Germinal centre protein HGAL promotes lymphoid hyperplasia and amyloidosis via BCR-mediated Syk activation

Abstract

The human germinal centre-associated lymphoma gene is specifically expressed in germinal centre B-lymphocytes and germinal centre-derived B-cell lymphomas, but its function is largely unknown. Here we demonstrate that human germinal centre-associated lymphoma directly binds to Syk in B cells, increases its kinase activity on B-cell receptor stimulation and leads to enhanced activation of Syk downstream effectors. To further investigate these findings in vivo, human germinal centre-associated lymphoma transgenic mice were generated. Starting from 12 months of age these mice developed polyclonal B-cell lymphoid hyperplasia, hypergammaglobulinemia and systemic reactive amyloid A (AA) amyloidosis, leading to shortened survival. The lymphoid hyperplasia in the human germinal centre-associated lymphoma transgenic mice are likely attributable to enhanced B-cell receptor signalling as shown by increased Syk phosphorylation, ex vivo B-cell proliferation and increased RhoA activation. Overall, our study shows for the first time that the germinal centre protein human germinal centre-associated lymphoma regulates B-cell receptor signalling in B-lymphocytes which, without appropriate control, may lead to B-cell lymphoproliferation.

Full text

ARTICLE
Received 16 Jul 2012 |Accepted 26 Nov 2012 |Published 8 Jan 2013
Germinal centre protein HGAL promotes lymphoid
hyperplasia and amyloidosis via BCR-mediated Syk
activation
Isabel Romero-Camarero1,2,*, Xiaoyu Jiang3,*, Yasodha Natkunam4, Xiaoqing Lu3, Carolina Vicente-Duen
˜as1,2,
Ines Gonzalez-Herrero1,2, Teresa Flores1,2, Juan Luis Garcia1,2, George McNamara5, Christian Kunder4,
Shuchun Zhao4, Victor Segura6, Lorena Fontan6, Jose A. Martı
´
nez-Climent6, Francisco Javier Garcı
´a-Criado7,
Jason D. Theis8, Ahmet Dogan8, Elena Campos-Sa
´nchez9, Michael R. Green10, Ash A. Alizadeh10,
Cesar Cobaleda9, Isidro Sa
´nchez-Garcı
´a1,2 & Izidore S. Lossos3,11
The human germinal centre-associated lymphoma gene is specifically expressed in germinal
centre B-lymphocytes and germinal centre-derived B-cell lymphomas, but its function is
largely unknown. Here we demonstrate that human germinal centre-associated lymphoma
directly binds to Syk in B cells, increases its kinase activity on B-cell receptor stimulation and
leads to enhanced activation of Syk downstream effectors. To further investigate these
findings in vivo, human germinal centre-associated lymphoma transgenic mice were gener-
ated. Starting from 12 months of age these mice developed polyclonal B-cell lymphoid
hyperplasia, hypergammaglobulinemia and systemic reactive amyloid A (AA) amyloidosis,
leading to shortened survival. The lymphoid hyperplasia in the human germinal centre-
associated lymphoma transgenic mice are likely attributable to enhanced B-cell receptor
signalling as shown by increased Syk phosphorylation, ex vivo B-cell proliferation and
increased RhoA activation. Overall, our study shows for the first time that the germinal centre
protein human germinal centre-associated lymphoma regulates B-cell receptor signalling in
B-lymphocytes which, without appropriate control, may lead to B-cell lymphoproliferation.
DOI: 10.1038/ncomms2334
1Experimental Therapeutics and Translational Oncology Program, Instituto de Biologı
´a Molecular y Celular del Ca
´ncer, CSIC/Universidad de Salamanca,
37007 Salamanca, Spain. 2Institute of Biomedical Research of Salamanca (IBSAL), 37007 Salamanca, Spain. 3Division of Hematology-Oncology,
Department of Medicine, University of Miami, Sylvester Comprehensive Cancer Center, Miami, Florida 33156, USA. 4Department of Pathology, Stanford
University School of Medicine, Stanford, California 94305, USA. 5Analytical Imaging Core Facility (AICF), Department of Cell Biology and Anatomy,
University of Miami, Miami, Florida 33156, USA. 6Division of Oncology, Center for Applied Medical Research (CIMA), University of Navarra, 31008
Pamplona, Spain. 7Department of Surgery, University of Salamanca 37007, Salamanca, Spain. 8Department of Laboratory Medicine and Pathology, Mayo
Clinic, Rochester, Minnesota 55905, USA. 9Centro de Biologia Molecular Severo Ochoa, CSIC/Universidad Autonoma, 28049 Madrid, Spain. 10 Department
of Medicine, Divisions of Hematology and Oncology, Stanford University School of Medicine, Stanford, California 94305, USA. 11 Department of Molecular
and Cellular Pharmacology, University of Miami, Miami, Florida 33156, USA. *These authors contributed equally to this work. Correspondence and requests
for materials should be addressed to I.S.-G. (email: isg@usal.es) or to I.S.L. (email: ILossos@med.miami.edu).
NATURE COMMUNICATIONS | 4:1338 | DOI: 10.1038/ncomms2334 | www.nature.com/naturecommunications 1
&2013 Macmillan Publishers Limited. All rights reserved.

The human germinal centre-associated lymphoma (HGAL)
gene is expressed in germinal centre (GC) B-lymphocytes
and GC-derived B-cell lymphomas1,2. In diffuse large
B-cell lymphoma and classic Hodgkin lymphoma patients, HGAL
expression is associated with improved survival1,3,4. The function
of its murine homologue M17 is unknown; knockout mice
demonstrated reduced-size Peyer’s patches but M17 protein was
dispensable for GC formation and function5.In vitro studies in
human lymphocytes demonstrated that HGAL decreases cell
motility by interacting with F-actin, myosin II and RhoA-specific
guanine nucleotide exchange factors6–8. HGAL induced RhoA
effects not only on cell migration but also on gene expression7.
These findings suggest that HGAL may contribute to the control
of GC lymphocyte motility but do not explain the biological
relevance of GC-specific HGAL expression.
The HGAL protein harbours a modified immunoreceptor
tyrosine-based activation motif (ITAM) frequently used for B-cell
receptor (BCR) signal transduction. BCR signalling is initiated on
antigen binding to membrane Ig, inducing receptor aggregation
and Src kinase family-mediated tyrosine phosphorylation of
ITAMs in signal-transducing elements Ig-aand Ig-b9. ITAM
phosphorylation creates docking sites for Syk SH2 domains.
Recruitment to the Ig-a/bfacilitates Syk phosphorylation, leading
to the activation of signalling molecules that couple the BCR to
multiple downstream signalling pathways. Consequently, Syk has
a key role in BCR signalling and its disruption leads to a block in
B-cell development10–12.
The presence of the ITAM, whose tyrosines can be phos-
phorylated by Lyn2,6, in the HGAL protein, raised a hypothesis
that it might be involved in BCR signalling. We demonstrate that
in vitro HGAL enhances BCR signalling by binding and
increasing Syk activation. To further investigate these findings
in vivo, HGAL transgenic mice were generated. Although young
mice exhibited normal B-cell development, older HGAL
transgenic animals progressively developed polyclonal lymphoid
hyperplasia and reactive amyloid A (AA) amyloidosis. Overall,
our findings implicate HGAL in regulating BCR signalling,
suggesting that it may have a role in humoral immune responses.
Results
HGAL enhances intracellular BCR signalling. We analysed
HGAL’s effect on the BCR pathway in response to surface Ig
stimulation (Fig. 1a). In unstimulated lymphoma cells there was
no evidence for Syk, Btk and PLCg2 phosphorylation (not
shown). Following BCR stimulation, Syk, Btk and PLCg2 phos-
phorylation was markedly reduced in lymphoma cell lines on
knockdown of endogenous HGAL by different short interfering
RNA (Fig. 1a and Supplementary Fig. S1a), while Ig-a/bphos-
phorylation was not affected (not shown). Concordantly, ectopic
HGAL expression in lymphoma cells and human peripheral
B-lymphocytes lacking endogenous HGAL protein resulted in
increased Syk phosphorylation (Fig. 1a,b).
In activated B-lymphocytes, Btk, PLCg2 and BLNK constitute
the calcium (Ca2þ) initiation complex, inducing release of
intracellular Ca2þand activating protein kinase C, nuclear factor
of activated T cells, nuclear factor kB and mitogen-activated
protein kinase signalling9,13. Consequently, we examined the
effects of HGAL gain and loss of function on the ability of
lymphoma cells to mobilize Ca2þin response to BCR
stimulation. Following Ig stimulation, HGAL expression in
HBL-1 cells led to increased Ca2þmobilization (not shown),
while HGAL knockdown in Raji, VAL and BJAB cells markedly
reduced Ca2þmobilization (Fig. 1c and Supplementary Fig. S1b).
HGAL knockdown in lymphoma cells transfected with constructs
containing either nuclear factor of activated T cells or nuclear
factor kB responsive luciferase reporter genes resulted in
significantly decreased reporter activity (Fig. 1d and
Supplementary Fig. S1c). HGAL knockdown in lymphoma cells
also led to decreased extracellular signal-regulated kinase1/2 and
p38 phosphorylation, while Jun N-terminal kinases
phosphorylation was unaffected (Fig. 1e and Supplementary
Fig. S1d), consistent with previous reports14. Overall, these
findings show that HGAL affects BCR signalling downstream of
the Ig-a/bproteins.
HGAL interacts with Syk and enhances its kinase activity. Syk
has a major role in coupling BCR activation to downstream
effectors. As Syk selectively binds to phosphorylated ITAM10,we
examined for a potential interaction between HGAL and Syk.
Endogenous Syk was detected in immunoprecipitates of
endogenous HGAL from unstimulated Raji and VAL
lymphoma cells (Fig. 2a and not shown). BCR stimulation
increased HGAL and Syk coimmunoprecipitation (coIP)
(Fig. 2a). HGAL protein did not coimmunoprecipitate with
Lyn, Ig-a/bp, Btk, BLNK and PLCg2 in unstimulated and BCR-
stimulated lymphoma cells (not shown). Nanoscope microscopy
of unstimulated Raji cells confirmed HGAL and Syk co-
localization in the cell membrane (Fig. 2b); BCR stimulation
further increased HGAL/Syk co-localization, in concordance with
the coIP findings. A glutathione S-transferase (GST) pull-down
assay using purified GST-Syk, GST-Grb2 or GST and TRX-
HGAL proteins demonstrated a direct, specific interaction. GST-
Syk pulled-down TRX-HGAL; however, non-phosphorylated
HGAL did not bind to GST-Grb2 and GST proteins (Fig. 2c).
To determine whether the Syk tandem SH2 domains mediate
interaction with HGAL protein, we transiently co-transfected
wild-type HGAL with truncated Syk encoding the tandem SH2
domains (amino acids 1–261) into Raji and 293T cells and
repeated the coIP experiments. Truncated Syk protein was
detected in the HGAL immunoprecipitates (Fig. 2d and
Supplementary Fig. S2a), indicating that HGAL protein
interacts with the Syk tandem SH2 domains.
To determine the HGAL interacting domain, HGAL mutants
encoding N-terminal 79 (1–79) or 118 (1–118) aa segments and
tyrosine mutants (Y128F, Y148F and Y106AY107A)
(Supplementary Fig. S2b) were co-transfected with the wild-
type Syk into VAL and 293T cells and coIP experiments were
repeated (Fig. 2e and Supplementary Fig. S2c). Individual
mutations of HGAL tyrosines and HGAL truncations did not
affect Syk binding, suggesting that the first 79aa of HGAL were
sufficient to mediate Syk binding.
We next examined whether the HGAL protein stimulates Syk
kinase activity. Syk protein was immunoprecipitated from BCR-
stimulated or unstimulated Raji cells and used in the kinase assay,
either alone or with purified HGAL protein (Fig. 2f). No Syk
kinase activity was observed in unstimulated cells, even in the
presence of HGAL protein. In stimulated cells, addition of HGAL
protein markedly increased Syk kinase activity, suggesting that
HGAL binding to Syk enhances its kinase activity in BCR-
stimulated cells, underlying the observed increased Syk
autophosphorylation and activation of its downstream effectors.
HGAL gain-of-function mouse model. To examine HGAL
effects on the immune system and BCR signalling in vivo, a Sca1-
HGAL plasmid encoding human HGAL complementary DNA
(cDNA) under the control of the mouse Ly-6E.1 promoter15
(Fig. 3a,b) was used to generate a transgenic mouse model in
which HGAL is expressed in Sca1 þhaematopoietic stem (HSC)/
progenitor cells and Sca1 þfraction of mature B cells of C57BL/
6xCBA mice16. A similar approach recapitulated gene functions
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2334
2NATURE COMMUNICATIONS | 4:1338 | DOI: 10.1038/ncomms2334 | www.nature.com/naturecommunications
&2013 Macmillan Publishers Limited. All rights reserved.

