Negative Feedback Regulation of Antigen
Receptors through Calmodulin Inhibition of
E2A
This information is current as
of June 18, 2017.
Jiyoti Verma-Gaur, Jannek Hauser and Thomas Grundström
J Immunol published online 11 May 2012
http://www.jimmunol.org/content/early/2012/05/11/jimmun
ol.1103105
http://www.jimmunol.org/content/suppl/2012/05/11/jimmunol.110310
5.DC1
Subscription
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Permissions
Email Alerts
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 2012 by The American Association of
Immunologists, Inc. All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
Supplementary
Material
Published May 11, 2012, doi:10.4049/jimmunol.1103105
The Journal of Immunology
Negative Feedback Regulation of Antigen Receptors through
Calmodulin Inhibition of E2A
Jiyoti Verma-Gaur,1 Jannek Hauser,1 and Thomas Grundström
Signaling from the BCR is used to judge Ag-binding strengths of the Abs of B cells. BCR signaling enables the selection for
successive improvements in the Ag affinity over an extremely broad range of affinities during somatic hypermutation. We show
that the mouse BCR is subject to general negative feedback regulation of the receptor proteins, as well as many coreceptors and
proteins in signal pathways from the receptor. Thus, the BCR can downregulate itself, which can enable sensitive detection of
successive improvements in the Ag affinity over a very large span of affinities. Furthermore, the feedback inhibition of the BCR
signalosome and most of its proteins, as well as most other regulations of genes by BCR stimulation, is to a large extent through
inhibition of the transcription factor E2A by Ca2+/calmodulin. The Journal of Immunology, 2012, 188: 000–000.
regulation. Activation of signaling leads to downregulation of the
receptor proteins, many coreceptors and proteins participating in
signal pathways from the receptor. Thus, the BCR can downregulate itself, which can enable sensitive detection of successive
improvements in Ag affinity of the receptor over a very large span
of affinities during SH. We also show that the mechanism of the
feedback inhibition of the BCR signalosome and most of its proteins, as well as most other regulations of genes by BCR stimulation, is to a large extent through inhibition of the transcription
factor E2A by Ca2+/calmodulin (CaM).
Materials and Methods
Activation and infection of B lymphocytes from mouse spleens
1
Primary B lymphocytes were purified from mouse spleens, maintained as
previously described (5), and stimulated with 5 ng/ml IL-4 (PeproTech) and
either 200 ng/ml CD40L (R&D Systems) or 10 mg/ml LPS (Calbiochem) for
48 h. The BCR of stimulated cells at 106 cells/ml, where indicated, was
activated by incubation with goat F(ab9)2 anti-mouse IgM at 2.5 mg/ml (5 mg/
ml for Fig. 3A and 3B, and various concentrations, as indicated, for Fig. 3C)
for the indicated times. The MSCV-IRES-GFP–based retroviruses encoding
wild-type and CaM-resistant m8N47 mutant of the E12 splice form of mouse
E2A were described previously (5). Retrovirus concentrated by centrifugation was added with 5 mg/ml polybrene to 0.5 3 106 purified B cells after
activation with CD40L plus IL-4 for 14 h (Figs. 4, 6B) or 24 h (Fig. 7). After
12 h of incubation, the infection was repeated for 12 h, followed by incubation for an additional 22 h postinfection in fresh complete medium with the
stimulants to allow for expression of GFP and E12. In addition to CD40L plus
IL-4, the medium was supplemented with LPS (200 ng/ml) during retroviral
infection incubations to enable efficient infection. Where indicated, antimouse IgM (2.5 mg/ml) was added for the indicated times. For DNA
microarray and Western blot analyses of infected B lymphocytes, the B cells
were purified from spleens of mice heterozygous for knockout of the E2A
gene (6). Infected cells were purified for DNA microarray and real-time RTPCR analyses with a FACSAria cell sorter (BD Biosciences), using their
green fluorescence from expression of the GFP.
Received for publication November 1, 2011. Accepted for publication April 12, 2012.
DNA microarray analysis
This work was supported by grants from the Swedish Research Council and the
Swedish Cancer Society.
Gene expression changes upon Ag receptor activation with anti-mouse IgM
were analyzed using the Illumina BeadChip system (Illumina, San Diego,
CA). For in vitro transcription amplification, 200 ng RNA from the time
course and 100 ng RNA from the comparison of gene expression in B cells
expressing wild-type and CaM-resistant E12 was used with the Illumina
RNA Amplification Kit (Ambion). Amplified RNA (1.5 mg) was hybridized
to the Sentrix MouseRef-6 Expression BeadChip containing 47,667 probes.
The primary data were collected from the BeadChip using the manufacturer’s BeadArray Reader and analyzed using the supplied scanner
software. Data normalization was performed by cubic spline normalization
using Illumina’s BeadStudio v3 software; fold changes $ 1.5 with p ,
0.05 were considered significant. Clustering of certain selected genes was
Department of Molecular Biology, Umeå University, SE-901 87 Umeå, Sweden
J.V.-G. and J.H. contributed equally to this work.
The datasets presented in this article have been submitted to the National Center for
Biotechnology Information Gene Expression Omnibus (http://www.ncbi.nlm.nih.
gov/geo) under accession numbers GSE35747 and GSE35748.
Address correspondence and reprint requests to Prof. Thomas Grundström, Department of Molecular Biology, Umeå University, SE-901 87 Umeå, Sweden. E-mail
address: [email protected]
Abbreviations used in this article: CaM, calmodulin; GC, germinal center; PLA,
proximity-ligation assay; SH, somatic hypermutation.
Copyright Ó 2012 by The American Association of Immunologists, Inc. 0022-1767/12/$16.00
www.jimmunol.org/cgi/doi/10.4049/jimmunol.1103105
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
D
uring B lymphocyte development, the cells have to pass
several checkpoints verifying the functionality of their
Ag receptors. The expression of a BCR with membranebound Ig is used for selection of cells for receptor editing, permission to leave the bone marrow, extrafollicular plasma cell
differentiation, and somatic hypermutation (SH) in the germinal
center (GC) reaction, as well as for selection against autoreactive
cells (1–4). To optimize the initial wave of Ab defense against
infections, the primary B cells with the highest Ag affinity are
selected to generate the extrafollicular plasmablast responses. B
cells with somewhat weaker Ag binding are selected for the GC,
where they undergo SH and selection for improved Ag affinity (1–
3). High-affinity B cell clones from the GC typically have 5–10
mutations in the variable regions of their Ig genes. They have
increased their Ag-binding strength by many orders of magnitude
(from ∼106 M21 or less to ∼1010 M21 and higher) through several
rounds of SH. Each round leads to one or a few mutations, followed by selection for increased affinity; after a number of such
cycles, the B cells with highest Ag affinity are selected for plasma
cell differentiation, resulting in mass production of the final
high-affinity Ab (1–3). Therefore, BCR signaling has to detect
improvements in Ag affinity over an extremely broad range of
affinities. The receptor should detect differences in affinity, even
when relatively small (e.g., for Ag-binding strengths ∼106 M21),
as well as for Ag-binding strengths to $1010 M21. In this study,
we examined whether this ability could be due, at least in part, to
an effect of activation of BCR signaling on the BCR signalosome
itself. We show that the BCR is subject to negative feedback
2
FEEDBACK INHIBITION OF Ag RECEPTORS BY CALMODULIN
done using Cluster 3 software, and data were visualized using TreeView
v1.2 software. All raw .CEL dataset files are available at http://www.ncbi.
nlm.nih.gov/geo under accession numbers GSE15606, GSE35747, and
GSE35748.
