Microsoft Word - 48208159-file00

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
Prepublished online Feb 3, 2009;
doi:10.1182/blood-2008-07-171389
Green tea polyphenols block the anticancer effects of bortezomib and other
boronic acid-based proteasome inhibitors
Encouse B. Golden, Philip Y. Lam, Adel Kardosh, Kevin J. Gaffney, Enrique Cadenas, Stan G. Louie,
Nicos A. Petasis, Thomas C. Chen and Axel H. Schonthal
Information about reproducing this article in parts or in its entirety may be found online at:
http://bloodjournal.hematologylibrary.org/misc/rights.dtl#repub_requests
Information about ordering reprints may be found online at:
http://bloodjournal.hematologylibrary.org/misc/rights.dtl#reprints
Information about subscriptions and ASH membership may be found online at:
http://bloodjournal.hematologylibrary.org/subscriptions/index.dtl
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published semimonthly
by the American Society of Hematology, 1900 M St, NW, Suite 200, Washington
DC 20036.
Copyright 2007 by The American Society of Hematology; all rights
.
reserved

Blood First Edition Paper, prepublished online February 3, 2009; DOI 10.1182/blood-2008-07-171389
From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
Running head
GOLDEN et al.
GREEN TEA NEUTRALIZES BORTEZOMIB THERAPY
Journal section designation
MYELOID NEOPLASIA
Title
Green tea polyphenols block the anticancer effects of bortezomib and other boronic acid-
based proteasome inhibitors
Bylines
Encouse B. Golden1, Philip Y. Lam2, Adel Kardosh3, Kevin J. Gaffney5, Enrique Cadenas2,
Stan G. Louie4, Nicos A. Petasis5, Thomas C. Chen1,6, Axel H. Schönthal3
Affiliation lines
1Department of Pathology, University of Southern California (USC) Keck School of Medicine
(KSOM), Los Angeles, CA; 2Department of Molecular Pharmacology and Toxicology, USC
School of Pharmacy (SoP), Los Angeles, CA; 3Department of Molecular Microbiology and
Immunology, USC KSOM, Los Angeles, CA; 4Department of Clinical Pharmacy and
Pharmaceutical Economics and Policy, USC SoP, Los Angeles, CA; 5Department of
Chemistry, USC College of Letters, Arts and Sciences, Los Angeles, CA; 6Department of
Neurosurgery, USC KSOM, Los Angeles, CA.
Abbreviations
BZM: bortezomib; CFA: colony formation assay; EGCG: epigallocatechin gallate; EGC:
epigallocatechin; ECG: epicatechin gallate; EC: epicatechin; GRP78: glucose-regulated
protein of molecular weight 78 kDa; GTE: green tea extract; MM: multiple myeloma; NVF:
nelfinavir.
Copyright © 2009 American Society of Hematology

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
Abstract
The anticancer potency of green tea and its individual components is being intensely
investigated, and some cancer patients already self-medicate with this `miracle herb' in hopes of
augmenting the anticancer outcome of their chemotherapy. Bortezomib (Velcade®) is a
proteasome inhibitor in clinical use for multiple myeloma. Here, we investigated whether the
combination of these compounds would yield increased antitumor efficacy in multiple myeloma
and glioblastoma cell lines in vitro and in vivo. Unexpectedly, we discovered that various green
tea constituents, in particular (-)-epigallocatechin gallate (EGCG) and other polyphenols with
1,2-benzenediol moieties, effectively prevented tumor cell death induced by bortezomib in vitro
and in vivo. This pronounced antagonistic function of EGCG was only evident with boronic acid-
based proteasome inhibitors (bortezomib, MG-262, PS-IX), but not with several non-boronic
acid proteasome inhibitors (MG-132, PS-I, nelfinavir). EGCG directly reacted with bortezomib
and blocked its proteasome inhibitory function; as a consequence, bortezomib could not trigger
endoplasmic reticulum stress or caspase-7 activation, and did not induce tumor cell death.
Taken together, our results indicate that green tea polyphenols may have the potential to negate
the therapeutic efficacy of bortezomib and suggest that consumption of green tea products may
be contraindicated during cancer therapy with bortezomib.
Page 2

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
Introduction
Herbal supplements are commonly perceived as "innocent" or "holistic" and have
become hugely popular as unrestrictedly available, over-the-counter "cure-all" remedies. Cancer
patients in particular may be tempted to self-medicate with such supplements in hopes to delay
the progression of their disease and/or reduce the side effects associated with conventional
chemotherapy,1-3 and they may do this unbeknownst to their health care providers.2,3 One very
popular herb is green tea, brewed from the leaves of the plant Camellia sinensis, which appears
to be an ideal alternative medicine because it is non-toxic and shown to have cardioprotective,
neuroprotective, anti-infective, and anti-tumoral properties.4-10 Green tea products are also
available as highly concentrated extracts, which can be easily accessed by the public at local
grocery, pharmacy, and health food stores throughout the United States.
Green tea is a heterogeneous product that contains a number of anti-oxidant
compounds, known as polyphenols. (-)-Epigallocatechin gallate (EGCG), (-)-epicatechin gallate
(ECG), (-)-epigallate catechin (EGC), and (-)-epicatechin (EC) are some of the polyphenol
compounds found in green tea.9 EGCG, the most bioactive green tea polyphenol, was
demonstrated to be a multipotent chemo-preventive and anti-cancer agent in several animal
models, including leukemia, lung, prostate, colon, and breast cancer.9,11-15 It was shown to
interact with numerous protein targets and to disrupt biological/biochemical reactions involved in
cancer progression. The EGCG compound can vary its chemical conformation, thus allowing it
to behave as a chemical chaperone that is able to interact with an assortment of biological
molecules including DNA, RNA, lipids, and proteins.16 However, it is not entirely clear which of
the many functions of EGCG are the most critical with regards to this compound's anticancer
activity.
Bortezomib (BZM; Velcade®, PS-341) has been approved for the treatment of multiple
myeloma and mantle cell lymphoma, and is being investigated for the potential therapy of other
tumors as well.17,18 It functions as a selective inhibitor of the 26S proteasome, a multisubunit
Page 3

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
protein complex that is responsible for the degradation of ubiquitinated proteins that are marked
for deletion from the cell's inventory.19 Proteasome inhibition generates several consequences.
For example, it may cause increased levels of biologically active proteins, such as IB, which is
the inhibitor of nuclear factor-kappaB (NF-B).20 Additionally, misfolded and other obsolete
proteins accumulate as well and trigger the unfolded protein response (UPR), which entails
endoplasmic reticulum (ER) stress.21
ER stress represents a cellular "yin-yang" system. On one side, it restores homeostasis
and thereby supports cell survival during low to moderate stress conditions; on the other side, it
initiates apoptosis in case of severe stress, such as in response to the effective inhibition of
proteasome activity by bortezomib.22,23 A key pro-survival regulator of the ER stress response is
GRP78 (glucose-regulated protein 78 kD; also called BiP), a chaperone protein with several
anti-apoptotic functions, which among them features binding to and prevention of caspase 7
activation.24 Earlier reports have clearly established that increased levels of GRP78 provide
chemoresistance to tumor cells, and therefore, inhibition of this protein may provide an attractive
means to achieve chemosensitization.25
One of several recognized functions of EGCG is its ability to bind to and inhibit GRP78
activity,26 and this interaction may underlie EGCG's reported ability to sensitize tumor cells to
killing by the topoisomerase II inhibitor etoposide or the alkylating agent temozolomide.26,27
Based on these encouraging earlier findings, we initiated a study to investigate whether EGCG
and other green tea constituents would be able to enhance the antitumor effects of bortezomib.
However, to our surprise, we discovered severe antagonism during combination drug
treatments, i.e., EGCG or green tea extract effectively blocked proteasome inhibition by
bortezomib and thereby prevented ER stress and subsequent tumor cell death in vitro and in
vivo. Because the extremely potent blockage of bortezomib's tumoricidal potential was
Page 4

