Synthesis and SAR of pyrrolotriazine-4-one based Eg5 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2006.05.037

Synthesis and SAR of substituted pyrrolotriazine-4-one analogues as Eg5 inhibitors are described. Many of these analogues displayed potent inhibitory activities in the Eg5 ATPase and A2780 cell proliferation assays. In addition, pyrrolotriazine-4-one anal

Full text of DOI:10.1016/j.bmcl.2006.05.037

Bioorganic & Medicinal Chemistry Letters 16 (2006) 3937–3942
Synthesis and SAR of pyrrolotriazine-4-one based Eg5 inhibitors
Kyoung Soon Kim,* Songfeng Lu, Lyndon A. Cornelius, Louis J. Lombardo,
Robert M. Borzilleri, Gretchen M. Schroeder, Christopher Sheng, George Rovnyak,
Donald Crews, Robert J. Schmidt, David K. Williams, Rajeev S. Bhide,
Sarah C. Traeger, Patricia A. McDonnell, Luciano Mueller, Steven Sheriff,
John A. Newitt, Andrew T. Pudzianowski, Zheng Yang, Robert Wild,
Frances Y. Lee, Roberta Batorsky, James S. Ryder, Marie Ortega-Nanos,
Henry Shen, Marco Gottardis and Deborah L. Roussell
Department of Drug Discovery, Bristol-Myers Squibb Pharmaceutical Research Institute, Princeton, NJ 08543-4000, USA
Received 11 April 2006; revised 8 May 2006; accepted 8 May 2006
Available online 30 May 2006
Abstract—Synthesis and SAR of substituted pyrrolotriazine-4-one analogues as Eg5 inhibitors are described. Many of these ana-
logues displayed potent inhibitory activities in the Eg5 ATPase and A2780 cell proliferation assays. In addition, pyrrolotriazine-
4-one analogue 26 demonstrated in vivo efficacy in an iv P388 murine leukemia model. Both NMR and X-ray crystallographic stud-
ies revealed that these analogues bind to an allosteric site on the Eg5 protein.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Antimitotic agents such as taxol and the vinca alkaloids
have been successfully used for the treatment of human
cancer.1 These agents bind to the tubulin component of
the mitotic spindle microtubules, disrupting microtubule
dynamics, inducing mitotic arrest, and inhibiting tumor
cell proliferation.2,3 Because microtubules play impor-
tant roles throughout the cell cycle, disruption of micro-
tubule dynamics by taxol and vinca alkaloids also
produces undesirable side effects, such as toxicity in
non-dividing cells like peripheral neurons, leading to
peripheral neuropathy in patients.4
bition of Eg5 activity induces the formation of an
aberrant mitotic spindle (monopolar spindle or monoas-
tral array of spindle microtubules), induces cell cycle
arrest at M-phase, and inhibits cell proliferation.8–10
Small molecule inhibitors of the ATPase activity of the
human Eg5 (KSP) motor domain have recently been
reported.10–12
Herein, we report on the synthesis of pyrrolotriazin-4-
one based Eg5 inhibitors and SAR as shown in Figure 1.
The 5-chloropyrrolotriazine analogue 7 was prepared
starting from 3-chloropyrrole methylester 113 which
was aminated using 2,4-dinitrophenoxyamine14 to afford
Antimitotic agents that specifically bind and modulate
the activity/function of other components of the mitotic
spindle apparatus may provide a novel mechanism by
which to target the clinically validated mitotic spindle
and reduce undesirable side effects. Human Eg5, a Kine-
sin 5 family member (also known as KIF 11 or KSP), is
a motor protein that specifically localizes to the mitotic
spindle apparatus and plays an important role in forma-
tion and function of the bipolar mitotic spindle.5–7 Inhi-
O
CH2C6H5
N
N
C2H5
X
N
H2N
N
O
Keywords: Eg5; Pyrrolotraizine; Microtubule; Kinesin; Anti-tumor;
Cytotoxic.
CH3
*
Corresponding author. Tel.: +1 609 252 5181; fax: +1 609 252
6601; e-mail: kyoung.kim@bms.com
Figure 1. General structure of the pyrrolotriazine-4-one analogues.