Real-time RT-PCR
FACS analysis and Western blot
Intracellular immunostaining of harvested purified mouse B cells was done
with 2% paraformaldehyde and ethanol, as previously described (8). The
immunostainings for FACS analysis were performed using the following
primary Abs: anti-IgM, goat-PE (1021-09; Southern Biotech); anti-Igl1,
rat (R11-153; BD Biosciences); anti-Igk, goat-FITC (1050-06; Southern
Biotech); anti-CD19, mouse (MB19-1; Santa Cruz Biotechnology); antiCD21, mouse (A-3; Santa Cruz Biotechnology); anti-CD22, goat (M-20;
Santa Cruz Biotechnology); anti-CD72, mouse (G-5; Santa Cruz Biotechnology); anti-CD81, rabbit (H-121; Santa Cruz Biotechnology); antiCD79a, mouse-FITC (LS-C44953; Lifespan Biosciences); anti-CD79b,
hamster-PE (LS-C43941; Lifespan Biosciences); anti-BLK, rabbit (K-23;
Santa Cruz Biotechnology); anti-BLNK, rabbit (H-80; Santa Cruz Biotechnology); anti-BTK, rabbit, raised against a GST-fusion protein of the
N-terminal part of human BTK by Dr. J. Forssell, Umeå University; antiGAB2, goat (M-19; Santa Cruz Biotechnology); anti-PKCb1, rabbit (C16; Santa Cruz Biotechnology); anti-PLCg2, rabbit (Q-20; Santa Cruz
Biotechnology); anti-LYN, rabbit (H-70; Santa Cruz Biotechnology); and
anti-SYK, rabbit (C-20; Santa Cruz Biotechnology).
The secondary Abs were donkey anti-rabbit IgG (H+L)-FITC, donkey
anti-rabbit IgG (H+L)-PE, donkey anti-rabbit IgG (H+L)-allophycocyanin,
goat anti-rat IgG (H+L)-allophycocyanin, donkey anti-mouse IgG (H+L)PE, and donkey anti-mouse IgG (H+L)-allophycocyanin [all F(ab9)2
Proximity-ligation assay
Harvested lymphocytes were fixed with 4% paraformaldehyde, permeabilized with 0.2% Triton X-100, and blocked with 5% FCS in PBS.
Proximity ligation assay (PLA) (10–13) was performed with the Duolink
system using in situ PLA kits from Olink Bioscience. In brief, cells were
stained for intracellular proteins using one primary mouse or goat Ab that
was biotinylated using an Ab biotinylation kit (Miltenyi Biotec) together
with a primary rabbit Ab. The Abs used were the same as for the determination of protein levels by FACS, except that the goat anti-CD19 Ab C20, the mouse anti-SYK Ab SYK-01, and the mouse anti-CD79B (antiIgb) Ab B29/123 (all from Santa Cruz Biotechnology) were used. Antibiotin IgG and anti-rabbit IgG secondary Abs conjugated with oligonucleotides (PLA probes) were subsequently used, according to the manufacturer’s protocol, to generate fluorescence signals only when the two
PLA probes were in close proximity (#40 nm). The fluorescence signal
from each detected pair of PLA probes was visualized as a distinct individual red dot (10–13). Nuclei were counterstained with Hoechst 33342
dye. To detect infected cells, anti–GFP-FITC (Novus Biologicals) Ab was
added during the PLA signal-detection step.
Results
Receptor activation downregulates BCR signaling components
We recently performed DNA microarray analysis of the genomewide transcriptional changes in 3 h when stimulating the BCR
of purified mouse splenic B cells activated to SH by CD40L plus
IL-4 (14). The expression of a large set of genes (.4600) differed
significantly between the BCR-stimulated and nonstimulated
B cells, and the numbers of upregulated and downregulated genes
were approximately equal, at 7% of the total number of genes (14).
Interestingly, the DNA microarray data for receptor components,
coreceptors, and downstream signaling factors of the BCR showed
a reduction in the expression of almost all these genes after the
BCR stimulation of the cells (Fig. 1). This reduction indicates very
broad negative feedback regulation of almost the entire receptorsignaling network of the BCR at the mRNA level. This general
downregulation was specific to the BCR and not a general property
of all of the receptors and signaling proteins, because analysis of
other signaling networks (Jak/Stat signaling, TLR signaling, receptor tyrosine kinases, TGF-b signaling, Erb/HER signaling, and
BAFF receptor signaling) showed no significant effect of BCR
stimulation on the expression of the vast majority of those genes.
Furthermore, and also in sharp contrast with the BCR-signaling
network, upregulations and downregulations were approximately
equally common among the genes that did change expression in
those other signaling networks.
