Ribosome-mediated incorporation of hydrazinophenylalanine into modified peptide and protein analogues

Article abstract of DOI:10.1021/ja974066e

(S)-α-Hydrazinophenylalanyl-tRNA(Phe), an aminoacyl-tRNA derivative containing the unnatural amino acid (S)-α-hydrazinophenylalanine, was prepared in an effort to examine the stereochemical requirements of the A- site of the ribosome during in vitro protein synthesis. The (S)-α- hydrazinophenylalanine moiety was of interest because it contains two nucleophilic centers, the secondary nitrogen attached to C(α), which is normally acylated during the course of peptide bond formation, and the sterically less hindered primary nitrogen. To determine the position of acylation, (S)-α-hydrazinophenylalanyl-tRNA(Phe) was tested in an Escherichia cull in vitro protein biosynthesizing system lacking elongation factor G, such that only dipeptide products were formed. The dipeptide product mixture was analyzed by HPLC in direct comparison with authentic synthetic standards. The dipeptide assay utilizing (S)-α- hydrazinophenylalanyl-tRNA(Phe) as the A-site tRNA established that the analogue functioned well as an acceptor tRNA; HPLC analysis of the products showed that both dipeptides were formed in approximately equal amounts. When attached to a suppressor tRNA transcript, (S)-α-hydrazinophenylalanine was also incorporated into position 27 of dihydrofolate reductase in an E. coli protein synthesizing system by readthrough of a nonsense codon. This finding expands the currently accepted model of peptide bond formation at the ribosome and adds to the repertoire of peptide-like products shown to form at the peptidyltransferase center of the ribosome.

Full text of DOI:10.1021/ja974066e

3032
J. Am. Chem. Soc. 1998, 120, 3032-3042
Ribosome-Mediated Incorporation of Hydrazinophenylalanine into
Modified Peptide and Protein Analogues
Jennifer A. Killian, Mark D. Van Cleve, Yuda F. Shayo, and Sidney M. Hecht*
Contribution from the Departments of Chemistry and Biology, UniVersity of Virginia,
CharlottesVille, Virginia 22901
ReceiVed December 1, 1997
Abstract: (S)-R-Hydrazinophenylalanyl-tRNAPhe, an aminoacyl-tRNA derivative containing the unnatural amino
acid (S)-R-hydrazinophenylalanine, was prepared in an effort to examine the stereochemical requirements of
the A-site of the ribosome during in vitro protein synthesis. The (S)-R-hydrazinophenylalanine moiety was of
interest because it contains two nucleophilic centers, the secondary nitrogen attached to CR, which is normally
acylated during the course of peptide bond formation, and the sterically less hindered primary nitrogen. To
determine the position of acylation, (S)-R-hydrazinophenylalanyl-tRNAPhe was tested in an Escherichia coli
in vitro protein biosynthesizing system lacking elongation factor G, such that only dipeptide products were
formed. The dipeptide product mixture was analyzed by HPLC in direct comparison with authentic synthetic
standards. The dipeptide assay utilizing (S)-R-hydrazinophenylalanyl-tRNAPhe as the A-site tRNA established
that the analogue functioned well as an acceptor tRNA; HPLC analysis of the products showed that both
dipeptides were formed in approximately equal amounts. When attached to a suppressor tRNA transcript,
(S)-R-hydrazinophenylalanine was also incorporated into position 27 of dihydrofolate reductase in an E. coli
protein synthesizing system by readthrough of a nonsense codon. This finding expands the currently accepted
model of peptide bond formation at the ribosome and adds to the repertoire of peptide-like products shown to
form at the peptidyltransferase center of the ribosome.
The catalysis of peptide bond formation between the ami-
One very specific way to probe peptidyltransferase function
would be to utilize natural tRNA substrates that contain altered
aminoacyl components, i.e., misacylated tRNAs. The first
example of the use of misacylated tRNA in protein biosynthesis
was reported in 1962 by Chapeville et al.5 These researchers
synthesized alanyl-tRNACys from cysteinyl-tRNACys by Raney
nickel-mediated desulfurization and utilized the resulting misa-
cylated tRNA to incorporate an alanine residue in the place of
a cysteine residue in the protein globin. This finding supported
the original hypothesis that the fidelity of protein synthesis was
controlled largely at the level of tRNA activation by aminoacyl-
tRNA synthetases. Fahnestock and Rich also explored the
capabilities of peptidyltransferase by the use of a misacylated
tRNA obtained via chemical deamination of enzymatically
noacyl moiety of the A-site tRNA and the peptidyl moiety of
the P-site tRNA may be regarded as the primary catalytic
function of the ribosome.1 Catalysis occurs at the peptidyl-
transferase site which is located on the 50S ribosomal subunit
in prokaryotes. Many fundamental aspects of peptide bond
formation at the peptidyltransferase site remain incompletely
defined. This includes the actual nature of catalysis, which
could in principle occur either as a result of induced proximity
of aminoacyl- and peptidyl-tRNA substrates or by some more
active role of the peptidyltransferase center.2
Studies of the peptidyltransferase center have utilized probes
such as the antibiotic puromycin, which disrupts peptide
synthesis by substituting for the A-site tRNA during peptide
bond formation.3 The peptidyltransferase site has also been
studied using aminoacyl-tRNA analogues with chemically or
photochemically activated components.2c,4
activated phenylalanyl-tRNAPhe 6
. A more general procedure for
the elaboration of misacylated tRNA’s was developed in our
laboratory7,8 and has been employed by a number of laboratories
for the introduction of modified amino acids into peptides8,9
and proteins.10
(1) (a) Hill, W. E.; Dahlberg, A. E.; Garrett, R. A.; Moore, P. B.;
Schlessinger, D.; Warner, J. R. In The Ribosome. Structure, Function and
EVolution; Am. Soc. Microbiol.: Washington, DC, 1990. (b) Noller, H. F.
Annu. ReV. Biochem. 1991, 60, 191.
(2) (a) Stade, K.; Riens, S.; Bochkariov, D.; Brimacombe, R. Nucleic
Acids Res. 1994, 22, 1394. (b) Stade, K.; Ju¨nke, N.; Brimacombe, R. Nucleic
Acids Res. 1995, 23, 2371. (c) Rinke-Appel, J.; Ju¨nke, N.; Osswald, M.;
Brimacombe, R. RNA 1995, 1, 1018. (d) Porse, B. T.; Garrett, R. A. J.
Mol. Biol. 1995, 249, 1. (e) Samaha, R. R.; Green, R.; Noller, H. F. Nature
1995, 377, 309. (f) Green, R.; Samaha, R. R.; Noller, H. F. J. Mol. Biol.
1997, 266, 40.
(3) (a) Fahnestock, S.; Neumann, H.; Shashoua, V.; Rich, A. Biochemistry
1970, 9, 2477. (b) Petska, S. In Molecular Mechanisms of Protein
Biosynthesis; Weissbach, H., Pestka, S., Eds.; Academic Press: New York,
1977; Chapter 10.
(4) Pellegrini, M.; Cantor, C. R. In Molecular Mechanisms of Protein
Biosynthesis; Weissbach, H., Pestka, S., Eds.; Academic Press: New York,
1977; Chapter 4.
A commonly used in vitro protein biosynthesizing system
for the study of peptidyltransferase activity involves polynucle-
otide templates (e.g., poly(U)) in the presence of reconstituted
ribosomes, elongation factors, GTP, K+, and Mg2+ 11
This
.
system forms polypeptides (e.g., polyphenylalanine) in response
to specific aminoacyl-tRNA’s (i.e., phenylalanyl-tRNAPhe). The
(5) (a) Chapeville, F.; Lipmann, F.; von Ehrenstein, G.; Weisblum, B.;
Ray, W. J.; Benzer, S. Proc. Natl. Acad. Sci. U.S.A. 1962, 48, 1086. (b)
von Ehrenstein, G.; Weisblum, B.; Benzer, S. Proc. Natl. Acad. Sci. U.S.A.
1963, 49, 669.
(6) (a) Fahnestock S.; Rich, A. Nature New Biol. 1971, 229, 8. (b)
Fahnestock, S.; Rich, A. Science 1971, 173, 340. See also: (c) Scolnick,
E.; Milman, G.; Rosman, M.; Caskey, T. Nature 1970, 225, 152. (d) Koh,
J. T.; Cornish, V. W.; Schultz, P. G. Biochemistry 1997, 36, 11314.
S0002-7863(97)04066-3 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/13/1998

Ribosome-Mediated Incorporation of HydrazinoPhe
J. Am. Chem. Soc., Vol. 120, No. 13, 1998 3033
Scheme 1. Preparation of
A- and P-site aminoacyl-tRNA analogues have been constructed
and used.7h,i,8 P-site aminoacyl-tRNA analogues studied have
included those acylated with (R)-amino acids, (R,S)-â-amino
acids, and non-amino acid substrates. (R)-Amino acids bound
to the ribosome but did not function well in peptide bond
formation.7h This indicates a binding pocket at the ribosomal
P-site sensitive to chirality. (R,S)-â-amino acids did function
well, however, indicating that no such restriction exists at the
â-position. A particularly interesting result was found using
N-(chloroacetyl)phenylalanyl-tRNAPhe at the P-site.7f It was
shown that two dipeptide products were formed, one resulting
from the expected acylation of the activated ester and one
resulting from the SN2 displacement of the chloro group of the
N-chloroacetyl moiety. The formation of such a product of
altered connectivity indicated a capability for chemical trans-
formations at the peptidyltransferase center not previously
anticipated.
The present study was designed to determine whether an
activated tRNA whose aminoacyl moiety contained more than
one nucleophilic functionality could form products from each
when bound at the A-site of the peptidyltransferase center. The
specific aminoacyl-tRNA analogue employed contained tRNA
activated with (S)-2-hydrazino-3-phenylpropionic acid (“(S)-R-
hydrazinophenylalanine”) which contains two nucleophilic N
atoms within the amino acid moiety. We demonstrate that this
aminoacyl-tRNA forms a product derived from each nucleo-
philic center, and that (S)-R-hydrazinophenylalanyl-tRNACUA
can be used to introduce hydrazinophenylalanine into positions
10 and 27 of Escherichia coli dihydrofolate reductase (DHFR)
by suppression of a nonsense (UAG) codon at the corresponding
positions in DHFR mRNA. This modified amino acid, there-
fore, must function with reasonable facility in each of the partial
reactions of protein biosynthesis.
(S)-R-Hydrazinophenylalanyl-tRNAPhe (II) by T4 RNA
Ligase-Mediated Coupling of tRNAPhe-COH + Chemically
Protected (S)-R-Hydrazinophenylalanyl-pdCpA
resulting peptide products are of varying length due to the lack
of normal stop and start codons. The formation of peptides of
varying length does not allow for the precise definition of the
effect of altering aminoacyl-tRNA substrates, however, and a
modified assay using high-salt washed ribosomes lacking the
elongation factor G has been used for this purpose instead.7f,h,i,8
This modified assay system forms only dipeptide products due
to the lack of elongation factor G, allowing for the quantification
and structural analysis of the dipeptide products.
Misacylated tRNA’s constructed via the strategy outlined in
Scheme 1 have been used to study the structural requirements
of the peptidyltransferase center using this dipeptide assay. Both
(7) (a) Hecht, S. M.; Alford, B. L.; Kuroda, Y.; Kitano, S. J. Biol. Chem.
1978, 253, 4517. (b) Pezzuto, J. M.; Hecht, S. M. J. Biol. Chem. 1980,
255, 865. (c) Heckler, T. G.; Zama, Y.; Naka, T.; Hecht, S. M. J. Biol.
