Published OnlineFirst March 20, 2013; DOI: 10.1158/1078-0432.CCR-12-2754
Clinical
Cancer
Research
Cancer Therapy: Preclinical
The Proteasome Inhibitor Carfilzomib Functions
Independently of p53 to Induce Cytotoxicity and an Atypical
NF-kB Response in Chronic Lymphocytic Leukemia Cells
Sneha V. Gupta1, Erin Hertlein2, Yanhui Lu4, Ellen J. Sass2, Rosa Lapalombella2, Timothy L. Chen2,
Melanie E. Davis2, Jennifer A. Woyach2, Amy Lehman3, David Jarjoura4, John C. Byrd1,2, and David M. Lucas1,2
Abstract
Purpose: The proteasome consists of chymotrypsin-like (CT-L), trypsin-like, and caspase-like subunits
that cleave substrates preferentially by amino acid sequence. Proteasomes mediate degradation of regulatory
proteins of the p53, Bcl-2, and nuclear factor-kB (NF-kB) families that are aberrantly active in chronic
lymphocytic leukemia (CLL). CLL remains an incurable disease, and new treatments are especially needed in
the relapsed/refractory setting. We therefore investigated the effects of the proteasome inhibitor carfilzomib
(CFZ) in CLL cells.
Experimental Design: Tumor cells from CLL patients were assayed in vitro using immunoblotting, realtime polymerase chain reaction, and electrophoretic mobility shift assays. In addition, a p53 dominantnegative construct was generated in a human B-cell line.
Results: Unlike bortezomib, CFZ potently induces apoptosis in CLL patient cells in the presence of
human serum. CLL cells have significantly lower basal CT-L activity compared to normal B and T cells,
although activity is inhibited similarly in T cells versus CLL. Co-culture of CLL cells on stroma protected from
CFZ-mediated cytotoxicity; however, PI3K inhibition significantly diminished this stromal protection. CFZmediated cytotoxicity in leukemic B cells is caspase-dependent and occurs irrespective of p53 status. In CLL
cells, CFZ promotes atypical activation of NF-kB evidenced by loss of cytoplasmic IkBa, phosphorylation of
IkBa, and increased p50/p65 DNA binding, without subsequent increases in canonical NF-kB target gene
transcription.
Conclusions: Together, these data provide new mechanistic insights into the activity of CFZ in CLL and
support phase I investigation of CFZ in this disease. Clin Cancer Res; 19(9); 2406–19. 2013 AACR.
Introduction
Chronic lymphocytic leukemia (CLL) is characterized by
the growth and accumulation of malignant B-lymphocytes
in the blood, bone marrow, lymph nodes, and spleen.
Despite the introduction of new therapies, most CLL patients
eventually relapse and succumb to their disease. Identification of new strategies for relapsed and refractory CLL is
essential to make a positive impact on patient survival.
Authors' Affiliations: 1College of Pharmacy; 2Department of Internal
Medicine; 3Center for Biostatistics, The Ohio State University, Columbus,
Ohio; and 4Department of Pharmacology and Molecular Sciences, Johns
Hopkins University School of Medicine, Baltimore, Maryland
Note: Supplementary data for this article are available at Clinical Cancer
Research Online (http://clincancerres.aacrjournals.org/).
J.C. Byrd and D.M. Lucas contributed equally to this article.
Corresponding Authors: John C. Byrd, 410 W. 12th Avenue, Room 455
OSUCCC Building, Columbus, OH 43210. Phone: 614-293-8330; Fax: 614292-3312; E-mail: [email protected]; and David M. Lucas. Phone:
614-685-5667; Fax: 614-292-3312; E-mail: [email protected]
doi: 10.1158/1078-0432.CCR-12-2754
2013 American Association for Cancer Research.
2406
Protein homeostasis is critical for the processes that
govern cell survival, and the multicatalytic 26S proteasome regulates the turnover of key proteins in these
processes. Proteasomes consist of a catalytic 20S core
capped by regulatory 19S subunits (1). There are 2 forms
of the 20S core: c20S and i20S. The former is ubiquitiously expressed in all cells, and the latter specifically in
hematopoietic cells and cells exposed to inflammatory
cytokines (2). The active site in the c20S and i20S core has
6 N-terminal threonine proteases consisting of chymotrypsin-like (CT-L), trypsin-like (T-L), and caspase-like
(C-L) subunits (3, 4). Because selective inhibition of the
CT-L site has a minor effect on total cellular protein
turnover, proteasome inhibitors that block only the
CT-L subunit are considered optimal (5). Proteasomes
mediate degradation of a variety of key regulatory factors
including cell-cycle control proteins (e.g., cyclins (6), p21
(7) and p27 (8), p53 (9), p53 target proteins Puma, Noxa,
and Bax of the Bcl-2 family (10), and the inhibitor of
nuclear factor-kB (NF-kB; IkB; ref. 11). Imbalanced
expression of Bcl-2 family proteins, constitutive NF-kB
activation, and variable p53 function are hallmarks of
CLL cells (12–14).
Clin Cancer Res; 19(9) May 1, 2013
Downloaded from clincancerres.aacrjournals.org on June 18, 2017. © 2013 American Association for Cancer Research.
Published OnlineFirst March 20, 2013; DOI: 10.1158/1078-0432.CCR-12-2754
Carfilzomib in Chronic Lymphocytic Leukemia
Translational Relevance
Standard therapies for chronic lymphocytic leukemia
(CLL) are often ineffective in p53-mutated cases and
detrimental to existing T cells, leaving patients at risk of
opportunistic infections. A major focus of therapeutic
development has been the investigation of agents that
are cytotoxic to B cells independent of p53 but have
minimal effects on T cells. Proteasome inhibitors have
known activity in multiple myeloma, but to date have
not shown clinical efficacy in CLL. This manuscript
shows B-cell selective efficacy of the proteasome inhibitor carfilzomib (CFZ) that is independent of p53, and
provides new mechanistic insights to its mode of action
in CLL patient cells. This work supports an ongoing
phase I clinical trial of CFZ in CLL.
Bortezomib (BTZ, Velcade) is a proteasome inhibitor
approved for the treatment of multiple myeloma and mantle cell lymphoma (15). Concentrations of BTZ that produce an antitumor response in vitro inhibit activities of the
CT-L and C-L subunits of the proteasome (2). In spite of a
high degree of cytotoxicity in vitro in CLL cells, BTZ failed to
produce objective responses in CLL patients in a phase II
clinical trial at the achieved doses (16). The lack of BTZ
efficacy in vivo has been attributed to the inactivation of its
boronate moiety by dietary flavonoids in human plasma
(17). Carfilzomib (CFZ, PR-171) is a novel proteasome
inhibitor that specifically and irreversibly inhibits the CTL activity of the proteasome (18). Unlike BTZ, CFZ has
minimal activity against off-target enzymes including serine
proteases, while at the same time inhibiting the CT-L
subunit of the proteasome more potently (19–21). Importantly, CFZ lacks the boronate moiety of BTZ that is potentially responsible for that agent’s inactivity in CLL patients.
Here, we investigate the effects of CFZ on CLL patient cells.
This work shows that CFZ irreversibly inhibits the CT-L
activity, has potent activity in CLL including cases with del
(17p13.1), and promotes an atypical activation of NF-kB
that may lack the classical prosurvival effect of this pathway.
Materials and Methods
Reagents
CFZ was provided by Onyx Pharmaceuticals. Boc-D-FMK
(Enzyme Systems Products) was used at 100 mmol/L. BTZ
was obtained from Millennium Pharmaceuticals Inc., and
TNF from R&D Systems. CD40L was purchased from PeproTech. 2-Fluoro-ara-A (active metabolite of fludarabine),
G418, doxycycline, and puromycin were purchased from
Sigma. CpG DSP30 (22) was purchased from Eurofins/
Operon.