To examine how fast the Ag receptor-driven transcriptional
changes of genes for BCR-signaling proteins occur, we performed
a time course analysis of activated primary splenic B cells after
BCR activation. The mRNA expression levels of almost all analyzed BCR proteins, coreceptors, and downstream-signaling factors decreased progressively (Fig. 2). The mRNA levels for IgH
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
Total RNA was extracted using TRIzol reagent (Invitrogen), and real-time
PCR analysis was performed, as previously described, using GAPDH as an
internal control (5). The specificity of the real-time PCR was analyzed by
melt-curve analysis. Some of the primer pairs used were described previously (7); the primer pairs that were not described previously were as
follows:
Igh-6: 59-ACAACCAGAAGTTCAAGGGCAAGG-39 and 59-ATCGAAGTACCAGTGGGACCGTAG-39; VpreB3: 59-AGCAGCAGCCAGGAAGTG-39 and 59-ACAGAAGTAATCGGCATCATCC-39; Igl-v1: 59-AGGTGATGTGGTGAGTGCAGATTCCTGAGTC-39 and 59-CGCATCTTGTCTCTGATTTGCTACTGATGACTGGATTTC-39; Igk-v3: 59-AGAAACCAGGACAGCCACCCAAACTC-39 and 59-GTCCCAGACCCACTGCCACTAAACC-39; Iga: 59-CCTTCCCAGTGCTTGCTCCAG-39 and
59-TGACACTAACGAGGCTGCTGTAATC-39; Igb: 59-GGAGGTAAAGTGGTCGGTAGGAGAG-39 and 59-GGCTAGAATGAGAGGTGGTATTGATGG-39; Cd19: 59-ACAGGACTGGAAGAAGAAG-39 and 59ACTGAATTGAGTGGAGCTG-39; Cd20: 59-TCAAACTTCCAAGCCGTATG-39 and 59-GAAGAAGGCAGAGATCAGCA-39; Cd21: 59-AGGCTTCTTTAGGGTGTCAGTCTCC-39 and 59-TGTCATCCTGGCACTGGCTCTG-39; Cd22: 59-TGTCCTGTGAGAGTGATGCCAATCC-39 and 59GTGAGGGTGCTGGGTGGTGAC-39; Cd72: 59-GCAGGTGTCTCGGCAGTTCC-39 and 59-AGCGTCCTCGTGAGTCCTCTG-39; Cd81: 59-TCATTCTCCGCAACGCAGTTC-39 and 59-AGGCTGGTGGTCTGTGGATC-39; Syk: 59-GCAGCAGAACAGGCACATTAAGGATAAGAAC-39
and 59-GCACGCAGGGCTTTGGAAAGACCGAAATCATCATAGAG-39;
Lyn: 59-CCAAGGAGGAGCCCATCTACATC-39 and 59-TTGATAGTGAAGCAGCCGAAGTTG-39; Btk: 59-ACCAAACAACGCCCCATCTTCATCATCAC-39 and 59-CGACTCCAAGTATTCCATTGCTTCACAGACATC-39; Blk: 59-GAGCAGCAAGAGGCAGGTCAG-39 and 59-TCACAGCGGCATAGTCAAACAGAG-39; Blnk: 59-AGGCTGAACTGCGTAAG-39 and 59-GTGTACGGCTGCTTGGAATC-39; Plcg2: 59-ACGCCCTCATCCAGCACTAC-39 and 59-GATCCGCATCAGCATGTCCTC-39;
Prkcb: 59-CGAGCGAGGGCGAGGAGAG-39 and 59-AGCAGCAGACTTGACACTGGAATC-39; Gab2: 59-GTACTGCCGTCCTATCTCCACAC-39
and 59-CCGCACATCTGTCCACTCCTG-39; Jnk2: 59-ACAGTGTGCAGGTGGCGGACTCAAC-39 and 59-GGCTCTCTTTGCGTGCGTTTGGTTCTG-39; p38a: 59-GGCGGACCTGAACAACATCGTGAAGTG-39 and
59-CAGTGTGCCGAGCCAGCCCAAAATC-39; p38b: 59-CGCCGACCTGAATAACATCGTCAAGTG-39 and 59-GCAGTCCTCGTTCACCGCTACATTG-39; p38g: 59-TGAGGAATGGAAGCGTGTGACTTACAAGGAAG-39 and 59-AGAAGGGAAGCCAAGGTGCCGTGAAG-39; Erk: 59CTTCCAACCTCCTGCTGAACACCACTTG-39 and 59-AGGATGCAGCCCACAGACCAAATATCAATG-39; Shp-1: 59-CTGGGCACTGTGTGAAGTCT-39 and 59-CCTGGAATGTTCTTGAGGGT-39; Ship-1: 59-GGAACATTCGCATAGTGGTG-39 and 59-CATGAAGGACACTCCCACTG39; and Pten: 59-CCAGTCAGAGGCGCTATGTA-39 and 59-CGCCACTGAACATTGGAATA-39.
fragments were from Jackson ImmunoResearch]. When using primary Ab
biotinylated using an Ab biotinylation kit (Miltenyi Biotec), the secondary
Ab was streptavidin-PerCP (BD Pharmingen) or streptavidin-allophycocyanin
(eBioscience). Stainings were done for $30 min at room temperature in
the dark. Flow cytometry was performed with a FACSCalibur, and analysis
was done using CellQuest software (Becton Dickinson Biosciences).
Nuclear extracts were prepared, as previously described (9), and Western
blot of purified mouse B cells was performed using the SuperSignal Chemiluminescence Substrate (Pierce), according to the manufacturer’s instructions. The Western blot analyses were performed using the following Abs:
anti-BTK, mouse (E-9; Santa Cruz Biotechnology); anti-p38a, rabbit
(C-20; Santa Cruz Biotechnology); anti-JNK/SAPK, rabbit (56G8; Santa
Cruz Biotechnology); anti-ERK1/2, rabbit (anti-p44/42 MAPK, 137F5;
Cell Signaling); anti–phospho-p38 MAPK, rabbit (D3F9; Cell Signaling);
anti–phospho-JNK/SAPK, rabbit (Cell Signaling); anti–phospho-ERK1/2,
rabbit (anti–phospho-p44/42 MAPK; Cell Signaling); and anti-a-tubulin,
mouse (B-5-1-2; Sigma).
The Journal of Immunology
3
We performed flow cytometry analyses to determine whether
the rapid reductions in the mRNA levels of analyzed genes after
BCR activation resulted in corresponding changes in the protein
levels. Most mRNA reductions led, with a delay, to reductions
in the protein (Fig. 3A). At 3 h, when the mRNA levels for most
genes were already reduced 50–90% (Figs. 1, 2), the protein levels
were rarely reduced .30%; several proteins showed very small
reductions or no reductions at all (Fig. 3A). At 5 h after BCR
activation, the protein levels were reduced 30–60%, with few
exceptions. CD81 was an exception for which the reduction in
mRNA was not followed by a reduction in protein. Surprisingly,
despite the increase in CD21 mRNA when activating the BCR
(Fig. 2), the CD21 protein level was reduced ∼2-fold. Thus, the
mechanism of CD21 downregulation after triggering the BCR
appears to be posttranscriptional. With regard to BLK, for which
the reduction in mRNA was small or absent (Figs. 1, 2), no significant reduction was found in the protein level (Fig. 3A).
BCR signaling reduces the sensitivity of the BCR
and the signaling proteins BLK and JNK2 decreased ,2-fold.
However, in most cases, the reductions were 2–10-fold after 3 h:
a minority of the reductions occurred within 1 h, but a majority of
the reductions occurred within 2 h (Fig. 2). The signaling protein
Shp-1 and the coreceptors CD21 and CD72 were exceptions
whose mRNA level was unchanged or even showed an increase
(Figs. 1, 2). Thus, activation of BCR signaling leads to negative
feedback regulation of almost all receptor-signaling components
at the mRNA level. To analyze whether the corresponding negative feedback regulations were seen when stimulating the BCR
of freshly isolated follicular B cells, we sorted follicular B cells
(CD19+ CD23hi CD21/35int) from mouse spleens and stimulated
the BCR with anti-IgM for 3 h (Supplemental Fig. 1A). Indeed,
BCR stimulation of these cells had the corresponding downregulation effect on the mRNA of the Ag receptor-signaling components (Figs. 1, 2, Supplemental Fig. 1A). We also analyzed the
effect of activation of the BCR on the mRNA level of CD19 after
activation of the splenic mouse B cells with LPS plus IL-4, instead
of CD40L plus IL-4 during the cultivation. BCR stimulation
resulted in the corresponding downregulation of CD19 mRNA
when using LPS instead of CD40L in the activation (Supplemental
Fig. 1B). These results show that the feedback inhibition is not
unique to B cells stimulated by cultivation with CD40L plus IL-4.
Genome-wide role of inhibition of E2A by Ca2+/CaM in B cells
after BCR activation
BCR activation leads to a signaling complex below the receptor
that rapidly produces a combination of Ca2+ signaling and a cascade of phosphorylations (15). To investigate whether Ca2+ sig-
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 1. Downregulation of Ag receptor-signaling components in
mouse B cells upon BCR activation. Cluster plots of genes for BCR
components, coreceptors, and signaling proteins during 3 h of anti-IgM
stimulation of the BCR of mouse B cells are shown. The conditions for
microarray analysis of genes differentially expressed during 3 h of antiIgM stimulation of the BCR of CD40L plus IL-4–activated B cells from
mouse spleen on Illumina BeadChip were published previously (5, 14).
