| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
*
Division of Molecular Immunology, Department of Pathology,
Department of Microbiology and Immunology, Weill Medical College of Cornell University, and
The Immunology Program, Weill Graduate School of Medical Sciences of Cornell University, New York, NY 10021
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
J genes and
switch to IgG, IgA, and IgE upon B cell receptor engagement and contact
with CD4+ T cells through CD40:CD154 and CD80:CD28
coengagement. In this paper we showed that the same stimuli induce
somatic hypermutation of bcl-6 in CL-01 and normal
IgM+ IgD+ B cells. bcl-6
hypermutation was not accompanied by translocation of this
proto-oncogene or hypermutation of the ß-actin gene, and it did mimic
Ig hypermutation. It was associated with transcription initiation, in
that it targeted the first exon and a 696-bp sequence immediately
downstream (
0.6 kb) of the transcription initiation site while
sparing further downstream (2.5 kb) and upstream (0.1 kb) areas.
bcl-6 hypermutation displayed an overall rate of 2.2
x 10-4 changes/base/cell division with
characteristic nucleotide preferences and showed strand polarity. These
findings show that B cell receptor engagement promotes hypermutation in
genes other than Ig, and suggest that cis-regulating
elements similar to those of the Ig locus exist in
bcl-6. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ig somatic hypermutation has been shown to be dependent on the presence of the intronic Ig µ enhancer (iEµ) (10, 12, 13). Foreign sequences can be substituted for V(D)J and still function as targets for hypermutation, suggesting that cis-acting elements outside the mutable region regulate and target mutations (14, 15), as also further indicated by the high frequency of point mutations in c-myc and bcl-2 proto-oncogenes brought under the influence of the Ig promoter and iEµ as a result of chromosomal translocation (16, 17, 18, 19, 20). These findings, together with the demonstration that Ig V(D)J genes mutate when the Ig promoter is replaced with another promoter, imply the possibility that the hypermutation process may target non-Ig genes. Consistent with this is the recent suggestion that bcl-6, a proto-oncogene encoding a POZ/Zinc finger transcriptional repressor normally required for germinal center formation, can accumulate somatic point mutations not only in lymphoma B cells, where it is frequently translocated into the Ig locus, but also in normal B cells (21, 22, 23). Whether endogenous bcl-6 is targeted by the hypermutation machinery outside of the Ig gene locus (that is, it mutates without translocation) and whether bcl-6 hypermutation is induced by the same molecular trigger(s) of Ig hypermutation remains to be formally demonstrated.
We used CL-01 cells, our monoclonal model of germinal center B
lymphocyte differentiation, as well as normal human B cells to address
the issue as to whether the physiological induction of the Ig somatic
hypermutation machinery results in the targeting of DNA sequences
outside the Ig gene locus. As we have shown, CL-01 cells are surface
IgM+ IgD+ and can be
induced to effectively switch to IgG, IgA, and IgE, and mutate the
expressed VHDJH-Cµ,
VHDJH-C
,
VHDJH-C
,
VHDJH-C
,
VHDJH-C
, and
VJ-C transcripts before and after switching in a stepwise
fashion, upon BCR engagement and contact with activated T cells
(24, 25, 26, 27, 28). As we also have shown, in CL-01 cells, Ig
hypermutation and class switching are part of an overall program of
coordinated germinal center differentiation that includes phenotypic
changes, such as modulation of CD38 and CD77 expression, and increased
levels of cytoplasmic Ig leading to a plasmacytoid phenotype (24, 25). We show here that in both CL-01 cells and normal human B
cells the stimuli that induce Ig class switching, Ig hypermutation, and
plasmacytoid differentiation also induce high rate of mutations in
bcl-6. The same induction requirements, the same dependency
on transcription initiation and similar mutational modalities suggest
that certain cis-acting element(s), structurally or
functionally similar to the cis-acting element(s) in the Ig
locus, are present in the bcl-6 locus, and that these
elements are recruited by the same or vastly overlapping
trans-activating pathways by the Ig hypermutation
machinery.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The human CL-01 B cell line has been reported
(24, 25, 26, 27, 28). CL-01 cells are monoclonal, as shown by Southern
blotting with labeled JH probes, and by the
expression of unique
VHDJH-Cµ and
VHDJH-C transcripts.
They display on both chromosomes the switch (S)µ, 
, and S3,
S1, S1, S2, S4, S, and S2 regions in germline
configuration, and no VHDJH
rearrangement in the nonfunctional chromosome (24, 26).
CL-01 cells express the phenotype of germinal center founder
centroblasts, including surface IgM, IgD, CD38, and CD77
(24). These cells were cultured in RPMI 1640 medium (Life
Technologies, Gaithersburg, MD), supplemented with 10% heat
inactivated FCS, 2 mM L-glutamine, 100 U/ml penicillin, and
100 µg/ml streptomycin (FCS-RPMI) at less than
105/ml.
Normal human naive IgM+ IgD+ B cells
Normal human PBMCs were isolated from healthy subjects and depleted of T cells by sheep RBC rosetting, as described (24, 25, 26). T cell-depleted cells were first reacted with a biotinylated mouse mAb to human IgD (Southern Biotechnology Associates, Birmingham, AL), and then with magnetic MACS MicroBeads conjugated with streptavdin (Streptavidin MicroBeads; MACS Technology, Miltenyi Biotec, Auburn, CA). sIgD+ B cells were purified using a magnetic MiniMACS sorter (Miltenyi Biotec). FACS analysis demonstrated that virtually all these purified surface IgD+ B cells were also surface IgM+ (data not shown).