Cells and cell lines
Blood was obtained from patients following written,
informed consent under a protocol approved by the Institutional Review Board of The Ohio State University. All
www.aacrjournals.org
patients examined had immunophenotypically defined
CLL as outlined by IWCLL criteria (23) and were newly
diagnosed or without treatment for a minimum of 30 days
at time of collection. The occurrence of del(17p13.1) was
determined in CLL patient samples by fluorescence in situ
hybridization as described (24), and in each positive case at
least 30% of cells showed this deletion. Normal cells were
obtained from partial leukocyte preparations from the
American Red Cross. B- or T-lymphocytes and CLL cells
were negatively selected using RosetteSep reagents (StemCell Technologies). The HS-5-GFP stromal cell line was
provided by Dr. Beverly Torok-Storb (Fred Hutchinson
Cancer Research Center; ref. 25). 293 cells were obtained
from American Type Culture Collection (Manassas, VA) and
697 lymphoblastic cells were obtained from DSMZ. Cells
were incubated at 37 C and 5% CO2 in AIM-V medium
(Invitrogen) or in RPMI 1640 with 10% FBS or 10% human
serum (HS) supplemented with penicillin, streptomycin,
and L-glutamine (Sigma).
Viability assays
Cell viability was monitored by flow cytometry using
annexin V/propidium iodide (BD Biosciences) on a FC500
instrument (Beckman Coulter). CellTiter 96 (MTS) assays
were conducted to monitor growth inhibition per manufacturer’s instructions (Promega). LIVE/DEAD (Invitrogen)
staining was done to monitor cytotoxicity with drug treatments using the manufacturer’s instructions.
Immunoblot analyses
Nuclear and cytoplasmic lysates were prepared with NEPER Nuclear and Cytoplasmic Extraction kit (Pierce). Antibody to polyADP-ribose polymerase (PARP) was from EMD
Biosciences, p21 (OP64) and p53 (OP43) from Calbiochem, p27 (88264) from Abcam, and pIkBa from Cell
Signaling Technologies. Remaining antibodies were from
Santa Cruz Biotechnology. Bands were quantified on an
AlphaImager system (Proteinsimple).
Electrophoretic mobility shift assay
A probe containing an NF-kB consensus binding site
(50 AGTTGAGGGGACTTTCCCAGGC 30 ; Santa Cruz Biotechnology) was 32P-labeled using the Nick Translation
System (Invitrogen). Five micrograms of nuclear protein
was incubated 30 minutes at room temperature in binding buffer (10 mmol/L Tris-HCl, pH 7.5, 1.0 mmol/L
ethylenediaminetetraacetic acid (EDTA), 4% Ficoll, 1.0
mmol/L dithiothreitol, 75 mmol/L KCl) plus 250 ng poly
dI-dC (Sigma). Complexes were separated on polyacrylamide gels in Tris-borate EDTA buffer (89 mmol/L Trisbase, 89 mmol/L boric acid, 2.0 mmol/L EDTA), dried,
and autoradiographed. For supershift experiments, antibodies to p65, p50, or c-Rel (Santa Cruz Biotechnology)
were incubated with nuclear extract for 10 minutes before
addition of probe.
Real-time reverse transcription-PCR
RNA was extracted using TRIzol (Invitrogen). cDNA was
prepared with SuperScript First-Strand Synthesis System
Clin Cancer Res; 19(9) May 1, 2013
Downloaded from clincancerres.aacrjournals.org on June 18, 2017. © 2013 American Association for Cancer Research.
2407
Published OnlineFirst March 20, 2013; DOI: 10.1158/1078-0432.CCR-12-2754
Gupta et al.
p53DN system. Mutations in the p53 DNA-binding domain
at codon 273 (Arg!His, p53DN818) and codon 281
(Asp!Glu, p53DN843) were selected to generate 2 different
p53DN constructs using the QuikChange mutagenesis kit
(Stratagene). DNA was sequenced to confirm mutations.
Using primer sequences provided by Clontech, the presence
and orientation of p53DN818 and p53DN843 sequences in
the pRetro vector were similarly confirmed. Retrovirus particles were produced by cotransfecting plasmid DNA and
ecotropic helper plasmids (pVSV and pGPZ) into the 293 cell
line using calcium phosphate precipitation. The inducible
Tet activator 697 cell line (pRetrox-tet-on) was established
according to the manufacturer’s protocol (Clontech).
697pTet-on cells were infected with retrovirus by culturing
for 10 hours in conditioned media with 8 mg/mL polybrene.
Cells were then washed and incubated 48 hours before
selection with 1 mg/mL puromycin and 100 mg/mL G418.
(Invitrogen). Real-time reverse transcription PCR (RT-PCR)
was done on an ABI 7900 (Applied Biosystems) using
TaqMan Universal Master Mix, primers, and labeled probes
(Applied Biosystems) and TBP as an endogenous control.
Mean threshold cycle (Ct) values were calculated by PRISM
software (Applied Biosystems) to determine fold differences
according to manufacturer’s instructions.
CT-L activity assay
Whole cell extracts without protease inhibitors were
analyzed using a CT-L–specific 20S proteasome assay kit
(Chemicon) following the manufacturer’s instructions. The
assay is based on detection of the fluorophore 7-amino-4methylcoumarin (AMC) after cleavage from the labeled
substrate. Free AMC was measured using a 380/460 nm
fluorometer filter set. Relative activity was standardized by
protein concentration, determined by BCA assay.
Statistics
For all experiments, linear mixed effects models were
used to account for dependencies among the data. From
the models, estimated differences between experimental
conditions were calculated along with 95% confidence
intervals (CI). Log transformations of the data were applied
when necessary to stabilize variances. Data from real-time
RT-PCR experiments were first normalized to internal controls, and then linear mixed models were applied to the logtransformed fold changes. All analyses were done using
SAS/STAT software, v9.2 (SAS Institute).
Gamma irradiation
Cells were irradiated using 5 or 8 Gy as indicated. RNA
and lysates were collected from a subset of cells after 4
hours, and viability was determined in remaining cells at
24 hours by annexin V/propidium iodide or LIVE/DEAD
staining.
Retroviral vectors and generation of cell line with
inducible dominant-negative (DN) p53 activity
The retroviral construct pRetroX-Tight-Puro (pRetro;
Clontech) was used to stably transfect 697 cells with the
B
*
**
6
2.5 10
6
2.0 10
6
1.5 10
6
1.0 10
5
5.0 10
0
*
150
% Mitochondrial activity
(normalized to vehicle)
Relative fluorescence units (RFU)
A
100
50
0
CFZ
(nmol/L)
T cells
B cells
CLL cells
Figure 1. T cells express more chymotrypsin-like (CT-L) proteasome activity than B cells. A, basal CT-L activity in T cells (n ¼ 10) is higher than B cells (n ¼ 11)
from healthy donors and in primary CLL cells (n ¼ 10; , P < 0.0001 for both). CT-L activity in B cells was also higher than in CLL cells ( , P ¼ 0.0003). Results are
displayed in relative fluorescence units (RFU). B, B and T cells from healthy donors (n ¼ 8 each) and CLL patient samples (n ¼ 10) were incubated in serum-free
AIM-V media with various concentrations of CFZ for 1 hour. Mitochondrial activity as a surrogate for cell viability was determined by MTS assay at 48 hours,
and is shown relative to time-matched untreated controls. Horizontal lines represent the mean. CFZ across all doses is more cytotoxic to CLL cells than T cells
( , P < 0.0001). CFZ is also more cytotoxic to CLL cells than to B cells at 33 and 100 nmol/L (P < 0.0001 for both) but is not different at 300 nmol/L.
2408
Clin Cancer Res; 19(9) May 1, 2013
Clinical Cancer Research
Downloaded from clincancerres.aacrjournals.org on June 18, 2017. © 2013 American Association for Cancer Research.
Published OnlineFirst March 20, 2013; DOI: 10.1158/1078-0432.CCR-12-2754
Carfilzomib in Chronic Lymphocytic Leukemia
centrations used were based on pharmacokinetics data from
a phase I clinical trial in CLL now underway at OSU. In this
trial, the maximum concentration achieved was 0.876 0.099 mg/mL (1.22 0.138 mmol/L) for the 45 mg/m (2)
dose (unpublished results). This is similar to concentrations
previously reported to be clinically achievable in patients
with hematologic malignancies (27). As shown in Fig. 1B,
normal B cells were significantly more sensitive to CFZ than
normal T cells at 48 hours using 100 and 300 nmol/L
concentrations (P < 0.0001 for both). This finding is consistent with the elevated CT-L activity observed in T cells
relative to B cells.
