ã
Oncogene (2000) 19, 3404 ± 3410
2000 Macmillan Publishers Ltd All rights reserved 0950 ± 9232/00 $15.00
www.nature.com/onc
The c-MYC allele that is translocated into the IgH locus undergoes
constitutive hypermutation in a Burkitt's lymphoma line
Mats Bemark*,1 and Michael S Neuberger1
1
MRC Laboratory of Molecular Biology, Hills Road, Cambridge, UK
Burkitt's lymphomas harbour chromosomal translocations bringing c-MYC into the vicinity of one of the
immunoglobulin gene loci. Point mutations have been
described within c-MYC in several Burkitt's lymphomas
and it has been proposed that translocation into the Ig
loci might have transformed c-MYC into a substrate for
the antibody hypermutation mechanism. Here we test
this hypothesis by exploiting a Burkitt's lymphoma line
(Ramos) that we have previously shown to hypermutate
its immunoglobulin genes constitutively. We ®nd that,
during in vitro culture, Ramos mutates the c-MYC allele
that is translocated into the IgH locus whilst leaving the
untranslocated c-MYC and other control genes essentially una€ected. The mutations are introduced downstream of the c-MYC transcription start with the pattern
of substitutions being characteristic of the antibody
hypermutation mechanism; the mutation frequency is 2 ±
3-fold lower than for the endogenous functional IgH
allele. Thus chromosomal translocations involving the Ig
loci may not only contribute to transformation by
deregulating oncogene expression but could also act by
potentiating subsequent oncogene hypermutation. Oncogene (2000) 19, 3404 ± 3410.
Keywords: somatic
genes; c-MYC
hypermutation;
immunoglobulin
Introduction
Burkitt's lymphomas (BLs) are characterized by
chromosomal translocations [t(8;14), t(8;2) or t(8;22)]
that bring c-MYC into proximity with the immunoglobulin heavy or light chain loci (Dalla-Favera et al.,
1982; Taub et al., 1982; Hamlyn and Rabbitts, 1983).
These translocations appear to lead to a dysregulation
of c-MYC expression (for review see Johnston and
Carroll, 1992). In addition, it was early noted that the
translocated c-MYC allele in several Burkitt's lymphoma lines had incorporated multiple mutations with
these mutations being largely focused within the ®rst
(non-translated) exon and ®rst intron of c-MYC
although mutations also extended into exon 2 (Rabbitts et al., 1983b, 1984; Taub et al., 1984). Two, nonmutually exclusive models have been forward to
account for these mutations. Either the intrinsic
mutation frequency of the translocated c-MYC is no
higher than that of other expressed genes in the cell:
rare c-MYC mutations could simply have been selected
by virtue of a conferred growth advantage. Alterna-
tively, translocation of c-MYC into the immunoglobulin loci could have transformed c-MYC into becoming
a substrate for the antibody hypermutation mechanism
and this is postulated to have been active at some stage
following the creation of the translocation (Rabbitts et
al., 1984; Taub et al., 1984; Cesarman et al., 1987).
The hypermutation of immunoglobulin V genes is a
process that is largely restricted to germinal centre B cells
and plays a central role in the anity maturation of
antibodies (reviewed in Wagner and Neuberger, 1996;
Parham, 1998). Although the molecular mechanism of
the hypermutation remains unidenti®ed, its target
requirements have been extensively studied. Thus, the
mutations (which are largely single nucleotide substitutions) are introduced into roughly 2 Kb regions located
downstream of the transcription start sites of the
productively rearranged immunoglobulin heavy and
light chain genes (Lebecque and Gearhart, 1990). The
V gene, whilst the physiological target of the mutation, is
not needed for mutation recruitment: other sequences
can be mutated in its place (Yelamos et al., 1995).
