Structure of Phosphorus-Containing Peptide Ligands. X-ray and NMR Structural Study of Free Ligand and Rhodium Complex

Article abstract of DOI:10.1021/jo970773e

The structure of a series of phosphorus-containing peptides was studied. The X-ray structure of a phosphorus-containing dodecapeptide is reported. Analysis of the solution structure of similar phosphorus-containing peptides before and after coordination of rhodium is also reported. In both the solid state and in solution, the peptides were found to exist in a mixture of α-helical and 3₁₀ helical conformations. Coordination of rhodium to the i, i + 4 orientated phosphine groups appears to alter the conformational preference.

Full text of DOI:10.1021/jo970773e

J . Org. Chem. 1997, 62, 5557-5566
5557
Str u ctu r e of P h osp h or u s-Con ta in in g P ep tid e Liga n d s. X-r a y a n d
NMR Str u ctu r a l Stu d y of F r ee Liga n d a n d Rh od iu m Com p lex
Scott R. Gilbertson,*,† Guohua Chen,† J eff Kao,† Alicia Beatty,† and Charles F. Campana‡
Department of Chemistry, Washington University, St. Louis, Missouri 63130 and Siemens Analytical
X-ray Systems, Inc., 6300 Enterprise Lane, Madison, Wisconsin 53719-1173
Received April 30, 1997X
The structure of a series of phosphorus-containing peptides was studied. The X-ray structure of a
phosphorus-containing dodecapeptide is reported. Analysis of the solution structure of similar
phosphorus-containing peptides before and after coordination of rhodium is also reported. In both
the solid state and in solution, the peptides were found to exist in a mixture of R-helical and 310
helical conformations. Coordination of rhodium to the i, i + 4 orientated phosphine groups appears
to alter the conformational preference.
In tr od u ction
containing amino acids were positioned in an i, i + 4
relationship which requires an R-helical secondary struc-
ture for the side chains to be in position to bind rhodium.
While metal coordination was evidence for helical sec-
ondary structure it was not proof. It is conceivable that
some other structure could have existed that allowed
coordination to rhodium. Despite the binding of rhodium
in one location, the structure of the remainder of the
peptide chain was not conclusively demonstrated. This
paper reports the X-ray crystal structure of our original
phosphorus-containing peptide. Also reported is an NMR
study of the secondary structure of a related peptide and
a peptide-metal complex.
We recently reported the synthesis of phosphine-
containing derivatives of serine.1,2 In those papers we
developed chemistry that allows for the protection of the
potentially air sensitive phosphine group and its depro-
tection after incorporation into peptides. This chemistry
was developed with two goals in mind. We feel that
peptides with stable secondary structures should be
excellent scaffolds upon which catalytically active transi-
tion metals can be placed. Such an arrangement has
potential in the design and synthesis of catalysts that
exhibit a variety of selectivities. The other motive behind
the development of such a system is that large libraries
of peptides can be made by combinatorial methods.3-9
Consequently, the system we reported allows for the
synthesis of large libraries of ligands which can coordi-
nate transition metals and then be screened for unique
activity and selectivity. This approach to ligand develop-
ment cannot only lead to better ligands, but also add in
an empirical way to the understanding of what features
are important in a given catalytic system. Before either
of these goals can be achieved, it is necessary to be able
to characterize the phosphine-peptide ligand systems
and their metal complexes. This paper reports the
structural characterization of the first peptide system we
developed.
The peptide reported in our first paper was designed
to possess a helical structure through the predominant
use of alanine and the incorporation of a number of
aminoisobutyric acid (Aib) residues. After synthesis, the
phosphine-sulfide side chains of this peptide were
converted to free phosphines, and rhodium was bound
into this unique environment. The secondary structure
of that peptide and peptide-metal complex was assumed
to be helical based on primary sequence and the peptide’s
ability to act as a bis-chelating ligand. The phosphine-
Resu lts
Syn th esis of P ep tid es. The desired peptides were
synthesized by solid phase peptide chemistry. A seg-
mental approach was taken, where the phosphine sulfide
containing amino acid was incorporated as a dimer with
alanine (1), and amino isobutyric acid was incorporated
as either a dimer with alanine (2) or a trimer with
alanine (3). The couplings were performed using either
Wang or Rink resin depending on whether the acid or
amide terminal peptide was desired.
X-r a y Cr ysta l Str u ctu r e of P ep tid e 4. The inital
peptide that was synthesized was a dodecamer that
contained three Aib residues (4). This peptide was
synthesized by standard FMOC peptide chemistry, in-
corporating two of the unnatural amino acid (diphen-
ylphosphino)serine (Pps), in an i, i + 4 orientation.1
X-ray quality crystals of this peptide were obtained by
slow evaporation of a mixture of methanol, ethyl acetate,
and hexane.
The X-ray crystallographic structure (Figure 2) shows
peptide 4 existed in a right-handed helical conformation.
This conformation was more apparent when the molecule
was viewed along its axis from the amine terminus
(Figure 2). A detailed study of the crystal structure
revealed that the peptide was a mixture of 310 and R-helix
conformations. From Figure 2 one can see that the first
several residues were in a 310-helical conformation. The
residues in the middle of the sequence were R-helical,
which was indicated by the location of the phosphine
sulfides on the same side of the helix, given their i, i +
4 relationship. Figure 2 also clearly illustrates that the
last two residues, at the acid terminus, were in an
extended form.
† Washington University.
‡ Siemens Analytical X-ray Systems.
X Abstract published in Advance ACS Abstracts, J uly 15, 1997.
(1) Gilbertson, S. R.; Chen, G.; McLoughlin, M. J . Am. Chem. Soc.
1994, 116, 4481.
(2) Gilbertson, S. R.; Wang, X. J . Org. Chem. 1996, 61, 434.
(3) Gilbertson, S. R.; Wang, X. Tetrahedron Lett. 1996, 37, 6475.
(4) Borman, S. Chem. Eng. News 1996, Feb 12, 29.
(5) Baum, R. M. Chem. Eng. News 1994, Feb 7, 20.
(6) Gordon, E. M.; Gallop, M. A.; Patel, D. V. Acc. Chem. Res. 1996,
29, 144.
(7) DeWitt, S. H.; Czarnik, A. W. Acc. Chem. Res. 1996, 29, 114.
(8) Nielsen, J . Chem. Ind. 1994, 22, 902.
(9) Ellman, J . A. Acc. Chem. Res. 1996, 29, 132.
S0022-3263(97)00773-1 CCC: $14.00 © 1997 American Chemical Society

5558 J . Org. Chem., Vol. 62, No. 16, 1997
Gilbertson et al.
F igu r e 1.
F igu r e 2. X-ray structure of 4 Ac-Ala-Aib-Ala-Pps(S)-Ala-Ala-Aib-Pps(S)-Ala-Ala-Aib-Ala-OH. The hydrogen atoms were omitted
for clarity.
Ta ble 1. Tor sion An gles O a n d ψ fr om X-r a y Cr ysta llogr a p h ic Da ta of 4
Ala1
Aib2
Ala3
Pps4
Ala5
Ala6
Aib7
Pps8
Ala9
Ala10
Aib11
Ala12
φ (deg)
ψ (deg)
-52.5
-36.7
-56.5
-29.5
-60.4
-26.2
-74.1
-24.9
-64.1
-39.9
-65.9
-47.2
-54.8
-46.3
-71.7
-32.2
-69.8
-31.3
-82.0
-15.0
+51.9
+40.7
-167
a
/
a
Not applicable.
These general observations are in agreement with
analysis of the torsion angles φ and ψ of each residue
and the hydrogen bonding pattern in the peptide back-
bone. The torsion angles φ and ψ for each residue from
the X-ray crystallographic data of 4 are summarized in
Table 1. The torsion angles φ and ψ for residues 1 to 10
were within the ranges defined for R-helix or a 310 helix.
The values of φ and ψ for the first three residues were
close to the defined value for an ideal 310-helix, {φ, ψ} )
{-60°, -30°}, and the values for residues in the middle
(residues 5, 6, and 7) were close to the defined value for
an ideal R-helix, {φ, ψ} ) {-60°,-45°}.10-12
310-helix and a left-handed R or 310-helix, respectively.
Toniolo et al. found that in an Aib decamer, the average
value of φ was 54° and that of ψ was 31°.10 The values,
from the X-ray, of φ and ψ for Aib(2) and Aib(7) are
consistent with a helical conformation. The positive
value for φ and ψ observed for Aib(11) indicates a helix
reversal, as is often the case at the carboxy end of Aib-
containing peptides. This causes the end of the molecule
to be in an extended conformation.16,17
The hydrogen bonds between the amide protons and
carbonyl oxygen atoms also indicate a mixture of helix
types. From the acetyl group to Ala(10) there were four
4-1 type (i, i + 3 type) hydrogen bonds interspersed with
five 5-1 type (i, i + 4 type) hydrogen bonds. This was
evidence that the peptide was a mixture of 310 and R-helix
structures. Three of the four 4-1 type hydrogen bonds
were at the amine end and the other one was near the
acid terminus, further indicating that the peptide had
310-helical character at the amine end.
Marshall13,14 and Burgess15 have shown that the al-
lowable φ and ψ angles for the Aib residue occur in two
very restricted regions near -57°, -47° and +57°, +47°.