Results
T cells express increased CT-L proteasome activity and
decreased sensitivity to CFZ compared to B cells
A common disadvantage of current therapies for CLL is
the negative impact on normal lymphocytes, especially T
cells, which leaves patients at increased risk of opportunistic
infection. To address the impact of CT-L inhibition on CLL
and normal cell types, we first measured the basal levels of
CT-L activity in CLL cells compared to B and T cells from
healthy volunteers. Interestingly, normal T cells showed
significantly higher CT-L activity compared to normal B
cells and CLL cells (P < 0.0001 for both). CT-L activity in
normal B cells was also significantly higher than in CLL cells
(P ¼ 0.0003; Fig. 1A). We next tested the relative cytotoxicity
of CFZ in these cell types. Because CFZ exhibits a short in vivo
half-life (approximately 30 minutes; refs. 26 and 27), cells
in all experiments were incubated with CFZ for 1 hour,
washed, and further cultured in fresh media for the indicated times to more closely mimic in vivo exposure. Con-
A
100
100
100
50
50
50
0
0
0
CFZ (nmol/L)
BTZ (nmol/L)
CLL cells
T cells
200
Veh
CFZ 100
CFZ 300
150
CFZ (nmol/L)
C
Veh
CFZ 100
100
50
0
CFZ
(nmol/L)
0
100 300
1 hour
0
100 300
4 hours
0
100 300
12 hours
% Annexin and propidium iodide negative
(normalized to vehicle at 1 hour)
% AnnexinV-/propidium iodide(normalized to vehicle)
RPMI with 10% FBS
RPMI with 10% HS
AIM-V media
B
% CT-L activity inhibition
(normalized to vehicle at 1 hour)
CFZ irreversibly inhibits the CT-L subunit and
promotes apoptosis in CLL cells
Because of the lower CT-L activity in CLL cells, we
hypothesized that the CT-L–specific inhibitor CFZ would
show particular efficacy in these cells. We therefore tested
the cytotoxicity of CFZ in CLL patient samples. As BTZ was
BTZ (nmol/L)
CFZ (nmol/L)
CLL cells
150
Veh
CFZ 100
CFZ 300
BTZ (nmol/L)
T cells
Veh
CFZ 100
CFZ 300
100
50
0
- 100 300 - 100 300 - 100 300
CFZ
(nmol/L)
4 hours
12 hours
1 hour
-
100
300
24 hours
Figure 2. Carfilzomib irreversibly inhibits the CT-L subunit and promotes selective apoptosis in CLL cells cultured in human serum. A, CLL cells (n ¼ 8) were
treated with various concentrations of CFZ or BTZ for 1 hour in media as indicated. Viability was determined by annexin V/propidium iodide flow cytometry at
24 hours. Horizontal lines represent the mean. B, CLL cells or normal T cells (n ¼ 6 each) were incubated with CFZ in RPMI with 10% HS for 1 hour
and cells were harvested at several time points to determine CT-L activity. C, viability of cells from B was determined by annexin/propidium iodide flow
cytometry. Because of cell number limitations, T cells (striped bars) were investigated for viability at 24 hours only.
www.aacrjournals.org
Clin Cancer Res; 19(9) May 1, 2013
Downloaded from clincancerres.aacrjournals.org on June 18, 2017. © 2013 American Association for Cancer Research.
2409
Published OnlineFirst March 20, 2013; DOI: 10.1158/1078-0432.CCR-12-2754
Gupta et al.
of the lack of external signals from the microenvironment
provided by soluble factors and contact with a variety of
cells in the bone marrow and lymph nodes (28). We
therefore sought to determine the impact of microenvironment factors on CFZ-mediated cytotoxicity using a
previously established stromal cell line, HS-5-GFP (29).
CFZ did not substantially affect viability of HS-5-GFP
stromal cells at 48 hours at concentrations up to 300
nmol/L (data not shown). To test the protective effects of
stroma, CLL cells were incubated with CFZ or vehicle for 1
hour and transferred to plates with or without HS-5-GFP
cells. Using flow cytometry, GFP-negative CLL cells were
distinguished from GFP-positive HS-5 cells and cytotoxicity was assessed by propidium iodide uptake. CFZ
caused significantly more cytotoxicity in CLL cells alone
as compared to CLL cells on stroma, although CLL cells
on stroma remained sensitive to CFZ at the tested concentrations (P < 0.0001; Fig. 3A). The importance of the
PI3K signaling pathway to the survival of mature B cells,
and specifically co-culture of CLL on stromal cells, has
been established (29). Therefore, we sought to determine
if the stromal cell protection of CLL cells was mediated by
PI3K signaling. As shown in Fig. 3B, the addition of the
pan-PI3K inhibitor LY294002 significantly diminished
the stromal cell–mediated protection of CFZ-treated CLL
cells (P ¼ 0.0083). These data indicate that CFZ can
mediate cytotoxicity in CLL cells despite the protective
effect of stromal cells, and that stromal cell protection
from CFZ-mediated cytotoxicity can be diminished by
PI3K inhibition.
previously shown to be inactivated by human plasma
components (17), we also included this agent and tested
in the presence of 3 different media: serum-free AIM-V
medium, RPMI 1640 with 10% HS, and RPMI 1640 with
10% FBS. In serum-free or FBS-containing media, CFZ was
more cytotoxic to CLL cells than BTZ (approximate LC50 in
AIM-V and FBS: CFZ ¼ 40 and 80 nmol/L; BTZ ¼ 200 and
170 nmol/L, respectively). In media containing HS, however, CFZ maintained efficacy but BTZ was largely inactive
(approximate LC50 of CFZ in HS ¼ 80 nmol/L; Fig. 2A). To
more closely approximate in vivo conditions, media with HS
was used for the remaining work. Next, CT-L activity was
assessed in CLL patient cells and in normal T cells after total
incubations of 1, 4, or 12 hours (Fig. 2B). Viability was also
determined by propidium iodide flow cytometry (Fig. 2C).
CFZ rapidly (within 1 hour) suppressed the CT-L proteasome activity both in CLL samples and in T cells. This was
followed by a time-dependent increase in apoptosis (as
determined by annexin positivity) and cytotoxicity (as
determined by propidium iodide uptake) that was greater
in CLL versus T cells. Despite drug washout, CT-L inhibition
was not reversible within the 12-hour period following CFZ
treatment. Proteasome activity was not determined at later
time points because of the high degree of CFZ-induced cell
death.
Stromal protection of CFZ-treated CLL cells is
diminished by a pan-PI3K inhibitor
CLL cells have dysregulated apoptosis in vivo, but in
vitro they rapidly undergo spontaneous apoptosis because
B
*
*
33 100 300 +
CLL suspension
0
33 100 300
+
0
CFZ
0
CFZ 0
(nmol/L)
20
LY
20
40
Vehicle
40
60
CFZ + LY
60
80
CFZ
80
100
LY
100
120
Vehicle
120
% Propidium iodide negative
(normalized to vehicle in suspension)
% Propidium iodide negative
(normalized to vehicle in suspension)
NS
CFZ + LY
A
CLL cells on stroma
CLL suspension
CLL cells on stroma
Figure 3. Stromal protection of CLL cells treated with CFZ is reversed by pan-PI3K inhibitor. A, CLL cells (n ¼ 5) were incubated with various concentrations of
CFZ for 1 hour, washed, and transferred to plates with or without HS-5-GFP stromal cells. Dinaciclib (indicated by þ) was included as a positive
control (52). Viability of GFP negative (CLL) cells was determined by propidium iodide uptake at 48 hours and the differences were significant (P < 0.0001).
B, addition of 25 mmol/L pan propidium iodide-3K inhibitor LY294002 to 100 nmol/L CFZ treated CLL cells (n ¼ 5) decreased stromal protection as determined
by propidium iodide uptake at 48 hours ( , P ¼ 0.0083). Horizontal lines represent the mean.
2410
Clin Cancer Res; 19(9) May 1, 2013
Clinical Cancer Research
Downloaded from clincancerres.aacrjournals.org on June 18, 2017. © 2013 American Association for Cancer Research.
