NEOPLASIA

Expression of the AID protein in normal and neoplastic B cells

Laura Pasqualucci, Roberta Guglielmino, Jane Houldsworth, Jessica Mohr, Said Aoufouchi, Roberto Polakiewicz,
R. S. K. Chaganti, and Riccardo Dalla-Favera

Somatic hypermutation (SHM) targets pri- GC-derived lymphomas, but the pattern the AID protein appeared predominantly
marily the immunoglobulin variable re- of expression of the AID protein is not localized in the cytoplasm. These results
gion (IgV) genes in germinal center (GC) known. Using 2 specific antibodies, here indicate that the AID protein is specifi-
B cells, thereby allowing antibody affinity we show that the AID protein can be cally expressed in normal and trans-
maturation. A malfunction of SHM, termed detected in GC centroblasts and their formed GC B cells; nonetheless, its pre-
aberrant somatic hypermutation (ASHM), transformed counterpart (Burkitt lym- dominantly cytoplasmic localization
was found in about 50% of diffuse large phoma) but not in pre-GC B cells and suggests that additional mechanisms may
B-cell lymphomas (DLBCLs), leading to post-GC neoplasms, including B-cell regulate its function and may be altered
mutations in the 5ⴕ sequences of multiple chronic lymphocytic leukemia and mul- during lymphomagenesis. (Blood. 2004;
genes, including oncogenes. Although tiple myeloma. DLBCLs displayed vari- 104:3318-3325)
the SHM mechanism is largely unknown, able levels of AID expression, which did
it was shown to require the activation- not correlate with IgV ongoing hypermuta-
induced cytidine deaminase (AID) gene. tion, ASHM, or disease subtype. Finally,
AID mRNA is expressed in GC B cells and both in normal and malignant B cells © 2004 by The American Society of Hematology

Introduction
Somatic hypermutation (SHM) is a specialized process that takes of the high homology with the RNA-editing enzyme apolipoprotein
place in germinal center (GC) B cells in response to T cell– B editing catalytic subunit 1 (APOBEC-1), it has been proposed
dependent antigen stimulation.1,2 This process introduces single that AID may function as a cytidine deaminase to modify a
nucleotide substitutions, with occasional deletions and duplica- preexisting mRNA into a new one, possibly encoding an endonucle-
tions, primarily into the variable region of the immunoglobulin ase.14,17 However, experimental evidence in Escherichia coli
(IgV) heavy and light chain genes, resulting in the production of indicated that AID may act directly on DNA and convert deoxycy-
high-affinity antibodies and allowing affinity maturation of the tidines to uracils, which are then processed by uracil-DNA
humoral immune response.3 SHM requires active transcription of glycosylase (UNG) and endonucleases.18 AID would thereby lead
the target locus but is not IgV sequence specific and does not to the creation of abasic sites, which may be repaired by base
depend on the V region promoter.4-7 In fact, at least 3 non-Ig genes, excision repair and putative error-prone mechanisms. Indeed, in
including BCL6, the FAS/CD95 gene, and the genes encoding the 2 vitro AID exhibits cytidine deamination activity on single-stranded
components of the BCR (B29 and mb1) have been shown to acquire DNA, with a base specificity similar to that reported for SHM.19-22
somatic mutations during the normal GC reaction, indicating that Recently, the SHM process has been shown to malfunction in
this mechanism may target more genes than originally suspected.8-12 about 50% of diffuse large B-cell lymphomas (DLBCLs) as well as
The molecular basis of SHM remains largely unknown. How- in about 20% of AIDS-related non-Hodgkin lymphomas (NHLs)
ever, studies from the past 4 years have identified the AID gene, and in a significant fraction of primary central nervous system
encoding for activation-induced cytidine deaminase, as an absolute lymphomas derived from non-HIV patients.23-25 In these tumors,
requirement for both SHM and class switch recombination (CSR) multiple somatic mutations are introduced into the 5⬘ region,
in humans and mice.13,14 Activation-induced cytidine deaminase including coding sequences, of several genes that do not represent
(AID) expression is also sufficient to initiate both events in physiologic SHM targets. These comprise the well-known proto-
fibroblasts expressing transcribed artificial constructs.15,16 Because oncogenes PIM1, PAX5, RhoH/TTF, and cMYC, all of which have

From the Institute for Cancer Genetics and the Departments of Pathology and Two of the authors (R.P. and J.M.) are employed by a company (Cell Signaling
Genetics & Development, Columbia University, New York, NY; Laboratory of Technology, Inc, Beverly, MA) whose potential product was studied in the
Cancer Genetics and the Department of Medicine, Memorial Sloan-Kettering present work.
Cancer Center, New York, NY; Cell Signaling Technology, Beverly, MA; and
INSERM U373 Faculté de Médecine Necker - Enfants Malades, Paris, France. The online version of this article contains a data supplement.

Submitted April 26, 2004; accepted June 28, 2004. Prepublished online as An Inside Blood analysis of this article appears in the front of this issue.
Blood First Edition Paper, August 10, 2004; DOI 10.1182/blood-2004-04-1558. Reprints: Laura Pasqualucci, Institute for Cancer Genetics, Columbia
Supported by a Leukemia and Lymphoma Society Specialized Center of University, 1150 St Nicholas Ave, New York, NY 10032; e-mail: lp171@
Research (SCOR) grant with contributions from the Cathy and Scott Zeilinger columbia.edu.
Philanthropic Fund of the Jewish Community Federation of Cleveland, the Carl The publication costs of this article were defrayed in part by page charge
and Ruth Shapiro Family Foundation, Lesley Goldwasser and Jonathan Plutzik payment. Therefore, and solely to indicate this fact, this article is hereby
and Friends, and Marla Kaminski and Friends. L.P. is a Special Fellow of the marked ‘‘advertisement’’ in accordance with 18 U.S.C. section 1734.
Leukemia and Lymphoma Society. Also supported by the Ligue Nationale
Contre le Cancer (Equipe labellisée) (S.A.). © 2004 by The American Society of Hematology

