close

Вход

Забыли?

вход по аккаунту

?

art.33311

код для вставкиСкачать
ARTHRITIS & RHEUMATISM
Vol. 64, No. 1, January 2012, pp 141–152
DOI 10.1002/art.33311
© 2012, American College of Rheumatology
Cyclin-Dependent Kinase Inhibitor p21,
via Its C-Terminal Domain, Is Essential for
Resolution of Murine Inflammatory Arthritis
Melissa Mavers,1 Carla M. Cuda,2 Alexander V. Misharin,2 Angelica K. Gierut,2
Hemant Agrawal,2 Evan Weber,2 Deborah Veis Novack,3 G. Kenneth Haines III,4
Dimitrios Balomenos,5 and Harris Perlman2
Objective. The mechanism responsible for persistent synovial inflammation in rheumatoid arthritis
(RA) is unknown. Previously, we demonstrated that
expression of the cyclin-dependent kinase inhibitor p21
is reduced in synovial tissue from RA patients compared
to osteoarthritis patients and that p21 is a novel suppressor of the inflammatory response in macrophages.
The present study was undertaken to investigate the role
and mechanism of p21-mediated suppression of experimental inflammatory arthritis.
Methods. Experimental arthritis was induced in
wild-type or p21ⴚ/ⴚ (C57BL/6) mice, using the K/BxN
serum–transfer model. Mice were administered p21
peptide mimetics as a prophylactic for arthritis development. Lipopolysaccharide-induced cytokine and sig-
nal transduction pathways in macrophages that were
treated with p21 peptide mimetics were examined by
Luminex-based assay, flow cytometry, or enzyme-linked
immunosorbent assay.
Results. Enhanced and sustained development of
experimental inflammatory arthritis, associated with
markedly increased numbers of macrophages and severe articular destruction, was observed in p21ⴚ/ⴚ mice.
Administration of a p21 peptide mimetic suppressed
activation of macrophages and reduced the severity of
experimental arthritis in p21-intact mice only. Mechanistically, treatment with the p21 peptide mimetic led to
activation of the serine/threonine kinase Akt and subsequent reduction of the activated isoform of p38 MAPK
in macrophages.
Conclusion. These are the first reported data to
reveal that p21 has a key role in limiting the activation
response of macrophages in an inflammatory disease
such as RA. Thus, targeting p21 in macrophages may be
crucial for suppressing the development and persistence
of RA.
Supported by the NIH (grants AR-060169 to Dr. Cuda,
AR-007611 to Dr. Gierut, AR-052705 to Dr. Novack, and AR-050250,
AR-054796, and AI-067590 to Dr. Perlman) and by Northwestern
University Feinberg School of Medicine (funding to Dr. Perlman). The
NIH provided funding to the Washington University School of Medicine Center for Musculoskeletal Biology and Medicine (grant P30AR-057235) and to the Northwestern University Feinberg School of
Medicine Cell Imaging Core Facility and the Methodology and Data
Management Core (grant P60-AR-048098).
1
Melissa Mavers, MD, PhD: Saint Louis University School of
Medicine, St. Louis, Missouri; 2Carla M. Cuda, PhD, Alexander V.
Misharin, MD, PhD, Angelica K. Gierut, MD, Hemant Agrawal, PhD,
Evan Weber, BS, Harris Perlman, PhD: Northwestern University
Feinberg School of Medicine, Chicago, Illinois; 3Deborah Veis Novack, MD, PhD: Washington University School of Medicine, St. Louis,
Missouri; 4G. Kenneth Haines III, MD: Yale University School of
Medicine, New Haven, Connecticut; 5Dimitrios Balomenos, PhD:
Centro Nacional de Biotecnologı́a, CSIC, Madrid, Spain.
Dr. Perlman has received speaking fees from Novartis (less
than $10,000).
Address correspondence to Harris Perlman, PhD, Northwestern University, Feinberg School of Medicine, Department of
Medicine/Rheumatology, 240 East Huron Street, Room M338, Chicago, IL 60611. E-mail: [email protected]
Submitted for publication December 6, 2010; accepted in
revised form August 25, 2011.
Macrophages play a central role in the pathogenesis of rheumatoid arthritis (RA). Conventional therapies, including methotrexate and cytokine inhibitors,
block proinflammatory cytokines produced primarily by
macrophages (1). Importantly, synovial macrophage infiltration correlates with subsequent radiographic joint
destruction (2). In addition, reduction of RA synovial
sublining macrophage numbers by various therapeutic
strategies correlates with clinical improvement, making
it a sensitive biomarker for disease activity (3,4). Intensive research is ongoing to elucidate the mechanisms
responsible for the increased synovial macrophage numbers in RA. Thus far, increased chemotaxis (5), reduced
141
142
emigration (6), and decreased apoptosis of macrophages
(7) have all been implicated in RA pathogenesis. In
addition to their increased numbers, the response of RA
synovial macrophages to Toll-like receptor (TLR) stimulation is amplified compared to macrophages from
patients with other inflammatory joint diseases or normal circulating monocytes differentiated in vitro (8). As
is the case in human patients, monocytes and macrophages are necessary for pathology to occur in various
experimental models of RA, such as collagen-induced
arthritis and K/BxN serum–transfer arthritis (9–11).
Despite the abundant data supporting the crucial role of
macrophages in RA, little is known about the factors
that control their state of activation.
Cyclin-dependent kinase inhibitors are of central
importance in suppressing cell cycle activity. To this end,
the vast majority of studies on cell cycle machinery in
RA have focused on the suppression of synovial fibroblast proliferation or production of inflammatory cytokines by cell cycle inhibitors such as retinoblastoma
protein, or cyclin-dependent kinase inhibitors p16,
p18, and p21 (12–17). Additionally, administration of a
replication-defective adenovirus expressing p16 or p21
via multiple intraarticular injections leads to suppression
of inflammatory arthritis in rodents (12–14,16). While
those studies focused on the role played by p16 or p21
in synovial fibroblasts, two recent studies have shown
that the presence of p21 reduces serum cytokine levels
and confers a protective effect on survival during lipopolysaccharide (LPS)–mediated endotoxic shock in
mice (18,19), a macrophage-dependent model. Expression of the activation markers CD40 and class II major
histocompatibility complex and secretion of the proinflammatory cytokines interleukin-6 (IL-6), tumor necrosis factor ␣, and IL-1␤ are enhanced in p21-deficient
macrophages following TLR ligation, even though these
cells have terminally withdrawn from the cell cycle
(18,19). These data suggest that p21 may function as a
novel suppressor of inflammation in macrophages.