Published OnlineFirst March 20, 2013; DOI: 10.1158/1078-0432.CCR-12-2754
B
Patient 3
Hours
CFZ
1.5
–
+
4
–
+
8
–
18
+
–
+
UV Jurkat
A
PARP
Actin
% AnnexinV-/propidium iodide-
Carfilzomib in Chronic Lymphocytic Leukemia
100
80
Patient 1
Patient 2
Patient 3
60
40
20
0
C
D
F-ara + BocD
F-ara A
BocD
CFZ + BoCD
CFZ
Untreated
Patient 6
PARP
% AnnexinV-/propidium iodide(normalized to vehicle)
4 hours
200
150
8 hours
18 hours
*
Patient 4
Patient 5
Patient 6
100
50
0
Actin
Figure 4. CFZ-mediated cytotoxicity in CLL cells is caspase-dependent. A, CLL samples (n ¼ 3) were treated with CFZ 350 nmol/L for 1 hour. Lysates were
collected at 1.5, 4, 8, and 18 hours, and cleavage of PARP was assessed by immunoblot. UV Jurkat was included as positive control to assess PARP cleavage
products. A representative immunoblot is shown. B, viability of cells from A was determined by annexin V/propidium iodide at 4, 8 and 18 hours. C, CLL cells
(n ¼ 3) were incubated with or without CFZ and pan-caspase inhibitor 100 mmol/L Boc-D-fmk (BocD) for 18 hours. PARP cleavage was assessed by
immunoblot. A representative immunoblot is shown. D, viability of cells from C was determined by annexin/propidium iodide at 24 hours. Differences in
cytotoxicity of CFZ with the addition of BocD were significant (P ¼ 0.0003). F-ara A (active metabolite of fludarabine) was included as a positive control for
PARP cleavage and caspase-dependent cytotoxicity.
Cytotoxicity induced by CFZ in CLL cells is caspase
dependent
CFZ has been reported to mediate caspase-dependent
apoptosis in mature B-cell malignancies such as Waldenstrom’s macroglobulinemia (30) and multiple myeloma
(31). We sought to determine if similar death pathways
were used in less-differentiated malignant B cells. CFZ
caused a time-dependent cleavage (Fig. 4A) of the caspase
substrate PARP and cytotoxicity (Fig. 4B) that was rescued
by the pan-caspase inhibitor Boc-D-fmk (Fig. 4C). The
cytotoxic effect of CFZ on CLL cells was also diminished
with the addition of Boc-D-fmk (Fig. 4D; P ¼ 0.0003). These
data indicate that the primary mechanism of the cytotoxicity by CFZ in primary CLL cells is caspase dependent.
CFZ is cytotoxic to cells independent of p53
Mutations or deletions of p53 are far more common in
advanced cases of CLL and are associated with resistance to
alkylating agents, disease progression in 1 or 2 years, and
www.aacrjournals.org
overall significantly poorer outcome (32). The proteasome
regulates the turnover of p53 as well as several key p53
targets. We therefore sought to determine the effects of
proteasome inhibition on p53 and its targets. CLL cells
incubated with CFZ were analyzed by immunoblot or realtime RT-PCR. Gamma irradiation was included as a positive
control representing a classical p53-inducing treatment. As
shown in Fig. 5A, CFZ treatment caused only modest accumulation of p53 and p21 proteins compared to irradiated
cells, and p27 was unchanged. Puma and Bax proteins were
similarly unchanged (data not shown). Significant increases
in p21 transcription were noted in all patient samples
following CFZ treatment (average fold change of 5.78;
P < 0.0001), and Puma transcription was slightly increased
(average fold change of 1.97; P ¼ 0.0323; Fig. 5B). However,
p53 was found to be downregulated by an average of 0.75
fold (P ¼ 0.0193), and no significant changes were observed
in the transcription of Noxa (P ¼ 0.2232) or Bax (P ¼
0.1531). These findings show that CFZ does not induce a
Clin Cancer Res; 19(9) May 1, 2013
Downloaded from clincancerres.aacrjournals.org on June 18, 2017. © 2013 American Association for Cancer Research.
2411
Gupta et al.
Pt 2
Veh CFZ
B
γ
Fold change (relative to untreated)
p53
p21
p27
Pt 4
–
+
Jurkat
Pt 3
+
300 nmol/L CFZ: –
UV
Jurkat
Actin
Noxa
del(17p)
100
80
60
40
20
0
/L
ol
m
***
m
/L
ol
/L
ol
m
*
**
10
1
0.1
iodide-
C
% AnnexinV-/propidium iodide-
Actin
100
Non-del(17p)
100
80
%
γ
Veh CFZ
AnnexinV-/propidium
Pt 1
A
0
**
*
60
40
20
m
/L
/L
/L
ol
ol
m
ol
m
Figure 5. Carfilzomib is cytotoxic to CLL cells independent of p53. CLL cells were treated with CFZ (300 nmol/L) for 1 hour and whole cell lysates were collected
at 8 hours. g-Irradiation (5 Gy) was included as a positive control for p53-dependent treatment. Untreated and UV-irradiated Jurkat cells were included as
negative and positive controls for expression of Noxa protein. Changes in p53 and its targets were assessed by (A) immunoblot (n ¼ 6) and (B) real-time RTPCR (n ¼ 6-14). CFZ treatment induced p21 ( , P < 0.0001) and PUMA ( , P ¼ 0.0323) and significantly downregulated p53 by 0.75-fold ( , P ¼ 0.0193)
compared to vehicle. Horizontal lines represent the geometric mean. C, CLL samples from patients with or without del(17p13.1) (n ¼ 9) were incubated
with 100 and 300 nmol/L of CFZ for 1 hour. No significant differences were observed (P ¼ 0.3924). A total of 5 Gy g-irradiation and fludarabine
(Fara A) were included as controls for p53-dependent treatments and were more cytotoxic in non-del(17p13.1) patients than del(17p13.1) patients
( , P < 0.0001; , P ¼ 0.0002). Viability was determined by annexin V/propidium iodide flow cytometry at 24 hours. Horizontal lines represent the mean.
typical p53 response in CLL cells, and suggests that CFZmediated p21 induction may be because of an indirect
mechanism rather than direct p53 activity. As was shown
previously with BTZ (33), we observed an increase in Noxa
protein (Fig. 5A), but not mRNA, with CFZ treatment.
Next, we analyzed CLL samples from patients with or
without del(17p13.1), the chromosomal site of the p53
gene (Fig. 5C). CFZ was similarly cytotoxic to patient cells
irrespective of del(17p13.1) status (P ¼ 0.3924). In contrast, significant differences in cytotoxicity were observed
between patient samples with and without this deletion for
both g-irradiation and fludarabine treatments (P < 0.0001
and P ¼ 0.0002, respectively). As p53 is only one of the genes
deleted on del(17p13.1), we next sought to test the cyto-
2412
Clin Cancer Res; 19(9) May 1, 2013
toxicity of CFZ in cells with wild-type versus defective p53
activity. In CLL, a transition mutation in the p53 DNA
binding domain at codon 273 that converts Arg to His has
been described as a gain-of-function dominant negative
mutation (34). In addition, mutations at codons 281
(Asp!Glu) and 273 (Arg!His) are found in certain leukemias (35). Doxycycline-inducible p53DN constructs
representing these mutations were transfected into 697
cells, an acute B-lymphoblastic leukemia cell line that has
wild-type p53 and responds to CFZ treatment similarly to
CLL cells (Supplementary Fig. S1B). As shown in Fig. 6A,
p53 protein was induced in cell lines carrying the p53DN
constructs, but not the empty vector, as early as 18 hours
after doxycycline addition. Induction of p53DN activity
Clinical Cancer Research
Carfilzomib in Chronic Lymphocytic Leukemia
increased p53 transcription by more than 5-fold in both
p53DN818 and p53DN843 cells but not in empty-vector
transfected cells (P < 0.0001; Fig. 6B). Addition of doxycycline also caused a significant reduction in cytotoxicity (Fig.
6C) with g-irradiation, but not with CFZ treatment, in the
cells with p53DN activity (P ¼ 0.0378 and 0.0243 for
p53DN818 and p53DN843, respectively). Furthermore, in
g-irradiated cells, although doxycyline caused a 1.5-fold
upregulation of p21 transcription in empty-vector control
cells, p21 was reduced (0.7-fold) in doxycycline-treated
p53DN cells (Fig. 6B). Although these changes did not reach
statistical significance, they further support the functional
activity of the p53DN constructs in the 697 cells. Cumulatively, these experiments indicate that cytotoxicity of CFZ is
not dependent on functional p53 activity in leukemic cells.