The fold differential expression of genes in the individual mice is color
coded, with increases marked in red and decreases marked in green
(dataset available at http://www.ncbi.nlm.nih.gov/geo accession number
GSE15606).
To analyze whether there is indeed a negative feedback regulation
of Ag receptor function as indicated by the rapid reductions in
many mRNA and protein levels, we compared the activating
phosphorylation of JNK, ERK, and p38 proteins downstream of the
BCR after an initial BCR stimulation with anti-IgM and a second
stimulation 6 h later. These proteins were chosen because, in
contrast to most other BCR signalosome proteins, none or very
little reduction was seen in Jnk2 and Erk mRNA, and the different
p38 mRNA showed opposite responses to BCR stimulation (Fig. 1).
Thus, for these proteins, phosphorylation upon a second BCR
stimulation would show the ability of the whole receptor to
function the second time, rather than reflecting only the level of
the individual protein, provided that the protein level reflects the
mRNA level. The activating phosphorylation was strongly induced 15 and 30 min after the initial BCR stimulation for all three
proteins (Fig. 3B, quantifications in Supplemental Fig. 2). After 4
h with anti-IgM (5 mg/ml), followed by 2 additional hours of incubation without anti-IgM, the phosphorylation level of all three
was close to background (Fig. 3B, quantifications in Supplemental
Fig. 2). Importantly, after these 6 h, a second BCR stimulation was
unable to induce any of the activating phosphorylations. This was
not due to a corresponding reduction in the individual protein,
because JNK and ERK protein levels did not change, in contrast to
the BTK control (Fig. 3B, quantifications in Supplemental Fig. 2).
With regard to p38, for which only the p38a mRNA was reduced
(Figs. 1, 2), there was a reduction in p38a protein; however, it was
too small to account for the complete block of induction of
phospho-p38 upon the second BCR stimulation (Fig. 3B, quantifications in Supplemental Fig. 2). Thus, activation of the BCR
leads to a strong reduction in signaling through the receptor (see
also below).
Next, we analyzed whether BCR signaling reduces the sensitivity of the BCR. A weaker stimulation of the BCR (0.75 mg/ml
anti-IgM) inhibited its ability to respond to a second stimulation of
the same strength (Fig. 3C). Importantly, however, BCR signaling
was also induced the second time when the strength of the stimulation was increased (cf. BCR restimulations at concentrations
$0.75 mg/ml of anti-IgM in Fig. 3C). Thus, BCR signaling
reduces the sensitivity of the BCR.
4
FEEDBACK INHIBITION OF Ag RECEPTORS BY CALMODULIN
naling and/or serine protein kinases were essential for negative
feedback after BCR stimulation, we first analyzed the activation
in the presence of various inhibitors of the signaling pathways
for one selected downregulated gene, CD19. The Ca2+ chelator
BAPTA-AM completely blocked the reduction in CD19 mRNA
after BCR stimulation, and inhibitors of L-type or IP3 receptor
Ca2+ channels partially or completely blocked the effect, especially when combined (Supplemental Fig. 1C). Thus, Ca2+ signaling is essential, even though inhibitors of Ca2+/CaM-dependent
protein kinase (KN93) or the Ca2+/CaM-dependent phosphatase
calcineurin (cyclosporin A) did not block this negative feedback
(Supplemental Fig. 1C). The MAPK inhibitor PD98059, but not
the protein kinase C inhibitor bisindolylmaleimide, blocked the reduction in CD19 mRNA (Supplemental Fig. 1C), suggesting that, as
with Ca2+ signaling, MAPK signaling is important. We decided to
analyze the Ca2+-signaling dependence for the entire group of transcriptionally downregulated proteins.
Previous studies showed that the main Ca2+-sensor protein
CaM, when Ca2+ loaded, can bind to and inhibit transcriptional activities of E2A transcription factors (5, 7, 9, 14, 16, 17). The E2A
proteins have important roles in the transcriptional regulatory network that promotes commitment to and differentiation of B cells
(18, 19). We reported a series of mutants of the E2A isoform E12
that gains resistance to CaM to different extents (9). To elucidate
the genome-wide role of Ca2+/CaM inhibition of E2A in the regulatory effects of BCR activation, we used the CaM-resistant E12
mutant m8N47, which has three amino acid substitutions in the
CaM binding site and retains the sensitivity of E2A to Id protein
inhibitors (20). Purified primary splenic B cells from mice heterozygous for the E2A gene were infected with retroviruses expressing either wild-type E12 or CaM-resistant E12, together with
GFP. Infected B cells were purified by cell sorting based on their
GFP expression, and the BCR was activated for half of these
cells by anti-IgM treatment for 3 h. Comparison of the changes in
the DNA microarray profile upon BCR-stimulation of wild-type
E2A-infected B cells with those of noninfected B cells (Figs. 1, 2)
revealed a very strong overlap (data not shown), which validated
our experimental conditions. Comparison of the expression levels
of BCR-regulated genes in wild-type E12-infected B cells with
those of CaM-resistant E12-infected B cells showed that 83% of
the genes that were downregulated and 87% of the genes that were
upregulated when infected to express wild-type E12 lost .50% of
the BCR stimulation effect on their expression when infected to
express CaM-resistant E12 (Fig. 4A). Thus Ca2+/CaM inhibition
of E2A is needed for a majority of the regulatory effect for most of
the genes that are rapidly up- or downregulated by activation of
the BCR of the splenic B cells. In summary, Ca2+/CaM inhibition
of E2A is very broadly used to shift the transcriptional program
upon activation of the BCR of these cells.
Mechanism of feedback regulation of BCR-signaling
components
BCR activation downregulated the mRNA levels of most of the
BCR-signaling components in the cells expressing wild-type E12
as in the case of uninfected B cells (Figs. 2, 4B), although the
extent of this effect was reduced in some cases, presumably due
to the infection, the overexpression of wild-type E12, or both. In
contrast, the mRNA expression levels of Igl, Iga, Cd20, Cd22,
Cd72, Cd81, Lyn, Btk, Blk, Plcg2, Prkcb, Gab2, and p38a showed
no reduction after BCR activation in the infected cells expressing
CaM-resistant E12 (Fig. 4B). Thus, the downregulation of all of
these genes by BCR activation was dependent on CaM inhibition of E2A. For all other genes analyzed, with the exception of
BLNK, most of the reducing effect of anti-IgM on the expression
in infected cells expressing wild-type E12 had disappeared in the
mutant (Fig. 4B), indicating that the expression of these genes is
regulated, to a large extent, by CaM inhibition of E12. Taken together, these results suggest that CaM sensitivity of E2A regulates
the inhibitions of expression of most BCR-signaling components
after BCR activation.
