Synthesis of (dicyclohexylphosphino) serine, its incorporation into a dodecapeptide, and the coordination of rhodium

Full text of DOI:10.1021/jo951327z

434
J . Org. Chem. 1996, 61, 434-435
Syn th esis of
(Dicycloh exylp h osp h in o)ser in e, Its
In cor p or a tion in to a Dod eca p ep tid e, a n d
th e Coor d in a tion of Rh od iu m
Scott R. Gilbertson* and Xifang Wang
F igu r e 1.
Sch em e 1a
Department of Chemistry, Washington University,
St. Louis, Missouri 63130
Received J uly 25, 1995
November 20, 1995 )
(Revised Manuscript Received
The coordination of metals to peptides has a number
of applications. Metal binding has been used to stabilize
and control peptide structure.1-8 The incorporation of
metal binding sites into proteins can facilitate protein
purification or control of enzyme activity.9,10 Metal-
protein conjugates can deliver medicinally important
metals for imaging.11,12 The majority of the metal-
ligating groups placed in peptides contain oxygen and
nitrogen.13 The chemistry of coordination complexes of
this type is rich, but there is a wide variety of transition
metal chemistry that is not accessible to this type of
complex. For example, transition metal complexes useful
for effecting hydrogenation,14 hydroformylation,15 and
π-allylpalladium16,17 reactions are often phosphine based
and hence not normally associated with biologically based
ligands. For this reason, we embarked on the develop-
ment of amino acids that contain phosphine ligands. With
a variety of phosphine amino acids available, one should
be able to build protein-metal conjugates with metals
such as rhodium, ruthenium, palladium, and platinum.
These conjugates could, in turn, be used to capitalize on
both the metal’s ability to influence peptide conformation
as well as the peptide’s ability to potentially control metal
reactivity by placing the metal in unique steric environ-
ments. In addition, the use of a peptide-based framework
for a metal should allow for the synthesis of diphosphine
ligands where the electronic properties of the two phos-
phines are different.18,19 In the past, the use of electroni-
cally different phosphines with small ligands has been
limited by the need for C2 symmetry.20 Because peptide-
based phosphine ligands should exist in one conforma-
a
Key: (a) S8, 1/8 equiv of C6H6, rt, 2 h; (b) nBuLi, 1.0 equiv,
THF, -78 °C; (c) acrylic acid, 0.5 equiv, -78 °C to rt; (d)
ClC(O)OtBu, THF, -78 to 0 °C, 1.5 h; (e) (S)-(-)-lithium, 4-benzyl-
2-oxazolidinone, THF, -78 to 0 °C, 2 h; (f) KHMDS, THF, -78
°C, 30 min; (g) Tris-N3, THF, -78 °C, 5 min; (h) HOAc, -78 to 0
°C, 8 h; (i) LiOH, THF/H2O, 0 °C; (j) SnCl2, MeOH, 24 h; (l)
FmocOSU, NaHCO3, acetone/H2O; (m) (BOC)2O, NaHCO3, diox-
ane/H2O.
Sch em e 2
(1) Ghadiri, M. R.; Chong, C. J . Am. Chem. Soc. 1990, 112, 1630.
(2) Ghadiri, M. R.; Soares, C.; Choi, C. J . Am Chem. Soc. 1992, 114,
825.
(3) Ghadiri, M. R.; Fernholz, A. K. J . Am Chem. Soc. 1990, 112,
9633.
(4) Imperiali, B.; Fisher, S. L. J . Am Chem. Soc. 1991, 113, 8527.
(5) Imperiali, B.; Fisher, S. L. J . Org. Chem. 1992, 57, 757.
(6) Ghadiri, M. R.; Case, M. A. Angew. Chem., Int. Ed. Engl. 1993,
1594.
(7) Imperiali, B.; Kapoor, T. M. Tetrahedron 1993, 49, 3501.
(8) Imperiali, B.; Fisher, S. L.; Moats, R. A.; Prins, T. J . J . Am. Chem.
Soc. 1992, 114, 3182.
(9) Todd, R. J .; Vandam, M. E.; Casimiro, D.; Haymore, B. L.; Arnold,
F. H. 1991, 10, 156.
(10) Arnold, F. H.; Haymore, B. L. Science 1991, 252, 1796.
