Optimization of α-acylaminoketone ecdysone agonists for control of gene expression

Article abstract of DOI:10.1016/S0960-894X(03)00315-9

Fifteen new α-acylaminoketones were prepared by four different routes in an initial effort to optimize the potency of these compounds as ecdysone agonists. The compounds were assayed in mammalian cells expressing the ecdysone receptors from Bombyx mori (BmEcR) and Choristoneura fumiferana (CfEcR) for their ability to cause expression of a reporter gene downstream of an ecdysone response element. A new α-acylaminoketone was identified which had activity equal to that of the standard dibenzoylhydrazine ecdysone agonist GS-E in the assay based on CfEcR.

Full text of DOI:10.1016/S0960-894X(03)00315-9

Bioorganic & Medicinal Chemistry Letters 13 (2003) 1883–1886
Optimization of ꢀ-Acylaminoketone Ecdysone Agonists for
Control of Gene Expression
Colin M. Tice,* Robert E. Hormann, Christine S. Thompson, Jennifer L. Friz,
Caitlin K. Cavanaugh and Jessica A. Saggers
RHeoGene, PO Box 949, 727 Norristown Road, Spring House, PA 19477-0949, USA
Received 30 December 2002; accepted 24 February 2003
Abstract—Fifteen new a-acylaminoketones were prepared by four different routes in an initial effort to optimize the potency of
these compounds as ecdysone agonists. The compounds were assayed in mammalian cells expressing the ecdysone receptors from
Bombyx mori (BmEcR) and Choristoneura fumiferana (CfEcR) for their ability to cause expression of a reporter gene downstream
of an ecdysone response element. A new a-acylaminoketone was identified which had activity equal to that of the standard diben-
TM
zoylhydrazine ecdysone agonist GS -E in the assay based on CfEcR.
# 2003 Elsevier Science Ltd. All rights reserved.
We recently reported the discovery of novel ecdysone
agonist 1a (Fig. 1), an a-acylaminoketone derived syn-
thetically from 1-aminocyclopentane-1-carboxylic acid.1
however, similar treatment of 5a and 5f gave less satis-
factory results and more extensive purification was
required to isolate cyclopentane 1a and cyclohexane 1f
in 40 and 28% yields, respectively.13 Method B, in
which azlactones 4 were converted into 1 in four steps
via aldehydes 6, proved more satisfactory for the pre-
paration of 1a and 1f despite its greater length. Method
B was also applied to the synthesis of geminal dimethyl
compound 1c and cycloheptane 1g. All target com-
TM
Non-steroidal ecdysone agonists, notably GS -E (2b),
are important components in systems to control gene
expression2,3 based on engineered ecdysone receptors.4ꢀ10
We were therefore interested in discovering more potent
analogues of 1a.
1
Our previous SAR study of a-acylaminoketones of
general structure 1 had shown that 1a, in which R1 and
R2 form a cyclopentane ring, was significantly more
potent than 1b (Table 1), in which R1 and R2 are both
ethyl groups. Thus our initial efforts were directed
towards exploration of the effect of ring size on potency.
Two previously described synthetic routes1 were used
(Scheme 1). In Method A, a,a-disubstituted amino acids
3 were converted to the corresponding azlactones 4
which were opened with N,O-dimethylhydroxylamine to
afford Weinreb amides 5.12 Weinreb amides 5 were
reacted with 3,5-dimethylphenylmagnesium bromide to
afford new a-acylaminoketones 1d–f. Desired cyclopro-
pane and cyclobutane products 1d and 1e were the
major components of the crude products derived from
5d and 5e isolated in 68 and 66% yields, respectively;
pounds were characterized by H NMR and either 13C
NMR or electrospray MS. Certain analogues were more
fully characterized.14
The initial set of compounds 1a–g was screened in dose
response assays in two different cell lines engineered to
express ecdysone receptors from lepidopteran species
and reporter genes under the control of ecdysone
response elements. The first assay used a previously
reported HEK-293 cell-line engineered to express the
ecdysone receptor from Bombyx mori (BmEcR) and a
b-galactosidase reporter gene.6,15 The second assay used
CHO cells expressing a modified Choristoneura fumiferana
*Corresponding author at current address: Concurrent Pharmaceu-
ticals, 502 West Office Center Drive, Fort Washington, PA 19034,
USA; e-mail: ctice@ concurrentpharma.com
Figure 1. Non-steroidal ecdysone agonists.
