Structure-based design and synthesis of phosphinate isosteres of phosphotyrosine for incorporation in Grb2-SH2 domain inhibitors. Part 2

Article abstract of DOI:10.1016/S0960-894X(00)00476-5

A series of novel phosphinates, derived from 4-phosphonomethylphenylalanine, are described as isosteres of phosphotyrosine. Benzyl (or alkyl) phosphinomethylphenylalanine derivatives were prepared by alkylation of an amino acid P-H phosphinate. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00476-5

Bioorganic & Medicinal Chemistry Letters 10 (2000) 2343±2346
Structure-Based Design and Synthesis of Phosphinate Isosteres
of Phosphotyrosine for Incorporation in Grb2-SH2 Domain
Inhibitors. Part 2
Clive V. Walker,a,* Giorgio Caravatti,b Alastair A. Denholm,a John Egerton,a
Alex Faessler,a Pascal Furet,b Carlos Garcõa-Echeverrõa,b Brigitte Gay,b
Ed Irving,a Ken Jones,a Amanda Lambert,a Neil J. Pressa
and John Woodsa
aNovartis Horsham Research Centre, Wimblehurst Road, Horsham, West Sussex, RH12 5AB, UK
bNovartis Pharmaceuticals, Inc., Therapeutic Area Oncology, CH-4002 Basel, Switzerland
Received 15 March 2000; revised 17 July 2000; accepted 8 August 2000
AbstractÐA series of novel phosphinates, derived from 4-phosphonomethylphenylalanine, are described as isosteres of phos-
photyrosine. Benzyl (or alkyl) phosphinomethylphenylalanine derivatives were prepared by alkylation of an amino acid P-H
phosphinate. # 2000 Elsevier Science Ltd. All rights reserved.
The signal transduction processes mediated by Grb2
(growth factor receptor bound protein 2) are needed for
the activity of growth factors and/or oncogenes that are
responsible for cell growth.1 Deregulation of growth
factor signalling leads to rapid cell growth and is central
to many tumours.2 Thus, control of this signal trans-
duction process, by antagonising the protein±protein
interactions of Grb2-SH2 with activated receptors, was
pursued to identify novel anti-tumour agents.
(Fig. 1) where one charge has been replaced by a neutral
group. The rationale, molecular modelling and biologi-
cal results are described in Part 1 of this work. In Part 2
we describe the synthesis of these novel phosphinate
isosteres 2±6 suitable for incorporation into peptido-
mimetic sequences.
Initially, we chose our synthetic strategy based on hav-
ing a protected amino acid P-H synthon which could be
alkylated at a late stage with di€erent electrophiles.
Thus, alkylation (Scheme 1) of the hypophosphorus
acid derivative 77 with 4-iodobenzyl bromide gave 8.
Zinc mediated Jackson coupling8,6b with either Boc-
iodoAla-OMe9 or Z-iodoAla-OBn10 gave the protected
hypophosphorus amino acid synthons 9a,b in good
yields. The latent P-H functionality was revealed using
TMSCl7 to give 10a,b suitable for reaction with di€erent
electrophiles. Although no special precautions were
taken the P-H compounds were routinely used immedi-
ately after deprotection to avoid any possibility of oxi-
dation to the phosphonates.
Central to this mechanism is the binding of the Grb2-
SH2 domain to a peptide sequence containing phospho-
tyrosine (pY).3 The design of novel peptidomimetic
sequences that are highly potent antagonists of this
interaction in vitro has been described.4 However,
replacement of the central pY recognition element
remains a highly desirable target and the design and
synthesis of pY mimics with improved properties have
been investigated increasingly over recent years.5 The
known amino acid phosphonomethyl Phe (Pmp) is a
stable phosphotyrosine mimic6 which still contains most
of the key recognition elements responsible for activity.
The reaction of 10a,b with electrophiles (Scheme 2)
proceeded readily via either the P(III) species11 or by
direct alkylation7b using NaH as base to give the phos-
phinates 11b±d. The hydroxymethyl compound 11a was
prepared by heating with paraformaldehyde.12 For 11b
a slightly higher yield (68% versus 54%) was obtained
using the P(III) method. Reaction of 10b with benzal-
However, this compound and its synthetically useful
derivatives (for example 1) still represent highly charged
species. We were interested in phosphinate derivatives
*Corresponding author. Tel.: +44-1403-272827; fax: +44-1403-
323837; e-mail: clive.walker@pharma.novartis.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00476-5

2344
C. V. Walker et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2343±2346
Figure 1.
Scheme 1. (a) 4-Iodobenzylbromide, NaH, THF, 10 ꢀC to rt; (b) (i) Boc-iodoAla-OMe, or Z-iodoAla-OBn, Zn, BrCH2CH2Br, TMSCl; (ii) (2-
furyl)3P, Pd2dba3, DMA±THF, 60 ꢀC; (c) TMSCl, CH2Cl2 (or CHCl3), EtOH, rt.
Scheme 2. (a) (CH2O)n, PhMe, Et3N, 100 ꢀC; (b) (i) NaH, MeI, THF, 10 ꢀC to rt or (ii) TMSCl, iPrEt2N, MeI, CH2Cl2; (c) TMSCl, iPrEt2N,
BnBr, CH2Cl2; (d) TMSCl, iPrEt2N, PhCHO, CH2Cl2.