These two regions corresponded to a right-handed R or
(10) Toniolo, C.; Crisma, M.; Bonora, G. M.; Benedetti, E.; Di Blasio,
B.; Pavone, V.; Pedone, C.; Santini, A. Biopolymers 1991, 31, 129-
138.
(11) Burgess, A. W.; Ponnuswamy, P. K.; Scheraga, H. A. Isr. J .
Chem. 1974, 12, 239.
(12) Ne´methy, G.; Scheraga, H. A. Quart. Rev. Biophys. 1977, 10,
239.
Since we ultimately desired to bind transition metals
to the phosphorus ligands in the side chains, we were
very interested in the distance between the two phos-
(13) Marshall, G. R.; Bosshard, H. E. Circ. Res. 1972, 30/ 31 (Suppl.
II), 143-150.
(16) Karle, I. L.; Balaram, P. Biochemistry 1990, 29, 6748.
(17) Marshall, G. R.; Hodgkin, E. E.; Langs, D. A.; Smith, G. D.;
Zabrocki, J .; Leplawy, M. T. Proc. Natl. Acad. Aci. U.S.A. 1990, 87,
487.
(14) Hodgkin, E. E.; Clark, J . D.; Miller, K. R.; Marshall, G. R.
Biopolymers 1990, 30, 533.
(15) Burgess, A. W.; Leach, S. J . Biopolymers 1973, 12, 2599-2605.

Phosphorus-Containing Peptide Ligands
J . Org. Chem., Vol. 62, No. 16, 1997 5559
Ta ble 2. NMR Assign m en ts for P ep tid e 5 fr om TOCSY
a n d NOESY Exp er im en ts
phorus atoms. That distance was found to be 6.8 Å. This
indicated that metal binding was possible, particularly
when one factors in that in the crystal structure the two
phosphine sulfides are turned away from each other. This
results in the distance in the crystal structure being
considerably longer than the potential minimum distance
when both phosphorus atoms are free to rotate toward
each other.
NMR Str u ctu r a l Da ta . To determine if the Pps(S)-
containing short peptides have the same secondary
structure in solution as in the solid phase, and to
determine what conformation the peptide backbone has
after chelation of transition metals, the secondary struc-
ture of peptide 5, and rhodium complex peptide 7, were
studied by NMR. Before such a study could take place
assignment of all the proton chemical shifts in the peptide
backbone had to be made.18-22
Seq u en t ia l Assign m en t of P r ot on R eson a n ces.
The acid terminal groups on the peptides studied were
varied for a variety of reasons. We were only able to
obtain crystals of the acid terminal peptides, and as a
consequence the crystal structure is of an acid terminal
peptide. The same peptide was not suitable for the NMR
work due to its solubility and the observation that the
NMR line shape was broad. For this reason the methyl
ester was investigated by NMR. Both the ester and the
amide-terminated metal complexes were evaluated. Ex-
cept for the differences mentioned below, the spectra were
qualitatively the similar. The amide data for the complex
is presented in the paper because of our ability to resolve
and assign all the protons in the molecule.
R-H (ppm)
N-H (ppm)
â-H (ppm)
Ala1
Aib2
4.21
7.98
8.39
7.97
7.77
7.59
7.58
7.95
7.82
7.73
7.39
7.52
7.22
7.77
1.18
1.29 and 1.33
1.04
3.20 and 3.20
1.23
Ala3
3.68
4.46
4.15
4.09
Pps(S)4
Ala5
Ala6
1.24
Aib7
1.12 and 1.08
3.28 and 3.19
1.26
1.20
1.23 and 1.26
1.17
Pps(S)8
Ala9
4.36
3.94
4.01
Ala10
Aib11
Ala12
Ala13
AcO
4.16
4.16
1.81
3.31
1.27
OMe
Ta ble 3. NMR Assign m en ts for P ep tid e 6 fr om TOCSY
a n d NOESY Exp er im en ts
R-H (ppm)
N-H (ppm)
â-H (ppm)
Ala1
Aib2
4.20
7.99
8.40
7.90
7.77
7.58
7.58
7.98
7.82
7.73
7.44
7.48
7.20
1.15
1.15 and 1.33
1.05
3.20 and 3.20
1.24
Ala3
3.68
4.46
4.15
4.10
Pps(S)4
Ala5
Ala6
1.23
Aib7
1.13 and 1.09
3.28 and 3.18
1.27
1.21
1.26 and 1.22
1.19
Pps(S)8
Ala9
4.34
3.92
3.96
Ala10
Aib11
Ala12
AcO
3.96
1.80
CONH2
6.91 and 6.94
Proton chemical shifts of peptides 5, 6, and 7 were
assigned by analysis of TOCSY23 and NOESY24 spectra.
The Pps(S)4 and Pps(S)8 residues were identified by the
spin propagation from NH through âH. The N-terminal
Ala(1) residue was assigned based on the NOE connection
between the acetyl methyl and NH proton of Ala(1).
Sequential assignment18 of all peptides was obtained by
using the NH-RH fingerprint region of the NOESY
spectrum. All of the dRN connections between adjacent
residues were observed. For the Aib residues at 2, 7, and
11, the lost connectivity was restored by observing NOE
between Aib-CH3 and the neighboring NHi+1 protons.
With the exception of the aromatic protons on the phenyl
rings attached to the phosphorus atoms, it was possible
to assign all of the proton resonances to specific protons
for each residue of the peptides. The complete sequence-
specific assignments for peptides 5, 6, and 7 are shown
in Tables 2, 3, and 4.
Ta ble 4. NMR Assign m en ts for Com p lex 7 fr om TOCSY
a n d NOESY Exp er im en ts
R-H (ppm)
N-H (ppm)
â-H (ppm)
Ala1
Aib2
4.32
8.48
9.04
8.26
7.89
8.02
7.67
8.28
8.24
8.14
7.71
7.57
7.33
1.25
1.39 and 1.39
1.15
2.84 and 3.11
1.46
Ala3
3.84
4.32
4.31
4.04
Pps(S)4
Ala5
Ala6
1.37
Aib7
1.29 and 1.37
3.03 and 3.18
1.41
1.34
1.42 and 1.50
1.21
Pps(S)8
Ala9
4.21
3.61
3.92
Ala10
Aib11
Ala12
AcO
4.00
1.96
CONH2
7.04 and 7.12
NMR Stu d y of th e Secon d a r y Str u ctu r e of P ep -
tid e 5. The helical region of a peptide can often be
characterized by sequential dNN and medium-range dNN
(i, i + 2) and dRΝ[(i, i + 2), (i, i + 3)] NOE connectivities
The structures for peptides 4-7 are as follows:
Ac-Ala-Aib-Ala-Pps(S)-Ala-Ala-Aib-Pps(S)-Ala-Ala-Aib-
Ala-OH (4).
Ac-Ala-Aib-Ala-Pps(S)-Ala-Ala-Aib-Pps(S)-Ala-Ala-Aib-
Ala-Ala-OCH3 (5).
3
and small values (<6.0 Hz) of J HN-RH. We have used
these methods along with the temperature dependence
of the N-H chemical shifts in the characterization of
peptides 5 and 7.
Ac-Ala-Aib-Ala-Pps(S)-Ala-Ala-Aib-Pps(S)-Ala-Ala-Aib-
Ala-NH2 (6).
Rhodium complex of Ac-Ala-Aib-Ala-Pps-Ala-Ala-Aib-
Pps-Ala-Ala-Aib-Ala-NH2 (7).
Table 5 summarizes the NOE data for peptide 5 and
Scheme 1 illustrates these tabulated NOEs. The sequen-
tial dNN (i, i + 1) observed along the peptide chain is
indicative of the peptide having a defined structure. A
number of the medium range dNN (i, i + 2) and dRΝ (i, i
+ 2) were also observed. Although dNN (1, 3) and dRN (8,
10) were missing, the corresponding dRΝ (1, 3) and dΝΝ
(8, 10) could be identified (Scheme 1). The dRΝ (i, i + 3)
for Ala(3), Ala(6), and Ala(10) were also observed. The
observed NOEs, discussed above, particularly the ob-
served dRN (i, i + 2) and the lack of dRN (i, i + 4), are
(18) Englander, S. W.; Wand, A. J . Biochemistry 1987, 26, 5953.
(19) Wu¨thrich, K.; Wider, G.; Wagner, G.; Braun, W. J . Mol. Biol.
1982, 155, 311.
(20) Wu¨thrich, K. Biopolymers 1983, 22, 131.
(21) Wu¨thrich, K.; Billeter, M.; Braun, W. J . Mol. Biol. 1984, 180,
715.
(22) Billeter, M.; Braun, W.; Wu¨thrich, K. J . Mol. Biol. 1982, 155,
321.
(23) Bax, A.; Davis, D. G. J . Magn. Reson. 1985, 65, 355.
(24) Kumar, A.; Ernst, R. R.; Wuthrich, K. Biophy. Res. Commun.
1980, 95, 1.

5560 J . Org. Chem., Vol. 62, No. 16, 1997
Ta ble 5. NMR-Ba sed In ter -r esid u e Dista n ce Con str a in ts for P ep tid e 5a
Gilbertson et al.
