Synthesis of pCpCpA-3′-NH-phenylalanine as a ribosomal substrate

Article abstract of DOI:10.1021/ol026560f

(equation presented) The trinucleotide cytidylyl(3′→5′phosphoryl)cytidylyl(3′- 5′phosphoryl)-3′-deoxy-3′-(L-phenylalanyl) amido adenosine (CpCpA-NH-Phe) was synthesized by phosphoramidite chemistry from 3′-amino-3′-deoxyadenosine as the ribosomal substrat

Full text of DOI:10.1021/ol026560f

ORGANIC
LETTERS
2002
Vol. 4, No. 21
3615-3618
Synthesis of
pCpCpA-3′-NH-Phenylalanine as a
Ribosomal Substrate
Biliang Zhang*, Lei Zhang, Lele Sun, and Zhiyong Cui
Program in Molecular Medicine, UniVersity of Massachusetts Medical School,
373 Plantation Street, Worcester, Massachusetts 01605
biliang.zhang@umassmed.edu
Received July 18, 2002
ABSTRACT
The trinucleotide cytidylyl(3′f5′phosphoryl)cytidylyl(3′f5′phosphoryl)-3′-deoxy-3′-(L-phenylalanyl) amido adenosine (CpCpA-NH-Phe) was
synthesized by phosphoramidite chemistry from 3′-amino-3′-deoxyadenosine as the ribosomal substrate. The 3′-amino-3′-deoxyadenosine
was first converted to 3′-(N-tert-butyloxycarbonyl-L-phenylalanine)amido-3′-deoxy-6-N,6-N,2′-O-tribenzoyl-adenosine and then coupled with cytidine
phosphoramidite to produce the fully protected CpCpA-NH-Phe-Boc. The title product was obtained after removing all protection groups and
then radiolabeled with 32P to yield p*CpCpA-NH-Phe, which demonstrated high activity for the peptidyl transferase reaction in the ribosome.
The 2′(3′)-O-aminoacyl-pCpCpA derivatives are the univer-
sally conserved terminal sequences of aminoacyl-tRNA and
potential substrates for ribosomal peptidyl transferase.1
Evidence from cross-linking and chemical footprinting
experiments has suggested specific tRNA-rRNA interac-
tions, in which the universally conserved CCA of the 3′-
end of tRNA and its attached aminoacyl moiety are involved
in the interactions between tRNA and 23S rRNA.2 Recently,
X-ray crystal structures of the ribosome suggested that the
23S ribosomal RNA is the peptidyl transferase, but its
mechanism is still unclear.3 Biochemical studies are impor-
tant in understanding the mechanism of the peptidyl trans-
ferase reactions, which might require 2′(3′)-O-aminoacylated
oligonucleotides or tRNAs. Most assays of rRNA function
involve in vitro reconstitution of 50S subunits. The peptidyl
transferase activity is measured by either “fragment reaction”4
or “poly (Phe) synthesis”.5 In these assays, the reaction
products are analyzed by high-voltage paper electrophoresis
or by quantitating the radioactivity in the ethyl acetate-
extracted fraction through scintillation counting. In fragment
reactions of the ribosome, CAACCA-(fMet), CCA-(fMet),
or even CA-fMet can replace the P-site tRNA, while
(3) (a) Ban, N.; Nissen, P.; Hansen, J.; Capel, M.; Moore, P. B.; Steitz,
T. A. Nature 1999, 400, 841-847. (b) Ban, N.; Nissen, P.; Hansen, J.;
Moore, P. B.; Steitz, T. A. Science 2000, 289, 905-920. (c) Cate, J. H.;
Yusupov, M. M.; Yusupova, G. Z.; Earnest, T. N.; Noller, H. F. Science
1999, 285, 2095-2104. (d) Nissen, P.; Hansen, J.; Ban, N.; Moore, P. B.;
Steitz, T. A. Science 2000, 289, 920-930. (e) Polacek, N.; Gaynor, M.;
Yassin, A.; Mankin, A. S. Nature 2001, 411, 498-501. (f) Muth, G. W.;
Chen. L.; Kosek, A.; Strobel, S. RNA 2001, 7, 1403-1415. (g) Moore, P.
B.; Steitz, T. A. Nature 2002, 418, 229-235.
(4) Monro, R. E.; Cerna, J.; Marker, K. A. Proc. Natl. Acad. Sci. U.S.A.
1968, 61, 1042-1049.
