Psoralenquinones as a novel class of proteasome inhibitors: Design, synthesis and biological evaluation

Article abstract of DOI:10.1002/cmdc.201100041

Proteasome subunit specificity: Psoralenquinones were identified as a novel class of nonpeptide proteasome inhibitors. Depending on the scaffold decoration, these compounds demonstrate interesting subunit specificity. Interactions with Thr1, Thr21 and Ser129 are critical for inhibition.

Full text of DOI:10.1002/cmdc.201100041

MED
DOI: 10.1002/cmdc.201100041
Psoralenquinones as a Novel Class of Proteasome Inhibitors: Design,
Synthesis and Biological Evaluation
Giovanni Marzaro, Valentina Gandin, Christine Marzano, Adriano Guiotto, and Adriana Chilin*[a]
Regulation of intracellular protein degradation plays a crucial
role in all cellular processes, including cell-cycle regulation, cell
differentiation and inflammation, signal transduction, and cell-
death pathways. In eukaryotic cells, intracellular protein degra-
dation is strictly regulated by the ubiquitin–proteasome path-
way. Proteins intended to be destroyed are marked by cova-
lent bond with a polyubiquitin chain;[1,2] this complex is then
recognized by the 26S proteasome machinery, which then
breaks it down in to smaller peptides. Structurally, 26S protea-
some is a cylinder-shaped multimeric protein complex com-
posed of a 20S proteasome core particle (responsible for cata-
lytic activity) and a 19S component (involved in activity regula-
tion).[3] To date, three proteolytic activities have been recog-
nized in the 20S proteasome depending on the cleavage pref-
erences: chymotripsin-like (CT-L), trypsin-like (T-L) and peptidyl-
glutamyl-peptide-hydrolising or caspase-like (C-L), cleaving
peptides after acidic, basic or hydrophobic amino acid resi-
dues, respectively.[4–6] All of these activities are located in well-
defined proteasome cavities and are mediated by the N-termi-
nal threonine residue (Thr1) of each catalytic pocket.[7] CT-L
sites, the primary target of most proteasome inhibitors, have
long been considered the most important for protein degrada-
tion. The biological roles of proteasome proteolitic sites and
their potential as targets of antineoplastic agents are still not
well defined. Recent work indicated that cytotoxicity of protea-
some inhibitors barely correlates with exclusive CT-L site inhibi-
tion and often requires co-inhibition of other sites.[8,9]
bortezomib (3),[18,19] and TMC-95A (4)[20] have been shown to
be useful proteasome inhibitors.
All of these compounds show different selectivity profiles
and, with the exception of TMC-95A, covalently bind the N-ter-
minal threonine residues.[7] Bortezomib (3) has received appro-
val from the US Food and Drug Administration (FDA) for use in
the treatment of advanced multiple myeloma, confirming the
proteasome as a valid target for the development of anticanc-
er agents. However, bortezomib has shown problems of ad-
verse side effects[21] and drug resistance.[14] Another disadvant-
age of all peptidomimetic inhibitors is their instability due to
the presence of functional groups prone to nucleophilic attack.
In an attempt to overcome this drawback, several non-pepti-
domimic inhibitors have recently been identified through virtu-
al screening processes.[22] Aiming to identify other, more “drug-
like” compounds, we present here the identification, synthesis
and biological evaluation of a novel class of non-peptidomimic
proteasome inhibitors derived from the psoralenquinone scaf-
fold.
Deregulation of proteasome pathways seems to be related
to inflammatory diseases,[10] cancer onset and progression,[11]
and drug resistance.[12] In particular, because their cell cycle
checkpoints are disordered, cancer cells are more closely de-
pendent on proteasome-mediated degradation of cell cycle
regulators.[13] Furthermore, the increased requirement for pro-
tein synthesis in proliferating cells makes cancer cells more vul-
nerable to proteasome inhibition.[14] Since the proteasome has
emerged as a promising molecular target for new cancer ther-
apeutics, greater attention has been paid to the identification
of proteasome inhibitors. In particular, several peptidomimic
derivatives, such as lactacystin (1),[15,16] calpain inhibitor I (2),[17]
Coumarin and its furan derivatives, psoralens and angelicins,
are attractive scaffolds for the design of novel compounds
with biological activity. For example, a number of compounds
have been synthesized and evaluated for their ability to bind
DNA or to inhibit CK2, topoisomerases, and NF-kB.[23–26] The
widespread usefulness of such scaffolds encouraged our re-
search group to find other suitable targets for psoralene deriv-
atives, paying particular interest to novel cancer-related tar-
gets, such as proteasome, due also to the possible correlation
between NF-kB, proteasome and topoisomerases.[26]
[a] Dr. G. Marzaro,+ Dr. V. Gandin,+ Dr. C. Marzano, Prof. A. Guiotto,
Prof. A. Chilin
Department of Pharmaceutical Sciences, University of Padova
Via Marzolo 5, 35131 Padova (Italy)
Fax: (+39)049-827-5366
E-mail: adriana.chilin@unipd.it
[+] These authors contributed equally to this work and should be considered
In order to evaluate whether psoralenquinone could be a
potential scaffold for novel proteasome inhibitors, several
docking studies have been performed. At this stage, the struc-
co-first authors.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201100041.
