Selective side chain introduction onto small peptides mediated by samarium diiodide: A potential route to peptide libraries

Article abstract of DOI:10.1021/ja0024800

A mild and simple method for the selective introduction of carbinol side chains onto glycine residues in peptides is presented as a potential route for the preparation of peptide libraries. A series of di- and tripeptides, as well as one tetrapeptide, eac

Full text of DOI:10.1021/ja0024800

VOLUME 122, NUMBER 50
DECEMBER 20, 2000
© Copyright 2000 by the
American Chemical Society
Selective Side Chain Introduction onto Small Peptides Mediated by
Samarium Diiodide: A Potential Route to Peptide Libraries
Marina Ricci, Peter Blakskjær, and Troels Skrydstrup*
Contribution from the Department of Chemistry, UniVersity of Aarhus, Langelandsgade 140,
8000 Aarhus C, Denmark
ReceiVed July 7, 2000
Abstract: A mild and simple method for the selective introduction of carbinol side chains onto glycine residues
in peptides is presented as a potential route for the preparation of peptide libraries. A series of di- and tripeptides,
as well as one tetrapeptide, each possessing one glycine residue, were first selectively functionalized at this
amino acid unit by a two-step sequence involving bromination with N-bromosuccinimide and then sulfide
formation by treatment of the unstable glycyl bromide with 2-mercaptopyridine. These modified peptides were
then reduced with samarium diiodide at room temperature in the presence of alkyl aldehydes and ketones,
affording a series of peptides in yields of 40-65% containing serine/threonine derivatives as new functionalities.
These reactions are quite efficient, considering the presence of as many as four amide protons in the enolate
intermediate. The diastereoselectivities of these reactions are low or nonexistent, which is ascribed to either
(a) the formation of single enolate, where the neighboring chiral centers impart no influence in the alkylation
step or (b) the generation of an enolate mixture, where each stereoisomer leads to opposite enantiomers with
respect to the newly formed amino acid upon alkylation. The successful nonselective double alkylation of the
tripeptide, Bz-Gly-Val-Gly-OMe, suggests the possibility that the reductive samariation approach to the
C-alkylation of peptides may be a viable route for the preparation of peptide libraries based on multiple serine/
threonine derivatives. Finally, a preliminary investigation on one peptide has shown that the addition of 1%
of nickel(II) iodide to these condensation reactions has a significant effect on the coupling yields.
Introduction
peptides through commercially available or synthetic nonnatural
amino acids.1 Such an approach is time-consuming and has led
to the development of techniques such as parallel synthesis or
combinatorial chemistry. A more convergent approach to the
modification of biologically active peptides would be to
incorporate the side chain directly onto the peptide backbone.
This route would be more attractive and in particular more
economical, as a multitude of analogues may be quickly
synthesized from a single and intact peptide, without having to
repeat the peptide synthesis for each modification.
Small peptides are implicated in a wide range of roles in both
plants and animals, acting as extracellular messengerss
hormones, neurotransmitters, and neuromodulatorsswhich in-
fluence essential functions such as metabolism, immune defense,
respiratory functions, and reproduction to name a few. Because
of these vital functions, bioorganic and medicinal chemists have
developed a long-standing interest in the application of such
peptides as pharmaceuticals for the treatment of diseases related
to the above-mentioned biological processes. However, for
efficient target screening and optimization of lead structures,
flexible synthetic approaches are required for modifying natural
peptides. Classical methods rely on stepwise syntheses of
(1) For a recent example on the modification of a peptide via the
incorporation of nonnatural amino acids resulting in the improvement of
its biological activity, see, De Filippis, V.; Quarzago, D.; Vindigni, A.; Di
Cera, E.; Fontana, A. Biochemistry 1998, 37, 13507.
10.1021/ja0024800 CCC: $19.00 © 2000 American Chemical Society
Published on Web 12/01/2000

12414 J. Am. Chem. Soc., Vol. 122, No. 50, 2000
Ricci et al.
Various reactive intermediates have been employed for the
C-C coupling step in a peptide chain including glycine
enolates,2-5 glycine cation equivalents,6,7 as well as glycyl
radicals.8-12 Perhaps the most noteworthy in this respect
represents the extensive work performed by Seebach and co-
workers on the selective alkylation of glycine residues in a
polylithiated peptide enolate.2 With this approach, linear and
cyclic peptides containing up to 11 amino acid residues could
be alkylated at a glycine residue via the initial generation of a
corresponding glycine enolate. The preparation of such enolates
first requires a low-temperature multiple deprotonation step of
the acidic amide protons with LDA, accompanied by a selective
deprotonation of one of the R-CH protons of glycine. Successful
enolate formation does, however, require that the C-terminal
amide linkage of the glycine unit be alkylated. Resolution of
this restriction was subsequently obtained by incorporating an
electron withdrawing substituent on the glycine R-carbon such
as an alkyl carboxylate or cyanide group.2e In this way, the pKa
of the R-C-H proton becomes sufficiently acidic such that
deprotonation is selective only at this site. An ensuing reduction
step is nevertheless necessary for removal of this assisting
functionality, which also results in loss of any stereochemical
information obtained in the alkylating step.
An alternative route to glycine enolates would be to perform
a reductive metalation step of an appropriately R-C-function-
alized glycine unit. Employing this route, the glycine enolate
may be prepared directly via a glycyl radical intermediate
without competitive deprotonation of amide protons with strong
bases (Scheme 1). However, successful application of this
(2) For a comprehensive review on the use of peptide enolates by Seebach
and co-workers, see: (a) Seebach, D.; Beck, A. K.; Studer, A. In Modern
Synthetic Methods; Ernst, B., Leumann, C., Eds.; VCH: Weinheim, 1995;
Vol. 7, p. 1. See also, (b) Seebach, D.; Bossler, H. G.; Gru¨ndler, G.; Shoda,
S.; Wenger, R. HelV. Chim. Acta 1991, 74, 197. (c) Seebach, D.; Beck, A.
K.; Bossler, H. G.; Gerber, C.; Ko, S. Y.; Murtiashaw, C. W.; Naef, R.;
Shoda, S.; Thaler, A.; Krieger, M.; Wenger, R. HelV. Chim. Acta 1993, 76,
1564. (d) Bossler, H. G.; Seebach, D. HelV. Chim. Acta 1994, 77, 1124. (e)
Seebach, D.; Bezenc¸on, O.; Jaun, B.; Pietzonka, T.; Matthews, J. L.; Ku¨hnle,
F. N. M.; Schweizer, W. B. HelV. Chim. Acta 1996, 79, 588. (f) Matt, T.;
Seebach, D. HelV. Chim. Acta 1998, 81, 1845 and references therein.
(3) Ager, D. J.; Froen, D. E.; Klix, R. C.; Zhi, B.; McIntosh, J.;
Thangarasa, R. Tetrahedron 1994, 50, 1975.
Scheme 1
approach would greatly depend on the stability of the enolate
species toward either intra- or intermolecular protonation from
the amide functionalities. The strong metal-complexing abilities
of lanthanide(III) metal ions to oxygen in addition to the low
basicity of alkyl lanthanide species suggested to us that the
single-electron reducing agents, such as the well-known sa-
marium diiodide (SmI2), might be suitable for this purpose.13
Evidence that SmI2 may be ideal for such studies came about
from our previous results in the divalent samarium-induced
C-glycosylations.14 In this work, we have demonstrated that
C-C bond formations with an anomeric organosamarium could
be accomplished even in the presence of an adjacent secondary
amide group, implying that coupling is sufficiently fast com-
pared to the protonation step.14c,f,g Whether these results could
be adapted to carbon-carbon bond-forming reactions on pep-
tides only protected at the terminal positions was therefore the
subject of this investigation.
In this paper, we disclose our findings on the use of a SmI2-
induced Reformatsky-type reaction for the selective introduction
of carbinol side chains onto glycine residues. We have found
that pyridyl sulfide glycyl units may indeed serve as an
appropriate precursor to the required glycine enolates. The
scopes and limitations on the application of such entities for
the C-alkylation of peptides are discussed. This work lays the
foundation for a possible route to the rapid and simple
preparation of peptide libraries containing serine/threonine
derivatives.15
(4) (a) O’Donnell, M. J.; Burkholder, T. P.; Khau, V. V.; Roeske, R.
W.; Tian, Z. Pol. J. Chem. 1994, 68, 2477. (b) O’Donnell, M. J.; Zhou, C.;
Scott, W. L. J. Am. Chem. Soc. 1996, 118, 6070. (c) O’Donnell, M. J.;
Drew, M. D.; Pottorf, R. S.; Scott, W. L. J. Comb. Chem. 2000, 2, 172.
(5) (a) Kazmaier, U. Angew. Chem., Int. Ed. Engl. 1994, 33, 998. (b)
Kazmaier, U. J. Org. Chem. 1994, 59, 6667. (c) Kazmaier, U.; Maier, S.
Chem. Commun. 1998, 2535. (d) Kazmaier, U.; Maier, S. J. Org. Chem.
1999, 64, 4575.
(6) Steglich, W.; Ja¨ger, M.; Jaroch, S.; Zistler, P. Pure Appl. Chem. 1994,
66, 2167.