T cells
CD4+ T cells were positively selected from normal human PBMCs, prepared by fractionation through Histopaque 1077 (Sigma, St. Louis, MO) using anti-CD4 mAb-conjugated magnetic beads (Miltenyi Biotec). CD4+ T cells were cultured in FCS-RPMI and expanded by weekly stimulation with irradiated (1200 rads) PBMCs, 100 µg/ml of PHA (Life Technologies), and 100 U/ml of human rIL-2 (Life Technologies). CD4+ T cells were used at least 2 wk after the last stimulation, and were incubated for 6 h with 20 ng/ml of PMA (Sigma), and 500 ng/ml of ionomycin (Calbiochem-Novabiochem, San Diego, CA) before culture with B cells.
B:T cell cocultures
CL-01 or normal naive IgM+IgD+ B cells (0.5 x 106) were cultured in the presence or absence of 2.5 x 106 irradiated and activated CD4+ T cells in a flat-bottom 6-well (5.0 ml FCS-RPMI volume) plate, in the presence of human rIL-2 (Life Technologies) and rIL-4 (100 U/ml) (Genzyme, Cambridge, MA). To cross-link the BCR, CL-01 or normal B cells were reacted for 2 h at 4°C with Sepharose-conjugated rabbit Ab to human Ig µ chain and rabbit Ab to human Ig (H + L) chain (Irvine Scientific, Santa Ana, CA; mixed 1:1 at 2 µg/ml) (anti-BCR Ab), and then washed with cold PBS. After 7 days of culture, the B cells were collected, freed of dead cells and debris by fractionation through Histopaque 1077 (Sigma), reacted again with anti-BCR Ab, washed, and re-seeded over a fresh layer of irradiated T cells in the presence of fresh cytokines. At day 14 of culture, the B cells were collected for genomic DNA or total RNA extraction after removal of residual T cells by absorption with anti-CD3 mAb-conjugated magnetic beads (Miltenyi Biotec). The CD40:CD154 and CD80:CD28 interactions were blocked by preincubating T or B cells with saturating amounts (30 µg/ml) of mouse 24-31 mAb to human CD154 (Ancell, Bayport, MN), mouse CD28.2 mAb to human CD28 (PharMingen, San Diego, CA), or mouse BB1/B7-1 mAb to human CD80 (PharMingen). Mouse Ber-H2 mAb to human CD30 (Dako, Carpinteria, CA) was used as a "control" mAb.
PCR amplification of bcl-6 genomic DNA
Genomic DNA was extracted from 2 x
106 CL-01 B cells using the QIAamp Tissue kit
(Qiagen, Chatsworth, CA). Three genomic DNA areas located upstream,
immediately downstream, or further downstream of the transcription
initiation site (residue +1) of the bcl-6 gene were
amplified by PCR (Fig. 1
). The
bcl-6 DNA encompassing residues -61 through -400 bp
upstream of the transcription initiation site (area A) was amplified
with the sense primer Bcl6 up2S (5'-GCTTGAGGGATTAGATAGGGAAAG-3')
and the antisense primer Bcl6 up3AS
(5'-GTAGGGTTAGAGAGGCGAGTAAG-3'). The bcl-6 DNA
encompassing residues +587 through +1282 bp downstream of the
transcription initiation site (area B) was amplified using the sense
primer Bcl6down1S (5'-CTGCCGCCGCCGCTGCTCATGA-3') together with the
antisense primer Bcl6down2AS (5'-CCTTCCAAATCTCGGTTCGG-3'). The
bcl-6 DNA encompassing residues +2525 through +3037 bp
further downstream of the promoter region (area C) was amplified using
the sense primer Bcl6down3S (5'-TGACCACCTCAGAGAGCCAATCTC-3') and the
antisense primer Bcl6down4AS (5'-TCCAGGAATAGCGTGTGGACTAGG-3'). The
PCRs were performed using Pfu Turbo DNA polymerase
(Stratagene, La Jolla, CA) (the error rate of the Pfu Turbo
is about 1.3 x 10-6
changes/base/cycle) and the reaction buffer provided by the
manufacturer with 30 cycles, each cycle consisting of a 1-min
denaturation at 94°C, a 1-min annealing at 60°C, and a 1-min
extension at 72°C. The amplifications were completed by an additional
10-min extension at 72°C. The PCR products were purified with the PCR
purification kit (Qiagen) and ligated into the pCR-Script SK vector
(pCR-Script Cloning kit, Stratagene), which was used to transfect into
XL1-Blue MRF Supercompetent cells (Stratagene). The (positive) white
bacterial colonies were screened by PCR using the primers described
above. The individual colonies that had been directly used as PCR
templates were seeded onto a fresh Luria-Bertani plate and expanded
overnight. The clones containing bcl-6 DNA were selected for
single-strand conformational polymorphism (SSCP) analysis.