Chem. 1983, 258, 4492. (d) Heckler, T. G.; Chang, L.-H.; Zama, Y.; Naka,
T.; Hecht, S. M. Tetrahedron 1984, 40, 87. (e) Heckler, T. G.; Chang, L.-
H.; Zama, Y.; Naka, T.; Chorghade, M. S.; Hecht, S. M. Biochemistry 1984,
23, 1468. (f) Roesser, J. R.; Chorghade, M. S.; Hecht, S. M. Biochemistry
1986, 25, 6361. (g) Payne, R. C.; Nichols, B. P.; Hecht, S. M. Biochemistry
1987, 26, 3197. (h) Heckler, T. G.; Roesser, J. R.; Cheng, X.; Chang, P.-I.;
Hecht, S. M. Biochemistry 1988, 27, 7254. (i) Roesser, J. R.; Xu, C.; Payne,
R. C.; Surratt, C. K.; Hecht, S. M. Biochemistry 1989, 28, 5185.
(8) Hecht, S. M. Acc. Chem. Res. 1992, 25, 545.
(9) (a) Bain, J. D.; Glabe, C. G.; Dix, T. A.; Chamberlin, A. R.; Diala,
E. S. J. Am. Chem. Soc. 1989, 111, 8013. (b) Bain, J. D.; Dean, A.; Wacker,
D. A.; Kuo, E. E.; Chamberlin, A. R. Tetrahedron 1991, 47, 2389. (c) Bain,
J. D.; Switzer, C.; Benner, S. A.; Chamberlin, A. R. Nature 1992, 356,
537.
(10) (a) Noren, C. J.; Spencer, J.; Anthony-Cahill, S. J.; Griffith, M. C.;
Schultz, P. G. Science 1989, 244, 182. (b) Ellman, J.; Mendel, D.; Anthony-
Cahill, S.; Noren, C. J.; Schultz, P. G. Methods Enzymol. 1991, 202, 301.
(c) Ellman, J. A.; Mendel, D.; Schultz, P. G. Science 1992, 255, 197. (d)
Ellman, J. A.; Volkman, B. F.; Mendel, D.; Schultz, P. G.; Wemmer, D. E.
J. Am. Chem. Soc. 1992, 114, 7959. (e) Mendel, D.; Ellman, J. A.; Chang,
Z.; Veenstra, D. L.; Kollman, P. A.; Schultz, P. G. Science 1993, 256, 1798.
(f) Judice, J. K.; Gamble, T. R.; Murphy, E. C.; deVos, A. M.; Schultz, P.
G. Science 1993, 261, 1578. (g) Chang, H.-H.; Benson, D. R.; Schultz, P.
G. Science 1993, 259, 806. (h) Mendel, D.; Ellman, J. A.; Schultz, P. G. J.
Am. Chem. Soc. 1993, 115, 4359. (i) Cornish, V. W.; Benson, D. R.;
Altenbach, C. A.; Hideg, K.; Hubbell, W. L.; Schultz, P. G. Proc. Natl.
Acad. Sci. U.S.A. 1994, 91, 2910. (j) Thorson, J. S.; Chapman, E.; Murphy,
E. C.; Schultz, P. G., Judice J. K. J. Am. Chem. Soc. 1995, 117, 1157. (k)
Cornish, V. W.; Mendel, D.; Schultz, P. G. Angew. Chem., Int. Ed. Engl.
1995, 34, 621. (l) Nowak, M. W.; Kearney, P. C.; Sampson, J. R.; Saks,
M. E.; Labarca, C. G.; Silverman, S. K.; Zhong, W.; Thorson, J.; Abelson,
J. N.; Davidson, N.; Schultz, P. G.; Dougherty, D. A.; Lester, H. A. Science
1995, 268, 439. (m) Mamaev, S. V.; Laikhter, A. L.; Arslan, T.; Hecht, S.
M. J. Am. Chem. Soc. 1996, 118, 7243. (n) Saks, M. E.; Sampson, J. R.;
Nowak, M. W.; Kearney, P. C.; Du, F.; Abelson, J. N.; Lester, H. A.;
Dougherty, D. A. J. Biol. Chem. 1996, 271, 23169. (o) Steward, L. E.;
Collins, C. S.; Gilmore, M. A.; Carlson, J. E.; Ross, J. B. A.; Chamberlin,
A. R. J. Am. Chem. Soc. 1997, 119, 6. (p) Karginov, V. A.; Mamaev, S.
V.; An, H.; Van Cleve, M. D.; Hecht, S. M.; Komatsoulis, G. A.; Abelson,
J. N. J. Am. Chem. Soc. 1997, 119, 8166. (q) Arslan, T.; Mamaev, S. V.;
Mamaeva, N. V.; Hecht, S. M. J. Am. Chem. Soc. 1997, 119, 10877. (r)
England, P. M.; Lester, H. A.; Davidson, N.; Dougherty, D. A. Proc. Natl.
Acad. Sci. U.S.A. 1997, 94, 11025.
Results and Discussion
Preparation of (S)-r-Hydrazinophenylalanyl-tRNAPhe. The
primary technical goal of this study was definition of the ability
of an aminoacyl-tRNA containing more than one nucleophilic
functionality within the aminoacyl moiety to form a product
derived from each at the ribosomal peptidyltransferase center.
The aminoacyl-tRNA analogue used for this purpose was (S)-
R-hydrazinophenylalanyl-tRNAPhe (II). As shown in Figure 1,
this analogue differs from (S)-phenylalanyl-tRNAPhe (I) in that
it contains two N nucleophiles at CR of the aminoacyl moiety.
Thus it seemed logical to think that tRNA II might exhibit dual
reactivity once bound at the ribosomal A-site. As illustrated
in Scheme 2 for a ribosome bearing N-acetylphenylalanyl-
tRNAPhe at the P-site and (S)-R-hydrazinophenylalanyl-tRNAPhe
at the A-site, nucleophilic attack of the N-atom connected to
CR of R-hydrazinophenylalanine on the activated carboxylate
of N-acetylphenylalanyl-tRNAPhe (pathway b) would afford
dipeptide 1 having connectivity analogous to those in naturally
occurring peptides. In contrast, nucleophilic attack by the other
N (pathway a) would produce a peptide analogue (2) of altered
connectivity.
To study the transformations outlined in Scheme 2, we
prepared (S)-R-hydrazinophenylalanyl-tRNAPhe by the route
outlined in Scheme 1. This involved removal of two nucleotides
from the 3′-terminus of mature yeast tRNAPhe by limited
digestion with venom exonuclease to afford tRNAPhe-COH, as
described previously.7 Condensation with NVOC-protected (S)-
R-hydrazinophenylalanyl pdCpA (12) via the agency of T4 RNA
ligase afforded NVOC-protected (S)-R-hydrazinophenylalanyl-
tRNAPhe, which was deblocked photochemically as described
(11) See, e.g., ref 7b and references therein.

3034 J. Am. Chem. Soc., Vol. 120, No. 13, 1998
Killian et al.
troveratryl chloroformate16 provided both the mono- and di-
NVOC protected compounds 8 and 9 (in 39% and 33% yields,
respectively). The assignment of position of monoacylation
shown for compound 8 (Nâ) was based on literature precedent
for Nâ-acylation of R-hydrazino acids.17 The mono- and
diprotected intermediates 8 and 9 were then treated separately
with ClCH2CN to give the corresponding cyanomethyl esters
10 and 11 in 71% and 61% yields, respectively.
Tetrabutylammonium pdCpA18 was aminoacylated by treat-
ment with monoprotected N-NVOC-(S)-R-hydrazinophenyla-
lanine (10) and diprotected N,N′-di-NVOC-(S)-R-hydrazinophe-
nylalanine (11) to give the corresponding aminoacylated
dinucleotides 12 and 13, respectively (Scheme 5). The progress
of the reactions was monitored by HPLC, and the compounds
were subsequently purified by HPLC. To permit control
reactions to be run, N-NVOC-(S)-phenylalanyl-pdCpA was
synthesized by the same method.19 (S)-Phenylalanine was
treated with nitroveratryl chloroformate to afford N-NVOC-(S)-
phenylalanine in 79% yield. This intermediate was treated with
chloroacetonitrile to afford N-NVOC-(S)-phenylalanine cya-
nomethyl ester in 78% yield. Tetrabutylammonium pdCpA was
then treated with N-NVOC-(S)-phenylalanine cyanomethyl ester
to give N-NVOC-(S)-phenylalanyl-pdCpA.19
To ensure the successful deprotection of the corresponding
N-NVOC-aminoacyl-tRNAs, attempts were made to deprotect
the N-NVOC-aminoacyl-pdCpA dinucleotides. The deprotec-
tion reactions were carried out in 1 mM potassium acetate buffer
solution, pH 4.5, at a concentration anticipated to be close to
that which would be used for the corresponding N-NVOC-
aminoacyl-tRNA’s (28 µM).7i Each sample was irradiated with
light from a high-pressure mercury lamp (1000 W), and time
points were taken every 2.5 min for analysis by HPLC. For
N-NVOC-(S)-R-hydrazinophenylalanyl-pdCpA (12) and N,N′-
di-NVOC-(S)-R-hydrazinophenylalanyl-pdCpA (13), HPLC analy-
sis showed that, after 2.5 min of irradiation, much of the starting
material had been consumed and after 7.5 min all of the starting
material was gone and many products were present. In
comparison, N-NVOC-(S)-phenylalanyl-pdCpA showed no such
behavior; after 5 min of irradiation most of the starting material
was consumed and a single product peak had formed. The
mechanism of photolytic NVOC removal is thought to involve
light-catalyzed intramolecular rearrangement in which the nitro
group is reduced to a nitroso function and the oxygen is inserted
into the C-H bond in the ortho position.20 The resulting
R-hydroxycarbamate functionality decomposes to give the
corresponding nitrosobenzaldehyde product, CO2, and the free
amine. It has been reported that photolytic removal of the
NVOC group from amino moieties are occasionally not
quantitative because of a side reaction between the released
amino functionality and the aldehyde byproduct, which forms
the corresponding imine.21 In the present case, it seemed
possible that the released hydrazinophenylalanyl dinucleotide
14 might condense with the benzaldehyde to form the corre-
Figure 1. Structures of (S)-phenylalanyl-tRNAPhe (I) and (S)-R-
hydrazinophenylalanyl-tRNAPhe (II).
below. The synthesis of the requisite aminoacylated dinucle-
otide 12 is outlined in Schemes 3-5 and involved initial
synthesis of (S)-R-hydrazinophenylalanine (7) itself (Scheme
3). (R)-phenylalanine (3) was converted to (R)-2-hydroxy-3-
phenylpropionic acid (4) with overall retention of stereochem-
istry in a diazotization/hydrolysis reaction believed to proceed
via an intermediate R-lactone (42% yield).12 The optical rotation
of (R)-2-hydroxy-3-phenylpropionic acid (4) was of essentially
the same magnitude but opposite sign relative to the published
value for the S enantiomer {[R]25 +29.8° (c 1.13, acetone);
D
literature value,12a [R]25D -27.8° (c 1.13, acetone)}. Compound
4 was converted to the respective methyl ester (5) using
diazomethane (80% yield). The enantiomeric purity (>98%)
of methyl ester 5 was verified by derivatization as the corre-
sponding Mosher ester.13 The R-hydroxy moiety was then
replaced with a BOC-hydrazino group with inversion of
stereochemistry by SN2 displacement of a trifluoromethane-
sulfonate intermediate, affording methyl (S)-2-[N-tert-butoxy-
carbonyl)hydrazino]-3-phenylpropionate (6) as a colorless wax
in 67% yield.14 The optical rotation of compound 6 was of the
same magnitude but opposite sign relative to the published value
for the corresponding R enantiomer {[R]25 -12.2° (c 1.19,
D
CHCl3); literature value,14 [R]25D +12.0° (c 1.0, CHCl3)}. The
assignment of the structure of compound 6 as resulting from
nucleophilic displacement of the triflate by the unsubstituted
nitrogen of tert-butylcarbazate (BocNHNH2) was made initially
based on the work of Legrel et al.15 who studied the acylation
of tert-butylcarbazate. Compound 6 was deprotected by suc-
cessive treatments with trifluoroacetic acid and NaOH, affording
(S)-R-hydrazinophenylalanine (7) in 52% overall yield.
