Synthesis and erythropoietin receptor binding affinities of N, N-disubstituted amino acids

Article abstract of DOI:10.1016/S0960-894X(00)00399-1

N,N-Dicinnamyl, N-benzyl-N-cinnamyl, and N,N-dibenzyl amino acids were prepared and evaluated in an EPO binding assay. Several derivatives of aspartic acid, glutamic acid, and lysine exhibited moderate (10-50 μM) affinity for EBP; 'dimerization' of the most potent analogues by coupling with linear diamines led to EPO competitors having 1-2 μM binding affinities. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00399-1

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1995±1999
Synthesis and Erythropoietin Receptor Binding Anities of
N,N-Disubstituted Amino Acids
Peter J. Connolly,* Steven K. Wetter, William V. Murray, Dana L. Johnson,
Frank J. McMahon, Francis X. Farrell, Jennifer Tullai and Linda K. Jolli€e
The R. W. Johnson Pharmaceutical Research Institute, 1000 Route 202, Raritan, NJ 08869, USA
Received 13 April 2000; accepted 30 June 2000
AbstractÐN,N-Dicinnamyl, N-benzyl-N-cinnamyl, and N,N-dibenzyl amino acids were prepared and evaluated in an EPO binding
assay. Several derivatives of aspartic acid, glutamic acid, and lysine exhibited moderate (10±50 mM) anity for EBP; `dimerization'
of the most potent analogues by coupling with linear diamines led to EPO competitors having 1±2 mM binding anities. # 2000
Elsevier Science Ltd. All rights reserved.
Introduction
Erythropoietin (EPO) is a 34 kD glycoprotein hormone
that is the principal growth factor responsible for the
regulation of red blood cell production in humans.1
Over the past decade, recombinant human EPO has
become a mainstay in the treatment of anemias resulting
from chronic renal failure, chemotherapy, and HIV-
infection. Because of the inconvenience and expense of
therapy involving an injectable protein like EPO, there is
considerable interest in an orally available, small mole-
cule EPO mimetic. However, designing a nonprotein
molecule with the ability to mimic the complex interac-
tion between a protein hormone and its receptor has
proved to be challenging. Literature references to small
molecule antagonists of cytokine receptors (i.e., small
molecule inhibitors of interleukin-2 binding to its recep-
tor2) predominate. On the other hand, the past three
years have witnessed reports of nonprotein cytokine
receptor agonists such as a nonpeptide mimic of
granulocyte-colony stimulating factor3 and an EPO
receptor-derived peptide that activates receptor signal-
ling.4 Previously, our group had disclosed a series of
cyclic peptides, exempli®ed by EMP1 (1), possessing
potent EPO-mimetic activity.5,6 A recent publication
describing a nonpeptide molecule having EPO-mimetic
activity (2b, derived from EPO receptor antagonist 2a)7
prompted us to disclose our work toward EPO-mimetic
small molecules.
Synthesis and Biological Results
Based on our experience with EPO-mimetic peptides
and the desire to explore easily accessible chemistry, we
synthesized a series of N,N-dicinnamyl amino acids and
their corresponding esters for screening in an immobi-
lized EPO receptor (EBP) binding assay (Scheme 1).8
Starting from side-chain-protected amino acid esters 3
(b- or g-t-butyl ester for aspartic and glutamic acid, t-
butyl ether for serine, e-Boc for lysine), alkylation with
the appropriately substituted cinnamyl bromide9 resul-
ted in easily separated mixtures of di- and monoalkylated
amino esters (4 and 5; X=trans-CHˆCHCH2). The
puri®ed mono- and dicinnamyl compounds were tested
directly in the EBP binding assay or were optionally
hydrolyzed to the respective amino acids using reagents
appropriate for the nature of the ester (50% TFA/
CH2Cl2 for a-t-butyl esters; NaOH for a-methyl esters).
t-Butyl and N-Boc side-chain protecting groups were
removed using 50% TFA/CH2Cl2, and the lysine e-amino
*Corresponding author. Tel.:+1-908-704-4928; fax:+1-908-526-6469;
e-mail: pconnoll@prius.jnj.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00399-1

1996
P. J. Connolly et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1995±1999
Table 1.