996
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 996 – 1000

ture of TMC-95A (4) bound in the CT-L pocket was selected as
a template.[27] The lowest root mean square deviation (RMSD)
value between the crystal structure and the docked structure
was reached employing Molegro software[28] and refining the
pose with the most favorable hydrogen-bonding energy
through the LigX application in the Molecular Operating Envi-
ronment (MOE) software (see Supporting Information for a de-
tailed description of the docking protocols used).[29] This two-
step docking approach has been used to virtually evaluate the
interactions of a series of synthetically accessible psoralenqui-
none derivatives (5–11) in each of the catalytic pocket of 20S
proteasome.
Scheme 1. Synthesis of compound 5. Reagents and conditions: a) Pyridinium
chloride, MW, 2008C, 7 min; b) CrO3, AcOH, 5 min.
ganic synthesis (MAOS) techniques in almost all steps
(Scheme 2). Benzopyran-2-ones 14a–b were condensed with
the appropriate a-chloroketone; this step was investigated to
both reduce reaction time and enhance yield. Taking advant-
age of MAOS, several reaction media and acid scavengers were
evaluated for the synthesis of 15, chosen as a representative
compound (Table 2).
Table 2. Effects of bases and solvents on etherification of compound
13.[a]
Entry
Solvent
Base
Time [min]
Yield [%][b]
1
2
3
4
acetone
acetone
H2O
K2CO3
TEA
K2CO3
TEA
30
30
30
20
65
40
0
H2O
95
[a] Reaction conditions: 7-hydroxy-8-methoxy-4-methylbenzopyran-2-one
(1.0 mmol) and chloroacetone (1.5 mmol) in the appropriate solvent
(0.25 mL) were irradiated under microwave conditions at 1308C in the
presence of the appropriate base (2.0 mmol). [b] Yield of isolated prod-
uct.
The best yield of 15 was achieved in water/triethylamine
(1:1) at 1308C for 20 min under microwave (MW) irradiation.
The water/triethylamine/MW protocol was also employed for
the synthesis of compounds 18a and 18b. In all cases, the
subsequent cyclization to the psoralenquinone structure was
also performed in a very short reaction time through MW irra-
diation in alkali medium. Subsequent demethylation followed
by oxidation with chromium oxide yielded the desired com-
pounds. Benzopsoralenquinone 9 was synthesized from com-
pound 20a as previously reported, as well as compound 10.[30]
Compound 11 was accessed from commercially available 4,9-
dimethoxy-7-methylfuro[3,2-g]chromen-5-one (khellin) through
oxidative demethylation in dilute nitric acid.[30]
As reported in Table 1, almost all of the designed com-
pounds showed a favorable interaction energy against each
catalytic pocket. All the designed compounds were synthe-
sized through very similar synthetic routes, choosing different
starting materials, depending on the scaffold substitution.