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.05.037

3938
K. S. Kim et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3937–3942
2 (Scheme 1). Reacting aminopyrrole 2 with acetonitrile
followed by cyclization and benzylation led to pyrrolo-
triazine-4-one 4. Conversion of the 2-methyl group to
the 2-formyl group was achieved via enamine formation
and oxidative cleavage using NaIO4 to obtain 2-formyl-
pyrrolotriazin-4-one 5. Reaction of compound 5 with
ethyl magnesiumbromide followed by chlorination using
thionyl chloride provided the 2-chloromethylpyrrolotri-
azine-4-one intermediate 6. Coupling of 6 with N-Boc-
1,3-diaminopropane, acylation with p-toluoyl chloride,
and deprotection provided 5-chloropyrrolotriazine-4-
one 7 as a HCl salt.
O
CH2C6H5
C2H5
N
N
N
NH
H2N
21
IC 50 = 100 μM, ATPase assay
1H and 13C chemical shifts of Ala, Ile, and Thr resi-
dues.19 Intermolecular NOEs were observed between
protons on the pyrrolotriazine-4-one ring and a single
Ala methyl group and a single Ile d-methyl group. No
The 6-chloropyrrolotriazine-4-one analogue 14 was pre-
pared starting from 4-chloropyrrole ester 915 (Scheme
2). The amination product of 9 was reacted with
Fmoc-protected thioamide A16 to afford benzylamino-
butylidene adduct 10. Thermal cyclization of 10 afforded
a mixture of 6-chloropyrrolotriazine-4-one 11 and
triazepine 12 which was easily separated by flash column
chromatography on silica gel. Deprotection of the Fmoc
of 11 followed by reductive alkylation, acylation with
p-toluoyl chloride, and deprotection of the N-Boc group
afforded 6-chloropyrrolotriazine-4-one 14.
˚
Ala or Ile residues were within 5 A of the ADP pocket,
eliminating this pocket as a potential binding site for 21.
Using the NOE information, we identified 4 possible
binding sites where an Ala and Ile were proximal and
could accommodate 21. The preferred candidate binding
site formed a pocket where numerous protein:ligand
interactions could occur and a model of Eg5:compound
21 using the NMR experimental data led to the tentative
assignments of the NOEs to Ala218 and Ile136.
The X-ray crystal structure20 of the ternary complex of
Eg5, compound 24, and ADP confirmed these assign-
ments from the NMR studies. This series of compounds
binds to the allosteric site (Fig. 2), which has also been
seen by others with monastrol,12a dihydropyrazoles,21
and tetrahydroisoquinolines.22 The binding of ligands
to the allosteric site causes considerable change in the
tertiary structure of Eg5. The center-to-center distance
The 7-chloro, 5,7-dichloro, and des-chloropyrrolotri-
azine analogues 18, 19, and 20 were prepared following
the similar procedures described in Scheme 1 starting
from 5-chloro and 3,5-dichloropyrrole esters 16 and
1717 (Scheme 3). Dehalogenation of 7-chloro-pyrrolotri-
azine-4-one 18 using PtO2 in MeOH–HOAc under 1 atm
of H2 gas afforded des-chloro-pyrrolotriazine-4-one 20.
˚
between compound 24 and the ADP is ꢀ13 A and the
The binding site on Eg5 for this series of compounds
was determined initially by NMR studies of 21, a trun-
cated analogue of 20.18
closest approach between atoms is the amino nitrogen
to a phosphate oxygen distance that is greater than
˚
7 A, indicating that beyond some possible long distance
electrostatic interactions, compound 24 and the ADP do
not directly interact with each other. The X-ray crystal
structure shows the inhibitor binds the protein in a fold-
ing conformation where two side-chain phenyl rings
interact with each other in a face-to-edge fashion.23
These two phenyl rings and the cyclopropyl interact
Although resonance assignments for the protein were
not determined, intermolecular nuclear Overhauser
effects (NOEs) observed between labeled alanine, isoleu-
cine, and threonine (‘AIT’) Eg5 protein and 21 were
tentatively assigned to amino acid type based on typical
Cl
Cl
O
Cl
CO2CH3
NH2
Cl
a
b
c,d
OCH3
N
CH2C6H5
CH3
OCH3
N
N
N
N
N
CH3
O
N
NH2
H
O
3
2
4
1
O
Cl
O
N
O
CH2C6H5
Cl
Cl
N
CH2C6H5
CHO
CH2C6H5
C2H5
Cl
e,f
g,h
i,j,k
N
N
N
C2H5
O
N
N
N
N
H2N
N
.HCl
7
5
6
CH3
Scheme 1. Reagents and conditions: (a) 2,4-dinitrophenoxyamine, NaH, DMF, rt; (b) CH3CN, HCl, dioxane, 82 ꢁC; (c) CH3CN, Et3N, 85 ꢁC; (d)
Cs2CO3, BnBr, dioxane, 95 ꢁC; (e) DMF dimethylacetal, MgSO4, DMF, 146 ꢁC; (f) NaIO4, THF, pH 7 buffer, rt; (g) ethyl magnesiumbromide,
CH2Cl2, À78 to À50 ꢁC; (h) SOCl2, pyridine, CH2Cl2, 0 ꢁC to rt; (i) N-Boc-1,3-diaminopropane, EtOH, reflux temp; (j) p-toluoyl chloride, Et3N,
CH2Cl2, rt; (k) 4 M HCl/dioxane, rt.

