Heterocyclic substituted cantharidin and norcantharidin analogues-synthesis, protein phosphatase (1 and 2A) inhibition, and anti-cancer activity

Article abstract of DOI:10.1016/j.bmcl.2007.03.093

Norcantharidin (3) is a potent PP1 (IC₅₀ = 9.0 ± 1.4 μM) and PP2A (IC₅₀ = 3.0 ± 0.4 μM) inhibitor with 3-fold PP2A selectivity and induces growth inhibition (GI₅₀ ～45 μM) across a range of human cancer cell lines including those of colorectal (HT29, SW480), breast (MCF-7), ovarian (A2780), lung (H460), skin (A431), prostate (DU145), neuroblastoma (BE2-C), and glioblastoma (SJ-G2) origin. Until now limited modifications to the parent compound have been tolerated. Surprisingly, simple heterocyclic half-acid norcantharidin analogues are more active than the original lead compound, with the morphilino-substituted (9) being a more potent (IC₅₀ = 2.8 ± 0.10 μM) and selective (4.6-fold) PP2A inhibitor with greater in vitro cytotoxicity (GI₅₀ ～9.6 μM) relative to norcantharidin. The analogous thiomorpholine-substituted (10) displays increased PP1 inhibition (IC₅₀ = 3.2 ± 0 μM) and reduced PP2A inhibition (IC₅₀ = 5.1 ± 0.41 μM), to norcantharidin. Synthesis of the analogous cantharidin analogue (19) with incorporation of the amine nitrogen into the heterocycle further increases PP1 (IC₅₀ = 5.9 ± 2.2 μM) and PP2A (IC₅₀ = 0.79 ± 0.1 μM) inhibition and cell cytotoxicity (GI₅₀ ～3.3 μM). These analogues represent the most potent cantharidin analogues thus reported. Crown Copyright

Full text of DOI:10.1016/j.bmcl.2007.03.093

Bioorganic & Medicinal Chemistry Letters 17 (2007) 3392–3397
Heterocyclic substituted cantharidin and norcantharidin
analogues—synthesis, protein phosphatase (1 and 2A)
inhibition, and anti-cancer activity
Timothy A. Hill,a Scott G. Stewart,a Benjamin Sauer,a Jayne Gilbert,b
Stephen P. Ackland,b Jennette A. Sakoffb and Adam McCluskeya,*
aDepartment of Chemistry, School of Environmental and Life Sciences, The University of Newcastle, Callaghan, NSW 2308, Australia
bDepartment of Medical Oncology, Newcastle Mater Misericordiae Hospital, NSW 2298, Australia
Received 10 February 2007; revised 21 March 2007; accepted 29 March 2007
Available online 2 April 2007
Abstract—Norcantharidin (3) is a potent PP1 (IC50 = 9.0 1.4 lM) and PP2A (IC50 = 3.0 0.4 lM) inhibitor with 3-fold PP2A
selectivity and induces growth inhibition (GI50 ꢀ45 lM) across a range of human cancer cell lines including those of colorectal
(HT29, SW480), breast (MCF-7), ovarian (A2780), lung (H460), skin (A431), prostate (DU145), neuroblastoma (BE2-C), and glio-
blastoma (SJ-G2) origin. Until now limited modifications to the parent compound have been tolerated. Surprisingly, simple hetero-
cyclic half-acid norcantharidin analogues are more active than the original lead compound, with the morphilino-substituted (9)
being a more potent (IC50 = 2.8 0.10 lM) and selective (4.6-fold) PP2A inhibitor with greater in vitro cytotoxicity (GI50
ꢀ9.6 lM) relative to norcantharidin. The analogous thiomorpholine-substituted (10) displays increased PP1 inhibition
(IC50 = 3.2 0 lM) and reduced PP2A inhibition (IC50 = 5.1 0.41 lM), to norcantharidin. Synthesis of the analogous cantharidin
analogue (19) with incorporation of the amine nitrogen into the heterocycle further increases PP1 (IC50 = 5.9 2.2 lM) and PP2A
(IC50 = 0.79 0.1 lM) inhibition and cell cytotoxicity (GI50 ꢀ3.3 lM). These analogues represent the most potent cantharidin ana-
logues thus reported.
Crown Copyright Ó 2007 Published by Elsevier Ltd. All rights reserved.