Reduction of in vivo proximity of BCR-signaling components
after BCR activation depends on CaM sensitivity of E2A
BCR activation leads to formation of a BCR-associated protein
complex composed of tyrosine kinases, adaptor proteins, lipidmetabolizing enzymes, and serine/threonine kinases that is often
referred to as a “signalosome” (15, 21, 22). We followed the interactions between proteins within the BCR signalosome after
BCR activation using in situ PLA with the Duolink system, which
enables visualization and quantification of protein interactions in
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 2. Changes in expression of
different mRNAs upon stimulation of the Ag
receptor in B lymphocytes from mouse
spleen activated with CD40L plus IL-4
activators. The effects of BCR activation
with anti-IgM on the expression of the indicated mRNA were determined by microarray analysis, as in Fig. 1. With regard to
CD21, Lyn, and Gab2, for which the signals
from the DNA microarray analysis were
weak, and CD81, with a relatively wide
range of microarray measurements, the
mRNA levels were determined by quantitative RT-PCR and normalized using the
mRNA levels of GAPDH. The levels before
addition of anti-IgM were set at 100%. Data
are representative of three different mice
and are expressed as mean 6 SD.
The Journal of Immunology
5
microscopy. Primary splenic mouse B cells were stained using
Abs against different combinations of two BCR-signaling proteins and PLA probes (secondary Abs conjugated with oligonucleotides). This generates fluorescence signals, detected as distinct
individual red dots, only when the two PLA probe-marked proteins are in close proximity (#40 nm) (10–13). We obtained specific signals of proximity for many protein pairs, and the levels
of background signals were minimal when one or both of the
primary Abs were omitted (Fig. 5, Supplemental Fig. 3). As expected for the formation of a signalosome (15, 21, 22), BCR activation for 15 min increased the number of signal dots for many
protein pairs, including CD19-LYN, CD19-SYK, CD19-BLNK,
CD22-SYK, CD22-BLNK, CD22-PLCg2, SYK-LYN, Igb-BTK,
and Igb-PLCg2 (Fig. 5, Supplemental Fig. 3). For each of these
protein pairs, the increase in proximity was followed by a reduction (within 5 h) in prolonged anti-IgM treatment (Fig. 5, Supplemental Fig. 3). The increases in the number of dots at 15 min
after BCR activation ranged from 2–5-fold; after 5 h of anti-IgM
treatment, the numbers of dots were reduced 2–5-fold (Fig. 6A).
All of the p values for the reductions that occurred between 15
min and 5 h after BCR activation were p , 0.015, with the exception of CD22-SYK (p = 0.07) and SYK-LYN (p = 0.08). The
reductions in the CD22-SYK and SYK-LYN pairs were greater
and were associated with lower p values in the analyses of infected
cells below, a finding that supports that these pairs are also reduced. In summary, the results show that BCR activation after
the initial formation of the BCR signalosome rapidly reduces the
number of proteins interacting in the signalosome, which further
supports the notion of negative-feedback regulation of the BCR
signalosome.
To further study the mechanism of the negative feedback of the
BCR signalosome, we performed PLA for purified primary splenic
B cells infected with retrovirus expressing either wild-type or CaMresistant E12 protein, together with GFP, which enabled selective
analysis of the infected cells. The protein pairs of CD19-LYN,
CD19-SYK, CD19-BLNK, CD22-SYK, CD22-BLNK, CD22PLCg2, SYK-LYN, Igb-BTK, and Igb-PLCg2 showed 2–5-fold
decreases in the numbers of dots/cell between 15 min and 5 h after BCR activation for the primary B cells infected with virus
expressing wild-type E12 (Fig. 6B). The reductions were not sig-
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 3. Negative feedback regulation of BCR function and changes in expression of Ag receptor, coreceptors, and signaling proteins in B cells upon
receptor activation. (A) B cells from mouse spleen were activated with CD40L and IL-4 for 48 h, followed by addition of anti-IgM for the indicated times,
and expression of indicated receptor and signaling proteins was determined by intracellular immunostaining and flow cytometry. Results are mean immunostaining fluorescence intensity 6 SD, expressed as a percentage of the levels before the addition of anti-IgM. Data are representative of three different
experiments. (B) B cells from mouse spleen were activated with CD40L and IL-4 for 48 h, followed by the addition of anti-IgM for 15 or 30 min or 4 h.
After 4 h, anti-IgM was washed away; 2 h later, the B cells were restimulated with anti-IgM for 15 or 30 min. The levels of the induced phosphorylations
and the expression levels of the indicated signaling proteins were determined by Western blot analysis. A representative experiment is shown; average
results 6 SD are shown in Supplemental Fig. 2. a-Tubulin (a-tub) was used as loading control. (C) B cells from mouse spleen were activated with CD40L
and IL-4 for 48 h, followed by the addition of 0.75 mg/ml of anti-IgM for 15 min or 4 h. After 4 h, anti-IgM was washed away; 2 h later, the B cells were
restimulated with the indicated concentrations of anti-IgM for 15 min. The levels of the induced phosphorylations and the expression levels of the indicated
signaling proteins p38 and ERK at the indicated times were determined by Western blot analysis. A representative experiment and average phosphorylation
results 6 SD of three independent experiments are shown. a-Tubulin (a-tub) was used as a loading control.
6
FEEDBACK INHIBITION OF Ag RECEPTORS BY CALMODULIN
FIGURE 4. Losses of effect of BCR stimulation in B cells with CaMresistant E12. Purified primary splenic B cells from mice heterozygous for
deletion of the E2A gene that encodes E12 were infected with a retrovirus
expressing either wild-type or CaM-resistant E12, together with GFP, followed by BCR activation of half of the cells by anti-IgM treatment for 3 h.
Infected B cells (30–50% of the cells) were purified by cell sorting based
on their GFP expression, and microarray analysis of the B cells infected
with virus expressing wild-type and CaM-resistant E12, with or without
BCR stimulation, was performed with three independent mice using the
Illumina Mouse BeadChip system. (A) Overview of genome-wide loss of
effect of BCR stimulation in B cells with CaM-resistant E12. The bar
graph shows the percentage of the genes regulated by BCR stimulation
($1.33-fold change and p , 0.05) in B cells infected with wild-type E12
that lost their BCR regulation in B cells infected with CaM-resistant E12
mutant to the extent indicated. The analysis included 1598 upregulated
genes and 1684 downregulated genes. (B) Loss of inhibition of expression
of mRNA for BCR components, coreceptors, and signaling proteins upon
BCR stimulation when E12 is CaM resistant. The changes in expression
levels of key genes in BCR signaling were analyzed as described in (A).
The mRNA levels before the addition of anti-IgM were set as 100% in both
wild-type and CaM-resistant E12-infected B cells (not represented in the
bar graphs). The mRNA levels of Iga, Cd72, Cd81, Lyn, Blk, Blnk, Plcg2,
and Gab2 were determined by real-time RT-PCR as for some genes in Fig.
2, because the signals were not clearly above background for Lyn and
Gab2, and the spread of the data was too great for the other six genes in the
DNA microarray analysis. Data are representative of three different mice
and are expressed as mean 6 SD.
nificantly different between noninfected and wild-type–infected
cells for any of the nine protein pairs (Fig. 6). Also, the largest
difference, from 50 6 24% to 79 6 13% for CD22-SYK, was not
statistically significant (p = 0.26). Importantly, and in sharp contrast to wild-type E12, the reductions that occurred between 15 min
and 5 h after BCR activation disappeared almost completely postinfection when using virus expressing CaM-resistant E12 (Fig.