|
RNA was extracted from 2 x 106 CL-01
or normal B cells using the RNeasy Mini Kit (Qiagen). mRNA was reverse
transcribed using the SuperScript Preamplification System (Life
Technologies) for first-strand cDNA synthesis. The bcl-6
gene first exon sequence encompassing residues +23 through +278 bp
downstream of the initiation of transcription (Fig. 2) was amplified using cDNA as a template
and the specific sense Bcl6ex1S (5'-AACCTCCAAATCCGAGACGC-3') and
antisense Bcl6ex1AS (5'-CTGGTGTCCGGCCTTTCCTAG-3') primers. The
ß-actin sequence encompassing residues 745-1095 bp (GenBank accession
number: NM001101) was amplified using cDNA as a template and the
specific sense ß-actinF (5'-GCAAGAGATGGCCACGGCTGCTTC-3') and
antisense ß-actinB (5'-ATCCACACGGAGTACTTGCGCTCA-3') primers. DNA
amplification and cloning were performed as described above.
|
Mutated bcl-6 genomic DNA or bcl-6 and
ß-actin cDNAs were identified by SSCP analysis, performed essentially
as previously reported (27). DNA for SSCP analysis was
amplified by PCR (30 cycles of a 1-min denaturation at 94°C, a 1-min
annealing at 60°C, and a 1-min extension at 72°C) using genomic DNA
or cDNA cloned into pCR-Script SK vector as template, in a 10 µl
reaction volume with Taq DNA polymerase (Life Technologies)
in the presence of 1 µCi [-32P]dCTP (NEN
Life Sciences, Boston, MA) (3000 Ci/mmol). The same primers used to
amplify bcl-6 genomic DNA and cDNA were used for SSCP
analysis. We previously demonstrated that a single nucleotide change in
a 200- to 400-bp DNA sequence is sufficient to alter DNA mobility in
the SSCP gel (27). Therefore, bcl-6 DNA
fragment A (340 bp) and bcl-6 exon 1 cDNA (256 bp) were
analyzed as such. DNA fragments B (696 bp) and C (513 bp) were digested
with MspI before analysis to yield 321/375- and 285/228-bp
fragments. The DNA fragments were diluted 1:15 in 10 mM EDTA and
0.1% SDS. The diluted DNAs were mixed with an equal volume of
sequencing stop solution containing 95% formamide, 20 mM NaOH, 20 mM
EDTA, 0.05% bromophenol blue, and 0.05% xylene cyanole. The samples
were denatured for 10 min at 98°C, chilled on ice, and immediately
loaded in 3 µl aliquots onto a 6% acrylamide gel (20:1
acrylamide:bisacrylamide) with 1x TBE containing 10% glycerol.
Electrophoresis was performed at room temperature for 18 h at 5 W.
The gels were autoradiographed using Kodak X-Omat AR films (Kodak,
Rochester, NY).
DNA Sequencing and mutational analysis
The clones displaying an altered electrophoretic mobility in the SSCP gel were analyzed by sequencing to confirm and characterize the nature of the mutations. Plasmids were extracted using the QIAprep Spin Plasmid Miniprep Kit (Qiagen), and sequenced on both strands using Taq DiDeoxy Terminator Cycle Sequencing Kit and a 373 Automatic Sequencer (Applied Biosystems, Foster City, CA). To identify mutations, the sequence of genomic DNA areas A, B, and C, as well as bcl-6 exon 1 or ß-actin cDNA from induced CL-01 cells, were compared with the respective DNA and cDNA sequences from CL-01 cells cultured in FCS-RPMI alone (nonstimulated and noninduced), using the MacVector v.5.0 software (International Biotechnologies, New Haven, CT).
Because bcl-6 mutations were found to occur at a significant
frequency in Burkitt’s lymphomas (22), we first
determined the bcl-6 sequence in CL-01 cells by analyzing 45
independent genomic DNA clones of areas A, 45 of area B, and 45 of area
C. The 45 area A and 45 area C bcl-6 clones were all
identical in sequence and identical to the reported sequence of the
human bcl-6 gene (GenBank accession no. Z79581)
(29) (Fig. 1
). The 45 area B bcl-6 clones were
also all identical but differed from the reported sequence of the human
bcl-6 gene by three nucleotides residues and one
single-nucleotide deletion, suggesting that unstimulated CL-01 cells
carried only one copy of bcl-6, possible consisting of an
allelic variant or slightly mutated form of the reported
bcl-6 sequence.
The census of the somatic point mutations was performed by counting
only once identical mutations in more than one collinear DNA sequence.
It was assumed that identical base changes in collinear DNA sequences
were due to the same mutational events (shared mutations), although
some of these mutations might have arisen as a result of hotspots, and
therefore could actually represent independent mutations. The
comparisons of the observed with the expected number of mutations for
each individual nucleotide residue to each of the three other
nucleotides were performed using a contingency table
(
2 test). The expected frequency of mutations
was calculated by taking into account the base composition of the
unmutated CL-01 bcl-6 genomic DNA sequence; that is, it was
corrected by considering the frequency of occurrence of the individual
nucleotides, or dinucleotides, within the CL-01 bcl-6
sequence assuming randomness.