3318 BLOOD, 15 NOVEMBER 2004 䡠 VOLUME 104, NUMBER 10

BLOOD, 15 NOVEMBER 2004 䡠 VOLUME 104, NUMBER 10 AID PROTEIN EXPRESSION IN B CELLS 3319

been already implicated in the pathogenesis of lymphoid malignan- The following Abs were also used: anti-BCL6 (N3, Santa Cruz Biotechnol-
cies. Analogous to the physiologic targets (IgV and BCL6), ogy, Santa Cruz, CA), anti-HA (3F10, Roche Diagnostics, Indianapolis,
mutations at these loci are scattered through the first approximately IN), anti–␤-actin (AC15, Sigma, St Louis, MO), and anti–␣-tubulin (clone
2 kb from the promoter with characteristic hotspot (RGYW/ B-5-1-2, Sigma).
WRCY) motifs and exhibit a bias for transitions over transversions,
with higher frequency at G:C pairs.23,24 Such distinctive features Protein extracts, cell fractionation, and Western blot analysis
strongly support the hypothesis that this phenomenon, termed
Total protein extracts were prepared from exponentially growing cell lines
“aberrant somatic hypermutation” (ASHM), results from an aber-
and snap-frozen lymphoma biopsies as described previously.36 To isolate
rant activity, possibly a loss of target specificity, of the physiologic
nuclear and cytoplasmic fractions, cells were incubated for 15 minutes at
SHM process and therefore of AID itself. Because the mutations 4°C in hypotonic buffer A (10 mM HEPES [N-2-hydroxyethylpiperazine-
affect both regulatory and coding sequences of the affected genes, N⬘-2-ethanesulfonic acid] pH 7.9, 10 mM KCl, 0.1 mM EDTA [ethylenedia-
with several amino acid substitutions predicting a change in the minetetraacetic acid], 1 mM dithiothreitol [DTT], and protease inhibitors
protein structure, ASHM may represent a major contributor to [Sigma]); Nonidet P-40 (NP40) was added to a final concentration of 0.4%,
DLBCL pathogenesis. Indeed, for some of the targeted genes, such and the samples were centrifuged for 5 minutes at 3000 rpm. Supernatants
as cMYC, the oncogenic effect of some mutations has been were used as cytoplasmic extracts after further centrifugation for 30
demonstrated.26-28 minutes at 14 000 rpm. The nuclear pellets were washed twice in buffer A,
Based on RNA expression data, AID appears selectively resuspended in about 10 volumes of buffer C (20 mM HEPES pH 7.9, 400
expressed in GC B cells and GC-derived malignancies14,29-31; mM NaCl, 1 mM EDTA, 0.4% NP40, 1 mM DTT, and protease inhibitors),
vortexed for 30 minutes at 4°C, and centrifuged for 30 minutes at 14 000
however, the pattern of expression of the AID protein has never
rpm. The final salt concentration in the nuclear extracts was adjusted to 300
been investigated. Using 2 specific antibodies, we examined the
mM by addition of buffer D (20 mM HEPES pH 7.9, 1 mM EDTA, 1 mM
distribution and levels of the AID protein in purified normal B-cell DTT, and protease inhibitors). For Western blot analysis, protein extracts
subpopulations (naive and centroblasts) as well as in B-lymphoid (30 ␮g whole-cell extract or 10 ␮g subcellular fraction) were gel
tumors derived from GC and post-GC B cells. These results have electrophoresed on 4% to 20% gradient sodium dodecyl sulfate–
implications for the mechanism regulating AID function and for its polyacrylamide gel electrophoresis (SDS-PAGE) gels (Invitrogen Life
possible role in lymphomagenesis. Technologies, Carlsbad, CA), transferred to nitrocellulose membranes, and
immunostained according to standard methods. Proteins were detected
using the enhanced chemiluminescence (ECL) reagents (Amersham Bio-
sciences, Freiburg, Germany) as recommended by the manufacturer.
Materials and methods To compare the AID protein expression levels in DLBCL samples from
Cell lines and primary lymphoma cases different experiments, equivalent amounts of GC centroblasts were loaded
in each gel and used as an internal reference. Band intensities were
All B-cell lines used were cultured in Iscove modified Dulbecco medium measured using the NIH Image program 1.62, and AID expression was
(IMDM) or RPMI (Gibco, Grand Island, NY) supplemented with 10% fetal normalized first for ␤-actin, as a control for loading, and then for the CB
calf serum (FCS), 100 U/mL penicillin, 100 U/mL streptomycin, and 2 mM sample loaded in the same gel.
L-glutamine. The lung carcinoma cell line H1299 was grown in Dulbecco
modified Eagle medium (DMEM)/10% FCS. Ramos and H1299 were
stably transduced with pBABE-puro retroviral vectors or with vectors RNA extraction, Northern blot analysis, and gene
expressing a C-terminal Flag-hemagglutinin (HA)–tagged human AID expression profiling
cDNA (AID-FH), followed by selection in 0.5 ␮g/mL puromycin. The 25
newly diagnosed DLBCL frozen tissue samples were obtained from the Total RNA was prepared using the Trizol reagent (Invitrogen Life
Department of Pathology of the Memorial Sloan-Kettering Cancer Center.32 Technologies) followed by RNeasy purification (QIAGEN, Valencia, CA),
Only cases in which the fraction of neoplastic cells corresponded to more and Northern blot analysis was performed according to standard proce-
than 70% (n ⫽ 21) were selected for the correlative analysis between AID dures, with radiolabeled probes corresponding to the full-length human
protein expression and DLBCL subgroup or ASHM. The B-cell chronic AID, BCL6, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH)
lymphocytic leukemia (B-CLL) primary cases and their characterization for cDNAs. The detailed characterization of normal B-cell subpopulations by
the presence of IgV mutations are described in detail elsewhere.33 gene expression profiling has been reported previously.34 The protocol for
microarray hybridization of the DLBCL samples to the Affymetrix HG95Av2
Isolation of normal human B-cell subpopulations from tonsils oligonucleotide array, as well as the expression data analysis, has also been
reported in detail.32
Naive and centroblast (CB) B-cell subpopulations were isolated by
magnetic cell separation using the MidiMACS system (Miltenyi Biotec,
Auburn, CA) as described.34 Briefly, naive B cells were isolated by PCR amplification, cloning, and sequencing analysis of PIM1,
depletion of GC B cells (CD10, CD27), memory B cells (CD27), plasma PAX5, RhoH/TTF, and cMYC
cells/blasts (CD27), and T cells (CD3), followed by a positive enrichment
Genomic DNA from 31 DLBCL samples, including 10 cell lines and 21
of IgD-positive cells. CBs were isolated in a single step by magnetically
primary biopsies, was extracted according to standard methods. Mutational
labeling CD77⫹ cells. The purity of the isolated fractions was determined
analysis of PIM1, PAX5, RhoH/TTF, and cMYC was performed on selected
on a FACSCalibur (Becton Dickinson, San Jose, CA) and by IgV region
genomic regions previously shown to contain more than 90% of the
gene analysis.34
mutations found in DLBCL.23 Following polymerase chain reaction (PCR)
amplification, purified amplicons were sequenced directly from both
Antibodies
strands and compared with the corresponding germ line sequences as
The method used for the generation of the anti-AID rabbit polyclonal described.23 All sequence variants were confirmed on independent PCR
antibody (Ab) is reported by Faili et al.35 The anti-AID mouse monoclonal products, and nucleotide changes due to previously reported polymor-
Ab (clone 7E7), an IgG1 ␬ light chain, was raised against a synthetic phisms or occurring more than once in separate cases were excluded from
peptide selected from the C-terminal half of the human AID protein. the assessment of the mutation frequencies unless their somatic origin could
Additional details are provided as Supplemental Materials (at the Blood be proven. Cases were scored as ASHM-positive if they harbored 1 or more
website, see the Supplemental Materials link at the top of the online article). somatic mutations distributed in 1 or more of the 4 proto-oncogenes.
3320 PASQUALUCCI et al BLOOD, 15 NOVEMBER 2004 䡠 VOLUME 104, NUMBER 10