Mutation recruitment depends upon the immunoglobulin transcription regulatory elements (Betz et al., 1994)
with the location of the mutation domain being de®ned
by the position of the transcription start site (Peters and
Storb, 1996; Rada et al., 1997; Tumas-Brundage and
Manser, 1997). Whilst the immunoglobulin loci are
certainly the preferred targets of the hypermutation, the
process is not wholly exclusive to them: BCL-6 (but not
several other genes analysed) has been found to be a
natural target for the hypermutation process ± albeit at
a much reduced mutation rate (Migliazza et al., 1995;
Pasqualucci et al., 1998; Shen et al., 1998).
In a search for an in vitro model in which to study
immunoglobulin gene hypermutation, we recently
identi®ed that the Ramos Burkitt's lymphoma cell line
constitutively mutates its rearranged immunoglobulin
V genes at high rate during in vitro culture (Sale and
Neuberger, 1998). This mutation shows the classic
hallmarks of physiological immunoglobulin gene
hypermutation. Since Ramos harbours a t(8;14)
translocation in which c-MYC has been brought into
the IgH locus (Figure 1a), this cell line provides us with
the means to ascertain whether the translocation itself
does indeed convert c-MYC into becoming a substrate
of the antibody hypermutation mechanism.
Results
Only the translocated c-MYC in Ramos is expressed
*Correspondence: M Bemark
Received 13 April 2000; revised 15 May 2000; accepted 16 May
2000
Ramos contains two c-MYC alleles: one that is
translocated into the IgH locus, the other being wild
type (Figure 1a). Although the chromosome 8 trans-
c-MYC hypermutation in Burkitt's lymphoma
M Bemark and MS Neuberger
3405
Figure 1 Sequence diversity in the translocated c-MYC allele Ramos culture (a) Schematic representation of the Ramos IgH, the cMYC gene and the t(8;14) c-MYC translocation; the data are taken from (Bernard et al., 1983; Wiman et al., 1984). (b) Analysis of
expressed c-MYC in Ramos by RT ± PCR. The sequence diversity amongst 19 c-MYC RT ± PCR products is illustrated by Pie
charts. The digit in the centre of the Pie indicates the number of sequences that were obtained from each of the c-MYC alleles and
the digits ¯anking each segment of the Pie indicate the number of mutations that were found in 250 base pairs around the exon 1/
exon 2 border (139 bp in exon 1, 111 in exon 2). The two c-MYC alleles could be discriminated by virtue of the A5234T mutation
on the translocated allele (Wiman et al., 1984). (c) Analysis of sequence diversity of the two Ramos c-MYC alleles by genomic PCR.
Separate PCR reactions were used to amplify the two c-MYC alleles; the locations of the primers used (arrows) and of the regions
sequenced (thin lines) are shown in the left-hand side of this panel. To the right, Pie charts depict the distribution of mutated
sequences. (d) Analysis of sequence diversity in BCL-6 and the ribosomal S14 gene
location breakpoint lies upstream of the c-MYC
promoter, the two c-MYC alleles can be distinguished
within the transcribed region since the translocated
allele has an A5234T mutation in exon 1 (Wiman et
al., 1984). This mutation allowed us to ask whether
both c-MYC alleles are transcribed in Ramos. They
are not. Of 19 cloned c-MYC cDNA RT ± PCR
products, all derived from the translocated allele
(Figure 1b).
The translocated c-MYC has accumulated mutations
Comparison of the 19 c-MYC RT ± PCR products also
revealed sequence heterogeneity. In addition to the
A5234T mutation common to all the sequences, ®ve
of the PCR products carried further mutations (Figure
1b). This frequency of mutation (16.8 bp71 1074) is far
in excess of the background error rate of the RT ± PCR
technique itself (50.4 bp71 1074 in Cm (Sale and
Neuberger, 1998)) and presumably re¯ects the fact
that Ramos has diversi®ed its translocated c-MYC
during culture.