Analysis of bcl-6 rearrangement by Southern blotting hybridization
The 1.7-kb BCL-6-MBR probe (Fig. 3C) encompassing residues
-400 through +1282 bp within the major break-point region (MBR) of
bcl-6 gene was prepared by PCR amplification from human
placenta genomic DNA (Sigma) using the sense primer Bcl6 up2S and the
antisense primer Bcl6down2AS, and verified in sequence. Genomic DNA
from human placenta and induced or noninduced CL-01 cells was digested
with the restriction enzymes BamHI or XbaI,
loaded on a 0.8% agarose gel for electrophoresis, and then transferred
to Immobilon-Ny+ transfer membrane (Millipore,
Bedford, MA). The blotted DNA was hybridized with the
BCL-6-MBR probe, labeled with
[-32P]dCTP using the Random Primer DNA
labeling kit (Boehringer Mannheim, Indianapolis, IN), washed, and then
detected by autoradiography using Kodak X-Omat AR film.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To determine whether the same stimuli that induced Ig V(D)J gene
somatic hypermutation, could induce hypermutation in the
bcl-6 gene, CL-01 cells were reacted with anti-BCR Ab
and then cultured with activated human CD4+ T
cells, according to the same protocol originally devised for the
induction of Ig hypermutation (27). After 14 days, genomic
DNA was extracted from the cultured cells. Selected genomic
bcl-6 DNA sequences were amplified using specific primers
and then cloned into appropriate vectors for PCR amplification in the
presence of [-32P]dCTP, and further analysis
by our ad hoc devised SSCP, in a fashion similar to that
previously used for the analysis of V(D)J mutations (27).
As we have shown (27), one single base change in the
analyzed DNA is sufficient to yield a shift in the SSCP gel band.
Using this approach, we analyzed a 696-bp DNA sequence encompassing
residues +587 through +1282 bp downstream of the initiation site of
transcription (Fig. 1, area B) that has been reported to be mutated in
follicular lymphoma B cells (30).
SSCP analysis revealed that 33 of 116 (>28%) DNA clones from CL-01
cells cultured with activated CD4+ T cells upon
BCR engagement were mutated in the 696-bp DNA sequence of
bcl-6 area B. Sequences of 26 of these 33 DNA clones
revealed a total of 92 point mutations in addition to two 1-nt
deletions and a 1-nt insertion (Fig. 4).
The overall 92 point mutations possibly resulted from 54 independent
mutational events (Fig. 1B) and corresponded to a frequency
of mutation of about 1.5 x
10-3 changes/base, about
40-fold (p < 0.001) the frequency of mutations
expected in this 696-nt sequence after 30 cycles of PCR amplification
taking into account a Pfu Turbo error rate (about 1.3
x 10-6
changes/base/cycle). Only 1 of 46 (2.2%) of the corresponding DNAs was
found mutated in this region in CL-01 cells cultured in FCS-RPMI alone
or with activated T cells without BCR engagement. The single DNA with
altered mobility in SSCP bore one single point mutation, yielding an
overall mutation frequency of 3.1 x
10-5 changes/base (Fig. 1B). Also, only 1 bcl-6 DNA area B fragment out
of 30 analyzed (3.3%) was found to be mutated in CL-01 cells cultured
with anti-BCR Ab in the absence of activated T cells (Fig. 1B), yielding an overall mutation frequency of 4.8 x
10-5 changes/base, which
was not significantly different from the overall mutation frequency
expected on the basis of the Pfu Turbo error rate. Thus, BCR
engagement and activated CD4+ T cells effectively
induce somatic hypermutation in human B cell bcl-6.
|
Our previous findings have determined that coengagement of the
CD40:CD154 and CD80:CD28 stimulatory pairs is necessary, in addition to
BCR engagement, for the induction of somatic Ig hypermutation in human
B cells (27). To explore the role of CD40 and CD80 in the
induction of bcl-6 hypermutation, CD154 and CD28 on the
surface of activated T cells were blocked using mouse mAbs to human
CD154 or CD28, respectively, in a 14 day T:B cell culture. To verify
that hypermutation of the bcl-6 intronic DNA area B
accompanied somatic hypermutation of the immediately adjacent exon(s),
as it is characteristic of the Ig locus, we chose to analyze the 256-bp
noncoding exon 1 mRNA sequence (encompassing residues +23 through +278
bp downstream of the transcription initiation site) (Fig. 2). Six of 47
(12.8%) of the 256-bp exon 1 transcripts were found to be mutated in
CL-01 cells cultured with activated T cells upon BCR engagement. These
6 mutated transcripts bore a total of 9 independent point mutations,
which corresponded to a frequency of 7.5 x
10-4 changes/base, and a
rate of about 1.9 x
10-4 changes/base/cell
division (Fig. 2). Blocking CD154 or CD28 on the surface of activated T
cells made these cells ineffective inducers of hypermutation (30
bcl-6 exon 1 cDNA clones from the anti-CD154 culture and
30 bcl-6 exon 1 cDNA clones from the anti-CD28 culture
analyzed) (Fig. 2). That CD28 is required to engage CD80 (the CD28
complement on the B cell surface) to induce bcl-6
hypermutation was further indicated by the lack of mutations (43
bcl-6 exon 1 cDNA clones analyzed) in CL-01 cells
cultured with activated CD4+ T cells in the
presence of blocking mouse anti-CD28 mAb. The control anti-CD30
mAb failed to interfere the induction of the mutational machinery
(data not shown), indicating that anti-CD154, anti-CD28, and
anti-CD80 mAb ablated Ig hypermutation not by merely reducing or
abrogating B:T cell contact, but rather by specifically interfering
with CD40:CD154 or CD80:CD28 coengagement and related signaling. These
experiments showed that CD40:CD154 and CD80:CD28 coengagement is
crucial for induction of bcl-6 hypermutation and that
bcl-6 hypermutation is not limited to the 5' region of the
first intron but also targets the exon 1 sequence.