In this animal study we demonstrated that p21
is important not only for limiting the development of
inflammatory arthritis, but also for induction of the
resolution or wound healing phase that occurs after
the arthritic stimulus is withdrawn. Similar to findings
in RA patients, the severity of inflammation and destruction of bone correlates with the number of macrophages in the pannus. Further, a p21 peptide mimetic
corresponding to the C-terminal domain of p21 is sufficient to reduce the severity of inflammatory arthritis and
lower the number of macrophages in the pannus. Isolated macrophages treated with the p21 peptide mimetic
MAVERS ET AL
displayed enhanced expression of phosphorylated Akt
and reduced p38 activation. Taken together, these data
are among the first to identify novel and essential roles
of p21 in suppression of inflammation and to translate
them into clinically significant information that may
shed light on the pathogenesis of RA.
MATERIALS AND METHODS
Mice. Male KRN mice were kindly provided by Dr.
Diane Mathis (Harvard Medical School, Boston, MA) and
were crossed with female NOD mice purchased from Taconic;
p21⫺/⫺ mice were backcrossed onto the C57BL/6 background
for at least 12 generations and tested for microsatellite markers
of background contribution (20). All experiments on mice were
approved by the Animal Care and Use Committee at Saint
Louis University and/or Northwestern University.
K/BxN serum–transfer arthritis. Serum was harvested
via cardiac puncture from 8-week-old male and female progeny of KRN and NOD mice (K/BxN) and injected intraperitoneally (IP) into 6–8–week-old mice. Ankle circumference was calculated from measurements obtained with a
caliper. Clinical scores were measured as follows: 0 ⫽ no
swelling, 1 ⫽ mild swelling in 1 or 2 limbs, 2 ⫽ mild swelling in
⬎2 limbs, 3 ⫽ moderate swelling in ⬎2 limbs, 4 ⫽ severe
swelling in ⬎2 limbs, and 5 ⫽ compromised mobility. At 7, 14,
or 25 days postinjection, mice were killed, serum was collected
via cardiac puncture, and ankles were harvested and fixed
in 10% formalin. Ankles were then subjected to microfocal
computed tomography (micro-CT) analysis (performed at the
core facility at Washington University School of Medicine)
and/or prepared for immunohistochemistry analysis. For peptide studies, wild-type (WT) mice (The Jackson Laboratory)
were injected IP with peptide (10 mg/kg) 30 minutes prior to
K/BxN serum injection and daily throughout the experiment.
Immunohistochemistry. Fixed ankles were decalcified
in EDTA (Sigma-Aldrich) in 10% formalin, embedded in
paraffin, and sectioned. Sections were stained with hematoxylin and eosin and with Safranin O–methyl green, and for
tartrate-resistant acid phosphatase (TRAP), CD45, proliferating cell nuclear antigen (PCNA), and F4/80 antigens. All
staining procedures were performed at the Saint Louis University core pathology facility except for TRAP staining, which
was performed at Washington University School of Medicine
Center for Musculoskeletal Biology and Medicine. Histopathologic scoring was performed, under blinded conditions, as
previously described (21), using an Olympus BX40CY microscope. Photographs were taken on an Olympus BX41 microscope
equipped with a DP20 Digital Camera (Olympus), at 100⫻
magnification.
Cell culture. Peritoneal cells from 6–8–week-old male
or female mice were harvested via lavage 3 days after IP
injection of 4% aged thioglycollate, adhered for 1 hour in
serum-free media, and then maintained in complete Dulbecco’s modified Eagle’s medium and used within 2 days. For
cytokine assays, peritoneal macrophages were incubated with
peptide (50 ␮M) for 2 hours, followed by stimulation with LPS
(10 ng/ml; Sigma-Aldrich) in the presence of peptide. For
IL-1␤ secretion assays, macrophages were further treated with
p21-MEDIATED SUPPRESSION OF INFLAMMATORY ARTHRITIS
ATP (5 mM; Sigma-Aldrich) for up to 30 minutes to induce
IL-1␤ release.
Luminex-based assays. Cytokine levels in serum or cell
supernatants were determined using Luminex-based assays according to the specifications of the manufacturer (Invitrogen).
Data were collected on a Luminex 200 using xPONENT software version 3.0 (Luminex) and fitted to a weighted 5-point
parameter log standard curve. For the peptide study (50 ␮M),
only the Tat-Ctrl peptide was used as a control, since Tat and
Tat-Ctrl showed no difference in arthritis development. For
phosphorylation assays, phosphorylated Akt (BD Biosciences),
phosphorylated p38 (Thr180/Tyr182), and total I␬B␣ protein
levels in whole cell lysates (prepared with a Cell Lysis Kit)
were measured using a Luminex-based assay according to the
instructions of the kit manufacturer (Bio-Rad). Data were
collected on a Luminex 200 with IS 2.3 software (Luminex).
Enzyme-linked immunosorbent assay (ELISA). For
detection of IL-1␤ in cell supernatants, sandwich ELISAs were
performed according to the instructions of the manufacturer
(R&D Systems). ELISA results were quantitated by absorbance at 450 nm on a microplate reader (Bio-Rad) and
normalized to the number of cells per well.
Peptides. A polycationic peptide derived from human
immunodeficiency virus 1 (HIV-1) Tat (22) was fused to the
p21 peptide mimetics, which were synthesized by and purchased from the Peptide Synthesis group at Tufts University
(Boston, MA). The peptides were as follows: aa 15–40
(Ac-rkkrr-orn-rrr-SKACRRLFGPVDSEQLSRDCDALMAG),
aa 46–65 (Ac-rkkrr-orn-rrr-RERWNFDFVTETPLEGDFAWOH), aa 63–77 (Ac-rkkrr-orn-rrr-AWERVRGLGLPKLY), and
aa 141–160 (Ac-rkkrr-orn-rrr-KRRQTSMTDFYHSKRRLIFS)
(numbers indicate amino acids [aa]). A negative control peptide, Tat-Ctrl (Ac-rkkrr-orn-rrr-SKACRRLKKPVDSEQLSRDCDALMAG), was designed by incorporating F22K and
G23K substitutions (23). A Tat (Ac-rkkrr-orn-rrr) peptide was
also used as a control. Fluorescein isothiocyanate (FITC)–
conjugated peptides were synthesized as above, with FITC
replacing the acetyl group on the amino terminus.