C. V. Walker et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2343±2346
2345
Scheme 3. (a) LiOH, EtOH±H2O, rt; (b) (i) H2, Pd±C; (ii) Fmoc-OSu, Na2CO3, dioxan; (c) (i) HBr±AcOH; (ii) Fmoc-OSu, Py, DMAP; (d) (i) TFA;
(ii) Fmoc-Cl, Dioxan±H2O; (e) (i) H2, Pd±C; (ii) Fmoc-OSu, dioxan; (f) TMSI±CH2Cl2; (g) (i) 1,4-cyclohexadiene, Pd±C; (ii) TMSI±CH2Cl2.
Scheme 4. (a) LDA, ClCO2Et, THF, 78 ꢀC; (b) LiBH4, Et2O, 10±20 ꢀC; (c) KHMDS, BnBr, THF, 78 ꢀC to rt; (d) TMSCl, CHCl3, EtOH, rt;
(e) KHMDS, 4-I-BnBr, THF, 78 ꢀC; (f) (i) Boc-iodoAla-OMe, Zn, BrCH2CH2Br, TMSCl; (ii) (2-furyl)3P, Pd2dba3, DMA±THF, 60 ꢀC; (g) H2,
Pd±C, EtOH; (h) LiOH, EtOH±H2O; (i) (i) TFA; (ii) Fmoc-OSu, 10% aq Na2CO3, dioxan; (j) TMSI, CH2Cl2, rt.