To de®nitively exclude any contribution by possible
reverse transcriptase-induced artefacts, we sequenced cMYC products that had been ampli®ed from genomic
DNA. The ampli®ed regions extended from upstream
of the promoter (using di€erent primers to amplify the
wild type and translocated alleles separately) to the
intron downstream of exon 1 (Figure 1c). With regard
to the translocated allele, about one third of the 79
ampli®ed sequences were mutated with between 1 and
4 mutations per sequence. In total, 25 distinct
mutations were identi®ed on this allele, three of which
were insertions of deletions (Table 1). These results are
most unlikely to re¯ect PCR artefact. Not only is the
frequency of mutation considerably in excess of that
obtained in the PCR control (Table 1) but also two of
the mutations were detected in independent PCR
ampli®cations (G4894C and C5034G). Moreover,
three of the mutations identi®ed in the genomic PCR
had also been identi®ed amongst the RT ± PCR
products (C4274T, G4894C and C5034G).
In contrast to these observations with regard to the
translocated c-MYC, only four of the 70 sequences
Oncogene
c-MYC hypermutation in Burkitt's lymphoma
M Bemark and MS Neuberger
3406
Table 1 The prevalence of mutations in a Ramos population
Gene
Translocated c-MYC
Exon 1
Upstream of promoter
Nontranslocated c-MYC
Exon 1
Upstream of promoter
BCL-6
Exon 1 and Intron 1
Upstream of promoter
S14
Intron 1
PCR control
(c-MYC exon 1)
Mutation prevalence
(bp7161074)
Mutations
detected
Deletions
insertions
Sequences
analysed
Base pairs
analysed
7.0
0
25
0
3
0
79
46
35500
14260
1.3
0
4
0
0
0
70
29
31500
11890
0.8
0
2
0
0
0
61
32
24400
12800
0.8
2
0
30+28
24600
0
0
0
35
15750
deriving from exon 1 of the untranslocated c-MYC
were mutated and, even then, there was a maximum of
one mutation per sequence. This degree of mutation
does not rise signi®cantly above the background
attributable to PCR error alone (Figure 1). Thus,
mutation of the Ramos c-MYC appears to be peculiar
to the translocated allele. Mutation is also peculiar to
the transcribed portion of the translocated c-MYC
since mutations were not detected in the region
immediately 5' of the c-MYC promoter on either allele
(Table 1 and Figure 1c).
We were interested in comparing the mutation
accumulation in c-MYC with that in two other genes:
BCL-6, which has earlier been shown to undergo
somatic hypermutation at a rate of about 1% that of
the immunoglobulin genes, and the ribosomal S14 gene
which has been found to be expressed but not mutated
in primary B cells (Pasqualucci et al., 1998; Shen et al.,
1998). The frequency of mutations detected at both
these loci in Ramos is not signi®cantly above the PCR
error rate.
The translocated c-MYC mutates constitutively in culture
These results revealed that Ramos had, at some stage,
accumulated mutations in its translocated c-MYC.
Considering the extended period over which our
Ramos culture had been serially subcultured, we
suspected that many of the mutations that we had
detected had probably been generated relatively
recently. To con®rm that the translocated c-MYC
was constitutively accumulating mutations during
routine culture, we analysed c-MYC diversity in a
Ramos subclone that had been generated by limiting
dilution and was then expanded and serially subcultured in RPMI/10%FBS for a period of 6 months.
As in the parental population, genomic PCR analysis
of the two c-MYC alleles in the Ramos subclone
revealed multiple mutations on the translocated allele
but only one on the germline allele (Figure 2a and
Table 2). Some of these sequences could be arranged
into a dynasty with the mutations presumably therefore
having accumulated stepwise (Figure 2b).