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
accomplished at concentrations of EGCG that are easily achievable in humans, our results
strongly suggest that patients undergoing bortezomib therapy avoid green tea products.
Materials & Methods
Materials
Bortezomib was obtained from the pharmacy as 3.5 mg Velcade® (Millennium
Pharmaceuticals, Cambridge, MA) and was suspended in 3.5 ml saline. Pure bortezomib
(>99%) was purchased from Chemie Tek, Indianapolis, IN. Nelfinavir was obtained from the
pharmacy and suspended in DMSO at 50 mM. MG-132 (Sigma-Aldrich, St. Louis, MO) was
dissolved in DMSO at 10.5 mM. MG-262 and proteasome inhibitor IX (PS-IX) (EMD Chemicals,
Gibbstown, NJ) were dissolved in DMSO at 203.5 M and 26.5 mM, respectively. Proteasome
inhibitor I (PS-I) (Bachem, Weil am Rhein, Germany) was dissolved in DMSO at 10 mM. EGCG,
EGC, ECG, and EC were purchased form Sigma-Aldrich (St. Louis, MO), dissolved in ultra-pure
water, and used fresh (or stored at -80°C). TEAVIGO-EGCG are capsules with caffeine-free,
highly concentrated green tea extract (94% EGCG) that are widely available at health food
stores;28 the capsule contents were dissolved in ultra-pure water and used fresh (or stored at -
80°C). Complete green tea extract in concentrated liquid form (Nature's Answer, Hauppauge,
NY) was purchased from a health food store; the amount of this liquid added to cell culture was
calculated based on its label claim of 30% EGCG.
Cell lines and culturing
Human MM cell lines RPMI/8226 and U266 were obtained from the American Tissue
Culture Collection (ATCC, Manassas, VA) and MM1 cells were kindly provided by Dr. Nancy
Krett. All MM cell lines were propagated in RPMI-1640 (Cellgro, Herndon, VA) supplemented
with 5% fetal bovine serum. Human glioblastoma cell lines (LN229 and U251) were kindly
supplied by Dr. Frank Furnari and propagated in DMEM (Cellgro, Herndon, VA) supplemented
Page 5

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
with 10% fetal bovine serum. All cell lines were grown with 100 U/ml penicillin and 0.1 mg/ml
streptomycin in a humidified incubator at 37°C and a 5% CO2 atmosphere.
MTT assay
MTT assays were performed in 96-well plates with the use of 3.0 x 103 cells per well for
glioblastoma cells and 2.5 x 104 cells per well for multiple myeloma cells as described in detail
elsewhere.29 In individual experiments, each treatment condition was set up in quadruplicate,
and each experiment was repeated one to five times independently.
Colony formation assay (CFA)
Cells were seeded into six-well plates at 200 cells per well. After complete cell
adherence, the cells were exposed to drug treatment for 48 h (in triplicate). Thereafter, the drug
was removed, fresh growth medium was added, and the cells were kept in culture undisturbed
for 12-14 days, during which time the surviving cells spawned a colony of proliferating cells.
Colonies were visualized by staining for 4 h with 1% methylene blue (in methanol) and then
counted.
Apoptosis measurements
Terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end labeling (TUNEL)
detection kit was purchased from Chemicon (Temecula, CA) and used according to the
manufacturer's instructions.
Immunoblots and antibodies
Preparation of cell lysates and determination of protein concentration was performed as
described before.29 50 g of lysate from each sample was run in parallel. The primary antibodies
were purchased from Cell Signaling Technologies (Beverly, MA) or Santa Cruz Biotechnology
Inc. (Santa Cruz, CA) and used according to the manufacturers' instructions. The blocking
buffer and fluorescent-conjugated secondary antibodies were purchased from LI-COR
Biosciences (Lincoln, NE) and used according to protocols supplied by the manufacturer. The
membranes were scanned and analyzed with the Odyssey Infrared Imaging System (LI-COR
Page 6

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
Biosciences) according to the manufacturer's instructions. All immunoblots were repeated at
least once.
Proteasome activity assay
Proteasome activity was determined in two different ways, either in lysates from drug-
treated cells (whole-cell activity) or via the direct addition of drugs to purified proteasomes. For
whole-cell proteasome activity, RPMI/8226 or LN229 cells were treated with various drug
combinations for up to 30 hours and then disrupted by vortexing in radioimmunoprecipitation
assay (RIPA) buffer.29 Proteasome activity in 100 g of total cell lysate was measured using a
20S proteasome activity assay kit (Millipore, Billerica, MA) according to the manufacturer's
instructions.
For purified proteasome activity, various concentrations of bortezomib were mixed with 5
M EGCG in proteasome assay buffer (25mM HEPES, pH 7.5, 0.5mM EDTA, 0.05% NP-40 and
0.001% SDS (w/v)), incubated for 30 min at RT, and then added to 0.2 g of isolated 20S
proteasomes (Sigma-Aldrich). After 5 min incubation at room temperature, proteasome activity
was measured using a 20S proteasome activity assay kit. The emitted fluorescence of cleaved
proteasome substrate (LLVY-AMC) was detected with a 380/460 nm filter. Cleaved LLVY-AMC
is indicative of chymotryptic activity.
Alizarin Red S Assay
The ARS assay is an optical reporter assay that is widely used to study the binding
events between boronic acids and a wide range of molecules harboring 1,2-benzene diol
(catechol) moieties.30 Its principle is based on the binding of boronic acid to the catechol unit of
the ARS molecule, which results in corresponding alterations of its color and fluorescence,
which can be recorded and quantitated. We used this assay as a three-component competition
system of ARS, boronic acid (i.e., bortezomib), and a second 1,2-benzene molecule (i.e.,
EGCG). Several repetitions with varying concentrations of the individual components were
performed, and binding intensity was measured via readings of optical density (450 nm) and/or
Page 7

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
fluorescence in a 96-well plate reader. Some of the repetitions were also performed with
TEAVIGO-EGCG (and yielded essentially the same results).
Drug treatment of nude mice
All animal protocols were approved by the Institutional Animal Care and Use Committee
(IACUC) of the University of Southern California, and all applicable policies were strictly
observed during the course of this study. Four- to six-week-old male athymic nu/nu mice were
obtained from Harlan (Indianapolis, IN) and implanted subcutaneously with 5 X 106 RPMI/8226
multiple myeloma cells. Once tumors of ~300 mm3 had developed, the mice received drug
treatments. Bortezomib was given as a single dose via i.p. injection. EGCG was given as a
single dose via direct administration into the stomach with a stainless steel ball-head feeding
needle (Popper and Sons, Inc., New Hyde Park, NY). After a total of 72 hours, the mice were
sacrificed and tumors were collected for analysis. The mice were closely monitored with regards
to body weight, food consumption, and clinical signs of toxicity; no differences between non-
drug-treated control animals and drug-treated animals were detected within these parameters.
Hematoxylin and eosin staining
7 µm-thick frozen tissue sections on glass slides (VWR, Media, PA) were fixed in
acetone (Sigma-Aldrich) for 10 minutes. The tissue sections were allowed to dry for 10 minutes
and then immediately stained with Mayer's hematoxylin (Sigma-Aldrich) for another 10 minutes.
Afterwards, the slides were rinsed in tap water and then stained with eosin (Surgipath,
Richmond, IL) for 10 seconds. The slides were then dehydrated in 95% and 100% ethanol in
succession for 10 seconds followed by a 30-second incubation in xylene (Sigma-Aldrich).
Finally, the slides were mounted on glass cover slips, sealed with nail polish, and analyzed
under the microscope.
Page 8