(7) (a) Renaud, P.; Seebach, D. Angew. Chem., Int. Ed. Engl. 1986, 25,
843. (b) Renaud, P.; Seebach, D. HelV. Chim. Acta 1986, 69, 1704. (c)
Seebach, D.; Charczuk, R.; Gerber, C.; Renaud, P.; Berner, H.; Schneider,
H. HelV. Chim. Acta 1989, 72, 401.
Results and Discussions
Initial efforts for the development of a reductive samariation
approach for the side-chain introduction onto peptides, first
(13) For recent reviews on the application of SmI2 in organic synthesis:
(a) Krief, A.; Laval, A.-M. Chem. ReV. 1999, 99, 745. (b) Molander, G.
A.; Harris, C. R. Chem. ReV. 1996, 96, 307. (c) Molander, G. A.; Harris,
C. R. Tetrahedron 1998, 54, 3321. (d) Skrydstrup, T. Angew. Chem., Int.
Ed. Engl. 1997, 36, 345. (e) Molander G. A. In Organic Reactions; Paquette,
L. A., Ed.; John Wiley: New York, 1994; Vol. 46, p 211. (f) Imamoto, T.
Lanthanides in Organic Synthesis; Academic Press: London, 1994. (g)
Molander, G. A. Chem. ReV. 1992, 92, 29. (h) Curran, D. P.; Fevig, T. L.;
Jaspersen, C. P.; Totleben, M. L. Synlett 1992, 943.
(8) (a) Elad, D.; Sperling, J. J. Chem. Soc. C 1969, 1579. (b) Elad, D.;
Sperling, J. J. Chem. Soc., Chem. Commun. 1969, 234. (c) Sperling, J.;
Elad, D. J. Am. Chem. Soc. 1971, 93, 967. (d) Sperling, J.; Elad, D. J. Am.
Chem. Soc. 1971, 93, 3839.
(9) (a) Easton, C. J.; Scharfbillig, I. M.; Tan, E. W. Tetrahedron Lett.
1988, 29, 1565. (b) Easton, C. J.; Tan, E. W.; Hay, M. P. J. Chem. Soc.,
Chem. Commun. 1989, 385. (c) Burgess, V. A.; Easton, C. J.; Hay, M. P.
J. Am. Chem. Soc. 1989, 111, 1047.
(10) Papadopoulos, A.; Lewall, B.; Steckhan, E.; Ginzel, K.-D.; Knoch,
F.; Nieger, M. Tetrahedron 1991, 47, 563.
(14) (a) Maze´as, D.; Skrydstrup, T.; Beau, J.-M. Angew. Chem., Int. Ed.
Engl. 1995, 34, 909. (b) Jarreton, O.; Skrydstrup, T.; Beau, J.-M. J. Chem.
Soc., Chem. Commun. 1996, 1661. (c) Urban, D.; Skrydstrup, T.; Riche,
C.; Chiaroni, A.; Beau, J.-M. Chem. Commun. 1996, 1883. (d) Jarreton,
O.; Skrydstrup, T.; Beau, J.-M. Tetrahedron Lett. 1997, 36, 303. (e)
Skrydstrup, T.; Jarreton, O.; Maze´as, D.; Urban, D.; Beau, J.-M. Chem.
Eur. J. 1998, 4, 655. (f) Urban, D.; Skrydstrup, T.; Beau, J.-M.; J. Org.
Chem. 1998, 63, 2507. (g) Andersen, L.; Mikkelsen, L. M.; Beau, J.-M.;
Skrydstrup, T. Synlett 1998, 1393. (h) Jarreton, O.; Skrydstrup, T.; Beau,
J.-M. Chem. Eur. J. 1999, 5, 430. (i) Krintel, S. L.; Jime´nez-Barbero, J.;
Skrydstrup, T. Tetrahedron Lett. 1999, 40, 7565. (j) Mikkelsen, L. M.;
Krintel, S. L.; Jime´nez-Barbero, J.; Skrydstrup, T. Chem. Commun. 2000,
2319.
(11) For a recent review on peptide radicals, see: Easton, C. J. Chem.
ReV. 1997, 97, 53.
(12) Nonnatural amino acids have also been incorporated into predeter-
mined sites of proteins employing a chemically misacylated suppressor
tRNA under in vitro protein translation conditions. This strategy has been
exploited for both the understanding of protein structure and function. See,
Short, G. F., III; Golovoine, S. Y.; Hecht, S. Biochemistry 1999, 38, 8808,
and Thorson, J. S.; Shin, I.; Chapman, E.; Stenberg, G.; Mannervik, B.;
Schultz, P. G. J. Am. Chem. Soc. 1998, 120, 451 and references therein.
(15) For a preliminary account of this work, see, Ricci, M.; Madariaga,
L.; Skrydstrup, T. Angew. Chem., Int. Ed. 2000, 39, 242.

Potential Route to Peptide Libraries
J. Am. Chem. Soc., Vol. 122, No. 50, 2000 12415
Table 1. Preparation of the Dipeptides 2-6
Table 2. SmI2-Mediated Coupling of the Dipeptides 2-6 with
Cyclohexanone
modified
dipeptide
entry
1
Nu
conditions
yield of 7a
S(2-Pyr)
(2-Pyr)SH (1.5 equiv)
DIPEA (1.5 equiv)
(a) PhSH (1.5 equiv)
DIPEA (1.5 equiv)
(b) RuCl3/NaIO4
AcOH (1.5 equiv)
DIPEA (1.5 equiv)
PhCO2H (1.5 equiv)
DIPEA (1.5 equiv)
MeOH
2 (87%)
entry
R
X
conditions
(%)
1
2
Bz S(2-Pyr) SmI2/THF 20 °C, 1 h
Bz S(2-Pyr) SmI2/THF HMPA, 20 °C,
then carbonyl cmpd
64
0
2
SO2Ph
3 (56%)
3
4
5
OAc
OBz
OMe
4 (46%)
5 (62%)
6 (26%)
3
Bz S(2-Pyr) SmI2/THF HMPA, -78 °C,
0
then carbonyl cmpd
4
5
6
7
8
9
Bz S(2-Pyr) Cp2TiCl2/Mn/THF 20 °C
Bz S(2-Pyr) Cp2TiCl2/Zn/THF 20 °C
Bz S(2-Pyr) SmI2/NiI2 (1%) 0 °C, few s
Bz S(2-Pyr) SmI2/NiI2 (1%) -78 °C, 10 min
Bz SO2Ph
Bz OAc
10
0
75
85
32
8
39
0
0
SmI2/THF 20 °C, few s
SmI2/THF 20 °C, few s
SmI2/THF 20 °C, few s
SmI2/THF 20 °C, 3 h
SmI2/THF 20 °C, 3 h
required the identification of a suitable glycine enolate precursor.
In our earlier work, we have established that a samarium
diiodide (SmI2)-induced Barbier reaction with glycosylpyridyl
sulfones is a rapid and mild approach for the stereoselective
construction of C-glycosides.14,16,17 Crucial for the successful
coupling of the intermediate anomeric organosamarium species
to carbonyl substrates are the observations that such pyridyl
sulfones are rapidly and selectively reduced by SmI2, which is
partly due to the generation of a stabilized anomeric radical.16
On the other hand, a glycyl radical would be expected to display
a greater stability due to the captodative effect (R-C-H bond
dissociation energy of approximately 80 kcal/mol),18 suggesting
that with an appropriate electrophore on the R-carbon of a
glycine residue, its SmI2-induced reduction would be even faster
than that for the corresponding glycosyl derivative.
10 Bz OBz
11 Bz OMe
12 Boc CN
a In all cases, a diastereoselectivity of 1:1 was observed.
coupling reactions. In contrast, the phenyl sulfide was oxidized
to its sulfone 3 in 75% with the RuCl3/NaIO4 combination (entry
2).
The results of the coupling experiments between the dipep-
tides 2-6 and cyclohexanone are shown in Table 2. Treatment
of a THF solution of the pyridyl sulfide 2 and an excess of
cyclohexanone (3 equiv) with 3 equiv of SmI2, resulted in the
consumption of 2 over a period of 0.5-1 h, affording a 1:1
diastereomeric mixture of alkylated dipeptides 7 in a promising
yield of 64% (entry 1). Also isolated from the reaction mixture
was the corresponding reduced peptide 1 in a 15% yield.
Reaction of the pyridyl sulfide with SmI2 prior to the addition
of cyclohexanone produced only the dipeptide 1. A similar result
was also obtained in attempts to either accelerate the electron
transfer or to stabilize the putative samarium enolate intermedi-
ate through the addition of the strong lanthanide complexing
agent, hexamethylphosphoramide (HMPA),20 as shown in entries
2 and 3. Substituting the reducing agent with the increasingly
popular biscyclopentadienyltitanium chloride (Cp2TiCl)21 gener-
ated from Cp2TiCl2/Mn or Cp2TiCl2/Zn only resulted in the slow
reduction of the pyridyl sulfide while affording a low yield of
7 in a single case (entries 4 and 5).
To test the reactivity of such modified glycine units, a variety
of electrophores were first introduced on the simple dipeptide
1 (Table 1). This required though that a procedure for the
selective functionalization of a glycine residue be available.