CFZ induces an atypical NF-kB response in CLL cells
CLL cells are characterized by constitutive activation of
NF-kB as indicated by high levels of nuclear NF-kB, p50,
and p65 (RelA) proteins. In cutaneous T-cell lymphoma,
BTZ was reported to inhibit the NF-kB pathway by preventing the degradation of the phosphorylated form of IkBa,
thus preventing the nuclear translocation of p50 and p65
and further activation of NF-kB targets (36). More recently it
was shown in multiple myeloma cell lines that IkBa can be
degraded by proteasome-independent mechanisms following BTZ treatment to allow for translocation and activation
of NF-kB subunits (37). Thus, NF-kB response to proteasome inhibition may vary by cell type or other factors not yet
identified. To determine the effects of CFZ on NF-kB activity
in CLL, CLL cells were incubated with CFZ for 1 hour and
examined by immunoblot. As shown in Fig. 7A–C, IkBa
protein decreases in both whole cell extracts as well as
nuclear and cytoplasmic fractions. However, phosphorylation of IkBa (p-IkBa) increased in all fractions. To determine if this decrease in IkBa was because of caspase activity
as was previously reported (38), CLL cells were incubated
with CFZ in the presence or absence of Boc-D-fmk (Fig. 7B).
Although Boc-D-fmk rescued cells from CFZ-mediated
PARP cleavage and apoptosis, it did not prevent degradation
of IkBa. Antibodies recognizing epitopes in both the N- and
C-terminal region of IkBa produced identical results (data
not shown). It also was reported that IkBa is a substrate of
calpains (37) but similar to caspase inhibitors, calpain
inhibitors did not prevent degradation of IkBa (data not
shown). As proteasome inhibitors were previously shown
to induce or inhibit both the canonical and noncanonical
NF-kB pathways (36, 37), nuclear and cytoplasmic fractions
of CFZ-treated CLL cells were probed for canonical (Fig. 7C)
and noncanonical (Fig. 7D) proteins. No significant
changes were observed in the nuclear or cytoplasmic levels
of NF-kB subunits p50, p65, p52, or p100 in the CFZ treated
samples. As controls, CpG and CD40L increased expression
of both canonical and noncanonical proteins in the nuclear
fraction.
To determine whether nuclear translocation of pIkBa
affects NF-kB activity, NF-kB DNA binding was measured in
nuclear extracts prepared from CLL cells treated with CFZ or
www.aacrjournals.org
CD40L. The constitutive NF-kB (p50/p65) DNA binding
activity in CLL cells was increased with CFZ treatment in
approximately 50% of the samples, whereas there was no
change in the remainder (Fig. 7E and F, respectively).
Importantly, CFZ was similarly cytotoxic to all CLL samples
tested (Fig. 7G). These results indicate that the impact of
CFZ on NF-kB is not relevant to its cytotoxic activity in CLL.
Finally, to determine whether the increased NF-kB activity
observed by EMSA resulted in changes in gene expression,
we analyzed several known NF-kB targets by real-time RTPCR (Fig. 7H). CFZ induced a subset of these targets,
including IkBa, IL-6, c-FLIP, and CXCL13, but with substantial interpatient variability (average fold changes for
target genes are indicated in Supplementary Table S1).
Notably, CFZ did not induce the transcription of several
classical NF-kB targets such as Bcl2A1, XIAP, Mcl-1, and
p53. Thus, our results show that CFZ induces rather than
inhibits NF-kB in CLL patient cells, but that this response is
atypical and does not contribute to the CFZ induced cytotoxicity or resistance to this therapy.
Discussion
CFZ represents a new class of irreversible proteasome
inhibitors that specifically target the CT-L subunit. It is
currently in clinical trials for B-cell malignancies, but its
mechanism of cell death is poorly understood. Here, we
show that CLL cells have a low level of CT-L proteasome
activity relative to normal B- and T-lymphocytes. We
hypothesized that an irreversible inhibitor of the CT-L
subunit would be most effective in cells having the least
baseline activity, and used this to justify the examination of
CFZ in CLL. In fact, we show that a short (1 hour) exposure
of 100 nmol/L CFZ is more cytotoxic to CLL cells compared
to normal lymphocytes, even though this concentration
similarly inhibits the CT-L subunit in CLL and normal T
cells. This finding indicates that the differential sensitivity of
CLL versus T cells to CFZ is not directly related to the extent
of CT-L inhibition. Unlike BTZ, the cytotoxicity of CFZ is
not diminished in media with HS. Importantly, our studies
indicate that CFZ induces cytotoxicity irrespective of del
(17p13.1) status or p53 function. Stromal cells partially
protected CLL cells from CFZ-mediated cytotoxicity,
although this protective effect was decreased by the addition
of a pan-PI3K inhibitor. Finally, we show that CFZ mediates
an atypical NF-kB activation that does not associate with
transcriptional induction of classical NF-kB targets or with
CLL cell death. As previously reported with other proteasome inhibitors (33), Noxa induction is observed with CFZ
treatment concurrent with the onset of apoptosis, suggesting that Noxa-mediated caspase induction is the mechanism of action of CFZ in CLL.
To our knowledge, this is the first study showing the
reduced CT-L activity in primary CLL cells versus normal
B and T cells. Although activity of proteasome inhibitors
is well documented, the mechanism by which these
agents preferentially target malignant cells is yet to be
elucidated. We speculate that differences in CT-L activity
explain the observed therapeutic window between CLL
Clin Cancer Res; 19(9) May 1, 2013
2413
Gupta et al.
Doxycycline
Vector
DN818
DN843
–
–
–
+
+
+
20
Fold change (relative to untreated)
B
+ Adr
A
p53
GAPDH
15
10
5
0
Doxycycline
–
–
15
10
5
0
Doxycycline
+
–
–
+
% Live
(normalized to Veh without Dox)
Gamma 8 Gy
–
–
C
% Live
(normalized to Veh without Dox)
–
+
+
+
p53
p21
Mcl1
10
5
0
–
–
+
–
100
80
60
40
20
0
*
120
100
80
60
40
20
/L /L l/L l/L
o
ol ol
o
m m
m m
Clin Cancer Res; 19(9) May 1, 2013
+
+
Vector
120
p53DN818
2414
–
+
p53DN843
15
+
+
/L
ol
m
0
+
–
20
Fold change (relative to untreated)
p53DN818
p53
p21
Mcl1
/L
/L
ol
m
ol
m
% Live
(normalized to Veh without Dox)
Fold change (relative to untreated)
Gamma 8 Gy
20
Vector
p53
p21
Mcl1
/L
ol
m
p53DN843
**
120
100
80
60
40
20
0
/L
ol
m
/L
/L
ol
m
ol
m
/L
ol
m
Clinical Cancer Research
Carfilzomib in Chronic Lymphocytic Leukemia
tumor cells and normal lymphocytes. This hypothesis
could be further tested by comparing the levels of inhibition of the CT-L activities in the constitutive (c20S) and
immunoproteasome (i20S). We observed that levels of
CT-L activity inhibition did not necessarily correlate with
CFZ cytotoxicity. One potential explanation is differing
levels of inhibition of the CT-L activity in c20S and i20S
by CFZ, although this remains to be tested.
The tumor suppressor protein p53 induces apoptosis
under conditions of cellular stress or DNA damage.
Although p53 is functional in most CLL patients at the time
of diagnosis, the gene becomes mutated or deleted in at least
one allele in approximately 40% of patients with advanced
CLL (24, 39), corresponding with significantly poorer outcome (32). Thus, agents that work independently of p53 are
of significant interest in the development of therapies for
drug-resistant CLL. Proteasome inhibitors have been shown
to induce apoptosis in both p53-independent (40) and p53dependent manners (41). Here, we show that CFZ induces
cytotoxicity irrespective of del(17p13.1) status or the
expression of a dominant negative p53 protein. Given the
poor outcome and lack of effective therapies for del
(17p13.1) CLL patients, these findings suggest CFZ therapy
may be effective in this disease subtype.
Recent advances in CLL biology show the importance of
both internal and external (microenvironment) factors in
the survival and drug resistance of CLL tumor cells. The PI3K
and NF-kB pathways are key regulators of differentiation
and survival in B cells and are induced by microenvironmental factors (42, 43). CLL cells cultured with stromal cells
were less sensitive to CFZ compared to CLL cells in suspension. This protective effect has been linked to the PI3K
pathway (29, 44), and our data show that PI3K inhibition
indeed reverses this protection. Further work is necessary to
determine the specific PI3K isoform responsible for this
effect, but our data suggest that combination therapy of CFZ
with PI3K inhibitors may provide added benefit.