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
Results
EGCG blocks the cytotoxic effects of BZM in multiple myeloma cells
To investigate whether the cytotoxic effects of bortezomib (BZM) could be affected by
the addition of EGCG (see Figure 1A for chemical structures), we treated different multiple
myeloma (MM) cell lines with these agents alone or in combination and determined cell growth
and survival by various assays. MTT assay of RPMI/8226 cells revealed that 10 nM BZM
effectively killed the entire cell culture within 48 hours, whereas over the same time period 10
µM EGCG alone had no effect on cell survival. Surprisingly, however, when BZM and EGCG
were added together, the cell killing by BZM was completely prevented and cell survival
remained at 100% (Figure 1B). A similar protective effect of EGCG was noticed when the extent
of apoptosis was investigated by TUNEL assay. Here, the addition of BZM increased the
amount of cell death to nearly 100%, and this increase was effectively and completely blocked
by the addition of EGCG (Figure 1C). Furthermore, cell cycle analysis of untreated cells
revealed the typical distribution of a proliferating cell culture across G1, S, G2/M, with a small
fraction (17%) undergoing cell death (indicated by the presence of cells within gate M1 to the
left of the dotted line in Figure 1D). Upon the addition of BZM, the normal cell cycle distribution
disappeared and the majority of cells (65%) shifted into M1. In stark contrast, however, this
effect was effectively blocked by EGCG; i.e., when EGCG and BZM were added together, the
treated cell culture maintained its normal cell cycle distribution and the fraction of dying cells
(M1) remained at the basal level (Figure 1D). This comprehensive cytoprotective effect of
EGCG was also verified by microscopic inspection of cell cultures, where the apoptotic
phenotype could be clearly observed in the presence of BZM, but was entirely absent when
EGCG was added as well (Figure 1E).
The sweeping cytoprotective effect of EGCG could also be documented at variable
concentrations of each compound and in different MM cell lines (Figure 1F-J). For example,
EGCG as low as 2.5 or 5.0 µM (which can easily be achieved in the blood of humans31)
Page 9

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
effectively blocked BZM (10 nM) cytotoxicity by 80100% in RPMI/8226, U266, and MM1 cell
lines (Figure 1F,G,H). The protective effect of EGCG could also be observed at increased
concentrations of each drug; for example, when BZM concentrations were quadrupled to 40 nM,
the presence of 10 µM EGCG reduced cell killing by 50% in RPMI/8226 cells; and even at ten-
fold increased BZM concentrations (100 nM), 20 µM EGCG was still able to significantly reduce
cell killing by 40% (Figure 1F). Thus, in this latter example, the cytotoxic IC50 of BZM increased
by an order of magnitude from 5 nM to >80 nM.
Earlier reports32,33 had indicated that EGCG alone may have cytotoxic effects on MM cell
lines when applied at concentrations of 10 µM or higher (which are much greater than the
typical concentrations achieved in humans). In comparison, we only noted weak cytotoxic
effects starting at 20 µM, which become somewhat more pronounced when drug exposure
times are increased from 48 to 72 hours (Figure 1H,I). Interestingly, even under conditions
where EGCG is slightly toxic, it is still able to potently antagonize the cytotoxic effects of BZM.
For example, in U266 cells, 20 µM EGCG alone reduces viability by 20% after 48 hours, and 10
nM BZM reduces viability by >95%--yet, when the two drugs are combined, viability is still only
reduced by 20% (Figure 1G).
Many of the experiments in this study were repeated with a different EGCG preparation,
called TEAVIGO-EGCG,28 that is sold in capsules from health food stores. In all cases,
TEAVIGO-EGCG yielded very similar results (Figure 1J). Moreover, when two primary tissue
samples from MM patients were analyzed, we obtained essentially the same outcome as shown
above with MM cell lines, namely that EGCG provided highly effective protection from the
cytotoxic effects of BZM (Figure 1K).
EGCG blocks the cytotoxic effects of BZM in glioblastoma cell lines
BZM is currently being investigated for the treatment of cancers other than multiple
myeloma. We therefore set out to determine whether EGCG could neutralize BZM cytotoxicity in
Page 10

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
two glioblastoma cell lines, U251 and LN229. In addition to the conventional MTT assay that
determines growth and survival of the entire cell culture during short-term treatments (48 hours),
we also performed colony-formation assays (CFA), which revealed the long-term (2 weeks)
survival of individual cells and their ability to spawn a colony of descendants. Figure 2A shows
the outcome of MTT assays with LN229 glioblastoma cells, which essentially recapitulated the
results obtained with the multiple myeloma cells shown in Figure 1, i.e., the presence of EGCG
effectively blocked the cytotoxic activity of BZM; in this case, the IC50 of BZM was increased
from 5 nM to 25, 50, and >100 nM in the presence of 10, 20, or 40 µM EGCG, respectively.
The cytoprotective effect of EGCG towards BZM was also verified in CFA. Figure 2B
shows an example of the actual colonies that formed after drug treatment, whereas Figure 2C
displays the quantitative analysis of several repetitions of this experiment. As can be seen in
both glioblastoma cell lines LN229 and U251, only a very small number of colonies formed after
treatment with BZM; but when cells were treated with BZM in the presence of EGCG, the
number of colonies was nearly as high as those in the untreated control group (Figures 2B, C).
Thus, the antagonistic characteristic of BZM and EGCG treatment could also be documented in
adherently growing cell lines representing a solid tumor type.
Other polyphenols and complete green tea extract neutralize the cytotoxic effects of BZM
In addition to EGCG, green tea contains various other polyphenol compounds, including
EGC, ECG, and EC (Figure 3A). To determine if complete green tea extract and individual
polyphenols other than EGCG can neutralize the cytotoxic effects of BZM, the effect of these
combinations on cell growth and survival was investigated by MTT assays. As shown in Figure
3B, each individual green tea polyphenol (EGC, ECG, EC) was able to neutralize the cytotoxic
effects of BZM. In comparison to EGCG, however, cytoprotection by these three polyphenols
against BZM was noticeable weaker. While 10 µM EGCG was able to provide 100% protection
Page 11

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
against 10 nM BZM (Figure 1 and 2A), these three polyphenols required higher concentrations
to achieve the same degree of cytoprotection (Figure 3B).
We next investigated the effects of a complete green tea extract (GTE) purchased from
a health food store as a concentrated liquid, which represents a mixture of all of the above
polyphenols and other ingredients naturally present in green tea. Liquid GTE was added to
LN229 glioblastoma, as well as U266 and RPMI/8226 MM cells, over a range of concentrations
providing EGCG from 0.17.5 µg/mL (equivalent to 0.2216.6 µM), and its impact on BZM
cytotoxicity was investigated by MTT assays. As seen before with individual polyphenols,
increasing concentrations of GTE exerted increasing and very effective protection from the
cytotoxic effects of BZM (Figure 3C-E). At the highest concentration of GTE (equivalent to 7.5
µg/ml EGCG), some cytotoxic effects of GTE became apparent; however, despite a 50%
reduction of viability by GTE alone, the further addition of BZM, even at 50 nM, could not reduce
viability further (Figure 3C). Thus, similar to the individual polyphenols tested, complex GTE was
able to provide effective protection against BZM, even when its concentration was increased to
the point where it became somewhat toxic by itself.
EGCG differentially affects cytotoxicity of various proteasome inhibitors
To determine whether EGCG can neutralize the cytotoxic effects of proteasome
inhibitors other than BZM, we included two different categories of such compounds in our
analysis (see Figure 4A for chemical structures). The first group contained MG-262 and PS-IX,
which represent compounds harboring a boronic acid moiety (like BZM). The second group
contained MG-132, PS-I, and nelfinavir (NFV), representing compounds without a boronic acid
moiety. (NFV is a clinically used HIV protease inhibitor that is known to also block proteasome
function.34) The inclusion of MG-132 and MG-262 was expected to be particularly revealing, as
these two compounds differ only in the presence/absence of the boronic acid moiety.
Page 12

