Estrogen receptor ligands: Design and synthesis of new 2-arylindene-1-ones

Article abstract of DOI:10.1016/j.bmcl.2005.04.013

The syntheses of a series of 2-arylindene-1-ones as potent ligands of ERβ and ERα are described. Several compounds exhibited high potency and moderate selectivity for the ERβ receptor. X-ray and modeling studies were used to understand ligand binding orientation and observed affinity.

Full text of DOI:10.1016/j.bmcl.2005.04.013

Bioorganic & Medicinal Chemistry Letters 15 (2005) 3137–3142
Estrogen receptor ligands: design and synthesis of
new 2-arylindene-1-ones
Robert E. McDevitt,a,* Michael S. Malamas,a Eric S. Manas,c Rayomand J. Unwalla,c
Zhang B. Xu,d Chris P. Millere and Heather A. Harrisb
aDepartment of Chemical and Screening Sciences, Wyeth Research, CN 8000, Princeton, NJ 08543-8000, USA
bWomen’s Health Research Institute,Wyeth Research, Collegeville, PA 19426, USA
cDepartment of Chemical and Screening Sciences, Wyeth Research, Collegeville, PA 19426, USA
dDepartment of Chemical and Screening Sciences, Wyeth Research, 200 Cambridge Park Drive, Cambridge, MA 02140, USA
eCorp. Intellectual Property, Glaxo SmithKline, 709 Swedeland Road, King of Prussia, PA 19406, USA
Received 3 February 2005; revised 6 April 2005; accepted 8 April 2005
Available online 3 May 2005
Abstract—The syntheses of a series of 2-arylindene-1-ones as potent ligands of ERb and ERa are described. Several compounds
exhibited high potency and moderate selectivity for the ERb receptor. X-ray and modeling studies were used to understand ligand
binding orientation and observed affinity.
Ó 2005 Elsevier Ltd. All rights reserved.
The effects of estrogens in mammalian tissues have been
well documented and it is now appreciated that estro-
gen affects many organ systems.1 Estrogen can exert
effects on tissues in several ways; the most characterized
mechanism of action is its interaction with estrogen
receptors leading to alterations in gene transcription.
Estrogen receptors (ER) are ligand-activated transcrip-
tion factors belonging to the nuclear hormone receptor
superfamily, which also includes the progesterone,
androgen, glucocorticoid, and mineralocorticoid
receptors.2
Met421.6 Early studies of ERb focused on defining its
affinity for a variety of ligands and some differences be-
tween that of ERa were observed. The tissue distribu-
tion of ERb has been well mapped in the rodent and
it has been shown not to completely overlap with
ERa. For example, tissues of the mouse and rat uterus
express predominantly ERa, whereas the mouse and
rat lung express predominantly ERb.8 It has also been
reported that within the same organ the distribution of
ERa and ERb can be compartmentalized. As example,
in the mouse ovary, ERb is highly expressed in the
granulosa cells and ERa is restricted to the thecal
and stromal cells.9
To date two estrogen receptors have been reported.
The first receptor was cloned in 1986 and has been
termed ERa.3 The second form of the receptor, ERb
was found in 1996 by Gustafsson and co-workers.4
The three dimensional structures of ERa5 and ERb6
have been solved by co-crystallization with various
ligands. The ligand binding domain of these receptors
are homologous, with 58% amino acid identity when
comparing their primary structures. However, the
well-recognized ligand-binding cavity differs by only
two amino acids.7 In the ERb receptor, Met336 replaces
the ERa Leu384 residue as well as Ile373 replacing
Recently it has been reported that the traditional utility
of ER ligands for such therapies as hormone replace-
ment and contraception is mediated primarily via
ERa.10 The most potent endogenous ligand for both
estrogen receptors is 17b-estradiol (E2).
The limited availability of ERb selective agents has led
to several groups exploring this unmet need. A prolifer-
ation of work in this area has been documented recently,
and some examples of new ligands include biphenols,11
aryl benzothiophenes,12 bicyclo[3.3.1] nonanes,13 and
aryl diphenolic azoles.14 Herein we report our findings
on the activity and selectivity of 2-substituted aryl
indenones.