and generated animal models highly similar to human
diseases17–19. Two independent Sca1-HGAL founder lines (102A
and 102B) exhibited normal embryonic and postnatal development
and were used to characterize the transgenic mice phenotype.
A Southern blot comparison of the endogenous M17 and
transgenic human HGAL hybridization signals indicated transgene
copy numbers ranging from 2 to 4 (Fig. 3b). Flow cytometry
studies revealed that only a fraction of B220 splenocytes expressed
Sca1. Similar fractions of B220 splenocytes in young Sca1-HGAL
(4- to 8-week-old) and control mice expressed Sca1 (Supple-
mentary Fig. S3a). While there was a tendency for a smaller Sca1
expressing fraction of B220 splenocytes in older (starting at 12
months of age) Sca1-HGAL mice compared with control animals,
it was not statistically significant (Supplementary Fig. S3a).
Immunofluorescence studies using antibody to the V5 tag, fused
to HGAL in the plasmid used to generate the transgenic animal,
detected ectopically expressed human HGAL in both bone marrow
(BM) and spleen cells (Fig. 3c), with no difference in expression
between young and old animals. HGAL expression was not
detected in either mature myeloid, monocyte and T-cell lineages or
in the wild-type controls. Overall, HGAL protein expression levels
were similar to levels observed in the human U2OS cell line
transfected with the same HGAL plasmid used to generate the
transgenic construct (Supplementary Fig. S3b). There was no
difference in the endogenous M17 messenger RNA (mRNA)
expression between Sca1-HGAL and littermate splenocytes
(Supplementary Fig. S3c).
Lymphoid hyperplasia and amyloidosis in Sca1-HGAL mice.
A total of 75 transgenic animals were analysed. Compared with
age-matched controls, 8-week-old Sca1-HGAL animals did not
show any visible changes within the major haematopoietic
compartments (BM, spleen, thymus, peripheral blood and
lymph nodes (LNs)) by flow cytometry and histological exam-
inations (Fig. 4a, Supplementary Fig. S4). Immunization with
sheep red blood cells led to GC formation in both transgenic and
wild-type mice. Flow cytometry analyses did not reveal statisti-
cally significant differences in the number of splenic B220 þ
PNA þFas þGL7 þGC B cells between the Sca1-HGAL and
control animals (Supplementary Fig. S5a). Immunohistochem-
istry also did not reveal differences in the size and number
of GCs in the spleens of immunized Sca1-HGAL and
control mice (Supplementary Fig. S5b), indicating that young
Raji HGAL siRNA
VAL HGAL siRNA
Actin
pBtk (pY551)
Btk
Syk
Raji control siRNA
VAL control siRNA
HBL-1 pcDNA3
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Raji control
siRNA
Raji HGAL
siRNA
NFKB
*
Raji control siRNA
Raji HGAL siRNA
HGAL (V5)
HGAL
28
0
0.2
0.4
0.6
0.8
1
1.2
Raji control
siRNA
Raji HGAL
siRNA
NFAT
*
B-cell pcDNA3
HBL-1 pcDNA3-HGAL
B-cell pcDNA3-HGAL
Actin
42 kDa
pSyk (pY352)
Syk
HGAL (V5)
28 kDa
1.0 2.2
p38
HGAL
Actin
pJNK (pT183/Y184)
JNK
p-p38 (pT180/Y182)
ERK
1.0
1.0
1.0
1.0 0.1 1.0 1.00.1 0.1
1.0 1.0
1.0 1.0 1.0 1.0 1.2
1.0 1.0 0.8
0.8 0.6 0.6
BJAB HGAL siRNA
BJAB control siRNA
BJAB HGAL siRNA
BJAB control siRNA
BJAB HGAL siRNA
BJAB control siRNA
kDa
43
43
45
45
40
40
25
42
kDa
72
72
72
72
130
130
25
42
0 200
20
400
40
600
60
800
80
100
FLUO-4 AM
Time (sec)
0
1K
0.5 0.5
α-IgM
p-ERK (pY204)
15′15′5′5′1′1′
pSyk (pY352)
pPLC γ2 (pY753)
PLC γ2
Luc/Rlu
1.0 0.5 1.0 0.4 1.0 >10
1.0 0.3 1.0 0.3 1.0 2.5
1.0
1.0 1.00.1 0.1
0.3 1.0 0.2 1.0 2.0
α-IgM
Figure 1 | HGAL enhances intracelular BCR signalling. Raji and VAL lymphoma cells were transfected with short interfering RNA (siRNA) for HGAL or
scrambled control siRNA and HBL-1 cells (a) and normal peripheral blood B-lymphocytes (b) transfected with pcDNA3.1-HGAL or pcDNA3.1-mock
plasmids for 48 h. After 1 min stimulation with goat F(ab’)2 anti-human IgM, western blot of BCR receptor effectors was performed. HGAL expression and
equal loading were analysed by immunoblotting with HGAL and actin antibodies. (c) Kinetic of calcium mobilization in Raji cells transfected with siRNA for
HGAL or scrambled control siRNA. Arrow indicated the time point of goat F(ab’)2 anti-human IgM stimulation. (d) Raji lymphoma cells were transfected
with firefly luciferase reporter plasmid pNFkB-Luc or pNFAT-Luc and renilla luciferase plasmid pRL-TK and with either siRNA for HGAL or scrambled control
siRNA. Forty-eight hours after transfection, the cells were stimulated for 10min with goat F(ab’)2 anti-human IgM and luciferase activities were detected
with the dual luciferase assay kit. Numbers refer to luciferase activities representing three independent experiments, each performed in triplicate. * Indicate
statistically significant difference (NFkBP¼0.0001 and nuclear factor of activated T cells P¼0.001 by two-tailed Student’s t-test). Data are presented as
mean±s.d. of the mean. (e) BJAB lymphoma cells were transfected with siRNA for HGAL or scrambled control siRNA for 48h. After stimulation with goat
F(ab’)2 anti-human IgM, cell lysates were used for western blot of mitogen-activated protein kinase/extracellular signal-regulated kinase pathway effectors
at indicated time points. HGAL knockdown or expression and equal loading were confirmed by immunoblotting with HGAL and actin antibodies.
Normalized densitometry measurements are shown below the corresponding blots. Results in (a–e) are representative of three independent experiments.
NFkB, nuclear factor kB.
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2334 ARTICLE
NATURE COMMUNICATIONS | 4:1338 | DOI: 10.1038/ncomms2334 | www.nature.com/naturecommunications 3
&2013 Macmillan Publishers Limited. All rights reserved.