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
All of the above compounds were added to cells at increasing concentrations, in the
presence or absence of EGCG, and cell survival was determined by CFA. As shown in Figure 4,
all five proteasome inhibitors displayed highly cytotoxic effects. In the case of the three
proteasome inhibitors lacking a boronic acid moiety (MG-162, PS-I, NFV), the addition of EGCG
did not at all interfere with their severe cytotoxic efficacy (Figure 4B-D). In contrast, and as
observed before with BZM, EGCG provided profound (close to 100%) protection against the
proteasome inhibitors that were based on boronic acid (MG-262, PS-IX) (Figure 4E,D). For
example, 10 nM MG-262 or 5 µM PS-IX reduced colony formation by 100%, but this extremely
severe cytotoxic effect was nearly completely prevented by the presence of EGCG. Thus, the
protective feature of EGCG is not a general effect towards all proteasome inhibitors, but rather
displays selectivity towards those compounds harboring a boronic acid moiety.
EGCG prevents proteasome inhibition and ER stress induction when combined with
boronic acid-based compounds
To gain insight into the molecular mechanisms by which EGCG exerted its robust
cytoprotective effect against boronic acid-based proteasome inhibitors, we directly investigated
the consequences for (i) proteasome inhibition and (ii) induction of ER stress, which represents
one of the critical pro-apoptotic processes mediating cell death initiated by proteasome
inhibition.22,23 Multiple myeloma RPMI/8226 cells and glioblastoma LN229 cells were
investigated in parallel.
The drug effects on cellular proteasome activity were determined by different
approaches: (1) directly via the measurement of proteasome activity, either in lysates from drug
treated cells or with purified 20S proteasomes, and (2) indirectly via Western blot analysis of
ubiquitinated proteins, which accumulate when proteasome activity is inhibited. As shown in
Figure 5A and B (top panels), treatment of cells with BZM, MG-132, or MG-262 alone resulted in
pronounced reduction of proteasome activity; at the same time, there was a marked
Page 13

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
accumulation of ubiquitinated proteins (middle panels of Figure 5A and B), confirming the
biological consequence of proteasome inhibition. When EGCG was included in these drug
treatments, proteasome inhibition and accumulation of ubiquitinated proteins was prevented in
the case of BZM and MG-262 (boronic acid-based compounds), but not in the case MG-132
(similar to MG-262, but lacking the boronic acid moiety).
EGCG and BZM were also added to highly purified 20S proteasomes in the absence of
any other cellular components, such as signaling proteins or organelles. Under these strictly
defined conditions as well, the proteasome inhibitory activity of BZM was diminished when
EGCG was present (Figure 5C), although the antagonism seemed less pronounced as
compared to conditions where intact cells were treated with these compounds (Figures 1-4 and
5A, B). Nonetheless, this result indicates that EGCG is able to inhibit BZM function in the
absence of additional cellular components.
The same drug combinations as above were investigated for any consequences on ER
stress and subsequent apoptosis. Expression levels of GRP78 protein represent a widely used
marker for the induction of ER stress.25 As shown in Figure 5B, this marker was strongly
induced by all three proteasome inhibitors, and the presence of EGCG effectively prevented this
induction in the case of BZM and MG-262, but not in the case of MG-132. Caspase 7 is an ER
stress-associated enzyme and important mediator of ER stress-induced apoptosis.24 As shown
in Figure 5A, caspase 7 is effectively activated by BZM; similarly, BZM causes cleavage of
PARP (poly (ADP ribose) polymerase), which represents one of the final steps of the proteolytic
caspase cascade and reliably indicates ongoing apoptosis.35 However, in the presence of
EGCG, BZM is unable to trigger caspase 7 activation or PARP cleavage, indicating that these
pro-apoptotic events are not set in motion when the polyphenol is present.
Page 14

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
EGCG neutralizes the anti-tumoral effects of BZM in vivo
We next investigated whether the antagonistic effects of EGCG and BZM would have in
vivo relevance. For this purpose, RPMI/8226 multiple myeloma cells were implanted
subcutaneously into nude mice and, after sizable tumors had formed, the animals received
treatment with BZM or EGCG, alone and in combination. Three days later, the animals were
sacrificed and the tumors were analyzed by TUNEL or hematoxylin/eosin (H&E) staining to
visualize the extent of apoptotic and necrotic cell death, respectively.
We found that treatment with BZM alone (0.5 mg/kg) doubled the number of TUNEL-
positive cells, i.e., there was a statistically significant increase in apoptosis. However, in tumors
from animals treated with EGCG alone (25 or 50 mg/kg) or with EGCG plus BZM, there was no
increase of apoptotic cell death as compared to tumors from non-treated controls (Figure 6A,B).
In a repetition of this in vivo experiment, we doubled the dosage of BZM to 1 mg/kg and
performed H&E staining to reveal any necrosis within the tumor tissue. As shown in Figure 6C,
BZM-treated animals showed signs of extensive subcapsular necrosis in their tumor tissue,
whereas none of the non-treated animals, EGCG-treated animals, or EGCG/BZM combination
treated animals displayed this outcome. Thus, the severe antagonistic effect of EGCG towards
the cytotoxic efficacy of BZM is not merely an in vitro phenomenon, but also takes place in
tumor tissues of drug-treated animals in vivo.
EGCG inactivates BZM via a direct reaction with its boronic acid group
As EGCG is known to have anti-oxidant function, we investigated whether this feature
might contribute to the observed antagonism between EGCG and BZM. However, as shown in
Figure 7A, BZM treatment did not generate increased levels of reactive oxygen levels (ROS)
and therefore, there was no basis for EGCG interference. As a positive control, cells were
challenged with hypoxia (a known inducer of ROS36) or with hydrogen peroxide (H2O2), and in
these cases, significantly (p=0.005) increased levels of ROS were generated, which were
Page 15

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
effectively squelched by EGCG, confirming its potent anti-oxidant function. Taken together,
these results indicated that EGCG's anti-oxidant function did not play a major role in its severe
antagonism of BZM.
As an alternative, and because all of the investigated molecular and cellular effects of
BZM were completely blocked in the presence of EGCG, we considered the possibility that
EGCG prevented BZM function via direct interactions between the two molecules, which might
prohibit BZM from binding to its target, the proteasome. This hypothesis is well supported by
long-known findings in the field of boron chemistry: namely that boronic acid derivatives form
covalent bonds with compounds having a 1,2-benzene diol moiety, resulting in the formation of
cyclic boronate esters and representing one of the strongest single-pair reversible functional
group interactions.37,38 In this regard, it should be noted that EGCG, ECG, EGC, and EC harbor
such 1,2-benzene diol moieties (see chemical structures in Figures 1A and 3A).
To investigate the postulated binding between BZM and EGCG, both compounds were
added to Alizarin Red S (ARS), a widely used reporter system to study the interaction between
boronic acids and molecules with 1,2-diols.30 As shown in Figure 7B, BZM readily bound to
ARS, but this interaction was effectively blocked in the presence of EGCG, revealing effective
competition by direct chemical binding of EGCG to BZM. In order to verify the reliability of the
ARS system for our purposes, we also included MG-132 and MG-262 in repetitions of this
experiment (Figure 7C). As expected, MG-132 (which lacks the boronic acid moiety) did not
bind to ARS, and there was no effect of added EGCG. However, MG-262 (which only differs
from MG-132 by the presence of a boronic acid moiety) displayed effective binding to ARS,
which could be blocked by EGCG. Together, these data indicate direct chemical interactions
between the EGCG molecule and boronic acid-based proteasome inhibitors.
Finally, in order to observe directly the interaction of EGCG with BZM, we carried out 1H
NMR and 13C NMR experiments of several ratios of BZM and EGCG in aqueous media (Figure
7D). By simply mixing the two compounds, the direct formation of a new adduct could be
Page 16