Flow cytometry. For evaluation of peptide entry, cells
were incubated with FITC-conjugated peptides for 2 hours at
4°C or 37°C. Cells were trypsinized to remove surface-bound
protein, and data were collected on a BD LSRII (BD Biosciences) using FACSDiva version 6.1.2 and analyzed using
FlowJo version 5.5.5 (Tree Star).
Cell imaging. Live cell imaging was performed at the
Northwestern University Feinberg School of Medicine Cell
Imaging Core Facility, on a Nikon C1Si laser scanning confocal
microscope fitted on a PerfectFocus stand to actively maintain
focal plane control. Cells were maintained at 37°C during
imaging, using a Tokai HIT stage top incubator. Confocal
imaging was performed using scan averaging of 12 and laser
dwell time of 1.92 ␮sec/pixel.
RESULTS
Increased severity of K/BxN serum–induced
arthritis in p21ⴚ/ⴚ mice. In p21⫺/⫺ mice backcrossed
onto the C57BL/6 background and verified for ⬎99%
C57BL/6 background contribution (20), there were no
143
obvious differences from WT mice in total leukocyte
numbers or in various populations of leukocytes (data
available from the corresponding author upon request).
Since p21 has been associated with suppression of
inflammatory disease, its role in the development of
inflammatory arthritis was evaluated using the K/BxN
serum–transfer model in WT and p21⫺/⫺ mice. Compared to WT mice, p21⫺/⫺ mice developed significantly
worse ankle swelling, as measured by the change in ankle
circumference following IP injection of K/BxN mouse
serum (Figure 1A). Additionally, p21⫺/⫺ mice displayed
evidence of more severe disease as assessed by a significant elevation in clinical score, most pronounced on day
25 (3.0-fold increase), when the disease had resolved in
WT mice but failed to fully resolve in p21⫺/⫺ mice
(Figure 1A).
Increased inflammatory cell numbers and articular destruction in the ankle joints of p21ⴚ/ⴚ mice. To
determine the extent of joint damage in WT and p21⫺/⫺
mice, ankles were harvested 7, 14, or 25 days following
injection of K/BxN mouse serum and examined histologically. Inflammation and development of pannus
were observed in the joints of both WT and p21⫺/⫺ mice,
particularly on days 7 and 14 (Figures 1B and C). At all
3 time points, pannus was more extensive in p21⫺/⫺ mice
than in control mice (by a mean of 2.9-, 1.5-, and 1.8-fold
on days 7, 14, and 25). The pannus formation was
associated with enhanced destruction of cartilage, particularly on day 14 (3-fold), and of bone at all time points
(3.7-, 1.5-, and 2.3-fold) in p21⫺/⫺ mice as compared to
WT mice (Figure 1C). Increased numbers of TRAPpositive cells, which were localized in pannus, were
found at all time points in p21⫺/⫺ mice compared to
control mice (Figures 1D and E). Long-term bone
destruction was further verified by micro-CT analysis
(Figure 1F), which showed increased bone damage in
p21⫺/⫺ mice. Since articular destruction persisted in
p21⫺/⫺ mice even at the late time point (25 days),
when there are clearly no arthritogenic antibodies remaining (11), these findings suggest that the deficiency
in p21 leads to sustained inflammation and/or a failure
of inflammation to resolve.
Because elevated cytokine production may contribute to the increased inflammation and destruction
observed in the ankles of p21⫺/⫺ mice, circulating cytokine levels were assessed. Seven days after the initiation
of arthritis, the serum level of the inflammatory cytokine
IL-6 was increased 3.0-fold in p21⫺/⫺ mice compared to
WT mice (Figure 1G). The level of IL-1␣ was significantly elevated in p21⫺/⫺ mice as well (Figure 1H).
144
MAVERS ET AL
Figure 1. Increased and prolonged inflammatory arthritis and elevated cytokine levels in p21⫺/⫺ mice. A, Change in ankle circumference and
clinical score in arthritic wild-type (WT) and p21⫺/⫺ mice (n ⫽ 15 per group on day 7, 10 per group on day 14, and 5 per group on day 25).
Data shown are representative of at least 2 independent experiments. B, Representative day 7 ankle sections stained with hematoxylin and eosin
(H&E). P ⫽ pannus; SL ⫽ synovial lining; C ⫽ cartilage; B ⫽ bone; BM ⫽ bone marrow. Original magnification ⫻ 100. C, Histologic scores of
H&E-stained ankle sections. Inflam. ⫽ inflammation; Synov. ⫽ synovial; Cartil. ⫽ cartilage, Lymph. ⫽ lymphocytes; PMN ⫽ polymorphonuclear
cells. Extraarticular refers to extraarticular inflammation. D, Representative day 25 ankle sections stained for tartrate-resistant acid phosphatase
(TRAP). Original magnification ⫻ 100. E, Number of TRAP-positive cells per 100⫻ field. Horizontal lines show the mean. F, Microfocal computed
tomographic imaging of WT and p21⫺/⫺ mouse ankles. AP ⫽ anteroposterior; ML ⫽ mediolateral. G and H, Levels of interleukin-6 (IL-6) (G) and
IL-1␣ (H) in serum collected on day 7 from arthritic ankles, measured using a Luminex-based assay. Values in A, C, and G are the mean ⫾ SEM.
ⴱ ⫽ P ⬍ 0.05 versus WT mice, by Student’s t-test.
p21-MEDIATED SUPPRESSION OF INFLAMMATORY ARTHRITIS
145
Figure 2. Increased inflammatory cell infiltration in p21⫺/⫺ mice. Arthritis was induced in wild-type (WT) and p21⫺/⫺ mice, and ankles (n ⫽ 10
per group) were prepared for histologic analysis. A–C, Representative antigen-retrieved paraffin-embedded ankle sections (day 7), stained with
antibody to CD45 (A), F4/80 (B), or proliferating cell nuclear antigen (PCNA) (C). Original magnification ⫻ 100. D, Number of positive cells in each
of the indicated regions. Values are the mean ⫾ SEM of at least 3 sections per ankle and 3 fields per section. ⴱ ⫽ P ⬍ 0.05 versus WT mice, by
Student’s t-test. Macs ⫽ macrophages; Prolif. ⫽ proliferating cells.