Protein phosphorylation and dephosphorylation is a
fundamental process for regulating cellular signaling
pathways and is mediated by a delicate interplay be-
tween kinases and phosphatases. The protein phospha-
tases 1 (PP1) and 2A (PP2A) comprise ꢀ1% of all
cellular proteins and account for 90% of all serine/thre-
onine protein phosphatase activity.1–4 PP2A has a het-
erotrimeric architecture with the core PP2A enzyme
consisting of a 65-kDa (A)-subunit and a 36-kDa cata-
lytic (C) subunit. This AC core enzyme binds to at least
18 different regulatory subunits (B) which are implicated
in regulating cellular localization, substrate specificity,
and enzymatic activity.5–9 Evidence is accumulating that
PP2A is a pivotal mediator of many, if not all, cellular
processes.10–13
There are numerous natural product inhibitors of PP1
and PP2A as exemplified by the okadaic acid class of
compounds. Okadaic acid, tautomycin, calyculin A,
and the microcystins are highly complex nanomolar
potent inhibitors of both PP1 and PP2A, typically dis-
playing modest levels of PP2A selectivity (ꢀ100-fold at
best), with tautomycin the exception being ꢀ4-fold
PP1 selective.3,4,14 Fostriecin (1), isolated from a bacte-
rial broth, is the most selective PP2A (vs PP1) inhibitor
yet reported (ꢀ40,000-fold PP2A selective).15,16 Interest-
ingly fostriecin has undergone considerable evaluation
as a potential anti-cancer agent, and as yet no successful
outcome has been reported, however most of these stud-
ies were compromised by poor compound supply, qual-
ity, and/or rapid in vivo deactivation. Cantharidin (2),
the active component of Spanish Fly, is another toxin
that inhibits both PP1 and PP2A albeit at higher con-
centrations.3 Like fostriecin, cantharidin has been eval-
uated and indeed utilized as an anti-cancer agent.
Mylabris, a natural source of cantharidin, was first
reported as an anti-cancer agent in 1264.3
Keywords: Cantharidin; Norcantharidin; Small molecule protein phos-
phatase inhibitors; PP1; PP2A; Anti-cancer.
*
Corresponding author. Tel.: +61 249216486; fax: +61 249215742;
e-mail: Adam.McCluskey@newcastle.edu.au
0960-894X/$ - see front matter Crown Copyright Ó 2007 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.03.093

T. A. Hill et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3392–3397
3393
tion of a second pyrrolidine (piperidine) nitrogen
adversely affects interactions in the hydrophobic domain
or (2) these analogues do not bind in the proximity of
the hydrophobic domain. In previous work, we devel-
oped a series of analogues that seemed to suggest that
accessing the hydrophobic domain was possible.18c
However, a report by Shan on the evaluation of a family
of norcantharidin analogues and enzyme kinetic experi-
ments supplied tantalizing evidence that cantharidin
may not bind in the same site as the okadaic acid class
of toxins.19 Our own enzymatic kinetic studies have been
ambiguous. We are in the process of completing a series
of synthesis and computer modeling studies in this area
and will report our findings shortly.
ONa
OH
O
P
OH
O
O
OH
O
O
O
O
O
OH
1
2
Structurally the simplest member of the okadaic acid
class of compounds, cantharidin, has been the subject
of multiple medicinal chemistry studies in efforts to im-
prove its PP1 and PP2A potency and to induce either
PP1 or PP2A selectivity.3,4,17,18 Our group has devel-
oped many such analogues and explored the correlation
between PP2A selectivity and in vitro anti-cancer activ-
ity. However we, as have others, have been frustrated by
the apparent inability to improve potency beyond that
of the native compound.17,18 In this paper, we report
on our recent efforts that have been somewhat more
successful.
Piperidine analogue 8 is by far the most intriguing entry
in Table 1, with the removal of the basic nitrogen atom
and drawing the hydrophobic entity closer to the amide
linkage we observed highly potent PP inhibitor, more
potent even than norcantharidin against PP1 and equi-
potent at PP2A, a first in our laboratory. In addition
to a return to PP inhibition we also observed an increase
in cytotoxicity across all cell lines evaluated, again at
least comparable or better than that observed for
norcantharidin.