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
observed in the NMR spectra (Figure 7D, part c). Furthermore, the amount of the new adduct
increased, relative to BZM, upon the increased addition of EGCG (Figure 7D, part d). Overall,
these results demonstrate that EGCG forms a stable cyclic boronate adduct with the boronic
acid moiety of BZM and, thereby, prevents BZM from acting as a proteasome inhibitor.
Page 17

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
Discussion
Green tea, concentrated GTE, and certain individual components of green tea, such as
EGCG, have widely recognized health benefits, which include purported anticarcinogenic and
antitumor effects. Besides being used by healthy people as a means to support continued
health, this "miracle herb" extract is also consumed by many cancer patients who follow popular
trends and self-medicate with complementary and alternative medicine (CAM) in hopes to
support their conventional therapy or to lessen the burden of side effects--sometimes without
the knowledge of their health care provider. In an effort to harness the therapeutic potential of
green tea in a more systematic fashion, numerous clinical trials are ongoing to establish the
framework for its future therapeutic use.
GTE or EGCG may also hold cancer therapeutic potential when combined with
established cancer treatments.39 In our current study presented here, we investigated whether
GTE or EGCG would be able to boost the anticancer efficacy of BZM, a proteasome inhibitor
that is in clinical use for the treatment of multiple myeloma and is under consideration for the
treatment of other cancers as well. To our surprise, we discovered that GTE, as well as some of
its individual components, effectively blocked the anticancer efficacy of BZM; i.e., in the
presence of polyphenolic green tea components, BZM was unable to exert proteasome
inhibitory function and, as a consequence, there was no trigger for pro-apoptotic ER stress and
subsequent cell death.
The severe antagonistic effect of EGCG appeared to require the presence of the boronic
acid moiety in BZM. Among the six proteasome inhibitors tested, the boronic acid containing
ones (BZM, MG-262, PS-IX) were similarly incapacitated by EGCG, whereas none of those
without this functional group (NFV, MG-132, PS-I) were affected by EGCG. In this context, it is
noteworthy that the chemical structures of MG-262 (effective inhibition by EGCG) and MG-132
(no inhibition by EGCG) differ only in the presence/absence of the boronic acid moiety (Figure
4A), indicating that the decisive mechanism of EGCG's antagonism resides with the chemical
Page 18

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
structure of the target molecule (i.e., the boronic acid moiety), not its function (i.e., proteasome
inhibition).
The ability of molecules with a 1,2-diol group to form covalent cyclic boronate moieties
with boronic acid in a tight, but reversible manner, is a well-described and long-established
chemical process, and is known to be one of the strongest single-pair reversible functional
group interactions in an aqueous environment.37,38,40 In retrospect, it was therefore not
surprising that we were able to verify direct molecular interactions between EGCG and the
bortezomib molecule (Figure 7), while similar behavior is expected for other polyphenols that
harbor 1,2-benzenediol groups, such as EGC, ECG, and EC. Interestingly, it has been reported
earlier that vitamin C, a 1,2-diol containing compound as well, is able to antagonize cell killing
by BZM.41 This latter report measured the direct interaction of vitamin C with 5-DMANBA (a
fluorescent model boronic acid compound) and provided evidence that the antagonism of
vitamin C against BZM was based on the direct reaction with the boronic acid moiety. Also,
another very recent report showed that dietary flavonoids, such as quercetin and myricetin,
underwent chemical reactions with the boronic acid group of BZM, thereby diminishing BZM's
ability to trigger apoptosis in tumor cells.42 The results reported in our present study regarding
the severe antagonism of EGCG towards BZM can be attributed similarly to a direct molecular
interaction between the two molecules. This view is in agreement with our finding that
proteasome inhibition by BZM does not take place in the presence of EGCG, i.e., BZM is
prevented from exercising the key step of its biological activity (Figure 5).
An earlier study had reported that EGCG and ECG--but not EGC or EC--exerted
protease inhibitory function.43 An earlier study had reported that EGCG and ECG--but not EGC
or EC--exerted protease inhibitory function.43 However, we were unable to detect such
inhibition, neither indirectly via the cellular accumulation of ubiquitinated proteins by Western
blot analysis (Figure 5A,B), nor directly via the measurement of proteasomal chymotryptic
activity (Figure 5C), and it is likely that the observed EGCG actions are due to the higher
Page 19

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
concentrations used. We note, however, that in our experiments all four of these polyphenols
were able to antagonize the cytotoxicity of BZM (Figure 3). Therefore, there is a lack of
correlation between the earlier reported proteasome inhibitory activity profile of these
compounds and their currently presented antagonism towards BZM, which would argue against
a major role of proteasome inhibition, if any, in the antagonistic events we are reporting. This
view is supported further by our observation that the severe antagonism of EGCG appears to be
dependent on the chemical structure of its targets (i.e., the boronic acid moiety of the
proteasome inhibitors), rather than their function (i.e., EGCG does not interfere with other
proteasome inhibitors lacking a boronic acid group) (Figure 4).
Earlier reports32,33 had indicated that EGCG alone may have cytotoxic effects on MM cell
lines when applied at concentrations of 10 µM or higher. In our own experimental setting,
however, we only noted weak cytotoxic effects starting at 20 µM (while at the same time EGCG
continued to exert profound antagonism towards BZM cytotoxicity; Figures 1-3). The absence of
cytotoxic effects at 10 µM, which contrasted with the earlier studies, was noted with EGCG from
two entirely different sources, i.e., from Sigma-Aldrich and from a health food store (TEAVIGO-
EGCG). We noticed, however, that during prolonged storage at room temperature, the clear
EGCG stock solution acquired a brownish color (presumably due to auto-oxidation) and became
somewhat more toxic to cells. This did not happen when EGCG solution was used fresh or after
storage at -80°C. Similary, there was comparatively higher cytotoxicity when GTE (purchased as
a dark-brown liquid from a health food store at room temperature) was used (Figure 3), further
pointing to the possibility that variable storage conditions may impact the cytotoxic potency of
solutions containing EGCG and that this may contribute to the reported discrepancies
concerning EGCG toxicity at 10 µM. In all cases, however, GTE and EGCG at low (non-toxic)
as well as at higher (weakly toxic) concentrations effectively abolished BZM toxicity.
Several of the above mentioned components (EGCG, EGC, ECG, vitamin C) are known
for their potent anti-oxidant function. In view of several earlier reports, which provided evidence
Page 20

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
that BZM triggered the formation of reactive oxygen species (ROS), it would be reasonable to
conclude that anti-oxidants antagonize BZM cytotoxicity via their radical scavenging function,
and that this would be the major antagonistic mechanism of the green tea compounds used in
our study. However, several lines of evidence argue against this conclusion. (i) Some earlier
conclusions about the role of ROS in BZM-induced cell death were based on the use of tiron (a
sulfonate-containing 1,2-benzene diol) as an anti-oxidant.23,44 Although tiron is able to prevent
BZM-induced cell death, it is also one of those 1,2-diol compounds that are able to directly react
with the boronic acid moiety of BZM; thus, the use of tiron does not allow to distinguish between
ROS scavenging and inactivation of BZM via direct binding. (ii) In previous side-by-side
comparisons, only those anti-oxidants with 1,2-diol groups (vitamin C, tiron, quercetin), but not
those without (N-acetyl-cysteine, butylated hydroxyanisole, MnTBAP, glutathione, vitamin E),
were able to antagonize cell killing by BZM.41,42,45,46 (iii) In head and neck squamous cell
carcinoma cells, induction of ROS was shown to be a consequence of BZM-induced ER
stress.23 In our cell system as well, markers of severe ER stress (GRP78, caspase 7) were
induced by BZM; however, in the presence of EGCG, ER stress was entirely prevented (Figure
5) and therefore could not possibly serve as a source of ROS. (iv) Anti-oxidants with 1,2-
benzene diol groups are able to antagonize BZM function even in cells where BZM treatment
does not cause ROS; for example, in our cell system, treatment with BZM did not increase ROS
(Figure 7A), yet EGCG and other anti-oxidants (EGC, ECG) were still able to block BZM
cytotoxicity; similarly, in several lung cancer cell lines, BZM did not increase ROS either, yet
vitamin C or tiron were able to prevent BZM cytotoxicity.41 In summary, these considerations
support our conclusion that EGCG and other components of GTE exert their antagonistic
function against BZM (and other boronic acid-based proteasome inhibitors) via direct
interactions between these molecules, rather than indirectly via ROS scavenging.
Compared to the previously identified antagonists of BZM41,42,45,46, we find that the
antagonism of EGCG towards BZM treatment is at least two orders of magnitude more potent
Page 21