Compound 5 was prepared from 8-methoxypsoralen (8-MOP;
12) through demethylation followed by oxidation of com-
pound 13 with chromium oxide (Scheme 1). Compounds 6–8
were synthesized by a reported procedure,[23] with slight but
useful modifications, such as adopting microwave-assisted or-
Table 1. Predicted interaction energies of psoralenquinone 5–10 in the
different 20S proteasome catalytic sites.[a]
To examine the ability of the newly synthesized psoralenqui-
none derivatives to inhibit proteasome activity, we tested the
functioning of each individual active site of 26S proteasome
purified from human ovarian 2008 cancer cells. Bortezomib (3)
was used as a reference drug under the same experimental
conditions. The results, summarized in Table 3, show that com-
pounds 8 and 9 were completely ineffective at inhibiting any
of the three catalytic proteasome activities (IC50 >50 mm for all
derivatives). While their potencies were lower than that ob-
served for bortezomib, compounds 5 and 6 efficiently inhibit-
ed human proteasome C-L activity (IC50 =40.23 and 12.09 mm
for 5 and 6, respectively), but they were essentially inactive
against CT-L and T-L activities (IC50 >50 mm). The C-L selectivity
Compd
MM/GBVI[b] [KcalmolÀ1]
C-L
CT-L
T-L
5
6
7
8
9
10
À10.50
À9.16
À9.79
À11.40
À6.40
À10.22
À3.70
À10.76
À0.20
À14.99
À5.86
À10.34
À12.55
À5.28
À12.35
À9.24
À15.38
À14.95
[a] 20S proteasome activity: chymotripsin-like (CT-L), trypsin-like (T-L), cas-
pase-like (C-L). [b] Molecular mechanics generalized Born interaction
(MM/GBVI) energy calculated using the LigX protocol (full details given in
the Supporting Information).
ChemMedChem 2011, 6, 996 – 1000
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
997

MED
of bortezomib was evaluated under the same experi-
mental conditions. IC50 values, calculated from dose–
survival curves, are shown in Table 4. Compounds 5,
7 and 9 exhibited moderate cytotoxicity against all
five cell lines, with average IC50 values of 29.5, 22.2
and 33.5 mm, respectively, whereas compounds 6, 8
and 10 showed the greatest in vitro antitumor activi-
ty, with average IC50 values of approximately 13 mm
for all three quinolones. Among all cancer cell lines,
2008 cells appeared to be the most sensitive to deriv-
atives 6, 8 and 10.
Taking the cytotoxicity and in vitro enzymatic in-
hibition results together, compounds 6 and 10 ap-
peared as the most promising antiproliferative
agents acting as proteasome inhibitors.
In order to further confirm their ability to inhibit
26S proteasome, 2008 cells were treated with in-
creasing concentrations of compounds 6 and 10 for
24 h and then submitted to CT-L and C-L activity
measurements. Concerning compound 6, at a con-
centration of 30 mm, the proteasomal C-L activity was
decreased by 50%; the IC50 value for CT-L activity cal-
culated in 2008 cells treated with compound 10 was
34 mm (Figure 1a). These data confirmed the results
obtained in the earlier experiments.
Scheme 2. Synthesis of compounds 6–8. Reagents and conditions: a) 2-Chloroacetone or
2-chlorocyclohexanone, H2O/Et3N (1:1), MW, 1308C, 20 min; b) NaOH (1m), propan-2-ol,
MW, 1208C, 5 min; c) pyridinium chloride, MW, 2008C, 7 min; d) CrO3, AcOH, 5 min.
To estimate intracellular polyubiquitinated protein
levels, 2008 cells treated with IC50 concentrations of
compound 6 or 10 were subjected to immunofluor-
escence analyses, which revealed an abnormal pres-
Table 3. In vitro inhibition of proteasome activities.[a]
Compd
IC50 [mM]
CT-L[a]
>50
T-L[b]
C-L[c]
5
6
7
8
9
10
11
>50
>50
40.23Æ2.00
12.09Æ1.73
>50
31.39Æ0.91
29.23Æ0.97
>50
>50
>50
>50
>50
>50
>50
>50
>50
12.33Æ1.46
>50
0.01Æ0.01
38.13Æ0.97
>50
>50
Bortezomib (3)
3.23Æ1.65
[a] Each proteasome activity was measured by the cleavage of the fluo-
rescent peptides: Boc-Gln-Ala-Arg-AMC, Z-Leu-Leu-Glu-AMC or succinyl-
Leu-Leu-Val-Tyr-AMC, used to reveal the trypsin-like (T-L), caspase-like (C-
L) and chymotrypsin-like (CT-L) activities, respectively. Values represent
the mean ÆSD of three independent experiments.
demonstrated by these derivatives is in contrast with that of
bortezomib, which, as is well known, preferentially inhibits the
CT-L activity and, to a lesser extent, C-L activity.[31] Compounds
7 and 10, besides mainly inhibiting CT-L activity (IC50 =31.39
and 12.33 mm for 7 and 10, respectively), have been found to
inhibit T-L activity more efficiently than bortezomib.