(11) Bakker, W. H.; Krenning, E. P.; Breeman, W. A.; Koper, J . W.;
Kooij, P. P.; Reubi, J .-C.; Klijn, J . G.; Visser, T. J .; Docter, R.; Lamberts,
S. W. J . Nucl. Med 1990, 31, 1501.
(12) Fischman, A. J .; Pike, M. C.; Kroon, D.; Fucello, A. J .; Rexinger,
D.; tenKate, C.; Wilkinson, R.; Rubin, R. H.; Strauss, H. W. J . Nuc.
Med. 1991, 32, 483.
tion, they offer a unique opportunity to take advantage
of diphos-ligands with two phosphines that are electroni-
cally different. This paper reports the synthesis of a new
phosphine amino acid precursor, (dicyclohexylphosphino)-
serine sulfide (1), which can be converted to the phos-
phine (2) (Cps) (Figure 1).21 This amino acid (1) was
incorporated into a 12-residue peptide along with another
phosphine derivative of serine [(diphenylphosphino)-
serine (Pps)].22 After synthesis of the peptide, the phos-
phine sulfides were converted to the corresponding
phosphines and rhodium was coordinated to the peptide.
(18) Chow, K. K.; Levason, W.; McAuliffe, C. W. In Transition Metal
Complexes of Phosphorus, Arsenic, and Antimony Ligands; McAuliffe,
C. A., Ed.; J . Wiley: New York, 1973.
(13) Ando, W.; Moro-oka, Y. E. The Role of Oxygen in Chemistry and
Biochemistry; Elsevier: Amsterdam, 1988.
(14) Takaya, H.; Ohta, T.; Noyori, R. In Catalytic Asymmetric
Synthesis; Ojima, I., Ed.; VCH: New York, 1993.
(15) Consiglo, G. In Catalytic Asymmetric Synthesis; Ojima, I., Ed.;
VCH Publishers, Inc.: New York, 1993.
(16) McKinstry, L.; Livinghouse, T. Tetrahedron 1994, 50, 6145.
(17) Trost, B. M.; Tanoury, G. J .; Lautens, M.; Chan, C.; MacPher-
son, D. T. J . Am. Chem. Soc. 1994, 116, 4255.
(19) Stelzer, O. In Topics in Phosphorus Chemistry; Griffith, J .,
Grayson, M., Eds.; J . Wiley: New York, 1977; Vol. 9, p 1.
(20) Whitesell, J . K. Chem. Rev. 1989, 89, 1581.
(21) We will use the three letter designation Cps to represent
(dicyclohexylphosphino)serine.
(22) Gilbertson, S. R.; Chen, G.; McLoughlin, M. J . Am. Chem. Soc.
1994, 116, 4481.
0022-3263/96/1961-0434$12.00/0 © 1996 American Chemical Society

Communications
J . Org. Chem., Vol. 61, No. 2, 1996 435
with RhCl(NBD)+ -ClO4 gave the cationic rhodium-
containing peptide. Upon complexation of rhodium, the
31P resonances shift downfield and coupling between spin
1/2 103Rh and the phosphines was observed (Figure 2c).
This ligand has the potential to exist in two different
conformations (Figure 2c,d), one in which the metal
points out of the page (Figure 2c) and one in which the
metal points back into the plane of the paper (Figure 2d).
Control of this type of conformation is difficult in the case
of small molecules and is generally overcome through the
use of C2 symmetric ligands. Since the peptide phosphine
ligands are not C2 symmetric, it is essential that the
metal complex exist in one conformation. As can be seen
in Figure 2, there is one set of 31P resonances for the
peptide. This means the peptide is either in one confor-
mation or is in rapid equilibrium between different
conformations. We have tested this by variable-temper-
ature NMR. The 1H and 31P NMR spectra of this complex
do not change over the temperature range of -40 to +40
°C. This indicates that the complex probably exists in
one conformation. We are currently attempting to de-
termine which conformation exists by looking for dipolar
exchange between the protons in the backbone of the
peptide and the phenyl rings on the phosphines.
F igu r e 2. Ac-Ala-Aib-Ala-Pps-Ala-Ala-Aib-Cps-Ala-Alqa-Aib-
Ala-OH.
The dodecapeptide synthesized here represents the first
example of a peptide-based diphos ligand with electroni-
cally different phosphines.
The first step in the synthesis of the dicyclohexylphos-
phinoserine was a conjugate addition of the anion of
dicyclohexylphosphine sulfide (4) to acrylic acid. After
difficulty was experienced with the addition of the
dicyclohexylphosphide anion, this proved to be an excel-
lent route to the desired phosphine sulfide (5).