Ala1
Aib2
Ala3
Pps4
Ala5
Ala6
Aib7
Pps8
Ala9
Ala10
Aib11
Ala12
Ala13
b
b
O
b
O
O
b
/
b
/
/
b
b
b
b
b
O
b
b
b
b
O
O
b
b
b
b
O
O
b
b
b
b
b
O
b
/
b
/
/
b
b
b
O
O
O
b
b
O
O
O
O
b
b
b
b
b
/
b
/
O
/
/
b
b
/
/
/
/
/
/
/
/
/
dNN (i, i + 1)
dRN (i, i + 1)
dNN (i, i + 2)
dRN(i, i + 2)
dRN (i, i + 3)
dRN(i, i + 4)
/
/
/
/
a
(b) NOE observed; (O) no NOE observed; (/) not applicable.
Sch em e 1. Med iu m -Ra n ge NOEs in P ep tid e 5
3
Ta ble 6. Tor sion An gle O Der ived fr om J rN for P ep tid e 5
Ala1
Aib2
Ala3
Pps4
Ala5
Ala6
Aib7
Pps8
Ala9
Ala10
Aib11
Ala12
Ala13
3J RN (Hz)
φ (deg)
6.5
-77
/
/
5.8
-72
6.5
-77
5.9
-73
?
?
/
/
6.0
-74
6.5
-77
6.0
-74
/
/
7.8
-88
6.5
-77
Ch a r t 1
consistent with this peptide existing predominately in 310-
helical secondary structure from Ala(1) to Ala(10).25
The vicinal coupling constants J HN-RH for Ala(1) to
ium complex 7 (Chart 1) was examined. To date, we have
not been able to obtain X-ray quality crystals of any of
our phosphine-containing peptides with a transition
metal coordinated, so we embarked on determining the
secondary structure in solution by NMR.
3
Ala(12) (except that of Ala(6) which could not be resolved)
1
were obtained from the one-dimensional H spectrum.
The average value of the coupling constants from Ala(1)
to Ala(10) was found to be 6.2 Hz (Table 6). This
corresponds to an average torsion angle φ of -70° (φ )
-54° was used for Aib(2) and Aib(7) in the calculation),
which was within the range of -60° to -90° typically
found in helical conformations.25,26 The absence of NOEs
dΝN (9, 11) and dRN (9, 11) suggested that Aib(11) may
have had a positive torsion angle φ, and consequently the
acid terminus of the peptide may have been in an
extended form. This result was analogous to the solid
state structure of peptide 4 in which the end of the
structure was found to be in an extended form. The
relatively large J HN-RH and torsion angle φ for Ala(12)
were in agreement with the NOE observations and this
conclusion.
2D NMR Stu d y of th e Secon d a r y Str u ctu r e of
Com p lex 7. Next, the secondary structure of the rhod-
Complex 7 was chosen for study because of the quality
of the NOE data obtained with this peptide. The
sequential assignment of the peptide backbone is listed
in Table 4. The NMR assignment of this molecule was
done in the same manner as described for peptide 5. The
sequential and medium range NOEs are listed in Table
7 and the medium range NOEs are illustrated in Scheme
2. All of the sequential dNN (i, i + 1) NOEs along the
peptide chain were observed, indicating that the peptide
backbone had a conformationally defined structure.
From Ala(1) to Ala(8), none of the medium range dNN (i,
i + 2), dRΝ (i, i + 2), or dâΝ (i, i + 2) NOEs were found,
while all of the dRΝ (i, i + 3) interactions were present.
This along with the absence of the dRN (i, i + 2) NOEs is
consistent with an ordered structure other than the 310
structure observed for this region in peptide 5. The lack
of dRN (i, i + 4) suggests a structure other than R-helical
as well. The presence of dRΝ (8, 11), dΝN (9, 11), and dΝN
(10, 12) indicated that Aib(11) had a negative torsion
angle φ (assumed here to be -54°). This data suggests
3
(25) Wu¨thrich, K. NMR of Proteins and Nucleic Acids; J . Wiley and
Sons: New York, 1986.
(26) Bystrov, V. Prog. Nucl. Magn. Spectrosc. 1976, 10, 41-82.

Phosphorus-Containing Peptide Ligands
J . Org. Chem., Vol. 62, No. 16, 1997 5561
Ta ble 7. NMR-Ba sed In ter -r esid u e Dista n ce Con str a in ts for Com p lex 7a
Ala1
Aib2
Ala3
Pps4
Ala5
Ala6
Aib7
Pps8
Ala9
Ala10
Aib11
Ala12
b
b
O
O
O
b
b
O
b
/
O
/
O
/
b
b
O
/
O
b
O
O
b
b
O
O
O
b
O
O
b
b
O
O
O
b
O
O
b
b
O
O
O
b
b
O
b
/
O
/
O
/
b
b
O
O
O
b
O
O
b
b
b
O
O
O
O
/
b
b
b
O
O
/
b
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
dNN (i, i + 1)
dRN (i, i + 1)
dNN (i, i + 2)
dRN (i, i + 2)
dâN (i, i + 2)
dRN (i, i + 3)
dâN (i, i + 3)
dRN (i, i + 4)
O
/
O
/
/
/
a
(b) NOE observed; (O) no NOE observed; (/) not applicable.
Sch em e 2. Med iu m -Ra n ge NOEs in Com p lex 7
3
Ta ble 8. Tor sion An gle O Der ived fr om J rN for P ep tid e 7
Ala1
Aib2
Ala3
Pps4
Ala5
Ala6
Aib7
Pps8
Ala9
Ala10
Aib11
Ala12
3J RN (Hz)
φ (deg)
5.14
-70
/
/
4.63
-66
6.68
-80
5.60
-73
6.17
-75
/
/
5.31
-70
4.96
-68
8.00
-90
/
/
6.85
-80
Ta ble 9. Va r ia ble Tem p er a tu r e Exp er im en t on Com p lex 7a
-10 °C
-5 °C
0 °C
5 °C
10 °C
15 °C
20 °C
-∆δ/∆T
H-bond
NH Ala1
NH Aib2
NH Ala3
NH Pps4
NH Ala5
NH Ala6
NH Aib7
NH Pps8
NH Ala9
NH Ala10
NH Aib11
NH Ala12
8.52
9.09
8.31
7.93
8.05
7.71
8.33
8.29
8.18
7.75
7.54
7.37
8.49
9.05
8.46
9.01
8.27
7.92
8.02
7.68
8.28
8.25
8.15
7.72
7.53
7.35
8.44
8.97
8.41
8.93
8.23
7.92
7.98
7.65
8.23
8.21
8.13
7.68
7.53
7.34
8.38
8.88
8.35
8.84
8.18
7.91
7.95
7.62
8.18
8.16
8.10
7.62
7.51
7.32
5.7
8.3
4.1
1.0
3.3
3.0
5.0
4.1
2.7
4.3
1.0
1.1
O
O
b
b
b
b
O
b
b
b
b
b
7.93
8.04
7.92
8.00
7.92
7.96
8.30
8.17
7.53
8.30
8.14
7.53
8.20
8.12
7.52
a
δ
ΝΗ (ppm) vs temperature (K) and temperature dependencies of amide proton -∆δ/∆T (10-3 ppm/K) in DMF-d7. (b) Hydrogen bonded;
(O) not hydrogen bonded.
that unlike the acid terminus of peptide 4 in the solid
state and the acid terminus of 5 in solution, the amide
terminus of complex 7 was in helical form. The dΝN (9,
11) and dΝN (10, 12) interactions found after the metal
binding site are consistent with either a 310 or an
R-helical structure.
shift of the amide protons, which is often used to indicate
the presence of hydrogen bonds in a peptide backbone,
was determined. The results are summarized in Table
9. As expected, it was found that amide protons of
Ala(1) and Aib(2) were exposed to solvent. In a helical
structure there is no internal hydrogen bond acceptor
available for these two amide protons, consequently they
exhibit a large change in their chemical shift as a
function of temperature. The temperature coefficient for
the amide proton of Aib(7) (-∆δ/∆T ) 5) indicated that
this proton was also exposed to solvent. This site is
located between the two metal binding amino acids and
it is possible that binding a metal between the two side
chains may strain the helix, resulting in exposure of the
amide proton of Aib(7). All the other N-Hs in the
molecule were found to be internally hydrogen bonded.
In terms of structural issues other than the peptide
secondary structure. The rhodium complex is character-
ized with norbornadiene as the other ligands in the
rhodium. Mass spectrometry shows a strong peak for the
3
Crude vicinal coupling constants J HN-RH of Ala(1) to
Ala(12), except that of Ala(5), were obtained from the one-
1
3
dimensional H spectrum at -10 °C. J HN-RH of Ala(5)
could not be resolved in this spectrum, but it was resolved
in the one-imensional 1H spectrum at 10 °C. All coupling
3
constants J HN-RH, together with their corresponding
torsion angles φ, were summarized in Table 8. All torsion
angles φ obtained were within the range of -60° to -90°.
From Ala(1) to Ala(12) the average of the torsion angle
φ was -69° (again, φ ) -54° was used for Aib(2) and
Aib(7) in the calculation), which was near -60°, the value
of φ defined for R or 310-helices.25,26
3
Since the NOE and J RN data for peptide 7 were
inconclusive, the temperature dependence of the chemical

5562 J . Org. Chem., Vol. 62, No. 16, 1997
Gilbertson et al.
molecular ion corresponding to this complex. On the
basis of the lack of a trans P-P coupling, we have
assigned the phosphines as being in a cis relationship
on the metal.
at the carboxy end of the molecule which is capable of
continuing the hydrogen bonding network of the helical
peptide.
Con clu sion
Discu ssion
The differentiation between R-helical and 310 helical
structures by NMR is not simple or always conclusive.