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00315-9

1884
C. M. Tice et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1883–1886
Table 1. Structure and transactivation assay results of a-acylaminoketone analogues 1
Compd
X
R1, R2
Y
Synthetic methoda
BmEcRb
EC50 (mM)d/RMFIe
CfEcRc
EC50 (mM)d/RMFIe
1a
1b
1c
1d
1e
1f
1g
1h
1i
1j
1k
1l
1m
1n
1o
1p
1q
2-Me-3-MeO
2-Me-3-MeO
2-Me-3-MeO
2-Me-3-MeO
2-Me-3-MeO
2-Me-3-MeO
2-Me-3-MeO
2-Me-3-MeO
2-Me-3-MeO
2-Me-3-MeO
2-Me-3-MeO
2-Me-3-MeO
2-Me-3-MeO
2-Et-3-MeO
–(CH2)4–
Et, Et
Me, Me
–(CH2)2–
–(CH2)3–
–(CH2)5–
–(CH2)6–
–(CH2)4–
–(CH2)5–
–(CH2)6–
3,5-diMe
3,5-diMe
3,5-diMe
3,5-diMe
3,5-diMe
3,5-diMe
3,5-diMe
2-MeO
A or B
B
B
A
A
A or B
B
B
B
B
1:87=0:85
4:53=0:61
10:00=0:17
20:36=0:22
8:52=0:69
1:41=0:99
1:61=0:69
5:16=0:70
0:98=0:84
1:94=0:78
10:00=0:18
1:60=0:98
15:00=0:10
1:81=0:85
3:68=0:81
0:74=0:89
1:51=0:63
1:91=0:86
6:23=0:54
41:95=0:29
33:30=0:09
10:00=0:89
1:11=0:96
2:97=0:88
6:61=0:67
1:03=0:71
3:36=0:66
10:00=0:96
1:80=0:85
30:00=0:04
1:25=0:89
3:22=0:91
2-MeO
2-MeO
–(CH2)2O(CH2)2–
–(CH2)2S(CH2)2–
–(CH2)2SO2(CH2)2–
–(CH2)5–
3,5-diMe
3,5-diMe
3,5-diMe
3,5-diMe
3,5-diMe
3,5-diMe
3,5-diMe
A
A
Af
C
C
C
2-Me-3,4-OCH2O
2-Et-3,4-OCH2CH2O
4-Et
–(CH2)5–
–(CH2)5–
–(CH2)5–
0:39=0:91
h
Bg or D
2a
2b
2-Me-3-MeO
2-Et-3-MeO
n/a
n/a
3,5-diMe
3,5-diMe
n/a
n/a
0:10=1:03
0:25=1:00
0:28=1:02
0:41=1:00
aSee Scheme 1 for Methods A and B and Scheme 2 for Method C.
bSee ref 15 for assay protocol.
cSee ref 16 for assay protocol.
dDose affording 50% of maximum transactivation.
eRatio of maximum level of gene expression of compound to maximum level of gene expression with 2b.
fPrepared from 1i by treatment with oxone in aqueous methanol.
g4-Ethylbenzoyl chloride (16b) was substituted for 3-methoxy-2-methylbenzoyl chloride in Scheme 1.
hCompound not tested.
Scheme 1. a-Acylaminoketone synthesis methods A (via 5) and B (via
6). (a) X-PhCOCl (2.5 equiv), pyridine, rt; (b) MeNHOMe.HCl, pyri-
dine, CH2Cl2, rt; (c) Y-PhMgBr (4 equiv), THF, rt, 5 h; (d) NaBH4,
THF, rt: (e) Dess–Martin periodinane, CH2Cl2, rt; (f) Y-PhMgBr
(4 equiv), THF, ꢀ70 ꢁC ! rt; (g) Dess–Martin periodinane, CH2Cl2, rt.
Figure 2. Overlap of the X-ray structures of 1f (grey) and 2b (black).
ecdysone receptor (CfEcR) and a luciferase reporter
gene.11,16 The results are reported in Table 1 in terms of
EC50 and maximum fold induction relative to the stan-
ether and sulfone functionality were 9- and 27-fold less
potent, respectively, than parent cyclohexane 1f. By
contrast, the more hydrophobic thioether 1l retained
most of the activity of 1f.
TM
dard ecdysone agonist GS -E (2b)17 which was used as
a positive control. Similar SAR trends were observed in
both assays. Cyclohexane 1f proved to be a more effec-
tive ligand (lower EC50 and higher relative maximum
fold induction) than either the previous lead compound,
cyclopentane 1a, or its homologue cycloheptane 1g.
