Design of an adenosine analogue that selectively improves the affinity of a mutant U1A protein for RNA

Article abstract of DOI:10.1021/ja021267w

The RNA recognition motif (RRM), one of the most common RNA binding domains, contains three highly conserved aromatic amino acids that participate in stacking interactions with RNA bases. We have investigated the contribution of these highly conserved aro

Full text of DOI:10.1021/ja021267w

Published on Web 02/07/2003
Design of an Adenosine Analogue that Selectively Improves
the Affinity of a Mutant U1A Protein for RNA
Ying Zhao and Anne M. Baranger*
Contribution from the Department of Chemistry, Wesleyan UniVersity,
Middletown, Connecticut 06459
Received October 9, 2002 ; E-mail: abaranger@wesleyan.edu
Abstract: The RNA recognition motif (RRM), one of the most common RNA binding domains, contains
three highly conserved aromatic amino acids that participate in stacking interactions with RNA bases. We
have investigated the contribution of these highly conserved aromatic amino acids to the affinity of the
complex formed between the N-terminal RRM of the U1A protein and stem loop 2 of U1 snRNA. Previously,
we found that substitution of one of these conserved aromatic amino acids, Phe56, with Ala resulted in a
large destabilization of the complex. Here, we have modified A6, the base in stem loop 2 RNA that stacks
with Phe56, to compensate for a portion of the destabilization caused by the Phe56Ala mutation. We have
designed two modified adenosines, A-3CPh and A-4CPh, in which a phenyl group is linked to the adenosine
such that it may replace the phenyl group that is eliminated by the Phe56Ala mutation in the complex. We
have found that incorporation of A-3CPh into stem loop 2 RNA stabilizes the complex formed with Phe56Ala
by 0.6 kcal/mol, while incorporation of A-4CPh into stem loop 2 RNA stabilizes this complex by 1.8 kcal/
mol. Either base modification destabilizes the wild-type complex by 0.8-0.9 kcal/mol. Experiments with
other U1A mutant proteins suggest that the stabilization of the complex between the Phe56Ala U1A protein
and stem loop 2 RNA is due to a specific interaction between the Phe56Ala U1A protein and A6-4CPh
stem loop 2 RNA.
Introduction
enables selective binding of diverse single-stranded target sites
and makes the RRM one of the most general RNA-binding
The RNA recognition motif (RRM), also known as the
ribonucleoprotein (RNP) domain or the RNA binding domain
(RBD), is one of the largest families of RNA binding domains
and is found in proteins that participate in almost every step of
gene expression.1-5 The target sites of RRMs are single-stranded
RNA oligonucleotides that vary in sequence, structure, and
flexibility. The RRM is comprised of an antiparallel â-sheet
flanked by two R-helices.6 The most conserved amino acids of
the RRM that contact RNA are found in the central two strands
of the â-sheet and contribute primarily to the nonspecific
recognition of RNA.3 Target site specificity is provided by the
variable regions of the RRM and the cooperation of mul-
tiple RRMs in the same protein.7 The modularity of the RRM
scaffolds.
An understanding of the recognition principles that enable
nonspecific and specific target site recognition in RRM-RNA
complexes is important to describe the biological processes that
involve RRM-RNA complexes, to develop small molecules
that can specifically modulate RRM-RNA binding, and to mod-
ify or redesign RRM-RNA complexes rationally. An effective
method for discovering and testing recognition and design
principles in biomolecular complexes is to recover the binding
energy lost from an initial destabilizing modification by re-
designing either component of the complex.8-14 Recently, this
approach has been used to engineer protein-ligand complexes
in order to probe and control biological pathways selectively.15-18
In this paper, we describe the design of a modified RNA
nucleoside to compensate for the destabilization caused by the
(1) Rubin, G. M.; Yandell, M. D.; Wortman, J. R.; Miklos, G. L. G.; Nelson,
C. R.; Hariharan, I. K.; Fortini, M. E.; Li, P. W.; Apweiler, R.; Fleischmann,
W.; Cherry, J. M.; Henikoff, S.; Skupski, M. P.; Misra, S.; Ashburner, M.;
Birney, E.; Boguski, M. S.; Brody, T.; Brokstein, P.; Celniker, S. E.;
Chervitz, S. A.; Coates, D.; Cravchik, A.; Gabrielian, A.; Galle, R. F.;
Gelbart, W. M.; George, R. A.; Goldstein, L. S. B.; Gong, F.; Guan, P.;
Harris, N. L.; Hay, B. A.; Hoskins, R. A.; Li, J.; Li, Z.; Hynes, R. O.;
Jones, S. J. M.; Kuehl, P. M.; Lemaitre, B.; Littleton, J. T.; Morrison, D.
K.; Mungall, C.; O’Farrell, P. H.; Pickeral, O. K.; Shue, C.; Vosshall, L.
B.; Zhang, J.; Zhao, Q.; Zheng, X. H.; Zhong, F.; Zhong, W.; Gibbs, R.;
Venter, J. C.; Adams, M. D.; Lewis, S. Science 2000, 287, 2204-2215.
(2) Lorkovic, Z. J.; Barta, A. Nucleic Acids Res. 2002, 30, 623-635.
(3) Birney, E.; Kumar, S.; Krainer, A. R. Nucleic Acids Res. 1993, 21, 5803-
5816.
(4) Krecic, A. M.; Swanson, M. S. Curr. Opin. Cell Biol. 1999, 11, 363-371.
(5) Varani, G.; Nagai, K. Annu. ReV. Biophys. Biomol. Struct. 1998, 27, 407-
445.
(6) Burd, C. G.; Dreyfuss, G. Science 1994, 265, 615-621.
(7) Pe´rez-Can˜adillas, J.-M.; Varani, G. Curr. Opin. Struct. Biol. 2001, 11, 53-
58.
(8) Atwell, S.; Ultsch, M.; De Vos, A. M.; Wells, J. A. Science 1997, 278,
1125-1128.
(9) Beuning, P. J.; Musier-Forsyth, K. Biopolymers 1999, 52, 1-28.
(10) Strobel, S. A.; Cech, T. R. Science 1995, 267, 675-679.
(11) Carter, P. J.; Winter, G.; Wilkinson, A. J.; Fersht, A. R. Cell 1984, 38,
835-840.
(12) Miller, W. T.; Hou, Y.-M.; Schimmel, P. Biochemistry 1991, 30, 2635-
2641.
(13) GuhaThakurta, D.; Draper, D. E. Biochemistry 1999, 38, 3633-3640.
(14) Dertinger, D.; Dale, T.; Uhlenbeck, O. C. J. Mol. Biol. 2001, 314, 649-
654.
