вход по аккаунту


Decreased Lyn expression and translocation to lipid raft signaling domains in B lymphocytes from patients with systemic lupus erythematosus.

код для вставкиСкачать
Vol. 52, No. 12, December 2005, pp 3955–3965
DOI 10.1002/art.21416
© 2005, American College of Rheumatology
Decreased Lyn Expression and Translocation to
Lipid Raft Signaling Domains in B Lymphocytes From
Patients With Systemic Lupus Erythematosus
Fabian Flores-Borja,1 Panagiotis S. Kabouridis,1 Elizabeth C. Jury,2 David A. Isenberg,2
and Rizgar A. Mageed1
Objective. B lymphocytes from patients with systemic lupus erythematosus (SLE) are hyperactive and
produce anti–double-stranded DNA (anti-dsDNA) autoantibodies. The cause or causes of B cell defects in SLE
are unknown. In this study, we determined the level and
subcellular distribution of Lyn protein, a key negative
regulator of B cell receptor signaling, and assessed
whether altered Lyn expression is characteristic of B
cells in the setting of SLE.
Methods. Negative selection was used to isolate B
lymphocytes from blood. Lipid raft signaling domains
were purified from B cells obtained from 62 patients
with SLE, 15 patients with rheumatoid arthritis, and 31
healthy controls, by gradient ultracentrifugation. The
total Lyn protein level was determined by Western
blotting, confocal microscopy, and fluorescein-activated
cell sorting (FACS). The distribution of Lyn into lipid
raft and nonlipid raft domains was determined by
Western blotting and confocal microscopy. Lyn content
in B cell subpopulations was determined by FACS. In
order to assess B lymphocyte activity, we used 3Hthymidine incorporation and enzyme-linked immunosorbent assay to measure spontaneous proliferation
and IgG and cytokine production by B cells.
Results. This study revealed that B lymphocytes
from a majority of patients with SLE have a reduced
level of Lyn and manifest altered translocation to lipid
rafts. An investigation into the mechanisms of Lyn
reduction suggested that increased ubiquitination is
involved. This was evident from increased ubiquitination of Lyn and translocation of c-Cbl into lipid rafts.
Studies of B cell responses showed that altered Lyn
expression was associated with heightened spontaneous
proliferation, anti-dsDNA autoantibodies, and increased interleukin-10 production.
Conclusion. This study provides evidence for altered Lyn expression in B cells from a majority of
patients with SLE. Altered Lyn expression in SLE may
influence the B cell receptor signaling and B cell hyperactivity that are characteristic of the disease.
Patients with systemic lupus erythematosus
(SLE) manifest immunologic abnormalities that include
spontaneous B lymphocyte proliferation, hyperresponsiveness to physiologic stimuli, and altered pattern of
production of and responses to cytokines (1–3). One
consequence of these abnormalities is the production of
pathogenic autoantibodies to nuclear antigens, including
double-stranded DNA (dsDNA). Although anti-dsDNA
autoantibodies have features indicating T lymphocyte–
driven responses, several studies suggest that the production of anti-dsDNA autoantibodies could also be
attributable to intrinsic B cell defects (4). Furthermore,
studies in gene-deficient mice have revealed that defects
in negative regulators of B cell receptor (BCR) signaling, CD22, Fc␥ receptor II (Fc␥RII), or CD72, result in
production of anti-dsDNA autoantibodies (5–7). In patients with SLE, there is evidence that a reduction in
expression of Fc␥RIIA and CD22 relates to anti-dsDNA
production (8,9). However, the molecular basis for the
relationship between BCR signaling defects and SLE
immunopathology remains unclear.
Supported by the Arthritis Research Campaign, UK. Dr.
Kabouridis’s work was supported by a Wellcome Trust Career fellowship (058408).
Fabian Flores-Borja, PhD, Panagiotis S. Kabouridis, PhD,
Rizgar A. Mageed, PhD: William Harvey Institute, Queen Mary
School of Medicine and Dentistry, London, UK; 2Elizabeth C. Jury,
PhD, David A. Isenberg, MD, FRCP: University College London,
London, UK.
Address correspondence and reprint requests to Rizgar A.
Mageed, PhD, Bone and Joint Research Unit, William Harvey Institute, Charterhouse Square, London EC1M 6BQ, UK. E-mail:
Submitted for publication February 24, 2005; accepted in
revised form August 12, 2005.
The BCR comprises multiple components, including membrane-bound immunoglobulins and the Ig␣
and Ig␤ proteins, which contain conserved motifs known
as immunoreceptor tyrosine-based activation motifs
(ITAMs) (10). Signaling through the BCR is initiated
upon crosslinking of membrane-bound immunoglobulins by antigen. This induces receptor aggregation and
Ig␣/Ig␤ ITAM phosphorylation by Lyn, Fyn, or Blk (11).
Lyn, however, also plays a negative regulatory role by
phosphorylating immunoreceptor tyrosine-based inhibitory motifs and recruiting and activating other negative
regulators of signaling (12). Phosphorylation of ITAMs
results in phosphorylation of Syk, which activates the B
cell linker protein and downstream signaling cascades,
leading to gene transcription (13). Thus, Lyn mediates
proximal signaling through Syk activation and recruitment, but in the absence of Lyn, receptor Syk complexes
are retained in the membrane, resulting in enhanced
activation of nuclear factor of activated T cells and B
lymphocyte hyperactivity (14).
Studies of lyn⫺/⫺ mice have revealed that B cells
show increased Ca⫹2 flux in response to antigen stimulation and proliferate spontaneously in the absence of
antigen (12). Furthermore, sustained activation of Lyn
in normal mice leads to lupus-like autoimmunity (15).
Liossis et al reported reduced Lyn messenger RNA
(mRNA) in blood mononuclear cells from patients with
SLE (16). However, because monocytes in mononuclear
cells also express Lyn, the study by Liossis et al has not
assigned this defect to B lymphocytes. Huck et al also
showed a decrease in total Lyn level and an increase in
CD45 protein expression in B cells from patients with SLE
The current model of BCR signaling suggests
that proximal signaling occurs in lipid raft domains (18).
Lipid rafts are specialized membrane domains enriched
in glycosphingolipids and cholesterol, which function as
platforms for the regulation of signaling and membrane
trafficking (19). The aim of the current study was to
determine the level of Lyn expression and translocation
to lipid rafts in B cells from patients with SLE and to
establish whether alterations in Lyn are a characteristic
feature of B lymphocytes in this disease. The results
reveal alterations in Lyn expression in SLE, which may
be due, at least partly, to increased ubiquitination.