In keeping with our philosophy that medicinal chemistry
should adopt elegantly simple synthetic approaches
where possible, and in an extension of earlier work,
our analogue development commenced with the readily
accessible norcantharidin (3).18a Treatment, Scheme 1,
of this active anhydride with a series of substituted
heterocyclic amines in THF at room temperature rap-
idly furnished us with the family of analogues shown
in Table 1. For comparative purposes cantharidin and
norcantharidin were also included in all assays
conducted.18a
Having established that the presence of the piperidine
substituent conferred highly favorable PP inhibition
and cytotoxicity, we extended our evaluations to include
a number of similar small heterocyclic amines. These
data are presented in Table 2.
Our synthesis and evaluation commenced with the
cyclopentylamine analogue 5.18a As shown in Table 1,
5 displayed modest levels of PP1 and PP2A inhibition,
with no enzyme selectivity, which is contrary to the par-
ent compounds which show modest levels of PP2A selec-
tivity. Indeed, 5 displayed comparable activity to
previously reported anilino derivatives from our
group,18a suggesting that the aromatic ring at this junc-
ture is not necessary for inhibition, but may play a role
in imparting PP2A selectivity. A moderate level of broad
spectrum cytotoxicity was also observed (GI50 ꢀ36 lM).
Commencing with a series of simple bioisosteric
replacements: 8 (40-CH2) ) 9 (40-O) ) 10 (40-S) ) 11
(40-NH) ) 12 (40-N-CH3) it is immediately apparent
that the 40-heterocyclic ring substituent has a pro-
nounced effect on PP inhibition and cytotoxicity. First
the introduction of the morphilino oxygen is detrimen-
tal to PP1 inhibition whilst improving PP2A inhibition
and hence PP2A specificity, with 9 now 4.6-fold PP2A
selective. Gratifyingly, given our contention that PP2A
specificity and potency correlates with cytotoxicity, we
observed across the board improvement in GI50 values
relative to
8 (8 mean GI50 = 14.5 lM; 9 mean
Current literature consensus is that cantharidin and its
analogues bind in the surface grooves of both PP1 and
PP2A, in the same area as the larger toxins, and several
groups including ours have made efforts to access these
hydrophobic domains to improve potency and specific-
ity.3,4,17,18 This was the rationale for examining the
chain extended analogues 6 and 7. No activity, either
phosphatase inhibition or growth inhibition, was
observed (Table 1). There are two logical explanations
for this unanticipated loss of activity: (1) the introduc-
GI50 = 9.6 lM). Bioisosteric replacement O ) S gener-
ates 10, which adversely affects PP2A inhibition whilst
positively affecting PP1 inhibition with 10 now PP1
selective (1.6-fold), and equipotent with 8, i.e., the
change from 40-CH2 to 40-S has no real effect on PP
inhibition, the effect on cytotoxicity is also negligible.
Given the change in PP2A selectivity we had expected
a decrease in cytotoxicity. Analogue 11 arises via
installation of a 40-NH and allows a return to PP2A
specificity (3.4-fold), although at the expense of po-
tency, and a modest level of cytotoxicity. Presumably
the decrease in cytotoxicity is a combination of reduced
PP inhibition and the bioavailability of 11. The former
effect could be due to a change in H-bonding character;
the change in heterocycle polarity based on an O ) N
replacement, our data do not distinguish between these
possibilities. N-methylation of 11 to 12 renders the
analogue insoluble in assay media as such no data were
obtained.
O
O
R
i
N
O
O
H
O
O
O
OH
3
4
Scheme 1. Reagents and condition: (i) RNH2, THF, rt.