The observation of short proton-proton distances by
NOE is the fundamental tool for this task. While the
destination between turns and helical structures is
relatively straight forward, the interactions used to
distinguish between an R-helix and a 310 helix are weak
and sometimes difficult to observe. The presence of
NOEs between the R proton of one amino acid and the
N-H two residues away is often found in 310 helices.
Those two protons are on opposite sides of an R-helix. A
similar interaction between the R proton of one amino
acid and the N-H of a residue four positions down the
peptide chain is often used to demonstrate the presence
of R-helical character. In both cases these interactions
are weak and can be difficult to observe. In the case of
our NMR data peptide 5 gives strong cross peaks from
the RN(i, i + 2) interaction, and we use this to assign its
structure as primarily a 310 helix. The structure of
complex 7 is not as clearly assigned. The NOEs expected
for a molecule with an organized structure, dRN(i, i + 1)
and dNN(i, i + 1), were found but the crossed peaks
necessary for the differentiation between an R or a 310
helix were not observed. To verify that the molecule
actually did possess secondary structure, the temperature
dependence of the N-H chemical shift was determined.
That study revelead that all but one of the expected
internal hydrogen bonds were present, and therefore this
molecule is likely to have had an organized secondary
structure. The reason the N-H of Aib(7) shifts with
temperature may be due to an interaction of its hydrogen
bond acceptor in the helical scaffold with rhodium or
simply that metal binding to that region of the peptide
may perturb the structure sufficiently to disrupt the
hydrogen bond between N-H and its carbonyl partner.
The structural characterization of phosphorus-contain-
ing peptides was carried out. The X-ray and NMR data
indicate these peptides have organized structures which
are helical in at least two cases. We are in the process
of exploring the use of these peptides as ligands for
transition metals with a wide variety of goals. These
ligands have potential in stereo- and regioselective
catalysis with transition metals, as delivery agents for
therapeutic and diagnostic metals, and as tools for the
investigation of protein structure and folding.
Peptides that contain Aib are known to have a pro-
pensity for the formation of either R-helical or 310 helical
secondary structures.13-17,27 We have found that the
phosphorus-containing peptides reported in this paper
have organized structures which are for the most part
helical. Peptides rich in amino isobutyric acid often favor
a 310 helical structure, particularly near the amine
terminus. Based on the X-ray data and NMR work, this
appears to be the case with our peptides as well. The
orientation of the amino acids with phosphine metal
binding groups in their side chains is i, i + 4 to each
other. As a consequence of this positioning of these
amino acids, the region of the peptide near those amino
acids must be in an R-helical conformation for the
phosphine to be able to bind rhodium. The crystal
structure of peptide 4 revealed that the amine end of the
peptide had a 310 helical structure with the secondary
structure changing to R-helical in the region where the
phosphine sulfides are located. It is interesting that the
region of the peptide crystal structure where the phos-
phine sulfides were located was R-helical. The reason
for the switch from 310 helix to R-helix is not clear. One
possibility is that an R-helical section is better able to
accommodate the extremely large diphenylphosphine side
chains. Another possibility is that crystal-packing forces
cause the molecule to adopt this conformation. The NMR
study of peptide 5 reveals that this peptide exists in
primarily a 310 helical structure. This includes the region
of the molecule that was R-helical in the X-ray structure
of peptide 4. Despite this molecule having 310 helical
structure, after removal of the sulfur atoms this molecule
readily acts as a bis-chelating ligand for rhodium. For
this to occur the molecule must adopt a conformation
other than a 310 helix. One obvious change would be for
the peptide to take on an R-helical structure. The NOE
data for complex 7 indicates that the region between
residues 9-12 is helical. The helical nature of this end
of complex 7 could be due to metal binding between the
i, i + 4 phosphines inducing a helical structure or due to
the amide terminus of this peptide. It has been demon-
strated that amide-terminated peptides have a greater
propensity for helix formation. While it is clear from the
NMR data the complex 7 has a stable secondary struc-
ture, we were not able to assign that structure as either
310 or R-helical.
Exp er im en ta l Section
The acid terminus of peptides 4 and 5 is in an extended
form. This effect has been seen before with Aib rich
peptides.16,17 Both positive and negative values of the
torsion angles φ and ψ are of equal energy for ami-
noisobutyric acid; consequently, there was no steric
hindrance to Aib(11) adopting positive torsion angles,
which lead to an extended structure. The NMR data of
complex 7 indicates that the acid terminus of that peptide
is helical. This could be due to metal binding but is most
likely caused by the fact that this peptide is terminated
as the amide. This provides a hydrogen bonding group
NMR Sp ectr oscop y. Samples were made by dissolving
approximately 10 mg of each peptide in 0.7 mL of DMSO-d6
or DMF-d7. The samples were degassed by several freeze-thaw
cycles and sealed in NMR tubes under vacuum. All 2D NMR
spectra were recorded with a Varian Unity-600 spectrometer
operating at 600 MHz, interfaced with a Sun SPARC I
computer. The data were processed off-line on a SPARC 10
station with VNMR software.
The total correlation TOCSY spectra were recorded using
an MELV-17 mixing sequence of 100 ms, flanked by two 2 ms
trim pulses. Phase-sensitive 2D spectra were obtained by
employing the Hypercomplex method. A total of 2 × 256 ×
1024 data matrix with 16 scans per t1 value were collected.
Gaussian line broadening and sinebell function were used in
weighting t2 and t1 dimensions, respectively. After two-
(27) Karle, I. L.; Flippen-Anderson, J . L.; Uma, K.; Balaram, P.
Biochemistry 1989, 28, 6696.

Phosphorus-Containing Peptide Ligands
J . Org. Chem., Vol. 62, No. 16, 1997 5563
dimensional Fourier transform, the spectra resulted in 1K ×
1K data points and were phase and baseline corrected in both
dimensions. Chemical shifts were measured in parts per
million (ppm) downfield from internal TMS, or with the center
line of the lower field quintet at 2.90 ppm for DMF-d7.
The NOESY spectra resulted from a 2 × 512 × 4096 data
matrix with 16 scans per t1 value. The spectra were recorded
with a 250 and 450 ms mixing time, but 450 ms generally gave
better results. Hypercomplex method was used to yield phase-
sensitive spectra. The time domain data were zero filled to
yield a 2K × 2K data matrices and were processed in the
similar to the TOCSY spectra.
Variable temperature experiments were performed on com-
plex 7. Chemical shifts were measured in parts per million
(ppm) downfield from internal TMS. The temperature de-
pendencies of the amide proton resonances were determined
by seven measurements over a range of -10 °C to 20 °C. Four
COSY spectra were also taken over this range at 10 °C
intervals to help identify amide proton peaks. Each COSY
spectrum was collected into a 512 × 4096 data matrix with
16 scans per t1 value. The time domain data was zero filled
to yield a 4096 × 4096 data matrix and was Fourier trans-
formed using sine-bell weighting function in both t2 and t1
dimension.
Sin gle Cr ysta l Gr ow th a n d X-r a y Da ta Collection a n d
Refin em en t of P ep tid e 4. A 50 mg amount of peptide 4 was
dissolved in 2 mL of methanol, and the solution was filtered
through a microfilter (pore size: 5 µm) into a 10 mL crystal-
lization dish. Mixed solvent ethyl acetate/hexane (2 mL, 8-2
by volume) was then added to the dish. The dish was then
covered with parafilm into which several holes were made. The
solvent was allowed to evaporate overnight without any
interruption and single crystals were obtained. The crystals
obtained were stable only when surrounded by mother liquor.
The crystal was mounted for X-ray with the mother liquor also
sealed in the capillary tube.
X-r a y An a lysis.30 Crystal data: [C65H88N12O16P2S2]‚EtOAc;
formula wt. 1475.64, 0.31 × 0.27 × 0.23 mm, triclinic, a )
37.842(6) Å, b ) 10.627(2) Å, c ) 20.037(5) Å, Z ) 4, density
(calcd) ) 1.216 g/cm3. F(000) ) 3136, space group P212121,
R1 ) 0.0893, wR2 ) 0.2266 for I > 2σ(I), goodness of fit )
1.22 for all data. An extinction coefficient of 0.001056 was
determined.
A single crystal of 4 was mounted on a glass fiber, cell
constants were determined, and intensity data were collected
on a Siemens SMART Charge Coupled Device (CCD) Detector
system single crystal X-ray diffractometer system. Data were
collected at 195 K using graphite monochromated Mo KR
radiation (λ ) 0.71073 Å) from a sealed tube X-ray source. A
total of 1271 frames of intensity data were collected, with a
frame width of 0.3˚ in ω, and counting time of 10 s/frame. The
crystal to detector distance was 6.14 cm, and the total
collection time was 6.0 h. Frame integration yielded 45914
reflections, with 12391 unique reflections and 10737 reflections
with I > 2σ(I). No absorption correction was applied.
Structure solution was determined using SHELXS 86 and
refined using SHELXL 93. The structure was solved using
Direct Methods, and refinement was carried out using suc-
cessive full matrix least-squares refinements and difference
Fourier map calculations. The acyl end of the peptide exhib-
ited two-fold disorder, with 65%/35% occupancy of the two
conformations. Disordered atoms include O9-O13, N12, and
C22-C24, and C61. This disorder was linked to a two-fold
disorder of S2 (also 65%/35% occupancy.) One phenyl ring
bonded to P1 also exhibited a two-fold disorder, where two
positions were idealized and refined at ∼60%/40% occupancy.