EBP bindinga
Compound
R1
R2
R3
% inh,
50 mM
IC50,
mM
4a
4b
4c
4d
4e
4f
Me
H
Me
H
Me
Me
Me
H
CH2CH2CO2-t-Bu
CH2CH2CO2-t-Bu
CH2CH2CO2H
CH2CH2CO2H
(CH2)4NHBoc
CH2O-t-Bu
PhO
PhO
PhO
PhO
PhO
PhO
PhO
PhO
6
0
6
0
20
8
ndb
nd
nd
nd
nd
nd
nd
nd
4g
4h
CH2OH
CH2OH
28
23
aInhibition of 125I-EPO binding to EBP.8
bnd=Not determined.
Table 2.
EBP bindinga
% inh, IC50,
Compound R1
R2
R3
50 mM
mM
5a
5b
5c
5d
5e
5f
5g
5h
5i
5j
5k
5l
5m
5n
5o
5p
Me CH2CH2CO2-t-Bu
PhO
PhO
3,4-Cl2PhO
4-t-Bu-PhO
3-CF3-PhO
PhO
0
78
32
78
70
20
81
0
73
77
0
18
0
ndb
42
nd
30
35
nd
36
nd
35
35
nd
nd
nd
145
40
H
Me
H
H
t-Bu
H
CH2CH2CO2H
CH2CH2CO2H
CH2CH2CO2H
CH2CH2CO2H
CH2CO2-t-Bu
CH2CO2H
Scheme 1. (a) R3C6H4-X-Br or R3C6H4-X-Cl/NaI, DIEA, DMF; (b)
50% TFA/CH2Cl2 (R1=t-Bu) or aqueous NaOH (R1=Me); (c)
R3C6H4-CHO, NaBH(OAc)3, (MeO)3CH; (d) R4C6H4-CH2Br or
R4C6H4-CH2Cl/NaI, DIEA, DMF; (e) 50% TFA/CH2Cl2; (f) MsCl,
Et3N or TMS-NCO, Et3N.
PhO
PhO
PhO
PhO
Me
(CH2)4NHBoc
Me (CH2)4NHCONH2
Me (CH2)4NHSO2Me
group was further converted to the corresponding urea
or sulfonamide.
Me
Me
Me
Me
H
(CH2)4NHBoc
(CH2)4NH2
CH2O-t-Bu
CH2OH
CH3
4-t-Bu-PhO
4-t-Bu-PhO
PhO
Many of the substituted amino acid derivatives were
only weakly competitive with 125I-EPO at EBP (Tables 1
and 2). In particular, the monocinnamyl amino acids and
esters (4) were uniformly inactive, as were dicinnamyl
derivatives (5) that contained protected side-chain func-
tional groups. However, signi®cant binding anities
(IC50<50 mM) were observed for dicinnamyl com-
pounds having a free carboxylate, with the most potent
analogues coming from the aspartic/glutamic acid family
(5b, 5d, 5e, and 5g). Alanine derivative 5o and lysine
derivatives having polar substituents on the side-chain
nitrogen (5i and 5j) were also potent. Interestingly, the
free e-amino compound (5l) was inactive.