(16) Patchornik, A.; Amit, B.; Zehavi, U. J. Org. Chem. 1974, 39, 192.
(17) (a) Niedrich, H. Chem. Ber. 1965, 98, 3451. (b) Grupe, R.; Niedrich,
H. Chem. Ber. 1967, 100, 3283. (c) Lecoq, A.; Marraud, M.; Aubry, A.
Tetrahedron Lett. 1991, 32, 2765. (d) Aubry, A.; Bayeul, D.; Mangeot,
J.-P.; Vidal, J.; Sterin, S.; Collet, A.; Lecoq, A.; Marraud, M. Biopolymers
1991, 31, 793.
(18) Robertson, S. A.; Noren, C. J.; Anthony-Cahill, S. J.; Griffith, M.
C.; Schultz, P. G. Nucleic Acids Res. 1989, 17, 9649.
(19) Robertson, S. A.; Ellman, J. A.; Schultz, P. G. J. Am. Chem. Soc.
1991, 113, 2722.
(20) Morrison, H. A. In The Chemistry of the Nitro and Nitroso Groups,
Part 1; Patai, S., Ed.; John Wiley and Sons: New York, 1969; Chapter 4.
(21) Woodward, R. B.; Patchornik, A.; Amit, B. J. Am. Chem. Soc. 1970,
92, 6333.
The conversion of (S)-R-hydrazinophenylalanine to the
respective N-NVOC cyanomethyl ester is outlined in Scheme
4. Treatment of (S)-R-hydrazinophenylalanine (7) with 2-ni-
(12) (a) Cohen, S. G.; Weinstein, S. Y. J. Am. Chem. Soc. 1964, 86,
5326. (b) Pirrung, M. C.; Brown, W. L. J. Am. Chem. Soc. 1990, 112, 6388.
(13) Dale, J. A.; Dull, D. L.; Mosher, H. S. J. Org. Chem. 1969, 34,
2543.
(14) Hoffman, R. V.; Kim, H.-O. Tetrahedron Lett. 1990, 31, 2953.
(15) Legrel, P.; Baudy-Floc’h, M.; Robert, A. Synthesis 1987, 306.

Ribosome-Mediated Incorporation of HydrazinoPhe
J. Am. Chem. Soc., Vol. 120, No. 13, 1998 3035
Scheme 2. Ribosomally Mediated Dipeptide Formation Using N-Acetyl-(S)-phenylalanyl-tRNAPhe and
(S)-R-Hydrazinophenylalanyl-tRNAPhe as the Donor and Acceptor, Respectively
Scheme 3.a Chemical Synthesis of
(S)-R-Hydrazinophenylalanine (7)
Scheme 5. Synthesis and Photodeprotection of
N-NVOC-(S)-R-hydrazinophenylalanyl-pCpAs
a Reaction Conditions: (a) NaNO2, H+3O; (b) CH2N2; (c) Tf2O and
then BocNHNH2; (d) 25% CF3COOH in CH2Cl2 and then NaOH.
Scheme 4.a Chemical Synthesis of N-NVOC Protected
Cyanomethyl Ester Derivatives of
(S)-R-Hydrazinophenylalanine
Table 1. Ribosome-Mediated Dipeptide Formationa
aminoacyl-tRNA
dipeptide formed (%)b
(S)-phenylalanyl-tRNAPhe (I)c
100
58
74
(S)-phenylalanyl-tRNAPhe (I)d
(S)-R-hydrazinophenylalanyl-tRNAPhe (II)d
a The peptidyl-tRNA analogue employed in each case was [3H]N-
acetyl-(S)-phenylalanyl-tRNAPhe b Relative yields of dipeptides, based
.
on incorporation of [3H]radiolabel. c Prepared by activation of yeast
tRNAPhe with (S)-phenylalanine in the presence of the cognate ami-
noacyl-tRNA synthetase. d Prepared by T4 RNA ligase-mediated
coupling of tRNAPhe-COH with an aminoacylated pdCpA.
a Reaction conditions: (a) 2-nitroveratryl chloroformate, Na2CO3;
(b) ClCH2CN.
ligated onto the enzymatically abbreviated tRNA (tRNAPhe-COH
;
Scheme 1) and deblocked, and the resulting (S)-R-hydrazi-
nophenylalanyl-tRNAPhe (II) was used as the acceptor tRNA
in an in vitro protein biosynthesizing system that contained
ribosomes, poly(U) and [3H]-N-acetylphenylalanyl-tRNAPhe but
no elongation factor G; this ensured that peptide bond formation
would be limited to the elaboration of dipeptides. The results
of the dipeptide assay are shown in Table 1. (S)-R-Hydrazi-
nophenylalanyl-tRNAPhe (II) functioned well as an acceptor
tRNA, forming dipeptide in slightly higher yield (74%) than
(S)-phenylalanyl-tRNAPhe constructed via the chemical ami-
noacylation procedure outlined in Scheme 1 (58% yield). The
lower yields of dipeptide obtained with both of these species,
relative to enzymatically activated phenylalanyl-tRNAPhe, is
consistent with experience and probably reflects damage
inflicted on the “chemically aminoacylated” tRNA’s during the
several steps employed for their preparation. While not
investigated in detail, adventitious ribonuclease activities present
in the enzymes used for tRNA modification, as well as
sponding hydrazone. The resulting hydrazone might well be
subject to decomposition, as hydrazones have been reported to
be unstable photochemically.17c
It was found that the addition of semicarbazide hydrochloride
to the photolysis mixture (as a scavenger for the formed
benzaldehyde derivative21) led to the formation of a single major
product (49%) having a retention time (14.4 min) expected for
the desired product 14 under the HPLC conditions employed.
To ensure that the addition of semicarbazide hydrochloride was
not interfering with NVOC removal per se, a parallel control
reaction was run containing NVOC-(S)-phenylalanyl-pdCpA.
Comparison of the results of the photolysis of NVOC-(S)-
phenylalanyl-pdCpA in the presence or absence of semicarba-
zide hydrochloride (200 equiv) showed that the same major
product was formed to the same extent in either case.
Dipeptide Formation and Analysis. The aminoacylated
dinucleotide N-NVOC-(S)-R-hydrazinophenylalanyl-pdCpA was

3036 J. Am. Chem. Soc., Vol. 120, No. 13, 1998
Killian et al.
Scheme 6.a Synthesis of the Dipeptide N-Acetylphenylalanyl-R-hydrazinophenylalanine Having Normal Peptide Connectivity
a Reaction conditions: (a) 1:1 CH2Cl2-10% NaHCO3; (b) DBU, N-acetoxysuccinimide; (c) NaOH; (d) 25% CF3COOH in CH2Cl2.
Scheme 7.a Synthesis of the Dipeptide
N-Acetylphenylalanyl-R-hydrazinophenylalanine Having
Altered Connectivity
photolysis with a high-intensity light source during NVOC
removal, are likely sources of the damage.
The participation of (S)-R-hydrazinophenylalanyl-tRNAPhe as
an acceptor tRNA having been established, the next consider-
ation was the determination of the connectivity of the dipeptide
product(s). Experimentally, this was carried out by HPLC
analysis of the ribosomally formed dipeptide(s) using authentic,
synthetic dipeptides as standards.
Scheme 6 illustrates the synthesis of the dipeptide acylated
at the nitrogen attached to CR of hyrazinophenylalanine, i.e.,
having connectivity analogous to that in normal peptides. It
was felt that protection of the primary amine of R-hydrazi-
nophenylalanine would allow for the regioselective acylation
of the other nitrogen. In fact, regioselective acylation of the
R-nitrogen of Nâ-protected hydrazino acid esters (CBz-NâNNR-
HCH2COOCH2CH3) has been reported using N-FMOC-pro-
tected amino acid chlorides.17c,d Accordingly, N-t-BOC-(S)-
R-hydrazinophenylalanine methyl ester (6) and the acid chloride
of N-FMOC-(S)-phenylalanine (15)22 were coupled in high yield
to afford putative dipeptide derivative 16. Conversion of
putative 16 to the required dipeptide standard (1) involved
removal of the FMOC group, acetylation, hydrolysis of the
methyl ester, and removal of the t-BOC protecting group. Initial
attempts to remove the FMOC group from compound 16 under
the usual basic conditions using piperidine or DBU (1,8-
diazabicyclo[5.4.0]undec-7-ene) did not afford the desired
dipeptide product 17 but instead resulted in quantitative forma-
tion of diketopiperazine 18. Since the next step would have
been to reacetylate the released amino function in any case, the
problem of diketopiperazine side product formation was cir-
cumvented by removing the FMOC group in the presence of
the acetylating reagent N-acetoxysuccinimide.23 Optimization
of the reaction conditions permitted the isolation of the desired
product 17 in 40% yield. Dipeptide derivative 17 was then
saponified to give the corresponding carboxylic acid 19 in 55%
yield. Treatment with trifluoroacetic acid then provided the
dipeptide of normal connectivity (1) in 88% yield.
Scheme 7 shows the synthetic route used for preparation of
the dipeptide acylated at the primary amine of R-hydrazinophe-
nylalanine, i.e., the dipeptide of altered connectivity (2). As
noted above, it has been reported that under standard peptide
coupling conditions, the coupling of R-hydrazino acid esters
(NâH2NRHCHRCOOCH2CH3) with N-protected amino acids
results primarily in Nâ-acylation.17 The strategy used for
synthesis of the dipeptide of altered connectivity (2) involved
the coupling of the deprotected (S)-R-hydrazinophenylalanine
methyl ester (21; accessible from 6 by treatment with trifluo-
roacetic acid) with N-CBz-(S)-phenylalanine (20) via the agency
of DCC and HOBT to afford putative dipeptide derivative 22
in 54% yield. The CBz group was then removed by hydro-
genolysis over 10% palladium-on-carbon to afford 23 in 83%
a Reaction Conditions: (a) DCC, HOBt; (b) 10% Pd-C, H2; (c)
N-acetoxysuccinimide; (d) NaOH.
yield; 23 showed no propensity toward cyclization. Dipeptide
23 was then N-acetylated using N-acetoxysuccinimide23 to afford
compound 24 (95% yield). Finally, dipeptide derivative 24 was
saponified to give the dipeptide of altered connectivity (2).
Beyond the assignment of connectivity in dipeptides 1 and 2
based on expectations from literature precedent involving the
coupling of hydrazino acid esters (NâH2NRHCHRCOOCH2CH3)
with N-protected amino acids,17 the facile formation of dike-
topiperazine 18 upon removal of the nitrogen protecting group
of peptide ester 16 (Scheme 6) constitutes strong inferential
evidence that dipeptide 16 was acylated on the R-nitrogen.