bcl-6 gene somatic hypermutation is associated with initiation of transcription and is not accompanied by hypermutation of ß-actin gene transcripts
Having shown the induction of bcl-6 hypermutation
mimics in requirements Ig somatic hypermutation, we addressed the issue
as to whether the bcl-6 hypermutation process is also, like
Ig hypermutation (3, 15, 31, 32, 33), associated with
initiation of transcription. To this end, we analyzed a 340-bp sequence
encompassing residues -61 through -400 bp upstream of the
transcription initiation site (Fig. 1, area A) and a 513-bp DNA
sequence encompassing residues +2525 through +3037 bp further
downstream of the transcription initiation site (Fig. 1, area C). We
reasoned that if bcl-6 hypermutation was associated with
initiation of transcription in the same fashion as the Ig locus, it
would target exon 1 and area B only, while sparing both areas A and C.
These two additional areas were chosen to reflect the topography of Ig
V(D)J hypermutation, which begins downstream of the gene transcription
initiation site and continues through a 1- to 2-kb region
(34, 35, 36, 37).
In contrast to what found in bcl-6 intronic area B DNA and
the exon 1 mRNA, no mutations were found in the area A upstream of
the bcl-6 gene transcription initiation site in CL-01
cells cultured with anti-BCR Ab in the presence or absence of
activated CD4+ T cells (30 and 24 DNA fragments
analyzed, respectively). Only one mutated DNA clone was found in this
region in CL-01 cells cultured in FCS-RPMI alone or in the presence of
activated CD4+ T cells but not anti-BCR Ab
(50 and 30 DNA fragments analyzed, respectively). Both mutated segments
carried one point mutation each to yield overall mutation frequencies
of 6.8 x 10-5 and
11.3 x 10-5
changes/base, respectively (Fig. 1), which were not significantly
different from the mutation frequency expected on the basis of the
Pfu Turbo error rate alone (about 4.0 x
10-5 changes/base). Only
one mutated DNA segment was found in the bcl-6 512-bp area C
region in CL-01 cells cultured upon BCR engagement and in the presence
of CD4+ T cells (50 DNA clones analyzed) or upon
BCR engagement and in the absence of CD4+ T cells
(36 DNA clones analyzed). No mutated DNA segments were found in the
analysis of the bcl-6 512-bp area C in CL-01 cells cultured
in FCS-RPMI alone or FCS-RPMI upon BCR-engagement but in the absence of
activated CD4+ T cells (24 and 29 DNA clones
analyzed, respectively).
That the hypermutation in Ig V(D)J and bcl-6 genes is not a general feature of all transcribed genes in CL-01 cells was revealed by the insignificant mutation frequency at the ß-actin locus, as judged by SSCP and cDNA sequencing analyses. Only 1 of 60 ß-actin cDNA clones derived from the CL-01 cells stimulated with activated CD4+ T cells upon BCR engagement displayed an altered mobility in SSCP (not shown). This shifted clone carried only one point mutation, yielding an overall frequency of mutations of about 5.0 x 10-5 changes/base, a frequency not significantly different from that expected (about 4.0 x 10-5 changes/base) on the basis of the Pfu Turbo error rate.
Point mutations in bcl-6 gene show a bias for transitions over transversions, G and C preference, and strand polarity
The 92-nt changes found in the bcl-6 area B DNA
sequence comprised 54 independent mutations, distributed over 51
positions throughout the whole region. They yielded an overall
independent point-mutation frequency of 8.6 x
10-4 changes/base, i.e.,
2.2 x 10-4
changes/base/cell division (Fig. 1), and showed a bias for
transitions over transversions. Randomly occurring point mutations
would be expected to be one-third transitions and two-thirds
transversions, but almost 60% of the 54 independent point mutations in
the 696-bp segment analyzed were transitions. G nucleotides were
mutated at a frequency (37.0% of total point mutations), almost 50%
higher than expected by chance alone (24.9%) after correcting for base
composition. G
A transitions accounted for 65% of the total G
mutations and 40.6% of all transitions (Table I). The preferential targeting of G by
mutations was associated with a scarcity of mutations in A
(p < 0.001) and T (p
< 0.001) (27). A preferential targeting of C nucleotides
was also observed in bcl-6 mutations, 33.3% of total point
mutations targeted C; that is, almost 30% higher than expected by
chance alone (25.9%), with CT transitions accounting for 61.1% of
the total C mutations. The proportion of T mutations was about 50%
higher than A, after correction for the base composition of the
bcl-6 gene sequence, suggesting a strand polarity in the
bcl-6 hypermutation, as further emphasized by the finding
that G bases were more frequently targeted mutations than C bases
(Table I).