To further investigate the increased inflammation and destruction observed on hematoxylin and
eosin–stained sections, infiltration of immune cells
into arthritic joints was analyzed. Ankle sections were
stained for CD45 (hematopoietic cells), F4/80 (macrophages), and PCNA (proliferation). On days 7 and 14,
respectively, the number of hematopoietic cells
(CD45⫹) was increased 2.8- and 1.4-fold in the pannus
of p21⫺/⫺ mouse ankles as compared to control mouse
ankles (Figures 2A and D). On day 25, the number of
CD45⫹ cells did not differ between WT and p21⫺/⫺
mice. In addition, peak PCNA staining occurred on day
7 in the ankles of p21⫺/⫺ mice and on day 14 in WT mice
(Figures 2B and D), likely representing increased proliferation of synovial fibroblasts as macrophages are
terminally differentiated. The number of PCNA⫹ cells
did not differ between WT and p21⫺/⫺ mice on day 25
(Figure 2D). This suggests that synoviocytes may be
important at the early stages of disease development and
that infiltration of inflammatory cells at later time points
is central for disease progression. As such, macrophage
numbers in the pannus were significantly increased in
p21⫺/⫺ mice as compared to WT mice on days 7
(3.7-fold), 14 (1.6-fold), and 25 (1.8-fold) (Figures 2C
and D).
Since there was no difference in total CD45⫹
cells on day 25 yet there were more macrophages, the
above data suggest that the failure of arthritis to resolve
in p21⫺/⫺ mice may be due to the persistence of
macrophages in the synovium. Thus, similar to findings
in patients with RA (4,24), increased numbers of macrophages in the arthritic p21-deficient mice are associated
with more severe articular destruction.
Protection against K/BxN serum–transfer arthritis in vivo by a peptide mimetic corresponding to aa
141–160 of p21. To decipher the critical domain of
p21 that plays a vital role in suppression of inflammation, we examined the impact of delivering established
domains of p21 to arthritic mice. Previous studies have
identified domains on p21 that are critical for interaction
with cyclins, cyclin-dependent kinases, and PCNA, as
well as modification of the activities of these peptides
leading to alterations in cell cycle progression (21,25).
We took advantage of these studies and designed 6 sets
of p21 peptide mimetics that were conjugated to a
polycationic peptide derived from HIV-1 transactivator
of transcription to allow cell entry (Tat, Tat-Ctrl, aa
15–40, aa 46–65, aa 63–77, and aa 141–160). We have
previously shown that this approach is viable in this
model of inflammatory arthritis using BH3 peptide
mimetics (21).
The functionality of p21 peptide mimetics in
suppression of K/BxN serum–transfer arthritis was evaluated. WT mice treated with aa 141–160 had a 3.6-fold
146
MAVERS ET AL
Figure 3. Reduced severity of arthritis in vivo after treatment with the p21 peptide mimetic aa 141–160. Arthritis was induced in 8 mice per group,
and the mice were injected intraperitoneally with peptide (10 mg/kg) 30 minutes prior to injection of K/BxN serum and daily throughout the
experiment. A and C, Change in ankle circumference in wild-type mice (A) and p21⫺/⫺ mice (C). B and D, Clinical score in wild-type mice (B) and
p21⫺/⫺ mice (D). Values are the mean ⫾ SEM. ⴱ ⫽ P ⬍ 0.05, aa 141–160 versus Tat-Ctrl (control); † ⫽ P ⬍ 0.05, aa 141–160 versus Tat, by Student’s
t-test. Data shown are representative of at least 2 independent experiments.
reduction in ankle swelling on day 2, a 6-fold reduction
on day 4, and a 4-fold reduction on day 7 as compared to
mice treated with control peptide (Figure 3A). Additionally, there was marked improvement in the clinical score
in WT mice treated with aa 141–160 as compared to
Tat-Ctrl–treated mice (Figure 3B). While mild clinical
improvement in ankle swelling and clinical score was
observed with aa 15–40 and aa 63–77, these were not
significant. Since aa 141–160 treatment led to reduced
arthritis in p21-intact mice, we also examined its effect in
mice lacking p21. There was no difference in ankle
swelling or clinical score in p21⫺/⫺ mice treated with Tat
only or with aa 141–160 (Figures 3C and D).
Suppression of articular destruction and inflammatory cell infiltration by the p21 peptide mimetic aa
141–160. Ankles were harvested and examined histologically 7 days following arthritis induction and treatment
with peptide. Only treatment with the peptide corresponding to aa 141–160 consistently caused a significant
decrease in pannus development, inflammation, synovial
lining thickness, bone erosion, extraarticular inflam-
mation, and infiltration of lymphocytes and polymorphonuclear cells as compared to treatment with control
peptides (Figures 4A and B). No difference in cartilage
destruction was detected in any of the groups of mice.
Furthermore, decreased infiltration of all hematopoietic
cells (CD45) and, in particular, macrophages (F4/80),
was observed within the pannus of aa 141–160–treated
mice (Figures 5A and B). No differences were noted
in areas of normal synovium. PCNA staining also revealed reduced proliferation in the pannus, although this
effect was elicited by several of the peptides (Figures 5A
and B).
The p21 peptide mimetic aa 141–160 suppresses
peritoneal macrophage production of cytokines following stimulation with TLR agonists. We have shown that
while Tat-conjugated peptides enter all cells, macrophages appear to preferentially uptake the peptides (21).
To explore the mechanism by which aa 141–160 reduces
the severity of arthritis, we focused on the effect of this
peptide on innate immune responses induced by TLR
agonists. Entry and cytoplasmic localization of FITC-
p21-MEDIATED SUPPRESSION OF INFLAMMATORY ARTHRITIS
147
Figure 4. Suppression of articular destruction by treatment with aa 141–160. Wild-type mice were treated with peptide, arthritis was induced,
and ankles (16 per group) were prepared for histologic analysis. A, Representative ankle sections stained with hematoxylin and eosin (H&E).
Original magnification ⫻ 100. B, Histologic scores of H&E-stained ankle sections. Scores were determined as previously described (11,21,45,46).