We also compared the diversity that had accumulated in the translocated c-MYC of the Ramos
subclone to that accumulated in the productively
rearranged VH segment (Table 2). The rate of VH
mutation was about twofold higher than for the
translocated c-MYC, a result which is in broad
Oncogene
Figure 2 The translocated c-MYC of Ramos diversi®es constitutively in culture. Sequence diversity was analysed by genomic
PCR from a Ramos culture that had been generated by limiting
dilution cloning (Sale and Neuberger, 1998) and expanded by
serial subculturing for 6 months in RPMI/10% FBS. (a) Pie
charts illustrate sequence diversity in the ®rst exon of the germline
and translocated c-MYC alleles as well as in the productively
rearranged VH. (b) A mutational tree that could be constructed
from the sequences of the translocated c-MYC
agreement with the relative prevalence of c-MYC and
VH mutations in our original oligoclonal Ramos
population (VH threefold higher than translocated cMYC; Figure 1c).
c-MYC mutations are characteristic of somatic
hypermutation
The mutations detected in the translocated c-MYC are
evenly spread over exon 1 (Figure 3a). Although we
have not de®ned the 3'-border of the mutation domain,
the 5'-border ± like that in immunoglobulin V genes ±
appears to be downstream of the transcription start. As
with immunoglobulin hypermutation, the c-MYC
mutations observed are largely nucleotide substitutions
(33 of the 38 independent events) with a preference for
transitions over transversions (52% rather than the
randomly expected 33%). The ®ve deletions or
c-MYC hypermutation in Burkitt's lymphoma
M Bemark and MS Neuberger
3407
Table 2 Mutations in a Ramos subclone acquired during in vitro culture
Gene
Translocated c-MYC
Exon 1
Nontranslocated c-MYC
Exon 1
Immunoglobulin VH
Exon 2
Mutation prevalence (bp7161074)
Mutations detected
Sequences analysed
Base pairs analysed
7.1
9
28
12600
0.8
1
33
14850
13.9
9
19
6460
Table 3 Nature of mutations in translocated Ramos c-MYC
From
A
C
G
T
A
C
±
1
9
0
0
±
4
0
To
G
T
Total
1
9
±
0
0
8
1
±
1
18
14
0
Most striking, however, is the preferential targeting
of the mutations to C or G residues (32/33 of the
nucleotide substitutions) with 52% of the mutations
occurring at AGC triplets. [Given the sequence of the
target, random mutational targeting would have
yielded 18% of the mutations at AGC]. Such highly
focused mutational targeting is also a feature of
immunoglobulin hypermutation in the Ramos cell line
as discussed below. In contrast, the rare mutations that
we observed in the ampli®ed BCL-6 and S14 genes
(where the mutation frequency did not rise signi®cantly
above the PCR error rate) showed no preference for
AGC hotspots.
Discussion
Figure 3 Mutations detected in the ®rst exon of the translocated
c-MYC. (a) The sequence of c-MYC is presented with position 1
being the start-site of transcripts initiated at the P1 promoter (the
major promoter used by the Ramos translocated c-MYC (Bernard
et al., 1983)). The transcription start site from the P2 promoter is
also indicated. Above the line (in upper case letters) are indicated
the independent mutations identi®ed downstream of position 98
in the translocated Ramos c-MYC in this work; all except two
(G4334A, G5254A) derive from the genomic PCR analysis.
Below the line (in lower case letters) are indicated c-MYC point
mutations that have been described in other BLs (data taken from
Rabbitts et al., 1983b, 1984; Taub et al., 1984; Showe et al., 1987;
Zajec-Kaye et al., 1988; Morse et al., 1989; Tachibana et al.,
1993). Note that in some of the cell lines only parts of exon 1 was
sequenced. Underlining highlights AGC triplets on either strand,
and the major transcriptional pausing/termination site Ta/TI is
boxed (Chung et al., 1987). (b) Deletions and insertions in the cMYC ®rst intron identi®ed in this work. Note that in the two last
cases, the exact location for the insertion/duplication events
within the oligo-dT stretch cannot be determined. The 2 bp
insertion event was identi®ed in the RT ± PCR analysis
duplications/insertion events detected were either found
adjacent to positions that had been identi®ed as point
mutated in other sequences or in a stretch of seven T's
located at the end of exon 1.