(5) Fahnestock, S.; Erdmann, V.; Normura, M. Method Enzymol. 1974,
30, 554-562.
(1) (a) Chladek, S.; Sprinzl, M. Angew. Chem., Int. Ed. Engl. 1985, 24,
371-391. (b) Bhuta, A.; Quiggle, K.; Ott, T.; Ringer, D.; Chladek, S.
Biochemistry 1981, 20, 8-15.
(2) (a) Moazed, D.; Noller, H. F. Cell 1989, 57, 585-597 (b) Porse, B.
T.; Thi-Ngoc, H. P.; Garrett, R. A. J. Mol. Biol. 1996, 264, 472-483. (c)
Saarma, U.; Spahn, C. M. T.; Nierhaus, K. H.; Remme, J. RNA 1998, 4,
189-194.
10.1021/ol026560f CCC: $22.00 © 2002 American Chemical Society
Published on Web 09/25/2002

puromycin serves as an aminoacyl acceptor.4,6 At the A site,
CCA-Phe and CACCA-Phe revealed similar binding activi-
ties.7 Therefore, only nucleotides of the CCA terminus play
an important role in the binding of aminoacyl-tRNA to the
A site. In most fragment reactions, puromycin was used as
a peptidyl acceptor and did not exhibit full acceptor activity
but required addition of alcohol. Recently, Starck and Roberts
reported that puromycin-oligonucleotides revealed steric
restrictions for ribosome entry and multiple modes of
translation inhibition.8a It was reported that tRNA bearing
3′-amino-3′-deoxyadenosine in the final sequence retained
full acceptor activity.8b Thus, the synthesized pCpCpA-NH-
Phe might likely exhibit full activity for the peptidyl
transferase reaction in the ribosome. Here, we report the first
synthesis of the new compound of CpCpA-3′-deoxy-3′-N-
phenylalanine and the enzymatic activity of 32pCpCpA-NH-
Phe.
Scheme 1. Synthesis of
2′-O-Benzoyl-N,N-dibenzoyl-3′-deoxy-3′-N-(Boc-phenylalanyl)
Adenosine with 5′-TBDPS Protectiona
The target compound was synthesized from 3′-amino-3′-
deoxy-adenosine 1 by phosphoramidite chemistry. We have
modified Robins’ nine-step route for the synthesis of
3′-amino-3′-deoxyadenosine 1 into a seven-step process with
55% overall yield.9 Briefly, adenosine was protected with
tert-butyldiphenylsilyl at the 5′-position and then treated with
R-acetoxyisobutyryl bromide to yield 2′-O-acetyl-3′-bromo-
3′-deoxy-5′-O-tert-butyldiphenylsilyl-adenosine. This was
treated with 0.5 N ammonia in methanol and then reacted
with benzylisocyanate to yield 3′-(benzylamino)-5′-O-(tert-
butyl)diphenylsilyl-3′-N,2′-O-carbonyl-3′-deoxyadenosine. This
product was reacted with sodium hydride and then with 1.0
N NaOH and finally deprotected by hydrogenation with
Pd-C (10%) to yield 3′-deoxy-3′-amino-3′-deoxyadenosine
1. Since compound 1 contains multiple functional groups,
finding a suitable protection group for each functional group
is critical toward the synthesis of CpCpA-NH-Phe.
Boc-L-phenylalanine was first introduced into the 3′-
position of 3′-amino-3′-deoxyadenosine 1, which also acted
as a protection group of the 3′-amino group. The reaction in
DMF was not successful because of the poor solubility of 2
in DMF and a high racemization of L-phenylalanine. When
N-(tert-butyloxycarbonyl)-L-phenylalanine N-hydroxy suc-
cinimide ester was stirred with 3′-amino-3′-deoxyadenosine
1 in anhydrous dimethyl sulfoxide (DMSO) at room tem-
perature for 4 h, an optically pure 3′-(N-tert-butyloxycarbo-
nyl-L-phenylalanine)amido-3′-deoxyadenosine 2 was ob-
tained in 95% yield.10 The tert-butyl-diphenylsilyl (TBDPS)
group was first used to protect the 5′-hydroxyl group of 2 to
yield 5′-O-(tert-butyl-diphenylsilyl)-3′-(N-tert-butyloxycar-
bonyl-L-phenylalanine) amido-3′-deoxy-adenosine 3 in 85%
yield (Scheme 1). This reaction was highly regioselective
a Reaction conditions: (a) Boc-Phe-NHS ester, DMSO, rt; (b)
TBDPS-Cl, pyridine, 2 days; (c) benzoyl chloride, pyridine, 0-25
°C; (d) TBAF, THF, 0-25 °C.