K. S. Kim et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3937–3942
3939
Cl
O
N
Cl
OCH3
CH2C6H5
C2H5
CCl3
a, b
c
ref 15
N
N
N
OCH3
N
Cl
O
H
N
N
C2H5
O
H
O
NHFmoc
11
C6H5H2CHN
NHFmoc
+
O
8
9
25 %
10
d,e
O
N
CH2C6H5
C2H5
NH
Cl
O
N
C2H5
Cl
N
CH2C6H5
C2H5
N
f,g
N
N
Cl
HN CH2C6H5
N
N
H2N
N
O
12
NH
NHBoc
56 %
13
14
CH3
Scheme 2. reagents and conditions: (a) 2,4-dinitrophenoxyamine, NaH, DMF, rt; (b) BnNHC(S)CH(NHFmoc)Et (A), EDCIÆHCl, CH2Cl2, rt; (c)
xylene, 130 ꢁC, 7 h; (d) piperdine, DMF; (e) (3-oxo-propyl)-carbamic acid tert-butyl ester, NaBH(OAc)3, HOAc, THF, rt; (f) p-toluoyl chloride,
Et3N, CH2Cl2, rt; (g) TFA, CH2Cl2, rt.
Cl
ref 17
OCH3
+
Cl
OCH3
Cl
OCH3
N
N
H
N
H
O
O
H
O
major
17
15
minor
16
O
N
O
O
Cl
Cl
CH2C6H5
C2H5
CH2C6H5
C2H5
CH2C6H5
C2H5
N
N
N
N
N
N
N
N
Cl
N
O
N
O
N
O
NH2
NH2
NH2
20
19
18
CH3
CH3
CH3
Scheme 3. Synthetic steps similar to those outlined in Scheme 1.
with Trp127, Tyr211, Pro137, and Arg119 of the pro-
tein. The pyrrole ring makes van der Waals contacts
with the surface of the pocket formed by Ile136,
Leu214, Phe239, and Leu160 of the protein.
of A2780 ovarian carcinoma cell proliferation is report-
ed here.26
Analogues having a chlorine substituent on the pyrrole
ring at either C-5 or C-7 (7 and 18) retained the same
activity as non-substituted analogue (20) both in ATP-
ase and human ovarian carcinoma A2780 cell line cyto-
toxicity assays (Table 1). However, chlorinated
analogues produced generally more favorable pharma-
cokinetic profiles than the non-chlorinated analogue.