*
Corresponding author. Tel.: +1 732 274 4554; fax: +1 732 274
4505; e-mail: mcdevib@wyeth.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.04.013

3138
R. E. McDevitt et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3137–3142
aryl-indenes and -indenones as ligands for the estrogen
receptor.17 This work provided insight into the spatial/
torsional requirements for binding to the estrogen recep-
tor of indene and indenone systems. Notably, the orient-
ation preference of 2-indene systems was determined
through varying a single hydroxyl group on the A and
C rings of the core systems. This work suggested that
the preferred mode of binding for the 2-aryl-3-ethyl
indenone core was that the pendant C-2 p-hydroxyphe-
nyl A-ring overlaps with that of the A ring in estradiol.
These studies focused on the orientation of the mono-
phenolic compounds and did not discuss selectivity for
the ER subtypes. They also found that dihydroxyl-2,3-
diarylindene systems bind with greater affinity than their
respective monohydroxyl counterparts.18 We utilized a
structure based X-ray approach with the advanced
SAR of our program to rapidly develop high affinity
estrogen receptor ligands through the 2-arylindenone
system.
Figure 1.
The indenone core was synthesized quickly and effi-
ciently by the method of Johnson and co-workers19
Methoxy phenylacetonitrile 4 was treated with commer-
cially available benzoates 5 in the presence of excess
lithium diisopropyl amine to produce substituted 3-ami-
no-2-indene-1-ones 6 (Scheme 1). Indenones 6 were then
deprotected with boron tribromide to give the diphen-
olic-3-amino-indenone 7 or converted to the indan-1,
3-diones 8 under acid hydrolysis. The indandiones 8 rep-
resented in their enolic forms were converted to the
diphenolic derivatives 9 upon treatment with pyridine
hydrochloride at 180 °C. Alternatively, conversion of
the indandiones 8 to the 3-bromoindenones 10 was
effected with carbon tetrabromide and triphenyl phos-
phine. The regioisomeric forms of the products were
separated by column chromatography and the position
of the C-5 versus C-6 methoxy group established
spectroscopically.
Our work stems from initial structure–activity relation-
ship (SAR) studies leading to benzofuran 1,15 which
possesses an ERb IC50 = 6 nM, and 30-fold selectivity
over ERa (Fig. 1). As part of our studies we undertook
replacing the central core of the benzofuran 1 with an
indenone core. With this change it was hypothesized
that analogs would mimic the binding activity of the
phytoestrogen genistein as well as allowing a handle to
incorporate the C-5 hydroxy group of genistein into
the indenone core. Maintaining the positions of the A
and C ring hydroxyls leads to two regioisomeric inde-
none forms, the 6-hydroxy series 2 and the 5-hydroxy
series 3 (resembling genistein).
Earlier work by Katzenellenbogen and co-workers dis-
closed 2,3-diaryl-indenes and -indenones16 as well as 2-
Scheme 1. Reagents and conditions: (a) LDA, THF, À10 °C; (b) BBr3, CH2Cl2; (c) 20% H2SO4, 100 °C; (d) pyridine–HCl; 180 °C; (e) NaHMDS,
THF, 0 °C; (f) R1MgBr or R1Li, THF, 0 °C–rt; R1 = Me, Ph; (g) CBr4, PPh3, CHCl3; (h) R2–NH2 or R2–SNa, DMF, 60–80 °C; R2 = Me, Et.

R. E. McDevitt et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3137–3142
3139
The 3-bromo-indenones 10 were doubly deprotected to
give 11 or alternatively subjected to nucleophilic dis-
placement with either substituted amines or the sodium
salts of alkyl thiols to give the substituted indenones.
The indenones were subsequently deprotected with
either pyridine hydrochloride or boron tribromide to af-
ford the final diphenolic indenones 12. In general indan-
diones 8 were converted to the 3-substituted indenones
upon treatment with sodium hexamethyldisilazane fol-
lowed by an alkyl Grignard or alkyl lithium reagent.
Demethylation with concomitant dehydration produced
the phenolic indenones 13. All of the compounds were
chemically characterized by melting point, infrared, nu-
clear magnetic resonance (1H NMR), and elemental
analysis or HRMS.
A competitive radioligand binding assay was used to as-
sess the relative binding affinity (IC50) of compounds for
the human ligand binding domains (LBD) of ERb and
ERa.20 Table 1 shows the binding affinities for analogs
of the 6-hydroxy series 2.