Analysis of IgV gene intraclonal variation expected based on RNA data (not shown). Western blot analysis
The rearranged IgVH genes were amplified from genomic DNA using the
with an anti-HA Ab, which recognizes only the tagged protein,
Pfu Turbo polymerase (Stratagene, La Jolla, CA) together with a set of 6 VH revealed the same 27-kDa band (Figure 1C, third panel), document-
gene family–specific primers hybridizing to sequences in the framework ing its exogenous nature. Similar results were obtained using the
region (FR)I and a JH primer mix in separate reactions for each VH primer.8 monoclonal and the polyclonal Abs, although the latter one gave
PCR products were gel purified and cloned into pGEM-T vector (Invitrogen significantly higher background. The reactivity in cells expressing
Life Technologies) as described.23 For each PCR product, 25 to 58 clones AID mRNA or a transfected AID cDNA indicates that both the
were sequenced and analyzed. Mutation frequencies were calculated by monoclonal and polyclonal Abs can specifically recognize the
dividing the total number of unique mutations (ie, changes observed in
AID protein.
individual subclones) by the total number of base pairs sequenced.
AID protein expression in normal B-cell subpopulations

To examine the expression pattern of the AID protein in normal

Results
B-cell subpopulations, pre-GC naive B cells (IgD⫹ CD27⫺) and
Identification of the AID protein by specific antibodies GC CBs (CD77⫹) were purified from the reactive tonsil of normal
individuals, and Western blot analysis was performed on total
Two antibodies were used throughout the study: one is a rabbit protein extracts as well as on nuclear and cytoplasmic cell
polyclonal Ab raised in the laboratory of one of the authors fractions. Elevated AID protein levels were detected in the CB
(S.A.)35; the second is a monoclonal Ab obtained against a peptide
population (ie, the cells deputed to SHM) but not in the pre-GC
sequence selected from the C-terminal half of the human AID
naive B cells, confirming RNA data generated by gene expression
polypeptide (see “Materials and methods”). To test the ability of
profiling (Figure 2A-B). Unexpectedly for a protein that is
these Abs to recognize the human AID protein, we first used 4
presumed to function as a DNA mutator, Western blot analysis of
B-cell lines that were documented by Northern blot (NB) analysis
GC B cells after cell fractionation revealed a predominantly
to express (ODHI I, Ly10) or lack (ODHI III, SUDHL6) endoge-
cytoplasmic localization of AID, with only traces being detected in
nous AID mRNA (Figure 1A). Western blot analysis on total
the nuclear fraction (Figure 2C). The purity of the extracts was
protein extracts detected a specific band of the expected 24-kDa
monitored using anti–␣-tubulin and anti-BCL6 Abs, respectively.
molecular weight in ODHI I and Ly10 but not in the 2 RNA-
Thus, the AID protein is expressed in GC B cells, where it resides
negative cell lines ODHI III and SUDHL6 (Figure 1B; the
predominantly in the cytoplasm.
polyclonal Ab is shown). We then engineered a human epithelial
cell line (H1299), lacking AID mRNA expression, and the B-cell AID protein expression in B-cell malignancies
line Ramos, which expresses low levels of AID mRNA, to stably
express an exogenous, Flag-HA–tagged human AID protein (AID- In B-cell neoplasms, the AID protein was found restricted to cells
FH) under the control of pBABE retroviral sequences. Figure 1C displaying a mature GC phenotype, including 10 of 10 group I
shows the presence of an approximately 27-kDa band (exo)- Burkitt lymphoma (BL) cell lines, where the levels were mostly
consistent with the predicted size of the double-tagged fusion comparable to CBs (Figure 3, top left panel). AID expression was
protein—only in AID-FH–transfected cells (Figure 1C, top 2 also examined in 3 pairs of BL lines (KEM, MUTU, ODHI), each
panels, lanes 2 and 4) but not in control vector–transfected cells one derived from the same tumor case but selected in vitro to
(Figure 1C, lanes 1 and 3). In addition, a band of smaller molecular represent either a group I GC phenotype or a group III post-GC
weight was recognized in both AID-FH– and control vector– immunoblastic phenotype.37 Elevated AID protein levels were
transfected Ramos cells (Figure 1C, lanes 3 and 4); this smaller detected in the group I but not in the group III BL lines (Figure 3,
band corresponded in size to the endogenous AID protein (endo) top left panel), suggesting that AID expression may be down-
and was not present in H1299 (Figure 1C, lanes 1 and 2), as regulated in this later stage of differentiation. Among DLBCL