While one chlorine substituent at either C-5 or C-7 is
tolerated, bis-chlorination at C-5 and C-7 (19) leads to
a reduction in potency, possibly because compound 19
Compound 24 inhibits the ATPase activity of Eg5, an
apparent consequence of changes in Eg5 tertiary struc-
ture that result from the binding of compound 24 (Table
2). Optimization of this series through the development
of structure–activity relationships monitored the activity
of new analogues in a microtubule-dependent Eg5 ATP-
ase assay.25 The cytotoxic activity of new analogues on
human tumor cell lines was also determined. Inhibition

3940
K. S. Kim et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3937–3942
Table 2. SAR of C-2 methyl substituent (R1) and phenyl substituent
(R2)
O
N
Cl
CH2C6H5
R1
N
N
N
O
NH2
R2
a
Compound R1
R2
IC50 (lM)
Figure 2. The binding site for compound 24 on Eg5. Compound 24 is
shown with thick rods and the protein is shown with thinner rods. The
initial 2Fo À Fc electron density map contoured at 1r is shown in cyan
mesh. Atomic coloring (N, blue; O, red; Cl, cyan) is used with green for
carbon atoms of the compound and gray for carbon atoms of the
protein. Figure created using PyMol.24
ATPase
A2780 cytotoxicity
7
22
23
24
25
C2H5
C2H5
H
CH3 0.07 (0.01) 0.14 (0.02)
Cl
Cl
0.19 (0.03) 0.44 (0.01)
>25 >10
Cyclopropyl CH3 0.10 (0.02) 0.18 (0.01)
Vinyl CH3 1.92 (0.12) 2.79 (0.47)
a IC50 values are reported as means of three independent determina-
tions with standard deviations shown (in parentheses).
Table 1. SAR of the pyrrole ring
O
R5
Analogues with a basic amine function on the C-2 side
chain displayed the best enzyme inhibitory activity (26,
28 vs 27, 31, Table 3). This is explained based on the
observation that the ammonium ion of the inhibitor
interacts with Glu116 (3.0 A, see Fig. 2). A relatively
small space in this region disfavors the gem-dimethyl
substituted amines (29, 30) or bulkier amine (32).
CH2C6H5
N
R6
N
C2H5
O
N
R7
N
˚
NH2
CH3
As shown in Tables 1–3, inhibition of Hs Eg5 ATPase
activity in vitro correlates with inhibition of A2780
human ovarian carcinoma cell proliferation in a 72 h
cytotoxicity assay. Examination of treated cells at earlier
time point shows a significant increase in the percentage
of cells arrested in mitosis compared to untreated con-
trols (3% mitotic cells). For example, treatment of
A2780 cells with compound 26 for 17 h, corresponding
to approximately one cell cycle, results in 65% of the cell
population in mitosis.27 The induction of mitotic arrest
is consistent with an Eg5 mechanism of action for the
cytotoxic activity of compound 26 in human cancer cell
lines.
Compound R5 R6 R7
IC50 (lM)
a
ATPase
A2780 cytotoxicity
7
14
18
19
20
Cl
H
H
Cl
H
H
Cl
H
H
H
H
H
Cl
Cl
H
0.07 (0.01) 0.14 (0.02)
0.33 (0.04) 0.98 (0.01)
0.06 (0.01) 0.13 (0.03)
0.29 (0.06) 0.47 (0.03)
0.08 (0.01) 0.11 (0.004)
a IC50 values are reported as means of three independent determina-
tions with standard deviations shown (in parentheses).
loses the flexibility to adjust itself to the compact space
of the hydrophobic pocket. Chlorination at C-6 (14) also
resulted in reduction in potency. The pyrrole ring’s C-6
is directed toward Leu160 (see Fig. 2) and the distance
from the side chain of Leu160 to C-6 of pyrrole ring is
In order to determine the in vivo antitumor effects of
pyrrolotriazine-4-one based Eg5 inhibitors, compounds
were evaluated in the iv P388 murine leukemia model.28
Compounds were evaluated for their effect upon life-
span, compared to untreated iv P388 control mice.
Compound 26, which has a good Vss with a 4.5 h half-
life (Table 4), was active in an iv P388 murine leukemia
model (Table 5, where T/C > 125% is considered active).
At its MTD of 20 mg/kg/inj, when given by intravenous
injection daily for five consecutive days (5· qd, iv), a
maximum increase in lifespan corresponding to a T/C
of 163% was achieved.
˚
only 3.7 A which would cause a steric clash when C-6
is substituted with chlorine.
Substitution of ethyl or cyclopropyl for the C-2 methyl
side chain produced analogues (7, 22, and 24) which dis-
played potent enzyme and cell activity, while non-substi-
tuted or vinyl-substituted analogues 23 and 25 possess
significantly less activity (Table 2). Loss of activity of
the vinyl analogue 25 is surprising in view of the fact
that ethyl and cyclopropyl analogues are potent inhibi-
tors. Generally, analogues with a p-methyl substituent
on the phenyl ring were slightly more potent than the
p-chloro-substituted analogues (7 vs 22).