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
than vitamin C or tiron; e.g., while the latter required concentrations in the range of 500-1,000
µM to be highly effective41, EGCG exerts pronounced (80-100%) protection at concentrations as
low as 2.5-5.0 µM (Figure 1). Even more so, when applied as complete GTE, very substantial
protection can be seen at EGCG concentrations of 1.0 µM--presumably because other
polyphenols that are present in GTE contribute as well (Figure 3). It is noteworthy that these low
concentrations are not merely diminishing the cytotoxic efficacy of BZM, but are able to
completely abolish it, which is also apparent in the in vivo model we used, where significantly
increased apoptosis and necrosis in tumor tissue in response to BZM therapy was entirely
absent when EGCG was added (Figure 6). Furthermore, in view of the underlying mechanism of
EGCG/BZM antagonism, namely chemical interactions between the molecules (Figure 7), we
envision that inactivation of BZM's therapeutic efficacy may not only apply to MM, but also to
any other cancer type where BZM might be considered for inclusion in the therapeutic regimen.
A case in point is our finding that EGCG abolishes BZM's cytotoxic potency in glioblastoma cell
lines as well (Figure 2).
In humans, EGCG concentrations of 58 µM can easily be achieved after the ingestion
of capsules containing GTE (polyphenon E).31 We therefore have no doubt that our discovery is
highly relevant for clinical considerations and would strongly urge patients undergoing BZM
therapy to abstain from consuming green tea products, in particular those widely available,
highly concentrated GTEs that are sold in liquid or capsule form. The significance of our view
may be further underscored by the following consideration: Based on the proposed mechanism
by which EGCG/GTE antagonizes BZM, namely via direct chemical inactivation of BZM and
thereby preventing its biological activity, one would expect that not only the therapeutic efficacy
of BZM is being neutralized, but perhaps also the emergence of some of the side effects that
usually accompany BZM therapy. As a consequence, patients consuming EGCG/GTE products
may experience improved well being, which may encourage further increased EGCG/GTE
Page 22

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
dosages, leading to even further improvements--while at the same time, and unknowingly, the
therapeutic efficacy of their BZM treatment is severely blunted, if not entirely obliterated.
Acknowledgements
We are grateful to Dr. Frank Furnari (Ludwig Institute of Cancer Research, La Jolla, CA)
and Dr. Nancy Krett (Northwestern University, Chicago, IL) for the various cell lines used in this
study. We thank Simon Coetzee for assisting with cell culturing. This project was supported by
funding from the Multiple Myeloma Research Foundation (to AHS). EBG wishes to dedicate this
work to the memory of JNB.
Authorship
Contribution: EBG, PYL, AK and KJG performed experiments and analyzed the results.
EBG, TCC, SGL, NAP, and EC designed experimental approaches and provided input towards
the manuscript. AHS directed the study and revised the manuscript. The authors declare no
financial conflict of interest.
Correspondence: Axel H. Schönthal, University of Southern California Keck School of
Medicine, Department of Molecular Microbiology and Immunology, 2011 Zonal Avenue, HMR-
405, Los Angeles, CA 90089-9094; e-mail: schontha@usc.edu
Page 23

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
References
1.
Sparreboom A, Cox MC, Acharya MR, Figg WD. Herbal remedies in the United States:
potential adverse interactions with anticancer agents. J Clin Oncol. 2004;22:2489-2503.
2.
Richardson MA, Sanders T, Palmer JL, Greisinger A, Singletary SE.
Complementary/alternative medicine use in a comprehensive cancer center and the implications
for oncology. J Clin Oncol. 2000;18:2505-2514.
3.
Burstein HJ, Gelber S, Guadagnoli E, Weeks JC. Use of alternative medicine by women
with early-stage breast cancer. N Engl J Med. 1999;340:1733-1739.
4.
Weber JM, Ruzindana-Umunyana A, Imbeault L, Sircar S. Inhibition of adenovirus
infection and adenain by green tea catechins. Antiviral Res. 2003;58:167-173.
5.
Yamaguchi K, Honda M, Ikigai H, Hara Y, Shimamura T. Inhibitory effects of (-)-
epigallocatechin gallate on the life cycle of human immunodeficiency virus type 1 (HIV-1).
Antiviral Res. 2002;53:19-34.
6.
Song JM, Seong BL. Tea catechins as a potential alternative anti-infectious agent.
Expert Rev Anti Infect Ther. 2007;5:497-506.
7.
Mandel SA, Avramovich-Tirosh Y, Reznichenko L, et al. Multifunctional activities of
green tea catechins in neuroprotection. Modulation of cell survival genes, iron-dependent
oxidative stress and PKC signaling pathway. Neurosignals. 2005;14:46-60.
8.
Yang CS, Lambert JD, Hou Z, Ju J, Lu G, Hao X. Molecular targets for the cancer
preventive activity of tea polyphenols. Mol Carcinog. 2006;45:431-435.
9.
Kurahashi N, Sasazuki S, Iwasaki M, Inoue M, Tsugane S. Green tea consumption and
prostate cancer risk in Japanese men: a prospective study. Am J Epidemiol. 2008;167:71-77.
10.
Cabrera C, Artacho R, Gimenez R. Beneficial effects of green tea--a review. J Am Coll
Nutr. 2006;25:79-99.
11.
Khan N, Mukhtar H. Tea polyphenols for health promotion. Life Sci. 2007;81:519-533.
Page 24

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
12.
Thangapazham RL, Singh AK, Sharma A, Warren J, Gaddipati JP, Maheshwari RK.
Green tea polyphenols and its constituent epigallocatechin gallate inhibits proliferation of human
breast cancer cells in vitro and in vivo. Cancer Lett. 2007;245:232-241.
13.
Yang CS, Lambert JD, Ju J, Lu G, Sang S. Tea and cancer prevention: molecular
mechanisms and human relevance. Toxicol Appl Pharmacol. 2007;224:265-273.
14.
Chen L, Zhang HY. Cancer preventive mechanisms of the green tea polyphenol (-)-
epigallocatechin-3-gallate. Molecules. 2007;12:946-957.
15.
Leong H, Mathur PS, Greene GL. Inhibition of mammary tumorigenesis in the
C3(1)/SV40 mouse model by green tea. Breast Cancer Res Treat. 2008;107:359-369.
16.
Kuzuhara T, Suganuma M, Fujiki H. Green tea catechin as a chemical chaperone in
cancer prevention. Cancer Lett. 2008;261:12-20.
17.
Richardson PG, Sonneveld P, Schuster MW, et al. Bortezomib or high-dose
dexamethasone for relapsed multiple myeloma. N Engl J Med. 2005;352:2487-2498.
18.
Ludwig H, Khayat D, Giaccone G, Facon T. Proteasome inhibition and its clinical
prospects in the treatment of hematologic and solid malignancies. Cancer. 2005;104:1794-
1807.
19.
Groll M, Berkers CR, Ploegh HL, Ovaa H. Crystal structure of the boronic acid-based
proteasome inhibitor bortezomib in complex with the yeast 20S proteasome. Structure.
2006;14:451-456.
20.
Fahy BN, Schlieman MG, Mortenson MM, Virudachalam S, Bold RJ. Targeting BCL-2
overexpression in various human malignancies through NF-kappaB inhibition by the
proteasome inhibitor bortezomib. Cancer Chemother Pharmacol. 2005;56:46-54.
21.
Meusser B, Hirsch C, Jarosch E, Sommer T. ERAD: the long road to destruction. Nat
Cell Biol. 2005;7:766-772.
22.
Fribley A, Wang CY. Proteasome inhibitor induces apoptosis through induction of
endoplasmic reticulum stress. Cancer Biol Ther. 2006;5:745-748.
Page 25