The proposition that translocation of c-MYC into the
IgH locus causes it to become a target of the antibody
hypermutation mechanism has previously been put
forward based on the detection of c-MYC mutations
and sequence heterogeneity in BL cell-lines, tumour
biopsies and primary tissue samples (Rabbitts et al.,
1983b, 1984; Taub et al., 1984; Johnston et al., 1991;
MuÈller et al., 1995). Our demonstration that the
translocated c-MYC (but not the germline c-MYC
allele or other control genes) undergoes constitutive
hypermutation in the Ramos cell line at a rate
comparable to that of IgV genes gives strong support
to such proposals. The pattern of c-MYC mutations
show the characteristic features of antibody hypermutation. The mutations are largely introduced into a
domain downstream of the transcription start-site: rare
mutations upstream of the canonical transcription start
could re¯ect occasional transcriptional initiation at
upstream c-MYC promoters (Bentley and Groudine,
1986) or within IgH sequences. Most mutations are
single nucleotide substitutions with a preference for
transitions, although, as in immunoglobulin hypermutation (particularly in humans Goossens et al., 1998;
Wilson et al., 1998) a minor proportion are deletions
and nucleotide insertions/duplications.
The distribution of c-MYC mutations that we have
detected in the cultured Ramos cell line is very similar
to that described in others BLs (KOBK1, Raji, PA682,
Daudi, BL2, BL22 and eBL3; Figure 3). Thus, 43% of
Oncogene
c-MYC hypermutation in Burkitt's lymphoma
M Bemark and MS Neuberger
3408
Oncogene
the positions in c-MYC that were mutated in Ramos
are also found to have been targeted in at least one of
the other BL lines (against a random expectation of
15%).
A non-random distribution of mutations could in
principle be attributed to a skewing e€ect of selection
or to intrinsic non-randomness of the mutational
process. In this case, we believe that the explanation
is largely provided by the intrinsic bias of the
mutational process. The intrinsic features of antibody
hypermutation have been extensively studied: mutations are introduced non-randomly with a marked
preference for hotspots, the sequences of many of these
hotpots conforming to an AGC consensus (Betz et al.,
1993). It is notable therefore that 52% of Ramos cMYC mutations and 42% of the mutations in the other
BLs are at AGC against a randomly expected 18%.
Even more striking is the G/C biased nature of the
mutational targeting. Thus, 32/33 nucleotide substitutions in Ramos c-MYC and 69/89 nucleotide substitutions in the c-MYCs of the other BLs are at G/C; this
is against a random expectation of 62% in the target
sequence. Such a marked preference for mutations at
G/C has previously been noted as a feature of antibody
hypermutation in Ramos and other BLs (Sale and
Neuberger, 1998) but is not, however, a feature of in
vivo hypermutation in man and mouse (Betz et al.,
1993; Wagner et al., 1996). This implies, as previously
discussed (Sale and Neuberger, 1998), that the antibody hypermutation process occurring in BLs, like that
in Msh2-de®cient mice (Rada et al., 1998; Frey et al.,
1998), exhibits some distinction from that taking place
in germinal centre B cells of higher organisms. The G/
C bias of the mutations in the BL lines therefore
suggests that few of their c-MYC mutations can have
been incorporated during the early, pre-transformation
life of the cell.
The mutations that we have detected in Ramos are
peculiar to the rearranged immunoglobulin V genes
and the translocated c-MYC. Mutability was not
detected at the germline c-MYC locus or in the
ribosomal S14 or BCL-6 genes. Our failure to detect
mutability at BCL-6 may at ®rst sight surprise in view
of the evidence indicating that BCL-6 acts as a
substrate for the antibody hypermutation process in
vivo. However, the in vivo mutability of BCL-6 has
been estimated as being one to two orders of
magnitude below that of the immunoglobulin loci; it
may well therefore have fallen below our level of
detection (Pasqualucci et al., 1998; Shen et al., 1998).
The results also give some insight into the cis-acting
sequences needed to confer mutability. Thus, from this
work it appears that IgH sequences downstream of Sm
suce for this purpose and that the IgH intronic
enhancer is not required. Interestingly, it was a similar
observation that translocated c-MYC transcription is
independent of the intronic enhancer that led to the
search for IgH 3'-enhancers (Neuberger and Calabi,
1983; Rabbitts et al., 1983a; Pettersson et al., 1990).