In summary, potent pyrrolotriazine-4-one based Eg5
inhibitors were synthesized and were shown to have
potent activity in ATPase and cell proliferation assays.
Compound 26 was efficacious in an iv P388 leukemia

K. S. Kim et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3937–3942
3941
Table 3. SAR of amine side chain (R2)
O
CH2C6H5
N
N
R1
N
R2
R7
N
O
R3
R7
R1
R2
R3
IC50 (lM)
A2780 cytotoxicity
a
Compound
ATPase
26
27
28
29
30
31
32
Cl
Cl
Cl
H
H
H
H
Cyclopropyl
Cyclopropyl
C2H5
CH2CH2NH2
CH2C(O)NH2
CH2CH2NH2
CH2C(CH3)2NH2
C(CH3)2NH2
CH2CH2OH
4-Piperdyl
4-C6H5CH3
4-C6H5CH3
4-C6H5Cl
0.06 (0.003)
0.98 (0.11)
0.06 (0.006)
0.65 (0.08)
3.72 (0.93)
6.88 (0.33)
6.06 (1.02)
0.05 (0.025)
0.33 (0.044)
0.16 (0.01)
1.14 (0.03)
2.72 (0.33)
2.86 (0.003)
4.46 (0.71)
C2H5
C2H5
4-C6H5Cl
4-C6H5Cl
C2H5
C2H5
4-C6H5CH3
4-C6H5CH3
a IC50 values are reported as means of three independent determinations with standard deviations shown (in parentheses).
Table 4. Summary of pharmacokinetic parameters of compound 26
(10 mg/kg in 50% PEG400/50% water) in mice (n = 3)
8. Blangy, A.; Lane, H. A.; d’Herin, P.; Harper, M.; Kress,
M.; Nigg, E. A. Cell 1995, 83, 1159.
9. Weil, D.; Garcon, L.; Harper, M.; Dumenil, D.; Dautry,
F.; Kress, M. Biotechniques 2002, 33, 1244.
Clearance (mL/min/kg)
16 2.7
Vss (L/kg)
MRT (h)
6.5 2.3
t1/2 (h)
5.9 0.92
4.5 1.6
10. Finner, J. T.; Bergnes, G.; Feng, B.; Smith, W. W.;
Chabala, J. C. PCT Int. Appl. WO 01/30768 A1.
11. Mayer, T. U.; Kapoor, T. M.; Haggarty, S. J.; King, R. W.;
Schreiber, S. L.; Mitchison, T. J. Science 1999, 286, 9171.
12. (a) Yan, Y.; Sardana, V.; Xu, B.; Jomnick, C.; Halczenko,
W.; Buser, C. A.; Schaber, M.; Hartman, G. D.; Huber, H.
E.; Kuo, L. C. J. Mol. Biol. 2004, 335, 547; (b) Sakowicz,
R.; Finer, J. T.; Beraud, C.; Crompton, A.; Lewis, E.,
et al. Cancer 2004, 64, 3276; (c) Cox, C. D.; Breslin, M. J.;
Mariano, B. J.; Coleman, P. J.; Buser, C. A., et al. Bioorg.
Med. Chem. Lett. 2005, 15, 2041; (d) Sunder-Plassmann,
N.; Sarli, V.; Gartner, M.; Utz, M.; Seiler, J., et al. Bioorg.
Med. Chem. 2005, 13, 6094; (e) Bergnes, G.; Brejc, K.;
Belmont, L. Curr. Top. Med. Chem. 2005, 5, 127.
13. Tehrani, K. A.; Borremans, D.; De Kimpe, N. Tetrahe-
dron 1999, 55, 4133.
Table 5. Summary of compound 26 in vivo antitumor activity against
P388 murine leukemia
Dose (mg/kg)
20
Route of administration
iv
Schedule
% T/C
163
5· qd
model. NMR and X-ray crystallographic studies re-
vealed that these analogues bind to an allosteric site
on the Eg5 protein.
Acknowledgments
14. (a) Sheradsky, T. Tetrahedron Lett. 1968, 9, 1909; (b)
Sheradsky, T. J. Heterocyclic Chem. 1967, 4, 413.