Diketopiperazine formation from dipeptide esters is a common
side reaction in solid-phase peptide synthesis.24 The lack of
cyclization of analogous intermediate 23 (which would involve
the formation of a seven-membered ring) may be adduced as
further support for the assignments.
The connectivity of the formed dipeptides was verified using
a dipeptide of altered connectivity, N-Ac-(S)-phenylalanyl-Nâ-
(S)-R-hydrazinophenylalanine methyl ester (24), for detailed
NMR spectral analysis. A COSY spectrum taken in CDCl3
indicated the presence of a cross-peak between the two
hydrazino NHs at δ 7.60 and 4.59, corresponding to NâH and
NRH, respectively. The latter exhibited a cross-peak to CRH
(δ 3.67) of the hydrazinophenylalanine moiety. Comparable
analysis of the NH signals for two dipeptides of normal
connectivity, including N-Ac-(S)-phenylalanyl-NR-(S)-R-hy-
drazinophenylalanine methyl ester (methyl ester of dipeptide
1; accessible by treatment of 17 with trifluorocetic acid in CH2-
Cl2) revealed a single signal at ∼δ 3.9 which had no cross-
peak to any other H in the COSY spectrum. The assignment
of signals to the hydrazino NH’s was supported by the ability
of each of these H’s to exchange with D2O; this exchange was
fast relative to that of the amide NH’s present.
(22) Carpino, L. A.; Cohen, B. J.; Stephens, K. E.; Sadat-Aalaee, S. Y.;
Tien, J.-H.; Langridge, D. C. J. Org. Chem. 1986, 51, 3732.
(23) Rappoport, S.; Lapidot, Y. Methods Enzymol. 1974, 29, 685.
(24) Bodansky, M. In Principles of Peptide Synthesis; Springer-Verlag:
Heidelberg, Germany, 1993.

Ribosome-Mediated Incorporation of HydrazinoPhe
J. Am. Chem. Soc., Vol. 120, No. 13, 1998 3037
Scheme 8. Products Formed from N-chloroacetylphenylalanyl-tRNAPhe and Phenylalanyl-tRNAPhe in the Presence of E. coli
Ribosomes
2 was 72% of theoretical. The yields of dipeptides 1 and 2
were 33 and 39%, respectively.
An analysis of the ribosomal dipeptide solution obtained by
using phenylalanyl-tRNAPhe as the acceptor tRNA in the
dipeptide assay was carried out as a control. A portion of this
dipeptide product was mixed with unlabeled synthetic standard
and subjected to HPLC analysis under the same conditions
employed for 1 and 2. The UV profile matched the radioactivity
profile closely; N-acetyl-(S)-phenylalanyl-(S)-phenylalanine was
formed in 41% of theoretical yield (data not shown).
In an earlier study, we demonstrated that a peptidyl-tRNA
analogue (N-chloroacetyl-(S)-phenylalanyl-tRNAPhe) containing
two electrophilic groups was capable of binding to the P-site
of E. coli ribosomes and reacting with a normal aminoacyl-
tRNA in the A-site to provide products derived from reaction
at both electrophilic centers (Scheme 8).7f The products were
identified by comparison with authentic synthetic standards, as
well as chemical degradation studies involving intermediates
radiolabeled on specific atoms in the amino acids.
The present study extends the earlier findings in that it
reinforces the concept that the peptidyltransferase reaction
leading to peptide bond formation is likely a direct chemical
reaction between the reactive partners at the ribosomal A- and
Figure 2. Superposition of the radioactivity and UV HPLC profiles
P-sites. As was also evident from our earlier report,7f the present
of a sample formed by admixture of authentic 1 and 2 (taller peaks at
study reflects a greater degree of flexibility of ribosomal function
17.7 and 19.2 min, respectively) to a dipeptide synthesized ribosomally
than might be anticipated if the peptidyltransferase center were
viewed as an active participant in the process of peptide bond
formation. Thus, the present findings reinforce the view that
the function of the peptidyltransferase center is simply to bring
into spatial proximity chemically reactive species intrinsically
capable of forming the products that are actually observed. To
the extent that the ribosome facilitates peptide bond formation
beyond the proximity effect, it seems likely that the microen-
vironment of the peptidyltransferase center is responsible, rather
than a specific set of functional groups that control the motion
of the reactive aminoacylated tRNA participants. Aside from
its mechanistic implications, this view leads to the prediction
that peptide bond formation could be accomplished on any
scaffold that can achieve a spatial organization of individual
aminoacyl-tRNA’s and microenvironment comparable to that
which obtains on the ribosome. In fact, the formation of peptide
bonds on deproteinized ribosomes,25 as well as by 23S ribosomal
RNA alone,26 is entirely consistent with this view.
in response to poly(U) in the presence of [3H]N-acetyl-(S)-phenylalanyl-
tRNAPhe and (S)-R-hydrazinophenylalanyl-tRNAPhe
.
In preparation for the HPLC analysis of the ribosomal
dipeptide mixture, dipeptide standards 1 and 2 were purified
by HPLC; optimization of the HPLC conditions to maximize
separation of the two was carried out and separation was
achieved (data not shown).
HPLC analysis was carried out on a portion of the ribosomally
derived dipeptide product(s) obtained using (S)-R-hydrazinophe-
nylalanyl-tRNAPhe (II) as the acceptor tRNA. An aliquot was
frozen, concentrated under diminished pressure, and combined
with a 1:1 mixture of dipeptide standards. HPLC analysis was
carried out on this mixture. Fractions were taken every 1 min
until the region where dipeptide(s) elution was anticipated and
then every 10 s. The radioactivity in each fraction was
determined, and a graph of time vs radioactivity was constructed.
Figure 2 shows the superposition of the UV and radioactivity
profiles. Comparison of the two profiles show that compounds
containing radioactivity coeluted with the authentic dipeptide
standards. The first and second peaks corresponded to the
dipeptides of normal (1) and altered (2) connectivity, respec-
tively; they were formed in a ratio of 1:1.2. On the basis of
the amount of [3H]N-acetyl-(S)-phenylalanyl-tRNAPhe that bound
to the ribosomes, the total dipeptide formation reflected in Figure
Incorporation of (S)-r-Hydrazinophenylalanine into Di-
hydrofolate Reductase. The foregoing data indicate that (S)-
R-hydrazinophenylalanyl-tRNAPhe (II) functions well as an
(25) Noller, H. F.; Hoffarth, V.; Zimniak, L. Science 1992, 256, 1416.
(26) (a) Nitta, I.; Ueda, T.; Watanabe, K. In 17th International tRNA
Workshop Abstracts; Chiba, Japan, 1997; p 9-18. (b) Nitta, I. In 22nd
Seminar on Frontier Technology Abstracts; Tokyo, 1997; p 57ff.

3038 J. Am. Chem. Soc., Vol. 120, No. 13, 1998
Killian et al.
An R-hydrazino acid moiety affords the opportunity to
introduce an additional substituent in a protein (attached to NR
or Nâ) without alteration of the primary amino acid sequence.
Also, an R-hydrazino acid might be used to exert a predefined
conformational effect in a peptide; Aubry et al. have reported
data on the geometry and conformational influence of the
R-hydrazino acid moiety in R-hydrazino acid-containing peptides
obtained from solid-state studies (X-ray crystallography) and
solution studies (1H NMR, IR spectroscopy).17c,d,28 The data
generally indicate that while the geometry (bond lengths, bond
angles) of the amide moiety remains largely unchanged, the
presence of the second nitrogen significantly alters the hydrogen
bonding characteristics of the peptide and, thus, the overall
conformation. Also, acylation of NR or Nâ showed very different
effects on the flexibility of the N-N bond and, thus, the
structural rigidity of the peptide.
Figure 3. Synthesis of E. coli dihydrofolate reductase containing (S)-
R-hydrazinophenylalanine at position 10 by in vitro suppression of a
nonsense codon in the presence of (S)-R-hydrazinophenylalanyl-
tRNAPhe: lane 1, mRNA from plasmid pTHD10, no suppressor tRNA;
lane 2, DHFR elaborated from wild-type mRNA; lanes 3-6, DHFR
expressed from pTHD10 mRNA in the presence of suppressor
tRNAPheCUA activated with (lane 3) (S)-valine (lane 4) (S)-phenylalanine,
(lane 5) (S)-R-hydrazinophenylalanine (prepared from 12); lane 6, (S)-
R-hydrazinophenylalanine (prepared from 13). A control was also run
using the mRNA from plasmid pTHD10 in the presence of full length
unacylated tRNAPheCUA. Negligible readthrough was observed.
acceptor in the peptidyltransferase reaction and that both possible
dipeptide products resulting from N-acylation are formed in
approximately equal amounts. Because virtually all misacylated
tRNA’s tested have functioned as donors in the peptidyltrans-
ferase reaction,7f,h,8 it seemed likely that (acylated) (S)-R-
hydrazinophenylalanyl-tRNA (II) would do so as well. Ac-
cordingly, we attempted to incorporate the (S)-R-
hydrazinophenylalanyl moiety into a single position in a protein.
This was carried out by attaching (S)-R-hydrazinophenylalanine
to a suppressor tRNACUA transcript, essentially as outlined in
Scheme 1, and including the misacylated suppressor tRNA in
an E. coli protein synthesizing system. Also included was the
mRNA for E. coli dihydrofolate reductase containing a nonsense
(UAG) codon at position 10. As shown in Figure 3, full length
DHFR was formed when (S)-R-hydrazinophenylalanyl-tRNACUA
was present in the protein synthesizing system. Relative to the
elaboration of wild-type DHFR from a plasmid lacking any
nonsense codon, (S)-R-hydrazinophenylalanine was incorporated
to the extent of about 10% when (S)-R-hydrazinophenylalanyl-
tRNACUA (II) was prepared from aminoacylated dinucleotide
12 (Scheme 1) and in about 3% yield when activated tRNA II
was prepared using dinucleotide 13 (Scheme 5). In comparison
valine was incorporated into position 10 in 53% yield and
phenylalanine in 12% yield (Figure 3, lanes 3 and 4, respec-
tively) in this particular experiment. (S)-R-Hydrazinophenyla-
lanine was also incorporated into position 27 of DHFR in
analogous fashion, albeit in somewhat lower yield (not shown).
Critically, the incorporation of hydrazinophenylalanine could
be demonstrated reproducibly. No full length DHFR was
obtained in the absence of this misacylated tRNA or when the
tRNA was not deblocked prior to admixture to the protein
synthesizing system.
While numerous amino acid analogues have now been
incorporated into peptides8,9 and proteins,10 only a few of these
have resulted in products of altered connectivity. While no
direct analysis of the connectivity of the hydrazinophenylalanine
moiety in DHFR has been carried out as yet, it seems not
unlikely that both N’s have participated in forming the backbone
of the derived DHFR. Another interesting facet of hydrazi-
nophenylanyl-tRNA’s not addressed in the current studies
involves the ability of aminoacyl-tRNA’s derived from mono-
protected R-hydrazino acids, (e.g. 8) to participate in peptide
bond formation. If successful, these would afford caged
proteins27 of known connectivity; the facile deblocking of such
species has recently been demonstrated,10p affording catalytically
active proteins.
Experimental Section
General Methods. Elemental analyses were carried out by Atlantic
Microlab, Inc. Melting points were taken on a Thomas-Hoover
apparatus and are not corrected. TLC Rf values reported were obtained
using 0.25 mm Merck glass-backed silica gel TLC plates. Optical
rotations were determined on a Perkin-Elmer model 141 polarimeter.
1H NMR spectra were recorded on a 300 MHz spectrophotometer.