Patients and controls. The study group comprised 62
patients with SLE (57 women and 5 men; mean age 36 years
[range 19–66 years]). Disease activity was assessed using the
British Isles Lupus Assessment Group (BILAG) index (20), in
which disease activity in each of 8 organs/systems is assigned a
score from A (disease is most active) to E (disease is not and
has never been active). A global score is determined by
assigning points to the grades, as follows: A ⫽ 9, B ⫽ 3, C ⫽
1, and D/E ⫽ 0. The BILAG index correlates well with clinical
disease activity and with other scoring systems such as the SLE
Disease Activity Index and the Systemic Lupus Activity Measure (21–23). According to the BILAG index, patients with a
global score of ⱖ6 have active disease, while those with a score
of ⱕ5 have inactive disease. Of the 62 SLE patients, 28 (45%)
had active disease, and 34 (55%) had inactive disease. B cells
from 31 healthy individuals (27 women and 4 men, mean age
34 years [range 19–61 years]) and 15 patients with RA (12
women and 3 men, mean age 53 years [range 36–70 years])
were also studied. The effect of medication was assessed by
dividing the patients into 2 groups. Group A comprised
patients receiving no treatment and patients being treated with
nonsteroidal antiinflammatory drugs, antimalarial drugs,
and/or low-dose steroids (⬍10 mg/day). Group B comprised
patients receiving high-dose steroids (⬎10 mg/day) with or
without immunosuppressive agents. All patients and control
subjects gave informed consent, and the study was approved by
the local ethics committee.
Antibodies. Monoclonal antibodies to human CD3,
CD4, CD8, CD14, CD16, and CD56 were purified from
supernatants of hybridoma lines (clones OKT3, RTF4, RTF8,
AML, 3G8, and B159). Purified rabbit and mouse anti-Lyn,
rabbit anti–c-Cbl, and anti–flotillin 2 antibodies were from
Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit
antiphosphotyrosine–Src (P-Y396) and antiphosphorylated
Lyn (P-Y507) antibodies were from New England Biolabs
(Hertfordshire, UK). Mouse anti–␤-actin, anti-rabbit IgG–
biotin, and horseradish peroxidase (HRP)–conjugated avidin
were from Sigma (St. Louis, MO). Monoclonal antibodies to
the lipid raft–associated non–T cell activation linker (NTAL)
molecule were provided by Dr. B. Schraven (Magdeburg,
Germany) (24). HRP-conjugated rabbit anti-mouse and goat
anti-rabbit antibodies were from Dako (Cambridge, UK).
Mouse anti-CD20, peridin chlorophyll protein–streptavidin,
and fluorescein isothiocyanate (FITC)–conjugated anti-CD3
(CD3–FITC), CD14–FITC, CD19–FITC, allophycocyanin
(APC)–conjugated CD19, phycoerythrin (PE)–conjugated
CD27 (CD27–PE), CD69–PE, and CD86–PE were from BD
Biosciences (Oxford, UK). Biotin–goat anti-mouse IgG2b was
from Southern Biotechnology (Birmingham, AL). Goat
F(ab⬘)2 anti-human IgG/IgA/IgM (anti-IgGAM) was from
Cappel (Aurora, OH).
B lymphocyte isolation. Negative isolation of B lymphocytes from blood mononuclear cells was performed using
monoclonal antibodies to CD3, CD4, CD8, CD14, CD16, and
CD56 and sheep anti-mouse–coated magnetic beads (Dynal,
Oslo, Norway). Fluorescence-activated cell sorting (FACS)
analysis showed that ⬎90% of the cells were B cells. Contaminant cells were mostly T cells (2–5%) and monocytes (0.1–
1.5%). T cells do not express Lyn, while monocytes do (at
levels similar to B cells).
Cell culture. B cells were cultured in RPMI containing
10% fetal calf serum (FCS), penicillin, and streptomycin. For
proliferation, phosphorylated B lymphocytes were cultured at
2 ⫻ 105 for 3 days, and incorporation of 3H-thymidine (1
␮Ci/ml) was determined after 18 hours. Supernatants were
analyzed for cytokine and antibody production.
Protein ubiquitination in B cells from 14 patients with
SLE and 8 controls was studied. The B cells were cultured with
or without ALLN (N-acetyl-Leu-Leu-Nle-CHO) proteasome
inhibitor (50 ␮M in dimethyl sulfoxide) in complete medium.
After overnight incubation, the cells were lysed in buffer
containing protease and phosphatase inhibitors and 5 mM
N-ethylmaleimide, to preserve ubiquitinated proteins. To compare data from the different Western blot experiments, B cells
from healthy controls were similarly treated and analyzed on
the same membrane. To assess the effect of BCR engagement,
B cells were rested for 1 hour and stimulated with 20 ␮g/ml
F(ab⬘)2 anti-IgGAM, lysed, and analyzed as described below.
Lipid raft extraction. B cells were lysed with 500 ␮l of
lysis buffer (1% Triton X-100 in MNE buffer [25 mM morpholinoethane sulfonic acid]) (pH 6.50), 2 mM EDTA, 150 mM
NaCl]) containing protease and phosphatase inhibitors and
lysate, used to make a 5–40% discontinuous gradient by mixing
with 500 ␮l of 80% sucrose. The resulting lysate–40% sucrose
mixture was overlaid with 1 ml of 30% sucrose and 0.5 ml of
5% sucrose in MNE and centrifuged at 200,000g for 16 hours
at 4°C. Nine fractions were collected; fractions 2 and 3
contained lipid rafts, and fractions 8 and 9 contained nonlipid
rafts. Lipid raft and nonlipid raft fractions were pooled and
Western blotting. Proteins were precipitated from lysates, lipid rafts, and nonlipid rafts with methanol:chloroform
(2:1) and resuspended in Laemmli buffer. Samples were
adjusted for equal protein loading and separated on reducing
10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) gels. All blottings were carried out in a
manner such that samples from patients with SLE and samples
from healthy controls were obtained on the same day and
analyzed on the same gel. For immunoprecipitation, cells were
lysed in 250 ␮l of precipitation buffer (1% Triton X-100, 150
mM NaCl, 10 mM Tris HCl [pH 7.4], 1 mM EDTA, 1 mM
EGTA, 0.5% Igepal, with protease and phosphatase inhibitors) and centrifuged to remove insoluble debris. The supernatants were adjusted to 100 ␮g of protein, and precleared
with 50% protein G–Sepharose (Pharmacia, St. Albans, UK).