15. Belanger, P. Tetrahedron Lett. 1979, 20, 2505.
16. Thioamide A was prepared by reacting the corresponding
amide with P2S5.
The authors wish to acknowledge Rolf Rysek for clon-
ing Eg5 DNA sequence, Gerald Duke for the prepara-
tion of Eg5 protein used in the Eg5 enzymatic assays,
and Valentina Goldfarb for making the protein for the
NMR studies.
17. Hodge, P.; Rickards, R. W. J. Chem. Soc. 1965, 459.
18. NMR experiments were performed on compound 21
because the resonances of compound 7 in complex with
Eg5 were broadened, indicating that the compound was in
intermediate exchange with Eg5. An ‘AIT’ Eg5 protein
sample was prepared in which the Ile d-methyl groups were
specifically labeled with 1H/13C, and Ala and Thr residues
were labeled with 1H/13C/15N; all other residues were
uniformly deuterated. By selectively labeling the protein
residues in this way, any inter-molecular NOEs observed in
a 3D 13C edited NOESY were ensured to arise from the
interaction of an Ala, Ile or Thr residue of the protein and
the compound. NOE data were collected on a sample of
2.0 mM 34, 0.8 mM ‘AIT’ Eg5, and 1.0 mM ADP.
19. Wishart, D. S.; Watson, M. S.; Boyko; Sykes, B. D.
J. Biomol. NMR 1997, 10, 329.
References and notes
1. Dutcher, J. P.; Novik, Y.; O’Boyle, K.; Marcoullis, G.;
Secco, C.; Wiernik, P. H. J. Clin. Pharmacol. 2000, 40,
1079.
2. Wilson, L.; Panda, D.; Jordon, M. A. Cell Struct. Funct.
1999, 24, 329.
3. Blagosklonny, M. V.; Fojo, T. Int. J. Cancer 1999, 83, 151.
4. Quasthoff, S.; Hartung, H. P. J. Neurol. 2002, 249, 9.
5. Sawin, K. E.; LeGuellec, K.; Philippe, M.; Mitchison, T. J.
Nature 1992, 359, 540.
6. Gaglio, T.; Saredi, A.; Bingham, J. B.; Hasbani, M. J.;
Gill, S. R.; Schroer, T. A.; Compton, D. A. J. Cell Biol.
1996, 135, 399.
20. (a) PDB Deposition number is 2GM1.; (b) Newitt, J. et al.,
manuscript in preparation.
21. Cox, C. D.; Breslin, M. J.; Mariano, B. J.; Coleman, P. J.;
Buser, C. A.; Walsh, E. S.; Hamilton, K.; Huber, H. E.;
7. Walczak, C. E.; Vernos, I.; Mitchison, T. J.; Karsenti, E.;
Heald, R. Curr. Biol. 1998, 8, 903.

3942
K. S. Kim et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3937–3942
Kohl, N. E.; Torrent, M.; Yan, Y.; Kuo, L. C.; Hartman,
G. D. Bioorg. Med. Chem. Lett. 2005, 15, 2041.
22. Tarby, C. M.; Kaltenbach, R. F., III, et al. Bioorg. Med.
Chem. Lett. 2006, 16, 2095.
23. Compound 24 itself might take the similar conformation
as that in the protein-bound structure. X-ray crystal
structure of 24 has not been determined.
manufacturer (Promega) Assay. It will be described in
greater detail in Burford, N. et al. manuscript in
preparation.
26. The 72 h cell proliferation assay was performed as
described in Kim, K. S.; Kimball, S. D.; Misra, R. N.;
Rawlins, D. B.; Hunt, J. T., et al. J. Med. Chem. 2002, 45,
3905.
24. DeLano, W. L. The PyMol Molecular Graphics System
(2002). DeLano Scientific, San Carlos, CA, US. <http://
www.pymol.org./>
25. ATPase assays with Eg5 motor domain protein were
performed with Kinase Glo Kit as described by the
27. Experimental details and additional data provided in
Roussell, D. et al., manuscript in preparation.
28. Rose, W. C.; Schurig, J. E.; Meeker, J. B. Anticancer Res.
1988, 8, 355, The activity criterion for increased lifespan
was a T/C of P125%..