Psoralenquinone derivatives were also tested for their cyto-
toxic activity toward a panel of five human tumor cell lines, in-
cluding examples of ovarian (2008 and C13*), cervical (HeLa),
colon (HCT-15) and breast (MCF-7) cancer. Cytotoxicity was
evaluated by means of an MTT assay after 72 h treatment with
increasing concentrations of the test compounds. Cytotoxicity
Figure 1. Inhibition of proteasome activity in 2008 cells. a) Inhibition of C-L
and CT-L activities of 26S proteasome in 2008 cells treated with increasing
concentrations of compound 6 or 10 estimated by means of specific fluoro-
genic substrates. IC50 values calculated by probit analysis (p<0.05, c2 test).
b) Detection of polyubiquitinated proteins in 2008 cells following 24 h treat-
ment with increasing concentrations of compound 6 or 10 by means of im-
munofluorescence analysis.
998
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 996 – 1000

some catalytic activity, thus
strengthening the initial hypoth-
esis (Table 3).
Table 4. Cytotoxicity of psoralenquinone 5–11 against five different cancer cell lines.
Compd
IC50 [mM][a]
2008
C13*
MCF-7
HeLa
LoVo
Moreover, on the basis of the
in vitro inhibitory activities
5
6
7
8
9
10
11
28.81Æ1.43
7.92Æ1.21
19.43Æ2.18
9.34Æ1.07
28.32Æ3.21
9.54Æ1.33
61.44Æ2.77
0.24Æ0.11
31.48Æ2.42
15.12Æ1.37
18.74Æ1.65
11.22Æ1.54
31.99Æ2.36
10.13Æ1.23
65.12Æ1.32
nd[b]
30.48Æ1.43
21.23Æ2.13
24.23Æ0.43
16.06Æ3.35
28.43Æ2.34
12.22Æ1.42
64.32Æ1.71
nd[b]
31.32Æ1.26
6.33Æ2.45
29.12Æ1.32
12.65Æ1.23
37.76Æ2.33
15.43Æ1.21
60.60Æ3.13
0.21Æ0.07
25.14Æ1.80
13.66Æ1.81
19.43Æ1.55
13.23Æ1.72
41.21Æ2.08
17.46Æ2.23
64.40Æ2.42
0.09Æ0.03
(Table 3),
a preliminary struc-
ture–activity relationship (SAR)
has been proposed. The inhibi-
tion of C-L activity requires no
substitution at the R4 position
(see Figure 2 for substituent po-
sitions), while CT-L and T-L inhib-
itions require bulky substituted
furans. The presence of a hy-
droxymethyl group at R1 instead
of a methyl group enhances the
Bortezomib (3)
[a] Values represent the mean ÆSD of three independent experiments. IC50 values were calculated by probit
analysis (p<0.05, c2 test). Cells (3–5ꢁ104 mLÀ1) were treated for 72 h with increasing concentrations of the test
compounds. Cytotoxicity was assessed by MTT assay. [b] nd=not determined.
ence of proteins polyconjugated to ubiquitin. These results,
presented in Figure 1b, clearly confirm that these psoralenqui-
nones are able to inhibit proteasome activities in intact cells,
thus inducing impairment of the ubiquitin–proteasome degra-
dation machinery.
CT-L inhibition, while it does not affect the T-L inhibition (cf. 7
and 10). Finally, the presence of a substituent at the R2 posi-
tion abolishes the inhibitory activity (compound 8). All of the
predicted binding poses have been found to be in agreement
with these considerations, and have been used to explain the
structural basis of this SAR (see Supporting Information for il-
lustrative figures).
Taken together, these data suggest that the ability of these
two newly synthesized psoralenquinones (6 and 10) to inhibit
the 26S proteasome activity may be responsible for their anti-
proliferative activity in tumor cells.