Previous work by Easton and co-workers has demonstrated that
glycine residues in small peptides readily undergo selective
bromination when subjected to N-bromosuccinimide.9,19 Hence,
the dipeptide, Bz-Gly-Phe-OMe (1), was irradiated in CH2Cl2
with a 150-W lamp in the presence of NBS to afford the
corresponding glycyl bromide derivative. Without isolation, this
unstable halide was immediately subjected to a variety of
nucleophiles, as depicted in Table 1. Good to modest yields of
the modified dipeptides 2-6 were obtained. The sulfide pyridyl
derivative 2 (entry 1), prepared in a satisfactory 87% yield, could
not be oxidized to the corresponding and desired sulfone as this
labile compound underwent spontaneous hydrolysis. As will be
discussed below, this has no consequence with the subsequent
The report in 1996 by the group of Namy and Kagan that as
low as 1 mol % NiI2 premixed with SmI2 accelerates the electron
transfer from the divalent lanthanide metal to substrates such
(20) (a) Otsubo, K.; Inanaga, J.; Yamaguchi, M. Tetrahedron Lett. 1986,
27, 5763. (b) Inanaga, J.; Ishigawa, M.; Yamaguchi, M. Chem. Lett. 1987,
1485. (c) Otsubo, K.; Inanaga, J.; Yamaguchi, M. Chem. Lett. 1987, 1487.
(d) Handa, Y.; Inanaga, J.; Yamaguchi, M. J. Chem. Soc., Chem. Commun.
1989, 298. (e) Ujikawa, O.; Inanaga, J.; Yamaguchi, M. Tetrahedron Lett.
1989, 30, 2837.
(16) (a) Maze´as, D.; Skrydstrup, T.; Doumeix, O.; Beau, J.-M. Angew.
Chem., Int. Ed. Engl. 1994, 33, 1383. (b) Skrydstrup, T.; Maze´as, D.;
Elmouchir, M.; Doisneau, G.; Riche, C.; Chiaroni, A.; Beau, J.-M. Chem.
Eur. J. 1997, 8, 1342.
(17) Skrydstrup, T.; Beau, J.-M. In AdVances in Free Radical Chemistry;
Zard, S. Z., Ed.; JAI Press: Stamford, 1999; Vol. 7, p 89.
(18) (a) Jonsson, M.; Wayner, D. D. M.; Armstrong, D. A.; Yu, D.; Rauk,
A. J. Chem. Soc., Perkin Trans. 2 1998, 1967. (b) Bordwell, F. G.;
Gallagher, T.; Zhang, X. J. Am. Chem. Soc. 1991, 113, 3495. (c) Clark, K.
B.; Wayner, D. D. M.; Demirjdji, S. H.; Koch, T. H. J. Am. Chem. Soc.
1993, 115, 2447. (d) Rauk, A.; Yu, D.; Taylor, J.; Shustov, G. V.; Block,
D. A.; Armstrong, D. A. Biochemistry 1999, 38, 9089.
(19) This free radical-based process was explained to proceed by a
hydrogen abstraction pathway taking place selectively at the glycine
R-carbon, owing to decreased nonbonding interactions of the generated
planar carbon radical compared to that of the branched amino acids (ref
11).
(21) For some recent application of Cp2TiCl, see: (a) RajanBabu, T.
V.; Nugent, W. A. J. Am. Chem. Soc. 1994, 116, 986. (b) Barden, M. C.;
Schwartz, J. J. Am. Chem. Soc. 1996, 118, 5484. (c) Spencer, R. P.;
Schwartz, J. J. Org. Chem. 1997, 62, 4204. (d) Gansa¨uer, A.; D. Bauer, J.
Org. Chem. 1998, 63, 2070. (e) Gansa¨uer, A.; Moschioni, M.; Bauer, D.
Eur. J. Org. Chem. 1998, 1923. (f) Gansa¨uer, A.; D. Bauer, Eur. J. Org.
Chem. 1998, 2673. (g) Dunlap, M. S.; Nicholas, K. M. Synth. Commun.
1999, 29, 1097. (h) Gansa¨uer, A.; Bluhm, H.; Pierobon, M. J. Am. Chem.
Soc. 1998, 120, 12849. (i) Gansa¨uer, A.; Lauterbach, T.; Bluhm, H.;
Noltemeyer, M. Angew. Chem., Int. Ed. 1999, 38, 2909. (j) Spencer, R. P.;
Cavallaro, C. L.; Schwartz, J. J. Org. Chem. 1999, 64, 3987. (k) Hansen,
T.; Krintel, S. L.; Daasbjerg, K.; Skrydstrup, T. Tetrahedron Lett. 1999,
40, 6087.

12416 J. Am. Chem. Soc., Vol. 122, No. 50, 2000
Ricci et al.
Scheme 2. Mechanistic Proposal for the SmI2-Induced
as alkyl halides,22 prompted us to examine the effect of this
catalyst on the yields of the SmI2-promoted C-alkylation. Hence,
a THF solution of the dipeptide 2 and cyclohexanone was
subjected to 3 equiv of SmI2 containing 1 mol % NiI2 at 0 °C
resulting in the fast consumption of 2 (few seconds). More
importantly, the coupling yield had increased from 64 to 75%
(entry 6).23 Decreasing the reaction temperature to -78 °C, also
led to a relatively fast reaction (few min), as well as a rise in
the yield of 7 to an impressive 85% yield (entry 7). Interestingly,
the diastereoselectivity was not effected by the temperature
change.
Alkylation of Pyridyl Sulfide-Modified Peptides
Although, we did not expect the phenyl sulfone derivative 3
to display an appreciable reactivity with the lanthanide-reducing
reagent in comparison to previous results obtained in our
C-glycosylation studies,16 we were nevertheless surprised to
observe that this compound reacted essentially instantaneously
with SmI2 (entry 8). Also unexpected, were the comparable and
fast reactivities (few seconds) of the acetate 4 (entry 9) and the
benzoate 5 (entry 10) with SmI2, while the methoxide 6
remained untouched even after several hours of subjection to
SmI2 (entry 11). Whereas, their characteristically low coupling
yields made these reactions synthetically undesirable, the
different reactivity between the pyridyl sulfide 2 with 3-5
suggested the possibility that two mechanisms may be operating
in the enolate formation step.
Scheme 3. Alternative Mechanisms for the SmI2-Promoted
Reduction of R-Substituted Glycine Residues
Finally, we also tested the cyanide derivative (entry 12), easily
prepared via a peptide-coupling step with a cyano-containing
amino acid building block.2e However, the peptide proved
unreactive to SmI2 alone at 20 °C and could only slowly be
reduced to the unfunctionalized peptide in the presence of
HMPA, conditions which are not adabtable for the succeeding
alkylation step.24
Although, an in-depth mechanistic investigation of the above
coupling experiments was not pursued, a tentative explanation
of these results is illustrated in Schemes 2 and 3. With the
pyridyl sulfide group, activation of the carbon-nitrogen double
bond through the coordination of the divalent lanthanide reagent
could lead to the electron tranfer into the heteroaryl nucleus as
with 8. C-S bond fragmentation is thereafter facilitated by the
formation of the captodatively stabilized glycyl radical 9.25
A
subsequent reduction step with a second equivalent of SmI2 then
leads to the SmIII enolate intermediate 10. Although, the
geometry of this enolate is unknown, the high oxophilicity of
the lanthanide metal ion, including its ability to form strong
complexes with amides, could suggest a coordination of the
SmIII ion to one of the adjacent amide groups of the peptide,
leadingtothepossibleformationofacyclicenolatestructure.14c,f,g,27
Unfortunately, owing to the absence of any diastereoselectivity
in this coupling event, it does not appear that the distal side
chain on the adjacent amino acid has any directing abilities with
respect to the incoming electrophile.28 Another explanation may
be that several enolate structures are produced under these
conditions which lead to products of opposing diastereoselec-
tivity in the coupling step to cyclohexanone.
It is not exactly clear how NiI2 exerts its rate-accelerating
effect on the electron transfer to the pyridyl sulfide group,
whether a SmI2/NiI2 complex or Ni(I) species is involved.30
Nevertheless, the improved coupling yield observed with NiI2,
(22) Machrouhi, F.; Hamann, B.; Namy, J.-L.; Kagan, H. B. Synlett 1996,
633.
(23) Support for this study was recently provided by the group of Beau,
demonstrating that the application of the NiII catalyst to the SmI2-induced
Barbier reaction of glycosylpyridyl sulfones with aldehydes has a dramatic
positive influence on the coupling yields. Miquel, N.; Doisneau, G.; Beau,
J.-M. Angew. Chem., Int. Ed., in press.
(24) For other examples concerning the reductive removal of a cyanide
group with SmI2/HMPA, see: (a) Kang, H.-Y.; Hong, W. S.; Cho, Y. S.;
Koh, H.-Y. Tetrahedron Lett. 1995, 36, 7661. (b) Molander, G. A.; Wolfe,
J. P. J. Brazilian Chem. Soc. 1996, 7, 335.
(25) Similar mechanisms have recently been proposed in the C-
glycosylation of glycosylpyridyl sulfones and the zinc-mediated reduction
of alkylpyridyl sulfides (refs 16, 26).
(28) An alternative mechanism, which may explain the low diastereo-
selectivity observed, involves the addition of the intermediate and uncom-
plexed glycyl radical to the aldehyde or ketone. Although direct intermo-
lecular addition to the carbonyl group is not likely due to the formation of
a highly reactive oxygen centered radical, activation of this functionality
by precomplexation to SmI2 prior to the radical addition step would result
in, (1) a lowering of the π*CdO energy level, thus favoring radical addition,
and (2) an immediate reduction of the reactive oxy-radical intermediate.
However, the high stability of the glycyl radical as observed by the relatively
weak R-C-H bond of a glycine unit in conformationally unrestricted
peptides (ref 18) suggests that the reactivity of this intermediate towards
addition reactions is probably unfavorable (ref 29).