PhO
PhO
PhO
57
72
45
H
H
85
aInhibition of 125I-EPO binding to EBP.8
bnd=Not determined.
substituted benzyl halides. This reaction also provided
the corresponding mono-N-benzyl derivatives
(X=6;CH2) which, after separation, were further
alkylated with other benzyl halides to give unsym-
4
metrically substituted N,N-dibenzyl analogues
6
(XˆCH2). The intermediate mono-N-benzyl amino
acid esters 4 (XˆCH2) were alternatively prepared by
reductive alkylation using a substituted benzaldehyde,
(MeO)3CH, and NaBH(OAc)3.10
Additional aspartic acid and lysine analogues were syn-
thesized using a modi®cation of the route described
above (Scheme 1). To prepare N-benzyl-N-cinnamyl
derivatives, N-cinnamyl amino acid esters 4 (X=trans-
CHˆCHCH2) were reacted with DIEA and an appro-
priately substituted benzyl halide in DMF. Treatment of
the resulting side-chain-protected products with TFA in
CH2Cl2 provided 6 (X=trans-CHˆCHCH2). Symme-
trically substituted N,N-dibenzyl analogues 5 (X=CH2)
were prepared by alkylation of amino acid esters 3 with
Several N,N-dibenzyl- and N-benzyl-N-cinnamyl deri-
vatives (6a, 6b, and 6d±f) were similar in potency to the
best dicinnamyl compounds in the EPO competition
assay (Table 3). However, inhibition of EPO binding at
the screening dose of 50 mM was only modest, indicating
relatively ¯at dose±response curves. Once again, side-
chain-protected compounds (6c and 6h±j) were inactive,

P. J. Connolly et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1995±1999
1997
Table 3.
EBP bindinga
% inh, 50 mM
Compound
X
R2
R3
R4
IC50 mM
6a
6b
6c
6d
6e
6f
6g
6h
6i
CHˆCHCH2
CHˆCHCH2
CHˆCHCH2
CHˆCHCH2
CH2
(CH2)4NH2
(CH2)4NH2
(CH2)4NHBoc
CH2CO2H
(CH2)4NH2
(CH2)4NH2
(CH2)4NH2
(CH2)4NHBoc
(CH2)4NHBoc
(CH2)4NHBoc
3-(4-t-Bu-PhO)
3-(PhO)
3-(4-t-Bu-PhO)
3-(PhO)
3-(BnO)
3-BnO
3-PhO
3-BnO
3-PhO
3-BnO
4-BnO
3-NO2
3-BnO
4-BnO
3-NO2
46
60
0
52
59
54
37
7
10
30
ndb
40
10
10
nd
nd
nd
nd
CH2
CH2
CH2
CH2
4-(BnO)
4-(BnO)
3-(BnO)
4-(BnO)
4-(BnO)
5
24
6j
CH2
aInhibition of 125I-EPO binding to EBP.8
bnd=Not determined.
but the free e-amino lysine analogues were fairly potent
competitors, in contrast to the poor anity exhibited by 5l.
were also inactive unless both the a- and o-carboxylates
were present as free acids (5b, 5d, 5e, and 5g). In the
lysine series, side-chain deprotection was also necessary
for EBP binding anity (6a, 6b, 6e, and 6f), although
there was no apparent need for a free a-carboxy group.
Interestingly, conversion of the e-amino group to a urea
(5i) or methanesulfonamide (5j) was accompanied by
moderately potent EBP binding anity, suggesting that
polar functionality, rather than positive charge, was
bene®cial. In order to determine whether the com-
pounds possessed EPO-mimetic activity, the highest
anity analogues were evaluated in an EPO-responsive
FDC-P1 cell proliferation assay but, unlike EMP1 (1),
none were found to be active.5
In an e€ort to improve binding anity, we next pre-
pared `dimeric' analogues in which two moderately
potent EPO competitors were connected by a hydro-
carbon or polyether linking group. For our purposes, a
series of commercially available diamines proved to be a
convenient set of reagents, allowing us to explore the
e€ects of distance and linker hydrophobicity on EBP
anity. Because of the relative potency of benzyloxy-
benzyl derivatives 6a, 6e, and 6f (Table 2), we chose to
make `dimeric' analogues of N,N-di-(4-benxyloxy-
benzyl) amino acids. Toward that end, 2 equiv of
Na-Cbz-protected aspartic acid (b-t-butyl ester), glutamic
acid (g-t-butyl ester), or lysine (Ne-Boc) (7) were cou-
pled with the appropriate diamine using water-soluble
carbodiimide (EDCI) and HOBt/DIEA in CH2Cl2
(Scheme 2). The Cbz group was removed by standard or
catalytic transfer hydrogenation; the resulting diamino
intermediate 8 was exhaustively N-alkylated with 4-
benzyloxybenzyl chloride and sodium iodide in DMF.