Values are the mean ⫾ SEM of at least 3 sections per ankle and 3 fields per section. ⴱ ⫽ P ⬍ 0.05 versus Tat-Ctrl (control); † ⫽ P ⬍ 0.05 versus
Tat, by Student’s t-test. Data shown are representative of at least 2 independent experiments.
conjugated Tat peptide into peritoneal macrophages was
confirmed using confocal microscopy (Figure 6A). The
percent and amount of incorporation was determined by
flow cytometry. One hundred percent of the cells equally
incorporated the Tat-Ctrl and Tat p21 peptide mimetics
(Figures 6B and C), and only a minor reduction in cell
survival was observed, even at 24 and 48 hours after
administration (Mavers M, et al: unpublished observations). The capacity of aa 141–160 to inhibit cytokine
production by peritoneal macrophages activated with
TLR ligation (LPS) was also assessed. The aa 141–160
peptide, which protected against K/BxN serum–induced
arthritis in vivo (Figures 3–5), suppressed production of
the inflammatory cytokines tumor necrosis factor ␣,
IL-6, and IL-1␤ in vitro (Figures 6D–F). There was no
inhibitory effect of the p21 peptide mimetic on production of macrophage inflammatory protein 1␣ (MIP-1␣),
MIP-1␤, or RANTES (Mavers M, et al: unpublished
observations).
The p21 peptide mimetic aa 141–160 increases
active Akt but reduces p38 activity in TLR-stimulated
macrophages. Because the p21 peptide mimetic aa 141–
160 induced a dramatic reduction in the production of
proinflammatory cytokines and development of arthritis,
we examined its effect on upstream signaling events,
using a Luminex-based assay. Treatment with aa 141–
160 led to a marked increase in activation of the
serine/threonine protein kinase Akt as compared to
Tat-Ctrl–treated macrophages beginning at 15 minutes
after stimulation with LPS, and Akt levels remained
higher throughout 2 hours of stimulation (Figure 6G). In
contrast, a reduction in the phosphorylation of p38 was
observed at 30 minutes and 60 minutes following TLR
ligation (Figure 6H). The peptide had little effect on the
degradation of I␬B as compared to control peptide–
treated cells (Figure 6I).
The effect of full-length p21 on intracellular
signaling pathways was also explored in WT and
148
MAVERS ET AL
Figure 5. Reduced proliferation and inflammatory cell infiltration in ankles from aa 141–160–treated mice. Wild-type mice were treated with
peptide, arthritis was induced, and ankles (16 per group) were prepared for histologic analysis. A, Representative antigen-retrieved paraffinembedded ankle sections stained with antibody to CD45, F4/80, or proliferating cell nuclear antigen (PCNA). Original magnification ⫻ 100. B,
Number of cells positive for CD45, F4/80, and PCNA. Values are the mean ⫾ SEM of at least 3 sections per ankle and 3 fields per section. ⴱ ⫽ P
⬍ 0.05 versus Tat-Ctrl (control); † ⫽ P ⬍ 0.05 versus Tat, by Student’s t-test. Data shown are representative of at least 2 independent experiments.
p21⫺/⫺ mouse peritoneal macrophages activated with
LPS, using flow cytometry. Cells from p21⫺/⫺ mice
displayed decreased phosphorylation and activation of
Akt 15 minutes after LPS stimulation, as compared to
control cells; subsequently, at 30 minutes and 60 minutes
following TLR ligation, an increase in phosphorylated p38 in p21⫺/⫺ mouse cells as compared to WT
mouse cells was observed (data available from the
corresponding author upon request). Taken together,
these results suggest that p21, via its C-terminal domain,
suppresses macrophage function by enhancing the phosphorylation of Akt, thereby reducing p38 activation and
subsequently limiting inflammatory cytokine production
in TLR-stimulated macrophages.
DISCUSSION
Over the last several years, p21 has gained attention in the fields of inflammation and autoimmunity.
Numerous studies have been performed using p21⫺/⫺
mice to examine the role of p21 in murine models of
sepsis, lupus, and RA. Recently, we and others have
shown that p21⫺/⫺ mouse macrophages, regardless of
background, display enhanced activation in response to
p21-MEDIATED SUPPRESSION OF INFLAMMATORY ARTHRITIS
149
Figure 6. The p21 peptide mimetic aa 141–160 enters into macrophages and reduces production of inflammatory cytokines following Toll-like
receptor stimulation in vitro. A, Peritoneal macrophages were incubated for 2 hours with no peptide, Tat-Ctrl (control), or fluorescein isothiocyanate
(FITC)–conjugated aa 141–160 and examined by confocal microscopy. Original magnification ⫻ 200. B and C, Peritoneal macrophages were
incubated for 2 hours at 37°C (B) or at 4°C or 37°C (C) with no peptide or with an FITC-conjugated Tat peptide corresponding to various domains
of p21. Shown in B is an overlay of results obtained with all of the cells treated with the Tat-conjugated, FITC-labeled peptides at 37°C. D–F, Levels
of tumor necrosis factor ␣ (TNF␣) (D), interleukin-6 (IL-6) (E), and IL-1␤ (F) in supernatants from lipopolysaccharide-stimulated peritoneal
macrophages incubated with aa 141–160 or control peptide were analyzed by enzyme-linked immunosorbent assay. G–I, Expression of
phosphorylated Akt (G), phosphorylated p38 (H), and total I␬B (I) in cell lysates was analyzed by Luminex-based assay. Data were normalized to
cell number. Values are the mean ⫾ SEM fold change relative to untreated cells. ⴱ ⫽ P ⬍ 0.05 versus Tat-Ctrl, by Student’s t-test. Data shown are
representative of at least 2 independent experiments.
TLR agonists, as compared to control macrophages
(18,19). Further, p21⫺/⫺ mice on either a mixed or an
inbred background are more susceptible to LPS-induced
endotoxic shock (18,19). These data suggest that in
sepsis, the background of the mice is not a contributing
factor. However, in spontaneous development of lupus,
the background of the mice may be more crucial (26,27).
The potentially conflicting results in the lupus studies
may be attributed to the notion that hybrid (C57Bl/6:
129) mice are more susceptible to spontaneous autoimmunity due to epistatic interactions between the two
genomes (28).
In support of the argument that p21 may be
considered a general inhibitor of autoimmune disease,
studies by Salvador and colleagues showed that loss of
p21 leads to early lethality due to lupus-like disease,
which is enhanced by the concomitant loss of GADD45a
(29). Further, genome-wide scanning studies have now
150
shown that p21 is a susceptibility locus for SLE (30,31).
Taken together, these results suggest that p21 may
be considered a negative regulator of spontaneous autoimmunity. However, its direct role in inhibiting macrophage function was not examined in those studies.