Figure 1. Identification of the AID protein by specific antibodies. (A) Northern blot analysis of endogenous AID expression in 2 BL (ODHI I, ODHI III) and 2 DLBCL (Ly10,
SUDHL6) cell lines. An approximately 2.7-kb message, corresponding to the AID transcript, can be detected in ODHI I and Ly10 but not in ODHI III and SUDHL6. Filters were
stripped and sequentially hybridized with probes for BCL6, as marker of differentiation stage, and GAPDH as control for loading. (B) Western blot analysis of the same cell lines
with anti-AID rabbit polyclonal Abs shows a specific band of the predicted 24-kDa molecular weight in ODHI I and Ly10, but not in ODHI III and SUDHL6, consistent with the
RNA data in panel A. Immunoblotting with anti-BCL6 (N3) and anti–␤-actin Abs is shown in the bottom panels. (C) Western blot analysis of AID expression in H1299 and Ramos
cells, stably transduced with pBABE retroviral vectors (H1299-V, Ramos-V) or with vectors expressing an AID-Flag-HA protein (H1299-AID-FH and Ramos-AID-FH). Both the
rabbit polyclonal antiserum (poly) and the mouse monoclonal (7E7) Abs were used (top 2 panels). An approximately 27-kDa band corresponding to the size of the exogenous
protein (exo) is detected in lysates from AID-FH–transduced cells (lanes 2 and 4) but not in cells transduced with the control vector (lanes 1 and 3). In the B-cell line Ramos, the
endogenous AID protein (endo) can also be seen (lanes 3 and 4). Membranes were stripped and sequentially probed with anti-HA, which recognizes the exogenous protein,
and with anti–␤-actin as control for loading (bottom 2 panels).
BLOOD, 15 NOVEMBER 2004 䡠 VOLUME 104, NUMBER 10 AID PROTEIN EXPRESSION IN B CELLS 3321

Figure 2. AID protein expression and subcellular localization in normal B lymphocytes. (A) AID mRNA expression in purified normal B-cell subpopulations (5 samples
each of N [naive], CB [centroblasts], CC [centrocytes], M [memory]). AID was represented by 2 probe sets in the Affymetrix U133Plus GeneChip array. BCL6, BCL7A, and the
proliferation-associated gene Ki67 are shown along with BCL2 and TOSO as markers for the identity of the purified subpopulations. (B) Western blot analysis of AID and ␤-actin
expression in whole cell extracts obtained from naive and CB populations. (C) Western blot analysis of AID expression in nuclear (nu) and cytoplasmic (cy) extracts from
tonsillar GC B cells. The purity of the fractions was monitored using anti-BCL6 and anti–␣-tubulin Abs, respectively.

cases, which included 25 primary biopsies and 10 cell lines, AID ABC-like, and 4 of 6 (66.6%) positive cases in the type 3 DLBCL
expression was very heterogeneous, ranging from complete ab- (Figure 4, top panel); however, when compared with the GCB type,
sence (3 of 10 cell lines and 10 of 25 biopsies; overall, 13 of 35 the ABC DLBCLs had on average higher expression levels (about
samples, corresponding to 37% of the cases) to levels comparable 3-fold) both in primary biopsies and in cell lines (Figure 4, bottom
to the ones observed in CBs (Figure 3, top 2 right panels, shows panel). Moreover, the overall higher levels of AID in ABC
representative results). No evidence of AID protein expression was DLBCLs tend to inversely correlate with the expression of BCL6, a
found in post-GC tumors such as MM (0 of 7 cell lines) and B-CLL well-established marker of GC (Figure 1A-B and data not shown),
(0 of 12 samples, including 10 primary cases and 2 cell lines) differently from what is observed in GC B cells or in BL, where the
regardless of their IgV mutational status (Figure 3, bottom panels). 2 proteins are coexpressed (see “Discussion”).
In all samples studied, NB analysis documented the complete
AID expression in DLBCL does not correlate with ongoing SHM
concordance between RNA and protein levels in terms of relative
distribution (data not shown). Collectively, these data indicate that To examine whether the differential expression of AID in DLBCL
expression of the AID protein is restricted to B-cell malignancies cases reflects the ongoing activity of the SHM machinery, we
derived from GC cells. investigated a number of samples for the presence of intraclonal
Expression of the AID protein in DLBCL subgroups variation (ICV) in their IgV sequences, an indicator that SHM has
been active during clonal expansion. The rearranged IgV genes of 6
Based on gene expression profiling studies, DLBCLs have been cell lines, representative of AID-positive (Ly3, Ly7, Ly10, RCK8)
classified into 3 biologically and clinically distinct subgroups: the and -negative (Ly8, SUDHL6) cases, were amplified from genomic
GC B-cell–like (GCB), the activated B-cell–like (ABC), and the DNA using the proofreading Pfu Turbo polymerase and then
type 3 DLBCL.38,39 To investigate whether the AID protein was cloned and sequenced (n ⫽ 25 to 58 clones per sample). The BL
differentially expressed among these phenotypic groups, we used Ramos clone 6, which is known to undergo SHM in culture
the Affymetrix U95Av2 oligonucleotide microarray to generate constitutively, was used as positive control, while Ramos clone 1,
gene expression profiles from 30 DLBCL samples, including 9 cell which has lost the ability to hypermutate, served as a negative
lines and the 21 primary biopsies that contained more than 70% control.40 Indeed, detectable AID protein levels were observed in
tumor cells, and then measured the AID protein levels in each Ramos clone 6 and were associated with significant intraclonal
group identified. The analysis classified 15 samples (50%) into the variation in its IgV sequences (22 mutational events distributed in
GCB subgroup, 9 samples (30%) into the ABC subgroup, and 6 14 of 27 clones, corresponding to a frequency of 0.239 ⫻ 10⫺2/bp;
samples (20%) into the type 3 group. In all of them, expression of P ⬍ .0001); conversely, only 1 nucleotide difference was found in
the AID protein was very heterogeneous, with 9 of 15 (60%) 31 DNA clones sequenced from the AID-negative Ramos clone 1
positive cases in the GCB-like, 7 of 9 (77.8%) positive cases in the (frequency, 0.009 ⫻ 10⫺2/bp, which is not significantly different

Figure 3. The AID protein is expressed in GC but not in post-GC–derived B-cell malignancies. Western blot analysis of AID expression in BL (top left panels), DLBCL (top
right panels), and post-GC–derived B-CLL and MM samples (bottom panels). Both IgV mutated and unmutated cases were included in the B-CLL filter. The top right panel
shows 2 pairs of isogenic BL cell lines representative of the group I/latency I and group III/latency III phenotype (see text for a detailed explanation); the B-cell lymphoma line
BJAB serves as a negative control, because it does not express AID mRNA (L.P., unpublished results, 2003). In each membrane, equivalent amounts of total protein extracts
from purified CBs were included; ␤-actin controls for loading.
3322 PASQUALUCCI et al BLOOD, 15 NOVEMBER 2004 䡠 VOLUME 104, NUMBER 10