to form the 5′-protected compound. After benzoylation, the
fully protected 3′-amino-3′-deoxyadenosine (98%) was treated
with tert-butylammonium fluoride (TBAF) in anhydrous
THF to remove the 5′-O-TBDPS group. Unfortunately, the
deprotection of TBDPS produced the desired compound 5
(57%) as well as 6-N-monobenzoyl compound 6 (13%) and
6-N,N-2′-O,5′-O-tetrabenzoyl compound 7 (8%), which was
formed via benzoyl migration.
We have also investigated the DMTr protection of the 5′-
hydroxy of 2. Reaction of 2 with 4,4′-dimethoxyltrityl
chloride in dry pyridine yielded a mixture of components
(Scheme 2). The mixture was separated by a flash column
of silica gel to give the desired product 3′-(N-tert-butyloxy-
carbonyl-L-phenylalanine)amido-3′-deoxy-5′-O-(4,4′-dimeth-
oxytrityl)-adenosine 10 in 47% yield,11 unreacted 2 (14%),
3′-(N-tert-butyloxycarbonyl-L-phenylalanine)amido-3′-deoxy-
2′-O-(4,4′-dimethoxytrityl)-adenosine 8, and 2′-O,5′-O-bis-
(4,4′-dimethoxytrityl)-3′-(N-tert-butyloxycarbonyl-L-phen-
ylalanine)amido-3′-deoxyadenosine 9. Compounds 8 and 9
were treated with 80% acetic acid at room temperature to
regenerate the starting material 2 without the cleavage of
(6) (a) Cerna, J.; Rychlik, I.; Krayevsky, A. A.; Gottikh, B. P. Febs Lett.
1973, 37, 188-191. (b) Moazed, D.; Noller, H. F. Cell 1989, 57, 585-
597. (c) Mercer, T. F. B.; Symons, R. H. Eur. J. Biochem. 1972, 28, 38-
45. (d) Parlato, G.; Guesnet, J.; Crechet, J.-B.; Parmeggiani, A. FEBS Lett.
1981, 125, 257-260.
(7) (a) Lessard, J. L.; Pestka, S. J. Biol. Chem. 1972, 247, 6901-8. (b)
Bhuta, P.; Kumar, G.; Chladek, S. Biochim Biophys Acta 1982, 696, 208-
11.
(8) (a) Starck, S. R.; Roberts, R. W. RNA 2002, 8, 890-903. (b) Fraser,
T. H.; Rich, A. Proc. Natl. Acad. Sci. U.S.A. 1973, 70, 2671-5.
(9) (a) Samano, M. C.; Robins, M. J. Tetrahedron Lett. 1989, 30, 2329-
2332. (b) Zhang, L.; Zhang, B. Unpublished results.
(10) Compound 2: TLC (85:15 chloroform/methanol), Rf ) 0.52; 1H
NMR (DMSO-d6) δ 8.39 (s, 1H), 8.14 (s, 1H), 8.02 (d, J ) 7.7 Hz, 1H),
7.32 (s, 2H), 7.28-7.15 (m, 5H), 6.93 (d, J ) 8.4 Hz, 1H), 6.04 (d, J )
3.3 Hz, 1H), 5.94 (d, J ) 2.9 Hz, 1H), 5.17 (t, J ) 5.5 Hz, 1H), 4.51 (br,
1H), 4.46 (m, 1H), 4.25 (m, 1H), 3.90 (m, 1H), 3.66-3.43 (dm, 2H), 2.98-
2.71 (dm, 2H), 1.28 (s, 9H); 13C NMR (DMSO-d6) δ 168.1, 152.2, 151.3,
148.7, 145.0, 135.3, 134.2, 125.4, 124.1, 122.3, 115.2, 85.4, 79.7, 74.2,
69.1, 57.2, 51.9, 46.5, 33.9, 24.2, 21.2. ESI-MS (m/z) calcd for C24H31N7O6
513.2, found 536.3 [M + Na]+.