Table 1. Inhibition of PP1 and PP2A, and growth inhibition against a number of human cancer cell lines by 2, 3, and the acid–amide analogues 5–8
a
a
R
IC50 (lM)
GI50 (lM)
PP1
PP2A
HT29b
SW480b
MCF-7c
A2780d
H460e
A431f
DU145g
BE2-Ch
SJ-G2i
2
3
11 2.0
9.0 1.4
1.2 0.1
3.0 0.4
3.3 0.1
48 4.0
4.6 0.2
35 1.1
7.0 0.2
64 3.4
5.8 0.2
59 0.0
4.0 0.1
45 1.8
3.4 0.2
33 0.8
2.4 0.1
30 1.1
4.9 0.2
56 1.8
3.0 0.1
35 1.1
H
N
5
18 1.5
17 1.0
44 3.0
(3.9 3.0)
(13 4.7)
27 4.5
(2.4 3.1)
(5.1 8.3)
48 2.5
(5.9 6.6)
(13 1.8)
56 1.0
(8.4 2.8)
(11 0.6)
29 4.3
(7.7 4.8)
(1.2 7.7)
32 2.0
(6.3 2.8)
(4.8 3.4)
33 4.0
31 2.0
31 1.5
(5.7 3.3)
(0)
O
O
N
6
(0)
(0)
(0)
(0)
N
H
O
7
(0)
(0)
(0)
(2.6 1.4)
N
N
H
N
8
3.0 0.49
4.4 0.18
14 1.2
9.3 0.40
14 0.9
12 1.0
19 1.7
14 0.2
16 0.3
14 0.3
18 1.8
O
Values in parentheses represent percentage growth inhibition generated at 100 lM drug concentration with IC50 or GI50 values not determined.
a Means SEM of three experiments conducted in triplicate.
b Colorectal.
c Breast.
d Ovarian.
e Lung.
f Skin.
g Prostate.
h Neuroblastoma.
i Glioblastoma human cancer cell lines.

Table 2. Inhibition of PP1 and PP2A, and growth inhibition against a number of human cancer cell lines by acid–amide analogues 9–18
a
a
R
IC50 (lM)
GI50 (lM)
PP1
13
PP2A
HT29b
SW480b
5.5 0.8
MCF-7c
7.7 1.8
A2780d
H460e
A431f
DU145g
14 0.3
BE2-Ch
SJ-G2i
O
S
2
2.8 0.1
7.4 2.1
8.3 0.4
18 2.8
8.5 1.3
6.0 0.6
11 0.6
N
N
9
O
O
10
3.2 0.0
32 2.0
Insol
5.1 0.4
9.3 0.8
Insol
7.7 1.6
17 0.0
Insol
5.3 0.6
17 0.9
Insol
7.0 0.1
20 1.7
Insol
6.2 0.1
24 2.5
13 1.2
21 1.6
8.5 0.7
23 3.0
Insol
11 1.0
29 1.5
Insol
5.6 0.2
20 0.7
Insol
12 0.9
13 3.0
Insol
NH
N
N
N
N
11
O
O
O
12
Insol
Insol
13
54 13
13 1.9
46 4.9
76 1.3
88 10
(18 0.9) > 100
(54 2.9) > 100
96 3.3
96 3.7
57 3.1
48 4.0
OH
O
14
15
9.3 1.2
4.4 0.07
19 0.9
15 1.3
24 2.7
24 1.5
24 0.9
21 0.0
25 0.0
18 3.6
20 0.7
N
O
H
N
N
44 6.3
31 2.9
6.5 2.4
18 0.3
43 1.0
7.9 0.4
(29 5.3)
(37 13)
(23 1.6)
(29 1.0)
(14 1.6)
(29 0.7)
(12 3.8)
(7.5 3.0)
(20 2.7)
(7.0 1.8)
(14 1.2)
(18 3.8)
(18 0.4)
(10 3.7)
(19 3.1)
(36 4.9)
(23 5.6)
(21 3.9)
(3.1 5.0)
(17 6.9)
(17 1.6)
(39 8.1)
(26 20)
(13 4.1)
(30 2.4)
(19 4.8)
(21 2.3)
O
O
O
O
16
N
N
H
O
N
O
O
17
N
H
HN
N
18
(35 5.5)
(56 8.7)
(21 4.6)
(33 5.3)
(17 2.0)
(20 1.9)
(25 5.1)
(32 2.5)
(14 3.0)
(23 2.3)
(27 3.7)
N
H
Values in parentheses represent percentage enzyme or percentage growth inhibition generated at 100 lM drug concentration with IC50 or GI50 values not determined.
a Means SEM of three experiments conducted in triplicate.
b Colorectal.
c Breast.
d Ovarian.
e Lung.
f Skin.
g Prostate.
h Neuroblastoma.
i Glioblastoma human carcinoma cell lines.