AID expression levels do not correlate with ASHM in DLBCL

To investigate whether AID overexpression is involved in the

pathogenesis of DLBCL-associated ASHM, we comparatively
analyzed AID protein levels and the presence of somatic mutations
in 4 genes previously identified as aberrant targets: PIM1, PAX5,
RhoH/TTF, and cMYC.23 To this end, 31 DLBCL samples were first
subjected to PCR amplification and direct sequencing of selected
regions representing the hypermutable area in these genes.23
Because PCR products were sequenced directly in all experiments,
the contribution of Taq polymerase error to mutations in the final
sequence is insignificant. The overall mutation load in each case
was then plotted against the normalized AID protein levels,
measured by Western blotting, and quantitated using the NIH
Image software.
PCR products of PIM1, PAX5, RhoH/TTF, and cMYC were
obtained from all DLBCL cell lines and from 20 of 21 primary
cases (30 total evaluable samples). The analysis revealed the
complete absence of somatic mutations in 2 of 10 cell lines (20%)
and 5 of 20 primary biopsies (25%), while the remaining 23
samples, including 8 cell lines and 15 biopsies, carried 1 or more
Figure 4. The AID protein is expressed in both GCB- and ABC-like DLBCLs. somatic mutations distributed in 1 or more of the 4 target genes
(Top) AID protein levels, measured by Western blotting and normalized for ␤-actin as
(range, 1 to 28 mutational events per case in the cell lines; 1 to 18
described in “Materials and methods,” are shown in individual DLBCL samples
classified by gene expression profiling as GCB-like (E), ABC-like (F), and type 3 (O). mutational events per case in the primary biopsies) (Supplemental
Two different scales are used for the primary biopsies (left) and the cell lines (right), Table S2). All the mutations, which included mostly single
because lower AID levels were consistently observed in the former ones, likely due to nucleotide substitutions but also 4 deletions and 1 duplication, were
the presence of contaminating nontumor cells. (Bottom) Histograms show the
average AID protein expression levels in the 3 groups. Error bars indicate standard confirmed on independent PCR products. These results are analo-
deviations. gous to those originally reported for an independent cohort of
DLBCL samples.23 When we compared the mutation status with
the AID protein levels, we found that, both in cell lines and in
from the expected error rate of the Pfu polymerase) (Figure 5A; see primary biopsies, AID expression did not correlate with the
also Supplemental Table S1). Of the DLBCL cell lines analyzed, presence of ASHM (Figure 6). In particular, AID was expressed in
only 1 (Ly7) displayed a significant number of IgV intraclonal 3 of 7 (43%) unmutated and 17 of 23 (74%) aberrantly mutated
variants (11 of 19 426 bp, corresponding to a mutation frequency of samples (Figure 6, top). Moreover, no direct correlation was
0.057 ⫻ 10⫺2/bp), while the frequency of mutations in the remain- observed in the mutated samples between AID protein amount
ing 5 lines was not significantly higher than the polymerase error and the total number of mutations (not shown), indicating that
rate (range, 2 of 5957 bp to 5 of 23 142 bp sequenced) (Figure 5B a simple increase in AID expression levels is not sufficient to
and Supplemental Table S1). These extremely low frequencies explain ASHM.
were detected regardless of the AID protein levels and even in cell
Predominant cytoplasmic localization of the AID protein in
lines, such as Ly3 and Ly10, which express protein amounts
normal and transformed B cells
comparable to GC B cells and higher than Ramos clone 6 (Figure
5B). Thus, differently from GC CB and BL cases, the heteroge- To examine whether abnormal subcellular localization, rather than
neous expression of AID in DLBCLs does not reflect the level of overexpression of AID, could be responsible for loss of target
activity of the SHM machinery. specificity in DLBCL cases with ASHM, we performed Western

Figure 5. AID protein expression is associated with

ongoing SHM in Ramos but not in DLBCL cell lines. (A)
Total cell extracts were prepared from 2 Ramos clones that
express (cl.6) or lack (cl.1) AID mRNA expression40 and, from
each of them, 3 dilutions were analyzed by immunoblotting
using anti-AID Abs and ␤-actin as control (top). In the bottom
panel, the AID protein levels from the top panel, quantitated
using the NIH Image software and normalized for ␤-actin (f),
are compared with the levels of ICV in the corresponding
rearranged IgV genes (o), expressed as frequency of muta-
tions and calculated as described in “Materials and meth-
ods.” The mutation frequency in the consensus sequence,
defined as the sequence identified by direct sequencing and
common to all DNAclones analyzed, is indicated in parenthe-
ses below the chart (also see Supplemental Table S1). (B)
AID protein expression in 10 DLBCL cell lines (top). The
correlation betweenAID protein levels (f) and the level of IgV
genes ICV (o), assessed as in panel A, is shown in the
bottom panel for 6 representative lines. Numbers in parenthe-
ses indicate the mutation frequency in the IgV consensus
sequence.
BLOOD, 15 NOVEMBER 2004 䡠 VOLUME 104, NUMBER 10 AID PROTEIN EXPRESSION IN B CELLS 3323