Figure 2. Unbiased 2fo À fc map contoured at r, showing the electron
density for compound 14 complexed with ERb.
molecules, and 167 water molecules. The corresponding
refinement values obtained were R = 22.6% and
Rfree = 28.6%. Atomic coordinates for ERb complexed
with compound 14 has been deposited in the Protein
Data Bank, with accession code 1ZAF.
One of the more selective derivatives, 14 was co-crystal-
lized with human ERb (space group P212121) as de-
˚
scribed in Ref. 21a. X-ray data (2.0 A) were collected
at 100 K using a Quantum-4 CCD area detector at the
Advanced Light Source (ALS, Berkeley, CA), and
processed using DENZO and SCALEPACK.22 The
crystal structure was solved by molecular replacement
AMORE23 using the ERb/genistein complex as a search
model. Crystallographic refinement was performed
using CNS.24 The refined models (with no ligand added)
were then used to calculate electron density difference
maps, which showed clear electron density for the bound
compound (see Fig. 2). The final model for the ERb/14
complex contained two protein molecules, two ligand
The crystal structure shows definitive orientation of the
phenolic hydroxyl group as the ÔA-ringÕ, forming hydro-
gen bonds to Glu305 and Arg346 similar to 17b-estra-
diol,25 genistein,26 and other ER ligands.21a,27 The
6-hydroxy group forms a hydrogen bond with His475
.
The remainder of the indenone core fills the rest of the
primarily hydrophobic pocket, forming a torsion angle
of approximately 39°(averaged over both monomers)
with respect to the A-ring.
Table 1. 6-Hydroxy binding affinities (IC50) for human ERa and ERb ligand binding domain
Compd
R4
R3
ERb IC50 (nM)a
ERa IC50 (nM)a
ERb/ERa ratio
E2
14
15
16
17
18
19
20
21
22
23
24
25
26
27
—
H
—
Br
3.6 1.6 (144)
20 19 (7)
11 (1)
3.2 1.0 (144)
290 250 (7)
55 (1)
1
15
5
H
CH3
SMe
SEt
Ph
H
3 (1)
8 (1)
12 (1)
10 (1)
4
H
1
H
9 1 (2)
150 (1)
9 1 (2)
200 (1)
1
OMe
H
Br
1
OH
OH
NH2
NHEt
Br
2625 (1)
1380 (1)
1150 (1)
2700 (1)
267 108 (5)
2 (1)
>5000 (1)
>5000 (1)
>5000 (1)
>5000 (1)
710 670 (4)
—
—
—
—
3
OH
H
OH
OH
OH
OH
OH
SMe
SEt
Ph
2
1 (2)
1
150 (1)
44 (1)
580 (1)
47 (1)
4
1
a Values are the mean of independent determinations (number in parenthesis) SD.

3140
R. E. McDevitt et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3137–3142
in improved potency at both receptor subtypes. This
observation is consistent with the fact that the pocket
residues are conserved in the vicinity of the C-3 position.
Increasing size at the C-3 position with a bulky phenyl
group (i.e., 18) resulted in similar potency at both
LBDÕs. Given the available space in this region of the
pocket, it is unclear whether or not the binding mode
shown in Figures 2 and 3 is maintained for the 3-phenyl
substituent. Polar groups (i.e., 20–23) at the 3-position
resulted in a dramatic loss of potency, most likely due
to desolvation of hydrophilic moieties without making
any significant compensating interactions. A similar
observation was made by Katzenellenbogen and co-
workers13a with a TrogerÕs base analog in their bicy-
clo[3.3.1]nonane series. They reasoned that the central
core polar bridgehead nitrogen atoms were responsible
for the observed low binding affinity. Except for analog
25, placing a hydroxyl group at the C-4 position of this
series also resulted in lower binding affinity (i.e., 24, 26,
and 27) versus their non C-4 hydroxylated counterparts
(i.e., 14, 17, and 18). This result suggests that a hydroxyl
group not intramolecularly hydrogen bonded as in geni-
stein at this position disrupts the lipophilic character re-
quired for tight binding to ERb.
Figure 3. ERb complexed with 14 (green), overlaid with the ERb/
genistein complex (white). Only key residues and a Connolly surface of
the ERb/14 binding site are shown for simplicity.