3396
T. A. Hill et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3392–3397
Table 3. Inhibition of PP1 and PP2A, and growth inhibition against a number of human cancer cell lines by cantharidin (2) and the morphilino-
substituted half-acid analogue 19
a
a
Analogue
IC50 (lM)
GI50 (lM)
PP1
PP2A
HT29b
SW480b MCF-7c A2780d
H460e
A431f
DU145g BE2-Ch
SJ-G2i
O
O
O 2
11.3 1.97
1.2 0.1 3.3 0.1 4.6 0.2 7.0 0.2 5.8 0.2 4.0 0.1 3.4 0.2 2.4 0.1 4.9 0.2 3.0 0.1
O
O
N
19
5.9 2.2
0.79 0.1 2.8 0.0 3.9 0.1 5.8 0.2 4.9 0.4 3.3 0.1 2.9 0.1 1.7 0.0 2.5 0.3 2.1 0.1
O
O
OH
O
a Means SEM of three experiments conducted in triplicate.
b Colorectal.
c Breast.
d Ovarian.
e Lung.
f Skin.
g Prostate.
h Neuroblastoma.
i Glioblastoma human cancer cell lines.
Introduction of additional substituents to the morphol-
ine ring of 8 with a 20-ethyl alcohol resulted in a decrease
in both PP inhibition and cytotoxicity (Table 2, com-
pound 13), whilst the 30,50-dimethylmorphilino analogue
14 displayed a slight decrease in PP1 inhibition but no
such adverse effect on either PP2A inhibition or cytotox-
icity suggesting that this may be a site suitable for fur-
ther modification. Insertion of an ethyl linker into
analogue 9, afforded analogue 16 which displayed very
poor PP inhibition and cytotoxicity. This suggests that
the proximity of the ring to the amide carbonyl is pivotal
for activity with 9 as one of the most potent protein
phosphatase norcantharidin analogues thus far re-
ported. This effect is further confirmed with the position
of the morpholine substituent as close to the amide car-
bonyl vital for high activity, with movement of even one
bond length detrimental to activity (Table 2, compounds
15 and 16). It is also likely that the type of nitrogen
atoms present in these analogues impacts on the ob-
served activity. The aniline-type nitrogen of 17 is less ba-
sic than the tertiary amine nitrogen of 16 and thus more
favorable for a hydrophobic environment. Also, drop in
potency of 16 compared to 9 may result from the in-
creased polarity of the basic nitrogen of the morpholine
ring in 16 (versus the amide nitrogen of 9).
Cantharidin has greater potency, selectivity, and cyto-
toxicity than norcantharidin, associated with the addi-
tional methyl groups. Thus, it was hypothesized that
the synthesis of a cantharidin based variant of 9
would generate an analogue with improved PP inhibi-
tion and cytotoxicity. Hence, we applied our synthetic
approach as shown in Scheme 1 and generated 19 and
the results of our biological screening are given in
Table 3.
The data in Table 3 clearly show that 19 is more potent
in all aspects than the parent 2. Given we observed no
decrease in inhibition with the 30,50-dimethylmorphili-
no-substituted 14 we feel that this result has the poten-
tial to allow the rapid development of novel highly
potent cantharidin analogues.
In this work, nucleophilic ring opening with a series
of heterocyclic amines has for the first time allowed
the generation of a series of novel (nor)cantharidin
analogues that are more potent than the original
lead compounds. Both PP inhibition and cytotoxicity
are enhanced. In particular, further evaluation of
substituted morphilino analogues should be investi-
gated, and are currently underway in our laborato-
ries and will be reported in due course.
Interestingly, the data generated suggest that the
consensus binding motif associated with analogues
of this kind warrants further investigation and gives
tentative support to the kinetic studies performed by
Shan et al.19
In this series the most potent and PP1 selective analogue
is 10 and the most potent and selective PP2A analogue is
9, and in both cases all the data generated indicate that
both these analogues are more potent than the lead nor-
cantharidin. The type of spacer (phenyl ring, 17) has an
effect on PP inhibition with the insertion of a phenyl ring
restoring potency analogous to our previously reported
anilino series, however analogue 18 possessing a simple
imidazole terminated ethyl chain displays low levels of
PP1/PP2A inhibition as well as very low levels of
cytotoxicity.
Acknowledgment
We are grateful for financial support from the Biotech-
nology Innovation Fund (Australia).

T. A. Hill et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3392–3397
3397
14. Sheppeck, J. E.; Gauss, C.-M.; Chamberlin, A. R. Bio.
Med. Chem. 1997, 5, 1739, and references cited
therein.