2346
C. V. Walker et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2343±2346
dehyde gave the TMS ether 11d, which was stable to
chromatography, as a mixture of diastereomers.12
C.; Rahuel, J.; Schoepfer, J.; Fretz, H. J. Med. Chem. 1998, 41,
3442. (c) Garcõa-Echeverrõa, C.; Furet, P.; Gay, B.; Fretz, H.;
Rahuel, J.; Schoepfer, J.; Caravatti, G. J. Med. Chem. 1998,
41, 1741. (d) Furet, P.; Garcõa-Echeverrõa, C.; Gay, B.;
Schoepfer, J.; Zeller, M.; Rahuel, J. J. Med. Chem. 1999, 42,
2358.
5. (a) For a recent review see: Burke, T. R.; Yao, Z.-J.; Smyth,
M. S.; Ye, B. Curr. Pharm. Design 1997, 3, 291. (b) For a
recent application to Grb2 see: Yao, Z.-J.; King, C. R.; Cao,
T.; Kelley, J.; Milne, G. W. A.; Voigt, J. H.; Burke, T. R. J.
Med. Chem. 1999, 42, 25.
6. (a) Marseigne, I.; Roques, B. P. J. Org. Chem. 1988, 53,
3621; (b) Dow, R. L.; Bechle, B. M. Synlett 1994, 293.
7. (a) Baylis, E. K. Tetrahedron Lett. 1995, 36, 9385. (b)
Froestl, W.; Mickel, S. J. et al. J. Med. Chem. 1995, 38, 3313.
8. Jackson, R. F. W.; Wishart, N.; Wood, A.; James, K.;
Wythes, M. J. J. Org. Chem. 1992, 57, 3397.
9. Smyth, M. S.; Burke, T. R. Tetrahedron Lett. 1994, 35, 551.
10. Z-Iodo-Ala-OBn: Literature preparation13 of this amino
acid is a two step procedure. We developed a one step proce-
dure (amenable to scale-up) as follows: Z-Ser-OBn (8.2 g,
25 mmol) was placed in a ¯ask under argon with triphenyl-
phosphine (8.56 g, 2.6 mmol) and imidazole (2.36 g, 35 mmol).
These were dissolved in acetonitrile (24.5 mL)/ether (40.9 mL)
and cooled to 0 ꢀC. Iodine (9.13 g, 36 mmol) was added slowly
and the mixture stirred at 0 ꢀC for 45 min, then diluted with
ether (50 mL) and washed with Na2S2O3 (aq) (20 mL), CuSO4
(aq) (20 mL) and water (20 mL). The organic layer was dried
(MgSO4), concentrated and puri®ed by ¯ash chromatography
eluting with 1:4 ethyl acetate:hexane. The title compound was
obtained as white crystals, 7.15 g (64%), mp=58±58.5 ꢀC
(lit.13 58 ꢀC).
For solid-phase peptide synthesis we chose an Fmoc
protection strategy. Thus, we deprotected both carboxyl
and phosphinate esters of 11a,b with LiOH to give 12a,b
and converted the Boc and Z groups to Fmoc to give 2
and 3 (Scheme 3). For 11c,d we reversed this sequence
and converted the N-terminus to Fmoc to give 12c,d
(the TMS ether of 11c was cleaved under the acidic
conditions used) followed by carboxyl and phosphinate
deprotection, using TMSI mediated cleavage, to give 5
and 6.
The Fmoc derivatives 2, 3, 5 and 6 were incorporated
into Grb2-SH2 peptidomimetic sequences and the
results are described in Part 1.
The previous route was not suitable for phosphinate 4
because, despite several attempts, alkylation of 10a,b
failed with b-hydroxy electrophiles. Thus, we used an
alternative `early alkylation' approach (Scheme 4) from
the hypophosphorus acid synthon 13.7 Acylation and
ester reduction gave the key hydroxyethyl phosphinate
15. Protection of the alcohol as its benzyl ether and
deprotection of the methyl ketal7 revealed the latent P-
H functionality in 16 which was alkylated with 4-iodo-
benzylbromide to give 17.
The central Jackson coupling reaction with Boc-
iodoAla-OMe9 gave 18 and subsequent deprotection
and functionalisation gave the target Fmoc phosphinic
acid 4. Although protection and deprotection reactions
take up a signi®cant part of this sequence it should be
noted that other protection strategies, that were able to
cope with the multi-functionality of 4, were unsuccessful.
11. Thottathil, J. K.; Przybyla, C. A.; Moniot, J. L. Tetra-
hedron Lett. 1984, 25, 4737.
12. Typical experimental procedures; 11a: paraformaldehyde
(1.10 g, 36.6 mmol) and triethylamine (0.51 mL, 3.7 mmol)
were added to a solution of 10a (2.82 g, 7.3 mmol) in toluene
(15 mL) and the reaction heated at 100 ꢀC for 2 h. The mixture
was allowed to cool to room temperature, and concentrated
in vacuo. The residue was pre-adsorbed onto silica gel and
puri®ed by ¯ash chromatography, eluting with 4% MeOH/
EtOAc to give 11a as a colourless gum; 31P NMR (CDCl3)
49.2 ppm, TLC RF=0.31 (10% MeOH/ethyl acetate). 11b:
NaH (50 mg, 60% dispersion) was washed with hexane and
In conclusion, we have prepared the ®rst examples of 4-(al-
kylphosphinomethyl)-phenylalanine derivatives, as phos-
photyrosine isosteres, in Fmoc protected form suitable for
solid phase synthesis of peptidomimetic sequences.
added to
a
cooled ( 10 ꢀC) solution of 10b (512 mg,
1.03 mmol) in THF (9 mL). The reaction mixture was treated
with MeI (0.2 mL, 3 equiv) and stirred at 10 ꢀC for 1 h and at
rt for 0.8 h. The reaction was treated with EtOAc, washed with
water, dried (MgSO4) and evaporated. The crude material was
puri®ed by ¯ash chromatography, eluting with EtOAc, to give
11b as a colourless oil; 31P NMR (CDCl3) 51.17 ppm, TLC
RF=0.20 (10% hexane/EtOAc). 11d: A solution of 10b
(519 mg, 1.05 mmol) in CH2Cl2 (10 mL) was treated with
TMSCl (3Â0.25 mL) and iPr2EtN (3Â0.35 mL) over 2 h. After
the ®nal addition, benzaldehyde (0.24 mL) was added and the
reaction stirred at rt for a further 2.5 h. The reaction was eva-
porated and the crude oil puri®ed by ¯ash chromatography,
eluting with 25% EtOAc/CH2Cl2, to give 11d as a colourless
oil; 31P NMR (CDCl3) 49.7 and 45.9 ppm, TLC RF=0.50
(25% EtOAc/CH2Cl2).
References and Notes
1. (a) Egan, S. E.; Weinberg, R. A. Nature 1993, 365, 781. (b)
Pawson, T. Nature 1995, 37, 573.
2. (a) Saltiel, A. R.; Sawyer, T. K. Chem. Biol. 1996, 3, 887.
(b) Gibbs, J. B.; Oli€, A. Cell 1994, 79, 193.
3. The minimum recognition sequence, with micromolar a-
nity, for the Grb2-SH2 domain is pTyr-X+1-Asn-NH2 where
asparagine at X+2 is critically important (Garcia-Echeverria,
C. et al., Novartis Pharma Inc., Oncology, unpublished
results).
4. (a) Furet, P.; Gay, B.; Garcõa-Echeverrõa, C.; Rahuel, J.;
Fretz, H.; Schoepfer, J.; Caravatti, G. J. Med. Chem. 1997, 40,
3551. (b) Furet, P.; Gay, B.; Caravatti, G.; Garcõa-Echeverrõa,
13. Adlington, R. M.; Baldwin, J. E.; Basak, A.; Kozyrod, R.
P. J. Chem. Soc., Chem. Commun. 1984, 944.