In the present study we have shown that
p21⫺/⫺ mice backcrossed onto a C57BL/6 background
for at least 12 generations and screened for more than
150 loci (20) develop a markedly more severe experimental RA-like disease (Figure 1). The arthritis in
p21⫺/⫺ mice fails to resolve as compared to that in WT
mice, with continuous articular destruction and a corresponding increase in macrophage numbers (Figures 1
and 2). While we previously obtained conflicting data
with regard to the role of p21 in the development of
K/BxN serum–transfer arthritis (11), these differences
are attributable to the mouse background. After extensive phenotyping of the mice, we discovered that p21⫺/⫺
mice on a mixed background, but not on an inbred
background, develop significantly fewer inflammatory
monocytes when compared to controls or even mice
(op/op) lacking macrophage colony-stimulating factor
(32). Further, injection of WT mouse macrophages into
p21⫺/⫺ mice restores their susceptibility to inflammatory
arthritis (11).
Thus, we were the first to show that p21 cooperates with 129 loci to produce inflammatory monocytes
and that inflammatory monocytes are crucial for
K/BxN serum–induced arthritis (11). However, p21 is
not required for the differentiation of mouse bone
marrow–derived macrophages, splenic-derived macrophages, or thioglycollate-elicited peritoneal macrophages
(17,33,34), regardless of the background. Further,
p21⫺/⫺ mice display similar numbers of tissue macrophages as compared to p21-intact mice, regardless of the
background (C57Bl/6 versus C57BL/6:129). Collectively,
these data are consistent with the results of immunohistochemical studies of p21 expression in RA synovium
(17) and therapeutic studies using adenoviral vectors
expressing p21 (12–14,16), which demonstrate that p21
may be an important inhibitor of inflammatory arthritis.
Despite the general success of biologic therapy,
many patients continue to experience the severely debilitating effects of RA. In addition, the mechanism
behind the persistent production of proinflammatory
cytokines targeted by these therapies remains to be
fully elucidated. We have now demonstrated that a
Tat-conjugated peptide mimetic corresponding to the
C-terminus of p21 significantly reduces the severity of
arthritis development in the K/BxN serum–transfer
MAVERS ET AL
mouse model (Figures 3–5). Further, the aa 141–160
peptide requires the presence of p21 to reduce arthritis
development. Previous studies have demonstrated the
ability of Tat to carry peptides and other cargo into
cells in vitro and in vivo (22,35), and in the current
study, entry into cells was confirmed by cell imaging,
flow cytometry, and functional assays (Figure 6). WT
mice were used for these peptide studies to more closely
replicate the condition encountered in humans. Goulvestre et al have shown that a peptide mimetic corresponding to aa 141–160 was able to reduce the severity
of lupus-like disease in mice (36). This effect was
attributed to suppression of lymphocyte proliferation via
inhibition of PCNA by the peptide, as this region at the
C-terminus of p21 encompasses the PCNA binding
domain (25).
While we found that the aa 141–160 peptide
mimetic reduces proliferation in ankles from WT mice
7 days after injection with K/BxN mouse serum (Figure 5), the other p21 peptide mimetics also reduced
proliferation of synovial fibroblasts in vivo, but had no
effect on clinical outcome. In vivo, macrophages are
terminally differentiated and also appear to preferentially take up Tat-conjugated peptides over other immune cells (21). Therefore, the cells most likely affected
by aa 141–160–mediated suppression of proliferation are
synovial fibroblasts. Our study also demonstrated a
significant decrease in hematopoietic cell infiltration,
particularly by macrophages, in the ankles of aa 141–
160–treated mice (Figure 5). Furthermore, we showed
that aa 141–160 suppresses inflammatory cytokine production in macrophages (Figure 6). These data are
consistent with the results of previous studies showing
p21-mediated reduction in macrophage activation following TLR ligation (18,19). Goulvestre and colleagues
further attributed the decreased lupus development in
aa 141–160–treated mice to significant proapoptotic
effects on lymphocytes (36). However, we observed no
difference in apoptosis between WT and p21⫺/⫺ mouse
macrophages, or following treatment with any of the
peptides (Mavers M, et al: unpublished observations).
In pursuing the mechanism by which the aa
141–160 peptide mimetic mediates a reduction in arthritis severity and suppression of cytokine production, we
found that it enhances Akt phosphorylation and inhibits
p38 activation (Figure 6). Similarly, p21⫺/⫺ mouse cells
display reduced Akt phosphorylation and increased p38
activation as compared to WT mouse cells (data available from the corresponding author upon request).
These data are consistent with our previous work demonstrating increased levels of IL-6 and IL-1␤ messenger
p21-MEDIATED SUPPRESSION OF INFLAMMATORY ARTHRITIS
RNA in p21⫺/⫺ mouse macrophages (18), which suggests that the mechanism by which p21 inhibits inflammatory cytokine production is likely to be suppression of
intracellular signaling pathways, leading to a reduction
in transcription of these cytokines.
This effect of p21 on activation of Akt and
suppression of p38, and subsequent reduction in secretion of proinflammatory cytokines, is not surprising
given previous studies showing that sustained Akt activation leads to decreased production of these cytokines
(37,38). Furthermore, enhanced signaling through the
MAPK pathway leads to increased production of inflammatory cytokines and has been shown to play a role in
RA pathogenesis (39). In fact, the targeting of intracellular signaling pathways is a prominent focus of
studies exploring novel mechanisms for the treatment
of arthritis (40). Our data suggesting that the peptide
mimetic corresponding to the C-terminus of p21 is
sufficient to enhance Akt activation and suppress p38
activation (Figure 6) are supported by previous studies
demonstrating that p21 interacts with Akt at the
C-terminus (41). Interestingly, Akt has been shown to
induce retention of p21 in the cytoplasm and enhance its
stability by phosphorylating p21 on its C-terminus (42).
Thus, one could envision a feedback loop in which
p21-mediated Akt activation further activates p21 cytoplasmic functions, thereby perpetuating p38 suppression
in order to turn off inflammatory reactions.
Additional work is needed in order to determine
whether p21 also regulates other intracellular signaling
pathways. Previous studies have shown that NF-␬B
DNA binding is increased in p21-deficient mouse macrophages, along with increased I␬B degradation and I␬B
kinase complex activity (19). Other studies have shown
that p21 may in fact promote transcriptional activation
by NF-␬B via activation of p300 through derepression of
a repression motif in a promoter-dependent manner;
however, those studies were not conducted in macrophages (43). We found that the aa 141–160 peptide
mimetic has little to no effect on I␬B␣ degradation
(Figure 6I). In addition, direct inhibition of JNK by p21
has been previously demonstrated in cell-free systems,
293 cells, and synovial fibroblasts (13,17,44).