We have synthesized both FMOC and tBOC protected
amino acids. The phosphine sulfide acid was first
converted to the R-azide (7) using chemistry developed
by Evans (Scheme 1).23-25 This compound (7) was the
branch point in the synthesis of the FMOC- and tBOC-
protected amino acids. Reaction of the azide oxazolidone
(7) with lithium hydroxide gave the azide acid (8).
Reduction of the azide with tin(II) chloride in methanol
afforded the amino methyl ester which was then pro-
tected as the FMOC carbamate to give 9. The tBOC-
protected amino acid (11) was synthesized by reduction
of the azide oxazolidone (7) with tin(II) chloride followed
by tBOC protection of the amine with tert-butyl anhy-
dride. Hydrolysis of 10 with lithium hydroxide then gave
the acid (11) ready for peptide coupling.
As stated earlier, one of the major goals of this work
is to ultimately develop new selective catalysts. We have
taken a preliminary look at the activity of the rhodium-
phosphine complex. We have found the reactivity of this
complex, in hydrogenation to be qualitatively the same
as simple DIPHOS-rhodium complexes. The hydroge-
nation of methyl 2-acetamidoacrylate with this catalyst
gives N-acyl alanine methyl ester in high yield and 8%
ee. While the enantiomeric excess is low, the inital goal
was to determine if this type of complex was capable of
performing catalytic chemistry. Positioning rhodium in
a peptide places the metal in the vicinity of both amide
nitrogens and carbonyls. Either of these potential ligands
could have coordinated to the rhodium and interrupted
the catalytic cycle. This apparently is not a problem,
since the metal ligand system turns over readily. Now
that we have established that these complexes catalyze
hydrogenation we are building a series of peptides with
(diphenylphosphino)serine and (dicyclohexylphosphino)-
serine in different positions along the peptide backbone.
This will enable us to evaluate a number of different
environments for transition metals.
To demonstrate that this phosphine-containing amino
acid was compatible with solid phase peptide synthesis
methods, we have incorporated it into a 12-residue
peptide (13) (Scheme 2).
The ability to incorporate phosphines into various
peptide structures allows for the facile synthesis of a wide
variety of phosphine ligands. In principle, a large
number of such ligands can be synthesized and screened
for unique structural characteristics, useful reactivity,
or important biological activity. We are currently using
commercially available technology to build such libraries
of ligands.
The peptide chosen was designed to be soluble in
organic solvents and have high helix content. The two
phosphine-containing amino acids were positioned with
three residues between them to allow for the proper
orientation for metal coordination assuming an R-helical
secondary structure. The peptide was synthesized by
standard FMOC chemistry on polystyrene resin.
The desulfurization and metalation reactions were
easily followed by 31P NMR.26 Figure 2a is the 31P spectra
of the diphosphine sulfide. Upon reaction with Raney
nickel, the 31P chemical shifts moved upfield to the
appropriate range for phosphines (Figure 2b). Reaction
Ack n ow led gm en t . We gratefully acknowledge
Mallinckrodt Chemicals, Inc., The Washington Univer-
sity High-Resolution NMR Facility, partially supported
by NIH 1S10R02004, and the Washington University
Mass Spectrometry Resource Center, partially sup-
ported by NIHRR00954, for their assistance.
(23) Evans, D. A.; Britton, T. C. J . Am. Chem. Soc. 1987, 109, 6881.
(24) Evans, D. A.; Britton, T. C.; Ellman, J . A.; Dorow, R. I. J . Am.
Chem. Soc. 1990, 112, 4011.
(25) Evans, D. A.; Ellman, J . A. J . Am. Chem. Soc. 1989, 111, 1063.
(26) Representative 31P chemical shifts are (a) Ph3PS δ ) 39.9-
43.5 ppm, (b) CH3PPh2 δ ) -28.0 ppm (Gorenstein D. G. Prog. NMR
Spectrosc. 1983, 16, 1), and (c) [Ph2(CH2)P]2RhCOD δ ) 14 to 18 ppm
(Descotes, G.; Lafont, D.; Sinou, D.; Brown, J . M.; Chaloner, P. A.;
Parker, D. Nouv. J . Chim. 1981, 5, 167).
Su p p or tin g In for m a tion Ava ila ble: Complete experi-
mental and spectroscopic data for all described compounds (10
pages).
J O951327Z