(15) Specht, K. M.; Shokat, K. M. Curr. Opin. Cell Biol. 2002, 14, 155-159.
(16) Koh, J. T. Chem. Biol. 2002, 9, 17-23.
(17) Doyle, D. F.; Mangelsdorf, D. J.; Corey, D. R. Curr. Opin. Chem. Biol.
2000, 4, 60-63.
(18) Clackson, T. Curr. Opin. Struct. Biol. 1998, 8, 451-458.
9
2480
J. AM. CHEM. SOC. 2003, 125, 2480-2488
10.1021/ja021267w CCC: $25.00 © 2003 American Chemical Society

Modified Adenosines Stabilize the U1A−RNA Complex
A R T I C L E S
elimination of a highly conserved aromatic amino acid from an
RRM-RNA complex.
There are three highly conserved stacking interactions
between aromatic amino acids and RNA bases in RRM-RNA
complexes.3,19-26 Because all of the nucleic acid bases can par-
ticipate in stacking interactions, and because these interactions
are highly conserved, stacking interactions are likely to con-
tribute significantly to nonspecific RNA binding by the RRM.
Stacking interactions between aromatic amino acid side chains
and nucleic acid bases are more common in the recognition of
nonhelical nucleic acids than helical nucleic acids. For example,
stacking interactions are found in complexes formed by other
RNA binding proteins, single-stranded DNA binding proteins,
and DNA repair proteins.27-39 In these complexes, Phe partici-
pates more often in stacking interactions than Tyr or Trp.40,41
Although Phe, Trp, and Tyr stack with all four bases, there is
a preference for Phe to stack with A. In particular, the stacking
interaction between Phe and A is more common than any other
stacking interaction in RRM-RNA complexes.41 Because
stacking interactions are highly conserved in RRM-RNA
complexes, their modification may reveal general recognition
principles for the formation of RRM-RNA complexes.
We have investigated RNA recognition by the N-terminal
RRM of the U1A protein as a model for RRM-RNA com-
plexes. The U1A protein is a spliceosomal protein that binds to
stem loop 2 of U1 snRNA with high affinity and specificity.42,43
Although the U1A protein contains two RRMs, only the
N-terminal RRM binds to RNA.44,45 Structures of the free
N-terminal RRM and the complex formed with stem loop 2
Figure 1. (A) Crystal structure of the complex formed between the
N-terminal RRM of the U1A protein and stem loop 2 RNA.19 Only part of
stem loop 2 RNA is shown. The stacking interactions among Phe56, A6,
and C7 and between Tyr13 and C5 of stem loop 2 RNA are shown. (B)
Secondary structure of stem loop 2 RNA used in these experiments. The
nucleotides that form the binding site for the U1A protein are shown in
boldface.
Figure 2. Close-up of the stacking interactions between Phe56 of the U1A
protein and A6 and C7 of stem loop 2 RNA.19
(19) Oubridge, C.; Ito, N.; Evans, P. R.; Teo, C. H.; Nagai, K. Nature 1994,
372, 432-438.
(20) Allain, F. H.-T.; Howe, P. W. A.; Neuhaus, D.; Varani, G. EMBO J. 1997,
16, 5764-5774.
RNA have been determined by NMR and X-ray crystal-
lography.19,46 The N-terminal RRM of the U1A protein contains
two of the three conserved aromatic amino acids found in
RRMs, Phe56 and Tyr13. In the X-ray crystal structure, Phe56
stacks with A6 and C7, and Tyr13 stacks with C5 of stem loop
2 RNA (Figure 1).19 Previously, we found that Phe56 is essential
for the stability of the U1A protein-RNA complex (Figure
2).47,48 Mutation of Phe56 to any other aromatic amino acid
did not destabilize the complex significantly, but mutation to
either Leu or Ala resulted in a large decrease in binding free
energy. It is unlikely that loss of the stacking interaction alone
is responsible for the large decrease in binding affinity observed
upon substitution of Phe56 with Leu or Ala. These mutations
may change the binding free energy by altering both direct
interactions and cooperative networks of interactions involving
Phe56 in both the free U1A proteins and the complexes formed
with stem loop 2 RNA.
(21) Deo, R. C.; Bonanno, J. B.; Sonenberg, N.; Burley, S. K. Cell 1999, 98,
835-845.
(22) Price, S. R.; Evans, P. R.; Nagai, K. Nature 1998, 394, 645-650.
(23) Ding, J.; Hayashi, M. K.; Zhang, Y.; Manche, L.; Krainer, A. R.; Xu, R.
J. Genes DeV. 1999, 13, 1102-1115.
(24) Handa, N.; Nureki, O.; Kuimoto, K.; Kim, I.; Sakamoto, H.; Shimura, Y.;
Muto, Y.; Yokoyama, S. Nature 1999, 398, 579-585.
(25) Wang, X.; Tanaka Hall, T. M. Nat. Struct. Biol. 2001, 8, 141-145.
(26) Allain, F. H.-T.; Bouvet, P.; Dieckmann, T.; Feigon, J. EMBO J. 2000,
19, 6870-6881.
(27) Antson, A. A.; Dodson, E. J.; Dodson, G.; Greaves, R. B.; Chen, X.;
Gollnick, P. Nature 1999, 401, 235-242.
(28) Slupphaug, G.; Mol, C. D.; Kavli, B.; Arvai, A. S.; Krokan, H. E.; Tainer,
J. A. Nature 1996, 384, 87-92.
(29) Obmolova, G.; Ban, C.; Hsieh, P.; Yang, W. Nature 2000, 407, 703-710.
(30) Morellet, N.; De´me´ne´, H.; Teilleux, V.; Huynh-Dinh, T.; Rocquigny, H.;
Fournie´-Zaluski, M.; Roques, B. P. J. Mol. Biol. 1998, 283, 419-434.
(31) Bruner, S. D.; Norman, D. P. G.; Verdine, G. L. Nature 2000, 403, 859-
866.
(32) Cavarelli, J.; Rees, B.; Ruff, M.; Thierry, J.-C.; Moras, D. Nature 1993,
362, 181-184.
(33) Qiocho, F. A.; Hu, G.; Gershon, P. D. Curr. Opin. Struct. Biol. 2000, 10,
78-86.
(34) Raghunathan, S.; Kozlov, A. G.; Lohman, T. M.; Waksman, G. Nat. Struct.
Biol. 2000, 7, 648-652.
In this paper, we describe experiments in which the RNA
target site is altered to compensate for the loss of binding affinity
caused by the substitution of Ala for Phe56. We have developed
the modified adenosines, A-3CPh and A-4CPh (Figure 3), which
possess a tethered phenyl group that can fill the cavity left by
the Phe56Ala mutation. We predicted that these modifications
would stabilize the complex with the Phe56Ala U1A protein
(35) Folmer, R. H. A.; Nilges, M.; Papavoine, C. H. M.; Harmsen, B. J. M.;
Konings, R. N. H.; Hilbers, C. W. Biochemistry 1997, 36, 9120-9135.