derange the SHM mechanism in DLBCL.23,24 Based on RNA data,

AID appears selectively expressed in GC B cells and GC-derived
malignancies. However, the expression pattern of the AID protein
has never been shown, mainly due to unavailability of specific Abs.
This is the first study to report the distribution of the AID protein in
normal and transformed human B cells, its relationship with IgV
SHM, and its correlation with ASHM. Our data have implications
for the understanding of the AID function in general and its specific
role in lymphomagenesis.
The results of our study complement available data on AID
mRNA expression by showing a close correlation between AID
transcripts and protein levels in GC B cells and GC-derived
malignancies29,30 (data not shown). Thus, no major control mecha-
nism appears to be involved in the expression of AID at the level of
RNA translation or protein stability. Nonetheless, neither RNA nor
protein studies on bulk cell populations can address the issue of
AID expression in cell subpopulations. Therefore, our results
cannot determine whether AID is expressed in all GC B cells or in
selected subpopulations, a question requiring the use of immunohis-
tochemistry techniques that, unfortunately, could not be performed
with the 2 Abs described.
Figure 6. Lack of correlation between AID protein levels and ASHM in DLBCL. Consistent with its distribution in normal B-cell subpopulations,
(Top) AID protein expression levels in DLBCL samples displaying (F) or lacking (E)
the AID protein was not detected in B-CLL samples. B-CLL
ASHM, as defined by the presence of mutations in PIM1, PAX5, cMYC, and
RhoH/TTF. AID protein levels were measured by Western blot analysis and represents a post-GC–derived malignancy in at least 60% of the
normalized for ␤-actin as described in “Materials and methods.” (Bottom) Histograms cases, which carry somatically mutated IgV genes; moreover,
show the average expression levels in the 2 groups. Error bars indicate standard based on gene expression profiling analysis, all B-CLL cases
deviations.
appear to originate from a post-GC memory B cell.41 Our data are
in apparent contradiction with recent studies reporting expression
blot analysis on nuclear and cytoplasmic fractions prepared from of the AID mRNA, mostly in the IgV unmutated subgroup of the
10 AID-expressing cell lines, including 7 DLBCL and 3 BL. The disease.29,42,43 However, these studies were performed using re-
DLBCL lines were representative of both the ABC (Ly3, Ly10) and verse transcriptase (RT)–PCR methods, while more sensitive
the GCB (Ly7, Ly1, Ly18, RCK8) type and displayed ASHM in 6 quantitative assays, including limiting dilution assays, revealed that
of 7 cases. The results shown in Figure 7 indicate that—with the only 0.01% to 0.2% of the B-CLL tumor cells are AID positive44
exception of the ABC-like, ASHM-positive Ly10 cell line, where and that the total mRNA levels in B-CLL cases correspond to only
AID appears to be evenly distributed in the nuclear and cytoplas- about 5% of those observed in GC B cells,45 the population where
mic fraction—AID is mostly retained in the cytoplasm (more than AID typically exerts its activity. Thus, our data are consistent with
90%), and only trace amounts of the protein can be detected in the previous reports in that the bulk of the B-CLL cells do not express
nuclear fraction, analogous to what observed in normal GC B cells. AID. The biologic relevance of these low and very sparse AID
This distribution was observed regardless of the histologic type of mRNA expression levels in B-CLL remains to be proven.
lymphoma (BL versus DLBCL) and also of the evidence for Our results show that the AID protein is expressed—in hetero-
ongoing SHM (see CB and Ramos clone 6). Thus, in steady state geneous fashion—in all phenotypic subtypes of DLBCL (GC-like,
conditions, the AID protein appears to be localized mostly in the ABC-like, and type 3) and that the AID amounts are tendentially
cytoplasm, both in normal and transformed B cells. higher in the ABC type. Analogous results have been obtained by
comparing AID mRNA levels in larger DLBCL databases gener-
ated by gene expression profiling at different institutions (http://
Discussion llmpp.nih.gov/lymphoma). The presence of AID—in some cases at
supraphysiological levels—in the ABC DLBCLs was somehow
In the past few years, the AID protein has been the object of unexpected. In fact, this tumor subtype has been typically defined
extensive studies due to its primary role in SHM and CSR. As a by the down-regulation of GC-specific expression markers (eg,
mutator, AID has also been implicated in the pathogenesis of B-cell BCL6) and up-regulation of post-GC markers such as IRF4 and
lymphomas, and indeed its deregulated expression may represent a XBP1, overall suggesting a derivation from a later, post-GC stage
good candidate to explain the loss of target specificity that seems to of differentiation.38,46 In addition, both the SHM machinery and the

Figure 7. Predominant cytoplasmic localization of the AID

protein in BL and DLBCL. Nuclear (N) and cytoplasmic (C)
fractions were prepared from 10 AID-positive lymphoma cell lines,
including 3 BL and 7 DLBCL, and immunoblotted with anti-AID
Abs, followed by anti-BCL6 (nuclear) and anti–␣-tubulin (cytoplas-
mic) as controls for the purity of the fractions. The presence or
absence of ASHM is also given; nd indicates not determined.
3324 PASQUALUCCI et al BLOOD, 15 NOVEMBER 2004 䡠 VOLUME 104, NUMBER 10

CSR mechanism have been presumably shut off in this disease cytoplasm is in disagreement with its presumed activity as a
type, because the tumor cells lack intraclonal diversity in their mutator, whether directly on DNA or on pre-mRNA, because the
rearranged IgV genes (our data and Lossos et al47) and express high RNA-editing process is also thought to take place in the nucleus.52
levels surface IgM.38,39,48 Thus, the presence of AID in the ABC While we cannot exclude the possibility that the biochemical
DLBCL subgroup may suggest deregulation of AID expression; procedure used may have allowed some leakage of AID into the
alternatively, this may reflect an activation status or the origin from cytoplasm, it is also conceivable that very small amounts of AID in
a distinct B-cell subpopulation characterized by the indicated the nucleus at any given time may be sufficient to exert its activity.
phenotypic features, possibly at the exit of the GC. Because the Nevertheless, one may wonder why these cells would produce such
presently available Abs are not suitable for immunohistochemistry/ a large amount of AID when not necessary. A more plausible
immunofluorescence studies of AID, this question remains explanation is that nuclear/cytoplasmic shuttling of AID constitutes
unanswered. a major point of regulation in its function, where sequestration of
We could not find a correlation between AID expression levels AID in the cytoplasm would serve as an active mechanism to
and evidence of ASHM in the DLBCL primary cases analyzed. protect genetic information from uncontrolled mutational activity.
Considering that no structural abnormalities of the AID coding Indeed, it has been recently reported that a GFP-AID fusion protein
sequence are present in DLBCL, based on DNA sequencing can accumulate in the nucleus of stably transfected cells following
data,23,29 several possibilities may explain this finding: (1) AID may treatment with the inhibitor of chromosome region maintenance/
have acted at earlier stages, before the biopsy was taken; however, exportin 1 (CRM-1)–dependent nuclear export leptomycin B
tumors like BL consistently express high levels of AID but do not (LMB).50,51 Notably, mutants that accumulate in the nucleus were
show evidence of ASHM23; (2) because the extent of the ASHM associated with higher mutation activity on cotransfected artificial
target is not known, it is possible that cases presently classified as target constructs, while mutants that could not shuttle to the nucleus
“unmutated” may in fact carry mutations in genes other than the upon LMB treatment lost their activities in both SHM and CSR.50,51
ones identified so far and therefore represent “false negatives”; and These results suggest that AID nuclear localization quantitatively
(3) regulation of AID activity (eg, gene target recognition and correlates with its biologic function. Further studies aimed at the
accessibility; shuttling of the protein to the target sequence) rather identification of signals controlling AID subcellular localization are
than the absolute expression levels may be implicated in ASHM necessary in order to understand its regulation in GC B cells and its
(see next paragraph). For instance, the ASHM defect may reside possible deregulation in DLBCL.
primarily in other molecules associated with AID.
One perhaps surprising finding of this study is represented by
the predominant cytoplasmic localization of the AID protein in all
samples analyzed, including CBs. This result is in agreement with Acknowledgments
the original report by Rada et al, who used confocal microscopy to
detect an exogenous GFP-AID fusion protein stably transfected in We thank U. Klein for the isolation of the normal B-cell subsets, V.
the Ramos B-cell line,49 and with 2 more recent studies also Miljkovic and A. Babiac for assistance in DNA sequencing, Claude-
performed by transduction of GFP-AID retroviral constructs in B Agnès Reynaud for kindly providing us with the anti-AID polyclonal
and non-B cells.50,51 However, the localization of AID in the antibody, and A. Martin for the Ramos clone 6 and clone 1.