Further research is necessary to determine the
exact role of p21 in regulating intracellular signaling,
particularly in the context of macrophage activation.
Elucidating the mechanisms by which macrophages
are activated or inhibited remains crucial to promoting
the development of new treatments for inflammatory
disease.
151
ACKNOWLEDGMENTS
The authors are grateful for the assistance of staff of
the Washington University School of Medicine Center for
Musculoskeletal Biology and Medicine and the Northwestern
University Feinberg School of Medicine Cell Imaging Core
Facility and the Methodology and Data Management Core.
AUTHOR CONTRIBUTIONS
All authors were involved in drafting the article or revising it
critically for important intellectual content, and all authors approved
the final version to be published. Dr. Perlman had full access to all of
the data in the study and takes responsibility for the integrity of the
data and the accuracy of the data analysis.
Study conception and design. Mavers, Cuda, Misharin, Agrawal,
Balomenos, Perlman.
Acquisition of data. Mavers, Cuda, Misharin, Gierut, Agrawal, Weber,
Novack, Haines, Balomenos, Perlman.
Analysis and interpretation of data. Mavers, Cuda, Misharin, Gierut,
Agrawal, Novack, Haines, Balomenos, Perlman.
REFERENCES
1. Lavagno L, Gunella G, Bardelli C, Spina S, Fresu LG, Viano I,
et al. Anti-inflammatory drugs and tumor necrosis factor-␣ production from monocytes: role of transcription factor NF-␬B and
implication for rheumatoid arthritis therapy. Eur J Pharmacol
2004;501:199–208.
2. Mulherin D, FitzGerald O, Bresnihan B. Synovial tissue macrophage populations and articular damage in rheumatoid arthritis.
Arthritis Rheum 1996;39:115–24.
3. Bresnihan B, Pontifex E, Thurlings RM, Vinkenoog M, ElGabalawy H, Fearon U, et al. Synovial tissue sublining CD68
expression is a biomarker of therapeutic response in rheumatoid
arthritis clinical trials: consistency across centers. J Rheumatol
2009;36:1800–2.
4. Wijbrandts CA, Vergunst CE, Haringman JJ, Gerlag DM, Smeets
TJ, Tak PP. Absence of changes in the number of synovial
sublining macrophages after ineffective treatment for rheumatoid
arthritis: implications for use of synovial sublining macrophages as
a biomarker. Arthritis Rheum 2007;56:3869–71.
5. Vergunst CE, van de Sande MG, Lebre MC, Tak PP. The role of
chemokines in rheumatoid arthritis and osteoarthritis. Scand
J Rheumatol 2005;34:415–25.
6. Polzer K, Baeten D, Soleiman A, Distler J, Gerlag DM, Tak PP,
et al. Tumour necrosis factor blockade increases lymphangiogenesis in murine and human arthritic joints. Ann Rheum Dis
2008;67:1610–6.
7. Hutcheson J, Perlman H. BH3-only proteins in rheumatoid arthritis: potential targets for therapeutic intervention. Oncogene
2008;27 Suppl 1:S168–75.
8. Huang Q, Ma Y, Adebayo A, Pope RM. Increased macrophage
activation mediated through Toll-like receptors in rheumatoid
arthritis. Arthritis Rheum 2007;56:2192–201.
9. Solomon S, Rajasekaran N, Jeisy-Walder E, Snapper SB, Illges H.
A crucial role for macrophages in the pathology of K/B ⫻ N
serum-induced arthritis. Eur J Immunol 2005;35:3064–73.
10. Campbell IK, Rich MJ, Bischof RJ, Hamilton JA. The colonystimulating factors and collagen-induced arthritis: exacerbation of
disease by M-CSF and G-CSF and requirement for endogenous
M-CSF. J Leukoc Biol 2000;68:144–50.
11. Scatizzi JC, Hutcheson J, Bickel E, Woods JM, Klosowska K,
Moore TL, et al. p21Cip1 is required for the development of
152
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
MAVERS ET AL
monocytes and their response to serum transfer-induced arthritis.
Am J Pathol 2006;168:1531–41.
Nasu K, Kohsaka H, Nonomura Y, Terada Y, Ito H, Hirokawa K,
et al. Adenoviral transfer of cyclin-dependent kinase inhibitor
genes suppresses collagen-induced arthritis in mice. J Immunol
2000;165:7246–52.
Nonomura Y, Kohsaka H, Nagasaka K, Miyasaka N. Gene
transfer of a cell cycle modulator exerts anti-inflammatory effects
in the treatment of arthritis. J Immunol 2003;171:4913–9.
Nonomura Y, Kohsaka H, Nasu K, Terada Y, Ikeda M, Miyasaka
N. Suppression of arthritis by forced expression of cyclindependent kinase inhibitor p21(Cip1) gene into the joints. Int
Immunol 2001;13:723–31.
Nonomura Y, Nagasaka K, Hagiyama H, Sekine C, Nanki T,
Tamamori-Adachi M, et al. Direct modulation of rheumatoid
inflammatory mediator expression in retinoblastoma protein–
dependent and –independent pathways by cyclin-dependent kinase
4/6. Arthritis Rheum 2006;54:2074–83.
Taniguchi K, Kohsaka H, Inoue N, Terada Y, Ito H, Hirokawa K,
et al. Induction of the p16INK4a senescence gene as a new
therapeutic strategy for the treatment of rheumatoid arthritis.
Nat Med 1999;5:760–7.
Perlman H, Bradley K, Liu H, Cole S, Shamiyeh E, Smith RC,
et al. IL-6 and matrix metalloproteinase-1 are regulated by the
cyclin-dependent kinase inhibitor p21 in synovial fibroblasts. J Immunol 2003;170:838–45.
Scatizzi JC, Mavers M, Hutcheson J, Young B, Shi B, Pope RM,
et al. The CDK domain of p21 is a suppressor of IL-1␤-mediated
inflammation in activated macrophages. Eur J Immunol 2009;39:
820–5.
Trakala M, Arias CF, Garcia MI, Moreno-Ortiz MC, Tsilingiri K,
Fernandez PJ, et al. Regulation of macrophage activation and
septic shock susceptibility via p21(WAF1/CIP1). Eur J Immunol
2009;39:810–9.