Although the mechanism of proteasome inhibitor–mediated cell death is unclear, these agents have commonly been
reported to be inhibitors of the NF-kB pathway (36, 37).
This notion is attributed to the observation that the NF-kB
inhibitory molecule IkBa is a proteasome substrate, and
that proteasome inhibitors should prevent its degradation.
More recent studies indicate that proteasome inhibitors can
either inhibit or induce NF-kB, again through interaction
with IkBa (36, 37). However, these reports are conflicting
and effects may vary by cell type. Our data in CLL show that
the NF-kB pathway is in fact activated by CFZ as evidenced
by degradation of total IkBa, accumulation of phosphor-
ylated IkBa in nuclear and cytoplasmic fractions, and
enhanced binding of NF-kB subunits to NF-kB consensus
sites. However, NF-kB activation does not correlate with
CLL cell death. This result is similar to what was reported in
multiple myeloma cell lines using BTZ (37), although our
studies in CLL primary cells extend these observations to
show that a classical NF-kB gene induction pattern is not
observed. Therefore, although the inhibition of proteasome
activity is not inhibitory to the NF-kB pathway per se, it
induces an atypical response that does not result in the
ultimate transcription of canonical NF-kB targets and thus is
unlikely to provide a prosurvival effect.
To characterize the mechanism of CFZ-mediated cell
death, we assessed the effects of CFZ on the Bcl-2 family
of proteins and subsequent caspase activation. We found
that inhibition of caspase activity caused a significant reduction in cytotoxicity of CFZ. In addition, of the Bcl-2 family of
proteins, only Noxa was consistently upregulated in all CFZtreated samples. This effect was recently shown to induce
caspase activity in CLL cells, further supporting a caspasedependent pattern of cytotoxicity (33). Proteasome inhibitors can also cause cytotoxicity because of endoplasmic
reticulum stress by blocking the degradation of regulatory
and misfolded proteins (33, 45). However in our studies
CFZ did not cause changes in conventional endoplasmic
reticulum stress markers, including splicing of XBP1 or
mRNA or protein levels of CHOP, GRP78, or pEIF2a (data
not shown). Thus, we concluded that in primary CLL cells,
CFZ-mediated cytotoxicity was caused by a caspase-dependent pathway and not via endoplasmic reticulum stress.
To date, BTZ is the only proteasome inhibitor approved
for treatment of malignant blood disorders and is now
being used in front-line therapy for multiple myeloma
(46). In spite of a high degree of cytotoxicity in vitro, BTZ
failed to produce any objective responses in CLL patients
in a phase II clinical trial (16). Its failure in vivo was
hypothesized to be because of the inactivation of the
boronate moiety in BTZ by flavonoids in HS (17), which
is consistent with our observation that BTZ lacks activity
against CLL cells in media containing HS. Despite the
initial clinical success of BTZ in hematological disorders,
significant populations of patients remain refractory to
treatment. Furthermore, toxicities with BTZ treatment
such as peripheral neuropathy and thrombocytopenia
have increased the intervals between dosing, allowing
recovery of the proteasome function (47, 48). This has
spurred development of a new class of proteasome inhibitors that lack these toxicities. CFZ is being investigated
through clinical trials in newly diagnosed, relapsed, or
Figure 6. Carfilzomib is similarly cytotoxic in cells expressing dominant negative p53 mutants. The 697 acute B-lymphoblast leukemia cell line was transfected
with an empty vector or with either of 2 different inducible p53 dominant negative constructs. After 18 hours induction with doxycycline (n ¼ 3): A, cells were
harvested for lysates to determine expression of p53 protein. Adriamycin (Adr) treated 697 cells were included as a positive control for p53 protein
expression. B, cells were g-irradiated (8 Gy) and RNA was collected at 4 hours. Changes in p53 and p21 were detected by real-time RT-PCR. Mcl-1 is not a p53target and was included as a negative control. Doxycycline induced p53 (P < 0.0001) in p53DN cells. C, cells were g-irradiated (8 Gy) or treated with CFZ for 1
hour. At 24 hours, viability was determined by LIVE/DEAD staining. In p53DN cells the difference in cytotoxicity in doxycycline-induced and uninduced cells
following g treatment was significantly different than with CFZ treatment ( , P ¼ 0.0378 and , P ¼ 0.0243 for p53DN818 and p53DN843, respectively).
Horizontal lines represent the mean.
www.aacrjournals.org
Clin Cancer Res; 19(9) May 1, 2013
2415
HeLa TNF
Gupta et al.
Pt 1
300 nmol/L CFZ:
−
+
Pt 2
−
Pt 3
+
−
+
HeLa
A
300 nmol/L CFZ
100 µmol/L BocD
pIκBα
PARP
IκBα
IκBα
Actin
GAPDH
NE
C
CE
Pt 5
Pt 4
B
−
−
+
−
−
+
+
+
D
+
−
−
+
+
+
CE
NE
Veh CFZ CD40L CpG Veh CFZ CD40L CpG
−
−
Veh CFZ CD40L CpG Veh CFZ CD40L CpG
p50
p100
p65
p52
pIκBα
RelB
IκBα
Brg1
Brg1
Tubulin
Tubulin
E
4
350
hours nmol/L
4 hours
0
70
F
+
–
4 hours
0
350 p65 p50 c-Rel CD40L BMS
p50/p65
p50/p65
p50/p50
p50/p50
70
4
350
hours nmol/L
–
+
350 p65 p50 c-Rel BMS CD40L
NF-κB unchanged
NF-κB induced
G
% Mitochondrial activity
(normalized to vehicle)
150
100
50
0
CFZ
(nmol/L)
0
300
NF-κB induced
0
300
NF-κB unchanged
Figure 7. CFZ induces a defective NF-kB response. A, CLL cells (n ¼ 6) were incubated 1 hour without or with 300 nmol/L CFZ, and lysates were collected at 8
hours. IkBa and phosphorylated IkBa (pIkBa) were assessed by immunoblot. Untreated and TNF-treated HeLa cells were included as negative and positive
controls for pIkBa protein expression, respectively. B, CLL cells (n ¼ 6) were treated as in A, with or without 100 mmol/L Boc-D-fmk. Lysates were isolated
at 18 hours and analyzed by immunoblot. CLL cells (n ¼ 6) were incubated with 0, 70, and 350 nmol/L CFZ for 1 hour or 500 ng/mL CD40L for 4 hours.
Nuclear (NE) and cytosolic (CE) extracts were collected at 4 hours and analyzed by immunoblot for NF-kB (C) canonical pathway and (D) noncanonical
pathway proteins. EMSAs were run using NE from cells treated with 350 nmol/L CFZ (1 hour treatment, 4 hours incubation) or 500 ng/mL CD40L
(4 hours; n ¼ 14), using a probe containing a consensus NF-kB binding site. Representative EMSAs from CFZ-treated NE that displayed (E) increased NF-kB
binding and (F) no changes are shown. CD40L and BMS-345541 (IKK inhibitor) treated NE were included as positive and negative controls for p50/p65 binding,
respectively. G, viability of all samples analyzed by EMSA was determined by MTS at 48 hours.
2416
Clin Cancer Res; 19(9) May 1, 2013
Clinical Cancer Research
Carfilzomib in Chronic Lymphocytic Leukemia
12
–Δ (Δ Ct)= log fold change
over vehicle
H
CFZ
9
**
*
6
** **
3
0
Iκ
B
α
–3
12
CD40L
–Δ (Δ Ct)= log fold change
over vehicle
–Δ (Δ Ct)= log fold change
over vehicle
12
9
6
3
0
–3
CpG
9
6
3
0
Iκ
B
Iκ
B
α
α
–3
Figure 7. (Continued ) H, CLL cells (n ¼ 12) were treated with 300 nmol/L CFZ (1 hour), 1.7 mmol/L CpG (4 hours), and 500 ng/mL CD40L (4 hours), and RNA was
collected at 8 hours from CFZ-treated samples. mRNA expression was analyzed using real-time RT-PCR. CFZ treatment significantly induced expression of
IkBa ( ) by 1.5-fold, and IL-6, c-FLIP, and CXCL-13 ( all 3) by greater than 2-fold. Horizontal lines represent the mean.
refractory multiple myeloma, and has seen promising
activity both as a single agent and in combination with
immunomodulators (49–51).