(36) Bochkarev, A.; Pfeutzner, R. A.; Edwards, A. M.; Frappier, L. Nature 1997,
385, 176-181.
(37) Hillier, B. J.; Rodriguez, J. M.; Gregoret, L. M. Folding Des. 1998, 3,
87-93.
(38) Murzin, A. G. EMBO J. 1993, 12, 861-867.
(39) Bogden, C. E.; Fass, D.; Bergman, N.; Nichols, M. D.; Berger, J. M. Mol.
Cell 1999, 3, 487-493.
(40) Jones, S.; Daley, D. T. A.; Luscombe, N. M.; Berman, H. M.; Thornton,
J. M. Nucleic Acids Res. 2001, 29, 943-954.
(41) Allers, J.; Shamoo, Y. J. Mol. Biol. 2001, 311, 75-86.
(42) Tsai, D. E.; Harper, D. S.; Keene, J. D. Nucleic Acids Res. 1991, 19, 4931-
4936.
(46) Avis, J. M.; Allain, F. H.-T.; Howe, P. W. A.; Varani, G.; Nagai, K.;
Neuhaus, D. J. Mol. Biol. 1996, 257, 398-411.
(43) Hall, K. B. Biochemistry 1994, 33, 10076-10088.
(44) Scherly, D.; Boelens, W.; van Venrooij, W. J.; Dathan, N. A.; Hamm, J.;
Mattaj, I. W. EMBO J. 1989, 8, 4163-4170.
(47) Nolan, S. J.; Shiels, J. C.; Tuite, J. B.; Cecere, K. L.; Baranger, A. M. J.
Am. Chem. Soc. 1999, 121, 8951-8952.
(48) Shiels, J. C.; Tuite, J. B.; Nolan, S. J.; Baranger, A. M. Nucleic Acids Res.
2002, 30, 550-558.
(45) Lu, J.; Hall, K. B. J. Mol. Biol. 1995, 247, 739-752.
9
J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003 2481

A R T I C L E S
Zhao and Baranger
zene. A Sonogashira coupling between 2-iodoadenosine and
4-phenyl-1-butyne formed 2-(4-phenyl-1-butyn-1-yl)adenosine
(3), according to the procedures of Abiru et al.53 Complete
reduction of the alkyne by 10% palladium on carbon under
hydrogen formed the butyl-linked product 4.
Protection of Modified Adenosines for Solid-Phase RNA
Synthesis. The modified adenosines 2 and 4 were appropriately
protected for solid-phase RNA synthesis (Scheme 2). Initially
dimethylformamidine (DMF) was chosen as the 6-NH2 protect-
ing group because removal of this group from the RNA
oligonucleotide is facile.54,55 However, the DMF group was not
stable under the conditions used to introduce the tert-butyldi-
methylsilyl (TBDMS) group onto the 2′-OH, described later.
Therefore, a benzoyl protection was used for the 6-NH2 of both
modified adenosines. The 5′-OH was then protected as the
dimethoxytrityl (DMT) ether to form compounds 7 and 8.56 The
2′-OH of compound 7 was silylated under standard conditions
with tert-butyldimethylsilyl chloride in THF in the presence of
AgNO3.57 Modest selectivity for 2′-O-silylation (9, 42%) over
3′-O-silylation (11, 28%) was observed.
Figure 3. Designed adenosine analogues.
by interacting favorably with amino acids in the cavity and by
helping to prevent structural changes in the complex that occur
as a result of the Phe56Ala mutation. We find that the complex
formed with the Phe56Ala U1A protein is stabilized by both
adenosine modifications, while the complex formed with the
wild-type protein is destabilized. Thus, a residue involved in
the conserved stacking interaction can be rationally modulated
to change the relative binding affinities of the wild-type and
mutant proteins.
Results
Compound 8 was not silylated under standard conditions.
Piccirilli and co-workers reported the silylation of highly hin-
dered 2′-hydroxyl groups with either tert-BuMgCl or KH with
18-crown-6.58 We observed no silylation products from the
reaction of tert-BuMgCl with either TBDMSCl or TBDMSOTf.
However, silylation was achieved using KH/18-crown-6 and
TBDMSCl. The reaction mixture, including KH and 18-crown-
6, was cooled to 0 °C before compound 8 was added. The
mixture was then cooled to -78 °C, and the TBDMSCl solution
was added dropwise. At -78 °C, a mixture of 2′,3′-O-disilyl,
2′-O-silyl, and 3′-O-silyl nucleosides were observed initially.
The 2′,3′-O-disilyl product quickly converted to the 2′-O- or
3′-O-silyl products. The reaction was quenched, and the silyl
derivatives were purified immediately to afford the 2′-O-isomer
(10, 30%) and the 3′-O-isomer (12, 30%) as white solids. Both
isomers are stable in the solid phase and can be stored at -20
°C under N2 for weeks without isomerization.
Design of Adenosines with Tethered Phenyl Groups. We
chose 2-(4-phenylbutyl)adenosine and 2-(3-phenylpropyl)aden-
osine as our target molecules (Figure 3). Experimental work
on stacking interactions with small molecules has shown that
stacking interactions between purines and aromatic rings can
occur with linkers as short as three or four methylene groups,
although a “herringbone” orientation between the rings can be
observed with the four-methylene-group linker.49-51 We per-
formed B3LYP/6-31G ab initio geometry optimization calcula-
tions with Gaussian 9852 on adenosines linked to phenyl with
propyl and butyl groups. As expected, the adenine and the
propyl-linked phenyl group exhibited a parallel stacked orienta-
tion, while the adenine and the phenyl group linked with the
butyl group exhibited a “herringbone” orientation. The phenyl
groups were linked to the C2 position of adenine to minimize
disruption of the hydrogen bonding network between A6 and
the U1A protein.19
Preparation of the Free Nucleoside. 2-(3-Phenylpropyl)-
adenosine (2) was prepared in two steps from 2-iodoadenosine
(Scheme 1). A Heck coupling of 2-iodobenzene with allylben-
zene led to the formation of compound 1. The Heck reaction
gave the highest yield when carried out in a sealed tube with 1
equiv of palladium and 2 equiv of tri-o-tolylphosphine. Sub-
sequent hydrogenation with 10% palladium on carbon under
hydrogen gave the propyl-linked product 2. 2-(4-Phenylbutyl)-
adenosine (4) was also prepared in two steps from 2-iodoben-
The 2′-O- and 3′-O-silylated products were identified by
1H,1H NOE difference spectroscopy and 1H,1H correlated NMR
spectroscopy (1H,1H COSY). Irradiation of H(1′) of the 2′-O-
silyl isomers of 2-(3-phenylpropyl)adenosine (9) and 2-(4-
phenylbutyl)adenosine (10) gave no NOE on the 3′-OH, while
irradiation of H(4′) gave an NOE of 4.9% for 9 and 4.0% for
10 on the 3′-OH. Irradiation of H(1′) gave an NOE of 4.8% on
the 2′-OH for the 3′-O-silyl isomers of 2-(3-phenyl-1-proplyl)-
adenosine (11) and 2-(4-phenylbutyl)adenosine (12), while
irradiation of H(4′) resulted in no NOE on the 2′-OH. The
expected coupling between the 3′-H and the 3′-OH was observed
1
in the H,1H COSY spectra of the 2′-O-silyl isomers of 2-(3-
(49) Newcomb, L. F.; Gellman, S. H. J. Am. Chem. Soc. 1994, 116, 4993-
4994.