(26) Boivin, J.; Lallemand, J.-Y.; Schmitt, A.; Zard, S. Z. Tetrahedron
Lett. 1995, 36, 7243.
(27) (a) Kawatsura, M.; Dekura, F.; Shirahama, H.; Matsuda, F. Synlett
1996, 373. (b) Molander, G. A.; McWilliams, J. C.; Noll, B. C. J. Am.
Chem. Soc. 1997, 119, 1265.
(29) We have recently observed the reluctancy of the glycyl radical to
add to alkenes under conditions which favor this addition step. Blakskjær,
P.; Pedersen, L.; Skrydstrup, T. J. Chem. Soc., Perkin Trans. 1, in press.

Potential Route to Peptide Libraries
J. Am. Chem. Soc., Vol. 122, No. 50, 2000 12417
Table 3. Introduction of the Pyridyl Sulfide and Subsequent
SmI2-Mediated Alkylation of a Series of Peptides with
Cyclohexanone
could be explained by its ability to allow the reductive coupling
to occur at low temperature, conditions where the rate of the
competitive proton transfer to the enolate intermediate is
substantially decreased compared to the room-temperature
reactions.31
In previous work with C-glycosides, we noted that the order
of decreasing reactivity of glycosyl arylsulfides and sulfones
with SmI2 was the following: (2-Pyr)SO2 > (2-Pyr)S >
PhSO2,14,16 whereas glycosyl benzoates were completely inert.32
The reversal of this order of reactivity with the modified
dipeptide (Table 2, entries 1, 8-10), therefore suggests the
possibility of two different mechanisms taking place in the initial
reduction step to the glycyl radical. For the phenyl sulfone-,
benzoate-, and acetate-derivative their complexation with SmI2
may lead to their rapid expulsion through the formation of an
iminium ion intermediate 11 (Scheme 3).33 This reactive species
is then immediately reduced to the R-carbon-centered radical.
We do not yet understand the reason for the low coupling yields
with these substrates, but reduction of the conjugated iminium
ion may not necessarily afford the required glycyl radical, if
the electron transfer occurs at one of the two neighboring
carbonyl groups.
Finally, the known capacity for SmI2 to promote the reductive
removal of R-heterosubstituted carbonyl compounds,34 including
amides,35 may suggest a completely different scenario for the
formation of enolate 10 (Scheme 3). Hence, reversible electron
transfer to the amide CdO bond leads to the carbon-centered
radical 12. Its reduction via a second electron transfer followed
by a â-elimination step would then generate the enolate 10. In
these examples, the heterosubstituent not only provides an extra
chelating site for the divalent samarium ion, but their electron-
withdrawing properties would increase the carbonyl group’s
propensity to undergo reduction.
With the identification of the pyridyl sulfide group as an
appropriate electrophore for C-C-bond formation reactions of
the dipeptide 2, we set out to examine the generality of this
approach. A series of dipeptides, tripeptides, and one tetrapeptide
were first subjected to the two-step bromination and thiopyridine
introduction sequence, affording modest to good yields of the
pyridyl sulfides 13-21 (Table 3, entries 1-9). In general, these
reactions were quite clean though the reactivity of the glycine
residue toward bromination diminished with increasing size of
the peptide. In all cases, an approximately 1.6-1.7 mixture of
diastereomers at the new stereocenter was obtained, the ratio
of which had no consequence in the subsequent coupling step.
As with 2, treatment of the peptides 13-21 with cyclohex-
anone and then SmI2 afforded good yields of the C-alkylated
peptides in the range of 40-63% (Table 3, entries 1-9), which
are satisfactory considering the simplicity of the reaction
a Ratio refers to the diastereoselectivity observed. b Yield based on
recovered starting material.
conditions. It was pleasing to observe that the coupling yields
appeared independent of the peptide chain length as exemplified
by the 54% yield obtained for the tetrapeptide 31 (entry 9) eVen
in the presence of four amide protons. The simple reduction
compounds were the major byproducts probably arising via the
proton transfer from the amide groups or THF. As with the
dipeptide 2, essentially no stereoselectivity was noted for these
coupling reactions, which is again quite remarkable, considering
the number of potential sites for complexation with the SmIII
metal ion, and the high diastereoselectivities recently seen in
the SmI2-promoted Reformatzky reaction of R-bromoacetyl-2-
oxazolidinones with aldehydes.36 Possibly, deviation of the
peptide from the normal linear conformation is energetically to
costly to provide an additional complexation site on the SmIII
metal ion of the enolate.
(30) Previous experiments in the laboratory of Namy have demonstrated
that mixing stoichiometric quantities of the nickel catalyst with SmI2 does
not lead to a reactive electron-donating agent. Machrouhi, F. Ph.D. Thesis,
Universite´ de Paris XI, 1999.
(31) The application of the NiI2 catalysis was discovered at a late stage
of the project and hence was not further pursued in the alkylation of other
peptides. We are currently examining the effect of this catalyst in other
systems, the work of which will be reported in a subsequent paper.
(32) Skrydstrup, T.; Beau, J.-M., unpublished results.
(33) For other samarium diiodide-promoted C-C bond forming reactions
with a putative iminium ion intermediate, see: Aurrecoechea, J. M.;
Ferna´ndez, A.; Gorgojo, J. M.; Saornil, C. Tetrahedron 1999, 55, 7345
and references therein.
(34) Molander, G. A.; Hahn, G. J. Org. Chem. 1986, 51, 1135. See also
ref 13e for a more elaborate account on the reduction of R-substituted
carbonyl compounds.
The potential for the application of this approach to the
multiple alkylation of a single peptide, was examined with the
tripeptide 22 containing two pyridyl sulfide groups (entry 10).
(35) (a) Molander, G. A.; Stengel, P. J. Tetrahedron 1997, 53, 8887. (b)
Concello´n, J. M.; Pe´rez-Andre´s, J. A.; Rodr´ıguez-Solla, H. Angew. Chem.,
Int. Ed. 2000, 39, 2773.
(36) Fukuzawa, S.; Matsuzawa, H.; Yoshimitsu, S. J. Org. Chem. 2000,
65, 1702.

12418 J. Am. Chem. Soc., Vol. 122, No. 50, 2000
Ricci et al.
Table 4. Coupling of Dipeptides 2 and 13 with Other Carbonyl
Compounds
Table 5. Effect of the N-Terminal Nitrogen Protecting Group on
the Introduction of the Pyridyl Sulfide Group and the Subsequent
SmI2-Promoted Coupling
pyridyl sulfide
yield (%)
alkylated dipeptide
yield (%)
entry
R
1
2
3
4
CH3CO
TolSO2
tBuO(CO)
BnO(CO)
39 (46)
40 (0)
41 (0)
43 (58)
42 (18)
44 (62)
of the readily available aldehyde 37 in 46% yield,38 although
the carbohydrate moiety imparted no sufficient diastereoselec-
tivity at either of the two newly created stereogenic centers.39
Finally, a small study was undertaken to investigate the
importance of the N-terminal nitrogen-protecting group on the
coupling yields (Table 5). As anticipated, the N-acetyl group
had little influence on either the introduction of mercaptopyri-
dine or the C-alkylation yield with cyclohexanone (entry 1).
However, in attempts to brominate the dipeptide R-Gly-Phe-
OMe possessing either the tosyl, tert-butyloxycarbonyl, or the
benzyloxycarbonyl group, extensive decomposition was ob-
served, affording either none or low yields of the corresponding
sulfide upon subjection with 2-mercaptopyridine (entries 2-4).
This was surprising in particular with the example possessing
a Boc group, considering the similar successful brominations
of Boc-protected glycine esters.40,41 Only in the latter case (entry
4) could we isolate the pyridyl sulfide 42, in a disappointing
18% yield. Its subsequent SmI2-promoted coupling to cyclo-
hexanone proved none the less satisfactory, affording the
branched peptide 44 in 62%. Although problems with the
introduction of the aryl sulfide group were encountered, the latter
result does suggest that alkyloxycarbonyl groups (including the
Boc group, see Table 2, entry 12) can serve as nitrogen-
protecting groups in the divalent lanthanide-promoted alkylation
step. In any case, these final results including the low yields
previously encountered in the bromination of other peptides (see
Table 3) only stress the point that an alternative and more
effective method for the introduction of the pyridyl sulfide group
is required.
a Ratio refers to the diastereoselectivity observed.
Whereas, the preparation of 22 from Bz-Gly-Val-Gly-OMe was
unexceptional, we were delighted to find that its coupling with
cyclohexanone led to the isolation of the dialkylated peptide
32 as a mixture of four stereoisomers in an unoptimized yield.
Other carbonyl substrates were tested with the dipeptides 2
and 13 as shown in Table 4. The simple alkyl aldehydes worked
well, leading to a mixture of stereoisomers (entries 2 and 3), as
detected by 1H NMR spectroscopy. On the other hand, benzal-
dehyde led to a poor coupling yield most likely due to the
competitive reduction of this compound to the corresponding
ketyl (entry 4). This was confirmed by the isolation of the
pinacol coupling product, hydrobenzoin, in addition to the
dipeptide 1 as the major compounds from this reaction.37
Entry 5 represents an interesting example of a potentially new
route for the direct introduction of a carbohydrate onto a peptide
backbone. The C-glycopeptide 38 was prepared by the coupling
Conclusions
We have demonstrated in this work that reductive samariation
can be a viable approach for the introduction of carbinol side-
chains onto small peptides containing pyridyl sulfide-modified
glycine residues. These Reformatsky-type couplings are quite
efficient, taking into consideration that the peptide substrates
tested contain up to four amide protons. In one example, we
have also shown that multiple alkylation may be possible. It is
not exactly clear why these coupling reactions lead to poor
diastereoselectivities, but none the less this characteristic allows
one to prepare peptides containing both an unnatural D- or
L-amino acid unit, which is a significant feature for the
application of this approach to the generation of peptide libraries,
although random in these cases. If this low selectivity is the
(37) A similar result was observed in the attempted coupling of either a
mannosyl or a glucosylpyridyl sulfone with benzaldehyde (refs 14a, e).