15. Tunac, J. B.; Graham, B. D.; Dobson, W. E. J. Antibiot.
1983, 36, 1595.
References and notes
1. Lin, X.-H.; Walter, J.; Scheidtmann, K.; Ohst, K.;
Newport, J.; Walter, G. Proc. Natl Acad. Sci. U.S.A.
1998, 95, 14693.
16. Stampwala, S. S.; Bunge, R. H.; Hurley, T. R.; Willmer,
N. E.; Brankiewicz, A. J.; Steinman, C. E.; Smitka, T. A.;
French, J. C. J. Antibiot. 1983, 36, 1601.
17. Sakoff, J. A.; McCluskey, A. Curr. Pharm. Des. 2004, 10,
1139.
2. Depaoli-Roach, A. A.; Park, I.-K.; Cerovsky, V.; Csortos,
C.; Durbin, S. D.; Kuntz, M. J.; Sitikov, A.; Tang, P. M.;
Verin, A.; Zolnierowicz, S. Adv. Enzyme Regul. 1994, 34,
199.
3. McCluskey, A.; Sakoff, J. A. Mini-Rev. Med. Chem. 2001,
1, 43.
4. McCluskey, A.; Sim, A. T. R.; Sakoff, J. A. J. Med. Chem.
2002, 45, 1151.
5. Ogris, E.; Gibson, D. M.; Pallas, D. C. Oncogene 1997, 15,
911.
18. (a) Hart, M. E.; Chamberlin, A. R.; Walklom, C.; Sakoff,
J. A.; McCluskey, A. Bioorg. Med. Chem. Lett. 2004, 14,
1969; (b) McCluskey, A.; Keane, M. A.; Walkom, C.;
Bowyer, M. C.; Sim, A. T. R.; Young, D. J.; Sakoff, J. A.
Bioorg. Med. Chem. Lett. 2002, 12, 391; (c) McCluskey,
A.; Bowyer, M. C.; Walkom, C.; Ackland, S. P.; Gardiner,
E.; Sakoff, J. A. Bioorg. Med. Chem. Lett. 2001, 11, 2941;
(d) Tatlock, J. H.; Linton, M. A.; Hou, X. J.; Kissinger, C.
R.; Pelletier, L. A.; Showalter, R. E.; Tempczyk, A.;
Villafranca, J. E. Bioorg. Med. Chem. Lett. 1997, 7, 1007;
(e) Sodeoka, M.; Baba, Y.; Kobayashi, S.; Hirukawa, N.
Bioorg. Med. Chem. Lett. 1997, 7, 1833; (f) Enz, A.; Zenke,
G.; Pombo-Villar, E. Bioorg. Med. Chem. Lett. 1997, 7,
2513; (g) McCluskey, A.; Taylor, C.; Quinn, R. J.;
Suganuma, M.; Fujiki, H. Bioorg. Med. Chem. Lett.
1996, 6, 1025.
6. Tolstykh, T.; Lee, J.; Vafai, S.; Stock, J. B. EMBO J. 2000,
19, 5682.
7. Wu, J.; Tolstykh, T.; Lee, J.; Boyd, K.; Stock, J. B.;
Broach, J. R. EMBO J. 2000, 19, 5672.
8. Yu, X. X.; Du, X.; Moreno, C. S.; Green, R. E.; Ogris, E.;
Feng, Q.; Chou, L.; McQuoid, M. J.; Pallas, D. C. Mol.
Biol. Cell 2001, 12, 185.
9. Wei, H.; Ashby, D. G.; Moreno, C. S.; Ogris, E.; Yeong,
F. M.; Corbett, A. H.; Pallas, D. C. J. Biol. Chem. 2001,
276, 1570.
10. Sontag, E. Cell Signal. 2001, 13, 7.
19. Shan, H.; Cai, Y.; Liu, Y.; Zeng, W.; Chen, H.; Fan, B.;
Liu, X.; Xu, Z.; Wang, B.; Xian, L. Anti-Cancer Drugs
2006, 17, 905.
11. Janssens, V.; Goris, J. Biochem. J. 2001, 353, 417.
12. Goldberg, Y. Biochem. Pharmacol. 1999, 57, 321.
13. Virshup, D. M. Curr. Opin. Cell Biol. 2000, 12, 180.