In summary, our study indicates that CFZ represents a
promising, p53-independent therapeutic for CLL and provides the rationale for development of CFZ in this disease.
Based on these data, we have initiated a phase I dose
escalation study of this agent in relapsed and refractory CLL.
Disclosure of Potential Conflicts of Interest
Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): S.V. Gupta, Y. Lu, E.J. Sass, J.C. Byrd,
D.M. Lucas
Study supervision: D.M. Lucas
Acknowledgments
CFZ was provided for these studies by Onyx Pharmaceuticals, South San
Francisco, CA. The HS-5-GFP stromal cell line was kindly provided by Dr.
Beverly Torok-Storb, Fred Hutchinson Cancer Research Center, Seattle, WA.
The authors thank the members of our laboratory for helpful comments and
the many patients who donated blood for our studies.
No potential conflicts of interest were disclosed.
Authors' Contributions
Conception and design: S.V. Gupta, Y. Lu, D. Jarjoura, J.C. Byrd, D.M. Lucas
Development of methodology: S.V. Gupta, Y. Lu, R. Lapalombella, J.C.
Byrd, D.M. Lucas
Acquisition of data (provided animals, acquired and managed patients,
provided facilities, etc.): S.V. Gupta, Y. Lu, E.J. Sass, T.L. Chen, M. Davis, J.
A. Woyach, J.C. Byrd
Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): S.V. Gupta, Y. Lu, A. Lehman, D. Jarjoura,
J.C. Byrd, D.M. Lucas
Writing, review, and/or revision of the manuscript: S.V. Gupta, Y. Lu, R.
Lapalombella, M. Davis, J.A. Woyach, A. Lehman, J.C. Byrd, D.M. Lucas
www.aacrjournals.org
Grant Support
This work was supported by The Leukemia & Lymphoma Society and the
National Cancer Institute (P50 CA140158, 1K12 CA133250). Mr. and Mrs.
Michael Thomas, The Harry Mangurian Foundation, and The D. Warren
Brown Foundation also provided support.
The costs of publication of this article were defrayed in part by the
payment of page charges. This article must therefore be hereby marked
advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate
this fact.
Received August 22, 2012; revised February 26, 2013; accepted March 5,
2013; published OnlineFirst March 20, 2013.
Clin Cancer Res; 19(9) May 1, 2013
2417
Gupta et al.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
2418
Peters JM, Franke WW, Kleinschmidt JA. Distinct 19-S and 20-S
subcomplexes of the 26-S proteasome and their distribution in the
nucleus and the cytoplasm. J Biol Chem 1994;269:7709–18.
Parlati F, Lee SJ, Aujay M, Suzuki E, Levitsky K, Lorens JB, et al.
Carfilzomib can induce tumor cell death through selective inhibition of
the chymotrypsin-like activity of the proteasome. Blood 2009;114:
3439–47.
Glickman MH, Ciechanover A. The ubiquitin-proteasome proteolytic
pathway: destruction for the sake of construction. Physiol Rev
2002;82:373–428.
Myung J, Kim KB, Crews CM. The ubiquitin-proteasome pathway and
proteasome inhibitors. Med Res Rev 2001;21:245–73.
Kisselev AF, Callard A, Goldberg AL. Importance of the different
proteolytic sites of the proteasome and the efficacy of inhibitors varies
with the protein substrate. J Biol Chem 2006;281:8582–90.
Yew PR. Ubiquitin-mediated proteolysis of vertebrate G1- and
S-phase regulators. J Cell Physiol 2001;187:1–10.
Blagosklonny MV, Wu GS, Omura S, El-Deiry WS. Proteasome-dependent regulation of p21WAF1/CIP1Expression. Biochem Biophys Res
Commun 1996;227:564–69.
Pagano M, Tam SW, Theodoras AM, Beer-Romero P, Del Sal G, Chau
V, et al. Role of the ubiquitin-proteasome pathway in regulating
abundance of the cyclin-dependent kinase inhibitor p27. Science
1995;269:682–85.
Batchelor E, Loewer A, Lahav G. The ups and downs of p53: understanding protein dynamics in single cells. Nat Rev Cancer. 2009;9:
371–77.
Rozan LM, El-Deiry WS. p53 downstream target genes and tumor
suppression: a classical view in evolution. Cell Death Differ 2007;
14:3–9.
Chen ZJ. Ubiquitin signalling in the NF-[kappa]B pathway. Nat Cell Biol
2005;7:758–65.
Chipuk JE, Moldoveanu T, Llambi F, Parsons MJ, Green DR. The BCL2 family reunion. Mol Cell 2010;37:299–10.
Hertlein E, Byrd JC. Signalling to drug resistance in CLL. Best Pract
Res Clin Haematol 2010;23:121–31.
Packham G, Stevenson FK. Bodyguards and assassins: Bcl-2 family
proteins and apoptosis control in chronic lymphocytic leukaemia.
Immunology 2005;114:441–49.
Bross PF, Kane R, Farrell AT, Abraham S, Benson K, Brower ME, et al.
Approval summary for bortezomib for injection in the treatment of
multiple myeloma. Clin Cancer Res 2004;10:3954–64.
Faderl S, Rai K, Gribben J, Byrd JC, Flinn IW, O'Brien S, et al. Phase II
study of single-agent bortezomib for the treatment of patients with
fludarabine-refractory B-cell chronic lymphocytic leukemia. Cancer
2006;107:916–24.
Liu FT, Agrawal SG, Movasaghi Z, Wyatt PB, Rehman IU, Gribben JG,
et al. Dietary flavonoids inhibit the anticancer effects of the proteasome
inhibitor bortezomib. Blood 2008;112:3835–46.
Groll M, Kim KB, Kairies N, Huber R, Crews CM. Crystal structure of
epoxomicin: 20S proteasome reveals a molecular basis for selectivity
of alpha-, beta-epoxyketone proteasome inhibitors. J Am Chem Soc
2000;122:1237–38.
Adams J, Behnke M, Chen SW, Cruickshank AA, Dick LR, Grenier L,
et al. Potent and selective inhibitors of the proteasome: dipeptidyl
boronic acids. Bioorganic Med Chem Lett 1998;8:333–38.
Demo SD, Buchholz TJ, Laidig GJ, Parlati F, Shenk KD, Smyth MS,
et al. Biochemical and cellular characterization of the novel proteasome inhibitor PR-171. Blood 2005;106:455a–55a.
Dorsey BD, Iqbal M, Chatterjee S, Menta E, Bernardini R, Bernareggi A,
et al. Discovery of a potent, selective, and orally active proteasome
inhibitor for the treatment of cancer. J Med Chem 2008;51:1068–72.
Liang H, Nishioka Y, Reich CF, Pisetsky DS, Lipsky PE. Activation of
human B cells by phosphorothioate oligodeoxynucleotides. J Clin
Invest 1996;98:1119–29.
Hallek M, Cheson BD, Catovsky D, Caligaris-Cappio F, Dighiero G,
Dohner H, et al. Guidelines for the diagnosis and treatment of chronic
lymphocytic leukemia: a report from the International Workshop on
Clin Cancer Res; 19(9) May 1, 2013
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
Chronic Lymphocytic Leukemia updating the National Cancer Institute-Working Group 1996 guidelines. Blood 2008;111:5446–56.
Lozanski G, Heerema NA, Flinn IW, Smith L, Harbison J, Webb J, et al.
Alemtuzumab is an effective therapy for chronic lymphocytic leukemia
with p53 mutations and deletions. Blood 2004;103:3278–81.
Graf L, Iwata M, Torok-Storb B. Gene expression profiling of the
functionally distinct human bone marrow stromal cell lines HS-5 and
HS-27a. Blood 2002;100:1509–11.
Demo SD, Kirk CJ, Aujay MA, Buchholz TJ, Dajee M, Ho MN, et al.
Antitumor activity of PR-171, a novel irreversible inhibitor of the
proteasome. Cancer Res 2007;67:6383–91.