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
23.
Fribley A, Zeng Q, Wang CY. Proteasome inhibitor PS-341 induces apoptosis through
induction of endoplasmic reticulum stress-reactive oxygen species in head and neck squamous
cell carcinoma cells. Mol Cell Biol. 2004;24:9695-9704.
24.
Reddy RK, Mao C, Baumeister P, Austin RC, Kaufman RJ, Lee AS. Endoplasmic
reticulum chaperone protein GRP78 protects cells from apoptosis induced by topoisomerase
inhibitors: role of ATP binding site in suppression of caspase-7 activation. J Biol Chem.
2003;278:20915-20924.
25.
Lee AS. GRP78 induction in cancer: therapeutic and prognostic implications. Cancer
Res. 2007;67:3496-3499.
26.
Ermakova SP, Kang BS, Choi BY, et al. (-)-Epigallocatechin gallate overcomes
resistance to etoposide-induced cell death by targeting the molecular chaperone glucose-
regulated protein 78. Cancer Res. 2006;66:9260-9269.
27.
Pyrko P, Schonthal AH, Hofman FM, Chen TC, Lee AS. The unfolded protein response
regulator GRP78/BiP as a novel target for increasing chemosensitivity in malignant gliomas.
Cancer Res. 2007;67:9809-9816.
28.
Wolfram S, Raederstorff D, Wang Y, Teixeira SR, Elste V, Weber P. TEAVIGO
(epigallocatechin gallate) supplementation prevents obesity in rodents by reducing adipose
tissue mass. Ann Nutr Metab. 2005;49:54-63.
29.
Kardosh A, Soriano N, Liu YT, et al. Multitarget inhibition of drug-resistant multiple
myeloma cell lines by dimethyl-celecoxib (DMC), a non-COX-2 inhibitory analog of celecoxib.
Blood. 2005;106:4330-4338.
30.
Springsteen G, Wang B. Alizarin Red S. as a general optical reporter for studying the
binding of boronic acids with carbohydrates. Chem Commun. 2001:1608-1609.
31.
Chow HH, Hakim IA, Vining DR, et al. Effects of dosing condition on the oral
bioavailability of green tea catechins after single-dose administration of Polyphenon E in healthy
individuals. Clin Cancer Res. 2005;11:4627-4633.
Page 26

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
32.
Baatout S, Jacquet P, Derradji H, Ooms D, Michaux A, Mergeay M. Study of the
combined effect of X-irradiation and epigallocatechin-gallate (a tea component) on the growth
inhibition and induction of apoptosis in human cancer cell lines. Oncol Rep. 2004;12:159-167.
33.
Shammas MA, Neri P, Koley H, et al. Specific killing of multiple myeloma cells by (-)-
epigallocatechin-3-gallate extracted from green tea: biologic activity and therapeutic
implications. Blood. 2006;108:2804-2810.
34.
Pyrko P, Kardosh A, Wang W, Xiong W, Schonthal AH, Chen TC. HIV-1 protease
inhibitors nelfinavir and atazanavir induce malignant glioma death by triggering endoplasmic
reticulum stress. Cancer Res. 2007;67:10920-10928.
35.
Duriez PJ, Shah GM. Cleavage of poly(ADP-ribose) polymerase: a sensitive parameter
to study cell death. Biochem Cell Biol. 1997;75:337-349.
36.
Guzy RD, Schumacker PT. Oxygen sensing by mitochondria at complex III: the paradox
of increased reactive oxygen species during hypoxia. Exp Physiol. 2006;91:807-819.
37.
Boeseken J. The use of boric acid for the determination of the configuration of
carbohydrates. Adv Carbohydr Chem. 1949;4:189-210.
38.
Lorand JP, Edwards JO. Polyol complexes and structure of the benzeneboronate ion. J
Org Chem. 1959;24:769-774.
39.
Zhang X, Zhang H, Tighiouart M, et al. Synergistic inhibition of head and neck tumor
growth by green tea (-)-epigallocatechin-3-gallate and EGFR tyrosine kinase inhibitor. Int J
Cancer. 2008;123:1005-1014.
40.
Springsteen G, Wang BA. Detailed examination of boronic acid-diol complexation.
Tetrahedron. 2002;58:5291-5300.
41.
Zou W, Yue P, Lin N, et al. Vitamin C inactivates the proteasome inhibitor PS-341 in
human cancer cells. Clin Cancer Res. 2006;12:273-280.
42.
Liu FT, Agrawal SG, Movasaghi Z, et al. Dietary flavonoids inhibit the anticancer effects
of the proteasome inhibitor bortezomib. Blood. 2008;112:3835-3846.
Page 27

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
43.
Nam S, Smith DM, Dou QP. Ester bond-containing tea polyphenols potently inhibit
proteasome activity in vitro and in vivo. J Biol Chem. 2001;276:13322-13330.
44.
Ling YH, Liebes L, Zou Y, Perez-Soler R. Reactive oxygen species generation and
mitochondrial dysfunction in the apoptotic response to Bortezomib, a novel proteasome
inhibitor, in human H460 non-small cell lung cancer cells. J Biol Chem. 2003;278:33714-33723.
45.
Fernandez Y, Miller TP, Denoyelle C, et al. Chemical blockage of the proteasome
inhibitory function of bortezomib: impact on tumor cell death. J Biol Chem. 2006;281:1107-1118.
46.
Llobet D, Eritja N, Encinas M, et al. Antioxidants block proteasome inhibitor function in
endometrial carcinoma cells. Anticancer Drugs. 2008;19:115-124.
47.
Schönthal A, Feramisco JR. Inhibition of histone H1 kinase expression, retinoblastoma
protein phosphorylation, and cell proliferation by the phosphatase inhibitor okadaic acid.
Oncogene. 1993;8:433-441.
48.
Hideshima T, Chauhan D, Hayashi T, et al. Antitumor activity of lysophosphatidic acid
acyltransferase-beta inhibitors, a novel class of agents, in multiple myeloma. Cancer Res.
2003;63:8428-8436.
Page 28

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
FIGURE LEGENDS
Figure 1: EGCG neutralizes the cytotoxic effects of BZM in multiple myeloma cells.
(A) Chemical structures of BZM and EGCG. (B) Percent cell viability of RPMI/8226 cells was
determined via MTT assay after 48 hours of treatment with 10 nM BZM in the presence or
absence of 10 µM EGCG. (C) Percent apoptotic cells were determined via TUNEL staining of
slides prepared from RPMI/8226 cells treated for 48 hours with 10 nM BZM in the presence or
absence of 10 µM EGCG. (D) Cell cycle distribution of cells treated with 20 nM BZM and/or 20
µM EGCG for 24 hours as determined by flow cytometry analysis.47 Distribution of cells left to
the dotted line (M1) indicates apoptotic or sub-G1/G0 cells. Cells in G1 display 2n DNA content
(x-axis) and those in G2/M display 4n. Y-axis is relative number of cells. (E) Microphotographic
images were taken of RPMI/8226 cells after 24 hours of treatment with 10 nM BZM in the
presence or absence of 10 µM EGCG. (FJ) Percent cell viability of RPMI/8226, U266, and
MM1 cells treated for 48 or 72 hours with increasing concentrations of BZM in the presence of
increasing concentrations of EGCG or TEAVIGO-EGCG as determined via MTT assay. The
percentage survival of non-treated cell cultures was set to 100%; shown is mean ±SE (n4).
Low concentrations of EGCG may increase the viability of a cell culture to >100%; this may be
due to a slight reduction of the high basal level of apoptosis, which is present even in non-
treated cells cultures (see parts C and D). (K) Patient MM cells were purified from bone marrow
aspirates as described48 and analyzed by MTT assay as detailed in (B) above. Shown is the
average of four measurements from two independent tissue samples (mean ±SE).
Figure 2: EGCG neutralizes the cytotoxic effects of BZM in glioblastoma.
(A) Percent cell viability of LN229 cells treated for 48 hours with increasing concentrations of
BZM in the presence of increasing concentrations of EGCG was determined via MTT assay. (B)
U251 and LN229 were treated with 10 nM BZM in the presence or absence of 10 or 20 µM
EGCG for 48 hours and the number of long-term surviving cells (2 weeks) was determined by
Page 29