phenylpropyl)adenosine (9) and 2-(4-phenylbutyl)adenosine
(10).
The 3′-phosphoramidites of both modified adenosines (13 and
14) were prepared under standard conditions and isolated in
(50) Newcomb, L. F.; Haque, T. S.; Gellman, S. H. J. Am. Chem. Soc. 1995,
117, 6509-6519.
(51) Seyama, F.; Akahori, K.; Sakata, Y.; Misumi, S.; Aida, M.; Nagata, C. J.
Am. Chem. Soc. 1988, 110, 2192-2201.
(52) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.
A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann,
R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin,
K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi,
R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.;
Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Salvador, P.;
Dannenberg, J. J.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B.
B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin,
R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara,
A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M.
W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople,
J. A. Gaussian 98, revision A.11; Gaussian, Inc.: Pittsburgh, PA, 2001.
(53) Abiru, T.; Miyashita, T.; Watanabe, Y.; Yamaguchi, T.; Machida, H.;
Matsuda, A. J. Med. Chem. 1992, 35, 2253-60.
(54) Froehler, B. C.; Matteucci, M. D. Nucleic Acids Res. 1983, 11, 8031-
8036.
(55) Zemlicka, J.; Holy, A. Collect. Czech. Chem. Commun. 1967, 32, 3159-
3168.
(56) Schaller, H.; Weimann, G.; Lerch, B.; Khorana, H. G. J. Am. Chem. Soc.
1963, 85, 3821-3827.
(57) Hakimelahi, G. H.; Proba, Z. A. Can. J. Chem. 1982, 60, 1106-1113.
(58) Tang X. Q.; Liao, X.; Piccirilli, J. A. J. Org. Chem. 1999, 64, 747-754.
9
2482 J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003

Modified Adenosines Stabilize the U1A−RNA Complex
A R T I C L E S
Scheme 1 a
a (i) allylbenzene, Pd(OAc)2, P(o-tolyl)3, Et3N, CH3CN. (ii) 4-phenyl-1-butyne, CuI, (Ph3P)2PdCl2, Et3N, DMF. (iii) Pd/C, H2, EtOH.
Scheme 2 b
b (i) TMSCl, PhCOCl, pyridine. (ii) DMTCl, pyridine. (iii) TBDMSCl, AgNO3, THF. (iv) KH, 18-crown-6, TBDMSCl, THF. (v) (i-Pr)2NP(OCH2CH2CN)Cl,
collidine, N-methylimidazole, THF.
sufficient material was obtained to perform the experiments
described in this paper. The resulting RNA was deprotected with
NH4OH/EtOH (3:1) at 55 °C for 12 h, followed by treatment
with a solution of TEA/3HF. The RNA was purified by de-
naturing polyacrylamide gel electrophoresis. To confirm that
the RNA was fully deprotected and that the modified bases were
incorporated into the RNA, the molecular weights of the RNA
oligonucleotides were determined by MALDI mass spectrometry
and correct composition was confirmed by enzymatic hydroly-
sis.60
The effect of the tethered phenyl groups on the stability of
stem loop 2 RNA was evaluated by melting curves obtained
by UV and CD spectroscopies. The UV and CD melting profiles
of the modified stem loop 2 RNAs and the wild-type stem loop
2 RNA were similar, and the calculated melting temperatures
(Tm) were within experimental error. In addition, the CD spectra
of the modified and wild-type stem loop 2 RNAs were similar.
These data suggest that the modification of A6 does not
significantly alter the structure or the stability of stem loop 2
RNA.
Figure 4. Secondary structure of stem loop 2 RNAs containing the designed
adenosine analogues.
62-68% yield.59 Two peaks in the 31P NMR spectrum that
correspond to a pair of diastereomers with chemical shifts of
approximately 150 ppm were observed.
Synthesis and Characterization of RNA. A6 was substituted
with 2-(4-phenylbutyl)adenosine (A6-4CPh) and with 2-(3-
phenylpropyl)adenosine (A6-3CPh) in stem loop 2 RNA (Figure
4). Coupling yields of the modified nucleotides were between
50% and 90%. The coupling reactions were not optimized, but
Equilibrium Binding of U1A Proteins to A6-3CPh and
A6-4CPh Stem Loop 2 RNAs. The affinities of the two
modified stem loop 2 RNAs for the wild-type U1A protein were
measured by gel mobility shift assays (Figure 5 and Table 1).48
Examples of plots illustrating the fraction RNA bound as a
(60) Gait, M. J.; Earnshaw, D. J.; Farrow, M. A.; Fogg, J. H.; Grenfell, R. L.;
Naryshkin, N. A.; Smith, T. V. In RNA: Protein Interactions; C. W. J.
Smith, Ed.; Oxford University Press: New York, 1998; pp 1-36.
(59) Scaringe, S. A.; Francklyn, C.; Usman, N. Nucleic Acids Res. 1990, 18,
5433-5441.
9
J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003 2483

A R T I C L E S
Zhao and Baranger
Figure 5. Examples of gel mobility shift analyses of the wild-type and
Phe56Ala U1A proteins binding to wild-type and A6-4CPh stem loop 2
RNAs. In each gel, the slower moving band is the complex and the faster
moving band is the free RNA.