|
AG dinucleotides and GAG trinucleotides were found to be
preferentially targeted by mutations in the Ig
VHDJH gene segments of
induced CL-01 cells (27). To verify whether a similar
preference was displayed by somatic hypermutation in bcl-6,
we considered the 16 possible dinucleotides identified in the 696-bp
sequence of bcl-6 gene area B and made a census of all the
point mutations targeting these dinucleotides (Table II). The AG and CG dinucleotides were
mutated at a frequency significantly higher than that expected by
chance alone (p < 0.001). This preference was
not merely due to G and C biases, as G and C were mutated at a
significantly lower frequency when occurring outside than within AG and
CG (45% of total G mutations targeted AG dinucleotides, although the
frequency of occurrences of the G nucleotides within the AG
dinucleotides is only 17.4%, and more than 30% of total C mutations
targeted the CG dinucleotides, although only 19.6% of total C
nucleotides are within CG).
|
The consensus sequence RGYW (where R = purine, Y = pyrimidine, and W = A or T) has been identified as a mutational hotspot in human and murine Ig genes (12, 38, 39), and in the bcl-6 gene of normal human germinal center B cells or in diffuse large cell lymphoma (22). In the bcl-6 gene area B 696-bp sequence, the RGYW motif occurred 14 times in six different iterations, which covered about 6.3% of the sequence. Nine point mutations were located in this motif, which corresponded to 16.4% of all independent mutations, a frequency more than 2.5-fold higher than expected by chance alone (p < 0.05).
bcl-6 somatic hypermutation is not associated with bcl-6 gene translocation
In human B cell lymphomas, bcl-6 mutations have been
associated with translocation of this proto-oncogene into the Ig locus.
In general, these translocations affect the band 3q27 and lead to
bcl-6 activation by bringing bcl-6 under the
effects of Ig promoters and other cis-acting elements of the
locus (30, 40). In bcl-6, these translocations
are associated with a MBR, as determined by Southern blot analysis
(41). Our previous findings showed that CL-01 cells bear
Ig switch regions in germline configuration on both chromosomes and
carry no translocated DNA sequence in the region between
JH and Sµ (24). Accordingly,
Southern blot hybridization analysis performed on genomic DNA digested
with the restriction enzymes BamHI or XbaI using
a 1.7-kb genomic probe within the MBR showed no evidence of
rearrangement of bcl-6 in a significant proportion of CL-01
cells before (Fig. 3A) or after (Fig. 3B)
induction of somatic hypermutation. The lack of bcl-6
translocation was further supported by a long range PCR assay spanning
2409 residues between JH and Sµ, and providing
no evidence of translocated DNA in the region between
JH and Sµ in both induced and uninduced CL-01
cells (data not shown).
BCR engagement and T cell contact induce bcl-6 hypermutation in normal human naive IgM+ IgD+ B cells
We extended our findings to normal human B cells by analyzing the
256-bp bcl-6 noncoding exon 1 mRNA sequence in freshly
isolated naïve IgM+
IgD+ B cells that were cultured in FCS-RPMI alone
or that were reacted with anti-BCR Ab and then cultured in FCS-RPMI
in the presence of activated human CD4+ T cells,
according to the same protocol used for the induction of
bcl-6 (present experiments) and Ig hypermutation
(27). SSCP analysis of the 256-bp bcl-6
noncoding exon 1 revealed that none of the 40 cDNA clones from normal B
cells cultured in FCS-RPMI alone without prior BCR engagement was
mutated (Fig. 5A). In
contrast, 7 of 40 (18%) cDNA clones from the B cells cultured with
activated CD4+ T cells upon BCR engagement were
mutated; that is, they displayed an altered gel mobility (Fig. 5B). Sequencing of these 7 cDNA clones displaying an altered
SSCP mobility demonstrated that each of them contained one point
mutation (Fig. 5C), which yielded an overall mutation
frequency of 6.8 x
10-4 changes/base.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
),
including: 1) association with initiation of transcription, 2) DNA
strand polarity, 3) preference for single nucleotide substitutions, 4)
bias for transitions over transversions, and 5) preferential targeting
of selected nucleotide residues and hot spots. They establish that
targeting of bcl-6 by the hypermutation machinery occurs in
the absence of bcl-6 translocation into the Ig locus,
indicating that the bcl-6 locus possesses
cis-regulatory elements that are similar to those of the Ig
locus and that may respond to the same trans-acting
factors.
|
Our results imply that the hypermutation of the 5'-regulatory region of bcl-6 is part of an overall germinal center differentiation program that also includes Ig hypermutation, class switching, and characteristic phenotypic changes, as we have shown in human IgM+ IgD+ CL-01 cells and freshly isolated IgM+ IgD+ B cells (24, 25, 26, 27, 28). bcl-6 encodes a POZ/Zinc transcriptional repressor that negatively regulates the STAT-6-dependent production of IL-4 by CD4+ T cells, as well as the expression of STAT-6-responsive genes in B cells (44, 45). As preliminary results suggest that certain mutations deregulate bcl-6 gene expression and its transcriptional regulatory activity (22), it is tempting to speculate that, under certain conditions, bcl-6 hypermutation contributes to the physiological regulation of IL-4:IL-4R:STAT-6-induced B cell responses, including class switching to IgG and IgE (28), as occurring during a germinal center reaction.