Not only might these downstream enhancers play an
important role in V gene mutation (as discussed by
Tumas-Brundage et al. (1996)) and in c-MYC
hypermutation; they might also make the intronic
enhancer dispensable for mutability.
A connection between mutability and transcription
has been noted in several studies (Betz et al., 1994;
Peters and Storb, 1996; Goyenechea et al., 1997;
Fukita et al., 1998) and Peters and Storb (1996) have
proposed that transcriptional pausing may play a role
in hypermutation. Transcriptional pause sites in cMYC are amongst the best characterized of those in
mammalian RNA polymerase II transcription units
(Bentley and Groudine, 1986; Chung et al., 1987;
Kerppola and Kane, 1988; Strobl and Eick, 1992;
Keene et al., 1999). There is no obvious correlation
between the single nucleotide mutations detected in the
c-MYC exon 1 and the major polymerase II pausing/
termination site described in this exon. Thus, the data
do not lend any strong support to a connection
between transcriptional pausing and antibody hypermutation although (I) the pause sites in the translocated c-MYC could di€er from those in the germline
allele (Spencer et al., 1990) and (II) two of the three
insertion/duplication events that we observed in the
Ramos c-MYC are within the homopolymeric dT
stretch that constitutes the major exon 1 pause site.
Finally, it is interesting to ask whether the
hypermutability of the translocated c-MYC contributes
to leukaemogenesis. Or does the chromosomal translocation simply lead to a deregulation of c-MYC
transcription with the increased mutability being an
inevitable consequence of the translocation event ± but
one of little relevance to the biology of the tumour?
With respect to the c-MYC mutations deleted in this
work, it is unlikely that any of them confer a
proliferative advantage or contribute to leukaemogenesis. The pattern of mutations detected conforms to the
intrinsic speci®city of the hypermutation mechanism
with no evidence of any skewing e€ect of selection.
Furthermore, the only consensus di€erence between the
expressed c-MYC gene in our starting Ramos culture
and its germline counterpart is an A5234T mutation
in exon 1 ± a mutation that is not observed in the
other BLs analysed. However, it remains fully possible
that mutations in c-MYC that lie outside the region
analysed in this work or that mutations in other protooncogenes that have been translocated into the
immunoglobulin loci could well contribute to leukaemogenesis. One example is the c-MYC box 1 motif that
is found at the NH2-terminus of the mature protein
and is encoded in exon 2. This motif is often mutated
and such mutations have recently been shown to
decrease the ubiquitin/proteasome-mediated turnover
of c-MYC protein (Bahram et al., 2000; Gregory and
Han, 2000).
Materials and methods
Cell culture
Ramos and the subclone Rc14 were cultured as described by
Sale and Neuberger (1998).
DNA preparation and analysis
Genomic DNA was prepared from between 4 ± 206106 cells
using either QIAamp DNA blood mini kit (Qiagen), and
cDNA was prepared according to standard procedures from
total RNA isolated using Trizol reagent (Life Technologies).
The PCRs were performed using pfuTurbo polymerase
(Stratagene) as recommended by the manufacturer with
between 1 ± 1/10 ml of DNA in a total reaction volume of
c-MYC hypermutation in Burkitt's lymphoma
M Bemark and MS Neuberger
50 ml. The PCR program was 30 cycles (958C for 1 min, 558C
for 30 s, and 728C for 2 min), except for the VH PCR which
was performed exactly as described (Sale and Neuberger,
1998). The primers used in the other PCRs were as follows:
c-MYC cDNA
5' primer:
CAGGGTACCGGCACTTTGCACTGGAACTT
3' primer:
CAGGAGCTCTCTGGTTCACCATGTCTCCT
C-MYC genomic
5' primer, translocated allele:
CAGGGTACCCTGTTCAGCACAGCACATC
5' primer, wild type allele:
CAGGGTACCTCCCTGGGACTCTTGATCAA
3' primer:
CAGGAGCTCTTCGGTGCTTACCTGGTTT
BCL-6
5' primer:
CAGGGTACCAAAATCCACCGATGCCGATC
3' primer:
CAGGAGCTCTCAATTCTCGGAATTTGAGC
S14
5' primer:
CGGGGTACCCGGGACAGACGTGGGCTCCCG
3' primer:
CAGGAGCTCTAAGGGAGAGAGAAACTGAC
3409
The PCR products were puri®ed using Qiagen PCR
puri®cation kit, digested with SacI/KpnI, repuri®ed with the
PCR puri®cation kit and ®nally cloned into pBluescript
(Stratagene) between the SacI and KpnI sites. Plasmids were
prepared using QIAprep 8 Turbo kits (Qiagen), and were
sequenced from the T3 and T7 promoters on an ABI 377
sequencer. The sequences were aligned and analysed using the
GAP4 alignment program.