(38) Hosomi, A.; Sakata, Y.; Sakurai, H. Carbohydrate Res. 1987, 171,
125.
(39) For some recent syntheses of C-glycosyl amino acids and peptides,
see (a) Burkhart, F.; Hoffmann, M.; Kessler, H. Angew. Chem., Int. Ed.
Engl. 1997, 36, 1191. (b) Lay, L.; Meldal, M.; Nicotra, F.; Panza, L.; Russo,
G. Chem. Commun. 1997, 1469. (c) Debenham, S. D.; Debenham, J. S.;
Burk, M. J.; Toone, E. J. J. Am. Chem. Soc. 1997, 119, 9897. (d) Dondoni,
A.; Marra, A.; Massi, A. Chem. Commun. 1998, 1741. (e) Dondoni, A.;
Marra, A.; Massi, A. Tetrahedron Lett. 1998, 39, 6601. (f) Dondoni, A.;
Marra, A.; Massi, A. Tetrahedron 1998, 54, 2827. (g) Ben, R. N.; Orellana,
A.; Arya, P. J. Org. Chem. 1998, 63, 4817. (h) Arya, P.; Ben, R. N.; Qin,
H. Tetrahedron Lett. 1998, 39, 6131. (i) Urban, D.; Skrydstrup, T.; Beau,
J.-M. Chem. Commun. 1998, 955. (j) Pearce, A. J.; Ramaya, S.; Thorn, S.
N.; Bloomberg, G. B.; Walter, D. S.; Gallagher, T. J. Org. Chem. 1999,
64, 5453.
(40) See ref 11 and references therein.
(41) The exact reasons for the poor bromination yields with the Boc-
protecting group are not fully understood, as attempted bromination with
an acid scavenger was likewise unsuccessful. However, in the case of the
tosyl and Cbz groups, possibly competitive bromination of these protecting
groups could be responsible for the low yields observed.

Potential Route to Peptide Libraries
J. Am. Chem. Soc., Vol. 122, No. 50, 2000 12419
added. The reaction mixture was stirred for 3 h at 0 °C and then diluted
with CH2Cl2 and washed with water (2×) and brine. The organic phase
was dried (MgSO4) and evaporated to dryness, whereafter the residue
was purified by flash chromatography (pentane/EtOAc, 7/3) to afford
a diastereomeric mixture (1.6:1.0 as determined by 1H NMR analysis)
of the pyridyl sulfide 2 (462 mg, 87%) as a light yellow foam. IR
(KBr) 3310, 1745, 1660 cm-1; 1H NMR δ ) 9.26 (1H, d, J ) 8.0 Hz),
9.15 (1H, d, J ) 8.2 Hz), 8.24 (1H, m), 8.06 (1H, m), 7.74-7.90 (6H,
m), 6.88-7.64 (22H, m), 6.42 (1H, d, J ) 7.6 Hz), 6.38 (1H, d, J )
7.0 Hz), 4.86-5.02 (2H, m), 3.75 (3H, s), 3.57 (3H, s), 2.99-3.20
(4H, m); 13C NMR δ ) 171.3, 171.2, 168.9, 168.6, 165.6, 157.5, 149.1,
136.8, 136.0, 135.4, 133.3, 132.1, 129.4, 129.1, 128.7, 128.0, 127.3,
127.2, 126.8, 123.2, 123.0, 120.5, 54.1, 53.9, 52.4, 52.2, 38.2, 38.0;
HR-MS (ES) calcd for C24H23N3O4SNa (M + Na): 472.1307, found:
472.1307.
result of an unbiased facial selectivity of a single enolate
structure present in solution, or from an enolate mixture, the
addition of asymmetric ligands to the lanthanide ion could
influence both the enolate structure and the facial selectivity in
the alkylation. In this way, access to either of the two
diastereomers may be possible by the choice of the appropriate
ligand. A number of enantiomerically pure ligands will be
examined in the future although little has been explored in
the realm of asymmetric synthesis with SmI2-promoted
reactions.27b,36,42
In this study, peptides possessing only lipophilic side chains
have so far been studied, where future work will have to
examine the compatibility of other amino acid side chains in
the SmI2-promoted C-alkylation step. Although, the functional
group selectivity of this mild one-reducing suggests that this
method will be adabtable even in the presence of a wide range
of side chains, more concern is put forth regarding the ability
to perform the initial bromination step, where various protected
amino acids may simply not be compatible to such conditions
(e.g., cysteine, methionine). Hence, work is currently underway
to identify a new electrophore, which may be incorporated into
an amino acid building block, such that no modification of the
peptide prior to the alkylation step would be required. The
successful identification of this group would likewise allow for
the multiple alkylation of peptides containing more than one
glycine residue, or for the selective alkylation at one or several
glycine sites in the presence of others not functionalized. This
method could allow for a rapid approach to a new series of
peptide libraries based on serine/threonine derivatives for
combinatorial chemistry.
General Procedure for the SmI2-Promoted Coupling with Car-
bonyl Compound. A 0.1 M solution of SmI2 in THF (5.5 mL, 0.55
mmol) was added to a stirred solution of the pyridyl sulfide (0.20 mmol)
and cyclohexanone (62 µL, 0.60 mmol) in THF (0.5 mL) at 20 °C.
After stirring for 1 h, saturated aqueous NH4Cl (2 mL) was added to
the reaction mixture, which was then extracted with CH2Cl2 (2 × 10
mL). The combined organic phases were washed with water, filtered
through Celite, dried with MgSO4, and evaporated to dryness. Flash
chromatography afforded the alkylated peptide 7, 23-35, 38, 43, 44.
Yields are described in Tables 3, 4, and 5. Diastereomeric ratios were
1
determined by H NMR analysis.
Bz-(1-hydroxycyclohexyl)Gly-Phe-OMe (7): Chromatography (pen-
tane/EtOAc, 2/1). Colorless solid (1.0:1.0 diastereomeric mixture). IR
(KBr) 3279, 1750, 1630 cm-1; 1H NMR δ ) 7.75-7.85 (4H, m), 7.38-
7.61 (6H, m), 7.00-7.33 (14H, m), 4.78-4.94 (2H, m), 4.57 (1H, d,
J ) 8.0 Hz), 4.53 (1H, d, J ) 8.4 Hz), 3.88 (1H, bs), 3.81 (1H, bs),
3.74 (3H, s), 3.67 (3H, s), 3.23 (1H, dd, J ) 5.2, 14.0 Hz), 3.16 (1H,
dd, J ) 5.2, 14.0 Hz), 3.02 (1H, dd, J ) 8.8, 14.0 Hz), 2.97 (1H, dd,
J ) 7.4, 14.0 Hz), 1.56-1.81 (20H, m); 13C NMR δ ) 171.5, 167.9,
167.7, 133.7, 132.2, 129.3, 128.8, 127.4, 127.3, 73.1, 72.9, 58.1, 53.4,
52.7, 37.9, 35.4, 33.7, 33.6, 25.7, 21.9, 21.6; HR-MS (ES) calcd for
C25H30N2O5Na (M + Na): 461.2052, found: 461.2055.
Bz-(1-Hydroxycyclohexyl)Gly-Val-OMe (23): Chromatography
(pentane/EtOAc, 3/1). Colorless solid (1.0:1.0 diastereomeric). IR (KBr)
3292, 1748, 1634 cm-1; 1H NMR (300 MHz) δ ) 7.77-7.84 (4H, m),
7.40-7.56 (6H, m), 7.05-7.21 (3H, m), 6.98 (1H, d, J ) 5.8 Hz),
4.59 (1H, d, J ) 5.6 Hz), 4.56 (1H, d, J ) 5.4 Hz), 4.50 (1H, dd, J )
5.9, 3.3 Hz), 4.46 (1H, dd, J ) 5.8, 3.2 Hz), 3.76 (3H, s), 3.64 (3H, s),
2.21 (2H, m), 1.19-1.86 (20H, m), 0.97 (3H, d, J ) 4.9 Hz), 0.94
(3H, d, J ) 4.9 Hz), 0.91 (3H, d, J ) 4.4 Hz), 0.90 (3H, d, J ) 4.4
Hz), 2 OH signals missing; 13C NMR (75 MHz) δ ) 172.0, 171.5,
168.1, 168.0, 133.7, 132.1, 128.8, 127.3, 127.2, 73.0, 72.8, 58.3, 57.8,
57.6, 57.4, 52.4, 52.3, 35.7, 35.4, 33.9, 33.6, 30.9, 25.7, 21.9, 21.8,
21.6, 19.2, 19.1, 18.0, 17.7; HR-MS (ES) calcd for C21H30N2O5Na (M
+ Na) 413.2052, found: 413.2043.
Experimental Section
General Methods. The SmI2-promoted coupling reactions were
performed under Ar, while all other reactions were carried out under
N2. THF was dried and freshly distilled over sodium/benzophenone.