Figure 3 shows an overlay of the ERb/genistein complex
with that of ERb/14. Notice that the A/C rings and the
corresponding hydroxyl groups overlap well. One inter-
esting difference is that the carbonyl groups of the two
ligands are oppositely oriented in the binding pocket,
which appears to be a consequence of interactions made
Table 2 shows the binding affinities for the 5-hydroxy
analogs (series 3). Our initial target 28 possessing the
C-3 Br substitution had high affinity for the ERb ligand
binding domain (IC50 = 1.7 nM). This is slightly more
potent than its regioisomer 14, which has an affinity of
20 nM. Docking studies, performed as previously descri-
bed11a,21a suggest that the 5-hydroxy group in series 3
forms a hydrogen bond to His475 and directs the car-
bonyl group in a similar orientation to that of genistein
(in contrast to what we observed for series 2). Replacing
the C-5 hydroxyl with a methoxy group 29 disrupted the
hydrogen bonding interaction with His475 resulting in a
117-fold decrease in potency, although maintaining a 9-
fold selectivity versus ERa. Subsequent C-3 replacement
with small alkyl groups (methyl) 30, or a thiomethyl
group 31 led to similar potency (2 nM) but no improve-
ment in selectivity. The C-3 thioethyl analog 32 showed
between the C-ring hydroxyl and His475
.
This will be discussed further below. Based on the over-
lay, it appears that the core selectivity of the indenone
scaffold is due to interactions of the aromatic B-C ring
with ERb Met336, similar to what has been described
for genistein,21b and other compounds.21a As such, it is
likely that variations in the indenone substituents (as
well as the substitution pattern) can lead to subtle elec-
tronic differences that modulate this methionine–aro-
matic interaction, leading to slight variations in
selectivity. In this series our goal was to exploit the
C-3 position by enhancing interaction with the lipophilic
pocket in which the bromine resides. Small changes to
substituents at the C-3 position (i.e., 15–17), resulted
Table 2. 5-Hydroxy binding affinities (IC50) for human ERa and ERb ligand binding domain
Compd
R5
R7
R3
ERb IC50 (nM)a
ERa IC50 (nM)a
ERb/ERa ratio
28
29
30
31
32
33
34
35
H
Me
H
H
H
H
H
H
H
Br
Br
1.7 0.4 (11)
199 22 (3)
2 (1)
5.2 0.21 (2)b
1800 900 (3)
6 (1)
3
9
H
H
Me
SMe
SEt
Ph
3
H
1.3 0.5 (2)
150 (1)
3.1 0.1 (2)
480 (1)
3
H
3
H
10 7 (2)
2.4 .0.3 (2)
3 (1)
10 7 (2)
31 15 (2)
3 (1)
1
OH
OH
Me
Ph
13
1
a Values are the mean of independent determinations (number in parenthesis) SD.
b Performance of 28 on the ERa receptor was atypical and the value reported are for the best curves.

R. E. McDevitt et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3137–3142
3141
Scheme 2. Reagents: (a) HI (57%), AcOH.
E. M.; Nilsson, O.; Holst, M.; Savendahl, L.; Wroblewski,
J. J. Steroid Biochem. Mol. Biol. 2000, 74, 383.
2. Tsai, M. J.; OÕMalley, B. W. Annu. Rev. Biochem. 1994, 63,
451.
3. Green, S.; Walter, P.; Kumar, V.; Krust, A.; Bornert, J.
M.; Argos, P.; Chambon, P. Nature 1986, 320, 134.
4. Kuiper, G. G. J. M.; Enmark, E.; Pelto-Huikko, M.;
Nilsson, S.; Gustafsson, J. A. Proc. Natl. Acad. Sci.
U.S.A. 1996, 93, 5925.
decreased affinity for both receptors. As observed in the
6-hydroxy series, placing a phenyl group at the 3-posi-
tion 33 gave equivalent potency (10 nM) at both ERa
and ERb. The 5,7-dihydroxy analog 34, being the most
structurally similar to genistein, bound tightly to ERb
(2.4 nM) and was 13-fold selective. Docking studies sug-
gest that 34 adopts a binding mode similar to that of
genistein.