Figure 6. Plots illustrating the fraction RNA bound as a function of protein
concentration: wild-type U1A protein/wild-type stem loop 2 RNA complex
(b), wild-type U1A protein/A6-4CPh stem loop 2 RNA complex (9),
Phe56Ala U1A protein/wild-type stem loop 2 RNA ([), Phe56Ala U1A
protein/A6-4CPh stem loop 2 RNA complex (1).
function of protein concentration are shown in Figure 6. The
A6-3CPh and A6-4CPh stem loop 2 RNAs bound the U1A
protein with 0.9 (( 0.3) and 0.8 (( 0.2) kcal/mol less binding
energy, respectively, than the wild-type stem loop 2 RNA (Table
2). To investigate whether the appended phenyl groups re-
store binding affinity lost upon substitution of Phe56 with
Ala, we measured the affinity of the Phe56Ala U1A protein
for A6-3CPh and A6-4CPh stem loop 2 RNAs (Table 1). Both
adenosine substitutions increased binding affinity, but greater
stabilization was observed for A6-4CPh, 1.8 (( 0.4) kcal/mol
(Table 2). Previously, we reported the binding of a series of
U1A proteins mutated at Phe56 to stem loop 2 RNA.47,48 We
measured the binding affinity of the Phe56Trp and Phe56Leu
U1A proteins for A6-3CPh and A6-4CPh stem loop 2 RNAs
to probe whether the amino acid at position 56 affects the
binding of the U1A protein to the modified RNAs. The affinity
of the Phe56Trp U1A protein for the wild-type stem loop 2
RNA was nearly equivalent to that of the wild-type protein,
but the affinity of the Phe56Leu U1A protein for the wild-type
stem loop 2 RNA was 4.3 kcal/mol less than that of the wild-
type protein. The substitution of A-3CPh or A-4CPh for A6
destabilized the Phe56Trp U1A protein-stem loop 2 RNA
complex by 0.5 (( 0.4) and 0.8 (( 0.3) kcal/mol, respectively.
This destabilization is within experimental error of that observed
when A-3CPh or A-4CPh was substituted for A6 in the wild-
type complex (Table 2). In contrast, the incorporation of either
A-3CPh or A-4CPh in stem loop 2 RNA improved binding of
the Phe56Leu U1A protein by 0.4 (( 0.2) and 0.3 (( 0.2) kcal/
mol, respectively.
disturb the structure surrounding A6 and any interactions that
depend on this structure. One possibility is that the hydrogen
bonding network formed between A6 and the U1A protein is
destabilized, even though the phenyl groups have been linked
to C2 of A6 in order to minimize disruption of these hydrogen
bonds. Our previous work has shown that binding is sensitive
to conservative modifications of both the RNA and the protein
at this position, which suggests that any changes in the geometry
of the interaction between A6 and amino acids in the U1A
protein caused by the base modifications will destabilize the
complex.47,48,61 As a result of the sensitivity of the complex to
modifications of A6, the wild-type U1A protein is selective for
the wild-type stem loop 2 RNA over A6-3CPh or A6-4CPh stem
loop 2 RNAs.
In contrast to the wild-type protein, the Phe56Ala U1A protein
binds with higher affinity to A6-3CPh and A6-4CPh stem loop
2 RNAs than to the wild-type stem loop 2 RNA. A6-3CPh and
A6-4CPh could improve the binding affinity of the Phe56Ala
U1A protein for stem loop 2 RNA by stabilizing the complex
or destabilizing the free RNA. However, any changes in the
free energy of the free RNA will also affect binding of the wild-
type U1A protein. Therefore, changes in the free energies of
the complexes must be responsible for the observed destabiliza-
tion of the wild-type complex and stabilization of the Phe56Ala
U1A protein-stem loop 2 RNA complex by the A-4CPh and
A-3CPh substitutions. A-3CPh and A-4CPh could stabilize the
complex between the Phe56Ala U1A protein and stem loop 2
RNA by participating in favorable interactions in the cavity
formed by the Phe56Ala mutation and by minimizing structural
changes that disrupt other binding interactions in the complex.
Binding would also be favored by placement of the hydrophobic
phenyl group and linker in the cavity formed by the Phe56Ala
mutation. These mechanisms of stabilization depend on the
complexes formed with Phe56Ala and wild-type U1A proteins
having similar structures. Although we do not know the structure
of the complex between the Phe56Ala U1A protein and stem
loop 2 RNA, molecular dynamics simulations and binding
experiments with stem loop 2 RNAs containing modified
adenosines at position 6 have suggested that large structural
Discussion
The modified adenosines A-3CPh and A-4CPh were designed
to stabilize the complex formed with the Phe56Ala U1A protein
but not the wild-type complex. In fact, the wild-type complex
is destabilized by both adenosine modifications. This destabi-
lization could result from free energy changes in either the free
RNA or the complex. Although the thermodynamic melts of
A6-4CPh and A6-3CPh stem loop 2 RNAs showed that their
stability was not affected by the modified adenosines, these
experiments primarily probe duplex stability and may not have
revealed changes in the stability or dynamics of the loop that
could affect binding to the U1A protein. In the complex with
the wild-type U1A protein, the adenine modifications may
(61) Tuite, J. B.; Shiels, J. C.; Baranger, A. M. Nucleic Acids Res. 2002, 30,
5269-5275.
9
2484 J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003

Modified Adenosines Stabilize the U1A−RNA Complex
A R T I C L E S
Table 1. Binding Affinities of Wild-Type and Mutant U1A Proteins for Wild-Type and Modified Stem Loop 2 RNAs
RNA
protein
wild-type
A6-3CPh
A6-4CPh
wild-type
Kd (M)
2.1 ((0.8) × 10-10
8.9 ((5.4) × 10-10
8.2 ((4.7) × 10-10
∆G° (kcal/mol)a
Kd (M)
-13.2 ( 0.2
-12.3 ( 0.3
-12.4 ( 0.3
Phe56Trp
Phe56Leu
Phe56Ala
6.3 ((3.2) × 10-10
1.7 ((0.8) × 10-9
2.5 ((1.4) × 10-9
∆G° (kcal/mol)a
Kd (M)
-12.5 ( 0.3
-12.0 ( 0.3
-11.7 ( 0.5
3.2 ((0.9) × 10-7
1.5 ((0.5) × 10-7
1.9 ((1.1) × 10-7
∆G° (kcal/mol)a
Kd (M)
-8.9 ( 0.2
-9.3 ( 0.2
-9.2 ( 0.3
2.4 ((2.1) × 10-5
-6.3 ( 0.5
8.2 ((5.0) × 10-6
-6.9 ( 0.3
1.2 ((0.7) × 10-6
-8.1 ( 0.4
∆G° (kcal/mol)a
a ∆G° is the free energy of association of the complex.