Arias CF, Ballesteros-Tato A, Garcia MI, Martin-Caballero J,
Flores JM, Martinez AC, et al. p21CIP1/WAF1 controls proliferation of activated/memory T cells and affects homeostasis and
memory T cell responses. J Immunol 2007;178:2296–306.
Scatizzi JC, Hutcheson J, Pope RM, Firestein GS, Koch AE,
Mavers M, et al. Bim–Bcl-2 homology 3 mimetic therapy is
effective at suppressing inflammatory arthritis through the activation of myeloid cell apoptosis. Arthritis Rheum 2010;62:441–51.
Gammon ST, Villalobos VM, Prior JL, Sharma V, Piwnica-Worms
D. Quantitative analysis of permeation peptide complexes labeled
with technetium-99m: chiral and sequence-specific effects on net
cell uptake. Bioconjug Chem 2003;14:368–76.
Chen J, Saha P, Kornbluth S, Dynlacht BD, Dutta A. Cyclinbinding motifs are essential for the function of p21CIP1. Mol Cell
Biol 1996;16:4673–82.
Bresnihan B, Gerlag DM, Rooney T, Smeets TJ, Wijbrandts CA,
Boyle D, et al. Synovial macrophages as a biomarker of response
to therapeutic intervention in rheumatoid arthritis: standardization and consistency across centers. J Rheumatol 2007;34:620–2.
Dotto GP. p21(WAF1/Cip1): more than a break to the cell cycle?
Biochim Biophys Acta 2000;1471:M43–56.
Balomenos D, Martin-Caballero J, Garcia MI, Prieto I, Flores JM,
Serrano M, et al. The cell cycle inhibitor p21 controls T-cell
proliferation and sex-linked lupus development. Nat Med 2000;6:
171–6.
Santiago-Raber ML, Lawson BR, Dummer W, Barnhouse M,
Koundouris S, Wilson CB, et al. Role of cyclin kinase inhibitor p21
in systemic autoimmunity. J Immunol 2001;167:4067–74.
28. Rivera J, Tessarollo L. Genetic background and the dilemma of
translating mouse studies to humans. Immunity 2008;28:1–4.
29. Salvador JM, Hollander MC, Nguyen AT, Kopp JB, Barisoni L,
Moore JK, et al. Mice lacking the p53-effector gene Gadd45a
develop a lupus-like syndrome. Immunity 2002;16:499–508.
30. Kim K, Sung YK, Kang CP, Choi CB, Kang C, Bae SC. A
regulatory SNP at position ⫺899 in CDKN1A is associated with
systemic lupus erythematosus and lupus nephritis. Genes Immun
2009;10:482–6.
31. Kong EK, Chong WP, Wong WH, Lau CS, Chan TM, Ng PK, et al.
p21 gene polymorphisms in systemic lupus erythematosus. Rheumatology (Oxford) 2007;46:220–6.
32. Brown NJ, Hutcheson J, Bickel E, Scatizzi JC, Albee LD, Haines
GK III, et al. Fas death receptor signaling represses monocyte
numbers and macrophage activation in vivo. J Immunol 2004;173:
7584–93.
33. Comalada M, Xaus J, Sanchez E, Valledor AF, Celada A.
Macrophage colony-stimulating factor-, granulocyte-macrophage
colony-stimulating factor-, or IL-3-dependent survival of macrophages, but not proliferation, requires the expression of p21(Waf1)
through the phosphatidylinositol 3-kinase/Akt pathway. Eur J
Immunol 2004;34:2257–67.
34. Xaus J, Comalada M, Cardo M, Valledor AF, Celada A. Decorin
inhibits macrophage colony-stimulating factor proliferation of macrophages and enhances cell survival through induction of p27(Kip1)
and p21(Waf1). Blood 2001;98:2124–33.
35. Gump JM, Dowdy SF. TAT transduction: the molecular mechanism and therapeutic prospects. Trends Mol Med 2007;13:443–8.
36. Goulvestre C, Chereau C, Nicco C, Mouthon L, Weill B, Batteux
F. A mimic of p21WAF1/CIP1 ameliorates murine lupus. J Immunol
2005;175:6959–67.
37. Fang H, Pengal RA, Cao X, Ganesan LP, Wewers MD, Marsh CB,
et al. Lipopolysaccharide-induced macrophage inflammatory response is regulated by SHIP. J Immunol 2004;173:360–6.
38. Luyendyk JP, Schabbauer GA, Tencati M, Holscher T, Pawlinski
R, Mackman N. Genetic analysis of the role of the PI3K-Akt
pathway in lipopolysaccharide-induced cytokine and tissue factor
gene expression in monocytes/macrophages. J Immunol 2008;180:
4218–26.
39. Sweeney SE, Firestein GS. Mitogen activated protein kinase
inhibitors: where are we now and where are we going? Ann
Rheum Dis 2006;65 Suppl III:iii83–8.
40. Mavers M, Ruderman EM, Perlman H. Intracellular signal pathways: potential for therapies. Curr Rheumatol Rep 2009;11:
378–85.
41. Franke TF. PI3K/Akt: getting it right matters. Oncogene 2008;27:
6473–88.
42. Li Y, Dowbenko D, Lasky LA. AKT/PKB phosphorylation of
p21Cip/WAF1 enhances protein stability of p21Cip/WAF1 and
promotes cell survival. J Biol Chem 2002;277:11352–61.
43. Perkins ND, Felzien LK, Betts JC, Leung K, Beach DH, Nabel GJ.
Regulation of NF-␬B by cyclin-dependent kinases associated with
the p300 coactivator. Science 1997;275:523–7.
44. Shim J, Lee H, Park J, Kim H, Choi EJ. A non-enzymatic p21
protein inhibitor of stress-activated protein kinases. Nature 1996;
381:804–6.
45. Scatizzi JC, Bickel E, Hutcheson J, Haines GK III, Perlman H.
Bim deficiency leads to exacerbation and prolongation of joint
inflammation in experimental arthritis. Arthritis Rheum 2006;54:
3182–93.
46. Scatizzi JC, Hutcheson J, Bickel E, Haines GK III, Perlman H.
Pro-apoptotic Bid is required for the resolution of the effector
phase of inflammatory arthritis. Arthritis Res Ther 2007;9:R49.
Документ
Категория
Без категории
Просмотров
2
Размер файла
2 110 Кб
Теги
art, 33311
1/--страниц
Пожаловаться на содержимое документа