6B). All of the differences in the reductions between wild-type and
CaM-resistant E12 were significant (p # 0.03). In summary, the
results show that CaM inhibition of E12 is essential for negativefeedback regulation of most of the BCR-signaling components.
Feedback inhibition of BCR signaling after BCR activation
depends on CaM sensitivity of E2A
To analyze whether negative-feedback regulation of BCR function after BCR stimulation depends on CaM sensitivity of E2A,
we compared the activating phosphorylation of JNK, ERK, and
p38 downstream of the BCR after an initial BCR stimulation with
anti-IgM and after a second BCR stimulation 6 h later. Purified
and activated primary splenic B cells from mice heterozygous
for deletion of the E2A gene that encodes E12 were infected
with a retrovirus expressing either wild-type or CaM-resistant
E12, together with GFP. The activating phosphorylations were all
strongly induced 15 and 30 min after the initial BCR stimulation
when infecting the B cells with virus expressing wild-type E12
or CaM-resistant E12 (Fig. 7A), as shown above for noninfected
B cells (Fig. 3B). In contrast, the result of the second BCR stimulation differed remarkably between B cells expressing wildtype or CaM-resistant E12 (Fig. 7A). As for noninfected B cells
(Fig. 3B), the second BCR stimulation was totally unable to induce any of the activating phosphorylations postinfection of the
B cells with wild-type E12, whereas all three phosphorylation
levels were high 6 h after the initial BCR stimulation, and especially upon 15 or 30 min of restimulation in the B cells in which
infection was with CaM-resistant E12 (Fig. 7A, quantifications
in Supplemental Fig. 4). This dramatic difference was especially
remarkable, because the majority of the cells in these Western blot
analyses were noninfected. Based on each quantified measurement of the level of the activating phosphorylation in the total cell
population postinfection, as well as in cells analyzed in parallel
that were not infected, and knowing the percentage of infected
cells in the individual infection (∼40%), the levels of the induced
phosphorylations in the infected cell populations were calculated.
Importantly, and in contrast to B cells infected with wild-type E12
that had downregulated all three phosphorylations and lacked response to the second BCR stimulation, this analysis showed that
the B cells infected with CaM-resistant E12 had lost the downregulation and were nevertheless, at least for phosphorylation of
p38 and ERK, responding to the second BCR stimulation by in-
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 5. Transitory in vivo proximity of CD19 and LYN upon BCR
stimulation. B cells from mouse spleen were activated with CD40L plus
IL-4 activators for 48 h, followed by stimulation of the BCR by addition of
anti-IgM for the indicated times. The detection of the proximal location of
CD19 and LYN (shown as red dots) by in situ PLA was as described in
Materials and Methods. The B cells used in controls with only the antiCD19 or the anti-LYN primary Ab or with no primary Ab (2Ab) were
stimulated with anti-IgM for 0.5 h. All samples were counterstained with
Hoechst 33342 dye (blue) to visualize nuclei. Original magnification
3100.
The Journal of Immunology
7
creased phosphorylation (Fig. 7B). The p values for the differences between wild-type and CaM-resistant E12 in the phosphorylations were all #0.04, #0.02, and #0.01 after 0, 15, and 30
min, respectively, of restimulation of the BCR. Thus, feedback
inhibition of BCR signaling after BCR activation depends on inhibition of E2A by CaM.
Discussion
Regulation of Ag receptor signaling is essential for the development of specific immunity while retaining tolerance to self. Although a great deal of progress has been made in the identification
of various signaling factors and their role in BCR signaling, the
precise molecular mechanisms of how this activation of B cells
regulates their differentiation remains poorly understood. In this
study, we show that activation of signaling by the BCR leads to
negative feedback regulation of most BCR-signaling components, including the BCR proteins, coreceptors, and downstreamsignaling factors. We also show that signaling through the BCR
is strongly reduced after activation of the receptor. Within 5 h after
BCR activation, the individual proteins were reduced 30–60%,
with a few exceptions. For each protein component, this can be
considered a relatively modest downregulation. However, together, .15 such downregulations (Fig. 3A) could explain the
very large downregulation of BCR signaling. Our PLA analysis of
interacting protein pairs showed 2–5-fold reductions for a large
number of pairs (Fig. 6A). The reductions in the protein pairs were
close to the products of the levels of the reductions of the in-
dividual proteins. The four pairs, CD19-BLNK, CD22-PLCg2,
SYK-LYN, and Igb-BTK, which have no protein in common, all
showed 2–5-fold reductions in PLA between 15 min and 5 h after BCR stimulation. This finding suggests that the reduction in
the complete BCR signalosomes with all proteins in appropriate
amounts is much greater and could exceed the product of these
four reductions, because at least as many other proteins of the
BCR signalosome were also downregulated. Furthermore, all signaling proteins of the BCR signalosome were not analyzed at
the protein level, and there is no reason why the downregulations should be limited to the studied ones. In addition, at least
one analyzed protein was downregulated posttranscriptionally.
Therefore, it is possible that other signalosome proteins are also
downregulated posttranscriptionally. Hence, the total downregulation of the BCR signalosome can be at least several orders of
magnitude and can explain that activation of the BCR leads to
a strong feedback reduction in signaling through the receptor
(Fig. 3B).
What is the physiological role of the strong negative feedback
regulation of the BCR signalosome in the immune system? The
reason for this feedback is most likely because the BCR has to
detect improvements in Ag affinity over an extremely broad range
of affinities. BCR signaling should detect differences in affinity,
even when relatively small, for Ag-binding strengths all the way
from ∼106 M21 or less to ∼1010 M21 and greater. The BCR is
used to select the B cells with highest Ag affinity for the extrafollicular plasmablast responses to optimize the initial wave of Ab
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 6. Reduction of in vivo proximity of BCR
signaling components after 5 h of BCR stimulation
in cells expressing wild-type E12 is blocked in cells
expressing CaM-resistant E12. (A) In vivo proximity of
BCR-signaling proteins was reduced with prolonged
BCR stimulation. The analyses of indicated protein
pairs were performed by PLA, as described in Fig. 5.
The bar graphs show relative numbers of proximities/
cell for each of the protein pairs derived by counting
photographs of four or five different fields for each
sample. A representative part of a field of one PLA
experiment for each protein pair, with the exception of
CD19-LYN, is shown in Supplemental Fig. 3. The Abs
were used at dilutions that yielded a low number of
dots/cell to avoid missing dots in the quantifications.