The phenyl ring (C53-C58) and the associated ethyl acetate
solvent molecule (O91-O92, C91-C94) were found to be
cooperatively disordered, with two orientations at 60%/40%
occupancy. The ethyl acetate position constituting 40% oc-
cupancy was unresolveable and therefore not included in the
structure refinement. All non-hydrogen atoms were refined
anisotropically, except for the ethyl acetate atoms, which were
refined isotropically. Hydrogen atoms were placed at idealized
positions and refined isotropically using the riding model.
Other restraints applied to the refinement were as follows:
restraint of the N11-C21, N12′-C22′, N12-C22, and N12′-
C23′ bonds to 1.40(1) Å, restraint of the C21-C22, C21-C62,
and C21-C61′ bonds to 1.50(1) Å, and restraint of the O11′-
C22′ bond to 1.23 Å. In addition, bonds of P2 with C53, C53′,
and C47 were restrained to the same value, as were the bonds
between P2 and the two disordered S2 positions. The disor-
dered phenyl ring (C53-C58) was refined as a flat regular
hexagon, and all bond distances between these atoms were
fixed at 1.39 Å. The absolute configuration of the molecule
was not determined with certainty from the refinement.
A complete list of displacement coefficients, structure fac-
tors, and bond lengths and angles is submitted as Supporting
Information.
F MOC-P p s(su lfid e)-Ala -OP NB. Ts-Ala-OPNB (1.80 g,
4.93 mmol) was weighed out in a flask. A 150 mL volume of
CH2Cl2 was added, followed by 3.2 mL (24.6 mmol) of triethyl-
amine. Once the solution became clear it was transferred to
a separatory funnel. The solution was washed with water (2
× 50 mL) and brine (50 mL) and dried (MgSO4). After
filtration, the solvent was removed on a rotovap. CH2Cl2 (100
mL) was then added to the flask, and the solution was cooled
to 0 °C. In a second flask, 2.0 g of FMOC-Pps(S)1 (3.79 mmol)
was dissolved in 100 mL of CH2Cl2. This solution was cooled
to 0 °C and then transferred to the flask containing the CH2-
Cl2 solution of Aib-OPNB. Next EDC (872 mg, 4.55 mmol) and
HOBT (614 mg, 4.55 mmol) were added to the reaction
mixture, and the reaction was kept at 0 °C under N2 for 4 h
and at rt for 20 h. The reaction mixture was then transferred
to a separatory funnel, washed with 1 N HCl (2 × 150 mL), 1
N NaHCO3 (150 mL), water (100 mL), and brine (100 mL) and
dried over MgSO4. After filtration and removal of the solvent,
the crude product was dried under vacuum. The material was
purified by dissolution in a minimum of warm ethyl acetate
followed by the addition of enough warm hexane to make the
solution cloudy. The solution was allowed to stand at room
temperature. Once crystals began to form, the flask was
placed in a -20 °C freezer for further crystallization. A 2.33
g amount (84% yield) of the product (FMOC-Pps(S)-Ala-OPNB)
was obtained. The product was sufficiently pure to be used
directly in the next step.
1H NMR (300 MHz, CDCl3) δ 8.2 (d, J HH ) 8.7 Hz, 2H), 7.9-
7.8 (m, 4H), 7.76 (d, J HH ) 7.4 Hz, 2H), 7.6 (dd, J HH ) 10.3
Hz, J HH ) 7.4 Hz, 2H), 7.5-7.4 (m, 10H), 7.34-7.28 (m, 2H),
7.15 (d, J HH ) 6.7 Hz, 1H), 6.19 (s, 1H), 5.21 (s, 2H), 4.65 (dt,
J HP ) 21.3 Hz, J HH ) 7.1 Hz, 1H), 4.40 (s, 1H), 4.3-4.1 (m,
3H), 3.10 (dd, J HP ) 10.5 Hz, J HH ) 4.9 Hz, 2H), 1.36 (d, J HH
) 7.2 Hz, 3H). 13C NMR (75 MHz, CDCl3) δ 171.7 (s), 170.0
(d, J CP ) 10.0 Hz), 155.4 (s), 147.5 (s), 143.5 (s), 142.4 (s), 141.0
(s), 131.8 (d, J CP ) 82.0 Hz), 131.7 (s), 130.7 (d, J CP ) 10.1
Hz), 128.5 (d, J CP ) 12.0 Hz), 128.1 (s), 127.6 (s), 126.9 (s),
125.1 (s), 123.6 (s), 119.8 (s), 67.2 (s), 65.3 (s), 51.0 (s), 48.4
(s), 46.7 (s), 33.4 (d, J CP ) 55.7 Hz), 17.56 (s). 31P NMR (120
MHz, CDCl3) δ 39.19. IR (thin film) 3350 (s), 3060 (s), 3003
(m), 2950 (m), 1710 (s), 1700 (s), 1607 (m), 1540 (s), 1438 (m),
1450 (m), 1380 (m), 1220 (m), 1250 (m), 1150 (m), 1105 (m),
1000 (w), 860 (m), 760 (s), 620 (m), 530 (m) cm-1. LRFAB (EI+)
m/ z (% rel intensity) 734 (MH+, 100), 718 (18.0), 702 (9.0),
656 (5.0), 635 (7.0), 613 (19.0). HRFAB calcd for C40H36N3O7-
PS (MH+) m/ z 734.20897, measured m/ z 734.20900.
F MOC-P p s(su lfid e)-Ala -OH (1). Zinc powder (8.93 g, 137
mmol) was added to a solution of 2.00 g (2.73 mmol) of dimer
(FMOC-Pps(S)-Ala-OPNB) in 150 mL of acetic acid in water
(9:1). After stirring at rt for 1.5 h the mixture was filtered
through a plug of Celite. Most of the acetic acid and water
were removed under vacuum at 45 °C. The resulting yellowish
slurry was then triturated with 500 mL of CHCl3. After
filtration, the yellow CHCl3 solution was washed with 1 N
aqueous HCl. The CHCl3 solution was further washed with
brine (100 mL) and dried (MgSO4). After filtration the solvent
was removed on a rotovop. The dimer acid 1 was purified by
silica gel chromatography (ethyl acetate:hexane:HOAc ) 50:
49:1). Product weighed 1.48 g, yield was 91%.
1H NMR (300 MHz, CDCl3) δ 9.30 (s, 1H), 7.86-7.79 (m,
4H), 7.74 (d, J HH ) 7.3 Hz, 2H), 7.5 (dd, J HH ) 11.2 Hz, J HH
)
7.3 Hz, 2H), 7.4-7.2 (m, 11H), 6.32 (d, J HH ) 7.8 Hz, 1H), 4.8

5564 J . Org. Chem., Vol. 62, No. 16, 1997
Gilbertson et al.
(dt J HP ) 13.5 Hz, J HH ) 6.5 Hz, 1H), 4.3-4.0 (m, 4H), 3.12
(dd, J HP ) 10.5 Hz, J HH ) 4.9 Hz, 2H), 1.36 (d, J HH ) 6.8 Hz,
Anal. Calcd for C29H30O5N2: C 69.12; H, 6.07. Found: C,
71.57; H, 6.22.
3H). 13C NMR (75 MHz, CDCl3) δ 175.4 (s), 170.6 (d, J CP
)
F MOC-Ala -Aib-OH (2). FMOC-Ala-Aib-OBn (10.0 g, 20.5
mmol) and 0.75 g of 10% Pd-on-carbon were added to a 500
mL round bottom flask containing degassed methanol (200
mL) and equipped with a magnetic stir bar. The air was
displaced by nitrogen followed by hydrogen (via a H2 filled
balloon), and the reaction mixture was vigorously stirred. The
reaction was stopped upon consumption of the starting mate-
rial, as judged by TLC. After replacement of hydrogen with
nitrogen, the flask was opened to air and the catalyst was
removed by filtration through Celite. The Celite was washed
with 200 mL of methanol, and the solvent was removed in
vacuo, yielding 7.9 g of 2 (97% yield without purification).
1H NMR (300 MHz, CD3OD) δ 7.76 (d, J HH ) 7.7 Hz, 2H),
7.62 (br m, 2H), 7.36 (dd, J HH ) 6.8 Hz, J HH ) 6.8 Hz, 2H),
7.28 (dd, J HH ) 6.4 Hz, J HH ) 6.4 Hz, 2H), 4.34 (m, 2H), 4.18
(t, J HH ) 7.1 Hz, 1H), 4.12 (q, J HH ) 7.3 Hz, 1H), 1.48 (s, 3H),
1.46 (s, 3H), 1.30 (d, J HH ) 7.1 Hz, 3H). 13C NMR (75 MHz,
CD3OD) δ 177.8, 174.6, 158.1, 145.2, 142.5, 128.7, 128.2, 126.2,
120.0, 67.9, 57.0, 51.8, 25.3, 25.0, 18.4. IR (thin film) 3350
(m), 2950 (w), 2900 (w), 2880 (w), 1750 (s), 1650 (s), 1550 (m),
1450 (m), 1280 (m), 1250 (m), 1180 (m), 1150 (w), 800 (m),
790 (m) cm-1. LRFAB (EI+) m/ z (% rel intensity) 397 [(M +
H)+, 100], 369 (70), 337 (10), 311 (11). HRFAB calcd for
C22H25O5N2 (MH+) m/ z 397.17633, measured m/ z 397.17470.