Sca1-HGAL mice respond normally to T-cell-dependent antigen
stimulation.
Starting at 12 months of age, the Sca1-HGAL mice demon-
strated increased-sized Peyer’s patches compared with control
animals (Fig. 3d) and expressed HGAL protein at levels similar to
ones observed in human Payer’s patches. These animals also
exhibited non-statistically significant decrease in number of BM
immature and pro- and pre-B-cells at the expense of recirculating
B cells (B220 þIgDhi) (Fig. 4a,b). All the Sca1-HGAL mice
became ill, showing decreased spontaneous movements in the
cage, increased respiratory rates, piloerection and shivering, and
died between 12 and 22 months, exhibiting statistically significant
shorter overall survival (Fig. 5a).
Macroscopic analysis of these animals revealed massive spleno-
megaly (Fig. 5b). The overall microscopic architecture of the spleen
was preserved; however, the splenic sinusoids were dilated and
distorted by large quantities of proteinaceous deposits, which were
also present surrounding small and medium-calibre splenic blood
vessels (Fig. 5c). These deposits were found adjacent to expanded
white-pulp areas showing lymphoid hyperplasia with preserved
periarteriolar lymphoid sheaths. Immunohistochemistry using B220
and PAX5 antibodies highlighted a significant proportion of the
expanded white pulp representing B cells, whereas CD3 highlighted
a slightly smaller proportion of T cells. B220 staining was weak
compared with that of PAX5, although both stains showed an
equivalent distribution of B cells. Staining for IRF4 highlighted
T-cell areas in addition to a few scattered plasmacytoid cells.
PNAbio showed no significant staining within the white-pulp,
suggesting that white-pulp expansions represented lymphoid
hyperplasia without GC formation (Fig. 5c). Control animals
showed a similar distribution of B and T cells within the white-
pulp, although without lymphoid hyperplasia (Supplementary
Fig. S6a). In both Sca1-HGAL and control animals, CD38 (not
shown) and CD138 showed weak expression in the B cells without
the intense expression typically seen in plasma cells (Fig. 5c and
Supplementary Fig. S6a). Flow cytometry also did not demonstrate
differences in the number and percentage of B220loCD138 þ
splenic cells between young and old Sca1-HGAL and control mice
(Supplementary Fig. S6b). In contrast, flow cytometry analysis
detected IgM þB-cells in peripheral blood and BM (Fig. 5d,e),
confirming that the observed splenic lymphoid hyperplasia was due
to follicular (FO) B cells (Fig. 5f). The percentage of Sca1þLin 
cells in BM did not increase in Sca1-HGAL mice compared with
normal littermates (data not shown). Analysis of the B-1 lympho-
cyte population in the older animals revealed a significant decrease
in B-1 cell number and percentage in the spleen and a significant
increase in the LN of the Sca1-HGAL mice compared with normal
littermates (Supplementary Fig. S4). There was no difference in the
number and percentage of B-1 and B-1a cells in the BM and
peritoneum between the Sca1-HGAL and control animals.
IP: HGAL Ab
Blot: Syk
HGAL
GST + TRX-HGAL
GST-Syk + TRX-HGAL
Blot: GST
Blot: HGAL 37 kDa
25 kDa
95 kDa
25 kDa
72 kDa
HGAL + Syk
HGAL + Syk-DN
Blot: Syk (Flag)
HGAL(V5)
Blot: Syk (Flag)
HGAL (V5)
HGAL + Syk
HGAL(Y148F) + Syk
HGAL(Y128F) + Syk
27 kDa
15 kDa
72 kDa
35 kDa
27 kDa
72 kDa
10 kDa
HGAL(Y106AY107A) + Syk
HGAL(1–79) + Syk
94
96
98
100
102
104
106
108
50 kDa
GST + TRX-HGAL
GST-Grb2 + TRX-HGAL
–++
––+
IP: V5 Ab – + + IP: V5 (HGAL) Ab – + + + + + +
HGAL + Syk
HGAL + Syk
HGAL HGAL
SykSyk
Syk +α-IgM
HGAL
Syk
Relative kinase activity
HGAL(1–118) + Syk
Time (40 min)
H2O
H2O+IP beads only
H2O+HGAL
H2O+Syk IP beads
H2O+Syk IP beads-IgM
HGAL+Syk IP beads
α-IgM Ab
HGAL +α-IgM +α-IgM
Figure 2 | HGAL interacts with Syk and stimulates its kinase activity. (a) Cell lysates extracted from unstimulated and goat F(ab’)2 anti-human IgM
stimulated (1 min) Raji lymphoma cells were subjected to immunoprecipitation with anti-HGAL or control antibodies followed by anti-HGAL and anti-Syk
western blot, respectively. (b) Unstimulated or stimulated Raji cells (as described in a) were stained with HGAL (green) and Syk (red) antibodies. Scale bar,
10 mm. (c) Purified GST-Syk, GCT-Grb2 or GST proteins were incubated with TRX-HGAL protein for 12h. The coprecipitated HGAL and Syk or Grb2 proteins
were detected by western blot with anti-GST and anti-HGAL antibodies. (d) Raji cells were transfected with wild-type Syk or its dominant negative Syk-DN
mutant plasmids and HGAL tagged with V5 plasmid. Cells lysates were immunoprecipitated with anti-V5 or control antibodies and analysed by western
blot with anti-Syk and anti-HGAL antibodies. (e) VAL cells were transfected with Syk plasmid and V5-tagged plasmids encoding wild-type HGAL or its
C-terminal truncated mutants HGAL (1–79) or HGAL(1–118), or tyrosine mutants HGAL(Y106AY107A), HGAL(Y128F) or HGAL(Y148F). Cells lysates were
immunoprecipitated with anti-V5 or control antibodies and analysed by western blot with anti-Syk and anti-HGAL antibodies. (f) Syk was
immunoprecipitated from unstimulated or stimulated Raji cells as described in aand used in Syk kinase activity measurement, either alone or with purified
HGAL protein. Immunoprecipitates with control antibody and beads only, purified HGAL alone and water were used as negative controls. Results in (a-f)
are representative of two (c) or three independent experiments.
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2334
4NATURE COMMUNICATIONS | 4:1338 | DOI: 10.1038/ncomms2334 | www.nature.com/naturecommunications
&2013 Macmillan Publishers Limited. All rights reserved.

Congo red and trichrome stains of the spleens for amyloid
showed distinctive positive staining of the proteinaceous deposits;
Congo red showed the characteristic muted orange colour on light
microscopy and exhibited apple-green birefringence under polar-
ized light. Similarly, the trichrome stain showed a greyish-blue
colour, consistent with amyloid deposition and in contrast to the
bright turquoise blue that is typically demonstrated by collagen
(Fig. 5c). Tissue sections from multiple organs including the kidney,
liver, lung and small intestine showed similar massive amyloid
deposition (Supplementary Fig. S7a,b). Sections of the kidney, liver
and lung showed preservation of the normal architecture with
modest mixed lymphoid infiltrates that were absent in control
animals. Massive amyloid deposition was observed in all renal
glomeruli, as well as the interstitium. Amyloid deposition was also
seen surrounding small and medium calibre blood vessels in the
kidneys, lung and liver, but not the heart. Electron microscopy of
spleen specimens confirmed the presence of amyloid deposits
(Supplementary Fig. S7c,d). Liquid chromatography tandem mass
spectrometry analysis of the kidney and spleen specimens showed
that the most abundant pathogenic protein in the amyloid deposits
was serum amyloid-associated protein-2, consistent with AA
amyloidosis (Fig. 6). Analysis of serum amyloid-associated protein-
2 peptide sequences revealed the absence of 33 aa of the C terminus,
suggesting these residues were cleaved during the amyloidogenesis,
similar to what has been described in human AA amyloidosis20.
Like in human amyloidosis, constituents of many human amyloid
types (apolipoprotein E, apolipoprotein A-IV, serum amyloid
P-component and apolipoprotein A-I) were also detected in the
amyloid deposits in the Sca1-HGAL mice.
To examine whether the observed lymphoid hyperplasia were of
clonal origin, DNA extracted from the splenic B-lymphocytes from
Sca1-HGAL mice were subjected to Southern blot using the 30J
H
4
probe.NoIgheavychain(IgH) gene rearrangements were
detected (Supplementary Fig. S8a). Polymerase chain reactions
followed by direct sequencing or cloning and sequencing of multiple
molecular clones also failed to detect predominant monoclonal
products (Supplementary Fig. S8b), further indicating that the
observed lymphoid hyperplasia were of polyclonal origin.
Cytogenetic and fluorescence in situ hybridization studies of the
splenic B-lymphocytes from lymphoid hyperplasia demonstrated
normal karyotypes without evidence for Bcl2-IgH translocation
(not shown).
Sca1-HGAL mice display hypergammaglobulinemia. The
presence of polyclonal FO B-cell lymphoid hyperplasia in the
Sca1-HGAL transgenic mice pointed to possible immune stimu-
lation, potentially leading to hypergammaglobulinemia. The
non-immunized Sca1-HGAL mice produced antibodies of all
isotypes and IgG titres tended to be higher than in the wild-type
controls, reaching statistical difference for IgG1 isotype (Fig. 7a).
Enhanced BCR signalling in lymphoid hyperplasia splenocytes.
The observed lymphoid hyperplasia in the Sca1-HGAL mice
might be attributed to increased B-cell proliferation due to
enhanced BCR signalling or to promotion of B-cell survival. To
differentiate between these mechanisms, we examined ex vivo
B-cell proliferation in response to BCR stimulation (Fig. 7b).
Stem cell antigen 1 gene (Sca1)
5′3′
ATG
EcoR I
EcoR I
EcoR I
1 kb
Not I
Not I
Human HGAL
Bone marrow Spleen
Control (wt) Sca1-HGAL
(kb)
6.4
23
9.1
4.3
2.0
1.0
0.8
–– –+ –– Human
Peyer’s
patch
Germinal
centres
PAX5
HGAL
WT
HGAL
SCA1-HGAL
+++++++
OldYoung
Figure 3 | HGAL gain-of-function mouse model. (a) Schematic representation of the genomic structure of the mouse Sca1 locus and the Sca1-HGAL
transgenic vector used to generate transgenic animals. (b) Identification of the transgenic mice (n¼75) by Southern analysis of tail snip DNA after EcoRI
digestion. Sca1-HGAL is indicated. (c) HGAL expression in Sca1-HGAL mice. BM and spleen cells of control (n¼3) and Sca1-HGAL mice (n¼3) were
stained with anti-V5 antibody (green) and 40,6-diamidino-2-phenylindole (blue) and examined under a microscope. Data are representative of three
independent experiments. Scale bar, 10 mm(d). Representative sections of ileum show increased-sized Peyer’s patches and prominent GCs stained with
PAX5 and HGAL in Sca1-HGAL compared with control mice. All mouse images were photographed at comparable magnifications: X100, Peyer’s patches
panels; X300, remaining panels. Human Peyer’s patches are shown for comparison (X40, Peyer’s patches panel, X80, remaining panels).
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2334 ARTICLE
NATURE COMMUNICATIONS | 4:1338 | DOI: 10.1038/ncomms2334 | www.nature.com/naturecommunications 5
&2013 Macmillan Publishers Limited. All rights reserved.