In conclusion, a novel class of non-peptidomimetic protea-
some inhibitors containing the psoralenquinone core structure
has been discovered through docking studies. A plausible
binding mode has been proposed, and supported by analysis
of structure–activity relationship data based on inhibitory activ-
ities and visual inspection of the docked structures. Interesting-
ly, two of the synthesized compounds exhibited selective in-
hibition of C-L activity, for which no effective inhibitors have
been developed to date. Furthermore, psoralenquinones inhib-
iting CT-L activity also induced significant inhibition of T-L ac-
tivity, which was more pronounced than that exhibited by bor-
tezomib. Recently, it has been shown that CT-L inhibition re-
duced degradation of unfolded proteins by a maximum of
50%, and a significant inhibition of protein breakdown can be
achieved only when the CT sites and either the T-L or C-L sites
are also inhibited.[33] Hence, the development of new protea-
some inhibitors acting on multiple sites is required to more
specifically target cancer cells. Moreover, to the best of our
knowledge, psoralenquinone is the first class of compounds
for which a selective inhibitory profile has been identified de-
pending on scaffold decoration. The synthesis of other ana-
logues and further investigations on the binding mode, along
with the understanding of biological and molecular pathways
involved in cancer cell death mechanism, are in progress.
While the preliminary docking studies predicted the ability
of these derivatives to inhibit all three catalytic activities of
proteasome, the biological evaluation revealed a selectivity
profile. However, visual inspection of the predicted interaction
poses revealed that all of the active compounds share the
same proposed binding mode (Figure 2), which is substantially
different to those of the inactive derivatives. Thus, the interac-
tion with Ser129, Thr1 and Thr21 seems to be fundamental
for the inhibition of proteasome activity by psoralenquinones.
It is noteworthy that the proposed binding mode described
here (Figure 2) is quite similar to the recently reported binding
mode for other benzoquinone derivatives.[32]
To further demonstrate our observations, we synthesized fu-
rochromone 11, which cannot form any hydrogen bonds with
Ser129, since the carbonyl group of the pyrone ring is in the g-
position rather than the a-position. The biological evaluation
on compound 11 revealed its inability to inhibit any protea-
Experimental Section
Experimental details on computational methodologies, chemical
synthesis including analytical details (mp, IR, NMR, HRMS, elemen-
tal analyses) for all compounds, and biological assays, as well as il-
lustrations of the docking studies performed, are available as Sup-
porting Information on the WWW under http://dx.doi.org/10.1002/
cmdc.201100041.
Figure 2. Proposed binding mode for the psoralenquinone derivatives.
ChemMedChem 2011, 6, 996 – 1000
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
999

MED
[19] R. C. Kane, A. T. Farrell, R. Sridhara, R. Pazdur, Clin. Cancer Res. 2006, 12,
2955.
[20] J. Kohno, Y. Koguchi, M. Nishio, K. Nakao, M. Kuroda, R. Shimizu, T.
Ohnuki, S. Komatsubara, J. Org. Chem. 2000, 65, 990.
Keywords: cytotoxicity · molecular docking · proteasome
activity · psoralenquinones
[21] E. Terpos, M. Roussou, M. A. Dimopoulos, Expert Opin. Drug Metab. Toxi-
col. 2008, 4, 639.
[1] A. Hershko, A. Ciechanover, Annu. Rev. Biochem. 1998, 67, 425.
[2] M. Hochstrasser, Nature 2009, 458, 422.
[22] N. Basse, M. Montes, X. Marechal, L. Qin, M. Bouvier-Durand, E. Genin, J.
Vidal, B. O. Villoutreix, M. Reboud-Ravaux, J. Med. Chem. 2010, 53, 509.
[23] O. Gia, A. Anselmo, A. Pozzan, C. Antonello, S. M. Magno, E. Uriarte,
Farmaco 1997, 52, 389.
[24] A. Chilin, R. Battistutta, A. Bortolato, G. Cozza, S. Zanatta, G. Poletto, M.
Mazzorana, G. Zagotto, E. Uriarte, A. Guiotto, L. A. Pinna, F. Meggio, S.
Moro, J. Med. Chem. 2008, 51, 752.
[25] L. Piccagli, M. Borgatti, E. Nicolis, N. Bianchi, I. Mancini, I. Lampronti, D.
Vevaldi, F. Dall’Acqua, G. Cabrini, R. Gambari, Bioorg. Med. Chem. 2010,
18, 8341.
[26] R. Ganapathi, S. A. Vaziri, M. Tabata, N. Takigawa, D. R. Grabowski, R. M.
Bukowski, M. K. Ganapathi, Curr. Pharm. Des. 2002, 8, 1945.
[27] M. Groll, Y. Koguchi, R. Huber, J. Kohno, J. Mol. Biol. 2001, 311, 543.