Activity dropped off sharply as the ring size was
decreased to cyclobutane 1e and cyclopropane 1d. The
superior potency of the cyclohexane ring was further
confirmed by synthesis and bioassay of 1h–j in which
the Y substituent is 2-methoxy rather than 3,5-dimethyl.
Finally, we turned our attention to optimization of the
substitution X on the benzamide ring of 1. As before we
were guided by the SAR of the related dibenzoylhy-
drazine ecdysone agonists 2 in the selection of 1n–q as
target compounds.18 Our confidence in the validity of
this analogy was bolstered by comparison of the X-ray
crystal structures of 1f19 and 2b.20 The unit cell of 1f
consisted of one molecule of methanol and two mole-
cules of the a-acylaminoketone in different conformers.
Figure 2 shows the overlap of one conformer each of 1f
and 2b. The RMSD between the backbone C–C(¼O)–
N–C–C(¼O)–C atoms of 1f and the C–C(¼O)–N–N–
C(¼O)–C atoms of 2b in these conformers was 0.081 A.
Next, compounds 1k–m were prepared to explore the
effect of introduction of heteroatoms into the cyclohexane
ring. 1k and 1m which display hydrogen bond accepting

C. M. Tice et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1883–1886
1885
products. Thus, 14 was treated overnight with an excess
of 16b to ensure complete N-acylation and the reaction
mixture was immediately saponified to cleave any
O-benzoylated products and afford clean alcohol 15.
Oxidation of 15 with Dess–Martin periodinane afforded
the desired a-acylaminoketone 1q in 16% yield from 14.
Figure 3. Acids used for preparation of 1o and 1p.
Compound 1n, with an ortho-ethyl group on the benza-
mide ring, was slightly less active than ortho-methoxy
compound 1f. The 2-methyl-3,4-methylenedioxy com-
pound 1o was less active than 1f; however, the 2-ethyl-
3,4-ethylenedioxy analogue 1p proved to be the most
potent a-acyalminoketone ligand tested in both the
BmEcR and CfEcR based assays. Its potency was equal
TM
Scheme 2. Method
C for synthesis of a-acylaminoketones. (a)
to that of the standard ligand GS -E (2b) in the CfEcR
based assay.
(Boc)2O, THF, rt, 18 h; (b) Dess–Martin periodinane, CH2Cl2, rt, 6 h;
(c) 3,5-diMe-PhMgBr, THF, ꢀ70 ꢁC, 1 h; (d) Dess–Martin period-
inane, CH2Cl2, rt, 20 h; (e) CF3CO2H, CH2Cl2, rt, 1 h; (f) X-PhCOCl
(16), pyridine, CH2Cl2; (g) X-PhCO2H (17), EDC.HCl, DMAP,
i-Pr2NEt, CH2Cl2, THF, rt, 24 h.
We have described the synthesis of fifteen new a-acyl-
aminoketones of general structure 1 by four different
routes and their activity as ligands for the control of
gene expression in systems based on lepidopteran EcRs.
We have determined the crystal structure of 1f, a repre-
sentative analogue. Compound 1p is the most potent
ecdysone agonist of general structure 1 described to
date.
Scheme 3. Method
D for synthesis of a-acylaminoketones. (a)
Acknowledgements
NaOMe, MeOH, 5 ꢁC>rt, 16 h; (b) Zn, conc HCl, MeOH, 5 ꢁC ! rt,
16 h; (c) 4-Et-PhCOCl (16b, 2.2 equiv), pyridine, THF, rt, 16 h; (d)
10% aq NaOH, MeOH, THF, rt, 16 h; (e) Dess–Martin periodinane,
CH2Cl2, rt, 6 h.
This work was supported by NIST Advanced Technology
Project Grant 70NANB0H3012.
In the second conformer of 1f (not shown) the torsional
angle between the 3-methoxy-2-methyl substituted ben-
zene ring was rotated by ca. 180ꢁ about the bond to the
carbonyl carbon.21
References and Notes
1. Tice, C. M.; Hormann, R. E.; Thompson, C. S.; Friz, J. L.;
Cavanaugh, C. K.; Michelotti, E. L.; Garcia, J.; Nicolas, E.;
Albericio, F. Bioorg. Med. Chem. Lett. 2003, 13, 475.
2. Albanese, C.; Hulit, J.; Sakamaki, T.; Pestell, R. G. Semi-
nars in Cell and Developmental Biology 2002, 13, 129.
3. DeMayo, F. J.; Tsai, S. Y. Trends in Endocrinology &
Metabolism 2001, 12, 348.