Acknowledgments
We thank Julian Sale for advice and discussions during the
work. M Bemark was supported by a post-doctoral
fellowship from the Swedish Cancer Society.
References
Bahram F, von der Lehr N, Cetinkaya C and Larsson L-G.
(2000). Blood, 95, 2104 ± 2110.
Bentley DL and Groudine M. (1986). Nature, 321, 702 ± 706.
Bernard O,Cory S, Gerondakis S, Webb E and Adams JM.
(1983). EMBO J., 2, 2375 ± 2383.
Betz AG, Milstein C, GonzaÂlez-FernaÂdez A, Pannell R,
Larson T and Neuberger MS. (1994). Cell, 77, 239 ± 248.
Betz AG, Rada C, Pannell R, Milstein C and Neuberger MS.
(1993). Proc. Natl. Acad. Sci. USA, 90, 2385 ± 2388.
Cesarman E, Dalla-Favera R, Bentley D and Groudine M.
(1987). Science, 238, 1272 ± 1275.
Chung J, Sussman DJ, Zeller R and Leder P. (1987). Cell, 51,
1001 ± 1008.
Dalla-Favera R, Bregni M, Erikson J, Patterson D, Gallo
RC and Croce CM. (1982). Proc. Natl. Acad. Sci. USA, 79,
7824 ± 7827.
Frey S, Bertocci B, Delbos F, Quint L, Weil JC and Reynaud
CA. (1998). Immunity, 9, 127 ± 134.
Fukita Y, Jacobs H and Rajewsky K. (1998). Immunity, 9,
105 ± 114.
Goossens T, Klein U and KuÈppers R. (1998). Proc. Natl.
Acad. Sci. USA, 95, 2463 ± 2468.
Goyenechea B, Klix N, YeÂlamos J, Williams GT, Riddell A,
Neuberger MS and Milstein C. (1997). EMBO J., 16,
3987 ± 3994.
Gregory MA and Han SR. (2000). Mol. Cell. Biol., 20, 2423 ±
2435.
Hamlyn PH and Rabbitts TH. (1983). Nature, 304, 135 ± 139.
Johnston JM, Yu MT and Carroll WL. (1991). Blood, 78,
2419 ± 2425.
Johnston JM and Carroll WL. (1992). Leuk. Lymph., 8, 431 ±
439.
Keene RG, Mueller A, Landick R and London L. (1999).
Nucleic Acids Res., 27, 3173 ± 3182.
Kerppola TK and Kane CM. (1988). Mol. Cell. Biol., 8,
4389 ± 4394.
Lebecque SG and Gearhart PJ. (1990). J. Exp. Med., 172,
1717 ± 1727.
Migliazza A, Martinotti S, Chen W, Fusco C, Ye BH,
Knowles DM, Ot K, Chaganti RS and Dalla-Favera R.
(1995). Proc. Natl. Acad. Sci. USA, 92, 12520 ± 12524.
Morse B, South VJ, Rothberg PG and Astrin SM. (1989).
Mol. Cell. Biol., 9, 74 ± 82.
MuÈller JR, Janz S, Goedert JJ, Potter M and Rabkin CS.