After puri®cation, the side-chain-protected compound 9
was treated with TFA in CH2Cl2 to provide the respec-
tive tetrabenzyl derivative 10 (Asp or Glu) or 11 (Lys).
For the lysine series, the side-chain amino group was
further derivatized by reaction with succinic or glutaric
anhydride to provide 12.
It is known that EPO exerts its e€ect at the molecular
level by bringing two molecules of EPO receptor toge-
ther on the cell surface. `Dimeric' targets 10±12 were
designed to exploit this property by linking together two
EPO competitors with the appropriate tether, in theory
converting a small molecule antagonist into an EPO-
mimetic receptor agonist. In practice, tethered com-
pounds 10±12 were highly e€ective EPO competitors,
especially in the lysine series, with many having IC50
values below 10 mM (Table 4). Linker length (11±20 A
between N atoms in the extended conformation) and
composition (hydrophobic vs hydrophilic) a€ected
binding anity, with the most potent derivatives 12a
and 12e (IC50=1.5 mM) arising from the shortest
hydrophilic linker, HNCH2CH2(OCH2CH2)2NH. The
hydrophobic linker, HN(CH2)12NH, was associated
with the lowest anity `dimers' (12b and 12f), possibly
due to poor access to the EPO binding site resulting
from hydrophobic collapse of the linker.
Discussion
As was mentioned previously, EBP binding of `mono-
meric' amino acid analogues 4±6 was a€ected by both
the degree of substitution at the a-nitrogen atom and by
the nature of the substituents on the side chain (Tables
1±3). While all of the mono-N-cinnamyl derivatives
were essentially inactive at 50 mM, many side-chain-
protected dicinnamyl derivatives were equally weak
EPO competitors. Aspartic and glutamic acid analogues
Unfortunately, despite EPO receptor binding anity on
par with that of EPO-mimetic peptide EMP1 (1), the
best `dimeric' analogues (11, 12a, 12c±e, 12g, and 12h)
did not promote proliferation in the FDC-P1 cell
assay.5 Apparently, the tethered construct exempli®ed

1998
P. J. Connolly et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1995±1999
Scheme 2. (a) EDCI, HOBt, DIEA, CH2Cl2; (b) Pd±C, H2 or Pd±C, NH4CO2, MeOH; (c) 4-BnOC6H4CH2Cl, NaI, DMF; (d) 50% TFA/CH2Cl2;
(e) DMAP, CH2Cl2.
Table 4.
acidic or polar residues on the amino acid side chains.
Coupling of two N,N-dibenzyl derivatives to a hydro-
philic diamine produced `dimers' like 12a and 12e hav-
ing 10-fold greater binding anity. Unfortunately,
neither the `dimers' nor the corresponding parent com-
pounds exhibited EPO-mimetic activity in a cell pro-
liferation assay.
References and Notes
EBP bindinga
Compound R1
R2
% inh,
50 mM
IC50,
mM
1. Krantz, S. B. Blood 1991, 77, 419.
2. Tilley, J. W.; Chen, L.; Fry, D. L.; Emerson, S. D.; Powers,
G. D.; Biondi, D.; Varnell, T.; Trilles, R.; Guthrie, R.; Men-
nona, F.; Kaplan, G.; LeMahieu, R. A.; Carson, M.; Han, R.-
J.; Liu, C.-M.; Palermo, R.; Ju, G. J. Am. Chem. Soc. 1997,
119, 7589.