Table 2. Comparison of the Destabilization Energy (∆∆G°)
Resulting from Each Phenyl-Tethered Adenine in Complexes
Formed with the Wild-Type and Mutant U1A Proteins
kcal/mol by the substitution of 4-methylindole for A6. Phe56Leu
and Phe56Ala U1A proteins bound with 1.8 and 1.0 kcal/mol
higher affinity, respectively, to the stem loop 2 RNA containing
4-methylindole than to the wild-type stem loop 2 RNA. These
experiments suggested that the hydrophobicity of 4-methylindole
contributed to the energetics of the interaction of stem loop 2
RNA with the U1A protein and that this effect was most
important for complexes formed with the Phe56Leu U1A
protein. In contrast, we find that the incorporation of A-4CPh
into stem loop 2 RNA is most stabilizing for the complex formed
with the Phe56Ala U1A protein, not the Phe56Leu U1A protein.
These data support the proposal that the favorable interaction
of the Phe56Ala U1A protein with A6-4CPh stem loop 2 RNA
is due to a specific interaction of this modified adenosine with
the Phe56Ala U1A protein.
RNA
A6-3CPh
∆∆G° (kcal/mol)a,b
A6-4CPh
∆∆G° (kcal/mol)a,b
protein
wild-type
Phe56Trp
Phe56Leu
Phe56Ala
0.9 ( 0.3
0.5 ( 0.4
-0.4 ( 0.2
-0.6 ( 0.2
0.8 ( 0.2
0.8 ( 0.3
-0.3 ( 0.2
-1.8 ( 0.4
a ∆∆G° is the difference in binding free energies between the complex
indicated in the table and the complex formed with the wild-type stem loop
2 RNA. b The individual ∆∆G° values were calculated from binding assays
performed simultaneously with the same set of protein dilutions for the
two RNAs being compared.
changes of the complex do not occur upon substitution of Phe56
with Ala.48,62
In conclusion, we have developed a novel modified adenosine,
A-4CPh, that improves the binding affinity of Phe56Ala U1A
protein for stem loop 2 RNA, but destabilizes the complex
formed with the wild-type protein. It is unlikely that the
improvement in binding is due to a nonspecific hydrophobic
effect because incorporation of A-C3Ph into stem loop 2 RNA
does not lead to as great an improvement in binding affinity
and the complex formed between the Phe56Leu U1A protein
and stem loop 2 RNA is not as stabilized by either base
substitution. When the experiments described in this paper and
our previously reported results are taken together,61 we have
developed two methods to specifically improve the stability of
complexes of stem loop 2 RNA with individual U1A mutant
proteins, while destabilizing the complex with the wild-type
U1A protein. Substitution of A-4CPh for A6 in stem loop 2
RNA stabilizes the complex with the Phe56Ala U1A protein,
while substitution of 4-methylindole for A6 in stem loop 2 RNA
stabilizes the complex with the Phe56Leu U1A protein. These
results show that the RNA base can be modified so that the
relative ability of wild-type and mutant proteins to recognize
RNA is altered. Since Phe-A stacking interactions are highly
conserved in RRM-RNA complexes and are also found in
many other protein-single-stranded nucleic acid complexes,27-39,41
the adenosine analogues A-4CPh and 4-methylindole may also
be used to alter the recognition interfaces of other protein-
nucleic acid complexes. More generally, these experiments
demonstrate that protein-RNA interfaces can be reengineered
to alter specificity,14 which may enable the extension of the
powerful applications of protein-ligand reengineering15-18 to
protein-RNA complexes.
If the increase in stabilization of the complex formed between
the Phe56Ala U1A protein and stem loop 2 RNA is due to a
specific, favorable interaction between A-3CPh or A-4CPh and
the Phe56Ala U1A protein, then the incorporation of A-3CPh
and A-4CPh into the A6 position of stem loop 2 RNA should
not improve the binding affinity of other U1A proteins mutated
at position 56, such as the Phe56Trp and Phe56Leu U1A
proteins. As expected, the substitution of A-3CPh and A-4CPh
for A6 destabilizes the complexes formed between stem loop 2
RNA and the Phe56Trp and wild-type U1A proteins nearly
equally. In contrast, the complex between the Phe56Leu U1A
protein and stem loop 2 RNA was stabilized by both adenosine
substitutions. However, the increase in binding affinity observed
when A-4CPh was incorporated into the complex formed with
the Phe56Ala U1A protein was significantly greater than the
increase in binding affinity observed when either modified
adenosine was incorporated into the complex formed with the
Phe56Leu U1A protein. These results suggest that recognition
of A6-4CPh stem loop 2 RNA by the Phe56Ala U1A protein is
distinct from the recognition of A6-3CPh stem loop 2 RNA by
the Phe56Ala U1A protein or the recognition of either modified
stem loop 2 RNA by the Phe56Leu U1A protein.
Favorable interactions between the hydrophobic tethered
phenyl groups and Leu may contribute to the greater affinity
of A6-3CPh and A6-4CPh stem loop 2 RNAs for the Phe56Leu
U1A protein. Previously, we substituted 4-methylindole for A6
in stem loop 2 RNA and measured the ability of the wild-type
and Phe56 mutant U1A proteins to bind to this modified stem
loop 2 RNA.61 In these experiments, the affinity of the wild-
type U1A protein for stem loop 2 RNA was decreased by 2
Experimental Procedures
General. Commercial solvents and reagents were used as received
unless otherwise noted. Before use, acetonitrile and pyridine were
(62) Blakaj, D. M.; McConnell, K. J.; Beveridge, D. L.; Baranger, A. M. J.
Am. Chem. Soc. 2001, 123, 2548-2551.
9
J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003 2485

A R T I C L E S
Zhao and Baranger
distilled from calcium hydride, THF was distilled from sodium and
benzophenone, and DMF was distilled from CaO and stored over
activated 4 Å molecular sieves. Flash chromatography was carried out
with Silicycle Ultrapure silica gel 60 (230-400 mesh). Prep TLC was
carried out on glass backed 1000 µm 60 Å silica gel with an F254
indicator (Analtech). Analytical TLC was carried out on glass backed
250 µm silica Gel GF (Analtech). Mass spectra were performed by the
Mass Spectrometry Facility at the University of Illinois at Urbana-
4.64 (app dd, J ) 11.5, 6.5 Hz, 1H, 2′-H), 4.13 (m, 1H, 3′-H), 3.97
(m, 1H, 4′-H), 3.51-3.69 (m, 2H, 5′-H, 5′′-H), 2.66 (t, J ) 8.2 Hz,
2H, CH2A), 2.62 (t, J ) 7.6 Hz, 2H, CH2Ph), 1.71 (m, 2H, CH2CH2A),
1.59 (m, 2H, CH2CH2Ph); HRMS(FAB) calcd for C20H25N5O4
[MH+], 400.198 480; found, 400.198 400.
General Procedure for Benzoylation of 6-NH2. Free nucleoside 2
or 4 (approximately 1.2 mmol) was dried by coevaporation with 3 ×
6 mL of dry pyridine and then suspended in dry pyridine (9.1 mL). To
the suspension was added trimethylsilyl chloride (1.17 mL, 9.22 mmol).