References
1. Jacob J, Kelsoe G, Rajewsky K, Weiss U. Intra- 10. Peng HZ, Du MQ, Koulis A, et al. Nonimmuno- tion in class-switch recombination. Proc Natl
clonal generation of antibody mutants in germinal globulin gene hypermutation in germinal center B Acad Sci U S A. 2003;100:2634-2638.
centres. Nature. 1991;354:389-392. cells. Blood. 1999;93:2167-2172. 18. Petersen-Mahrt SK, Harris RS, Neuberger MS.
2. Liu YJ, de Bouteiller O, Arpin C, et al. Normal hu- 11. Müschen M, Re D, Jungnickel B, Diehl V, Rajew- AID mutates E. coli suggesting a DNA deamina-
man IgD⫹IgM⫺ germinal center B cells can ex- sky K, Küppers R. Somatic mutation of the CD95 tion mechanism for antibody diversification. Na-
press up to 80 mutations in the variable region of gene in human B cells as a side-effect of the ger- ture. 2002;418:99-103.
their IgD transcripts. Immunity. 1996;4:603-613. minal center reaction. J Exp Med. 2000;192: 19. Chaudhuri J, Tian M, Khuong C, Chua K, Pinaud
3. Rajewsky K. Clonal selection and learning in the 1833-1840. E, Alt FW. Transcription-targeted DNA deamina-
antibody system. Nature. 1996;381:751-758. 12. Gordon MS, Kanegai CM, Doerr JR, Wall R. So- tion by the AID antibody diversification enzyme.
4. Peters A, Storb U. Somatic hypermutation of im- matic hypermutation of the B cell receptor genes Nature. 2003;422:726-730.
munoglobulin genes is linked to transcription ini- B29 (Igbeta, CD79b) and mb1 (Igalpha, CD79a). 20. Pham P, Bransteitter R, Petruska J, Goodman
tiation. Immunity. 1996;4:57-65. Proc Natl Acad Sci U S A. 2003;100:4126-4131. MF. Processive AID-catalysed cytosine deamina-
5. Tumas-Brundage K, Manser T. The transcrip- 13. Revy P, Muto T, Levy Y, et al. Activation-induced tion on single-stranded DNA simulates somatic
tional promoter regulates hypermutation of the cytidine deaminase (AID) deficiency causes the hypermutation. Nature. 2003;424:103-107.
antibody heavy chain locus. J Exp Med. 1997; autosomal recessive form of the hyper-IgM syn- 21. Ramiro AR, Stavropoulos P, Jankovic M, Nussen-
185:239-250. drome (HIGM2). Cell. 2000;102:565-575. zweig MC. Transcription enhances AID-mediated
6. Fukita Y, Jacobs H, Rajewsky K. Somatic hyper- 14. Muramatsu M, Kinoshita K, Fagarasan S, et al. cytidine deamination by exposing single-stranded
mutation in the heavy chain locus correlates with Class switch recombination and hypermutation DNA on the nontemplate strand. Nat Immunol.
transcription. Immunity. 1998;9:105-114. require activation-induced cytidine deaminase 2003;4:452-456.
7. Yelamos J, Klix N, Goyenechea B, et al. Targeting (AID), a potential RNA-editing enzyme. Cell. 22. Dickerson SK, Market E, Besmer E, Papavasiliou
of non-Ig sequences in place of the V segment by 2000;102:553-563. FN. AID mediates hypermutation by deaminating
somatic hypermutation. Nature. 1995;376:225- 15. Yoshikawa K, Okazaki IM, Eto T, et al. AID en- single stranded DNA. J Exp Med. 2003;197:1291-
229. zyme-induced hypermutation in an actively tran- 1296.
8. Pasqualucci L, Migliazza A, Fracchiolla N, et al. scribed gene in fibroblasts. Science. 2002;296: 23. Pasqualucci L, Neumeister P, Goossens T, et al.
BCL-6 mutations in normal germinal center B 2033-2036. Hypermutation of multiple proto-oncogenes in
cells: evidence of somatic hypermutation acting 16. Okazaki IM, Kinoshita K, Muramatsu M, Yo- B-cell diffuse large-cell lymphomas. Nature.
outside Ig loci. Proc Natl Acad Sci U S A. 1998; shikawa K, Honjo T. The AID enzyme induces 2001;412:341-346.
95:11816-11821. class switch recombination in fibroblasts. Nature. 24. Gaidano G, Pasqualucci L, Capello D, et al. Aber-
9. Shen HM, Peters A, Baron B, Zhu X, Storb U. Mu- 2002;416:340-345. rant somatic hypermutation in multiple subtypes
tation of BCL-6 gene in normal B cells by the pro- 17. Doi T, Kinoshita K, Ikegawa M, Muramatsu M, of AIDS-associated non-Hodgkin lymphoma.
cess of somatic hypermutation of Ig genes. Sci- Honjo T. De novo protein synthesis is required for Blood. 2003;102:1833-1841.
ence. 1998;280:1750-1752. the activation-induced cytidine deaminase func- 25. Montesinos-Rongen M, Van Roost D, Schaller C,
BLOOD, 15 NOVEMBER 2004 䡠 VOLUME 104, NUMBER 10 AID PROTEIN EXPRESSION IN B CELLS 3325