There was no difference in cell proliferation rates between control
and young Sca1-HGAL mouse groups at all analysed IgM con-
centrations. In contrast, B-cells from Sca1-HGAL mice with
lymphoid hyperplasia demonstrated increased proliferation to 0.5
and 2 mgml1anti-IgM antibody concentrations compared with
both wild-type controls and young transgenic mice. However,
B-cell proliferation was similar in all the three types of animals at
saturating 8 mgml1anti-IgM antibody concentrations. Splenic
B-cell surface Ig expression was similar in all the three types of
analysed animals (not shown). These findings suggest that
B-lymphocytes from Sca1-HGAL mice with lymphoid hyperpla-
sia show enhanced sensitivity to surface Ig stimulation. Con-
cordantly, increased Ca2þmobilization (Fig. 7c) and Syk
phosphorylation (Fig. 7d) were observed in the BCR-stimulated
B-cells from Sca1-HGAL mice with lymphoid hyperplasia com-
pared with control animals. In contrast, isolated B220 þsple-
nocytes from Sca1-HGAL transgenic and control mice
supplemented only with 5% FCS exhibited a similar in vitro
overtime death rate (Supplementary Fig. S9), suggesting that
HGAL does not promote B-cell survival in the Sca1-HGAL mice.
RhoA activation in Sca1-HGAL transgenic mice. Our previous
in vitro studies demonstrated that HGAL induces RhoA activa-
tion in human B-cells7. Consequently, we examined the levels of
GTP-bound RhoA in B-cell splenocytes isolated from Sca1-
HGAL transgenic mice with and without lymphoid hyperplasia.
RhoA-GTP levels were significantly higher in Sca1-HGAL B-cells
from spleens with versus without lymphoid hyperplasia
(P¼0.006, analysed by two-tailed Student’s t-test) (Fig. 7e),
corroborating the observed activation of this pathway in human
B cells.
Transcriptome changes in Sca1-HGAL transgenic mice.To
comprehensively assess the global effect of HGAL on lymphocyte
functions and the observed lymphoproliferation, we performed
Control Sca1-HGAL
100101102103104
FL1-H: FL1-B220
100
101
102
103
104
FL4-H: FL4-IgM
Immature B cells
5.4
Pro-B and Pre-B cells
13.9
Recirculating B cells
5.93
FL1-H: FL1-B220
FL4-H: FL4-IgM
Immature B cells
2.16
Pro-B and Pre-B cells
2.32
Recirculating B cells
9.6
FL1-H: FL1-CD19
100
101
102
103
104
FL2-H: FL2-C-KIT
Pro-B cells
4.14
FL1-H: FL1-CD19
FL2-H: FL2-C-KIT
Pro-B cells
1.55
FL1-H: FL1-B220
100
101
102
103
104
FL2-H: FL2-CD25
Pre-B cells
10.9
FL1-H: FL1-B220
FL2-H: FL2-CD25
Pre-B cells
1.76
100101102103104
FL2-H: FL2-B220
100
101
102
103
104
FL1-H: FL1-IgD
Pro-B and Pre-B cells
19.5
Recirculating B cells
4.61
FL2-H: FL2-B220
FL1-H: FL1-IgD
Pro-B and Pre-B cells
4.15
Recirculating B cells
9.79
2.5×1007
2.4×1007
2.3×1007
2.2×1007
2.1×1007
2.0×1007
1.9×1007
1.8×1007
1.7×1007
1.6×1007
1.5×1007
1.4×1007
1.3×1007
1.2×1007
1.1×1007
1.0×1007
9000000
8000000
7000000
6000000
5000000
4000000
3000000
2000000
1000000
0
Bone marrow
Young control
Old control
Young Sca1-HGAL
Old Sca1-HGAL
Cells/bone marrow
Spleen
Immature B cells
Recirculating B cells
Pro-B cells
Pre-B cells
Fo B cells
Granulocytes
Macrophages
Hematopoietic stem cells
Young control
Old control
Young Sca1-HGAL
Old Sca1-HGAL
1.5×1008
1.4×1008
1.4×1008
1.4×1008
1.3×1008
1.2×1008
1.2×1008
1.2×1008
1.1×1008
1.0×1008
1.0×1008
9.5×1007
9.0×1007
8.5×1007
8.0×1007
7.5×1007
7.0×1007
6.5×1007
6.0×1007
5.5×1007
5.0×1007
4.5×1007
4.0×1007
3.5×1007
3.0×1007
2.5×1007
2.0×1007
1.5×1007
1.0×1007
5.0×1006
0
Cess/spleen
CD4 T cells
CD8 T cells
Mature B cells
Transitional B cells
Immature B cells
FO B cells
MZ B cells
Granulocytes
100101102103104
100101102103104
100
101
102
103
104
100
101
102
103
104
100
101
102
103
104
100101102103104
100101102103104
100101102103104
100101102103104
100
101
102
103
104
Figure 4 | Analysis of the BM and spleen cell lineages in the Sca1-HGAL mice. Cells from the BM and spleen of young (8 weeks) and old (more than
7-month-old) age-matched Sca1-HGAL mice and wild-type controls (n¼11 in each group) were analysed by flow cytometry. (a) Total cell number of the
indicated cell sub-populations in the BM and spleens of the young and old Sca1-HGAL and control mice. Data are presented as mean±s.d. of the mean.
(b) Representative FACS plots of 12-month-old Sca1-HGAL mice are displayed. FACS definition of B-cell developmental stages was performed according to
Kwon et al.37
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2334
6NATURE COMMUNICATIONS | 4:1338 | DOI: 10.1038/ncomms2334 | www.nature.com/naturecommunications
&2013 Macmillan Publishers Limited. All rights reserved.

gene expression profiling using isolated CD22 þB220 þspleno-
cytes from adult Sca1-HGAL mice with hyperplastic follicles and
wild-type littermates (four samples each).
In supervised analyses using significance analysis of micro-
arrays21 at a 10% false discovery rate (FDR), we identified 981
genes with significantly increased expression in B-splenocytes
from adult Sca1-HGAL mice with lymphoid hyperplasia
compared with controls (Supplementary Data 1). Fig. 8a depicts
107 of these genes (Supplementary Data 2), focusing on those
with at least threefold change between the specimens from Sca1-
HGAL transgenic and control mice. In concordance with HGAL’s
role in regulating lymphocyte motility, gene expression profiles in
the Sca1-HGAL mice with lymphoid hyperplasia apparently
reflect ‘Cell Movement’ (P¼5.46e-9, IPA) and ‘Migration of
Cells’ (P¼0.0001, IPA) as evidenced by significant overlap with
the corresponding pathways using the Ingenuity Pathway
Analysis knowledge base. Sca1-HGAL B-splenocytes also
exhibited decreased mRNA expression of regulator of G-protein
signalling 13 and 18 (RGS13 and RGS18), whose downregulation
may enhance the signalling activity of G-proteins potentially
enhancing RhoA activation.
We also observed significant induction of BCR pathway
components in the Sca1-HGAL mice (Fig. 8a,b), including Ig
heavy chain genes (Ighg1, Igh-2 (IgA), Ighg2b, Ighg2c), corrobo-
rating our observation of serum hypergammaglobulinemia
(Fig. 7a). Within these lymphoid hyperplasias we also observed
significant induction of CD80 (Fig. 8b), which costimulates
T-lymphocytes, a finding which we confirmed at the protein level
(Fig. 8c), demonstrating CD80 expression in a fraction of
B-lymphocytes. No significant differences in expression levels of
mRNAs for CD86 (B7-2), CD69, CD5 or MHC-II antigens were
detected.
To evaluate the relative representation of normal
differentiation states within splenocytes from Sca1-HGAL mice
with hyperplastic follicles compared with B cells from wild-type
littermates, we utilized gene set enrichment analysis (GSEA) of
transcriptional profiling data. Classification signatures from
normal murine stages of B-cell differentiation were obtained
Control Sca1-HGAL
Control Sca1-HGAL
Peripheral blood
Bone marrow
FL1-H: FL1-B220
FL4-H: FL4-IgM
B cells
13.6
100101102103104
FL1-H: FL1-B220
100
101
102
103
104
100101102103104
100
101
102
103
104
100101102103104
100
101
102
103
104
100101102103104
100
101
102
103
104
100101102103104
100
101
102
103
104
100101102103104
100
101
102
103
104
FL4-H: FL4-IgM
B cells
20.5
FL1-H: FL1-B220
FL4-H: FL4-IgM
Immature
B cells
3.84
Pro B and Pre B cells
13
Recirc. B cells
5.69
FL1-H: FL1-B220
FL4-H: FL4-IgM
Immature
B cells
1.39
Pro B and Pre B cells
2.26
9.79
Spleen
FL2-H: FL2-CD21
FL1-H: FL1-CD23
Follicular B cells
79.3
FL2-H: FL2-CD21
FL1-H: FL1-CD23
Follicular B cells
67
MZ B cells
12.3
MZ B cells
3.59
Survival of Sca1-HGAL mice
Percent survival
100
50
0
Months
WT (n=8)
Sca1-HGAL (n=23)
10 20 30
0
H&E H&E
Lymphoid hyperplasia
H&E
Spleen
B220
Congo red Trichrome
PAX5 CD3
Lymphoid hyperplasia
IRF4 PNAbio
Pola rized
CD138
Recirc. B cells
Figure 5 | B-cell lymphoid hyperplasia in Sca1-HGAL mice. (a) Kaplan–Meier overall survival plots of Sca1-HGAL mice. The total numbers of mice
analysed in each group were eight wild-type control mice and 23 Sca1-HGAL transgenic mice. Differential survival in Sca1-HGAL and control mice was
analysed using the log-rank test (P¼0.0000003). (b) Sca1-HGAL mice develop massive splenomegaly. Data is representative of pathologic analysis of 36
Sca1-HGAL transgenic mice. (c) Representative sections of spleen of Sca1-HGAL mice show overall preservation of microscopic architecture; however, the
splenic sinusoids were dilated and distorted by large quantities of proteinaceous deposits (spleen haematoxylin and eosin (H&E) panel original
magnification I X40; lymphoid hyperplasia panels original magnifications X100 and X400). These deposits were found adjacent to lymphoid hyperplasia,
which was highlighted by B220 (weak positive staining) and PAX5 representing B cells, and a smaller proportion of CD3-positive T-cells. Staining for IRF4
highlighted T-cell areas in addition to a few scattered plasmacytoid cells while PNAbio (a GC marker) showed no significant staining to indicate GC
formation. CD138 showed weak expression in the B cells without the intense expression typically seen in plasma cells (B220, PAX5, CD3, IRF4, PNAbio and
CD138 panels, X200 original magnification). Congo red (orange staining) and trichrome (greyish-blue) stains highlight amyloid deposits, which under
polarized light, showed the characteristic apple-green birefringence (X400 original magnification). (d-f) Representative FACS plots of peripheral blood (d),
BM (e) and spleen (f) lymphocytes from Sca1-HGAL mice with lymphoid hyperplasia (n¼15) and control littermates (n¼15).
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2334 ARTICLE
NATURE COMMUNICATIONS | 4:1338 | DOI: 10.1038/ncomms2334 | www.nature.com/naturecommunications 7
&2013 Macmillan Publishers Limited. All rights reserved.