The CD28 activation pathway has been reported to be essential for the Ab response to T cell-dependent Ags in the mouse (46). Our previous results have determined that, in addition to BCR engagement, engagement of the CD40:CD154 and CD80:CD28 costimulatory pairs is necessary for the induction of Ig somatic hypermutation in human B cells (27). In this paper we addressed the role of these costimulatory molecules in the induction of bcl-6 somatic hypermutation, and provided evidence that B:T cell contact allowing for CD40:CD154 and CD80:CD28 coengagement is necessary, in addition to BCR engagement, to induce this process. Whether, after BCR engagement, CD40:CD154 and CD80:CD28 coengagement is sufficient to induce bcl-6 (and Ig) hypermutation or signals emanating from the coengagement of these costimulatory pairs mediate the induction of crucial T cell surface and/or soluble molecules and/or factors remains to be determined.
bcl-6 gene mutation had been observed in B cell lymphomas of the germinal center/postgerminal center type (47). In some tumors with bcl-6 mutations the bcl-6 gene is translocated into the Ig locus. In this case, the translocated bcl-6 gene has come under the control of the cis-acting elements of the Ig gene and a becomes natural target by the Ig hypermutation, as the c-myc and bcl-2 genes translocated into the Ig locus in Burkitt’s lymphomas do (18). The germline configuration of bcl-6 in CL-01 cells and normal naive IgD+ B cells indicates that the bcl-6 gene is regulated by cis-acting elements structurally and functionally similar to those of the Ig locus, such as the V promoter and/or the Ig iEµ present in the locus of this proto-oncogene. As shown by experiments, such bcl-6 cis-acting elements would respond to the same BCR and T cell signaling-dependent trans-acting factors that induce hypermutation on the Ig locus, thereby suggesting an additional new approach to the investigation of the mechanisms by which hypermutation targets Ig loci.
In germinal center B cells, the frequency of bcl-6 mutations was found to be 20- to 100-fold lower than that of the Ig V(D)J gene segment (21, 23), suggesting a relative inefficiency in the somatic hypermutation process of bcl-6. In our Ag-selection-free in vitro system, the frequency of mutations in bcl-6 was comparable to that of Ig V(D)J (2.2 x 10-4 vs 2.4 x 10-4 changes/base/cell division) in the same cells, suggesting that the somatic hypermutation process targets bcl-6 as efficiently as Ig V(D)J. This lends further support to the hypothesis of structural and functional overlap of the bcl-6 cis-acting elements with the corresponding elements in the Ig gene locus and their responsiveness to the same BCR and/or T cell surface receptor-emanating signaling pathways.
The hypermutation process targets rearranged V(D)J genes by an unknown
mechanism in germinal center B cells where it introduces point
mutations. To a lesser extent, it introduces insertions and deletions
into a 1.5-kb region immediately downstream of the transcriptional
initiation site with sparing of the area upstream of the leader
sequence, thereby displaying a strong association with transcription
initiation. Our findings demonstrate that the same features apply to
bcl-6 hypermutation. The published reports on in vivo
bcl-6 mutation have been limited to the analysis of the 5'
end sequence of the first intron (21, 22, 23). We showed here
that bcl-6 hypermutation targets in addition to the 5' end
of the first intron, the first exon of the bcl-6 gene, while
sparing the areas upstream or further downstream of the promoter. In
our experiments, the dependence of bcl-6 hypermutation on
initiation of transcription is further emphasized by the strand
polarity, as suggested by the G>C and T>A bias of bcl-6
mutations.
Both in vivo and in vitro studies have shown that Ig somatic
mutations are not inserted randomly, and certain DNA sequences and/or
structures have been recognized as possible hotspots for hypermutation
(27, 48, 49, 50, 51, 52). The identification of sequence-specific
preferences by the somatic hypermutation machinery is important as it
may provide clues in determining the molecular mechanism of this
process. Like Ig mutations in induced CL-01 cells (27) and
other human lymphoma cell lines (52, 53), bcl-6
gene mutations preferentially targeted G and C nucleotides. More than
70% of all nucleotide substitutions in the CL-01 bcl-6 were
of G/C residues (Fig. 4, Table I) compared with corresponding figures
of about 50% for bcl-6 mutations found in the human in vivo
(21, 22). Some of the target biases may in fact reflect
repair biases in the hypermutation process. The similar pattern of
bcl-6 and Ig mutation in CL-01 cells indicates that repair
in bcl-6 hypermutation resembles that in Ig hypermutation.
Preferential targeting of G/C bases has been observed in exotherms
(54) and in Chinese hamster cells (55),
suggesting that the presence of a germinal center microenvironment
might be crucial to recruit a non-G/C-biased stage of mutation
generation or the activity of certain trans-acting factor(s)
which act(s) to preferentially revert mismatches mainly at G and C in
both Ig and bcl-6 genes. The G/C biased pattern of
hypermutation is displayed by Msh2- and Pms2-deficient mismatch repair
protein-deficient mice (56, 57, 58, 59), indicating that the
mutations that preferentially target G/C nucleotides of
bcl-6 and Ig genes could likely result from the lack of a
component of the mismatch repair system in CL-01 cells.