Chemical shifts are referenced to CHCl3 at 7.26 ppm, CH3OH at 3.30
ppm, or HOD at 4.80 ppm. The following abbreviations are used; s,
singlet; d, doublet; t, triplet; m, multiplet; br, broad; dd, doublet of
doublets. Chemical ionization mass spectra (CIMS) were recorded on
a Finnigan MAT 4600 mass spectrometer using methane as the carrier
gas unless otherwise specified. Electron impact mass spectra (EIMS)
were also recorded on a Finnigan MAT 4600 mass spectrometer.
Liquid secondary ion mass spectra (LSIMS) were obtained on a
Quadrupole-FT mass spectrometer. High-resolution peak matching was
carried out at the Michigan State University Mass Spectrometry Facility
or at the Nebraska Center for Mass Spectrometry, University of
Nebraska.
(R)-2-Hydroxy-3-phenylpropionic Acid (4). To a cold (0-5 °C),
stirred suspension of 20 g (0.12 mol) of (R)-phenylalanine in 150 mL
of CHCl3 was added 4.4 mL (5.3 g; 0.15 mol) of concentrated HCl.
The solution was stirred at 0 °C for 10 min and concentrated under
diminished pressure. The residue was diluted with 350 mL of 5%
aqueous H2SO4 and cooled in an ice bath. To the cold (0-5 °C)
solution was added a solution containing 25 g (0.36 mmol) of NaNO2
in 100 mL of water. The solution was stirred at 0 °C for 6 h and then
extracted with three 200-mL portions of ether. The organic extract
was dried over Na2SO4 and concentrated under diminished pressure.
Crystallization from CH2Cl2 gave (R)-2-hydroxy-3-phenylpropionic acid
(4) as colorless needles: yield 8.4 g (42%); mp 120-121 °C, lit.12a
mp 126-127 °C; silica gel TLC Rf 0.83 (4:1:1 n-butanol-acetic acid-
water); [R]25 +29.8° (c 1.13, acetone), lit.12a value for S enantiomer,
D
[R]25D -27.8° (c 1.13, acetone); 1H NMR (CD3OD) δ 2.87 (dd, 1H, J
) 14, 8 Hz), 3.08 (dd, 1H, J ) 14, 4 Hz), 4.32 (dd, 1H, J ) 8, 4 Hz),
4.71 (br s, 1H), and 7.13-7.22 (m, 5H); mass spectrum (chemical
ionization) m/z 167, (M + H)+. Anal. Calcd for C9H10O3: C, 65.05;
H, 6.07. Found: C, 64.98; H, 6.10.
Methyl (R)-2-Hydroxy-3-phenylpropionate (5). To a cold (0-5
°C) solution containing 7.0 g (42 mmol) of (R)-2-hydroxy-3-phenyl-
propionic acid (4) in 60 mL of dry THF was added diazomethane until
a yellow color persisted. Excess diazomethane was destroyed by
dropwise addition of 10% aqueous acetic acid until bubbling ceased.
The solution was concentrated under diminished pressure, and the
residue was dissolved in 30 mL of CH2Cl2 and extracted with two 20-
mL portions of brine. The organic extract was dried over Na2SO4 and
concentrated under diminished pressure. Crystallization of the product
from ether gave methyl (R)-2-hydroxy-3-phenylpropionate (5) as
colorless needles: yield 6.1 g (80%); mp 43-44 °C; silica gel TLC Rf
0.78 (1:1 ethyl acetate-hexane); [R]25 +15.5° (c 1.0, CH2Cl2); H
1
D
NMR (CDCl3) δ 2.73 (br d, 1H, J ) 6 Hz), 2.97 (dd, 1H, J ) 14, 7
Hz), 3.13 (dd, 1H, J ) 14, 5 Hz), 3.77 (s, 3H), 4.46 (m, 1H), and
(27) For a review on caged compounds, see: Adams, S. R.; Tsien, R.
Y. Annu. ReV. Physiol. 1993, 55, 755.

Ribosome-Mediated Incorporation of HydrazinoPhe
J. Am. Chem. Soc., Vol. 120, No. 13, 1998 3039
7.20-7.33 (m, 5H); mass spectrum (chemical ionization) m/z 181, (M
+ H)+. Anal. Calcd for C10H12O3: C, 66.65; H, 6.71. Found: C,
66.50; H, 6.71.
colorless microcrystals: yield 95 mg (52%); mp 197-200 °C, lit.29a
mp 196-201 °C; silica gel TLC Rf 0.61 (5:2:3 butanol-acetic acid-
water); 1H NMR (D2O) δ 2.87 (m, 2H), 3.45 (m, 1H), and 7.21-7.32
(m, 5H); mass spectrum (LSIMS) m/z 181, (M + H)+.
Mosher Ester Derivative of Methyl (R)-2-Hydroxy-3-phenylpro-
pionate (5). A solution containing 20 mg (0.11 mmol) of methyl (R)-
2-hydroxy-3-phenylpropionate (5), 23 mg (0.11 mmol) of N,N′-dicyclo-
hexylcarbodiimide, 29 mg (0.24 mmol) of 4-(dimethylamino)pyridine,
and 34 mg (0.15 mmol) of (R)-(+)-R-methoxy-R-trifluoromethylphe-
nylacetic acid (MTPA) in 0.5 mL of CH2Cl2 was stirred at 25 °C for
8 h. The solution was filtered to remove the precipitate, and the filtrate
was concentrated under diminished pressure. The product was purified
by preparative silica gel TLC (development was with 15% ethyl acetate
in hexane) to provide the Mosher ester13 of methyl (R)-2-hydroxy-3-
phenylpropionate (5) as a colorless oil: yield 9 mg (20%); silica gel
N-NVOC-(S)-r-hydrazinophenylalanine (8) and N,N′-di-NVOC-
(S)-r-hydrazinophenylalanine (9). To a solution containing 70 mg
(0.39 mmol) of (S)-R-hydrazinophenylalanine (7) and 62 mg (0.58
mmol) of Na2CO3 in 1.0 mL of water was added a solution containing
128 mg (0.47 mmol) of 2-nitroveratryl chloroformate16 in 2 mL of
dioxane. The combined solution was stirred at 25 °C for 5 h. The
dioxane was removed under diminished pressure, and the aqueous layer
was acidified to pH 6.0 with 1 N NaHSO4. The aqueous layer was
extracted with three 20-mL portions of CH2Cl2. The organic layer was
dried over Na2SO4 and concentrated under diminished pressure. The
residue was purified by chromatography on a silica gel column (15 ×
1 cm). Elution with 1% CH3OH in CH2Cl2 gave N-NVOC-(S)-R-
hydrazinophenylalanine (8) and N,N′-di-NVOC-(S)-R-hydrazinophe-
nylalanine (9) as yellow foams: yield 63 mg (39%) of 8 and 54 mg
(33%) of 9. Compound 8: silica gel TLC Rf 0.12 (2:98:0.1 CH3OH-
1
TLC Rf 0.54 (25% ethyl acetate in hexane); H NMR (CDCl3) δ 3.10
(dd, 1H, J ) 14, 8 Hz), 3.22 (dd, 1H, J ) 14, 4 Hz), 3.55 (s, 3H), 3.77
(s, 3H), 5.42 (dd, 1H, J ) 9, 4 Hz), and 7.05-7.50 (m, 10H); mass
spectrum (chemical ionization) m/z 397, (M + H)+; mass spectrum
(electron impact), m/z 397.126 (C20H20O5F3 requires m/z 397.126).
1
CH2Cl2-HOAc); H NMR (CDCl3 + CD3OD) δ 2.90 (dd, 1H, J )
For comparison, compound 5 was prepared in racemic form starting
from phenylalanine; the Mosher ester of racemic 5 was synthesized. A
solution containing 20 mg (0.11 mmol) of methyl (R,S)-2-hydroxy-3-
phenylpropionate (racemic 5), 23 mg (0.11 mmol) of N,N′-dicyclo-
hexylcarbodiimide, 27 mg (0.22 mmol) of 4-(dimethylamino)pyridine,
and 33 mg (0.14 mmol) of (R)-(+)-MTPA in 0.5 mL of dry CH2Cl2
was stirred at 25 °C for 8 h. The solution was filtered to remove
precipitate, and the filtrate was concentrated under diminished pressure.
The product was purified by preparative silica gel TLC (development
with 15% ethyl acetate in hexane). The product was obtained as a
colorless oil: yield 21 mg (48%); silica gel TLC Rf 0.54 (25% ethyl
14, 7 Hz), 3.06 (dd, 1H, J ) 14, 6 Hz), 3.89 (br s, 7H), 5.43 (s, 2H),
6.99 (s, 1H), 7.21 (m, 6H) 7.65 (br s, 1H), and 7.67 (s, 1H); mass
spectrum (chemical ionization) m/z 420, (M + H)+; mass spectrum
(FABMS) m/z 420.139 (C19H22N3O8 requires m/z 420.141). Compound
9: silica gel TLC Rf 0.40 (2:98:0.1 CH3OH-CH2Cl2-HOAc); 1H NMR
(DMSO-d6) δ 2.95 (m, 1H), 3.12 (m, 1H), 3.69 (br s, 3H), 3.78 (br s,
3H), 3.81 (br s, 6H), 4.87 (br s, 1H), 5.41 (m, 4H), 6.50 (br s, 1H),
6.91 (s, 1H), 6.95 (s, 1H), 7.22 (m, 5H) and 7.64 (s, 2H); mass spectrum
(chemical ionization) m/z 659, (M + H)+; mass spectrum (FAB) m/z
658.171 (C29H30N4O14 requires m/z 658.176).
N-NVOC-(S)-r-hydrazinophenylalanine Cyanomethyl Ester (10).
A solution containing 22 mg (53 µmol) of N-NVOC-(S)-R-hydrazi-
nophenylalanine (8) in 200 µL of 1:1 triethylamine-CH3CN was cooled
in an ice bath and treated under argon with 13 µL (16 mg; 210 µmol)
of ClCH2CN. The solution was stirred at 0 °C for 30 min and then at
25 °C for 15 h. The reaction mixture was diluted with 5 mL of CH2-
Cl2 and washed with two 10-mL portions of 1 N aqueous Na2HSO4
and two 10-mL portions of water. The organic layer was dried over
Na2SO4 and concentrated under diminished pressure. The residue was
purified by preparative silica gel TLC (development with 1:1 ethyl
acetate-hexane) to give N-NVOC-(S)-R-hydrazinophenylalanine cya-
nomethyl ester (10) as a yellow foam: yield 17 mg (71%); silica gel
TLC Rf 0.55 (1:1 ethyl acetate-hexane); 1H NMR (CDCl3) δ 3.00 (dd,
1H, J ) 14, 8 Hz), 3.11 (dd, 1H, J ) 14, 6 Hz), 3.69 (br s, 1H), 3.95
(s, 3H), 3.97 (s, 3H), 4.07 (m, 1H), 4.73 (s, 2H), 5.50 (s, 2H), 6.47 (br
s, 1H), 6.93 (s, 1H), 7.21-7.35 (m, 5H), and 7.70 (s, 1H); mass
spectrum (chemical ionization) m/z 459, (M + H)+; mass spectrum
(FAB) m/z 459.147 (C21H23N4O8 requires m/z 459.152).
1
acetate in hexane); H NMR (CDCl3) δ 3.11 (dd, 1H, J ) 14, 9 Hz),
3.22 (dd, 0.5H, J ) 14, 4 Hz), 3.33 (dd, 0.5H, J ) 14, 4 Hz), 3.34 (s,
1.5H), 3.56 (s, 1.5H), 3.77 (s, 3H), 5.42 (m, 1H), and 7.04-7.50 (m,
10H); mass spectrum (chemical ionization) m/z 397, (M + H)+; mass
spectrum (electron impact) m/z 397.126 (C20H20O5F3 requires m/z
397.126).