After the mixture was stirred for 2 h at room temperature, it was cooled
to 0 °C and benzoyl chloride (0.42 mL, 3.6 mmol) was added dropwise
over 10 min. After the addition was complete, the mixture was stirred
at 0 °C for 5 min, warmed to room temperature, and stirred for an
additional 2 h. The reaction was quenched by the addition of water (3
mL) at 0 °C. The reaction mixture was stirred for 5 min at room
temperature, and then concentrated aqueous ammonia (3 mL) was
added. After the mixture was stirred for 15 min, it was poured into
water (50 mL) and extracted with CH2Cl2 (5 × 15 mL). The organic
phase was dried over Na2SO4, filtered, and concentrated under reduced
pressure. The residue was purified by flash chromatography using a
step gradient from 100% EtOAc to 10% MeOH/EtOAc.
1
Champaign. H NMR, 31P{1H}, COSY, and NOE difference spectra
were obtained on 300 and 500 MHz spectrometers. 1H NMR chemical
shifts are reported in δ (ppm) in reference to residual proton signals in
the deuterated solvent. 31P{1H} chemical shifts are reported in δ (ppm)
relative to an external standard of 85% H3PO4. All exchangeable protons
were detected by the addition of D2O. 2-Iodoadenosine and 2-(4-phenyl-
1-butyn-1-yl)adenosine (3) were prepared according to literature
procedures.53,63-68 The synthesis and purification of the N-terminal RRM
of the U1A protein, amino acids 2-102, and the Phe56Ala, Phe56Leu,
and Phe56Trp U1A mutant proteins has been reported previously.48
2-(Allylbenzene)adenosine (1). A solution of 2-iodoadenosine (0.243
g, 0.62 mmol), palladium (II) acetate (0.142 g, 0.63 mmol), tri-o-
tolyphosphine (0.387 g, 1.27 mmol), triethylamine (0.24 mL, 1.7 mmol),
and allylbenzene (0.80 mL, 6.0 mmol) in freshly distilled acetonitrile
(10 mL) was placed in a dry sealed tube and purged with N2 for 5
min. The reaction mixture was stirred in an 80 °C oil bath overnight.
After TLC showed that the 2-iodoadenosine was completely consumed,
the mixture was cooled to room temperature, filtered through a Celite
pad, washed with CH2Cl2 (200 mL), and concentrated under reduced
pressure. The residue was purified by preparatory TLC (10% MeOH/
CHCl3) to give compound 1 as a light yellow solid (0.092 g, 53%). 1H
NMR (DMSO-d6) (mixture of two isomers) δ 8.29 and 8.30 (s, 1H,
H8), 7.22-7.43 (m, 8H, HCdCHCH2, Ph, NH2), 6.53 (m, 1H, HCd
CHCH2), 5.88 (d, J ) 6.7 Hz, 1H, 1′-H), 5.63 (dd, J ) 7.7, 4.1 Hz,
1H, 5′-OH), 5.45 (m, 1H, 2′-OH), 5.21 (m, 1H, 3′-OH), 4.66 (m, 1H,
2′-H), 4.15 (m, 1H, 3′-H), 3.98 (m, 1H, 4′-H), 3.54-3.72 (m, 4H, 5′-
H, 5′′-H, CH2Ph). MS(ES) calcd for C19H21N5O4 [MH+], 384.16; found,
384.13.
N6-Benzoyl-2-(3-phenylpropyl)adenosine (5). Reaction of com-
1
pound 2 (0.424 g, 1.10 mmol) gave a white foam (0.365 g, 68%). H
NMR (DMSO-d6) δ 11.16 (s, 1H, NHCO), 8.64 (s, 1H, H8), 8.04 (d,
J ) 7.1 Hz, 2H, PhCO), 7.51-7.65 (m, 3H, PhCO), 7.15-7.31 (m,
5H, Ph), 6.03 (d, J ) 6.1 Hz, 1H, 1′-H), 5.56 (d, J ) 6.1 Hz, 1H,
2′-OH), 5.28 (d, J ) 4.7 Hz, 1H, 3′-OH), 5.20 (m, 1H, 5′-OH), 4.70
(app dd, J ) 11.3, 6.9 Hz, 1H, 2-′H), 4.21 (m, 1H, 3′-H), 4.01 (m, 1H,
4′-H), 3.57-3.75 (m, 2H, 5′-H, 5′′-H), 2.91 (t, J ) 7.7 Hz, 2H, CH2A),
2.68 (t, J ) 7.6 Hz, 2H, CH2Ph), 2.09 (m, 2H, CH2). HRMS(FAB)
calcd for C26H27N5O5 [MH+], 490.209 044; found, 490.208 900.
N6-Benzoyl-2-(4-phenylbutyl)adenosine (6). Reaction of compound
1
4 (0.485 g, 1.21 mmol) gave a light yellow foam (0.510 g, 83%). H
NMR (DMSO-d6) δ 11.16 (s, 1H, NHCO); 8.61 (s, 1H, H8), 8.03 (d,
J ) 7.6 Hz, 2H, PhCO), 7.52-7.67 (m, 3H, PhCO), 7.13-7.29 (m,
5H, Ph), 5.99 (d, J ) 6.4 Hz, 1H, 1′-H), 5.54 (d, J ) 6.1 Hz, 1H,
2′-OH), 5.27 (d, J ) 4.7 Hz, 1H, 3′-OH), 5.20 (m, 1H, 5′-OH), 4.69
(m, 1H, 2′-H), 4.19 (m, 1H, 3′-H), 3.98 (m, 1H, 4′H), 3.53-3.71 (m,
2H, 5′-H, 5′′-H), 2.92 (t, J ) 7.4 Hz, 2H, CH2A), 2.63 (t, J ) 7.4 Hz,
2H, CH2Ph), 1.81 (m, 2H, CH2CH2A), 1.64 (m, 2H, CH2CH2Ph).
HRMS(FAB) calcd for C27H29N5O5 [MH+], 504.224 694; found,
504.224 800.
General Procedure for Hydrogenation. A suspension of compound
1 or 3 (approximately 1 mmol) and 10% Pd/C (0.120 g) in EtOH (50
mL) was stirred under H2 at atmospheric pressure and room temperature
for 24-48 h until the starting material was completely hydrogenated
by NMR. The reaction mixture was filtered through a Celite pad and
washed with EtOH (100 mL). The filtrate and washings were
concentrated to dryness under reduced pressure and purified by flash
chromatography (10% MeOH/CHCl3).