Lyn was precipitated with 5 ␮g of rabbit anti-Lyn and isolated
with protein G–Sepharose, resuspended in Laemmli buffer,
and separated by SDS-PAGE. Separated proteins were transferred onto nitrocellulose membranes and probed with specific
antibodies (1 ␮g/ml) in blocking buffer. Bound antibodies were
revealed with HRP secondary antibodies and enhanced chemiluminescence. Lyn quantification was carried out by densitometry using Image-Pro Plus software (Media Cybernetics, Silver
Springs, MD). To enable reliable comparison of different
experiments, protein loading was first normalized relative to
flotillin 2 protein for lipid rafts, and actin for total lysates and
nonlipid rafts, and levels were expressed as the relative intensity of the Lyn bands compared with that of the reference
Confocal microscopy. Fixed B cells were attached onto
TESPA-treated coverslips (Sigma). The cells were permeabilized with 0.5% Triton X-100 and blocked with 0.5% fish skin
gelatin and anti-Lyn or anti c-Cbl (5 ␮g/ml), followed by
secondary FITC antibodies. B cells were stained for lipid rafts,
using PE-conjugated cholera toxin subunit B (CTB). Stained
cells were mounted and analyzed using the MRC-1024 Confocal Laser Scanning Imaging System (Bio-Rad, Herts, UK).
Images were acquired with Laser Sharp 3.2 software (BioRad). PE or FITC fluorescence was recorded using a 40⫻
objective. For overlays, images were adjusted to the same
output intensities and merged, using the Confocal Assistant 4.2
program (Bio-Rad), into a composite RGB image. Lyn–CTB
and c-Cbl–CTB colocalization in lipid rafts was quantified
using the colocalization tool in LSM-S10 software (Carl Zeiss,
Thornwood, NJ), and results were expressed in pixels.
FACS analysis. For membrane staining, 106 mononuclear cells or 2 ⫻ 105 B cells were incubated with either
FITC-, PE-, or APC-conjugated antibodies for 30 minutes at
4°C. For intracellular Lyn staining, cells were fixed and permeabilized in permeabilization buffer (2% FCS and 0.3%
saponin in phosphate buffered saline), and Lyn was detected
with primary antibodies followed by FITC-conjugated secondary antibodies. Cells were analyzed with a FACSCalibur
cytometer (BD Biosciences).
IgG, IgM, anti-dsDNA antibody, and cytokine measurement. Supernatants were harvested after day 3 of culture
without activation and added to enzyme-linked immunosorbent assay plates precoated with 2 ␮g/ml goat F(ab⬘)2 antihuman ␥ or ␮ (Sigma). Anti-dsDNA antibodies were detected
using plates coated with methylated bovine serum albumin and
sonicated dsDNA. Supernatants were titrated, and antibodies
were detected with HRP anti-human ␥ or ␮. Total and
anti-dsDNA IgM and IgG were quantified using standard
curves of known inputs of antibodies. Interleukin-6 (IL-6) and
IL-10 were determined using plates coated with anti-human
IL-6 or IL-10 antibodies (Diaclone, Besancon, France). Bound
cytokines were revealed with biotin–anti-IL-6 or anti–IL-10
antibodies and streptavidin–HRP.
Statistical analysis. Differences in Lyn levels were
assessed using the Mann-Whitney test. Differences in ubiquitinated proteins in B cells from patients and controls were
analyzed by Fisher’s exact test. P values less than 0.05 were
considered significant.
Levels and lipid raft/nonlipid raft partitioning of
Lyn. The total cell content of Lyn in lipid raft and
nonlipid raft fractions was assessed by Western blotting,
confocal microscopy, and FACS (Figures 1–3). All experiments were carried out such that B cells from a
number of patients and controls were obtained on the
same day and analyzed simultaneously (Figure 1). Total
Lyn was reduced in B cells from 32 (52%) of 62 patients
with SLE. The mean ⫾ SEM Lyn level in lipid raft
fractions in SLE B cells (0.66 ⫾ 0.05 relative units [RU])
was significantly reduced compared with that in healthy
controls (0.99 ⫾ 0.05 RU) (P ⫽ 0.0003) (Figure 1d).
Similarly, the level of Lyn was reduced in nonlipid raft
fractions of B cells from patients with SLE (0.82 ⫾ 0.04
RU) compared with the level in nonlipid raft fractions of
Figure 1. Lyn level and distribution. a, Representative Western blot of Lyn expression in B cells from 5 patients
with systemic lupus erythematosus (SLE) and 2 normal controls (N). Equal amounts of protein from the different
samples were loaded on the gels, and these were further adjusted relative to the amount of loaded flotillin 2
(Flot-2) (lipid raft [LR]) or actin (nonlipid raft [NLR] and cell lysates). Lyn was semiquantified by the expressing
intensity of the Lyn band in each sample relative to the mean intensity of Lyn bands in B cells from the healthy
controls run on the same gel (relative units [RU]). RUs for the Lyn bands in the Western blots shown are
presented in the histogram beneath the blots. b, Confocal microscopy of Lyn in lipid rafts from an SLE patient
with reduced Lyn levels and a healthy control. In the patient with SLE, reduced Lyn is indicated by decreased
yellow areas resulting from the colocalization of fluorescein isothiocyanate (FITC)–conjugated Lyn and
phycoerythrin-conjugated cholera toxin subunit B (CTB). Colocalization was quantified by determining the
percentage of green pixels (Lyn) colocalizing with red pixels (middle image in top panel). c, Confocal microscopy
showing a reduction in total Lyn in a patient with SLE (upper panel) compared with the level in a healthy control
(lower panel), as indicated by the altered intensity of the FITC (green) staining. Bars in b and c ⫽ 5 ␮m. d, Scatter
plot showing Lyn levels in lipid rafts and nonlipid rafts from patients with SLE, patients with rheumatoid arthritis
(RA), and normal controls. Horizontal lines represent the mean. e, Percent of SLE patients with reduced levels
of Lyn in lipid rafts, nonlipid rafts, and both lipid and nonlipid rafts, and the total number of patients with
reduced levels of cellular Lyn. f, Correlation between Lyn in lipid rafts and nonlipid rafts in patients with SLE.
g, Distribution and mean level of Lyn in lipid rafts and nonlipid rafts in SLE patients with active (}) and inactive
({) disease. h, Graph showing consistency of Lyn expression in 3 SLE patients at 3 different time points. The
solid horizontal line represents the mean normal Lyn level in nonlipid rafts, and the broken line shows the mean
normal level in lipid rafts.