3616
Org. Lett., Vol. 4, No. 21, 2002

The synthesis of 5′-HO-CpCpA-3′-N-Phe 17 depicted in
Scheme 3 was based upon a similar strategy of phosphor-
Scheme 2. Synthesis of
2′-O-Benzoyl-N,N-dibenzoyl-3′-deoxy-3′-N-(Boc-phenylalanyl)
Adenosine with 5′-DMTr Protection
Scheme 3. Synthesis of Cytidylyl(3′f5′Phosphoryl)
Cytidylyl(3′f5′Phosphoryl)-3′-deoxy-3′-(L-phenylalanyl) Amido
Adenosine
a Reaction conditions: (a) DMTr-Cl, pyridine; (b) benzoyl
chloride, pyridine; (c) 80% acetic acid.
N-Boc group of phenylalanine (30% yield). On the basis of
the recovery of the starting material, the calculated yield of
10 was 84%.
Benzoylation of 10 with a large excess (6 equiv) of
benzoyl chloride produced 3′-(N-tert-butyloxycarbonyl-L-
phenylalanine)amido-3′-deoxy-5′-O-(4,4′-dimethoxytrityl)-6-
N,6-N,2′-O-tribenzoyl-adenosine 11 in 96% yield.12 Depro-
tection of the DMTr group with 80% acetic acid at room
temperature afforded the desired compound 5 in 80% yield.13
a Reaction conditions: (a) (i) 1H-tetrazole, MeCN; (ii) t-Bu-
OOH; (iii) 80% AcOH, rt. (b) (i) 12, 1H-tetrazole, MeCN; (ii) t-Bu-
OOH. (c) 80% AcOH, rt. (d) 3:1 NH3 (aq)/ethanol. (e) (i) TFA;
(ii) TEA‚3HF, 1-methyl-pyrollidinone, 65 °C.
(11) Compound 10: TLC (9:1 chloroform/methanol), Rf ) 0.26; 1H NMR
(CDCl3) δ 8.31 (s, 1H), 8.05 (s, 1H), 7.32-6.73 (m, 18H), 6.39 (br, 1H),
5.70 (sh, 1H), 5.65 (br, 2H), 5.15 (br, 1H), 4.84 (br, 1H), 4.47 (br, 1H),
4.32-4.22 (m, 2H), 3.77 (s, 3H), 3.76 (s, 3H), 3.45-3.35 (m, 2H), 3.09-
2.92 (m, 2H), 1.42 (s, 9H); 13C NMR (CDCl3) δ 172.5, 158.7, 156.0, 152.8,
148.8, 144.6, 138.8, 136.9, 135.9, 135.8, 130.4, 129.5, 128.9, 128.5, 128.1,
127.2, 127.1, 120.1, 113.4, 91.2, 86.7, 82.8, 80.4, 74.6, 63.4, 56.2, 55.4,
53.7, 51.5, 39.5, 28.5; ESI-MS (m/z) calcd for C45H49N7O8 815.4, found
838.3 [M + Na]+.
amidite methodology. The 5′-hydroxy group of 5 was
coupled with a commercially available cytidine phosphor-
amidite 12 in the presence of 1H-tetrazole in anhydrous
acetonitrile, oxidized by tert-butyl hydroperoxide, and then
treated with 80% acetic acid to yield 1314 in 75% yield.
Dinucleotide 13 was coupled with 12 again to yield the fully
protected trinucleotide 14.15 After flash column purification,
14 was treated by 80% acetic acid to produce 1516 in 69%
yield. Deprotection of benzoyl and cyanoethyl groups with
ammonium 3:1 hydroxide/EtOH at 55 °C for 24 h afforded
(12) Compound 11: TLC (9:1 chloroform/methanol), Rf ) 0.62; 1H NMR
(CDCl3) δ 8.61 (s, 1H), 8.26 (s, 1H), 7.92-6.78 (m, 28H), 6.20 (d, J ) 2.0
Hz, 1H), 5.92 (d, J ) 8.6 Hz, 1H), 5.84 (dd, J ) 6.2, 2.0 Hz, 1H), 5.31 (dt,
J ) 8.6, 6.0 Hz, 1H), 5.04 (d, J ) 7.4 Hz, 1H), 4.21 (m, 1H), 3.95 (br,
1H), 3.75 (s, 6H), 3.48 (m, 2H), 3.04 (m, 1H), 2.73 (m, 1H), 1.32 (s, 9H);
ESI-MS (m/z) calcd for C66H61N7O11 1127.4, found 1150.4 [M + Na]+;
HRMS (m/z) calcd 1127.4429, found 1128.4376 [M + H]+.