[28] R. Thomsen, M. H. Christensen, J. Med. Chem. 2006, 49, 3315.
[29] The Molecular Operating Environment (MOE), version 2008.10, The
Chemical Computing Group Inc., (Montreal, Canada); http://www.chem-
corp.com.
[3] O. Coux, K. Tanaka, A. L. Goldberg, Annu. Rev. Biochem. 1996, 65, 801.
[4] M. Groll, L. Ditzel, J. Lowe, D. Stock, M. Bochtler, H. D. Bartunik, R.
Huber, Nature 1997, 386, 463.
[5] M. Groll, W. Heinemeyer, S. Jager, T. Ullrich, M. Bochtler, D. H. Wolf, R.
Huber, Proc. Natl. Acad. Sci. USA 1999, 96, 10976.
[6] W. Heinemeyer, M. Fischer, T. Krimmer, U. Stachon, D. H. Wolf, J. Biol.
Chem. 1997, 272, 25200.
[7] M. Groll, R. Huber, Biochim. Biophys. Acta 2004, 1695, 33.
[8] M. Screen, M. Britton, S. L. Downey, M. Verdoes, M. J. Voges, A. E. Blom,
P. P. Geurink, M. D. Risseeuw, B. I. Florea, W. A. van der Linden, A. A. Plet-
nev, H. S. Overkleeft, A. F. Kisselev, J. Biol. Chem. 2010, 285, 40125.
[9] M. Britton, M. M. Lucas, S. L. Downey, M. Screen, A. A. Pletnev, M. Ver-
does, R. A. Tokhunts, O. Amir, A. L. Goddard, P. M. Pelphrey, D. L. Wright,
H. S. Overkleeft, A. F. Kisselev, Chem. Biol. 2009, 16, 1278.
[10] N. Qureshi, S. N. Vogel, C. Van Way, 3rd, C. J. Papasian, A. A. Qureshi,
D. C. Morrison, Immunol. Res. 2005, 31, 243.
[11] R. Z. Orlowski, D. J. Kuhn, Clin. Cancer Res. 2008, 14, 1649.
[12] T. Hideshima, P. Richardson, D. Chauhan, V. J. Palombella, P. J. Elliott, J.
Adams, K. C. Anderson, Cancer Res. 2001, 61, 3071.
[30] A. Chilin, G. Pastorini, A. Castellin, F. Bordin, P. Rodighiero, A. Guiotto,
Synthesis 1995, 1190.
[31] M. J. Williamson, J. L. Blank, F. J. Bruzzese, Y. Cao, J. S. Daniels, L. R. Dick,
J. Labutti, A. M. Mazzola, A. D. Patil, C. L. Reimer, M. S. Solomon, M. Stir-
ling, Y. Tian, C. A. Tsu, G. S. Weatherhead, J. X. Zhang, M. Rolfe, Mol.
Cancer Ther. 2006, 5, 3052.
[32] H. R. Lawrence, A. Kazi, Y. Luo, R. Kendig, Y. Ge, S. Jain, K. Daniel, D. San-
tiago, W. C. Guida, S. M. Sebti, Bioorg. Med. Chem. 2010, 18, 5576.
[33] A. F. Kisselev, A. Callard, A. L. Goldberg, J. Biol. Chem. 2006, 281, 8582.
[13] G. I. Evan, L. Brown, M. Whyte, E. Harrington, Curr. Opin. Cell Biol. 1995,
7, 825.
[14] D. J. McConkey, K. Zhu, Drug Resist. Updat. 2008, 11, 164.
[15] S. Omura, T. Fujimoto, K. Otoguro, K. Matsuzaki, R. Moriguchi, H. Tanaka,
Y. Sasaki, J. Antibiot. 1991, 44, 113.
[16] G. Fenteany, R. F. Standaert, W. S. Lane, S. Choi, E. J. Corey, S. L. Schreib-
er, Science 1995, 268, 726.
[17] A. Vinitsky, C. Cardozo, L. Sepp-Lorenzino, C. Michaud, M. Orlowski, J.
Biol. Chem. 1994, 269, 29860.
[18] B. A. Teicher, G. Ara, R. Herbst, V. J. Palombella, J. Adams, Clin. Cancer
Res. 1999, 5, 2638.
Received: January 31, 2011
Revised: March 11, 2011
Published online on April 6, 2011
1000
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 996 – 1000