fluorophosphate was used as the model system. Displacement
of chlorine by a variety of nucleophiles, followed by
photolytic removal of the ruthenium complex (Scheme 1),
Scheme 2
Scheme 1
piperidine on 12 gave equal amounts of the 3- and 4-sub-
stituted aniline derivatives (13a + 13b). However, only
monosubstitution was observed, even with an excess of
piperidine.
led to the substituted phenylalanines shown in Table 1. In
this reaction the sodium anions of phenol, benzenethiol,
dimethyl malonate, methanol, and excess neutral piperidine
readily displaced chloride in typical fashion. Additionally,
the sodium thiolate of Boc-Cys-OMe and weakly nucleo-
philic sodium anions of succinimide and hydantoin displaced
chloride from 3, although no attempts were made to optimize
these three reactions.
Displacement of both chlorines from 12 could be ac-
complished with sodium benzenethiolate. Additionally, the
cyclopentadienyl ruthenium complexes of monochloro 13a
and 13b react with sodium benzenethiolate to give the
thioether and piperidine-substituted phenylalanine. Unfor-
tunately, the present irradiation technique did not remove
ruthenium from the complex, a problem previously noted
with highly heteroatom-substituted aromatic rings.5c Further
work on decomplexation is therefore required. The ease of
displacement of chlorine in both mono- and dichlorinated
phenylalanines and the compatibility of the protected amino
acid complex with a variety of nucleophiles indicates that
libraries of substituted peptides could be prepared by use of
combinatorial chemistry.
Table 1. Intermolecular SNAr Reactions
Our results with model intermolecular ruthenium-activated
SNAr reactions led us to see if an intramolecular version
could be used to synthesize novel cyclic systems. Previously,
only phenols3,4 have been used as nucleophiles in the
ruthenium-activated intramolecular SNAr reaction. The nu-
cleophilic heteroatoms sulfur and nitrogen in cysteine,
histidine, and lysine containing tripeptides were chosen to
explore this area of chemistry and provide entry to 14-, 15-,
and 17- member ring systems, respectively.
Formation of the histidine-containing cyclic tripeptide
began with complexation of ruthenium to Boc-(4Cl)Phe-OH7
(14) followed by peptide coupling with Leu-His-OMe to give
the ruthenium-containing linear tripeptide 16 (Scheme 3).
Cyclization under high dilution conditions (c ) 5 mM)
followed by photolytic decomplexation of ruthenium3 led to
the novel heteroaryl cyclized tripeptide 17.
The first attempts to synthesize the analogous cysteine and
lysine cyclic peptides were disappointing due to low yields
obtained in the cyclization step. Our first attempt, outlined
for cysteine (Scheme 4), began with peptide coupling of 15
to the free amine derived from Fmoc-Leu-Cys(Mmt)-OMe
(18) to give ruthenium-containing linear tripeptide 19. Highly
acid-labile protecting groups [4-methyoxytrityl (Mmt) for
cysteine and 4-methyltrityl (Mtt) for lysine] were chosen for
side-chain protection. The trityl group was selectively
a Five equivalents, no NaH for step 1. b Complete transesterification
observed. c DMF as solvent for step 1.
Dichloroaromatic rings are known to undergo stepwise
displacement of halogen.6 However, displacement was not
regioselective with ruthenium complex 3,4-dichlorophenyl-
alanine 12 (Scheme 2). For example, nucleophilic attack by
(6) Pearson, A. J.; Park, J. G.; Zhu, P. Y. J. Org. Chem. 1992, 57, 3583.
(7) Pearson, A. J.; Park, J. G. J. Org. Chem. 1992, 57, 1744.
1820
Org. Lett., Vol. 1, No. 11, 1999