Ac-Ala -Aib-Ala -OBn . A solution of the N-carboxyanhy-
dride of R-aminoisobutyric acid28 (10.1 g, 78.3 mmol) in 250
mL of THF was cooled, under N2, to -78 °C. This solution
was then slowly added via cannula to a stirred solution of Et3N
(22 mL, 156.6 mmol) and alanine benzyl ester p-toluene-
sulfonate (26.4 g, 78.3 mmol) in 300 mL of CH2Cl2 at -78 °C.
The mixture was stirred at -78 °C for 10 min and then slowly
warmed to room temperature. After stirring at room temper-
ature for about 30 min, approximately 80% of the solvent was
removed, and the oily mixture was placed in -20 °C freezer
for several hours. The triethylamine salt was removed by
filtration of the cold mixture through a fritted funnel. The
remaining solvent was removed under vacuum. The crude
product, Aib-Ala-OBn, was immediately used in the next
coupling reaction.
10.2 Hz), 155.6 (s), 143.5 (s), 141.1 (s), 131.5 (d, J CP ) 82.0
Hz), 131.7 (s), 131.0 (d, J CP ) 10.4 Hz), 128.5 (d, J CP ) 12.5
Hz), 127.7 (s), 127.0 (s), 125.2 (s), 119.9 (s), 67.3 (s), 51.0 (s),
48.5 (s), 46.7 (s), 33.7 (d, J CP ) 56.0 Hz), 17.6 (s). 31P NMR
(120 MHz, CDCl3) δ 39.16. IR (thin film) 3300 (s), 3050 (s),
3000 (w), 2930 (w), 2350 (m), 1710 (s), 1525 (s), 1480 (w), 1450
(m), 1430 (m), 1400 (w), 1360 (m), 1325 (w), 1250 (s), 1220 (s),
1150 (w), 1100 (m), 1050 (m), 1000 (w), 850 (w), 760 (m), 750
(m), 720 (m), 700 (m), 670 (w), 625 (w), 610 (w), 525 (w), 500
(m) cm-1. LRFAB (EI+) m/ z (% rel intensity) 599 (MH+, 100),
648 (6.0), 613 (29.0), 583 (15.0), 567 (12.0), 553 (5.0), 522 (5.0),
510 (18.0). HRFAB calcd for
C
33H31N2O5PS (MH+) m/ z
599.17694, measured m/ z 599.17680.
r-Am in oisob u t yr ic Acid Ben zyl E st er p -Tolu en e-
su lfon a te (Ts-Aib-OBn ). R-Aminoisobutyric acid (30.0 g, 290
mmol) and p-toluenesulfonic acid monohydrate (56.3 g, 296
mmol) were added to a mixture of benzyl alcohol (120 mL) and
benzene (60 mL) in a 500 mL round bottom flask. The mixture
was heated to reflux, and the water formed in the reaction
was trapped in a Dean-Stark receiver. When the reaction
stopped evolving water, the mixture was allowed to cool to
room temperature. Once at room temperature the mixture
was diluted with ether (1000 mL) and cooled in an ice bath
for 2 h. The crystalline p-toluenesulfonate salt of Aib benzyl
ester was collected by filtration, washed with ether (300 mL),
and dried in air. After recrystallization from methanol-ether
the salt weighed 86.5 g (82% yield).
1H NMR (300 MHz, CDCl3) δ 7.71 (d, J HH ) 8.25 Hz, 2H),
7.35 (m, 5H), 7.22 (d, J HH ) 7.8 Hz, 2H), 5.22 (s, 2H), 2.34 (s,
3H), 1.55 (s, 6H). 13C NMR (75 MHz, CDCl3) δ 172.59, 143.54,
141.67, 136.46, 129.79, 129.72, 129.53, 126.93, 126.70, 69.26,
57.94, 23.94, 21.29. IR (thin film) 3500 (br m), 3000 (br m),
1800 (s), 1675 (m), 1590 (m), 1350 (w), 1220 (s), 1200 (s), 1150
(m), 1050 (m), 1010 (m), 700 (m) cm-1. MS-FAB (EI+) m/ z (%
rel intensity) 559.4 (14), 387.3 (17), 366.2 (MH+, 9), 387.3 (20),
347.2 (10), 307.2 (8), 194.3 [(M - C7H7O3S)+, 100]. Anal.
Calcd for C18H23O5NS C, 59.16; H, 6.35. Found: C, 58.87; H,
6.51.
The crude Aib-Ala-OBn dimer was dissolved in 150 mL of
CH2Cl2 and cooled to 0 °C. After cooling, the solution was
transferred to a 0 °C CH2Cl2 (200 mL) solution of Ac-Ala-OH
(6.88 g, 52.5 mmol) in a 500 mL round bottom flask. Next,
15.2 g of DCC (73.5 mmol) and 8.5 g of HOBT (63.0 mmol)
were added to this mixture, and the reaction was kept at 0 °C
under N2 for 4 h and at rt for 20 h. The reaction mixture was
then transferred to a separatory funnel, washed with 1 N HCl
(2 × 50 mL), 1 N NaHCO3 (50 mL), water (50 mL), and brine
(50 mL), and dried over MgSO4. After filtration and removal
of the solvent under vacuum, the crude product was purified
by silica gel chromatography (75:25 ethyl acetate:hexane, Rf
) 0.3), yielding 11.9 g of Ac-Ala-Aib-Ala-OBn (60% yield).
1H NMR (300 MHz, CDCl3) δ 7.35 (br m, 5H), 6.98 (d, J HH
) 7.1 Hz, 1H), 6.70 (s, 1H), 6.13 (d, J HH ) 6.4 Hz, 1H), 5.16
(dd, J HH ) 12.0 Hz, J HH ) 24.0 Hz, 2H), 4.58 (dq, J HH ) 7.3
Hz, J HH ) 7.1 Hz, 1H), 4.32 (dq, J HH ) 6.8 Hz, J HH ) 7.1 Hz,
1H), 1.99 (s, 3H), 1.57 (s, 3H), 1.50 (s, 3H), 1.40 (d, J HH ) 7.1
Hz, 3H), 1.33 (d, J HH ) 6.8 Hz, 3H). 13C NMR (75 MHz, CDCl3)
δ 173.5, 172.9, 171.9, 170.5, 135.4, 128.6, 128.4, 128.1, 67.0,
57.2, 49.7, 48.4, 25.9, 24.5, 23.1, 17.9, 17.6. IR (thin film) 3350
(br w), 2950 (br m), 1750 (s), 1650 (m), 1550 (br w), 1400 (m),
F MOC-Ala -Aib-OBn . Ts-Aib-OBn (17.59 g, 48.2 mmol)
was weighed into a flask, and CHCl3 (300 mL) was added,
followed by 14 mL (24.6 mmol) of triethylamine. Once the
mixture turned clear it was transferred to a separatory funnel.
The solution was washed with water (2 × 100 mL) and brine
(50 mL) and dried (MgSO4). After filtration, the solvent was
removed on a rotovap. CH2Cl2 (300 mL) was then added to
the flask, and the solution was cooled to 0 °C. In a second
flask, 10.0 g of FMOC-Ala-OH (32.1 mmol, 1.0 equiv) was
dissolved in 200 mL of CH2Cl2. This solution was also cooled
to 0 °C and then transferred to the flask containing the CH2-
Cl2 solution of Aib-OBn. Next 7.38 g of EDC (38.5 mmol, 1.2
equiv) and 5.2 g of HOBT (38.5 mmol, 1.2 equiv) were added
to the reaction mixture. The reaction was kept at 0 °C under
N2 for 4 h and at rt for 20 h. The reaction mixture was then
transferred to a separatory funnel, washed with 1 N HCl (2 ×
150 mL), 1 N NaHCO3 (150 mL), water (100 mL), and brine
(100 mL), and dried over MgSO4. After filtration and removal
of the solvent, the crude product was dried under vacuum. A
14.4 g (92% crude yield) amount of the product (FMOC-Ala-
Aib-OBn) was obtained. This material was sufficiently pure
for use in the next step.
1H NMR (300 MHz, CDCl3) δ 7.76 (d, J HH ) 7.7 Hz, 2H),
7.58 (d, J HH ) 6.8 Hz, 2H), 7.3 (m, 9H), 6.95 (s, 1H), 5.70 (d,
J HH ) 7.8 Hz, 1H), 5.14 (s, 2H), 4.35 (d, J HH ) 6.9 Hz, 2H),
4.30 (m, 1H), 4.20 (t, J HH ) 6.4 Hz, 1H), 1.55 (s, 6H), 1.32 (d,
J HH ) 6.4 Hz, 3H). 13C NMR (75 MHz, CDCl3) δ 173.9, 171.5,
155.9, 143.6, 141.2, 135.5, 128.4, 128.2, 128.0, 127.7, 127.0,
125.0, 119.9, 67.1, 67.0, 56.4, 50.3, 47.0, 24.6, 18.7. IR (thin
film) 3350 (w), 2950 (m), 2830 (w), 1750 (s), 1710 (m), 1700
(s), 1550 (m), 1425 (m), 1380 (m), 1380 (m), 1250 (s), 1150 (w),
800 (w), 790 (m) cm-1. MS-FAB (EI+) m/ z (% rel intensity)
553.9 (8), 487.8 (MH+, 100), 461.7 (18.0). HRMS calcd for
C29H30O5N2 (MH+) m/ z 487.22328, measured m/ z 487.22385.