from Green et al.22 to assess expression bias in Sca1-HGAL
compared with control B cells. Gene expression signature
corresponding to both the normal FO/marginal zone (FO/MZ)
and the plasma cell B-cell differentiation states were significantly
enriched in Sca1-HGAL B-cells (Fig. 8d,e). Conversely, signatures
corresponding to the remaining differentiation states were
significantly enriched in control B cells, suggesting a relative
over-representation of B cells at the FO/MZ and plasma cell
differentiation states in Sca1-HGAL mice with hyperplastic
follicles. To further validate the role of HGAL in these B-cell
differentiation states, we performed a reciprocal analysis of Sca1-
HGAL signatures in transcriptional profiling data from normal
murine B cells22. Genes found to be highly expressed in Sca1-
HGAL B cells compared with control B cells (Fig. 8a) were tested
for enrichment in normal FO/MZ and plasma B cells and
compared with B cells of other differentiation states. This revealed
a significant enrichment of HGAL-induced genes in normal FO/
MZ B cells, but not in normal plasma cells, compared with other
differentiation states (Supplementary Fig. S10a–c). The lack of
significant enrichment in normal plasma cells is likely the result
of downregulation of a subset of HGAL-induced genes within this
differentiation state. Together, these results suggest that Sca1-
HGAL B cells possess transcriptional signatures indicative of an
over-representation of FO/MZ B cells and plasma cells. However,
immunohistochemistry and flow cytometry studies did not show
an increase in plasma cells in the lymphoid hyperplasia observed
in the Sca1-HGAL mice, suggesting that the B220 þcells may be
poised towards plasma cell differentiation; however, induced
transcriptional changes characteristic to plasma cells were
insufficient to lead to plasma cell differentiation. Indeed, the
observed transcriptional signatures lacked expression of PRDM1
and XBP1, genes necessary for plasma cell genesis. Furthermore,
normal murine FO/MZ B cells had significantly enriched
expression of HGAL-induced genes, suggesting that HGAL may
have a role in normal FO/MZ B-cell biology.
We also compared the transcriptional responses evoked by
HGAL in B-cell lymphoid hyperplasia of Sca1-HGAL mice to those
induced in other B-cell tumours at various stages of the B-cell
lineage, with a focus on those arrested at the GC and beyond23,24.
Notably, genes induced in the lymphoid hyperplasia in the Sca1-
HGAL mice significantly overlapped those upregulated in murine
GC-derived lymphomas driven by Bcl6 (ref. 24) (P¼2.48e 4by
hypergeometric test) (Supplementary Fig. S10d) and were also
enriched for those uniformly repressed within plasma cell tumours
driven by Myc/Bcl-xL23 (P¼1.8e 12 by hypergeometric test)
(Supplementary Fig. S10e), suggesting a distinct regulatory
programme in these hyperplasia leading to Ig secretion.
Discussion
The outcome of BCR stimulation is diverse and is determined by
the maturation state of the cell, the nature and quantity of the
IPI00118457 (100%), 13,622.6 Da
Serum amyloid A-2 protein
6 unique peptides, 8 unique spectra, 20 total spectra,
45/122 amino acids (37% coverage)
Probability legend: Kidney Spleen
Bio view:
Identified proteins (665)
Including six decoys
Over 95%
80% to 94%
50% to 79%
20% to 49%
0% to 19%
IPI00323571 36 k Da 61 61
32
32
40
5851
23
34
18
11
10
10
10
10
43
32
17
17
30
19
15
11
11
11
10
6
6
88
20
20
9
7
9
15
20
16
16
16
17
22
23
35
11
1
9
5
5
5
8
8
4
22
55
10
26
12
22
11 11
9
8
6
5
5
13
4
4
2
2
15
15
15
13
12
9
3
2
77
55
59
8
5
6
33
87
7
7
69 k Da
105 k Da
42 k Da
11 k Da
45 k Da
61 k Da
185 k Da
55 k Da
14 k Da
26 k Da
55 k Da
15 k Da
16 k Da
77 k Da
54 k Da
351 k Da
109 k Da
26 k Da
50 k Da
52 k Da
26 k Da
31 k Da
28 k Da
53 k Da
IPI00131695
IPI00119818
IPI00110850
IPI00869381
IPI00377351
IPI00115522
IPI00648042
IPI00129240
IPI00118457
IPI00131674
IPI00279079
IPI00469114
IPI00988950
IPI00139788
IPI00227299
IPI00124959
IPI00283934
IPI00130391
IPI00122312
IPI00320420
IPI00309214
IPI00121209
IPI00420761
IPI00177214
Starred?
#
1 Apolipoprotein E
Accession number
Molecular weight
Kidney sample 1
Kidney sample 2
Spleen sample 1
Spleen sample 2
Serum albumin
Inter alpha-trypsin inhibiton, heavy chain 4
Actin, cytoplasmic 1
Apolipoprotein A-II precursor
Apolipoprotein A-IV
Fibrinogen, alpha polypeptide isoform 2
Zinc finger, CCHC domain containing 11
Vitronectin
Serum amyloid A-2 protein
Trypsinogen 7
Fibrinogen beta chain
Haemoglobin subunit alpha
Haemoglobin subunit beta-1-like
Serotransferrin
Vimentin
Uncharacterized protein
Isoform 1 of uncharacterized protein C17
Uncharacterized protein
Clusterin
Serum amyloid P-component
Apolipoprotein A-I
Putative uncharacterized protein
Iq mu chain C region membrane-bound
Protease, serine, 1
2
4
6
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
7
3
5
Figure 6 | Analysis of amyloid fibrils. (a) Congo red-stained kidney specimen viewed under fluorescent light source with bright red areas representing
amyloid deposits. Purple coloured lines-areas microdissected for MS-based proteomic analysis. (b) Results of MS-based proteomic analysis of amyloid
plaques from the kidney and spleen specimens in two independent microdissections. The identified proteins are listed according to their relative abundance,
with top 25 from a total of 720 proteins shown. The columns show the protein name, the UnitProt identifier (protein accession number in the UniProt
database, http://www.uniprot.org/), the molecular weight of the protein, two microdissections of the kidney specimen and two microdissections of the
spleen specimen. The numbers indicate number of total peptide spectra identified for each protein. The amyloid-associated proteins are identified by yellow
stars. The peptides representing the serum amyloid A-2 protein (red box) are the primary cause of amyloid deposits in the Sca1-HGAL mouse model.
(c) serum amyloid-associated protein-2 (SAA2) protein coverage in the Sca1-HGAL mouse amyloidosis model. The amino-acid (aa) sequence of the
SAA2_MOUSE is provided. The yellow highlights indicate the parts of the protein that were identified by the liquid chromatography tandem mass
spectrometry. The first 19 aa represent the N terminus signal peptide of the SAA2_MOUSE (red box) cleaved before the mature SAA2_MOUSE is secreted
into the circulation, and thus is not expected to be part of the amyloid deposits. The peptides missing in the middle of the protein are unlikely to be detected
by liquid chromatography tandem mass spectrometry due to abundance of numerous trypsin cutting sites.
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2334
8NATURE COMMUNICATIONS | 4:1338 | DOI: 10.1038/ncomms2334 | www.nature.com/naturecommunications
&2013 Macmillan Publishers Limited. All rights reserved.