4. Christopherson, K. S.; Mark, M. R.; Bajaj, V.; Godowski,
P. J. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 6314.
5. No, D.; Yao, T.-P.; Evans, R. M. Proc. Natl. Acad. Sci.
U.S.A. 1996, 93, 3346.
6. Suhr, S. T.; Gil, E. B.; Senut, M.-C.; Gage, F. H. Proc.
Natl. Acad. Sci. U.S.A. 1998, 95, 7999.
7. Martinez, A.; Sparks, C.; Hart, C. A.; Thompson, J.; Jep-
son, I. The Plant Journal 1999, 19, 97.
8. Hoppe, U. C.; Marban, E.; Johns, D. C. Molecular Therapy
2000, 1, 159.
9. Saez, E.; Nelson, M. C.; Eshelman, B.; Banayo, E.; Koder,
A.; Cho, G. J.; Evans, R. M. Proc. Natl. Acad. Sci. U.S.A.
2000, 97, 14512.
While Method B would have been a workable synthetic
approach to 1n–q, we sought methods that would allow
incorporation of the benzamide ring later in the synthesis.
Two methods were developed. Firstly, in Method C
(Scheme 2) the known aminoalcohol 722 was protected as
its Boc derivative and oxidized with Dess–Martin peri-
odinane to afford aldehyde 8 in 89% yield. Reaction of
8 with 3,5-dimethylphenylmagnesium bromide afforded
a secondary alcohol which was oxidized with the Dess–
Martin periodinane to give ketone 9 in 73% yield. TFA
mediated removal of the Boc protecting group from 9
liberated the aminoketone 10 in 71% yield. Treatment
of 10 with 2-ethyl-3-methoxybenzoyl chloride (16a)
afforded 1n while carbodiimide coupling with acids
17a23 and 17b24 (Fig. 3) gave the desired a-acylamino-
ketones 1o and 1p, respectively.25,26
10. Kumar, M. B.; Fujimoto, T.; Potter, D. W.; Deng, Q.;
Palli, S. R. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 14710.
11. Palli, S. R.; Kapitsaya, M. Z.; Kumar, M. B.; Cress, D. E.
Eur J. Biochem. 2003, 270, 1308.
12. Kemp, A.; Ner, S. K.; Rees, L.; Suckling, C. J.; Tedford,
M. C.; Bell, A. R.; Wrigglesworth, R. J. Chem. Soc., Perkin
Trans. 2 1993, 741.
In Method D, nitrocyclohexane (11) was reacted with
benzaldehyde 12 under literature conditions27 to afford
13 in 14% yield (Scheme 3). Nitroaldol adduct 13 was
reduced to aminoalcohol 14 in 85% yield. Treatment of
14 with one equivalent of 4-ethylbenzoyl chloride (16b)
gave a mixture of unreacted 14, mono- and diacylated

1886
C. M. Tice et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1883–1886
13. The products of rearrangement of the Weinreb amides 4a
and 4e to the corresponding N-(hydroxymethyl)-N-methyl
amides were major byproducts: Graham, S. L.; Scholz, T. H.
Tetrahedron Lett. 1990, 31, 6269.
plate and incubated at 37 ꢁC under 5% CO2 for 24 h. Ligand
stock solutions were prepared in DMSO and diluted 300-fold
for all treatments. Dose–response testing consisted of 8 con-
centrations ranging from 33 to 0.01 mM. Luciferase reporter
gene expression was measured 48 h after cell treatment using
Bright-GloTM Luciferase Assay System from Promega (E2650).
Luminescence was detected at room temperature using a Dynex
MLX microtiter plate luminometer. EC50s were calculated
from dose response data using a three parameter logistic model.
17. Carlson, G. R.; Cress, D. E.; Dhadialla, T. S.; Hormann,
R. E.; Le, D. P. U.S. Patent 6,258,603, 2001; Chem. Abstr.
2001, 135, 72148.
14. Compound 1f: Mp 161–163 ꢁC. 1H NMR (CDCl3,
300 MHz) d 1.2–1.9 (8H), 1.96 (s, 3H), 2.05 (m, 2H), 2.31 (s,
6H), 3.80 (s, 3H), 6.25 (br s, 1H), 6.75 (d, J=7.6 Hz, 1H), 6.85
(d, J=8.2 Hz, 1H), 7.08 (s, 1H), 7.14 (m, 1H), 7.51 (s, 2H); 13
C
NMR (CDCl3, 75 MHz) d 11.9, 21.3, 21.7, 25.1, 35.4, 55.6,
64.0, 111.3, 118.2, 125.0, 125.8, 126.5, 133.0, 137.2, 137.3,
137.7, 158.0, 168.7, 203.1. IR (CDCl3) 1680, 2939 cmꢀ1. MS
(ESI-ve ion) m/z 380 (M-1). Anal. calcd for C24H29NO3: C,
75.96; H, 7.70; N, 3.69. Found: C, 75.77; H, 7.94; N, 3.83.