Methyl (S)-[2-(N-tert-butoxycarbonyl)hydrazino]-3-phenylpro-
pionate (6). To a cold (0-5 °C) solution of 1.44 g (8 mmol) of methyl
(R)-2-hydroxy-3-phenylpropionate (5) in 5 mL of CH2Cl2 under argon
was added 2.0 mL (3.39 g, 12 mmol) of trifluoromethanesulfonic
anhydride followed by 2.0 mL (1.84 g, 17 mmol) of 2,6-lutidine. The
solution was stirred at 0 °C under argon for 1.5 h. To the solution
was added 2.11 g (16 mmol) of t-BOC-hydrazine. The reaction was
stirred at 0 °C for 4 h and then at 25 °C for 15 h. The solution was
diluted with CH2Cl2, extracted with water and brine, dried over Na2-
SO4, and evaporated to dryness. The crude product was purified by
flash chromatography on a silica gel column (12 × 4 cm). Elution
with 15% ethyl acetate in hexane provided methyl (S)-2-[(N-tert-
butoxycarbonyl)hydrazino]-3-phenylpropionate (6) as colorless micro-
N,N′-di-NVOC-(S)-r-hydrazinophenylalanine Cyanomethyl Es-
ter (11). A solution containing 71 mg (0.17 mmol) of N,N′-di-NVOC-
(S)-R-hydrazinophenylalanine (9) in 400 µL of 1:1 triethylamine-
CH3CN was cooled in an ice bath and treated under argon with 54 µL
(64 mg; 0.85 mmol) of ClCH2CN. The solution was stirred at 0 °C
for 1 h and then at 25 °C for 15 h. The reaction mixture was diluted
with 30 mL of CH2Cl2 and washed with two 20-mL portions of 1 N
aqueous Na2HSO4 and two 20-mL portions of water. The organic layer
was dried over Na2SO4 and concentrated to dryness. The residue was
purified by flash chromatography on a silica gel column (12 × 1 cm).
Elution with 2:3 ethyl acetate-hexane gave N,N′-di-NVOC-(S)-R-
hydrazinophenylalanine cyanomethyl ester (11) as a yellow foam: yield
crystals: yield 1.40 g (60%); mp 52-54 °C; silica gel TLC Rf 0.80
1
(1:1 ethyl acetate-hexane); [R]25 -12.8° (c 1.19, CHCl3); H NMR
D
(CDCl3) δ 1.40 (s, 9H), 2.96 (dd, 1H, J ) 14, 7 Hz), 3.06 (dd, 1H, J
) 14, 5 Hz), 3.69 (s, 3H), 3.96 (m, 1H), 4.18 (br s, 1H), 6.17 (br s,
1H), and 7.20-7.30 (m, 5H); mass spectrum (chemical ionization) m/z
295 (M + H)+. Anal. Calcd for C15H22O4N2: C, 61.21; H, 7.53.
Found: C, 61.23; H, 7.56.
(S)-r-Hydrazinophenylalanine (7).29 To a cold (0-5 °C) solution
containing 300 mg (1.02 mmol) of methyl (S)-2-[(N-tert-butoxycar-
bonyl)hydrazino]-3-phenylpropionate (6) in 1.6 mL of dry CH2Cl2 under
argon was added 0.4 mL of trifluoroacetic acid. The solution was
stirred under argon at 0 °C for 2 h and then concentrated under
diminished pressure. The residue was triturated with hexane and dried
overnight under vacuum and then dissolved in 1.5 mL of CH3OH and
cooled in an ice bath. To the cold (0-5 °C) solution was added 4.0
mL (4 mmol) of 1 N aqueous NaOH. The solution was stirred at 0 °C
for 1 h and then at 25 °C for 1 h. The solution was concentrated under
diminished pressure, and the resulting precipitate was filtered out and
washed with cold water to give (S)-hydrazinophenylalanine (7) as
1
47 mg (61%); silica gel TLC Rf 0.20 (2:3 ethyl acetate-hexane); H
NMR (CDCl3) 3.14-3.55 (m, 2H), 3.86 (s, 3H), 3.89 (s, 3H), 3.94 (s,
(28) Dupont, V.; Lecoq, A.; Mangeot, J.-P.; Aubry, A.; Boussard, G.;
Marraud, M. J. Am. Chem. Soc. 1993, 115, 8898.
(29) (a) Darapsky, A. J. Prakt. Chem. 1971, 96, 251. (b) Takamura, N.;
Yamada, S.-I. Chem. Pharm. Bull. 1976, 24, 800. (c) Trimble, L. A.;
Vederas, J. C. J. Am. Chem. Soc. 1986, 108, 6397. (d) Gennari, C.; Colombo,
L.; Bertolini, G. J. Am. Chem. Soc. 1986, 108, 6394. (e) Evans, D. A.;
Britton, T. C.; Dorow, R. L.; Dellaria, J. F., Jr. Tetrahedron 1988, 44, 5525.
(f) Evans, D. A.; Britton, T. C.; Dorow, R. L.; Dellaria, J. F. J. Am. Chem.
Soc. 1986, 108, 6395.

3040 J. Am. Chem. Soc., Vol. 120, No. 13, 1998
Killian et al.
6H), 4.01 (m, 1H), 4.58-4.80 (m, 2H), 5.30-5.70 (m, 4H), 6.85 (br s,
2H), 6.95 (br s, 1H), 7.12-7.40 (m, 5H) and 7.69 (br s, 2H); mass
spectrum (chemical ionization) m/z 698, (M + H)+; mass spectrum
(FAB) m/z 697.190 (C31H31O14N5 requires m/z 697.187).
(m, 3H), 4.90 (br s, 1H), 5.32 (m, 1H), 5.50 (br s, 1H), 6.36 (br s, 1H),
7.13-7.43 (m, 14H), 7.55 (d, 2H), and 7.76 (d, 2H); mass spectrum
(chemical ionization, isobutane) m/z 662, (M - H)+, 590, and 500;
mass spectrum (electron impact) m/z 663.297 (C39H41N3O7 requires m/z
663.295).
N-NVOC-(S)-r-hydrazinophenylalanyl-pdCpA (12). A solution
containing 5.0 mg (11 µmol) of N-NVOC-(S)-R-hydrazinophenylalanine
cyanomethyl ester (10) and 2.75 µmol of tris(tetrabutylammonium)
pdCpA18 in 50 µL of dry DMF was stirred under argon at room
temperature for 3.5 h. The reaction mixture was treated with 0.95 mL
of 1:2 CH3CN-50 mM NH4OAc, pH 4.5. The crude product was
purified by C18 reverse phase HPLC (100 × 4.6 mm column). The
column was washed with 1 f 56% CH3CN in 50 mM NH4OAc, pH
4.5, over a period of 40 min at a flow rate of 1.0 mL/min (monitoring
at 260 nm). The fractions containing the desired product (retention
time 21.1 min) were combined and lyophilized to afford N-NVOC-
(S)-R-hydrazinophenylalanyl-pdCpA (12) as a colorless solid: yield
17.2 A260 units (33%); mass spectrum (FAB) m/z 1037.236 (M+)
(C38H45N11O20P2 requires m/z 1037.232).
N,N′-di-NVOC-(S)-r-hydrazinophenylalanyl-pdCpA (13). A so-
lution containing 6.5 mg (9.3 µmol) of N,N′-di-NVOC-(S)-R-hydrazi-
nophenylalanine cyanomethyl ester (11) and 2.75 µmol of tris-
(tetrabutylammonium) pdCpA18 in 50 µL of dry DMF was stirred under
argon at room temperature for 10 h. The reaction was quenched by
the addition of 0.95 mL of 1:2 CH3CN-50 mM NH4OAc, pH 4.5, and
the crude product was purified by C18 reverse phase HPLC (100 × 4.6
mm column). The column was washed with 1 f 64% CH3CN in 50
mM NH4OAc, pH 4.5, over a period of 45 min at a flow rate of 1.0
mL/min (monitoring at 260 nm). The fractions containing the desired
product (retention time 27.0 min) were combined and lyophilized to
afford N,N′-di-NVOC-(S)-R-hydrazinophenylalanyl-pdCpA (13) as a
colorless solid: yield 20.4 A260 units (30%); mass spectrum (LSIMS)
m/z 1275, (M - H)+; mass spectrum (FAB) m/z 1276.274 (M+)
(C48H54N12O26P2 requires m/z 1276.275).
(S)-r-Hydrazinophenylalanyl-pdCpA (14). The photolyses of
N-NVOC-(S)-R-hydrazinophenylalanyl-pdCpA (12) and N-NVOC-(S)-
phenylalanyl-pdCpA were carried out simultaneously and under the
same conditions. One A260 unit (40 nmol) of N-NVOC-(S)-R-
hydrazinophenylalanyl-pdCpA (12) and 1.0 A260 unit (40 nmol)
N-NVOC-(S)-phenylalanyl-pdCpA were each dissolved separately in
800 µL of 1 mM KOAc, pH 4.5. To each sample was added 180 µL
of a 36 mM solution (6.5 µmol) of semicarbazide hydrochloride in 1
mM KOAc, pH 4.5. The solutions were cooled in an ice bath and
irradiated with light from a high-pressure mercury lamp (1000 W). Time
points were taken at 0, 5, and 10 min by withdrawing 100-µL aliquots
of reaction solution and freezing the aliquots in dry ice. The aliquots
were analyzed by C18 reverse phase HPLC under the conditions
described above for dinucleotide 12.
HPLC analysis of the photolysis reaction containing N-NVOC-(S)-
R- hydrazinophenylalanyl-pdCpA (12) in the presence of semicarbazide
hydrochloride showed that the reaction proceeded with the formation
of a major product having a retention time of 14.4 min. HPLC analysis
of the photolysis reaction of the control compound N-NVOC-(S)-
phenylalanyl-pdCpA in the presence of semicarbazide hydrochloride
revealed that the deprotection proceeded as it did in the absence of
semicarbazide hydrochloride, with the formation of a single product
having a retention time of 13.5 min.
N-Acetyl-(S)-phenylalanyl-Nâ-t-BOC-Nr-(S)-r-hydrazinophen-
ylalanine Methyl Ester (17). To a solution containing 200 mg (0.30
mmol) of N-FMOC-(S)-phenylalanyl-Nâ-t-BOC-NR-(S)-R-hydrazinophe-
nylalanine methyl ester (16) and 615 mg (3.9 mmol) of N-acetoxysuc-
cinimide in 2 mL of dry DMF under argon was added successively
135 µL (138 mg; 0.90 mmol) of 1,8-diazabicyclo[5.4.0]undec-7-ene
and 84 µL (64 mg; 0.60 mmol) of triethylamine. The reaction mixture
was stirred at 25 °C for 15 h and then concentrated under diminished
pressure, diluted with 30 mL of CH2Cl2, and washed with three 50-
mL portions of brine. The organic layer was dried over Na2SO4 and
concentrated under diminished pressure. The residue was purified by
flash chromatography on a silica gel column (12 × 2 cm). Elution
with 35% ethyl acetate in hexane gave N-acetyl-(S)-phenylalanyl-Nâ-
t-BOC-NR-(S)-R-hydrazinophenylalanine methyl ester (17) as a yellow
oil: yield 58 mg (40%); silica gel TLC Rf 0.25 (1:1 ethyl acetate-
1
hexane); H NMR (CDCl3, 50 °C) δ 1.47 (s, 9H), 1.90 (br s, 3H),
3.01-3.09 (m, 4H), 3.62 (s, 3H), 5.08 (br s, 1H), 5.29 (m, 1H), 6.06
(br m, 1H), and 7.19-7.30 (m, 11H); mass spectrum (chemical
ionization, isobutane) m/z 484, (M + H)+; mass spectrum (electron
impact), m/z 484.246, (M + H)+ (C26H34N3O6 requires m/z 484.245).