In human B cells, bcl-6 hypermutation is likely part of an overall regulatory germinal center differentiation program, in which bcl-6 acts to prevent apoptotic cell death, and may contribute to modulating the withdrawal from the cell cycle (60). Hypermutation of the 5' putatively regulatory region of the bcl-6 gene may lead to alterations of crucial steps of the B lymphocyte cell cycle and lymphomagenesis (61). As emphasized by our findings, significant differences exist between the human and the mouse B cell hypermutation machinery. In contrast with human bcl-6, mouse bcl-6 has been shown not to be the target of hypermutation (12), indicating that different cis-acting elements are associated with the human and mouse bcl-6 genes and/or that different levels of transcription activity exist in the B cells of the two spices. Understanding these differences will possibly further our understanding of the mechanisms of hypermutation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Paolo Casali, Division of Molecular Immunology, Department of Pathology, Weill Medical College of Cornell University, 1300 York Avenue, New York, NY 10021.
3 Abbreviations used in this paper: BCR, B cell Ag receptor; MBR, major break-point region; iEµ, intronic Ig µ enhancer; SSCP, single-strand conformational polymorphism.
Received for publication August 30, 1999. Accepted for publication April 18, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
gene utilization, junctional VJ and VJ diversity, and somatic hypermutation. J. Immunol. 1612895..
transgenes show clonal recruitment of hypermutation: a role for both MAR and the enhancers. EMBO J. 16:3987.[Medline]
S and sequential SµS, SS DNA recombination. J. Immunol. 161:5217.3 sequence upstream of the human Ig H chain S3 region is a functional promoter: synergistic activation by CD40L and IL-4 via cooperative NF-B and STAT-6 binding sites. J. Immunol. 162:5327. gene is in the leader intron. Eur. J. Immunol. 24:1453.[Medline]
This article has been cited by other articles:
|
|
 |
|
  Y.-W. Chen, X.-T. Hu, A. C. Liang, W.-Y. Au, C.-C. So, M. L. Wong, L. Shen, Q. Tao, K.-M. Chu, Y.-L. Kwong, et al. High BCL6 expression predicts better prognosis, independent of BCL6 translocation status, translocation partner, or BCL6-deregulating mutations, in gastric lymphoma Blood, October 1, 2006; 108(7): 2373 - 2383. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Yang, K. Lee, J. Said, X. Gong, and K. Zhang Association of Ig/BCL6 translocations with germinal center B lymphocytes in human lymphoid tissues: implications for malignant transformation Blood, September 15, 2006; 108(6): 2006 - 2012. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Kurosu, M. D. Weiden, Y. Takiguchi, W. N. Rom, N. Yumoto, J. Jaishree, K. Nakata, Y. Kasahara, N. Tanabe, K. Tatsumi, et al. BCL-6 Mutations in Pulmonary Lymphoproliferative Disorders: Demonstration of an Aberrant Immunological Reaction in HIV-Related Lymphoid Interstitial Pneumonia J. Immunol., June 1, 2004; 172(11): 7116 - 7122. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Jardin, C. Bastard, N. Contentin, F. Parmentier, J.-M. Picquenot, H. Tilly, F. K. Stevenson, and S. S. Sahota Intronic BCL-6 mutations are preferentially targeted to the translocated allele in t(3;14)(q27;q32) non-Hodgkin B-cell lymphoma Blood, September 1, 2003; 102(5): 1872 - 1876. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Schaffer, E. C. Kim, X. Wu, H. Zan, L. Testoni, S. Salamon, A. Cerutti, and P. Casali Selective Inhibition of Class Switching to IgG and IgE by Recruitment of the HoxC4 and Oct-1 Homeodomain Proteins and Ku70/Ku86 to Newly Identified ATTT cis-Elements J. Biol. Chem., June 13, 2003; 278(25): 23141 - 23150. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Wang, Z. Li, A. Naganuma, and B. H. Ye Negative autoregulation of BCL-6 is bypassed by genetic alterations in diffuse large B cell lymphomas PNAS, November 12, 2002; 99(23): 15018 - 15023. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Martin and M. D. Scharff Somatic hypermutation of the AID transgene in B and non-B cells PNAS, September 17, 2002; 99(19): 12304 - 12308. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Gurrieri, P. McGuire, H. Zan, X.-J. Yan, A. Cerutti, E. Albesiano, S. L. Allen, V. Vinciguerra, K. R. Rai, M. Ferrarini, et al. Chronic Lymphocytic Leukemia B Cells Can Undergo Somatic Hypermutation and Intraclonal Immunoglobulin VHDJH Gene Diversification J. Exp. Med., September 2, 2002; 196(5): 629 - 639. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Zhang, P. D. Bardwell, C. J. Woo, V. Poltoratsky, M. D. Scharff, and A. Martin Clonal instability of V region hypermutation in the Ramos Burkitt's lymphoma cell line Int. Immunol., September 1, 2001; 13(9): 1175 - 1184. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Cerutti, A. Schaffer, R. G. Goodwin, S. Shah, H. Zan, S. Ely, and P. Casali Engagement of CD153 (CD30 Ligand) by CD30+ T Cells Inhibits Class Switch DNA Recombination and Antibody Production in Human IgD+ IgM+ B Cells J. Immunol., July 15, 2000; 165(2): 786 - 794. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