B cells from controls (1.02 ⫾ 0.03 RU) (P ⫽ 0.0082). To
assess whether Lyn levels were consistent or varied from
experiment to experiment, Lyn levels in 7 SLE patients
were determined on at least 3 occasions (at 6–8-month
intervals) and in 3 controls. The results showed that the
pattern of Lyn expression remained relatively consistent
in individual patients (Figure 1h) and controls. Selectivity of the changes in Lyn expression was verified by the
finding that levels of flotillin 2 and actin in B cells from
patients with SLE and patients with RA were similar to
those in B cells from the healthy controls. In addition,
the level of NTAL, a signaling protein constituent of
lipid rafts, was normal in most patients who had reduced
Lyn levels (ref. 24, and data not shown).
The reduction of Lyn in lipid rafts/total Lyn was
verified by confocal microscopy (Figures 1b and c). A
reduction of Lyn in lipid rafts is indicated by reduced
colocalization of Lyn with the raft marker GM1, as
Figure 2. Lyn expression in relation to coreceptor expression. Coreceptor expression was determined by
fluorescence-activated cell sorting (FACS) with allophycocyanin (APC)–conjugated anti-CD19 and phycoerythrin (PE)–conjugated anti-CD23, anti-CD27, anti-CD69, or anti-CD86. The Lyn level was determined by
staining with APC-conjugated anti-CD19 and PE-conjugated anti-CD23, anti-CD27, anti-CD69, or anti-CD86
and for intracellular Lyn with rabbit anti-Lyn and fluorescein isothiocyanate (FITC)–conjugated secondary
antibody. a, Representative results of FACS analysis of B cells from a healthy control, showing the percentage
of cells expressing CD19 and coreceptors. b, Percentage of CD19⫹ cells expressing coreceptors in 6 patients with
systemic lupus erythematosus (SLE) and 6 controls. c, Representative results of FACS analysis for Lyn
expression. The left contour shows the binding of anti-Lyn antibody in primary T lymphocytes (CD3⫹) and Lyn
in B lymphocytes (CD19⫹) from a normal control (N). The right contour depicts cellular Lyn in B cells from an
SLE patient and a healthy control. d, Mean fluorescence intensity (MFI) for total Lyn expression in B cells from
healthy controls and patients with SLE coexpressing CD23, CD27, CD69, or CD86. NS ⫽ nonstained cells; Isot ⫽
binding of the isotype control. Values in b and d are the mean and SEM. Color figure can be viewed in the online
issue, which is available at
detected by CTB–PE binding in the SLE patients compared with the healthy controls (Figure 1b, yellow). The
mean ⫾ SEM percentage of green pixels (Lyn) colocalizing with red pixels (CTB) in 100 B cells from each of
the SLE patients examined was 32.9 ⫾ 4.9, compared
with 63.7 ⫾ 5.2 for the controls (Figure 1b).
To assess the effect of disease activity on Lyn
expression, the Lyn levels in patients with active disease
were compared with the levels in patients with inactive
disease. The mean ⫾ SEM level of Lyn in lipid rafts
from patients with active disease (0.63 ⫾ 0.08 RU) was
only slightly reduced compared with that in patients with
inactive disease (0.68 ⫾ 0.06 RU; P ⫽ 0.1928). The
levels of Lyn in nonlipid rafts from patients with active
disease and patients with inactive disease were almost
identical (0.81 ⫾ 0.07 RU and 0.82 ⫾ 0.05 RU, respectively; P ⫽ 0.9428) (Figure 1g). Medication did not
influence Lyn expression: the levels in lipid rafts and
nonlipid rafts in the 2 treatment groups were not significantly different (P ⫽ 0.84 and P ⫽ 0.11, respectively)
(data not shown).
B lymphocyte activation and Lyn expression. The
activity of Src kinases is regulated by autophosphorylation of regulatory tyrosine residues and by ubiquitination (25,26). Autophosphorylation modulates the conformation of Src kinases and is reversible, but
ubiquitination is not reversible and leads to degradation
by the proteasome. Activation of lymphocytes results in
the phosphorylation of Src kinases, ubiquitination, and
degradation (27). Because there is evidence for B lym-
Figure 3. Lyn expression in B lymphocyte subpopulations. B lymphocytes were
stained with phycoerythrin-conjugated IgG1 antibody to CD27 and unconjugated
IgG2b anti-CD20, followed by biotin–anti-mouse IgG2b antibodies and peridin
chlorophyll protein–streptavidin. After permeabilization, the cells were stained for
fluorescein isothiocyanate–conjugated rabbit anti-Lyn. a, Representative
fluorescence-activated cell sorting contours of B lymphocyte subpopulations in 1
healthy control and 1 patient with systemic lupus erythematosus (SLE). b, Frequency
of B lymphocyte subpopulations in 12 patients with SLE and 5 normal subjects. c,
Mean fluorescence intensity (MFI) of Lyn expression in B cell subpopulations. ⴱ ⫽
P ⬍ 0.05, SLE patients versus controls. Values in b and c are the mean and SEM.
Color figure can be viewed in the online issue, which is available at http://
phocyte activation in SLE, we sought to assess whether
altered Lyn expression is attributable to activation. First,
we determined the relationships of expression of various
coreceptors that are up-regulated following activation in
vivo with total Lyn. Second, we assessed the effect of in
vitro activation on the Lyn level and the pattern of
phosphorylation of the regulatory tyrosines. Third, we
assessed the level of protein and Lyn ubiquitination in
patients with SLE and healthy controls.
To assess whether activation in vivo resulted in
reduced Lyn, percentages of B lymphocytes positive for
(and the intensity of) CD23, CD27, CD69, and CD86
expression in 6 SLE patients with reduced Lyn and 6
healthy controls were determined (Figure 2). No significant differences in the number of B cells expressing
CD23, CD27, CD69, or CD86 were noted (Figure 2b).