antigen, signals from coreceptors and the magnitude and
duration of BCR signalling. An ‘on-off’ mode of BCR signal
transduction would not allow for diverse cellular responses. Fine-
tuning the magnitude and duration of BCR signalling and
preferential activation of specific downstream signalling compo-
nents by adaptor proteins selectively expressed during B-cell
development is of paramount importance for B-cell functional
plasticity25,26. We demonstrate that the GC-specific adaptor
HGAL regulates BCR signalling by direct binding and activation
of Syk. While HGAL is not necessary for GC formation5, HGAL
transgenic animals exhibited increased-size Peyer’s patches, a
finding corroborating the reduced-size Peyer’s patches in the M17
knockout mice5. This observation suggests a specific function of
HGAL in Peyer’s patches. Furthermore, it is possible that the
increased-size HGAL expressing Peyer‘s patches may contribute
to systemic lymphoid hyperplasia and subsequent development of
AA amyloidosis, which was never observed in the absence of
lymphoid hyperplasia. HGAL-mediated enhancement of BCR
signalling may be important for efficient and timely completion
of the GC reaction and selective activation of signalling pathways
(for example, extracellular signal-regulated kinase and p38 but
not Jun N-terminal kinases). Based on the presented mouse
model, HGAL expression predisposes to lymphoid hyperplasia,
polyclonal hypergammaglobulinemia and enhanced proliferation
in response to BCR stimulation, culminating in systemic AA
amyloidosis. B-lymphocytes in HGAL transgenic mice showed
enhanced activation of Syk and RhoA signalling, suggesting this
model recapitulates the biological effects of HGAL protein in
human B cells. The observed polyclonal lymphoid hyperplasia
was characterized by an accumulation of FO lymphocytes in the
absence of GC lymphocytes and terminally differentiated plasma
cells. The FO B cells in the lymphoid hyperplasia exhibited
transcriptional signatures characteristic to FO/MZ B cells and
plasma cells and were accompanied by polyclonal hyper-
gammaglobulinemia in the absence of mature plasma cells,
suggesting a block in terminal plasma cell differentiation whose
nature is currently unknown. Studies examining HGAL’s effects
on transcriptional signatures in human B cells are ongoing. In
addition, whether the observed phenotype was due to expression
of HGAL in the transgenic lines starting from progenitor and not
specifically in the GC cells will need to be addressed.
B-cell activation in the Sca1-HGAL mice led to reactive
systemic AA amyloidosis not observed in the control animals.
The molecular features and anatomical location of the observed
amyloid deposits precisely recapitulated human AA amyloidosis
complicating chronic inflammatory diseases (for example,
rheumatoid arthritis). In these diseases reactive FO hyperplasia
and lymphadenopathy are commonly observed27 and frequently
precede development of systemic amyloidosis. Heretofore,
temporary AA amyloid deposits were induced in mice by
exogenous agents (for example, casein, lipopolysaccharide), that
would resolve without continued stimulation. Progressive and
permanent AA amyloid deposits were observed in interleukin-6
and serum amyloid A activating factor 1 transgenic mice starting
0
200
400
600
800
1,000
499,600.0
998,600.0
Antibody concentration
(μg ml–1)
*
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
RhoA activity (OD490)
pSyk (pY352)
Syk
Actin
Control
Sca1-HGAL
Control
α-IgM
Sca1-HGAL
kDa
72
72
42
80,000
*** ***
60,000
40,000
20,000
0
0
–
+
Control-1
Control-2
Control-3
Scal-HGAL-1
Scal-HGAL-3
Scal-HGAL-2
0.5 μg ml
–1
anti-lgM
2 μg ml
–1
anti-lgM
8 μg ml
–1
anti-lgM
100
80
60
40
Fluo-4 AM
c.p.m. ([3H] thymidine
incorporation)
20
0
0200 400 600 800 1K
Time (sec)
IgA
IgG1 IgG2a IgG2b IgG3 IgM
You ng Sca1-HGAL
Hyperplasic Sca1-HGAL
Control wild-type
Figure 7 | Hypergammaglobulinemia and enhanced BCR and RhoA signalling in the Sca1-HGAL mice. (a) Ig isotype titres in the serum of 14-month-old
unimmunized control (open circles) and Sca1-HGAL (black circles) mice analysed by enzyme-linked immunosorbent assay. n¼5 per group. * Indicates
P¼0.0007 by two-tailed Student’s t-test. (b)Ex vivo B-cell proliferation in response to surface Ig stimulation. Purified splenic B cells from wild-type mice,
young Sca1-HGAL transgenic mice without lymphoid hyperplasia and Sca1-HGAL transgenic mice with lymphoid hyperplasia (three animals in each group)
were stimulated with the indicated concentrations of an anti-IgM antibody and cell proliferation assessed by [3H]thymidine incorporation in triplicates.
Shown is mean±s.e.m. *** Indicates statistically significant difference: for 0.5 mgml1anti-IgM P¼0.000004 and for 2 mgml1anti-IgM
P¼0.00000007 by two-way analysis of variance test. (c) Kinetic of calcium mobilization measured by flow cytometry in purified spleen B-lymphocytes
from three Sca1-HGAL transgenic mice with lymphoid hyperplasia (black) and three wild-type control mice (grey). Arrow indicates the time point of
stimulation with goat F(ab’)2 anti-mouse IgM. (d) Purified spleen B-lymphocytes from Sca1-HGAL transgenic or control mice stimulated with goat F(ab’)2
anti-mouse IgM for 1 min were used for western blot with pSyk (pY352) and total Syk antibodies. Equal loading was confirmed by immunoblotting with
actin antibodies. (e) Purified spleen B cells from Sca1-HGAL transgenic or wild-type control mice stimulated with 1 mgml1lysophosphatidic acid for
45 seconds were used for RhoA activity measurement in triplicates; lysis buffer served as negative control and RhoA protein as a positive control. Shown is
mean±s.d. Results in dand eare representative of three independent experiments.
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2334 ARTICLE
NATURE COMMUNICATIONS | 4:1338 | DOI: 10.1038/ncomms2334 | www.nature.com/naturecommunications 9
&2013 Macmillan Publishers Limited. All rights reserved.

form 3–14 months of age28,29. In these models interleukin-6 and
serum amyloid A activating factor directly stimulated serum
amyloid A production, bypassing the preceding chronic
inflammatory state observed in human disease. Conversely,
HGAL mice develop AA amyloid deposits in a similar
timeframe and show systemic lymphoid hyperplasia and
hypergammaglobulinemia preceding amyloidosis onset,
therefore, authentically recapitulating human amyloidosis
associated with chronic inflammatory disorders. These mice
may thus serve as an excellent model for this disease.
The polyclonal lymphoid hyperplasia observed in the Sca1-
HGAL mice were similar to the lymphoid hyperplasia detected in
the BCL2 transgenic mice models, which also displayed
polyclonal hypergammaglobulinemia. However, the mechanisms
underlying these phenomena were different (increased cell
proliferation and decreased apoptosis, respectively)30.
Additional differences between these models are also noticeable.
BCL2 transgenic mice develop lymphoid hyperplasia earlier than
Sca1-HGAL mice, display autoimmune complex glomeru-
lonephritis and a proportion of these animals develop
Wild-type Sca1-HGAL
lghg1
2Wild-type
Sca1-HGAL
1
0
–1
–2
120 82.3% 17.7%
90
60
30
0
120
90
60
30
0
102103104105
102
Cell surface CD80 expression
Enriched in control Enriched in Sca1-HGAL FDR
0.002
<0.001
0.049
0.188
0.196
0.043
Pro/pre
Trans.
Foll./MZ
GCB
P.blast
P.cell
–2
0.30
0.20
0.10
0
–0.10
–0.20
–0.30
0.30
0.35
0.20
0.25
0.15
0.10
0.05
–0.05
–0.10
–0.15
Enriched in Sca1- HGAL Enriched in Sca1- HGAL
Foll./MZ signature
Enrichment score
FDR = 0.049
Plasma cell signature
FDR = 0.043
Normalized enrichment score
–1 012
No. among B220+ cells
103104105
Relative expressiorn (log2)
lgh2 lghg1 lghg2b Cd80
lghg2b
Cd80
Anxa2
lgh-2 [lgA]
Mx1
Rgs13
Rgs18
Cd93
Myof
Mx2
58.3% 41.7%
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2334
10 NATURE COMMUNICATIONS | 4:1338 | DOI: 10.1038/ncomms2334 | www.nature.com/naturecommunications
&2013 Macmillan Publishers Limited. All rights reserved.