Cyclo((S)-phenylalanyl-Nâ-t-BOC-Nr-(S)-r-hydrazinophenylala-
nine) (18). To a cold (0-5 °C) solution containing 75 mg (0.11 mmol)
of N-FMOC-(S)-phenylalanyl-Nâ-t-BOC-NR-(S)-R-hydrazinophenyla-
lanine methyl ester (16) in 0.5 mL of dry CH2Cl2 was added 34 µL
(29 mg; 0.34 mmol) of piperidine. The reaction mixture was stirred
at 0 °C for 2 h and then at 25 °C for 15 h. The reaction mixture was
concentrated under diminished pressure, and the residue was purified
by flash chromatography on a silica gel column (10 × 1.5 cm). Elution
with 1:1 ethyl acetate-hexane gave cyclo((S)-phenylalanyl-Nâ-t-BOC-
NR-(S)-R-hydrazinophenylalanine) (18) as a colorless foam: yield 44
mg (94%); silica gel TLC Rf 0.25 (1:1 ethyl acetate-hexane); 1H NMR
(CDCl3) δ 1.50 (s, 9H), 2.86 (dd, 1H, J ) 13, 3 Hz), 3.29 (m, 2H),
3.98 (br dd, 1H), 4.64 (m, 1H), 5.71 (br s, 1H), and 6.89-7.42 (m,
12H); mass spectrum (chemical ionization, isobutane) m/z 410, (M +
H)+, and 543; mass spectrum (electron impact) m/z 409.199 (C23H27N3O4
requires m/z 409.200).
N-Acetyl-(S)-phenylalanyl-Nâ-t-BOC-Nr-(S)-r-hydrazinopheny-
lalanine (19). To a cold (0-5 °C) solution containing 58 mg (0.12
mmol) of N-acetyl-(S)-phenylalanyl-Nâ-t-BOC-NR-(S)-R-hydrazinophe-
nylalanine methyl ester (17) in 0.5 mL of 1:19 H2O-CH3OH was added
260 µL (0.26 mmol) of 1 N aqueous NaOH. The reaction mixture
was stirred at 0 °C for 1 h and then 25 °C for 1 h. The reaction mixture
was neutralized with 10% aqueous citric acid and concentrated under
diminished pressure. The residue was dissolved in 15 mL of CH2Cl2
and extracted with three 10-mL portions of 5% aqueous NaHCO3. The
aqueous layer was then cooled in an ice bath, acidified to pH 6 with 1
N HCl, and extracted with three 10-mL portions of CH2Cl2. The
organic layer was dried over Na2SO4 and concentrated to give N-acetyl-
(S)-phenylalanyl-Nâ-t-BOC-NR-(S)-R-hydrazinophenylalanine (19) as a
colorless foam: yield 31 mg (55%); silica gel TLC Rf 0.55 (10:90:0.1
1
CH3OH-CH2Cl2-HOAc); H NMR (CDCl3, 50 °C) δ 1.43 (s, 9H),
N-FMOC-(S)-phenylalanyl-Nâ-t-BOC-Nr-(S)-r-hydrazinophen-
ylalanine Methyl Ester (16). To a solution containing 496 mg (1.2
mmol) of N-FMOC-(S)-phenylalanine acid chloride (15)22 in 2.0 mL
of CH2Cl2 under argon was added successively a solution containing
300 mg (1.02 mmol) of N-t-BOC-(S)-R-hydrazinophenylalanine methyl
ester (6) in 1.0 mL of CH2Cl2 and 3.0 mL of 10% aqueous NaHCO3.
The reaction solution was stirred at 25 °C for 45 min and then diluted
with CH2Cl2 and washed with three 30-mL portions of brine. The
organic layer was dried over Na2SO4 and concentrated under diminished
pressure. The residue was purified by flash chromatography on a silica
gel column (12 × 1.5 cm). Elution with 25% ethyl acetate in hexane
gave N-FMOC-(S)-phenylalanyl-Nâ-t-BOC-NR-(S)-R-hydrazinopheny-
lalanine methyl ester (16) as a colorless foam: yield 621 mg (92%);
silica gel TLC Rf 0.42 (30% ethyl acetate in hexane); 1H NMR (CDCl3,
50 °C) δ 1.47 (s, 9H), 2.99-3.11 (m, 4H), 3.63 (s, 3H), 4.17-4.36
1.86 (br s, 3H), 2.85-3.10 (m, 4H), 3.32 (m, 1H), 3.72 (br s, 1H),
4.94 (br s, 1H), and 7.11-7.26 (m, 11H); mass spectrum (chemical
ionization) m/z 468, (M-H)+; mass spectrum (chemical ionization) m/z
470.232, (M + H)+ (C25H32N3O6 requires m/z 470.229).
N-Acetyl-(S)-phenylalanyl-Nr-(S)-r-hydrazinophenylalanine (1).
To a cold (0-5 °C) solution of 31 mg (0.66 mmol) N-acetyl-(S)-
phenylalanyl-Nâ-t-BOC-NR-(S)-R-hydrazinophenylalanine (19) in 750
µL of dry CH2Cl2 under argon was added 250 µL of trifluoroacetic
acid. The reaction mixture was stirred at 0 °C for 2 h and then at 25
°C for 30 min. The reaction mixture was concentrated under diminished
pressure and coevaporated repeatedly with portions of benzene to afford
N-acetyl-(S)-phenylalanyl-NR-(S)-R-hydrazinophenylalanine (1) as a
colorless foam: yield 28 mg (88%); silica gel TLC Rf 0.42 (15:85:0.1
1
CH3OH-CH2Cl2-HOAc); H NMR (CDCl3, 50 °C) δ 1.87 (s, 3H),
2.85 (dd, 1H, J ) 14, 8 Hz), 3.04-3.18 (m, 2H), 3.36 (m, 1H), 5.29

Ribosome-Mediated Incorporation of HydrazinoPhe
J. Am. Chem. Soc., Vol. 120, No. 13, 1998 3041
(m, 1H), 5.56 (m, 1H), 6.65 (d, 1H, J ) 8 Hz), 7.11-7.29 (m, 10H),
and 7.77 (br s, 2H); mass spectrum (chemical ionization, isobutane)
m/z 370, (M + H)+; mass spectrum (FAB) m/z 370.175 (C20H24N3O4
requires m/z 370.177).
mmol) of (S)-phenylalanyl-Nâ-(S)-R-hydrazinophenylalanine methyl
ester (23) in 1 mL of dry DMF under argon was added 66 mg (0.42
mmol) of N-acetoxysuccinimide. The reaction mixture was stirred at
0 °C for 15 min and then at 25 °C for 2 h. The reaction mixture was
concentrated under diminished pressure, and the residue was purified
by flash chromatography on a silica gel column (12 × 1.5 cm). Elution
with 4:1:0.1 ethyl acetate-hexane-triethylamine gave N-acetyl-(S)-
phenylalanyl-Nâ-(S)-R-hydrazinophenylalanine methyl ester (24) as
colorless foam: yield 153 mg (95%); silica gel TLC Rf 0.10 (80:20:
N-Acetyl-(S)-phenylalanyl-Nr-(S)-r-hydrazinophenylalanine Meth-
yl Ester (Methyl Ester of 1). To a cold solution (0-5 °C) containing
11 mg (24 µmol) of N-acetyl-(S)-phenylalanyl-Nâ-t-BOC-NR-(S)-R-
hydrazinophenylalanine methyl ester (17) in 0.5 mL of dry dichlo-
romethane was added 100 µL of trifluoroacetic acid under argon. The
reaction mixture was stirred at 0 °C for 2 h and then at 25 °C for 30
min and then concentrated under diminished pressure. The residue
was dried under vacuum for 2 h and then dissolved in a minimum
amount of dichloromethane and purified by flash chromatography on
a silica gel column (15 × 1 cm). Elution with 4% CH3OH in CH2Cl2
gave N-acetyl-(S)-phenylalanyl-NR-(S)-R-hydrazinophenylalanine meth-
yl ester as a colorless oil: yield 11 mg (97%); silica gel TLC Rf 0.21
1
0.1 ethyl acetate-hexane-triethylamine); H NMR (CD3OD) δ 1.86
(s, 3H), 2.86 (m, 2H), 2.99 (m, 2H), 3.52 (s, 3H), 3.66 (dd, 1H, J )
14, 7 Hz), 4.55 (t, 1H, J ) 8 Hz), and 7.13-7.22 (m, 10H); mass
spectrum (chemical ionization) m/z 384, (M + H)+; mass spectrum
(electron impact) m/z 383.185 (C21H25N3O4 requires m/z 383.185).
N-Acetyl-(S)-phenylalanyl-Nâ-(S)-r-hydrazinophenylalanine (2).
To a cold (0-5 °C) solution containing 46 mg (0.12 mmol) of N-Ac-
(S)-phenylalanyl-Nâ-(S)-R-hydrazinophenylalanine methyl ester (24) in
2.5 mL of 4:1 CH3OH-H2O was added 12.5 mg (0.3 mmol) of
LiOH‚H2O. The reaction was stirred at 25 °C for 2 h. At the end of
this time, no reaction had occurred. To this solution was added 6 µL
of aqueous 0.02 M NaOH. The solution was stirred at 25 °C for 4 h,
at which time no starting material remained. The reaction mixture was
concentrated under diminished pressure, diluted with 25 mL water, and
extracted with three 20-mL portions of ethyl acetate. The aqueous
phase was then acidified to pH 5.0 with 1 N HCl and extracted with
three 30-mL portions of ethyl acetate. The ethyl acetate layer was dried
over Na2SO4 and concentrated under diminished pressure. The residue
was purified by flash silica chromatography on a silica gel column (10
× 1 cm). Elution with 8:92:0.1 CH3OH-CH2Cl2-HOAc gave
N-acetyl-(S)-phenylalanyl-Nâ-(S)-R-hydrazinophenylalanine (2) as a
colorless foam: yield 28 mg (44%); silica gel TLC Rf 0.17 (8:92:0.1
CH3OH-CH2Cl2- HOAc); 1H NMR (CD3OD) δ 1.85 (s, 3H), 2.85 (m,
2H), 2.97 (m, 2H), 3.63 (dd, 1H, J ) 6 Hz), 4.51 (dd, 1H, J ) 7 Hz),
and 7.19-7.23 (m, 10H); mass spectrum (chemical ionization) m/z 370,
(M + H)+; mass spectrum (electron impact) m/z 369.171 (C20H23N3O4
requires m/z 369.169).
1H NMR COSY Experiments To Determine Dipeptide Con-
nectivity. DFQ COSY30 experiments were performed on a Varian
Unity 1 INOVA 500 spectrometer (1H frequency was 500.018 MHz).
Complex points were collected in the F-2 dimension (1024) and in the
F-1 dimension (256). The F-1 dimension was zero-filled to 1024 points.
The adaption function used for the DFQ COSY was 45° shifted sin2
function in both dimensions. COSY experiments were phase-sensitive.
The time between scans (recycle time) for the COSY was 2 s.