However, when the mean fluorescence intensity was
assessed, there were marginal increases (P ⬎ 0.05) in
CD27, CD69, and CD86 expression in the SLE patients
(data not shown). To assess the direct relationship
between coreceptor up-regulation and Lyn, we carried
out a 3-color FACS analysis for the 4 coreceptors with
CD19 and intracellular Lyn. The specificity of FACS
staining was confirmed by the lack of staining with the
isotype control and the lack of staining of T cells with
anti-Lyn (Figure 2c, left). When total Lyn was determined in healthy controls and patients with SLE, there
was a clear reduction in Lyn in the SLE patients
compared with the controls (Figure 2c, right). When the
level of Lyn in B cell subpopulations was determined,
the results showed a reduction in total Lyn in all
subpopulations of SLE patients (Figure 2d). Furthermore, expression of the coreceptors was related to an
increase rather than a decrease in Lyn levels.
Because of evidence for the accumulation of
lymphoblasts/plasma cells (encompassed within
CD19⫹,CD20⫺,CD27⫹ cells) in the blood of SLE
patients (28), we next investigated whether such accumulation explains altered Lyn expression. We determined the frequency of Lyn and total Lyn in
CD20⫹,CD27⫺, CD20⫹,CD27⫹, CD20⫺,CD27⫺, and
CD20⫺,CD27⫹ B lymphocytes from 12 SLE patients
with reduced levels of Lyn and from 5 controls (Figure
3a). The percentage of CD20⫺,CD27⫹ cells was signif-
Figure 4. Level and pattern of Lyn phosphorylation. a, Representative Western blot
showing expression of Lyn, phosphorylated Y-396 and Y-507, and flotillin 2 (Flot-2)
in the lipid raft (LR) and nonlipid raft (NLR) fractions of B lymphocytes from 3
patients with systemic lupus erythematosus (SLE) and 2 normal controls (N). The
histogram shows the mean and SEM intensity of bands representing the 2 phosphorylated forms of Lyn relative to that of total Lyn, after normalization for equal
loading. b, Western blot showing the level of expression of Lyn and its phosphorylated forms in B cell receptor–activated B cells in vitro. Rested B cells (106) were
stimulated with F(ab⬘)2 anti-IgG/IgA/IgM for 2 or 5 minutes. The cells were lysed,
and aliquots corresponding to 0.25 ⫻ 106 cells were analyzed. Unstimulated B cells
from the same patients were lysed and used to represent time 0 stimulation. B cells
from 3 SLE patients (with detectable Lyn levels) and 3 healthy controls were
analyzed. The histogram shows the level of phosphorylated Lyn forms relative to that
of total Lyn, after normalization for equal loading. P-Tyr ⫽ phosphorylated tyrosine.
icantly increased in SLE patients compared with controls (Figure 3b). This increase, however, correlated with
disease activity but not with reduced Lyn (Figure 3c).
To further investigate the relationship between
Lyn expression and B lymphocyte activation, we studied
changes in Lyn levels and the phosphorylation pattern
after BCR engagement. Lyn has 2 regulatory carboxylterminal tyrosines, at positions 396 (Y396) and 507
(Y507), that determine Lyn activation status. Y507 is
phosphorylated in inactive Lyn, while Y396 is phosphorylated in active Lyn (29). When the phosphorylation
status of Lyn in ex vivo B lymphocytes from 10 SLE
patients and 4 controls was analyzed, the results showed
that in both groups, Lyn was phosphorylated mostly at
Y507 (Figure 4a). There was a slight increase (P ⬎ 0.05)
in the level of P-Y396 in the nonlipid raft fractions in
SLE (mean ⫾ SEM 0.24 ⫾ 0.1 RU) compared with
controls (0.16 ⫾ 0.2 RU). In addition, there was a
notable decrease in the overall level of P-Y507 in ex vivo
SLE B cells. We then studied the effect of BCR engagement on the level and phosphorylation status of the 2
tyrosines in Lyn in vitro. B lymphocytes from 3 controls
and 3 SLE patients were rested for 1 hour and activated
with F(ab⬘)2 anti-IgGAM for 2 or 5 minutes. BCR
ligation led to an increase in the phosphorylation of both
Y396 and Y507 but did not lead to a reduction in the
level of Lyn (Figure 4b).
Ubiquitination and Lyn expression in SLE.
There is evidence that c-Cbl targets Lyn for ubiquitination and degradation in activated murine B cells (30).
Because our experiments showed a reduction in Lyn
P-Y507 in ex vivo SLE B cells, we sought to determine
the degree of total protein and Lyn ubiquitination in
these cells. Lyn ubiquitination and the level of total
protein were determined in B cells from 11 patients with
SLE who had reduced Lyn levels and from 3 SLE
patients and 8 healthy controls who had normal levels of
Lyn. Ubiquitination in the presence of ALLN, a potent
inhibitor of the proteasome, was compared with ubiquitination in cells cultured in the absence of ALLN. In
all 11 SLE patients with reduced Lyn levels, there was
visible accumulation of ubiquitinated proteins in the
presence of ALLN. In the presence of ALLN, there was
a marginal increase in the level of ubiquitinated proteins
in the 3 SLE patients with normal Lyn, compared with
the level in the absence of ALLN. Two of the 8 control
subjects also showed slight accumulation of ubiquitinated proteins (Figure 5a). The difference between the
Figure 5. Ubiquitination and Lyn expression. B cells (106) from 14 patients with systemic lupus erythematosus
(SLE) and 8 normal controls (N) were cultured with or without N-acetyl-Leu-Leu-Nle-CHO (ALLN) for 16
hours. The cells were collected, washed, lysed, and analyzed by Western blotting. a, Equally loaded proteins were
analyzed for ubiquitination (Ubiq; upper blots) and for total Lyn (middle blots) in the absence or presence of
ALLN in B cells from 3 SLE patients (lanes 1–6) and 1 healthy control (lanes 7 and 8). The number beneath each
lane represents a measure of ubiquitinated protein in relative units (RU), as determined (after adjusting for equal
protein loading based on actin band) by measuring the intensity of the smear divided by its area and expressed
relative to that of similarly treated B cells from the normal controls included on the same blot (lanes 1, 3, and
5 relative to lane 7, and lanes 2, 4, and 6 relative to lane 8). Total Lyn levels were measured as described in Figure
1. b, Representative blot showing ubiquitinated Lyn immunoprecipitated (IP) with anti-Lyn antibody from B cells
cultured in the absence or presence of ALLN. The membranes were probed with mouse antibody to human
ubiquitin (upper panel) or antibody to Lyn (lower panel). The intensity of the ubiquitin smear and Lyn bands in
each sample was measured by densitometry and expressed as RU, as described above. c, Confocal microscopic
profiles of B cells showing c-Cbl (fluorescein isothiocyanate) expression in lipid rafts (LRs) (cholera toxin subunit
B [CTB]–phycoerythrin). Translocation of c-Cbl into lipid rafts is shown as yellow patches. Bars ⫽ 5 ␮m. d,
Scatter graph showing the percentage of B cells with detectable c-Cbl in lipid rafts. Horizontal lines show the
mean. e, B cell receptor ligation with F(ab⬘)2 anti-IgG/IgA/IgM–coated magnetic beads causing polarization of
c-Cbl (green) to the region where activation of Lyn is presumed to take place. In the top photomicrographs
(confocal analysis), the position of the magnetic bead is indicated by an X. The bottom photomicrographs
(phase-contrast microscopy) represent the same cells/beads. rel. ⫽ relative; Tx ⫽ transmission (phase-contrast)
11 SLE patients with reduced Lyn and the 2 other
groups was significant (P ⬍ 0.05 by Fisher’s exact test).