lymphomas that were not detected in the Sca1-HGAL mice30–33.
In contrast, infiltration of normal organs by B-lymphocytes and
amyloidosis was observed only in the Sca1-HGAL mice.
HGAL is expressed in the GC-derived B-cell lymphomas, but
whether it has a role in lymphomagenesis is unknown. The Sca1-
HGAL mice developed lymphoid hyperplasia, but not lympho-
mas. However, it is possible that HGAL may facilitate the
transformation process. Gene expression arrays demonstrated
significant overlap between genes upregulated in the lymphoid
hyperplasia of the Sca1-HGAL mice and lymphomas driven by
Bcl6/lMyc24. It is possible that the BCR-mediated proliferation
detected in the Sca1-HGAL mice may cooperate with the
antiapoptotic effects induced by BCL2 or differentiation arrest
induced by BCL6 during the transformation process. Previous
studies demonstrated similar cooperation between BCR and Myc
in the genesis of B-cell lymphomas34. Use of the Sca1-HGAL
mouse model to generate bitransgenic mice will help to elucidate
the potential role of HGAL in lymphomagenesis.
Methods
Mice and cell culture. Human non-Hodgkin lymphoma cell lines Raji, VAL, BJAB
and HBL-1 were grown as previously reported35. Human B-lymphocytes were
isolated from healthy donor blood using EasySep Human B cell enrichment kit
(StemCell Technologies, Vancouver, BC, Canada).
A Sca1-HGAL plasmid in which HGAL cDNA is under the control of the mouse
Ly-6E.1 promoter was generated as follows: the fragment containing the human
HGAL-V5 cDNA, was inserted into the ClaI site of the pLy6 vector15, resulting in
Sca1-HGAL vector. The transgene fragment (Fig. 3a) was excised from the vector
by restriction digestion with NotI, purified for injection (2 ng ml1) and injected
into CBAxC57BL/6J fertilized eggs. Transgenic mice were identified by Southern
analysis of tail snip DNA after EcoRI digestion. HGAL cDNA was used as a probe
for detection of the transgene. The transgene copy number was determined by
Southern blot analysis using copy standards prepared by mixing non-transgenic tail
DNA with a known amount of transgene DNA as described in Camper36.
A total of 47 and 28 animals derived from founder lines 102A and 102B,
respectively, housed in pathogen-free conditions were used for the analyses. There
were no differences between the founder lines. All animal experiments were
performed in accordance with the guidelines of the Committees on Animal
Research at University of Salamanca and University of Miami, and the National
research Council.
For cell sorter separation, B220 and CD22 lineage markers were used to purify
cells from the spleen of both Sca1-HGAL or control wild-type mice by
fluorescence-activated cell sorting (FACS) (FACSVANTAGE; Becton Dickinson).
Purity of the sorted cells was over 98%, as determined by FACS reanalysis.
Alternatively, mouse B-lymphocytes were isolated from mice spleen using EasySep
mouse B-cell enrichment kit (StemCell Technologies).
For analysis of B1 and B2 lymphocyte populations, single-cell suspensions were
obtained from the spleens, LN, BM and peritoneal cavities of individual Sca1-
HGAL and control mice. To obtain peritoneal cells, 6 ml of ice-cold PBS 1% fetal
calf serum (FCS) were injected into the peritoneal cavity using a 25-g needle
attached to a syringe. After injection, the peritoneum was gently massaged and cells
were collected using a 5-ml syringe. Cells were washed with PBS 1% FCS and
incubated for at least 5 min with CD16/CD32 Fc Block (clone 2.4G2) to minimize
background staining. A total of 2 106cells were stained for 30 min on ice with
IgD-FITC (clone 11-26c.2a), CD5-PE (clone 53–7.3), Mac1-PerCP-Cy5.5 (clone
M1/70) and IgM-APC (II/41) to identify the B2 and B1 cell subsets. All the
antibodies were purchased from the BD Pharmingen. At least 1 105cells were
acquired using a BD C6 Accuri (BD Biosciences) and analysed by the Flowjo
software (Tree Star).
Antibodies, plasmids and chemicals, as well as procedures for cell tranfection,
western blotting, immunoprecipitation, RhoA activity, GST pull-down and
luciferase reporter assays are described in the Supplementary Methods.
Flow cytometry and immunofluorescence microscopy. Nucleated cells were
obtained from total BM (flushing from the long bones), peripheral blood, thymus,
LN and spleen. Contaminating red blood cells were lysed with RCLB lysis buffer
and the remaining cells were then washed in PBS with 1% FCS. After staining, all
cells were washed once in PBS with 1% FCS containing 2mgml1propidium
iodide (PI) to allow dead cells to be excluded from both analyses and sorting
procedures. The samples and the data were analysed in a FACSCalibur using
CellQuest (Becton Dickinson) or FlowJo (Tree Star) software. Specific fluorescence
of FITC, PE, PI and allophycocyanin excited at 488 nm (0.4 W) and 633 nm
(30 mW), respectively, as well as known forward and orthogonal light scattering
properties of mouse cells were used to establish gates. For each analysis a total of at
least 50,000 viable (PI ) cells were assessed.
FACS definition of B-cell developmental stages was performed according to
Kwon et al.37 with minor modifications: BM pro-B cells (CD19 þc-Kitþ), BM pre-
B cells (B220 þCD25 þIgM–), BM immature B-cells (B220 þIgMhiIgD ), BM
recirculating B-cells (B220 þIgDhi), peripheral transitional B-cells
(B220 þIgMhiIgDhi), peripheral mature-B cells (B220 þIgMloIgDhi), MZ B cells
(B220 þCD21hiCD23lo), FO B cells (B220 þCD21intCD23hi).
The subcellular localization of HGAL and Syk was assessed by stimulated
emission depletion fluorescence nanoscope images38, as described in detail in the
Supplementary Methods, together with the methodology for the immunohisto-
chemistry and analysis of amyloid fibril protein.
Gene expression analysis. Using flow cytometry, CD22 þB220 þsplenocytes
from adult Sca1-HGAL mice with hyperplastic follicles and wild-type littermates
(four samples each) were sorted to B98% purity, and profiled the corresponding
gene expression patterns using Affymetrix GeneChip Mouse Gene 1.0 ST Arrays.
Raw array microarray data files were submitted to GEO and are available under the
accession number GSE35400. For probeset summarization, microarray CEL data
files were subjected to a custom Chip Definition File (MoGene10stv1_Mm_EN-
TREZG_13.0.0) with a mapping of probes to Entrez Gene Identifiers as previously
described39. Gene level data were normalized using robust multichip average
within the BioConductor environment40. Significance analysis of microarrays21 at a
10% FDR was used to identify the probe sets with significant differential expression
between experimental conditions. We used DAVID41, Ingenuity Pathway Analysis
tools (IPA 9/2011 Release), and the Molecular Signatures Database (MSigDB,
v3.0)42, to assess whether differentially expressed genes show functional
enrichment.
For assessment of differentiation state, normal murine B-cell differentiation state
signatures corresponding to Pro-B/Pre-B, transitional, FO/MZ, GC, plasmablast
and plasma cell stages of differentiation were used as previously defined22. Over-
representation of differentiation states within Sca1-HGAL compared with control
B cells was inferred by assessing comparative enrichment of normal differentiation
state signatures using GSEA42. GSEA was performed using a weighted enrichment
statistic, signal-to-noise ratio ranking metric, and corrected for multiple hypothesis
testing using 10,000 permutations to provide a multiple hypothesis testing-
corrected FDR q-value. Reciprocal analysis was performed in order to validate
positive enrichments within Sca1-HGAL B cells. Therein, publicly available gene
expression data of normal murine B-cell differentiation states (GSE26408) was used
to probe for enrichment of HGAL-induced genes within the FO/MZ and plasma
cell differentiation states compared with all other differentiation states (1-versus-all
analysis) using GSEA.
Figure 8 | Transcriptional signatures evoked in vivo within B-cell lymphoid hyperplasia in the Sca1-HGAL mice. (a) Genes significantly induced or
repressed within purified B220 þsplenocytes of Sca1-HGAL mice in comparison with wild-type littermates, as determined by significance analysis of
microarrays using an FDR 10% and demonstrating at least threefold change. Each row represents a separate gene and each column a separate mRNA
sample from each of eight mice, with four mice from within each group. The level of expression of each gene in each sample, relative to the mean level of
expression of that gene across all the samples, is represented using a red–green colour scale, extending from fluorescence ratios of 0.25 to 4 ( 2to þ2in
log base 2 units). Selected genes are highlighted. (b) Depicted are the relative mRNA expression levels for selected genes exhibiting significant differential
expression from a, including Igh-2 (IgA), Ighg1 (IgG1), Ighg2b (IgG2b) and Cd80 (CD80). These genes were independently assessed for differential
expression of the corresponding proteins (in parentheses), as depicted in Fig. 7a, and panel c.(c) Induction of CD80 cell surface protein expression in B-cell
splenocytes of the Sca1-HGAL mice measured by flow cytometry. Data are representative of three mice. (d) GSEA of the relative representation of
differentiation states within Sca1-HGAL compared with control B-cells. Gene expression data from Sca1-HGAL B cells showed significant enrichment for
signatures corresponding to both FO/MZ and plasma cell normal murine B-cell differentiation states, suggesting an over-representation of these states
within Sca1-HGAL cells. Signatures corresponding to other differentiation states were significantly enriched in control B cells, suggesting a relative under-
representation in Sca1-HGAL B cells. (e) Detailed view of GSEA results, showing a strong bias of genes from normal FO/MZ B cells (left panel) and plasma
cells (right panel) in Sca1-HGAL B cells compared with control B cells.
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2334 ARTICLE
NATURE COMMUNICATIONS | 4:1338 | DOI: 10.1038/ncomms2334 | www.nature.com/naturecommunications 11
&2013 Macmillan Publishers Limited. All rights reserved.