The number of dots in B cell samples, after 0.25 h of
anti-IgM treatment, was set as 100% for all nine pairs:
CD19-LYN (7.23 dots/cell), CD19-SYK (2.77 dots/
cell), CD19-BLNK (1.06 dots/cell), CD22-SYK (1.08
dots/cell), CD22-BLNK (1.08 dots/cell), CD22-PLCg2
(3.63 dots/cell), SYK-LYN (0.73 dots/cell), Igb-BTK
(1.7 dots/cell), and Igb-PLCg2 (1.2 dots/cell). The B
cells used in controls with only one primary Ab or no
primary Ab (2Ab) were stimulated with anti-IgM for
0.5 h. All results are mean 6 SD of three independent
experiments. (B) Purified primary splenic B cells were
infected with retroviruses encoding either wild-type or
CaM-resistant E12, together with GFP, and the BCR
was stimulated with anti-IgM for 0.25 or 5 h. The
detection of indicated protein pairs was done by PLA,
as described in Fig. 5. The relative number of proximities/successfully infected cell, as detected by staining
for GFP using FITC-labeled anti-GFP (Novus Biologicals), is shown for each of the protein pairs. The numbers of proximities/cell after 0.25 h of anti-IgM treatment were set as 100% in infected B cells expressing
wild-type E12 and those expressing CaM-resistant E12
(not represented in the bar graphs). Data are mean 6 SD
of three independent experiments.
8
FEEDBACK INHIBITION OF Ag RECEPTORS BY CALMODULIN
defense against infections, whereas B cells with less strong Ag
binding are selected for the GC, where they undergo SH and selection for improved Ag affinity. Finally, the GC B cells with
highest affinity are selected for plasma cell differentiation (1–3).
High-affinity GC B cell clones have increased their Ag-binding
strength by many orders of magnitude (from ∼106 M21 to at least
1010 M21) and typically have 5–10 mutations in the variable
regions of their Ig genes. They typically have gone through several
rounds of SH. Each round leads to one or a few mutations, followed by selection for increased affinity; after a number of such
cycles, the B cells with highest Ag affinity are selected for plasma
cell differentiation and, thereby, mass production of their highaffinity Ab (1–3). The threshold for BCR signaling has to increase
successively during the GC reaction, because, during this process,
B cells with successively higher and higher affinities will compete
for the signal enabling continued improvement, instead of apoptosis, and, finally, memory cell or plasma cell differentiation.
Competition for Ag is believed to enable successively more
stringent requirements for BCR affinity at later stages of the GC
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 7. Loss of negative feedback regulation of BCR function in B
cells expressing CaM-resistant E12. (A) Purified primary splenic B cells
from mice heterozygous for deletion of the E2A gene that encodes E12
were activated with CD40L and IL-4. After 24 h, they were infected with
a retrovirus expressing either wild-type or CaM-resistant E12, together
with GFP. After 22 h of additional incubation postinfection, their BCR was
activated by addition of anti-IgM for 15 or 30 min or 4 h. After 4 h, antiIgM was washed away; 2 h later, the B cells were restimulated with antiIgM for 15 or 30 min. Approximately 40% of the cells was infected, as
determined by their green fluorescence in FACS analysis. The levels of the
induced phosphorylations and the expression levels of the signaling proteins JNK, p38, and ERK were determined by Western blot analysis at the
indicated times. A representative experiment is shown; average results 6
SD from three independent experiments are shown in Supplemental Fig. 4.
(B) Comparison between the levels of the induced phosphorylations of the
signaling proteins JNK, p38, and ERK after the second stimulation of the
BCR in splenic B cells expressing either wild-type or CaM-resistant E12 in
the experiments shown in (A) and Supplemental Fig. 4. The levels of the
induced phosphorylations in the infected cells were calculated from the
levels of these phosphorylations in the total cell population [shown in (A)
and Supplemental Fig. 4], as well as in cells analyzed in parallel that were
not infected. The calculations were based on the percentage of infected
cells after each infection (∼40%) determined by their green fluorescence in
FACS analysis. Data are from three independent experiments.
reaction when the competitors have improved their Ag affinity (2,
3). However, the experimental data supporting this remain weak
(3), and mathematical modeling suggests that competition for Ag
binding is not the most efficient selection mechanism, even when
including limiting Ag binding sites on the follicular dendritic cells
in the simulations (2). In this study, we examined whether the
increasingly stringent requirement for BCR affinity during the GC
reaction could be due, at least in part, to an effect of activation of
BCR signaling on the BCR signalosome itself. We show that the
BCR can downregulate itself. Furthermore, BCR restimulations of
different strengths after a weak initial stimulation directly showed
that BCR signaling reduces the sensitivity of the BCR to the next
stimulation. This could enable sensitive detection of successive
improvements in Ag affinity of the receptor over a very large span
of affinities during SH. This discovered property of the B cellselection system does not exclude the previously suggested selection mechanisms. Instead, it is likely that the impressive efficiency of affinity maturation in the GC is due to a combination
of selection mechanisms (2). One such mechanism is affinitydependent T cell help (2, 3, 23). It is notable that in our analysis
of the genome-wide transcriptional changes upon stimulation of
the BCR of mouse splenic B cells activated to SH, three of the
most strongly upregulated genes are those for the chemokines
XCL1/lymphotactin-a, CCL3 (MIP1-a), and CCL4 (MIP1-b),
which are all chemotactic for T cells. Their mRNA levels increased 50-, 32-, and 24-fold, respectively, in 3 h. For CCL3 and
CCL4, BCR stimulation was also shown to induce secretion of the
proteins from tonsillar B cells (24). Thus, BCR stimulation leads
to a concerted downregulation of the BCR signalosome and upregulation of chemokines recruiting T cell help, which could provide an efficient system for selection of additional successive improvements in Ag affinity.
Autoreactive B cells are eliminated by apoptosis, receptor editing, and anergy. In the last case, autospecific cells persist but
have become unresponsive to Ag. Desensitized BCR of anergic
B cells exhibit a defect at the earliest events in BCR signaling,
consistent with failed transmission of signals through the receptor
complex (25, 26). Importantly, a consistent feature of anergic T
and B lymphocytes is increased basal intracellular free Ca2+ (26),
which indicates the possibility of a long-term reduction in E2A
activity by Ca2+-loaded CaM in such cells. Therefore, our findings
could indicate that the mechanism reported in this article is also
important in desensitization of the BCR in anergy. This mechanism could function together with other mechanisms of BCR desensitization in anergy, as discussed by Yarkoni et al. (26).
We show that inhibition of the transcription factor E2A by Ca2+/
CaM is important for the feedback inhibition of BCR signaling
and most of the proteins of the BCR signalosome. How much
of the downregulation is through inhibition of this transcription
factor? Most of the negative feedback of the BCR appears to be
through inhibition of E2A, because mutation of this protein to
CaM resistance leads to loss of the downregulation of BCR signaling, as well as loss of most of the negative feedback effect of
BCR stimulation on expression of BCR signalosome components.
We previously reported that AID protein in SH and repressors of
initiation of plasma cell development were regulated through inhibition of the transcription factor E2A by Ca2+/CaM (5, 14). Our
global analysis of the effects of mutation of E2A to CaM resistance on mRNA levels after BCR stimulation showed that the
previously analyzed genes were no exceptions. In fact, .80% of
all significantly downregulated genes lost .50% of their downregulation in cells expressing a CaM-resistant E2A.