To directly investigate the level of Lyn ubiquitination, total protein extracts from B lymphocytes were
immunoprecipitated with anti-Lyn antibody. B cells
from the 11 SLE patients with reduced Lyn showed a
visible accumulation of high and low molecular weight
ubiquitinated Lyn species in the presence of ALLN
(Figure 5b). The level of ubiquitinated Lyn did not
change or changed only marginally in the 3 SLE patients
with normal Lyn and in the 8 healthy controls.
To verify the notion of increased ubiquitination,
the expression of c-Cbl in lipid rafts from 7 SLE patients
and 5 healthy controls was studied by confocal microscopy. The results showed an increase in the number of B
cells with high c-Cbl ligase levels in lipid rafts from SLE
patients compared with the controls (Figures 5c and d).
The ubiquitin ligase c-Cbl also copolarized with the
Figure 6. Altered Lyn expression and B cell activity. a, B cells (2 ⫻ 105/ml) were
obtained from 12 patients with systemic lupus erythematosus (SLE) in whom Lyn
expression was altered and from 6 patients with rheumatoid arthritis (RA) and 9
healthy controls with normal Lyn expression. B cells were cultured in 96-well plates
with no stimulation for 3 days, 1 ␮g of 3H-thymidine was added, and uptake was
determined after 16 hours. b–d, Supernatants from nonpulsed cells were collected
for measurement of IgG production, interleukin-10 (IL-10) production, and IL-6
production. ⴱ ⫽ P ⬍ 0.05, SLE patients versus healthy controls (and RA patients in
a). Values are the mean and SEM.
BCR upon F(ab⬘)2 anti-IgGAM activation (Figure 5e).
Colocalization of c-Cbl to lipid rafts was confirmed by
Western blotting (results not shown).
Lyn expression and B lymphocyte responses. To
determine whether altered Lyn expression coincides
with B lymphocyte responsiveness, we studied spontaneous proliferation and IgG and cytokine production by B
cells from 12 SLE patients with reduced Lyn, 6 patients
with RA, and 9 healthy controls. Figure 6a shows that
spontaneous proliferation was significantly higher in B
cells from SLE patients compared with RA patients and
healthy controls (P ⬍ 0.05). In addition, only B cells
from the SLE patients produced IgG anti-dsDNA antibodies (results not shown) and higher levels of total IgG
and IL-10 compared with controls (Figures 6b and c).
There was no correlation between proliferation and
increased IgG in the SLE patients (r ⫽ ⫺0.3124, P ⫽
0.498), implying that the increase in the SLE B cell
numbers in culture was not the cause of increased IgG
This study revealed that B cells from a majority of
patients with SLE had reduced levels of Lyn and diminished translocation to lipid raft domains. These alterations appeared to be specific to Lyn, because there
were no parallel alterations in the expression of other
proteins that translocate lipid raft domains, such as
flotillin 2 and the signaling molecule NTAL (24). The
alterations in Lyn expression in SLE, however, did not
correlate with disease activity. This lack of correlation
with disease activity and the relatively consistent nature
of Lyn expression suggest that the noted alterations in
Lyn expression could be a phenotypic expression of
genetic variability rather than causally related to the
disease. Alternatively, it is possible that altered Lyn
expression could be relevant to specific clinical and/or
immunologic features of SLE. Thus, alterations in Lyn
expression could be attributable to long-term activation
of B cells in patients with SLE. Formal confirmation of
the cause(s) of altered Lyn expression in SLE requires
further investigation, especially considering that reduced
levels of cellular Lyn were also observed in some patients with RA.
To identify possible causes of altered Lyn expression in SLE, we studied the effect of B lymphocyte
activation. These experiments were prompted by established evidence for perturbed B lymphocyte homeostasis
in SLE and by the manner in which Src kinases are
regulated (25,26,28). In T lymphocytes, activation induces Lck phosphorylation, ubiquitination, and degradation (31). Thus, the level of Src kinases at a given
point in the life of a lymphocyte is controlled by synthesis, autophosphorylation, and degradation by ubiquitination. To assess whether reduced cellular Lyn in SLE B
lymphocytes was attributable to activation in vivo, ex-
pression of coreceptors that were up-regulated upon
activation was studied. The results showed that modest
up-regulation of CD27, CD69, and CD86 on SLE B
lymphocytes, or an increased frequency of lymphoblast/
plasma cells, did not relate to the alterations in Lyn
expression. In vitro activation of B cells by engaging the
BCR with F(ab⬘)2 anti-human IgGAM antibody showed
that short-term activation does not lead to a reduced
level of cellular Lyn. In addition, incubation of B cells
with interferon-␥, which is increased in patients with
SLE, did not affect Lyn levels (data not shown). Interestingly, however, the in vitro activation experiments
showed a decrease in the level of the inactive form of
Lyn (i.e., P-Y507). This finding suggests that long-term
BCR engagement in SLE patients might cause a shift in
the phosphorylation pattern of Lyn and promote degradation.