5. Brzozowski, A. M.; Pike, A. C. W.; Dauter, Z.; Hubbard,
R. E.; Bonn, T.; Engstrom, O.; Ohman, L.; Greene, G.;
Gustafsson, J. A.; Carlquist, M. Nature 1997, 389, 753.
6. Pike, A. C. W.; Brzozowski, A. M.; Hubbard, R. E.;
Bonn, T.; Thorsell, A.-G.; Engstrom, O.; Ljunggren, J.;
Ohman, L.; Gustafsson, J. A.; Carlquist, M. EMBO J.
1999, 18, 4608.
Finally, we investigated the planarity requirements of
the indenone core. As shown in Scheme 2, enone 10
could be simultaneously deprotected and reduced with
HI (57%) in acetic acid to afford the indane 36. This ana-
log displayed a significant decrease in potency at ERb
(700 nM) and ERa (690 nM), with loss of selectivity as
compared to 14. While there may be some contribution
to this decrease in affinity due to removal of the bromo
group, it is hypothesized that the primary contribution
is most likely disruption of the core scaffold shape, pre-
venting optimal orientation of the key hydroxyl groups.
7. Mosselman, S.; Polman, J.; Dijkema, R. FEBS Lett. 1996,
392, 49.
8. (a) Couse, J. F.; Lindzey, J.; Grandien, K.; Gustafsson, J.
A. Endocrinology 1997, 138, 4613; (b) Kuiper, G. G. J. M.;
Carlsson, B.; Grandien, K.; Enmark, E.; Haggblad, J.;
Nilsson, S.; Gustafsson, J. A. Endocrinology 1997, 138,
863.
9. (a) Sar, M.; Welsch, F. Endocrinology 1999, 140, 963; (b)
Fitzpatrick, S. L.; Funkhouser, J. M.; Sindoni, D. M.;
Stevis, P. E.; Deecher, D. C.; Bapat, A. R.; Merchenthaler,
I.; Frail, D. E. Endrocrinology 1999, 140, 2581.
10. (a) Sun, J.; Meyers, M. J.; Fink, B. E.; Rajendran, R.;
Katzenellenbogen, J. A.; Katzenellenbogen, B. S. Endo-
crinology 1999, 140, 800; (b) Harris, H. A.; Katzenellen-
bogen, J. A.; Katzenellenbogen, B. S. Endocrinology 2002,
143, 4172.
11. (a) Edsall, R. J.; Harris, H. A.; Manas, E. S.; Mewshaw,
R. E. Bioorg. Med. Chem. 2003, 11, 3457; (b) Yang, C.;
Edsall, R. J.; Harris, H. A.; Zhang, X.; Manas, E. S.;
Mewshaw, R. E. Bioorg. Med. Chem. 2004, 12, 2553.
12. Schopfer, U.; Schoeffter, P.; Bischoff, S. F.; Nozulak, J.;
Feuerbach, D.; Floersheim, P. J. Med. Chem. 2002, 45,
1399.
In summary, we have demonstrated that both the 5- and
6-hydroxy indenone cores bind potently to ERb and
ERa and show moderate ERb selectivity. Compound
14 was co-crystallized with ERb. The binding mode
was found to be similar to that of genistein, with the
exception that the direction of the carbonyl group is re-
versed. This appears to be influenced by the way the 6-
hydroxy group is directed toward its hydrogen bonding
partner His475. A more genistein-like binding mode,
with the carbonyl moiety directed toward a similar re-
gion of the binding pocket as that of genistein, is pre-
dicted for 5-hydroxy analog 34 based on docking
calculations. Analog 34 showed similar binding potency
but 2–3-fold less selectivity over its genistein
counterpart.
13. (a) Muthyala, R. S.; Carlson, K. E.; Katzenellenbogen, J.
A. Bioorg. Med. Chem. Lett. 2003, 13, 4485; (b) Shibley,
R.; Hatoum-Mokdad, H.; Schoenleber, R.; Musza, L.;
Stirtan, W.; Marrero, D.; Carley, W.; Ziao, H.; Dumas, J.
Bioorg. Med. Chem. Lett. 2003, 13, 1919.
Acknowledgements
14. Malamas, M. S.; Manas, E. S.; McDevitt, R. E.; Guna-
wan, I.; Xu, Z. B.; Collini, M. D.; Miller, C. P.; Dihn, T.;
Bray, J.; Henderson, R. A.; Keith, J. C.; Harris, H. A. J.