HPLC Analysis of the Dipeptide Mixture Obtained by Utilizing
(S)-r-Hydrazinophenylalanyl-tRNAPhe (II) as the Acceptor for
Ribosomally Mediated Dipeptide Formation. A portion of the
ribosomal dipeptide mixture (one-sixth, 200 µL, 1045 cpm) obtained
by utilizing (S)-R-hydrazinophenylalanyl-tRNAPhe (II) as the acceptor
tRNA in an in vitro protein biosynthesis system7h,i was frozen and
concentrated under diminished pressure to remove the water and
ethanol, affording 30 µL of a cloudy solution. The dipeptide standards
(0.15 mg each) were added to this solution in 20 µL of 2:1 CH2Cl2-
CH3OH. The resulting clear solution was concentrated under dimin-
ished pressure and then treated with 40 µL of formamide. HPLC
analysis was carried out on a C18 reverse phase column (3 µ, 100 ×
4.6 mm). The column washed with 15 f 25% CH3CN in 0.1% aqueous
CF3COOH over a period of 10 min at a flow rate of 1.0 mL/min and
then with 25 f 42% CH3CN in 0.1% aqueous CF3COOH over a period
of 24 min at the same flow rate (monitoring at 260 nm).
1
(4% CH3OH in CH2Cl2); H NMR (CDCl3) δ 1.87 (s, 3H), 2.92 (dd,
1H, J ) 14, 6 Hz), 3.09 (dd, 2H, J ) 14, 8 Hz), 3.35 (m, 1H), 3.75 (s,
3H), 3.99 (br s, 2H), 5.59 (m, 2H), 6.05 (d, 1H, J ) 8 Hz), and 7.19
(m, 10H); mass spectrum (FAB) m/z 406.173, (M + Na)+ (C21H25N3O4Na
requires m/z 406.174).
N-CBz-(S)-phenylalanyl-Nâ-(S)-r-hydrazinophenylalanine Meth-
yl Ester (22). To a cold (0-5 °C) solution of 400 mg (1.36 mmol) of
methyl (S)-2-BOC-hydrazino-3-phenylpropionate (6) in 1.5 mL of CH2-
Cl2 was added 250 µL of trifluoroacetic acid. The solution was stirred
at 0 °C for 3.5 h. The solution was concentrated under diminished
pressure to give the crude trifluoroacetic acid salt of (S)-R-hydrazi-
nophenylalanine methyl ester (21) as a yellow oil. The residue was
triturated with hexane and dried under vacuum for 15 h.
To a cold (0-5 °C) solution containing 407 mg (1.36 mmol) of
N-CBz-(S)-phenylalanine (20) and 184 mg (1.36 mmol) of hydroxy-
benzotriazole in 0.8 mL of dry DMF under argon was added 280 mg
(1.36 mmol) of N,N′-dicyclohexylcarbodiimide. The solution was
stirred under argon at 0 °C for 15 min. To the reaction solution was
added a solution containing 264 mg (1.36 mmol) of the trifluoroacetic
acid salt of (S)-R-hydrazinophenylalanine methyl ester and 380 µL (275
mg; 2.72 mmol) of triethylamine in 1 mL of dry DMF. The reaction
mixture was stirred at 0 °C for 2 h and then concentrated under
diminished pressure. The residue was dissolved in 30 mL of CH2Cl2
and filtered, and then the filtrate was washed with two 30-mL portions
of saturated aqueous NaHCO3, two 30-mL portions of brine, and two
30-mL portions of water. The organic layer was dried over Na2SO4
and concentrated under diminished pressure. The residue was purified
by flash chromatography on a silica gel column (12 × 1.5 cm). Elution
with 35:65:0.1 ethyl acetate-hexane-triethylamine gave N-CBz-(S)-
phenylalanyl-Nâ-(S)-R-hydrazinophenylalanine methyl ester (22) as
colorless microcrystals: yield 346 mg (54%); mp 121-124 °C; silica
gel TLC Rf 0.47 (1:1 ethyl acetate-hexane); 1H NMR (CD3OD) δ 2.86
(dd, 2H, J ) 16, 6 Hz), 2.99 (dd, 2H, J ) 13, 7 Hz), 3.53 (s, 3H), 3.64
(t, 1H, J ) 7 Hz), 4.27 (t, 1H, J ) 8 Hz), 4.99 (s, 2H) and 7.14-7.23
(m, 15H); mass spectrum (chemical ionization) m/z 476, (M + H)+,
and 342; mass spectrum (chemical ionization) m/z 475.210 (M+)
(C27H29N3O5 requires m/z 475.211).
(S)-Phenylalanyl-Nâ-(S)-r-hydrazinophenylalanine Methyl Ester
(23). A solution containing 250 mg (0.53 mmol) of N-CBz-(S)-
phenylalanyl-Nâ-(S)-R-hydrazinophenylalanine methyl ester (22) and
50 mg of 10% palladium-on-carbon in 2 mL of dry DMF was stirred
under an atmosphere of H2 at 25 °C for 2.5 h. The catalyst was
removed by filtration, and the filtrate was concentrated under diminished
pressure. The residue was purified by flash silica chromatography on
a silica gel column (12 × 1.5 cm). Elution with 5% CH3OH in CH2-
Cl2 gave (S)-phenylalanyl-Nâ-(S)-R-hydrazinophenylalanine methyl ester
(23) as a colorless oil: yield 149 mg (83%); silica gel TLC Rf 0.20
Fractions were collected in scintillation vials once every 1 min and
then once every 10 s over the dipeptide elution range. Scintillation
fluid (5 mL) was added to each sample, and the samples were analyzed
for tritium content. The profile of elution vs time was compared to
the UV profile for the experiment. A comparison of the two profiles
showed that compounds containing radioactivity coeluted with the
authentic dipeptide standards: 345 cpm for the dipeptide of normal
connectivity (1) having a retention time of 17.7 min (33%) and 408
1
(5% CH3OH in CH2Cl2); H NMR (CD3OD) δ 2.77 (dd, 2H, J ) 13,
7 Hz), 2.89 (dd, 2H, J ) 12, 7 Hz), 3.44 (dd, 1H, J ) 7 Hz), 3.54 (s,
3H), 3.62 (t, 1H, J ) 7 Hz), and 7.14-7.26 (m, 10H); mass spectrum
(chemical ionization) m/z 342, (M + H)+; mass spectrum (FAB) m/z
364.164, (M + Na)+ (C19H23N3O3Na requires 364.164).
N-Acetyl-(S)-phenylalanyl-Nâ-(S)-r-hydrazinophenylalanine Meth-
yl Ester (24). To a cold (0-5 °C) solution containing 143 mg (0.42
(30) Piantini, U.; Sorensen, O. W.; Ernst, R. R. J. Am. Chem. Soc. 1982,
104, 6800.

3042 J. Am. Chem. Soc., Vol. 120, No. 13, 1998
Killian et al.
cpm for the dipeptide of altered connectivity (2) having a retention
time of 19.2 min (39%). These yields are based on the theoretical
yield of dipeptide determined from a nitrocellulose filter binding assay
that showed the extent of peptidyl-tRNA binding to the ribosomal
P-site.7i
HPLC Analysis of the Dipeptide Mixture Obtained by Utilizing
(S)-Phenylalanyl-tRNAPhe (I) as the Acceptor for Ribosomally
Mediated Dipeptide Formation. A portion of the ribosomal dipeptide
solution obtained by utilizing phenylalanyl-tRNAPhe as the acceptor
tRNA in the dipeptide assay (150 µL, 1335 cpm) was frozen,
concentrated under diminished pressure to 40 µL, mixed with unlabeled
synthetic standard (0.1 mg), and analyzed by C18 reverse phase HPLC
as described above. The fractions were analyzed for tritium content,
and a graph of elution time vs radioactivity was constructed. The UV
profile matched the radioactivity profile, with the majority of radio-
activity coeluting with the authentic dipeptide standard at a retention
time of 20.7 min.
Preparation of Misacylated tRNAs. Chemical misacylation reac-
tions were carried out in 100 µL (total volume) of 50 mM K Hepes,
pH 7.5, containing 0.5 mM ATP, 15 mM MgCl2, 50 µg suppressor
tRNA-COH, 0.5 A260 unit NVOC-protected aminoacyl-pdCpA (5-10-
fold molar excess), 20% dimethyl sulfoxide, and 100 units of T4 RNA
ligase (Promega). After incubation at 37 °C for 30 min, the reaction
was quenched by the addition of 0.1 volume of 2 M sodium acetate,
pH 4.5, and the tRNA was precipitated with 2.5 volumes of ethanol.
Deprotection of NVOC-containing aminoacyl-tRNAs was carried out
in 5 µL (total volume) of 1 mM potassium acetate, pH 4.5. The
aminoacyl-tRNAs were cooled to 1 °C and irradiated with a 500 W
mercury-xenon lamp using both Pyrex and water filters. Typically,
monosubstituted NVOC aminoacyl-tRNAs were deblocked for 5 min.
(S)-R-Hydrazinophenylalanyl-tRNAs were deblocked in the presence
of 200 equiv of semicarbazide hydrochloride.
Dihydrofolate reductase was synthesized in a reaction mixture (28
µL final volume) that contained 2 µg of plasmid DNA (pTHD10,
containing the gene for DHFR with a TAG codon at position 10)
dissolved in diethyl pyrocarbonate-treated water, 10 µL of premix (35
mM Tris-acetate, pH 7.0, 190 mM potassium glutamate, 30 mM
ammonium acetate, 2 mM dithiothreitol, 11 mM magnesium acetate,
0.8 mg/mL E. coli tRNA, 2 mM isopropyl-â-D-thiogalactopyranoside,
35 mg/mL poly(ethylene glycol) 6000, 20 µg/mL folinic acid, 20 mM
ATP and GTP, 5 mM CTP and UTP, and 10 mM cAMP),33 100 µM
of amino acids minus methionine, 10 µCi of [35S]-(S)-methionine, and
7.5 µL of (S)-30 extract. Suppression reactions (100 µL) contained
50 µg of deblocked misacylated suppressor tRNA dissolved in 10 µL
of diethylpyrocarbonate-treated water and were incubated at 37 °C for
2 h. Protein was precipitated from the reaction mixtures by the addition
of 5 volumes of acetone, followed by centrifugation. The protein pellets
were dried and resuspended in sodium dodecyl sulfate loading buffer
prior to polyacrylamide gel electrophoresis.34 Dried gels were visual-
ized and analyzed using a Molecular Dynamics Phosphorimager.
Acknowledgment. We thank Drs. Vladimir Karginov,
Serguei Golovine, and Guy Zuber for preliminary experiments
that employed hydrazinophenylalanyl-tRNA in protein biosyn-
thesizing systems and Dr. Jeffrey Ellena for assistance with
NMR measurements. This work was supported by Research
Grant GM43328 from the National Institues of Health and by
Grant BIO-96-008 from Virginia’s Center for Innovative
Technology.
JA974066E
(31) Pratt, J. M. In Transcription and Translation: A Practical Approach,
IRL Press: Oxford, U.K., 1984; pp 179-209.
(32) Short, G. F., III; Golovine, S. Y.; Hecht, S. M. Manuscript in
preparation.
(33) Lesley, S. A.; Brow, M. D.; Burgess, R. R. J. Biol. Chem. 1991,
266, 2632.
In Vitro Synthesis of Dihydrofolate Reductase. An (S)-30 extract
was prepared from an E. coli strain (XAC-RF) containing a temperature
sensitive release factor 1. The (S)-30 extract was prepared according
to Pratt31 with some modifications.32
(34) Laemmli, U. K. Nature 1970, 227, 680.