Dichloromethane was freshly distilled over P2O5. DMF was distilled
and stored over molecular sieves (4 Å). 1H and 13C NMR spectra were
recorded on a Varian Gemini 2000 (200 and 50 MHz, respectively)
using CDCl3 as solvent unless otherwise stated and TMS as an internal
1
standard for H NMR. High-resolution mass spectral analyses were
performed with a Micromass LC-TOF instrument. Analytical thin-layer
chromatography was performed on Merck silica gel 60 F254 plates.
Compounds were visualized using a cerium sulfate-ammonium mo-
lybdate solution, blue stain solution, or p-anisaldehyde solution followed
by heating. Flash column chromatography was performed using Fluka
silica gel 60 (0.04-0.063 mm). All peptides were synthesized using
standard peptide coupling procedures with 1-[3-(dimethylamino)propyl]-
3-ethylcarbodiimide (EDC) and 1-hydroxy-1H-benzotriazole (HOBt).
For details on the oligopeptide syntheses see the standard procedure.43
Samarium diiodide was prepared according to the literature.44
Bz-(1-Hydroxycyclohexyl)Gly-Asp-OMe (24): Chromatography
(pentane/EtOAc, 2/3). Colorless solid (1.0:1.0 diastereomeric mixture).
1
IR (KBr) 3278, 1743, 1633 cm-1; H NMR δ ) 7.78-7.87 (4H, m),
7.73 (1H, d, J ) 8.0 Hz), 7.61 (1H, d, J ) 8.4 Hz), 7.36-7.56 (6H,
m), 7.15-7.28 (2H, m), 4.90 (1H, ddd, J ) 4.8, 5.0, 8.6 Hz), 4.84
(1H, ddd, J ) 5.0, 5.2, 8.0 Hz), 4.67 (1H, d, J ) 8.6 Hz), 4.66 (1H, d,
J ) 8.8 Hz), 3.74 (3H, s), 3.66 (3H, s), 3.65 (3H, s), 3.62 (3H, s), 3.50
(2H, bs), 2.74-3.07 (4H, m), 1.12-1.85 (20H, m); 13C NMR δ ) 171.4,
171.3, 171.1, 171.0, 170.8, 170.6, 167.9, 167.7, 133.7, 133.5, 132.0,
128.7, 127.3, 73.3, 73.0, 58.7, 58.4, 53.6, 52.2, 48.9, 48.6, 36.0, 35.9,
35.4, 35.3, 33.7, 33.4, 25.6, 21.8, 21.6; HR-MS (ES) calcd for
C21H28N2O7Na (M + Na): 443.1794, found: 443.1767.
Bz-(1-Hydroxycyclohexyl)Gly-Ala-OMe (25): Chromatography
(pentane/EtOAc, 1/1). Pale yellow solid (1.1:1.0 diastereomeric mix-
ture). IR (KBr) 3290, 1750, 1635 cm-1; 1H NMR δ ) 7.78-7.88 (4H,
m), 7.39-7.58 (6H, m), 7.06-7.30 (4H, m), 4.45-4.67 (4H, m), 4.04
(1H, bs), 3.90 (1H, bs), 3.75 (3H, s), 3.65 (3H, s), 1.19-1.86 (26H,
m); 13C NMR δ ) 172.9, 172.6, 171.4, 171.3, 168.0, 167.8, 133.7,
132.1, 128.8, 128.5, 127.3, 73.2, 72.9, 58.3, 57.9, 52.7, 48.3, 35.5, 35.4,
33.8, 33.5, 25.7, 21.8, 21.6, 18.0, 17.8; HR-MS (ES) calcd for
C19H26N2O5Na (M + Na): 385.1739, found: 385.1736.
General Procedure for the Preparation of the Pyridyl Sulfide
Derivatives. Bz-((2-Pyr)S)Gly-Phe-OMe (2). A solution of N-bromo-
succinimide (226 mg, 1.29 mmol) and the dipeptide 1 (400.0 mg, 1.18
mmol) in CH2Cl2 (14 mL) was irradiated with a 150-W lamp for 3 h.
The heat from the lamp was sufficient to maintain refluxing. The
solution was then cooled to 0 °C, after which 2-mercaptopyridine (198.0
mg, 1.77 mmol) and diisopropylethylamine (308 µL, 1.77 mmol) were
(42) (a) Fukuzawa, S.; Seki, K.; Tatsuzawa, M.; Mutoh, K. J. Am. Chem.
Soc. 1997, 119, 1482. (b) Mikami, K.; Yamaoka, M. Tetrahedron Lett. 1998,
39, 4501. (c) Mikami, K.; Yamaoka, M.; Yoshida, A. Synlett 1998, 607.
(d) Taniguchi, N.; Kaneta, N.; Uemura, M. Tetrahedron 1998, 54, 12775.
(e) Kim, S. M.; Byun, I. S.; Kim, Y. H. Angew. Chem., Int. Ed. 2000, 39,
728. (f) Fang, J.-M.; Chen, M.-Y.; Shiue, J.-S.; Lu, L.; Hsu, J.-L.
Tetrahedron Lett. 2000, 41, 4633. (g) Riber, D.; Hazell, R.; Skrydstrup, T.
J. Org. Chem. 2000, 65, 5382.
(43) Bodanszky, M. Practice of Peptide Synthesis; Spring Verlag: Berlin,
1984.
(44) Namy, J. L.; Girard, P.; Kagan, H. B. J. Am. Chem. Soc. 1980,
102, 2693.

12420 J. Am. Chem. Soc., Vol. 122, No. 50, 2000
Ricci et al.
Bz-(1-Hydroxycyclohexyl)Gly-Pro-OMe (26): Chromatography
23.0, 22.7, 22.3, 21.9, 21.3; HR-MS (ES) calcd for C37H52N4O7Na
(M + Na): 687.3734, found: 687.3736.
(pentane/EtOAc, 3/2). Colorless solid (1.8:1.0 diastereomeric mixture).
1
IR (KBr) 3431, 1752, 1630 cm-1; H NMR δ ) 7.77-7.85 (4H, m),
A small fraction of one of the two diastereomers could be isolated
1
7.38-7.57 (6H, m), 7.12 (1H, d, J ) 9.0 Hz), 7.04 (1H, d, J ) 9.5
Hz), 4.93 (1H, d, J ) 9.5 Hz), 4.86 (1H, d, J ) 9.5 Hz), 4.46-4.54
(1H, m), 3.68-3.91 (1H, m), 4.14 (2H, s), 3.75 (3H, s), 3.67 (3H, s),
1.90-2.36 (8H, m), 1.20-1.90 (20H, m); 13C NMR δ ) 172.2, 171.33,
171.28, 167.4, 167.3, 133.4, 133.5, 131.9, 128.6, 127.22, 127.17, 73.3,
73.0, 59.0, 58.8, 55.4, 55.2, 52.4, 47.9, 34.9, 33.9, 33.8, 29.3, 25.7,
24.9, 24.6, 21.7, 21.6, 21.5, 21.4; HR-MS (ES) calcd for C21H28N2O5-
Na (M + Na): 411.1896, found 411.1903.
from the same chromatographic separation. H NMR δ ) 7.75-7.85
(2H, m), 7.69 (1H, m), 7.00-7.63 (10H, m), 6.94 (1H, d, J ) 8.2 Hz),
4.88 (1H, m), 4.65 (1H, m), 4.21-4.42 (2H, m), 3.64 (3H, s), 3.10
(1H, dd, J ) 6.0, 14.0 Hz), 2.94 (1H, dd, J ) 8.0, 13.6 Hz), 1.04-
1.86 (16H, m), 0.68-1.03 (12H, m).
Bz-(1-Hydroxycyclohexyl)Gly-Val-(1-hydroxycyclohexyl)Gly-
OMe (32). A 0.1 M solution of SmI2 in THF (2.6 mL, 0.26 mmol)
was added to a stirred solution of the pyridyl sulfide 22 (24.7 mg, 0.043
mmol) and cyclohexanone (27 µL, 0.26 mmol) in THF (0.5 mL) at 20
°C. After workup according to the general procedure for the preparation
of 7, flash chromatography (pentane/EtOAc 7/3 to 1/1) gave the
tripeptide 32 (8.9 mg, 38%) as a colorless solid in a 3.4:1.5:1.4:1.0
Bz-Leu-(1-hydroxycyclohexyl)Gly-OMe (27): Chromatography
(pentane/EtOAc, 3/2). Colorless solid (1.1:1.0 diastereomeric mixture).
1
IR (KBr) 3332, 1745, 1637 cm-1; H NMR δ ) 7.72-7.82 (4H, m),
7.33-7.54 (6H, m), 7.15-7.29 (2H, m), 6.85-6.96 (2H, m), 4.73-
4.88 (2H, m), 4.60 (1H, d, J ) 9.0 Hz), 4.57 (1H, d, J ) 9.0 Hz), 3.75
(3H, s), 3.69 (3H, s), 2.98 (1H, bs), 2.79 (1H, bs), 1.12-1.82 (26H,
m), 0.91-1.02 (12H, m); 13C NMR δ ) 172.7, 172.4, 171.9, 171.6,
167.7, 167.5, 134.0, 131.9, 128.7, 127.2, 73.1, 72.8, 59.4, 52.4, 52.3,
42.7, 41.4, 35.3, 35.1, 34.5, 25.5, 25.1, 25.0, 23.2, 23.0, 22.4, 22.3,
21.7; HR-MS (ES) calcd for C22H32N2O5Na (M + Na): 427.2209,
found: 427.2209.