Med. Chem. 2004, 47, 5021.
We thank the Wyeth Discovery Analytical Chemistry
department for physical analysis and Dr. Al Robichaud
for advice and discussion.
15. Collini, M. D.; Kaufman, D. H.; Manas, E. S.; Harris, H.
A.; Henderson, R. A.; Xu, Z. B.; Unwalla, R. J.; Miller, C.
P. Bioorg. Med. Chem. Lett. 2004, 14, 4925.
References and notes
16. (a) Anstead, G. M.; Altenbach, R. J.; Wilson, S. R.;
Katzenellenbogen, J. A. J. Med. Chem. 1988, 31, 1316; (b)
Anstead, G. M.; Peterson, C. S.; Pinney, K. G.; Wilson, S.
R.; Katzenellenbogen, J. A. J. Med. Chem. 1990, 33, 2726.
17. (a) Anstead, G. M.; Ensign, J. L.; Peterson, C. S.;
Katzenellenbogen, J. A. J. Org. Chem. 1989, 54, 1485;
(b) Anstead, G. M.; Wilson, S. R.; Katzenellenbogen, J. A.
J. Med. Chem. 1989, 32, 2163.
1. (a) Mendelsohn, M. E.; Karas, R. H. New Engl. J. Med.
1999, 340, 1801; (b) Epperson, C. N.; Wisner, K. L.;
Yamamoto, B. Psychosomatic Med. 1999, 61, 676; (c)
Crandall, J. Womens Health Gender Based Med. 1999, 8,
1155; (d) Hurn, P. D.; Macrae, I. M. J. Cerebral Blood
Flow Metab. 2000, 20, 631; (e) Toran-Allerand, C. D. J.
Steroid Biochem. Mol. Biol. 1996, 56, 169–178; (f) Ritzen,

3142
R. E. McDevitt et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3137–3142
18. Anstead, G. M.; Peterson, C. S.; Katzenellenbogen, J. A.
J. Steroid Biochem. 1989, 33, 887.
19. Kayaleh, N. E.; Gupta, R. C.; Morrissey, J. F.; Johnson,
F. Tetrahedron Lett. 1997, 38, 8121.
20. Harris, H. A.; Bapat, A. R.; Gonder, D. S.; Frail, D. E.
Steroids 2002, 67, 379.
21. (a) Manas, E. S.; Unwalla, R. J.; Xu, Z. B.; Malamas, M.
S.; Miller, C. P.; Harris, H. A.; Hsiao, C.; Akopian, T.;
Hum, W.-T.; Malakian, K.; Wolfrom, S.; Bapat, A.; Bhat,
R. A.; Stahl, M. L.; Somers, W. S.; Alvarez, J. C. J. Am.
Chem. Soc. 2004, 126, 15106; (b) Manas, E. S.; Xu, Z. B.;
Unwalla, R. J.; Somers, W. S. Structure 2004, 12, 2197.
22. Otwinowski, Z.; Minor, W. Methods Enzymol. 1994, 276,
307.
23. Bailey, S. Acta Crystallogr., Sect. D 1994, 50, 760.
24. Brunger, A. Acta Crystallogr., Sect. D 1998, 55, 941.
25. (a) Brzozowski, A. M.; Pike, A. C. W.; Dauter, Z.;
Hubbard, R. E.; Bonn, T.; Engstrom, O.; Ohman, L.;
Greene, G. L.; Gustafsson, J.-A.; Carlquist, M. Nature
1997, 389, 753; (b) Warnmark, A.; Treuter, E.; Gustafs-
son, J. A.; Hubbard, R. E.; Brzozowski, A. M.; Pike, A. C.
W. J. Biol. Chem. 2002, 277, 21862.
26. Pike, A. C. W.; Brzozowski, A. M.; Hubbard, R. E.;
Bonn, T.; Thorsell, A. G.; Engstrom, O.; Ljunggren, J.;
Gustafsson, J. K.; Carlquist, M. EMBO J. 1999, 18, 4608.
27. Shiau, A. K.; Barstad, D.; Loria Paula, M.; Cheng, L.;
Kushner Peter, J.; Agard David, A.; Greene Geoffrey, L.
Cell 1998, 95, 927.