3. Tian, S.-S.; Lamb, P.; King, A. G.; Miller, S. G.; Kessler,
L.; Luengo, J. I.; Averill, L.; Johnson, R. K.; Gleason,
J. G.; Pelus, L. M.; Dillon, S. B.; Rosen, J. Science 1998, 281,
257.
4. Naranda, T.; Wong, K.; Kaufman, R. I.; Goldstein, A.;
Olsson, L. Proc. Natl. Acad, Sci. U.S.A. 1999, 96, 7569.
5. Wrighton, N. C.; Farrell, F. X.; Chang, R.; Kashyap, A.
K.; Barbone, F. P.; Mulcahy, L. S.; Johnson, D. L.; Barrett,
R. W.; Jolli€e, L. K.; Dower, W. J. Science 1996, 273, 458.
6. Johnson, D. L.; Farrell, F. X.; Barbone, F. P.; McMahon,
F. J.; Tullai, J.; Hoey, K.; Livnah, O.; Wrighton, N. C.; Mid-
dleton, S. A.; Loughney, D. A.; Stura, E. A.; Dower, W. J.;
Mulcahy, L. S.; Wilson, I. A.; Jolli€e, L. K. Biochemistry
1998, 37, 3699.
7. Qureshi, S. A.; Kim, R. M.; Konteatis, Z.; Biazzo, D. E.;
Motamedi, H.; Rodrigues, R.; Boice, J. A.; Calaycay, J. R.;
Bednarek, M. A.; Gao, Y.-D.; Chapman, K.; Mark, D. F.
Proc. Natl. Acad, Sci. U.S.A. 1999, 96, 12156.
8. Test compounds were assayed for their ability to displace 125I-
EPO from immobilized EPO receptor extracellular domain
(EPO binding protein, or EBP), as described in ref 6.
9. Substituted cinnamyl bromides were prepared by published
procedures; i.e. Jackson, W. P.; Islip, P. J.; Kneen, G.; Pugh,
A.; Wates, P. J. J. Med. Chem. 1988, 31, 499.
10a
10b
10c
11
12a
12b
12c
12d
12e
12f
12g
12h
1
C
D
A
A
A
B
C
D
A
B
CH2CO2H
CH2CO2H
CH2CH2CO2H
(CH2)4NH2
75
66
79
51
88
55
82
78
93
56
65
77
Ð
18
28
75
2
1.5
85
4
10
1.5
100
6
7
(CH2)4NH-CO(CH2)3CO2H
(CH2)4NH-CO(CH2)3CO2H
(CH2)4NH-CO(CH2)3CO2H
(CH2)4NH-CO(CH2)3CO2H
(CH2)4NH-COCH2CH2CO2H
(CH2)4NH-COCH2CH2CO2H
(CH2)4NH-COCH2CH2CO2H
(CH2)4NH-COCH2CH2CO2H
EMP16
C
D
5
aInhibition of 125I-EPO binding to EBP.8
by 10±12 was not ¯exible enough or could not span the
distance required to bring together two EPO receptor
molecules. The additional binding anity gained by the
`dimer' construct was likely due to added nonspeci®c
hydrophobic interactions outside the EPO binding
pocket of EBP or the EPO receptor.
In conclusion, moderate EBP binding anity was
observed for N,N-disubstituted analogues of lysine,
aspartic acid, and glutamic acid, with a preference for

P. J. Connolly et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1995±1999
1999
10. McNamara, D. J.; Dobrusin, E.; Leonard, D. M.; Shuler,
K. R.; Kaltenbronn, J. S.; Quin, J.; Bur, S.; Thomas, C. E.;
Doherty, A. M.; Scholten, J. D.; Zimmerman, K. K.; Gibbs,
B. S.; Gowan, R. C.; Latash, M. P.; Leopold, W. R.; Przy-
branowski, S. A.; Sebolt-Leopold, J. S. J. Med. Chem. 1997,
40, 3319.