single base mutation of that site which decreases binding of
LR1 about 15-fold (18). The nonspecific end-binding factor
Ku, which is present in all these extracts and binds to both duplexes,
provided an internal control on activity of the extracts. As shown in
Fig. 5, LR1 DNA binding activity (lower
band) was evident in all the EBV-positive cell lines assayed, including
Raji, Akata, and BJAB-B1. LR1 DNA binding activity was much reduced in
BJAB and its EBNA-2-transfected derivative, BJE2, and was completely
absent from the human fibroblast line HT1080. The absence of binding to
the G14A duplex (right) confirmed that the observed binding is due to
LR1. These data show that enhanced LR1 DNA binding activity is
correlated with the presence of EBV genome in the cells and that both
levels and targeting of switch substrate recombination correlate with
the levels of LR1 DNA binding activity.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Comparison of recombination in BJAB and its EBV-infected derivative, BJAB-B1, showed that EBV infection induces recombination activities. The EBV-infected lymphoma, BJAB-B1, supported 5-fold higher levels of recombination of the substrate pML22 than did its EBV-negative parent line, BJAB. Modification of the recombination cassette to include a strong transcriptional activator upstream of Sµ enhanced recombination in both cell lines but did not boost recombination in BJAB to the level observed in BJAB-B1. Recombination was highest in the two BL lines, particularly in Raji. In this group III EBV-transformed BL, recombination levels approached 20% of recovered plasmids, comparable with that observed in transfection of primary murine B cells activated by LPS.
Mapping of recombinants showed that the activities that promote switch
substrate recombination in EBV-infected cells are very similar or
identical to those that promote S region recombination in activated B
cells. The acceptor endpoints in recombinants were almost exclusively
within the S3 region, while a significant proportion of donor
endpoints mapped upstream of the Sµ region sequences, as also occurs
in chromosomal switch recombination (27, 28). A
considerably higher fraction of recombination junctions were targeted
within the S regions in recombinants recovered from the EBV-positive
lines.
The results reported here show that EBV infection can induce host activities essential to switch recombination. We have assayed for one such activity, LR1, a B cell-specific, sequence-specific duplex DNA binding protein with multiple sites in the G-rich S regions and at the c-myc locus (19). In the cell lines examined, switch substrate recombination correlated with levels of LR1 duplex DNA binding activity. LR1 binding activity is absent from resting B cells but is induced by EBV infection or activation of switch recombination. Because EBV infection induces LR1 activity, it is an intriguing possibility that the host factor LR1 plays some role in the EBV life cycle. This possibility is consistent with the fact that there are multiple potential binding site for LR1 in the EBV BamW repeats. In addition, the BamW repeats are frequently sites of EBV genomic reorganization during lytic replication, and just as translocations contribute to human disease, EBV genomic rearrangements have the potential to alter viral gene regulation or gene function (3, 4). EBV therefore appears to induce host activities that diminish stability of the viral genome and are also responsible for the c-myc translocations characteristic of sporadic BL.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Department of Hematology/Bone Marrow Transplantation, City of Hope National Medical Center, 1500 E. Duarte Road, Duarte, CA 91010.
3 Address correspondence and reprint requests to Dr. Nancy Maizels, Department of Molecular Biophysics and Biochemistry, Yale University School of Medicine, 333 Cedar Street, New Haven, CT 06520-8024. E-mail address:
4 Abbreviation used in this paper: BL, Burkitt’s lymphoma; EBNA, Epstein-Barr nuclear Ag; LMPs, latent membrane proteins.
Received for publication June 24, 1999. Accepted for publication October 7, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. L. Wang and M. Wabl DNA Acrobats of the Ig Class Switch J. Immunol., May 15, 2004; 172(10): 5815 - 5821. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. He, N. Raab-Traub, P. Casali, and A. Cerutti EBV-Encoded Latent Membrane Protein 1 Cooperates with BAFF/BLyS and APRIL to Induce T Cell-Independent Ig Heavy Chain Class Switching J. Immunol., November 15, 2003; 171(10): 5215 - 5224. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Sun, A. Yabuki, and N. Maizels A human nuclease specific for G4 DNA PNAS, October 23, 2001; 98(22): 12444 - 12449. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Hanakahi and N. Maizels Transcriptional activation by LR1 at the E{micro} enhancer and switch region sites Nucleic Acids Res., July 15, 2000; 28(14): 2651 - 2657. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