Wiestler OD, Deckert M. Primary diffuse large tion. Proc Natl Acad Sci U S A. 2003;100:2639- 44. Albesiano E, Messmer BT, Damle RN, Allen SL,
B-cell lymphomas of the central nervous system 2644. Rai KR, Chiorazzi N. Activation-induced cytidine
are targeted by aberrant somatic hypermutation. 35. Faili A, Aoufouchi S, Gueranger Q, et al. AID-de- deaminase in chronic lymphocytic leukemia B
Blood. 2004;103:1869-1875. pendent somatic hypermutation occurs as a DNA cells: expression as multiple forms in a dynamic,
26. Gregory MA, Hann SR. c-Myc proteolysis by the single-strand event in the BL2 cell line. Nat Immu- variably sized fraction of the clone. Blood. 2003;
ubiquitin-proteasome pathway: stabilization of nol. 2002;3:815-821. 102:3333-3339.
c-Myc in Burkitt’s lymphoma cells. Mol Cell Biol. 36. Niu H, Ye BH, Dalla-Favera R. Antigen receptor 45. McCarthy H, Wierda WG, Barron LL, et al. High
2000;20:2423-2435. signaling induces MAP kinase-mediated phos- expression of activation-induced cytidine deami-
27. Bahram F, von der Lehr N, Cetinkaya C, Larsson phorylation and degradation of the BCL-6 tran- nase (AID) and splice variants is a distinctive fea-
LG. c-Myc hot spot mutations in lymphomas re- scription factor. Genes Dev. 1998;12:1953-1961. ture of poor-prognosis chronic lymphocytic leuke-
sult in inefficient ubiquitination and decreased 37. Rowe M, Khanna R, Jacob CA, et al. Restoration mia. Blood. 2003;101:4903-4908.
proteasome-mediated turnover. Blood. 2000;95: of endogenous antigen processing in Burkitt’s 46. Wright G, Tan B, Rosenwald A, Hurt EH, Wiestner
2104-2110. lymphoma cells by Epstein-Barr virus latent mem- A, Staudt LM. A gene expression-based method
28. Yeh E, Cunningham M, Arnold H, et al. A signal- brane protein-1: coordinate up-regulation of pep-
to diagnose clinically distinct subgroups of diffuse
ling pathway controlling c-Myc degradation that tide transporters and HLA-class I antigen expres-
large B cell lymphoma. Proc Natl Acad Sci U S A.
impacts oncogenic transformation of human cells. sion. Eur J Immunol. 1995;25:1374-1384.
2003;100:9991-9996.
Nat Cell Biol. 2004;6:308-318. 38. Alizadeh AA, Eisen MB, Davis RE, et al. Distinct
types of diffuse large B-cell lymphoma identified 47. Lossos IS, Alizadeh AA, Eisen MB, et al. Ongoing
29. Greeve J, Philipsen A, Krause K, et al. Expres- immunoglobulin somatic mutation in germinal
by gene expression profiling. Nature. 2000;403:
sion of activation-induced cytidine deaminase in center B cell-like but not in activated B cell-like
503-511.
human B-cell non-Hodgkin lymphomas. Blood. diffuse large cell lymphomas. Proc Natl Acad Sci
2003;101:3574-3580. 39. Rosenwald A, Wright G, Chan WC, et al. The use
U S A. 2000;97:10209-10213.
of molecular profiling to predict survival after che-
30. Smit LA, Bende RJ, Aten J, Guikema JE, Aarts 48. Shaffer AL, Rosenwald A, Staudt LM. Lymphoid
motherapy for diffuse large-B-cell lymphoma.
WM, van Noesel CJ. Expression of activation- malignancies: the dark side of B-cell differentia-
N Engl J Med. 2002;346:1937-1947.
induced cytidine deaminase is confined to B-cell tion. Nat Rev Immunol. 2002;2:920-932.
non-Hodgkin’s lymphomas of germinal-center 40. Zhang W, Bardwell PD, Woo CJ, Poltoratsky V,
phenotype. Cancer Res. 2003;63:3894-3898. Scharff MD, Martin A. Clonal instability of V re- 49. Rada C, Jarvis JM, Milstein C. AID-GFP chimeric
gion hypermutation in the Ramos Burkitt’s lym- protein increases hypermutation of Ig genes with
31. Hardianti MS, Tatsumi E, Syampurnawati M, et al. phoma cell line. Int Immunol. 2001;13:1175-1184. no evidence of nuclear localization. Proc Natl
Activation-induced cytidine deaminase expres-
41. Klein U, Tu Y, Stolovitzky GA, et al. Gene expres- Acad Sci U S A. 2002;99:7003-7008.
sion in follicular lymphoma: association between
sion profiling of B cell chronic lymphocytic leuke-
AID expression and ongoing mutation in FL. Leu- 50. Ito S, Nagaoka H, Shinkura R, et al. Activation-
mia reveals a homogeneous phenotype related to
kemia. 2004;18:826-831. induced cytidine deaminase shuttles between
memory B cells. J Exp Med. 2001;194:1625-
32. Houldsworth J, Olshen AB, Cattoretti G, et al. Re- nucleus and cytoplasm like apolipoprotein B
1638.
lationship between REL amplification, REL func- mRNA editing catalytic polypeptide 1. Proc Natl
42. Oppezzo P, Vuillier F, Vasconcelos Y, et al. Acad Sci U S A. 2004;101:1975-1980.
tion, and clinical and biologic features in diffuse Chronic lymphocytic leukemia B cells expressing
large B-cell lymphomas. Blood. 2004;103:1862- AID display dissociation between class switch 51. McBride KM, Barreto V, Ramiro AR, Stavropoulos
1868. recombination and somatic hypermutation. Blood. P, Nussenzweig MC. Somatic hypermutation is
33. Pasqualucci L, Neri A, Baldini L, Dalla-Favera R, 2003;101:4029-4032. limited by CRM1-dependent nuclear export of
Migliazza A. BCL-6 mutations are associated with activation-induced deaminase. J Exp Med. 2004;
43. Heintel D, Kroemer E, Kienle D, et al. High ex-
immunoglobulin variable heavy chain mutations 199:1235-1244.
pression of activation-induced cytidine deami-
in B-cell chronic lymphocytic leukemia. Cancer nase (AID) mRNA is associated with unmutated 52. Wedekind JE, Dance GS, Sowden MP, Smith HC.
Res. 2000;60:5644-5648. IGVH gene status and unfavourable cytogenetic Messenger RNA editing in mammals: new mem-
34. Klein U, Tu Y, Stolovitzky GA, et al. Transcrip- aberrations in patients with chronic lymphocytic bers of the APOBEC family seeking roles in the
tional analysis of the B cell germinal center reac- leukaemia. Leukemia. 2004;18:756-762. family business. Trends Genet. 2003;19:207-216.