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
colony formation assays. Images were taken of representative 6-well plates. (C) The average
number of colonies of U251 and LN229 cells treated with BZM in the presence or absence of
EGCG is presented as the percent of colonies from untreated cells, which was set at 100%
(mean ±SE; n3).
Figure 3: Complete green tea extract and its polyphenols EGC, ECG, and EC neutralize
the cytotoxic effects of BZM.
(A) Chemical structures of the green tea polyphenols EGC, ECG and EC. (B) Percent cell
viability of LN229 cells treated for 48 hours with increasing concentrations of BZM in the
presence of increasing concentrations of EGC, ECG, and EC was determined by MTT assay.
(C) Percent cell viability of LN229 and (D) U266 cells treated for 48 hours with increasing
concentrations of BZM in the presence of increasing concentrations of complete green tea
extract (GTE) was determined via MTT assay. In all cases, the average percentage of non-
treated cell cultures was set to 100%; shown is mean ±SE (n4). (E) Typical result of MTT
survival assay of RPMI/8226 cells treated for 48 hours with increasing concentrations of BZM in
the presence of increasing concentrations of complete GTE. Shown are wells from a 96-well
plate, and each condition is shown in quadruplicate. Deep purple color represents fully viable
cell culture; pink(ish) color indicates reduced viability; yellow color indicates absence of viable
cells. The concentration of EGCG noted in CE was calculated based on the percentage of
EGCG as listed on the label of the liquid GTE (see materials and methods).
Figure 4: EGCG neutralizes the cytotoxic effect of MG-262 and proteasome inhibitor IX
but not NFV, MG-132, and proteasome inhibitor I
(A) Chemical structures of the proteasome inhibitors NFV, proteasome inhibitor I (PS-I), MG-
132, MG-262, and proteasome inhibitor IX (PS-IX), and their grouping according to the
presence of a boronic acid. (B-F) LN229 cells were treated with increasing concentrations of the
Page 30

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
individual proteasome inhibitors in the presence or absence of 20 µM EGCG for 48 hours;
thereafter, the number of long-term surviving cells that were able to spawn colonies during the
following 2 weeks (in the absence of drug treatment) was determined by colony formation
assays. Shown is the mean ±SE (n3).
Figure 5: EGCG prevents proteasome inhibition and ER stress induction when combined
with BZM and MG-262, but not MG-132
(A) LN229 cells were cultured for 30 hours with 20 nM BZM, 1.5 µM MG-132, or 20 nM MG-262
in the presence or absence of 20 µM EGCG. Top: Percent proteasome activity in total cell
lysates was determined via 20S proteasome activity assay. Bottom: The relative amount of
polyubiquitinated proteins and the ER stress marker GRP78 were determined via Western blot
analysis with specific antibodies against ubiquitin and GRP78, respectively. Western blot to
actin was used as a loading control. (B) RPMI/8226 cells were cultured for 30 hours with 20 nM
BZM in the presence or absence of 20 µM EGCG. Top: Percent proteasome activity from total
cell lysates was determined via 20S proteasome activity assay. Bottom: The relative amount of
polyubiquitinated proteins, activated caspase-7 (cleaved casp-7), and cleavage of PARP (which
indicates ongoing apoptosis) in RPMI/8226 cells were determined via Western blot analysis with
specific antibodies against ubiquitin, caspase-7, and PARP, respectively. (C) Highly purified
proteasomes were treated with increasing concentrations of BZM in the presence or absence of
5 µM EGCG. Percent activity was determined via 20S proteasome activity assay. Shown is
percent proteasome activity (mean ±SE; n3), where activity in the absence of drug was set to
100%.
Figure 6: EGCG neutralizes the anti-tumoral effects of BZM in multiple myeloma in vivo
Tumor-bearing mice were treated with bortezomib and EGCG individually or in combination, or
remained untreated. 72 hours later, the animals were sacrificed and the tumors were analyzed
Page 31

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
by TUNEL (apoptosis indicator) or H&E (morphology). (A) Representative microphotographic
images were taken of the TUNEL staining (original magnification 400x for all panels).
Arrowheads, representatively selected TUNEL-positive cells. (B) The percentage of TUNEL-
positive cells (reddish brown stain) was determined in 10 randomly chosen microscopic fields
from each treatment group. Columns, mean; bars, SEM. Statistically significant differences in
the extent of tumor cell death between individual and combination drug treatments are indicated
in the chart. (C) Representative microphotographic images were taken of the H&E staining
(original magnification 200x for all panels). Open arrowheads, area of subcapsular necrosis.
Figure 7: BZM does not generate ROS, but directly reacts with EGCG to form a new
boronate product
(A) RPMI/8226 cells were treated with 20 nM BZM, 5 mM H2O2, or exposed to hypoxia for 18
hours in the presence or absence of 20 µM EGCG. Shown is the fold-increase of intracellular
ROS levels (mean ±SE; n3) from one representative experiment; two asterisks (**) indicate
statistically significant (p=0.005) differences between hypoxia-treated or H2O2-treated cells vs.
untreated (Control) cells; n.s.: not statistically significantly different from untreated control. The
entire experiment was repeated in different variations and with different incubation times; yet,
BZM consistently failed to increase the levels of ROS. The generation of intracellular ROS was
measured via labeling with 2',7'-dichlorofluorescein diacetate (H2DCFDA) and subsequent
analysis by flow cytometry.23 In some of the repetitions, total ROS production was determined
by UV spectrometry after incubating cells with H2DCFDA. (B) The direct binding of EGCG to
BZM was investigated with the Alizarin Red S (ARS) reporter assay. In the absence of EGCG,
there is increasing binding between ARS and rising (0150 µM) concentrations of BZM.
However, the addition of EGCG (250 µM) effectively reduces this interaction via direct
competive binding of EGCG to BZM. (C) Results from the ARS assay with MG-132 (65 µM) and
MG-262 (65 µM) in the presence or absence of EGCG (250 µM). Note that MG-132 does not
Page 32

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.
bind to ARS and there is no interference by EGCG, but that MG-262 strongly binds to ARS and
this binding is effectively prevented by EGCG. All results are presented as mean ±SD (arbitrary
units) of relative binding increase. These measurements were also repeated with TEAVIGO-
EGCG and yielded very similar outcomes. (D) 400 MHz 1H NMR spectra (left) and 100 MHz 13C
NMR spectra (right) of combinations of BZM and EGCG in 20% D2O in CD3CN. (a) pure BZM,
(b) pure EGCG, (c) a 2:1 mixture of BZM and EGCG, and (d) a 1:1 mixture of BZM and EGCG.
Selected peaks from the NMR spectra indicate the presence of the new adduct from the
reaction of a 1,2-diol unit of EGCG with the boronic acid group of BZM. By increasing the
amount of EGCG relative to BZM, from 2:1 in (c) to 1:1 in (d), the product peaks (red arrows in
the 1H NMR and red dots in the 13C NMR) incease relative to the BZM peaks (black arrows in
the 1H NMR and black dots in the 13C NMR).
Page 33

From www.bloodjournal.org at American Society of Hematology on February 6, 2009. For personal use only.