1350 (s), 1150 (w), 900 (w) cm-1
. LRFAB (EI+) m/ z (% rel
intensity) 378 (MH+, 100), 369 (50), 336 (5), 317 (12). HRFAB
calcd for C19H27O5N3 (MH+) m/ z 378.20288, measured m/ z
378.20340.
Ac-Ala -Aib-Ala -OH (3). Ac-Ala-Aib-Ala-OBn (1.83 g, 4.85
mmol) and 0.3 g of 10% Pd-on-carbon were added to a 250 mL
round bottom flask of degassed methanol (100 mL) equipped
with a magnetic stirrer. The air was displaced by nitrogen
followed by hydrogen (via a H2 filled balloon), and the reaction
mixture was vigorously stirred. The reaction was stopped
(28) Hanby, W. E.; Waley, S. G.; Watson, J . J . Chem. Soc. 1950,
3009.

Phosphorus-Containing Peptide Ligands
J . Org. Chem., Vol. 62, No. 16, 1997 5565
upon consumption of the starting material, as judged by TLC.
After replacement of hydrogen with nitrogen the flask was
open to air, and the reaction mixture was filtered through
Celite to remove the catalyst. The Celite was washed with
200 mL of methanol. The solvent was removed in vacuo,
yielding 1.36 g of 3 (crude yield 98%).
1H NMR (300 MHz, DMSO-d6) δ 8.14 (br s, 1H), 8.12 (s,
1H), 7.41 (d, J HH ) 6.5 Hz, 1H), 4.10 (m, 2H), 1.81 (s, 3H),
1.34 (s, 3H), 1.31 (s, 3H), 1.19 (d, J HH ) 6.3 Hz, 3H), 1.14 (d,
J HH ) 6.4 Hz, 3H). 13C NMR (75 MHz, DMSO-d6) δ 174.0,
173.5, 172.3, 169.6, 55.8, 49.0, 47.8, 26.0, 23.9, 22.3, 17.2, 17.1.
IR (thin film) 3300 (w), 3000 (w), 2950 (m), 2880 (w), 1750 (s),
1650 (m), 1550 (br w), 1450 (br w), 1350 (s), 1210 (s), 1120 (w)
cm-1. LRFAB (EI+) m/ z (% rel intensity) 378 (MH+, 100), 369
(50), 336 (5), 317 (12). HRFAB calcd for C12H21O5N3 (MH+)
m/ z 288.15593, measured m/ z 288.15870.
fractions weighed 0.6 g. Total yield was 81% relative to 2.4
mmol resin used.
1H NMR (300 MHz, CDCl3) δ 7.90 (br m, 19H), 7.40 (br m,
13H), 4.35 (br m, 4H), 4.10(m, 1H), 3.95 (m, 2H), 3.62 (m, 1H),
3.40 (m, 1H), 3.10 (br m, 4H), 2.05 (s, 3H), 1.50 (br m, 39H).
31P NMR (120 MHz, DMSO-d6) δ 39.98, 39.59. IR (thin film)
3350 (w), 3325 (br s), 3075 (w), 2990 (w), 2950 (w), 1675 (s),
1550 (s), 1425 (m), 1220 (m), 1180 (s), 1110 (w), 710 (w), 700
(w), 600 (w) cm-1. LRFAB (EI+) m/ z (% rel intensity) 1409
[(M + Na)+, 100], 1387 (MH+, 10), 1379 (30), 1332 (40), 1298
(32). HRFAB calcd for C65H88O14N12P2S2 (MNa+) m/ z
1409.53570, measured m/ z 1409.53650.
Ac-Ala -Aib-Ala -P p s(S)-Ala -Ala -Aib-P p s(S)-Ala -Ala -Aib-
Ala -Ala -OCH3 (5). Peptide 4 (1.69 g, 1.22 mmol) was dis-
solved in 100 mL of CH2Cl2 in a 200 mL round bottom flask,
and the solution was cooled to 0 °C. To this solution were
added HCl‚Ala-OMe (0.85 g, 1.22 mmol), morpholine (1.35 mL,
6.01 mmol), EDC (1.17 g, 6.01 mmol) and HOBT (0.66 g, 6.01
mmol). The reaction was stirred under N2 at 0 °C for 4 h and
at room temperature for 20 h. The reaction mixture was then
transferred to a separatory funnel, washed with 1 N HCl (2 ×
30 mL), 1 N NaHCO3 (30), water (30 mL), and brine (30 mL),
and dried over MgSO4. The solvent was removed under
vacuum. The crude product was then dissolved in a minimum
amount of CH3CN in a 100 mL flask. The solvent was allowed
to evaporate at room temperature until a white solid started
to precipitate. The flask was capped and moved into a -20
°C freezer for further crystallization. The crystals were filtered
and placed in vacuum, and the mother liquor was subjected
to HPLC purification. HPLC conditions were the following:
column: 22.5-mm × 25-cm ECONOSIL C18 10U; eluent:
gradient flow of 95% of solvent B (99% H2O/1% TFA) and 5%
of solvent A (10% H2O/89% CH3CN/1% TFA) to 5% of solvent
B and 95% of solvent A over 60 min.; flow rate: 8 mL/min.;
retention time of peptide 5: 55.4 min. The total peptide
weighed 1.17 g (66% yield).
Gen er a l Clea va ge P r oced u r e (Ra p id Am id e, Wa n g a n d
Rin k Resin s ba sed on 0.1 m m ol of r esin ). The resin was
soaked in methanol for 1 min and then drained (three times).
For cleavage of RapidAmide and Wang resins, after drying for
10 min, 90% trifluoroacetic acid (TFA), 5% 1,2 ethanedithiol
(EDT), and 5% water were added. The reaction mixture was
agitated for 16 h for RapidAmide resin or 1 h for Wang resin.
For Rink resin, after drying for 10 min, 50% (v/v) TFA/CH2-
Cl2 was added, and the resin was stirred for 2 min. After
reaction, the TFA solution was removed by filtration, and the
resin was washed with 0.5 mL of reaction solution. The
volume of TFA was reduced to 1-2 mL under a stream of
nitrogen. The crude peptide was then precipitated by addition
of 25 mL of diethyl ether followed by cooling to -78 ˚C. The
ether layer was removed and crude peptide was dried under
vacuum. The crude peptide was then purified by reverse-
phase HPLC.
Ac-Ala -Aib-Ala -P p s(S)-Ala -Ala -Aib-P p s(S)-Ala -Ala -Aib-
Ala -OH (4). Sm a ll Sca le Syn th esis. This peptide was
synthesized by the standard solid phase FMOC protocol.29
FMOC-Ala-Wang resin (0.2 mmol equiv total) was used. All
couplings were complete in 2 h. Cleavage of the completed
peptide from resin was performed according to the procedures
described above. The peptide was purified by HPLC. HPLC
conditions were the following: column: 22.5-mm × 25-cm
ECONOSIL C18 10U; eluent: gradient flow of 95% of solvent
B (99% H2O/1% TFA) and 5% of solvent A (10% H2O/89% CH3-
CN/1% TFA) to 5% of solvent B and 95% of solvent A over 60
min.; flow rate: 8 mL/min.; retention time of peptide 1: 49.0
min. After HPLC purification, 144 mg of the peptide 4 was
obtained. Yield was 52% relative to 0.2 mmol resin used.
La r ge Sca le Syn th esis. A 4.45 g amount of FMOC-Ala-
Wang resin was used (2.4 mmol equiv). A 4.8 mmol amount
of each short peptide residue was used per coupling (1.9 g of
FMOC-Ala-Aib-OH; 2.9 g of FMOC-Pps(S)-OH; 1.4 g of Ac-
Ala-Aib-Ala-OH). A 4.5 h period was allowed for each coupling
step. After all the couplings were complete, the peptide was
cleaved from the resin by adding 60 mL of a solution of 90%
TFA, 5% 1, 2-ethanedithiol (EDT), and 5% water to the schlink
funnel which was then shaken for 30 min. The solution was
filtered into a 250 mL round bottom flask, and the resin was
rinsed with 20 mL of TFA. The filtrate was concentrated on
a rotovap. The residue was dissolved in 100 mL of CH2Cl2
and washed with water (50 mL) and brine (50 mL). The CH2-
Cl2 was then removed under vacuum. The oily product was
dissolved in a minimum amount of CH3CN. The solvent was
then allowed to evaporate at room temperature until a white
solid started to precipitate. The flask was then capped and
moved into a -20 °C freezer for further crystallization. The
crystals were filtered and vacuum dried (weight 2.1 g). The
mother liquor was subjected to HPLC purification (HPLC
conditions were the same as those used in purification of
peptide 4 from small scale synthesis). The combined peptide
1H NMR (600 MHz, DMSO-d6) δ 8.39 (s, 1H), 7.98 (d, J HH
) 6.5 Hz, 1H), 7.95 (s, 1H), 7.90 (d, J HH ) 5.8 Hz, 1H), 7.86
(m, 6H), 7.82 (d, J HH ) 6.0 Hz, 1H), 7.77 (d, J HH ) 6.5 Hz,
2H), 7.73 (d, J HH ) 6.5 Hz, 1H), 7.59 (d, J HH ) 6.5 Hz, 1H),
7.58 (m, 1H), 7.51 (m, 15H), 7.39 (d, J HH ) 6.0 Hz, 1H), 7.22
(d, J HH ) 7.8 Hz, 1H), 4.46 (m, 1H), 4.36 (m, 1H), 4.21 (dq,
J HH ) 6.5 Hz, J HH ) 7.2 Hz, 1H), 4.16 (m, 3H), 4.09 (m, 1H),
4.01 (dq, J HH ) 6.0 Hz, J HH ) 6.5 Hz, 1H), 3.94 (dq, J HH ) 6.5
Hz, J HH ) 7.6 Hz, 1H), 3.68 (dq, J HH ) 5.8 Hz, J HH ) 7.5 Hz,
1H), 3.31 (s, 3H), 3.20 (m, 2H), 3.24 (m, 2H), 1.81 (s, 3H), 1.33
(s, 3H), 1.29 (s, 3H), 1.27 (m, 6H), 1.26 (s, 3H), 1.24 (m, 9H),
1.20 (d, J HH ) 7.2 Hz, 3H), 1.18 (m, 6H), 1.12 (s, 3H), 1.08 (s,
3H), 1.04 (d, J HH ) 7.5 Hz, 3H). 31P NMR (120 MHz, CDCl3)
δ 39.86, 39.48. IR (thin film) 3350 (w), 3325 (br s), 3075 (w),
2990 (w), 2950 (w), 1675 (s), 1550 (s), 1425 (m), 1220 (m), 1110
(w), 710 (w), 700 (w), 600 (w) cm-1. LRFAB (EI+) m/ z (% rel
intensity) 1494.6. [(M + Na)+, 100], 1478.7 (24), 1426.7 (13),
1451.6 (5). HRFAB calcd for C69H95O15N13P2S2 (MNa+) m/ z
1494.58845, measured m/ z 1494.58850.