(1995). Proc. Natl. Acad. Sci. USA, 92, 6577 ± 6581.
Neuberger MS and Calabi F. (1983). Nature, 305, 240 ± 243.
Parham P. (ed). (1998). Immunol. Rev. Munksgaard,
Copenhagen.
Pasqualucci L, Migliazza A, Fracchiolla N, William C, Neri
A, Baldini L, Chaganti RS, Klein U, KuÈ ppers R,
Rajewsky K and Dalla-Favera R. (1998). Proc. Natl.
Acad. Sci. USA, 95, 11816 ± 11821.
Peters A and Storb U. (1996). Immunity, 4, 57 ± 65.
Pettersson S, Cook GP, BruÈggemann M, Williams GT and
Neuberger MS. (1990). Nature, 344, 165 ± 168.
Rabbitts TH, Forster A, Baer R and Hamlyn PH. (1983a).
Nature, 306, 806 ± 809.
Rabbitts TH, Forster A, Hamlyn P and Baer R. (1984).
Nature, 309, 592 ± 597.
Rabbitts TH, Hamlyn PH and Baer R. (1983b). Nature, 306,
760 ± 765.
Rada C, YeÂlamos J, Dean W and Milstein C. (1997). Eur. J.
Immunol., 27, 3115 ± 3120.
Rada C, Ehrenstein MR, Neuberger MS and Milstein C.
(1998). Immunity, 9, 135 ± 141.
Sale JE and Neuberger MS. (1998). Immunity, 9, 859 ± 869.
Shen HM, Peters A, Baron B, Zhu X and Storb U. (1998).
Science, 280, 1750 ± 1752.
Showe LC, Moore RC, Erikson J and Croce CM. (1987).
Proc. Natl. Acad. Sci. Acad. USA, 84, 2824 ± 2828.
Spencer CA, LeStrange RC, Novak U, Hayward WS and
Groudine M. (1990). Genes Dev., 4, 75 ± 88.
Strobl LJ and Eick D. (1992). EMBO J., 11, 3307 ± 3314.
Tachibana K, Takayama N, Matsuo K, Kato S, Yamamoto
K, Ohyama K, Umezawa A and Takano T. (1993). Gene,
124, 231 ± 237.
Taub R, Kirsch I, Morton C, Lenoir G, Swan D, Tronick S,
Aaronson S and Leder P. (1982). Proc. Natl. Acad. Sci.
USA, 79, 7837 ± 7841.
Taub R, Moulding C, Battery J, Murphy W, Vasicek T,
Lenoir GM and Leder P. (1984). Cell, 36, 339 ± 348.
Tumas-Brundage K, Vora KA, Giusti AM and Manser T.
(1996). Semin. Immunol., 8, 141 ± 150.
Tumas-Brundage K and Manser T. (1997). J. Exp. Med.,
185, 239 ± 250.
Wagner SD, Elvin JG, Norris P, McGregor JM and
Neuberger MS. (1996). Int. Immunol., 8, 701 ± 705.
Wagner SD and Neuberger MS. (1996). Annu. Rev.
Immunol., 14, 441 ± 457.
Wilson PC, de Boutelier O, Liu Y-J, Potter K, Banchereau J,
Capra JD and Pascual V. (1998). J. Exp. Med., 187, 59 ±
70.
Oncogene
c-MYC hypermutation in Burkitt's lymphoma
M Bemark and MS Neuberger
3410
Oncogene
Wiman KG, Clarkson B, Hayday AC, Saito H, Tonegawa S
and Hayward WS. (1984). Proc. Natl. Acad. Sci. USA, 81,
6798 ± 6802.
YeÂlamos J, Klix N, Goyenechea B, Lozano F, Chui YL,
Gonzalez Fernandez A, Pannell R, Neuberger MS and
Milstein C. (1995). Nature, 376, 225 ± 229.
Zajec-Kaye M, Gelmann EP and Levens D. (1988). Science,
240, 1776 ± 1780.