Statistical analysis. To test the differences in responses between cells, we used the
two-way analysis of variance test for ex vivo B-cell proliferation assay and two-
tailed Student’s t-test for remaining experiments. Mice survival curves were esti-
mated using the product-limit method of Kaplan–Meier and were compared using
the log-rank test. P-values o0.05 were considered statistically significant.
References
1. Lossos, I. S., Alizadeh, A. A., Rajapaksa, R., Tibshirani, R. & Levy, R. HGAL is a
novel interleukin-4-inducible gene that strongly predicts survival in diffuse
large B-cell lymphoma. Blood 101, 433–440 (2003).
2. Pan, Z. et al. Studies of a germinal centre B-cell expressed gene, GCET2,
suggest its role as a membrane associated adapter protein. Br. J. Haematol. 137,
578–590 (2007).
3. Natkunam, Y. et al. Expression of the human germinal center-associated
lymphoma (HGAL) protein, a new marker of germinal center B-cell derivation.
Blood 105, 3979–3986 (2005).
4. Natkunam, Y. et al. Expression of the human germinal center-associated
lymphoma (HGAL) protein identifies a subset of classic Hodgkin lymphoma of
germinal center derivation and improved survival. Blood 109, 298–305 (2007).
5. Schenten, D., Egert, A., Pasparakis, M. & Rajewsky, K. M17, a gene specific for
germinal center (GC) B cells and a prognostic marker for GC B-cell
lymphomas, is dispensable for the GC reaction in mice. Blood 107, 4849–4856
(2006).
6. Lu, X. et al. HGAL, a lymphoma prognostic biomarker, interacts with the
cytoskeleton and mediates the effects of IL-6 on cell migration. Blood 110,
4268–4277 (2007).
7. Jiang, X. et al. HGAL, a germinal center specific protein, decreases
lymphoma cell motility by modulation of the RhoA signaling pathway. Blood
116, 5217–5227 (2010).
8. Lu, X. et al. Germinal center-specific protein human germinal center associated
lymphoma directly interacts with both myosin and actin and increases the
binding of myosin to actin. FEBS. J. 278, 1922–1931 (2011).
9. Niiro, H. & Clark, E. A. Regulation of B-cell fate by antigen-receptor signals.
Nat. Rev. Immunol. 2, 945–956 (2002).
10. Sada, K., Takano, T., Yanagi, S. & Yamamura, H. Structure and function of Syk
protein-tyrosine kinase. J. Biochem. 130, 177–186 (2001).
11. Turner, M. et al. Perinatal lethality and blocked B-cell development in mice
lacking the tyrosine kinase Syk. Nature 378, 298–302 (1995).
12. Cheng, A. M. et al. Syk tyrosine kinase required for mouse viability and B-cell
development. Nature 378, 303–306 (1995).
13. Koretzky, G. A., Abtahian, F. & Silverman, M. A. SLP76 and SLP65: complex
regulation of signalling in lymphocytes and beyond. Nat. Rev. Immunol. 6,
67–78 (2006).
14. Jiang, A., Craxton, A., Kurosaki, T. & Clark, E. A. Different protein tyrosine
kinases are required for B cell antigen receptor-mediated activation of
extracellular signal-regulated kinase, c-Jun NH2-terminal kinase 1, and p38
mitogen-activated protein kinase. J. Exp. Med. 188, 1297–1306 (1998).
15. Miles, C., Sanchez, M. J., Sinclair, A. & Dzierzak, E. Expression of the Ly-6E.1
(Sca-1) transgene in adult hematopoietic stem cells and the developing mouse
embryo. Development 124, 537–547 (1997).
16. Fossati, V., Kumar, R. & Snoeck, H. W. Progenitor cell origin plays a role in fate
choices of mature B cells. J. Immunol. 184, 1251–1260 (2010).
17. Perez-Caro, M. et al. Cancer induction by restriction of oncogene expression to
the stem cell compartment. EMBO. J. 28, 8–20 (2009).
18. Vicente-Duen
˜asa, C. et al. Expression of MALT1 oncogene in hematopoietic
stem/progenitor cells recapitulates the pathogenesis of human lymphoma in
mice. Proc. Natl Acad. Sci. USA 109, 10534–10539 (2012).
19. Vicente-Duenas, C. et al. A novel molecular mechanism involved in multiple
myeloma development revealed by targeting MafB to haematopoietic
progenitors. EMBO. J. 31, 3704–3717 (2012).
20. van der Hilst, J. C. Recent insights into the pathogenesis of type AA
amyloidosis. ScientificWorld Journal 11, 641–650 (2011).
21. Tusher, V. G., Tibshirani, R. & Chu, G. Significance analysis of microarrays
applied to the ionizing radiation response. Proc. Natl Acad. Sci. USA 98,
5116–5121 (2001).
22. Green, M. R. et al. Signatures of murine B-cell development implicate Yy1 as a
regulator of the germinal center-specific program. Proc. Natl Acad Sci. USA
108, 2873–2878 (2011).
23. Boylan, K. L. M. et al. A transgenic mouse model of plasma cell malignancy
shows phenotypic, cytogenetic, and gene expression heterogeneity similar to
human multiple myeloma. Cancer. Res. 67, 4069–4078 (2007).
24. Pasqualucci, L. et al. AID is required for germinal center-derived
lymphomagenesis. Nat. Genet. 40, 108–112 (2008).
25. Marshall, A. J. et al. A novel B lymphocyte-associated adaptor protein, Bam32,
regulates antigen receptor signaling downstream of phosphatidylinositol 3-
kinase. J. Exp. Med. 191, 1319–1332 (2000).
26. Shlapatska, L. M. et al. CD150 association with either the SH2-containing
inositol phosphatase or the SH2-containing protein tyrosine phosphatase is
regulated by the adaptor protein SH2D1A. J. Immunol. 166, 5480–5487 (2001).
27. Kojima, M., Motoori, T. & Nakamura, S. Benign, atypical and malignant
lymphoproliferative disorders in rheumatoid arthritis patients. Biomed.
Pharmacother. 60, 663–672 (2006).
28. Solomon, A. et al. Transgenic mouse model of AA amyloidosis. Am. J. Pathol.
154, 1267–1272 (1999).
29. Ray, A., Shakya, A., Kumar, D., Benson, M. D. & Ray, B. K. Inflammation-
responsive transcription factor SAF-1 activity is linked to the development of
amyloid A amyloidosis. J. Immunol. 177, 2601–2609 (2006).
30. McDonnell, T. J. et al. Bcl-2-immunoglobulin transgenic mice demonstrate
extended B cell survival and follicular lymphoproliferation. Cell 57, 79–88
(1989).
31. Xiang, H. et al. The immunoglobulin heavy chain gene 3’ enhancers induce
Bcl2 deregulation and lymphomagenesis in murine B cells. Leukemia 25,
1484–1493 (2011).
32. Egle, A., Harris, A. W., Bath, M. L., O’Reilly, L. & Cory, S. VavP-Bcl2 transgenic
mice develop follicular lymphoma preceded by germinal center hyperplasia.
Blood 103, 2276–2283 (2004).
33. Strasser, A. et al. Enforced BCL2 expression in B-lymphoid cells prolongs
antibody responses and elicits autoimmune disease. Proc. Natl Acad. Sci. USA
88, 8661–8665 (1991).
34. Refaeli, Y. et al. The B cell antigen receptor and overexpression of MYC can
cooperate in the genesis of B cell lymphomas. PLoS. Biol. 6, e152 (2008).
35. Sarosiek, K. A. et al. Novel IL-21 signaling pathway up-regulates c-Myc and
induces apoptosis of diffuse large B-cell lymphomas. Blood 115, 570–580
(2010).
36. Camper, S. A. Research applications of transgenic mice. Biotechniques 5,
638–650 (1987).
37. Kwon, K. et al. Instructive role of the transcription factor E2A in early B
lymphopoiesis and germinal center B cell development. Immunity 28, 751–762
(2008).
38. Willig, K. I., Harke, B., Medda, R. & Hell, S. W. STED microscopy with
continuous wave beams. Nat. Methods 4, 915–918 (2007).
39. Dai, M. et al. Evolving gene/transcript definitions significantly alter the
interpretation of GeneChip data. Nucleic Acids Res. 33, e175 (2005).
40. Irizarry, R. A. et al. Summaries of affymetrix genechip probe level data. Nucleic
Acids Res. 31, e15 (2003).
41. Dennis, Jr G. et al. DAVID: database for annotation, visualization, and
integrated discovery. Genome Biol. 4, P3 (2003).
42. Subramanian, A. et al. Gene set enrichment analysis: a knowledge-based
approach for interpreting genome-wide expression profiles. Proc. Natl Acad.
Sci. USA 102, 15545–15550 (2005).
Acknowledgements
Grant support: I.S.L. is supported by National Institutes of Health (NIH) grants NIH
CA109335 and NIH CA122105, and the Dwoskin Family Foundations. Y.N. is supported
by NIH P01 CA34233. Research in ISG group was partially supported by FEDER and by
MICINN (SAF2009-08803 and SAF2012-32810 to ISG), by Junta de Castilla y Leo
´n
(REF. CSI007A11-2 and Proyecto Biomedicina 2009-2010), by MEC OncoBIO Con-
solider-Ingenio 2010 (Ref. CSD2007-0017), by Sandra Ibarra Foundation, by Group of
Excellence Grant (GR15) from Junta de Castilla y Leon and the ARIMMORA project
(FP7-ENV-2011, European Union Seventh Framework Programme), and by Proyecto en
Red de Investigacio
´n en Celulas Madre Tumorales en Cancer de Mama, supported by
Obra Social Kutxa y Conserjerı
´a de Sanidad de la Junta de Castilla y Leon.
Author contributions
I.R.-C, X.J., Y.N.-performed experiments and analysed the data and wrote the paper; X.L.,
C.V., I.G.-H., T.F., J.L.G., G.M., C.K., S.Z., V.S., L.F., J.A.M.-C., F.J.G.-C., J.D.T., A.D.,
E.C.-S., M.R.G., A.A.A., C.C.-performed experiments and analysis of data; I.S.-G. and
I.S.L.- designed the study, analysed the data and wrote the paper. All the authors
reviewed the manuscripts and agree with its content.
Additional information
Supplementary Information accompanies this paper at http://www.nature.com/
naturecommunications
Competing financial interests: The authors declare no competing financial interests.
Reprints and permissions information is available at http://npg.nature.com/
reprintsandpermissions
How to cite this article: Romero-Camarero, I. et al. Germinal centre protein HGAL
promotes lymphoid hyperplasia and amyloidosis via BCR-mediated Syk activation.
Nat. Commun. 4:1338 doi: 10.1038/ncomms2334 (2013).
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2334
12 NATURE COMMUNICATIONS | 4:1338 | DOI: 10.1038/ncomms2334 | www.nature.com/naturecommunications
&2013 Macmillan Publishers Limited. All rights reserved.