The ubiquitination experiments revealed increased accumulation of Lyn protein in B cells from SLE
patients, in the presence of a proteasome inhibitor
(Figure 5). The ubiquitin/proteasome system is the
major proteolytic system involved in selective degradation of short-lived regulatory proteins in eukaryotic cells
(32). Increased degradation of Lyn by the proteasome is
consistent with results of recent studies showing increased ubiquitination of signaling molecules in patients
with SLE (33,34). To gain further insight into Lyn
ubiquitination, we assessed c-Cbl translocation to lipid
rafts (Figure 5c). The E3 ubiquitin ligase c-Cbl targets
protein tyrosine kinases for degradation (35). In addition, recent studies have shown that c-Cbl negatively
regulates Lyn in mast cells (36) and B cells (30). Our
results showed that compared with healthy controls, the
patients with SLE had higher numbers of B cells with
increased Cbl in the lipid raft domains. Furthermore, the
data showed that BCR ligation led to c-Cbl colocalization in lipid rafts with the BCR at the contact sites
(Figure 5e). The translocation of c-Cbl to lipid rafts in
SLE B cells may represent additional indirect evidence
for increased ubiquitination, although it does not explain
the phenotype noted in these B cells. Further investigation
is necessary to verify this possibility.
Increased degradation of Lyn by the proteasome
in SLE B cells is consistent with a model of Lyn and Syk
regulation in Epstein-Barr virus–transformed cells (37).
However, it is also possible that other factors contribute
to altered Lyn expression. For example, Liossis et al
suggested that reduced levels of Lyn messenger RNA
(mRNA) in mononuclear cells from SLE patients could
be attributable to reduced gene transcription (16). There
is also evidence for reduced mRNA translation in T
lymphocytes from some patients with SLE (38). The
investigators in the latter study demonstrated that a
protein kinase, RNA-dependent protein kinase, which is
responsible for the phosphorylation and consequent
inhibition of 2 eukaryotic initiation factors (eIF4E and
eIF2␣), was overexpressed in SLE T cells. There is also
evidence for reduced synthesis of signaling molecules in
SLE T lymphocytes. For example, Khan and colleagues
revealed that ⬃80% of patients with SLE had a deficiency in type I protein kinase A phosphotransferase
activity (39). Nevertheless, our data provide evidence for
reduced Lyn in lipid rafts and changes in the pattern of
Lyn phosphorylation. These data, while not totally excluding other mechanisms for Lyn reduction, suggest
that an alteration in Lyn expression in SLE B cells could
be, at least partly, due to increased ubiquitination and
The relevance of altered Lyn expression to SLE
immunopathology is not clear. However, altered Lyn
expression could weaken negative regulation of proximal
signaling, leading to the expansion of autoreactive B
cells. In lyn⫺/⫺ mice there is evidence for uncontrolled
expansion of B lymphocytes that produce nephritiscausing anti-dsDNA autoantibodies. Furthermore, in
mice with a targeted gain-of-function mutation
(Lynup/up), in which there is sustained Lyn activation,
Hibbs and colleagues noted exaggerated positive signaling, as evidenced by constitutive phosphorylation of Syk
and phospholipase C2, in resting B cells (15). Furthermore, B cells had enhanced Ca⫹2 flux upon BCR
ligation and developed autoantibodies and nephritis. It
could be possible, therefore, that increased degradation
of Lyn in SLE patients reduces negative signaling,
facilitates responses to autoantigens, and enhances autoantibody production. An alternative possibility (although the 2 are not mutually exclusive) is that reduced
Lyn could disrupt the constitutive phosphorylation of
proteins involved in maintaining B lymphocyte homeostasis. Experiments in which B cell responses were
studied showed that the SLE B lymphocytes with reduced Lyn proliferated at higher rates than did B
lymphocytes from healthy controls and produced IgG
anti-dsDNA antibodies. This finding is consistent with B
lymphocyte responses in lyn⫺/⫺ mice and could, perhaps,
be associated with the reduced phosphorylation of
CD22, src homology phosphatase 2, and Fc␥RII
1. Tsokos GC, Liossis SN. Immune cell signaling defects in lupus:
activation, anergy and death. Immunol Today 1999;20:119–24.
2. Tanaka Y, Shirakawa F, Ota T, Suzuki H, Eto S, Yamashita U.
Mechanism of spontaneous activation of B cells in patients with
systemic lupus erythematosus: analysis with anti-class II antibody.
J Immunol 1988;140:761–7.
3. Mageed RA, Prud’homme G. Immunopathology and the gene
therapy of lupus. Gene Ther 2003;10:861–73.
4. Reininger L, Winkler T, Kalberer CP, Jourdan M, Melchers F,
Rolink AG. Intrinsic B cell defects in NZB and NZW mice
contribute to systemic lupus erythematosus in (NZB ⫻ NZW)F1
mice. J Exp Med 1996;184:853–61.
5. O’Keefe TL, Williams GT, Batista F, Neuberger MS. Deficiency in
CD22, a B cell-specific inhibitory receptor, is sufficient to predispose to development of high affinity autoantibodies. J Exp Med
6. Pritchard NR, Cutler A, Uribe S, Chadban S, Morley B, Smith KG.
Autoimmune-prone mice share a promoter haplotype associated
with reduced expression and function of the Fc receptor Fc␥RII.
Curr Biol 2000;10:227–30.
7. Pan C, Bamgarth N, Parnes JR. CD72-deficient mice reveal
non-redundant roles of CD72 in B cell development and activation. Immunity 1999;11:495–506.
8. Salmon JE, Millard S, Schachter L, Arnett F, Ginzler EM, Gourley
M, et al. Fc␥RIIA alleles are heritable risk factors for lupus
nephritis in African Americans. J Clin Invest 1996;97:1348–54.
9. Hatta Y, Tsuchiya N, Matsushita M, Shiota M, Hagiwara K,
Tokunaga K. Identification of the gene variations in human CD22.
Immunogenetics 1999;49:280–6.
10. Kurosaki T. Molecular dissection of B cell antigen receptor
signaling. Int J Mol Med 1998;1:515–27.
11. Pao LI, Famiglietti S, Cambier JC. Asymmetrical phosphorylation
and function of immunoreceptor tyrosine-based activation motif
tyrosines in B cell antigen receptor signal transduction. J Immunol
12. Cornall R, Cyster J, Hibbs M, Dunn A, Otipoby K, Clark A, et al.
Polygenic autoimmune traits: Lyn, CD22, and SHP-1 are limiting
elements of a biochemical pathway regulating BCR signaling and
selection. Immunity 1998;8:497–508.