There were approximately as many genes with upregulation that
is dependent on CaM sensitivity of E2A as there were genes with
The Journal of Immunology
downregulation that is dependent on this sensitivity. How can this
occur? It could be due to CaM-dependent relief of repression by
E2A, because E2A represses some of its target genes (27). Inductions could also be the result of a shift from CaM-sensitive Eprotein homodimers to a CaM-resistant E-protein heterodimer with
another bHLH transcription factor, analogous to the Ca2+/CaM
dependent shift to E-protein heterodimers with MyoD, which is
important in muscle development (20). In summary, we report that
inhibition of E2A by Ca2+/CaM plays a major role in the responses
of B cells to BCR stimulation and, in particular, in a broad negative
feedback regulation of the BCR signalosome. The role of defects
in Ca2+/CaM inhibition of E2A, leading to dysregulated BCR signaling responses in development of humoral immune disorders and
B cell lymphomas, will be important to explore.
Acknowledgments
We thank Drs. J. Forssell and P. Sideras for the kind gift of the Ab against
BTK.
The authors have no financial conflicts of interest.
References
1. Brink, R., T. G. Phan, D. Paus, and T. D. Chan. 2008. Visualizing the effects of
antigen affinity on T-dependent B-cell differentiation. Immunol. Cell Biol. 86:
31–39.
2. Meyer-Hermann, M., M. T. Figge, and K. M. Toellner. 2009. Germinal centres
seen through the mathematical eye: B-cell models on the catwalk. Trends
Immunol. 30: 157–164.
3. Tarlinton, D. M. 2008. Evolution in miniature: selection, survival and distribution of antigen reactive cells in the germinal centre. Immunol. Cell Biol. 86: 133–
138.
4. Tussiwand, R., N. Bosco, R. Ceredig, and A. G. Rolink. 2009. Tolerance
checkpoints in B-cell development: Johnny B good. Eur. J. Immunol. 39: 2317–
2324.
5. Hauser, J., N. Sveshnikova, A. Wallenius, S. Baradaran, J. Saarikettu, and
T. Grundström. 2008. B-cell receptor activation inhibits AID expression through
calmodulin inhibition of E-proteins. Proc. Natl. Acad. Sci. USA 105: 1267–1272.
6. Zhuang, Y., P. Soriano, and H. Weintraub. 1994. The helix-loop-helix gene E2A
is required for B cell formation. Cell 79: 875–884.
7. Hauser, J., A. Wallenius, N. Sveshnikova, J. Saarikettu, and T. Grundström.
2010. Calmodulin inhibition of E2A stops expression of surrogate light chains of
the pre-B-cell receptor and CD19. Mol. Immunol. 47: 1031–1038.
8. Krutzik, P. O., and G. P. Nolan. 2003. Intracellular phospho-protein staining
techniques for flow cytometry: monitoring single cell signaling events. Cytometry A 55: 61–70.
9. Saarikettu, J., N. Sveshnikova, and T. Grundström. 2004. Calcium/calmodulin
inhibition of transcriptional activity of E-proteins by prevention of their binding
to DNA. J. Biol. Chem. 279: 41004–41011.
10. Jarvius, M., J. Paulsson, I. Weibrecht, K. J. Leuchowius, A. C. Andersson,
C. Wählby, M. Gullberg, J. Botling, T. Sjöblom, B. Markova, et al. 2007. In situ
detection of phosphorylated platelet-derived growth factor receptor beta using
a generalized proximity ligation method. Mol. Cell. Proteomics 6: 1500–1509.
11. Söderberg, O., M. Gullberg, M. Jarvius, K. Ridderstråle, K. J. Leuchowius,
J. Jarvius, K. Wester, P. Hydbring, F. Bahram, L. G. Larsson, and U. Landegren.
2006. Direct observation of individual endogenous protein complexes in situ by
proximity ligation. Nat. Methods 3: 995–1000.
12. Söderberg, O., K. J. Leuchowius, M. Gullberg, M. Jarvius, I. Weibrecht, L. G.
Larsson, and U. Landegren. 2008. Characterizing proteins and their interactions
in cells and tissues using the in situ proximity ligation assay. Methods 45: 227–
232.
13. Söderberg, O., K. J. Leuchowius, M. Kamali-Moghaddam, M. Jarvius,
S. Gustafsdottir, E. Schallmeiner, M. Gullberg, J. Jarvius, and U. Landegren.
2007. Proximity ligation: a specific and versatile tool for the proteomic era.
Genet. Eng. (N. Y.) 28: 85–93.
14. Hauser, J., J. Verma-Gaur, A. Wallenius, and T. Grundström. 2009. Initiation of
antigen receptor-dependent differentiation into plasma cells by calmodulin inhibition of E2A. J. Immunol. 183: 1179–1187.
15. Dal Porto, J. M., S. B. Gauld, K. T. Merrell, D. Mills, A. E. Pugh-Bernard, and
J. Cambier. 2004. B cell antigen receptor signaling 101. Mol. Immunol. 41: 599–
613.
16. Corneliussen, B., M. Holm, Y. Waltersson, J. Onions, B. Hallberg, A. Thornell,
and T. Grundström. 1994. Calcium/calmodulin inhibition of basic-helix-loophelix transcription factor domains. Nature 368: 760–764.
17. Onions, J., S. Hermann, and T. Grundström. 1997. Basic helix-loop-helix protein
sequences determining differential inhibition by calmodulin and S-100 proteins.
J. Biol. Chem. 272: 23930–23937.
18. Lazorchak, A., M. E. Jones, and Y. Zhuang. 2005. New insights into E-protein
function in lymphocyte development. Trends Immunol. 26: 334–338.
19. Murre, C. 2007. Regulation and function of the E2A proteins in B cell development. Adv. Exp. Med. Biol. 596: 1–7.
20. Hauser, J., J. Saarikettu, and T. Grundström. 2008. Calcium regulation of
myogenesis by differential calmodulin inhibition of basic helix-loop-helix
transcription factors. Mol. Biol. Cell 19: 2509–2519.
21. Brezski, R. J., and J. G. Monroe. 2008. B-cell receptor. Adv. Exp. Med. Biol. 640:
12–21.
22. Jun, J. E., and C. C. Goodnow. 2003. Scaffolding of antigen receptors for immunogenic versus tolerogenic signaling. Nat. Immunol. 4: 1057–1064.
23. Allen, C. D., T. Okada, and J. G. Cyster. 2007. Germinal-center organization and
cellular dynamics. Immunity 27: 190–202.
24. Krzysiek, R., E. A. Lefèvre, W. Zou, A. Foussat, J. Bernard, A. Portier,
P. Galanaud, and Y. Richard. 1999. Antigen receptor engagement selectively
induces macrophage inflammatory protein-1 alpha (MIP-1 alpha) and MIP-1
beta chemokine production in human B cells. J. Immunol. 162: 4455–4463.
25. Vilen, B. J., T. Nakamura, and J. C. Cambier. 1999. Antigen-stimulated dissociation of BCR mIg from Ig-alpha/Ig-beta: implications for receptor desensitization. Immunity 10: 239–248.
26. Yarkoni, Y., A. Getahun, and J. C. Cambier. 2010. Molecular underpinning of Bcell anergy. Immunol. Rev. 237: 249–263.
27. Greenbaum, S., A. S. Lazorchak, and Y. Zhuang. 2004. Differential functions for
the transcription factor E2A in positive and negative gene regulation in preB lymphocytes. J. Biol. Chem. 279: 45028–45035.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
Disclosures
9