(13) Compound 5: TLC (9:1 chloroform/methanol), Rf ) 0.52; 1H NMR
(CDCl3) δ 8.68 (s, 1H), 8.43 (s, 1H), 7.97-7.18 (m, 20H), 6.48 (d, J ) 3.3
Hz, 1H), 6.22 (d, J ) 2.7 Hz, 1H), 5.75 (dd, J ) 6.1, 2.9 Hz, 1H), 5.10 (dt,
J ) 7.0, 6.6 Hz, 1H), 4.95 (br, 1H), 4.33 (dt, J ) 7.4, 7.0 Hz, 1H), 4.23
(br, 1H), 4.11 (m, 1H), 4.02 (m, 1H), 3.76 (m, 1H), 3.05 (m, 2H), 1.32 (s,
9H); 13C NMR (CDCl3) δ 172.6, 172.5, 165.3, 152.6, 152.5, 152.3, 144.2,
136.7, 134.5, 134.2, 133.4, 130.2, 129.8, 129.4, 129.2, 129.1, 129.0, 128.5,
128.4, 127.5, 89.4, 84.8, 81.0, 76.7, 61.6, 56.3, 49.7, 38.2, 28.4; ESI-MS
(m/z) calcd for C45H43N7O9 825.3, found 848.3 [M + Na]+; HRMS (m/z)
calcd 825.3211, found 826.3189 [M + H]+.
(14) Compound 13: 1H NMR (CDCl3) δ 8.68 (s, 1H), 8.33 (s, 1H), 8.27-
7.14 (m, 25H), 6.24 (br, 1H), 5.86 (br 1H), 5.60 (s, 1H), 5.15-3.70 (m,
13H), 3.02 (m, 2H), 2.70 (br, 2H), 1.20 (s, 9H), 0.88 (s, 9H), 0.12 (s, 3H),
0.10 (s, 3H); 31P NMR (CDCl3) δ -0.7, -0.9; ESI-MS (m/z) calcd for
C70H76N11O17PSi 1401.5, found 1424.1 (M + Na)+.
Org. Lett., Vol. 4, No. 21, 2002
3617

trinucleotide 16 in 89% yield. The N-Boc group of 16 was
removed by stirring with trifluoroacetic acid at 0 °C for 30
min.
After removal of trifluoroacetic acid, the solid compound
was heated with a mixture of triethylamine/triethylamine
trifluoric acid/1-methyl-pyrollidinone (v/v/v ) 3:4:6) at 65
°C for 1.5 h. The final product was purified by a C18
reversed-phase column eluted with a gradient of water and
methanol to give a white solid 1717 in 61% yield from 15.
The overall yield from 5 was 28%.
Labeling the 5′-end of trinucleotide 17 was accomplished
through phosphorylation using T4 polynucleotide kinase
(PNK) and [γ-32P]ATP. The 5′-end labeled p32CpCpA-NH-
Phe was used as a peptidyl acceptor for the peptidyl
transferase reactions in the ribosome. These reactions were
monitored by polyacrylamide gel electrophoresis (PAGE)
(Figure 1). No product was formed in the absence of tRNA-
ribosome under the same reaction buffer, a similar product
band was formed at an approximate 90% yield at the
reaction’s final extent (lanes 8-13, Figure 1). After treatment
of streptavidin, the product band migrated much slower (the
band not shown in Figure 1, lane 7), indicating that a biotin
moiety was transferred to *pCpCpA-NH-Phe. This suggested
that a peptide bond was formed between pCpCpA-NH-Phe
and tRNA-O-Met-biotin. These results demonstrated that
pCpCpA-NH-Phe is fully active for the peptidyl transferase
reaction in the ribosome. It has been well-known that
puromycin is a universal protein synthesis inhibitor by
covalent attachment to the nascent chain of peptide. Recently,
a report8a suggested that puromycin was shown to be more
complicated than thought previously and, in some cases, to
not form peptide bonds. Therefore, pCpCpA-NH-Phe is a
new compound and should be a useful peptidyl acceptor for
studying the peptide bond formation in the ribosome.
Acknowledgment. This research was partially supported
by a startup fund from the University of Massachusetts
Medical School.
Supporting Information Available: Experimental details
and NMR spectra. This material is available free of charge
via the Internet at http://pubs.acs.org.