13. De Franco AL. The complexity of signaling pathways activated by
the BCR. Curr Opin Immunol 1997;9:296–308.
14. Ma H, Yankee T, Hu J, Asai D, Harrison M, Geahlen RL.
Visualization of Syk-antigen receptor interactions using green
fluorescent protein: differential roles for Syk and Lyn in the
regulation of receptor capping and internalization. J Immunol
15. Hibbs ML, Harder KW, Armes J, Kountouri N, Quilici C,
Casagranda F, et al. Sustained activation of Lyn tyrosine kinase in
vivo leads to autoimmunity. J Exp Med 2002;196:1593–604.
16. Liossis SN, Solomou E, Dimopoulos M, Panayiotidis P, Mavrikakis
M, Sfikakis PP. B-cell kinase lyn deficiency in patients with
systemic lupus erythematosus. J Investig Med 2001;49:157–65.
17. Huck S, le Corre R, Youinou P, Zouali M. Expression of B cell
receptor-associated signaling molecules in human lupus. Autoimmunity 2001;33:213–24.
18. Cheng PC, Dykstra M, Mitchell R, Pierce SK. A role for lipid rafts in
BCR signaling and antigen targeting. J Exp Med 1999;190:1549–60.
19. Simons K, Ikonen E. Functional rafts in cell membranes. Nature
20. Hay EM, Bacon P, Gordon C, Isenberg DA, Madison P, Snaith M,
et al. The BILAG index: a reliable and valid instrument for
measuring clinical disease activity in systemic lupus erythematosus. Q J Med 1993;86:447–58.
21. Bombardier C, Gladman DD, Urowitz MB, Caron D, Chang CH,
and the Committee on Prognosis Studies in SLE. Derivation of the
SLEDAI: a disease activity index for lupus patients. Arthritis
Rheum 1992;35:630–40.
22. Liang MH, Socher SA, Larson MG, Schur PH. Reliability and
validity of six systems for the clinical assessment of disease activity in
systemic lupus erythematosus. Arthritis Rheum 1989;32:1107–18.
Gladman DD, Goldsmith CH, Urowitz MB, Bacon P, Bombardier
C, Isenberg D, et al. Sensitivity to change of 3 systemic lupus
erythematosus disease activity indices: international validation.
J Rheumatol 1994;21:1468–71.
Brdicka T, Imrich M, Angelisova P, Brdickova N, Horvath O,
Spicka J, et al. Non-T cell activation linker (NTAL): a transmembrane adaptor protein involved in immunoreceptor signaling. J
Exp Med 2002;196:1617–26.
Yamaguchi H, Hendrickson WA. Structural basis for activation of
human lymphocyte kinase Lck upon tyrosine phosphorylation.
Nature 1996;384:484–9.
Hakak Y, Martin GS. Ubiquitin-dependent degradation of active
Src. Curr Biol 1999;9:1039–42.
Oda H, Kumar S, Howley PM. Regulation of the Src family
tyrosine kinase Blk through E6AP-mediated ubiquitination. Proc
Natl Acad Sci U S A 1999;96:9557–62.
Jacobi AM, Odendahl M, Reiter K, Bruns A, Burmester G,
Radbruch A, et al. Correlation between circulating CD27high
plasma cells and disease activity in patients with systemic lupus
erythematosus. Arthritis Rheum 2003;48:1332–42.
DeFranco AL, Chan VW, Lowell CA. Positive and negative roles
of the tyrosine kinase Lyn in B cell function. Semin Immunol
Shao Y, Yang C, Elly C, Liu YC. Differential regulation of the B
cell receptor-mediated signaling by the E3 ubiquitin ligase Cbl.
J Biol Chem 2004;279:43646–53.
Rao N, Miyake S, Reddi AL, Douillard P, Ghosh AK, Dodge IL,
et al. Negative regulation of Lck by Cbl ubiquitin ligase. Proc Natl
Acad Sci U S A 2002;99:3794–9.
Hershko A, Ciechanover A. The ubiquitin system. Annu Rev
Biochem 1998;67:425–79.
Jury EC, Kabouridis PS, Abba A, Mageed RA, Isenberg DA.
Increased ubiquitination and reduced expression of LCK in T
lymphocytes from patients with systemic lupus erythematosus.
Arthritis Rheum 2003;48:1343–54.
Nambiar MP, Enyedy E, Fisher CU, Krishnan S, Warke VG,
Gilliland WR, et al. Abnormal expression of various molecular
forms and distribution of T cell receptor ␨ chain in patients with
systemic lupus erythematosus. Arthritis Rheum 2002;46:163–74.
Thien CB, Langdon WY. Cbl: many adaptations to regulate
protein tyrosine kinases. Nat Rev Mol Cell Biol 2001;2:294–307.
Kyo S, Sada K, Qu X, Maeno K, Miah SM, Kawauchi-Kamata K,
et al. Negative regulation of Lyn protein-tyrosine kinase by c-Cbl
ubiquitin-protein ligase in Fc␥RI-mediated mast cell activation.
Genes Cells 2003;8:825–36.
Winberg G, Matskova L, Chen F, Plant P, Rotin D, Gish G, et al.
Latent membrane protein 2A of Epstein-Barr virus binds WW
domain E3 protein-ubiquitin ligases that ubiquitinate B-cell tyrosine kinases. Mol Cell Biol 2000;20:8526–35.
Grolleau A, Kaplan MJ, Hanash SM, Beretta L, Richardson B.
Impaired translational response and increased protein kinase PKR
expression in T cells from lupus patients. J Clin Invest 2000;106:
Khan IU, Laxminarayana D, Kammer GM. Protein kinase A RI ␤
subunit deficiency in lupus T lymphocytes: bypassing a block in RI
␤ translation reconstitutes protein kinase A activity and augments
IL-2 production. J Immunol 2001;166:7600–5.
Smith KG, Tarlinton D, Doody G, Hibbs M, Fearon D. Inhibition
of the B cell by CD22: a requirement for Lyn. J Exp Med
Aman MJ, Tosello-Trampont A, Ravichandran K. Fc␥RIIB1/
SHIP-mediated inhibitory signaling in B cells involves lipid rafts.
J Biol Chem 2001;279:46371–8.
Без категории
Размер файла
358 Кб
expressions, domain, patients, systemic, erythematosus, raft, decrease, lupus, lyn, signaling, translocation, lymphocytes, lipid
Пожаловаться на содержимое документа