The proton assignments for the peptide sequence were make
by 2D NMR (TOCSY, NOESY) and are listed in Table 2.
Ac-Ala -Aib-Ala -P p s(S)-Ala -Ala -Aib-P p s(S)-Ala -Ala -Aib-
Ala -NH2 (6). Both RapidAmide and Rink resins were used
in the synthesis of this peptide (0.2 mmol). The peptide was
synthesized by the standard solid phase coupling using pen-
tafluorophenol-activated amino acids. All couplings were
complete in 2 h. Cleavage of the completed peptide from the
resins was according to the procedures described above. The
peptide was purified by HPLC under the following condi-
tions: column: 22.5-mm × 25-cm ECONOSIL C18 10U; eluent:
gradient flow of 95% of solvent B (99% H2O/1% TFA) and 5%
of solvent A (10% H2O/89% CH3CN/1% TFA) to 5% of solvent
B and 95% of solvent A over 60 min; flow rate: 8 mL/min.;
retention time of peptide 6: 54.6 min. After HPLC purifica-
tion, 139 mg of the peptide 6 was obtained using FMOC-Ala-
RaMPS resin (50% yield), and 101 mg of the same peptide was
obtained using Rink resin (36% yield).
(29) Bodanszky, M.; Bodanszky, A. The Practice of Peptide Synthesis;
Springer-Verlag: Berlin, 1984.
(30) The author has deposited atomic coordinates for this structure
with the Cambridge Crystallographic Data Centre. The coordinates
can be obtained, on request, from the Director, Cambridge Crystal-
lographic Data Centre, 12 Union Road, Cambridge, CB2 1EZ, UK.
1H NMR (600 MHz, DMSO-d6) δ 8.40 (s, 1H), 7.99 (d, J HH
) 6.9 Hz, 1H), 7.98 (s, 1H), 7.82 (m, 11H), 7.77 (d, J HH ) 7.6
Hz, 1H), 7.73 (d, J HH ) 6.5 Hz, 1H), 7.60 (m, 15H), 7.44 (d,

5566 J . Org. Chem., Vol. 62, No. 16, 1997
Gilbertson et al.
J HH ) 6.5 Hz, 1H), 7.20 (d, J HH ) 7.2 Hz, 1H), 6.94 (s, 1H),
6.91 (s, 1H), 4.47 (br m, 1H), 4.34 (m, 1H), 4.22 (m, 1H), 4.15
(m, 1H), 4.10 (m, 1H), 3.97 (m, 2H), 3.93 (m, 1H), 3.68 (m,
1H), 3.30 (br m, 4H), 1.81 (s, 3H), 1.20 (br m, 39H). 31P NMR
(120 MHz, CDCl3) δ 39.46, 39.16. IR (thin film) 3350 (w), 3325
(br s), 3075 (w), 2990 (w), 2950 (w), 1800 (m), 1775 (w), 1675
(s), 1550 (s), 1425 (m), 1220 (m), 1180 (s), 1110 (w), 710 (w),
complex was purified by silica gel chromatography. A column
of 15 mL of silica gel was capped with a septum, and 100 mL
of degassed CH2Cl2 was passed through the column before
crude complex 7 was loaded. The column was then eluted with
30 mL of degassed CH2Cl2 followed by 50 mL of degassed CH3-
OH. The yellow methanol fraction was collected, and solvent
was removed in vacuo.
700 (w), 600 (w) cm-1
.
LRFAB (EI+) m/ z (% rel intensity)
1H NMR (600 MHz, DMF-d7) δ 9.04 (s, 1H), 8.48 (d, J HH
)
1408.5 [(M + Na)+, 28], 1386.4 (MH+, 24), 1369.4 [(M - H2O)+,
100],1298.3 (65), 1213.4 (25), 1142.5 (13), 784.4 (87), 699.3 (67),
628.3 (52), 557.3 (38). HRFAB calcd for C65H89O13N13P2S2
(MNa+) m/ z 1408.55168, measured m/ z 1408.55180.
The assignment of the protons was made by 2D NMR
(TOCSY, NOESY) and is listed in Table 3.
Ac-Ala -Aib -Ala -P p s-Ala -Ala -Aib -P p s-Ala -Ala -Aib -Ala -
NH2. Raney nickel (300 mg) was washed with degassed
methanol (3 × 2 mL) and then transferred, using 3 mL of
methanol, into a 25 mL Schlenk tube containing 21 mg (15
µmol) of peptide 6. The mixture was pump-thaw degassed
(three cycles) and was stirred for 12 h at 48 °C. After cooling
to room temperature, the reaction mixture was filtered under
N2 into a 10 mL two-neck round bottom flask. Degassed
methanol (10 mL) was used to rinse the Raney nickel, and
the solvent was removed in vacuo. 31P NMR was used to
monitor the course of the desulfurization. The reduced peptide
was immediately used in the next step. 31P NMR (120 MHz,
CD3OD) δ -20.85, -23.79.
5.14 Hz, 1H), 8.28 (s 1H), 8.26 (d, J HH ) 4.63 Hz, 1H), 8.24 (d,
J HH ) 5.31 Hz, 1H), 8.14 (d, J HH ) 4.96 Hz, 1H), 8.02 (d, J HH
) 5.60 Hz, 1H), 7.89 (d, J HH ) 6.68 Hz, 1H), 7.71 (d, J HH
)
8.00 Hz, 1H), 7.67 (d, J HH ) 6.17 Hz, 1H), 7.62 (br m, 11H),
7.57 (s, 1H), 7.45 (m, 14H), 7.33 (d, J HH ) 6.85 Hz, 1H), 7.04
(s, 1H), 7.12 (s, 1H), 4.32 (m, 1H), 4.31 (m, 1H), 4.21 (m, 1H),
4.04 (m, 1H), 4.00 (m, 1H), 3.92 (m, 1H), 3.86 (m, 1H), 3.84
(m, 1H), 3.64 (s, 3H), 3.61 (s, 3H), 3.61 (m, 1H), 3.18 (m, 1H),
3.03 (m, 1H), 3.11 (m, 1H), 2.84 (br m, 1H), 1.96 (s, 3H), 1.20
(br m, 39H) cm-1
.
31P NMR (120 MHz, CD3OD) δ 27.24 (d,
J Rh-P ) 168.0 Hz, 1P), 23.85 (d, J Rh-P ) 168.0 Hz, 1P). MS-
FAB (EI+) m/ z (% rel intensity) 1516.9 [(M + H)+ with NBD
coordinated to the rhodium (43)], 1426.5 (87), 1334.7 (40),
1241.5 (20), 1179.5 (35), 1108.5 (30), 978.5 (100), 868.5 (22),
609.2 (23), 522.5 (37).
The assignment of the protons was made by 2D NMR
(TOCSY, NOESY) and is listed in Table 4.
Rh od iu m Com p lex of Ac-Ala -Aib-Ala -P p s-Ala -Ala -Aib-
P p s-Ala -Ala -Aib-Ala -NH2 (7). [Rh(NBD)Cl]2 (3.46 mg, 7.5
µmol) and AgClO4 (3.11 mg, 15.0 µmol) were weighed into a 1
mL vial containing a small stir bar. Degassed MeOH (0.50
mL) was added, and the mixture was stirred under N2 for 15
min. The precipitated AgCl was then filtered off through a
microfilter (pore size: 45 µm), and the filtrate was added to
the two-neck round bottom flask containing the diphosphine
peptide. The solution was stirred under N2 for 10 min during
which time the color of the solution turned from yellow to
orange. Methanol solvent was then removed in vacuo. The
Ack n ow led gm en t. This work was supported by the
donors of the Petroleum Research Fund, administered
by the American Chemical Society. We gratefully
acknowledge the Washington University High-Resolu-
tion NMR Facility, partially supported by NIH
1S10R02004, and the Washington University Mass
Spectrometry Resource Center, partially supported by
NIHRR00954, for their assistance.
J O970773E