1
diastereomeric mixture as determined by H NMR analysis. IR (KBr)
3200, 1720, 1661 cm-1; 1H NMR (300 MHz at 45 °C) δ ) 7.78-7.87
(8H, m), 7.40-7.59 (12H, m), 6.84-7.32 (12H, m), 4.30-4.70 (12H,
m), 3.76 (3H, s), 3.70 (3H, s), 3.67 (3H, s), 3.62 (3H, s), 2.18-2.60
(4H, m), 1.09-1.96 (80H, m), 0.83-1.03 (24H, m), 8 OH signals
missing; 13C NMR (75 MHz) δ ) 172.1, 172.0, 171.9, 171.8, 171.7,
171.6, 171.5, 171.4, 171.1, 170.9, 170.7, 167.8, 167.7, 133.5, 132.2,
132.1, 131.9, 131.8, 128.7, 128.5, 127.3, 127.2, 127.1, 73.5, 73.1, 72.8,
72.6, 72.4, 72.1, 72.0, 71.8, 59.5, 59.1, 59.0, 58.9, 58.8, 58.7, 58.5,
58.4, 58.2, 52.1, 51.9, 35.6, 35.4, 35.2, 35.0, 34.9, 34.8, 34.6, 34.4,
34.2, 34.1, 34.0, 30.2, 29.9, 29.4, 29.3, 25.4, 25.3, 25.2, 21.8, 21.7,
21.6, 21.5, 21.4, 21.3, 19.7, 19.4, 19.3, 19.2, 19.1, 17.6, 17.3, 17.1,
17.0, 16.8; HR-MS (ES) calcd for C29H43N3O7Na (M + Na): 568.2998,
found: 568.2993.
Bz-Leu-(1-hydroxycyclohexyl)Gly-Val-OMe (28): Chromatogra-
phy (pentane/EtOAc, 3/2). Colorless solid (1.2:1.0 diastereomeric
1
mixture). IR (KBr) 3334, 1745, 1640 cm-1; H NMR δ ) 7.75-7.84
(4H, m), 7.33-7.54 (8H, m), 7.02-7.22 (4H, m), 4.72-4.88 (2H, m),
4.48-4.62 (2H, m), 4.36-4.47 (2H, m), 4.05 (1H, bs), 3.80 (1H, bs),
3.74 (3H, s), 3.54 (3H, s), 0.78-2.34 (52H, m); 13C NMR δ ) 173.3,
173.1, 172.5, 172.0, 171.6, 167.4, 167.3, 133.9, 131.8, 128.5, 127.4,
72.8, 72.5, 58.5, 57.9, 57.5, 57.1, 53.6, 52.4, 52.2, 52.1, 41.8, 41.7,
35.6, 35.2, 34.3, 33.7, 33.1, 32.1, 31.1, 30.6, 29.8, 25.7, 25.5, 25.2,
25.0, 23.0, 22.8, 22.5, 22.3, 21.8, 21.6, 21.4, 19.1, 17.9, 17.8.
Bz-(1-Hydroxycyclopentyl)Gly-Phe-OMe (33): Chromatography
(pentane/EtOAc, 7/3 to 3/2). Colorless solid (1.1:1.0 diastereomeric
1
mixture). IR (KBr) 3280, 1749, 1631 cm-1; H NMR δ ) 7.73-7.83
(4H, m), 7.09-7.61 (20H, m), 4.75-4.90 (2H, m), 4.50 (1H, d, J )
8.4 Hz), 4.43 (1H, d, J ) 8.2 Hz), 4.05 (1H, bs), 3.99 (1H, bs), 3.74
(3H, s), 3.64 (3H, s), 3.20 (1H, dd, J ) 5.2, 14.0 Hz), 3.16 (1H, dd, J
) 5.0, 14.0 Hz), 3.02 (1H, dd, J ) 8.4, 14.0 Hz), 2.97 (1H, dd, J )
8.0, 14.0 Hz), 1.44-1.94 (16H, m); 13C NMR δ ) 171.7, 171.6, 168.1,
167.7, 136.0, 135.7, 133.5, 132.2, 129.2, 128.7, 128.6, 127.3, 127.1,
82.8, 82.6, 58.9, 58.6, 53.4, 52.6, 38.8, 38.7, 37.8, 37.3, 24.5, 24.4,
24.0, 23.8; HR-MS (ES) calcd for C24H28N2O5Na (M + Na): 447.1896,
found: 447.1894.
Bz-(1-Hydroxy(cyclohexyl)methyl)Gly-Phe-OMe (34): Chroma-
tography (pentane/EtOAc, 2/1). Colorless foam (2.5:1.8:1.1:1.0 dia-
stereomeric mixture). IR (KBr) 3280, 1745, 1635 cm-1; 1H NMR δ )
7.72-7.84 (8H, m), 6.94-7.61 (40H, m), 4.62-4.97 (8H, m), 4.01-
4.18 (2H, m), 3.80-3.94 (2H, m), 3.74 (3H, s), 3.73 (3H, s), 3.66 (3H,
s), 3.64 (3H, s), 2.88-3.30 (8H, m), 0.78-2.15 (44H, m), 4 OH signals
missing; 13C NMR δ ) 171.6, 167.9, 167.5, 136.0, 135.7, 133.4, 132.1,
129.2, 128.7, 128.6, 128.5, 127.4, 127.2, 127.1, 74.6, 53.7, 53.4, 52.6,
41.0, 37.7, 29.7, 29.4, 29.1, 28.7, 26.3, 26.0, 25.8; HR-MS (ES) calcd
for C26H32N2O5Na (M + Na): 475.2209, found: 475.2200.
Bz-Leu-(1-hydroxycyclohexyl)Gly-Phe-OMe (29): Chromatogra-
phy (pentane/EtOAc, 1/1). Colorless solid (2.0:1.0 diastereomeric
1
mixture). IR (KBr) 3292, 1747, 1637 cm-1; H NMR δ ) 7.75-7.91
(5H, m), 7.05-7.65 (21H, m), 4.75-4.98 (4H, m), 4.63 (1H, d, J )
8.8 Hz), 4.52 (1H, d, J ) 8.8 Hz), 4.01 (1H, bs), 3.65 (3H, s), 3.60
(1H, bs), 3.51 (3H, s), 2.87-3.23 (4H, m), 1.10-1.80 (26H, m), 0.84-
1.02 (12H, m); 13C NMR δ ) 172.9, 172.1, 171.7, 171.1, 170.9, 167.5,
136.2, 135.9, 134.0, 131.7, 129.2, 128.7, 128.6, 128.5, 127.4, 127.2,
127.1, 73.0, 72.7, 58.9, 53.4, 52.4, 52.1, 41.9, 41.4, 38.1, 34.9, 33.5,
33.1, 25.5, 25.0, 23.2, 23.1, 22.4, 22.2, 21.4; HR-MS (ES) calcd for
C31H41N3O6Na (M + Na): 574.2893, found: 574.2869.
Bz-(1-Hydroxycyclohexyl)Gly-Leu-Phe-OMe (30): Chromatogra-
phy (pentane/EtOAc, 3/2). Colorless solid (1.0:1.0 diastereomeric
1
mixture). IR (KBr) 3316, 1743, 1624 cm-1; H NMR δ ) 7.74-7.82
(4H, m), 7.37-7.54 (6H, m), 7.07-7.36 (12H, m), 6.96 (1H, d, J )
7.6 Hz), 6.90 (1H, d, J ) 8.8 Hz), 6.89 (1H, d, J ) 8.4 Hz), 6.66 (1H,
d, J ) 7.8 Hz), 4.85 (1H, dt, J ) 6.0, 8.0 Hz), 4.76 (1H, dt, J ) 5.8,
8.0 Hz), 4.58 (1H, d, J ) 8.8 Hz), 4.33-4.49 (2H, m), 4.40 (1H, d,
J ) 7.8 Hz), 3.97 (1H, s), 3.94 (1H, s), 3.70 (3H, s), 3.49 (3H, s), 3.12
(2H, d, J ) 6.0 Hz), 3.10 (1H, dd, J ) 5.8, 14.0 Hz), 2.99 (1H, dd,
J ) 8.0, 14.0 Hz), 1.16-1.82 (26H, m), 0.90 (3H, d, J ) 6.2 Hz), 0.88
(3H, d, J ) 6.2 Hz), 0.83 (3H, d, J ) 6.2 Hz), 0.79 (3H, d, J ) 5.8
Hz); 13C NMR δ ) 171.7 (2), 171.6, 171.4 (2), 171.1, 168.5, 167.9,
136.3, 135.8, 133.6, 133.1, 132.2, 132.1, 129.3, 128.7, 128.6 (2), 128.4,
127.3, 127.2, 127.1, 126.9, 73.2, 72.4, 59.0, 58.7, 53.3 (2), 52.4, 52.2
(3), 40.6, 40.3, 37.9, 37.7, 35.4, 35.3, 34.0, 32.9, 29.7 (2), 25.5 (2),
24.9, 24.7, 23.0, 22.9, 21.8, 21.5 (2), 21.4, 21.3; HR-MS (ES) calcd
for C31H41N3O6Na (M + Na): 574.2893, found 574.2863.