OL026560F
(15) Compound 14: 1H NMR (CDCl3) δ 9.95-9.94 (br, 1H), 8.96-
6.88 (m, 47H), 6.39-3.54 (m, 24H), 3.81 (s, 6H), 3.76-3.54 (m, 4H), 3.05-
2.84 (m, 2H), 2.74-2.53 (m, 4H), 1.11-0.83 (m, 27H), 0.20-0.02 (m,
12H); 13C NMR (CDCl3) 173.2, 172.9, 172.5, 172.4, 167.1, 165.7, 163.2,
159.0, 156.1, 155.7, 155.4, 152.5, 152.1, 144.5, 144.1, 137.3, 135.1, 135.0,
134.1, 133.2 (m), 130.4 (m), 129.6, 128.9, 128.4 (m), 127.6, 126.9, 116.8,
116.7, 113.6, 113.5, 97.7, 97.3, 90.1-86.8 (m), 82.2, 81.5, 80.4-79.8 (m),
75.9-72.9 (m), 68.3, 67.5, 66.4, 62.7 (m), 59.7, 55.4, 50.9, 40.4, 29.8, 28.0,
25.8, 19.7, 18.2, -4.6 (m); 31P NMR (CDCl3) δ 0.7, -1.9 (m); ESI-MS
(m/z) calcd for C116H127N15O25P2Si2 2279.8, found 2303.4 [M + Na]+.
(16) Compound 15: 1H NMR (CDCl3) δ 9.96-9.94 (br, 1H), 8.68-
7.12 (m, 40H), 6.39-3.85 (m, 23H), 3.04-2.88 (m, 2H), 2.74-2.48 (m,
4H), 1.10 (m, 9H), 0.90-0.80 (m, 18H), 0.10-0.01 (m, 12H); 13C NMR
(CD3OD) δ 172.8, 172.5, 172.4, 167.2, 165.6, 163.5, 163.0, 156.3, 155.7,
152.6, 152.1, 146.1, 144.6, 137.1, 134.1, 133.9, 133.3, 130.2-127.8 (m),
127.0, 116.9, 116.8, 97.4, 91.8, 91.1, 89.1, 88.7, 83.1, 80.5, 78.9, 76.0,
75.1-73.2 (m), 68.4-65.5 (m), 65.9, 62.9, 60.6, 60.1, 59.4, 55.6, 50.4,
39.5, 28.1, 25.8, 19.7, 18.2, -4.6, -4.9 (m); 31P NMR (CDCl3) δ 0.2, -1.3
(m); HRMS (m/z) calcd for C95H110N15O25P2Si2 1978.6839 [M + H], found
1978.6713.
Figure 1. Autoradiogram of the peptidyl transferase reaction of
2′(3′)-biotin-methionyl-tRNA with 5′-32pCpCpA-NH-Phe in the
presence of E. coli 70S ribosome or S30 extract in the reaction
buffer [20 mM Tris‚HCl (pH 7.4), 6 µM Spermidine, 400 mM
NH4Cl, 4 mM MgCl2, 330 mM KCl]. Samples were run on 7.5 M
urea/20% polyacrylamide gel with 1 × TBE buffer at 30 W. The
bottom bands are 32pCpCpA-NH-Phe, and the top bands are
32pCpCpA-NH-Phe-Met-biotin.
O-Met-biotin substrate in E. coli S30 extract (Promega) with
p*CpCpA-NH-Phe (lanes 1-3, Figure 1). When p*CpCpA-
NH-Phe was incubated with 50 µM tRNA-O-Met-biotin in
the presence of 1.0 U of E. coli S30 extract (lanes 4-7,
Figure 1) under the reaction buffer, a new band was formed.
When p*CpCpA-NH-Phe was incubated with 50 µM tRNA-
O-Met-biotin in the presence of 1.0 U of E. coli 70S
(17) Compound 17: 1H NMR (D2O) δ 8.43 (s, 1H), 8.23 (s, 1H), 7.93
(d, J ) 8.0 Hz, 1H), 7.86 (d, J ) 8.0 Hz, 1H), 7.46-7.28 (m, 5H), 5.98 (d,
J ) 3.2 Hz, 1H), 5.94 (d, J ) 8.0 Hz, 1H), 5.85 (d, J ) 8.0 Hz, 1H), 5.74
(m, 2H), 4.61-3.80 (m, 16H), 3.31-3.15 (m, 2H); 31P NMR (D2O) δ 0.2,
0.0; HRMS (m/z) calcd for C37H48N13O18P2 1024.2716 [M + H], found
1024.2969.
3618
Org. Lett., Vol. 4, No. 21, 2002