O'Connor OA, Stewart AK, Vallone M, Molineaux CJ, Kunkel LA,
Gerecitano JF, et al. A phase 1 dose escalation study of the safety
and pharmacokinetics of the novel proteasome inhibitor carfilzomib
(PR-171) in patients with hematologic malignancies. Clin Cancer Res
2009;15:7085–91.
Meads MB, Hazlehurst LA, Dalton WS. The bone marrow microenvironment as a tumor sanctuary and contributor to drug resistance. Clin
Cancer Res 2008;14:2519–26.
Herman SE, Gordon AL, Wagner AJ, Heerema NA, Zhao W, Flynn JM,
et al. Phosphatidylinositol 3-kinase-delta inhibitor CAL-101 shows
promising preclinical activity in chronic lymphocytic leukemia by
antagonizing intrinsic and extrinsic cellular survival signals. Blood
2010;116:2078–88.
Sacco A, Aujay M, Morgan B, Azab AK, Maiso P, Liu Y, et al. Carfilzomib-dependent selective inhibition of the chymotrypsin-like activity of
the proteasome leads to antitumor activity in Waldenstrom's Macroglobulinemia. Clin Cancer Res 2011;17:1753–64.
Kuhn DJ, Chen Q, Voorhees PM, Strader JS, Shenk KD, Sun CM, et al.
Potent activity of carfilzomib, a novel, irreversible inhibitor of the
ubiquitin-proteasome pathway, against preclinical models of multiple
myeloma. Blood 2007;110:3281–3290.
Zenz T, Krober A, Scherer K, Habe S, Buhler A, Benner A, et al.
Monoallelic TP53 inactivation is associated with poor prognosis in
chronic lymphocytic leukemia: results from a detailed genetic characterization with long-term follow-up. Blood 2008;112:3322–29.
Baou M, Kohlhaas SL, Butterworth M, Vogler M, Dinsdale D, Walewska
R, et al. Role of NOXA and its ubiquitination in proteasome inhibitorinduced apoptosis in chronic lymphocytic leukemia cells. Haematologia 2010;95:1510–18.
Trbusek M, Smardova J, Malcikova J, Sebejova L, Dobes P, Svitakova
M, et al. Missense mutations located in structural p53 DNA-binding
motifs are associated with extremely poor survival in chronic lymphocytic leukemia. J Clin Oncol 2011;29:2703–08.
Sugawara W, Arai Y, Kasai F, Fujiwara Y, Haruta M, Hosaka R, et al.
Association of germline or somatic TP53 missense mutation with
oncogene amplification in tumors developed in patients with Li-fraumeni or Li-fraumeni-like syndrome. Genes Chromosomes Cancer
2011;50:535–45.
Juvekar A, Manna S, Ramaswami S, Chang TP, Vu HY, Ghosh CC,
et al. Bortezomib induces nuclear translocation of IkBa resulting in
gene-specific suppression of NF-kB–dependent transcription and
induction of apoptosis in CTCL. Mol Cancer Res 2011;9:183–94.
Hideshima T, Ikeda H, Chauhan D, Okawa Y, Raje N, Podar K, et al.
Bortezomib induces canonical nuclear factor-kB activation in multiple
myeloma cells. Blood 2009;114:1046–52.
Ravi R, Bedi A, Fuchs EJ. CD95 (Fas)-induced caspase-mediated
proteolysis of NF-kB. Cancer Res 1998;58:882–86.
Zenz T, Vollmer D, Trbusek M, Smardova J, Benner A, Soussi T, et al.
TP53 mutation profile in chronic lymphocytic leukemia: evidence for a
disease specific profile from a comprehensive analysis of 268 mutations. Leukemia 2010;24:2072–79.
Perez-Galan P, Roue G, Villamor N, Montserrat E, Campo E, Colomer
D. The proteasome inhibitor bortezomib induces apoptosis in mantlecell lymphoma through generation of ROS and Noxa activation independent of p53 status. Blood 2006;107:257–64.
Lopes UG, Erhardt P, Yao R, Cooper GM. p53-dependent induction of
apoptosis by proteasome inhibitors. J Biol Chem 1997;272:12893–96.
Clinical Cancer Research
Carfilzomib in Chronic Lymphocytic Leukemia
42. Cantrell DA. Phosphoinositide 3-kinase signalling pathways. J Cell Sci
2001;114:1439–45.
43. Baldwin AS Jr. The NF-kB and IkB proteins: new discoveries and
insights. Annu Rev Immunol 1996;14:649–83.
44. Niedermeier M, Hennessy BT, Knight ZA, Henneberg M, Hu J, Kurtova
AV, et al. Isoform-selective phosphoinositide 30 -kinase inhibitors inhibit CXCR4 signaling and overcome stromal cell-mediated drug resistance in chronic lymphocytic leukemia: a novel therapeutic approach.
Blood 2009;113:5549–57.
45. Obeng EA, Carlson LM, Gutman DM, Harrington WJ Jr., Lee KP, Boise
LH. Proteasome inhibitors induce a terminal unfolded protein response
in multiple myeloma cells. Blood 2006;107:4907–16.
46. Ruschak AM, Slassi M, Kay LE, Schimmer AD. Novel proteasome
inhibitors to overcome bortezomib resistance. J Natl Cancer Inst
2011;103:1007–17.
47. Orlowski RZ, Stinchcombe TE, Mitchell BS, Shea TC, Baldwin AS,
Stahl S, et al. Phase I trial of the proteasome inhibitor PS-341 in
patients with refractory hematologic malignancies. J Clin Oncol
2002;20:4420–27.
www.aacrjournals.org
48. Richardson PG, Sonneveld P, Schuster MW, Irwin D, Stadtmauer EA,
Facon T, et al. Bortezomib or high-dose dexamethasone for relapsed
multiple myeloma. N Engl J Med 2005;352:2487–98.
49. Jakubowiak AJ, Dytfeld D, Griffith KA, Lebovic D, Vesole DH, Jagannath S, et al. A phase 1/2 study of carfilzomib in combination with
lenalidomide and low-dose dexamethasone as a frontline treatment for
multiple myeloma. Blood 2012;120:1801–09.
50. Siegel DS, Martin T, Wang M, Vij R, Jakubowiak AJ, Lonial S, et al. A phase
2 study of single-agent carfilzomib (PX-171-003-A1) in patients with
relapsed and refractory multiple myeloma. Blood 2012;120:2817–25.
51. Vij R, Siegel DS, Jagannath S, Jakubowiak AJ, Stewart AK, McDonagh
K, et al. An open-label, single-arm, phase 2 study of single-agent
carfilzomib in patients with relapsed and/or refractory multiple myeloma who have been previously treated with bortezomib. Br J Haematol 2012;158:739–48.
52. Johnson AJ, Yeh Y-Y, Smith LL, Wagner AJ, Hessler J, Gupta S, et al. The
novel cyclin-dependent kinase inhibitor dinaciclib (SCH727965) promotes apoptosis and abrogates microenvironmental cytokine protection
in chronic lymphocytic leukemia cells. Leukemia 2012;26:2554–7.
Clin Cancer Res; 19(9) May 1, 2013
2419
Published OnlineFirst March 20, 2013; DOI: 10.1158/1078-0432.CCR-12-2754
The Proteasome Inhibitor Carfilzomib Functions Independently of
p53 to Induce Cytotoxicity and an Atypical NF- κB Response in
Chronic Lymphocytic Leukemia Cells
Sneha V. Gupta, Erin Hertlein, Yanhui Lu, et al.
Clin Cancer Res 2013;19:2406-2419. Published OnlineFirst March 20, 2013.
Updated version
Supplementary
Material
Cited articles
Citing articles
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
doi:10.1158/1078-0432.CCR-12-2754
Access the most recent supplemental material at:
http://clincancerres.aacrjournals.org/content/suppl/2013/03/20/1078-0432.CCR-12-2754.DC1
This article cites 52 articles, 30 of which you can access for free at:
http://clincancerres.aacrjournals.org/content/19/9/2406.full#ref-list-1
This article has been cited by 2 HighWire-hosted articles. Access the articles at:
http://clincancerres.aacrjournals.org/content/19/9/2406.full#related-urls
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at
[email protected].
To request permission to re-use all or part of this article, contact the AACR Publications Department at
[email protected].
Downloaded from clincancerres.aacrjournals.org on June 18, 2017. © 2013 American Association for Cancer Research.