Bz-(1-Hydroxyoctyl)Gly-Phe-OMe (35): Chromatography (pentane/
EtOAc, 2/1). Colorless solid (2.2:2.1:1.0 diastereomeric mixture). IR
(KBr) 3289, 1742, 1635 cm-1; 1H NMR δ ) 7.73-7.85 (6H, m), 7.02-
7.64 (30H, m), 4.76-4.92 (3H, m), 4.56-4.68 (3H, m), 4.02-4.28
(2H, m), 3.76 (1H, m), 3.74 (3H, s), 3.72 (3H, s), 3.67 (3H, s), 2.91-
3.26 (6H, m), 1.06-1.72 (33H, m), 0.78-0.94 (12H, m), 3 OH signals
missing; 13C NMR δ ) 171.7, 171.4, 170.9, 170.5, 168.1, 167.7, 136.0,
135.7, 133.4, 132.1, 129.2, 128.7, 127.4, 73.7, 73.6, 57.2, 56.6, 53.4,
52.6, 37.7, 33.9, 33.6, 31.9, 29.5, 29.3, 26.0, 22.7, 14.2; HR-MS (ES)
calcd for C27H36N2O5Na (M + Na): 491.2522, found: 491.2523.
Bz-(2-(2,3,4,6-Tetra-O-benzylglucopyranosyl)-1-hydroxy-ethyl)-
Gly-Val-OMe (38): Chromatography (pentane/EtOAc, 7/3). Thick
colorless syrup. Although 4 signals for the OMe group were observed,
their overlap with other signals made it impossible to determine the
diasteromeric ratio from the proton integrations. IR (film) 3301, 1742,
Bz-Leu-(1-hydroxycyclohexyl)Gly-Leu-Phe-OMe (31): Chroma-
tography (pentane/EtOAc, 7/3 to 1/1). Colorless solid (1.2:1.0 diastereo-
1
meric mixture). IR (KBr) 3293, 1748, 1635 cm-1; H NMR δ ) 8.11
(1H, d, J ) 8.4 Hz), 7.75-7.92 (7H, m), 7.01-7.53 (20H, m), 4.32-
5.10 (8H, m), 3.65 (3H, s), 3.64 (3H, s), 2.88-3.16 (4H, m), 1.07-
1.88 (32H, m), 0.73-1.05 (24H, m), 2 OH signals missing; 13C NMR
δ ) 173.8, 173.1, 172.1, 171.9, 171.7, 171.6, 171.0, 168.0, 167.5, 136.2,
135.9, 134.1, 132.1, 131.6, 129.3, 129.2, 128.7, 128.5, 128.4, 127.6,
127.5, 127.2, 127.0, 73.0, 72.4, 59.1, 58.6, 53.8, 52.9, 52.5, 52.3, 51.7,
41.5, 41.3, 40.5, 37.9, 35.4, 35.1, 33.9, 33.0, 29.8, 25.5, 25.1, 24.9,
1
1644 cm-1; H NMR (300 MHz) δ ) 7.76-7.91 (8H, m), 7.02-7.70
(100H, m), 4.00-4.94 (48H, m), 3.30-3.90 (24H, m), 3.70 (3H, s),
3.67 (3H, s), 3.65 (3H, s), 3.62 (3H, s), 1.80-2.34 (12H, m), 0.78-
1.03 (24H, m), 4 OH signals missing; 13C NMR (75 MHz) δ ) 171.8,
171.6, 171.5, 170.9, 170.8, 170.7, 168.0, 167.8, 167.7, 167.6, 138.5,

Potential Route to Peptide Libraries
J. Am. Chem. Soc., Vol. 122, No. 50, 2000 12421
138.4, 138.3, 138.2, 138.0, 137.9, 137.8, 137.7, 137.6, 137.5, 133.6,
133.5, 133.4, 131.9, 131.8, 131.7, 128.6, 128.5, 128.3, 128.2, 127.9,
127.8, 127.7, 127.6, 127.5, 127.1, 127.0, 81.9, 81.8, 81.6, 79.5, 79.2,
79.1, 78.4, 77.9, 77.7, 77.5, 75.6, 75.5, 75.3, 75.2, 75.0, 74.6, 74.5,
73.4, 73.3, 73.2, 73.1, 73.0, 72.9, 72.2, 71.9, 71.8, 71.6, 71.5, 71.3,
70.6, 70.1, 69.2, 69.1, 68.9, 68.6, 57.5, 57.4, 57.3, 56.6, 56.2, 56.0,
52.0, 30.7, 30.6, 30.2, 29.6, 29.4, 29.2, 29.1, 19.1, 19.0, 18.9, 17.7,
17.5, 17.4; HR-MS (ES) calcd for C51H58N2O10Na (M + Na): 881.3989,
found: 881.3996.
4.96 (2H, m), 4.02 (1H, d, J ) 8.8 Hz), 4.01 (1H, d, J ) 8.8 Hz), 3.72
(3H, s), 3.70 (3H, s), 3.48 (1H, bs), 3.41 (1H, bs), 2.93-3.26 (4H, m),
1.10-1.68 (20H, m); 13C NMR δ ) 171.7, 171.3, 156.7, 136.2, 135.9,
129.2, 128.7, 128.6, 128.2, 128.0, 127.3, 72.9, 72.7, 67.2, 60.2, 59.9,
53.2, 52.6, 37.9, 37.7, 35.1, 33.2, 33.0, 25.5, 21.6, 21.5, 21.4; HR-MS
(ES) calcd for C26H32N2O6Na (M + Na): 491.2158, found: 491.2138.
SmI2-Promoted Coupling of Dipeptide 2 with Cyclohexanone in
the Presence of 1% NiI2. A 0.01 M solution of NiI2 in THF (0.6 mL,
6.0 µmol) was added to a 0.1 M solution of SmI2 in THF (6.0 mL,
0.60 mmol) and then allowed to stir for 15 min, after which the solution
was cooled to 0 °C. This solution was then added dropwise to a stirred
solution of the pyridyl sulfide 2 (89.8 mg, 0.20 mmol) and cyclohex-
anone (62 µL, 0.60 mmol) in THF (0.5 mL) at -78 °C. TLC analysis
showed that the reaction was complete after 10 min. The reaction
mixture was then warmed to 20 °C and saturated aqueous NH4Cl (1 mL)
was added. Workup according to the general procedure and flash
chromatography gave the dipeptide 7 (74.9 mg, 85%) as a colorless
solid in a 1:1 diastereomeric mixture as determined by 1H NMR
analysis.
Ac-(1-Hydroxycyclohexyl)Gly-Phe-OMe (43): Chromatography
(pentane/EtOAc, 1/2). Colorless solid (2.7:1.0 diastereomeric mixture).
IR (KBr) 3328, 1741, 1645 cm-1; 1H NMR δ ) 7.08-7.35 (11H, m),
6.95 (1H, d, J ) 8.2 Hz), 6.41-6.52 (2H, m), 4.76-4.91 (2H, m),
4.32 (1H, d, J ) 8.8 Hz), 4.31 (1H, d, J ) 8.8 Hz), 3.78 (1H, bs), 3.73
(3H, s), 3.72 (3H, s), 3.21 (1H, dd, J ) 5.2, 14.0 Hz), 3.17 (1H, dd,
J ) 5.6, 14.0 Hz), 3.02 (1H, dd, J ) 7.8, 14.0 Hz), 2.98 (1H, dd, J )
8.2, 14.0 Hz), 2.01 (3H, s), 1.98 (3H, s), 1.12-1.70 (20H, m), 1 OH
signal missing; 13C NMR δ ) 171.5, 171.4, 170.7, 170.5, 136.0, 135.7,
129.3, 129.2, 128.7, 127.2, 72.7, 72.6, 57.8, 57.4, 53.3, 52.6, 38.0, 37.8,
35.3, 33.5, 25.6, 23.1, 21.8, 21.7, 21.5; HR-MS (ES) calcd for
C20H28N2O5Na (M + Na): 399.1896, found: 399.1888.
Acknowledgment. We are indebted to the Danish National
Science Foundation (THOR Program), the University of Aarhus,
the Carlsberg Foundation, the Leo Pharmaceutical Foundation,
and the Lundbeck Foundation for generous financial support.
A small fraction of one of the two diastereomers could be isolated
1
from the same chromatographic separation. H NMR δ ) 7.07-7.38
(5H, m), 6.79 (1H, bd), 6.36 (1H, bd), 4.84 (1H, m), 4.32 (1H, d, J )
8.8 Hz), 3.73 (3H, s), 3.17 (1H, dd, J ) 5.6, 14.0 Hz), 3.02 (1H, dd,
J ) 7.8, 14.0 Hz), 1.98 (3H, s), 1.12-1.70 (10H, m), OH signal missing;
13C NMR δ ) 171.5, 170.7, 170.5, 135.7, 129.2, 128.7, 127.2, 72.7,
57.8, 53.3, 52.6, 37.8, 35.3, 33.5, 25.6, 23.1, 21.7, 21.5.
Supporting Information Available: Full experimental detail
and characterization for compounds 13-22, 39, and 42 (PDF).
This material is available free of charge via the Internet at
http://pubs.acs.org.
Cbz-(1-Hydroxycyclohexyl)Gly-Phe-OMe (44): Chromatography
(pentane/EtOAc, 3/1). Colorless solid (1.2:1.0 diastereomeric mixture).
IR (KBr) 3339, 1734, 1693 cm-1; 1H NMR δ ) 7.04-7.41 (20H, m),
6.65-6.87 (2H, m), 5.66-5.79 (2H, m), 5.02-5.17 (4H, m), 4.79-
JA0024800