15. A bulk transformed HEK-293 cell line, created as descri-
bed in ref 6, was provided by Gage and Suhr. Dilution cloning
was used to isolate individual clones. Clone Z3 was selected
using 450 mg mLꢀ1 G418 and 100 ng mLꢀ1 puromycin. Cells
were trypsinized and diluted to a concentration of 2.5 ꢂ
104 cells mLꢀ1. Cell suspension 100 mL was placed in each well
of a 96 well plate (Dynex, 14-245-182) and incubated at 37 ꢁC
under 5% CO2 for 24 h. Ligand stock solutions were prepared
(10 mM in DMSO) and diluted 100-fold in media. 50 mL of
diluted ligand solution was added to each well. The final con-
centration of DMSO was maintained at 0.03% in both con-
trols and treatments. b-Galactosidase reporter gene expression
was measured 48 h after treatment of the cells using Gal
ScreenTM bioluminescent reporter gene assay system from
Tropix (GSY1000). Luminescence was detected at room tem-
perature using a Dynex MLX microtiter plate luminometer.
Fold inductions were calculated by dividing relative light units
(RLU) in ligand treated cells with RLU in DMSO treated
cells.
16. CHO cells were transiently transfected with transcription
cassettes for GAL4 DBD (1-147) CfEcR(DEF) and for VP16
TAD bRXR-LmUSP ECH9 controlled by ubiquitously active
cellular promoters (PGK and EF-1a, respectively) on a single
plasmid. Stably transfected cells were selected by Zeocin resis-
tance. Individually isolated CHO cell clones were transiently
transfected with a GAL4 RE-luciferase reporter (pFR Luc).
27-63 clone was selected using Hygromycin. Cells were trypsi-
nized and diluted to a concentration of 2.5 ꢂ 104 cells mL.
100 mL of cell suspension was placed in each well of a 96-well
18. (a) Dhadialla, T. S.; Carlson, G. R.; Le, D. P. Ann. Rev.
Entomology 1998, 43, 545. (b) Hormann, R. E. unpublished
results.
19. CCDC 199585 contains the supplementary crystal-
lographic data for compounds 1f. This data can be obtained
free of charge via www.ccdc.cam.ac.uk/conts/retrieving.html
(or from the CCDC, 12 Union Road, Cambridge CB2 1EZ,
UK; fax: +44 1223 336033; e-mail: deposit@ccdc.cam.ac.uk).
20. The unit cell of 2b consisted of one molecule of CH2Cl2
and two molecules of the dibenzoylhydrazine in slightly dif-
ferent conformers. Hormann, R. E. unpublished results.
21. X-ray structures of crystals of 2a crystallized from differ-
ent solvents have exhibited both rotamers observed for 1f.
Hormann, R. E. unpublished results.
22. Cremlyn, R. J. W.; Ellam, R. M.; Mitra, T. K. J. Chem.
Soc., Perkin Trans. 1 1972, 1727-1730.
23. Lidert, Z.; Hormann, R. E.; Le, D. P.; Opie, T. R. U.S.
Patent 5,344,958, 1994; Chem. Abstr. 1995, 122, 314557.
24. Munk, S. A.; Harcourt, D. A.; Arasingham, P. N.; Burke,
J. A.; Kharlamb, A. B.; Manlapaz, C. A.; Padillo, E. U.;
Roberts, D.; Runde, E.; Williams, L.; Wheeler, L. A.; Garst,
M. E. J. Med. Chem. 1997, 40, 18.
25. Garcia, J.; Nicolas, E.; Albericio, F.; Michelotti, E. L.;
Tice, C. M. Tetrahedron Lett. 2002, 43, 7495.
26. Spectral data for 1p: 1H NMR (CDCl3) d 1.01 (t,
J=7.6 Hz, 3H), 1.3–2.1 (10H), 2.32 (s, 6H), 2.45 (q, J=7.6 Hz,
2H), 4.25 (s, 3H), 6.20 (s, 1H), 6.69 (m, 2H), 7.08 (s, 1H), 7.51
(s, 2H); MS (ESI -ve ion) m/z 420 (M-1).
27. Bobowski, G.; Gottlieb, J. M.; West, B. J. Heterocyclic
Chem. 1980, 17, 1563.