General Procedure for DMT Protection of 5′-OH. Compound 5
or 6 (approximately 1 mmol) was dried by coevaporation with 3 × 5
mL of dry pyridine and suspended in dry pyridine (4.7 mL). To the
suspension was added 4,4′-dimethoxytrityl chloride (0.378 g, 1.11
mmol). The mixture was stirred under N2 atmosphere overnight. After
TLC (25% hexanes/EtOAc) showed the reaction was complete, it was
quenched by the addition of MeOH (3 mL) and stirred for 1 h. The
reaction mixture was then concentrated to dryness under reduced
pressure, coevaporated with toluene (5 mL) to remove pyridine, and
dissolved in CH2Cl2 (50 mL). The organic phase was washed with 5%
NaHCO3 and saturated NaCl. It was then dried over Na2SO4 and
concentrated under reduced pressure. The residue was purified by flash
chromatography. The residue was loaded onto the column with 100%
EtOAc and eluted with a step gradient of 100% hexanes, 50% hexanes/
EtOAc, 100% EtOAc, and 10% MeOH/EtOAc.
2-(3-Phenylpropyl)adenosine (2). Hydrogenation of compound 1
1
(0.752 g, 1.96 mmol) gave a yellow foam (0.642 g, 85%). H NMR
(DMSO-d6) δ 8.22 (s, 1H, H8), 7.15-7.30 (m, 7H, Ph, NH2), 5.84 (d,
J ) 6.3 Hz, 1H, 1′-H), 5.64 (dd, J ) 7.8, 3.9 Hz, 1H, 5′-OH), 5.41 (d,
J ) 6.3 Hz, 1H, 2′-OH), 5.20 (d, J ) 4.2 Hz, 1H, 3′-OH), 4.64 (app
dd, J ) 11.5, 5.8 Hz, 1H, 2′-H), 4.13 (m, 1H, 3′-H), 3.97 (m, 1H,
4′-H), 3.51-3.68 (m, 2H, 5′-H, 5′′-H), 2.58-2.67 (m, 4H, CH2Ph,
CH2A), 1.99 (m, 2H, CH2). HRMS(FAB) calcd for C19H23N5O4 [MH+],
386.182 830; found, 386.182 900.
2-(4-Phenylbutyl)adenosine (4). Hydrogenation of compound 3
1
(0.425 g, 1.07 mmol) gave a white foam (0.340 g, 79%). H NMR
(DMSO-d6) δ 8.23 (s, 1H, H8), 7.12-7.28 (m, 7H, Ph, NH2), 5.83
(d, J ) 6.7 Hz, 1H, 1′-H), 5.72 (dd, J ) 8.1, 3.9 Hz, 1H, 5′-OH),
5.40 (d, J ) 6.3 Hz, 1H, 2′-OH), 5.18 (d, J ) 4.3 Hz, 1H, 3′-OH),
5′-O-(4,4′-Dimethoxytrityl)-N6-benzoyl-2-(3-phenylpropyl)adeno-
(63) Matsuda, A.; Shinozaki, M.; Suzuki, M.; Watanabe, Y.; Miyasaka, T.
Synthesis 1986, 385-386.
sine (7). Reaction of compound 5 (0.279 g, 0.57 mmol) gave a yellow
1
foam (0.384 g, 85%). H NMR (CDCl3) δ 8.78 (s, 1H, NHCO), 8.24
(64) Robins, M. J.; Uznanski, B. Can. J. Chem. 1981, 59, 2601-2607.
(65) Gerster, J. F.; Jones, J. W.; Robins, R. K. J. Org. Chem. 1963, 28, 945-
948.
(s, 1H, H8), 8.02 (d, 2H, J ) 7.8 Hz, PhCO), 7.51-7.62 (m, 3H, PhCO),
6.74-7.31 (m, 18H, Ph), 6.29 (broad s, 1H, OH), 5.98 (d, J ) 6.0 Hz,
1H, 1′-H), 4.80 (m, 1H, 2′-H), 4.48 (m, 1H, 3′-H), 4.41 (m, 1H, 4′-H),
3.76 (s, 6H, OCH3), 3.24-3.47 (m, 2H, 5′-H, 5′′-H), 3.05 (t, J ) 7.5
Hz, 2H, CH2A, overlapping with a singlet, 1H, OH), 2.74 (t, J ) 7.5
(66) Nair, V.; Turner, G. A.; Buenger, G. S.; Chamberlain, S. D. J. Org. Chem.
1988, 53, 3051-3057.
(67) Matsuda, A.; Shinozaki, M.; Yamaguchi, T.; Homma, H.; Nomoto, R.;
Miyasaka, T.; Watanabe, Y.; Abiru, T. J. Med. Chem. 1992, 35, 241-252.
(68) Nair, V.; Young, D. A. J. Org. Chem. 1985, 50, 406-408.
9
2486 J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003

Modified Adenosines Stabilize the U1A−RNA Complex
A R T I C L E S
Hz, 2H, CH2Ph), 2.21 (m, 2H, CH2). HRMS(FAB) calcd for C47H45N5O7
[MH+], 792.339 724; found, 792.339 900.
5′-O-(4,4′-Dimethoxytrityl)-N6-benzoyl-2-(4-phenylbutyl)adeno-
to give both 2′-O-silyl (10, Rf )0.47, 0.047 g, 30%) and 3′-O-silyl (12,
Rf ) 0.28, 0.047 g, 30%) products as white solids.
5′-O-(4,4′-Dimethoxytrityl)-2′-O-(tert-butyldimethylsilyl)-N6-ben-
zoyl-2-(4-phenylbutyl)adenosine (10). 1H NMR (DMSO-d6) δ 11.13
(s, 1H, NHCO), 8.48 (s, 1H, H8), 8.02 (d, J ) 7.2 Hz, 2H, PhCO),
7.51-7.66 (m, 3H, PhCO), 6.80-7.40 (m, 18H, Ph), 6.03 (d, J ) 4.8
Hz, 1H, 1′H), 5.21 (d, J ) 5.7 Hz, 1H, 3′OH), 4.93 (app t, J ) 4.9 Hz,
1H, 2′H), 4.26 (app dd, J ) 10.5, 5.1 Hz, 1H, 3′H), 4.12 (app dd, J )
8.1, 3.9 Hz, 1H, 4′H), 3.703 (s, 3H, OCH3), 3.699 (s, 3H, OCH3), 3.21-
3.40 (m, 2H, 5′-H, 5′′-H, overlapping with broad water peak), 2.73
(m, 2H, CH2A), 2.53 (m, 2H, CH2Ph), 1.52-1.69 (m, 4H, CH2CH2Ph,
CH2CH2A), 0.75 (s, 9H, t-Bu), -0.05 (s, 3H, SiMe), -0.15 (s, 3H,
SiMe). HRMS(FAB) calcd for C54H61N5O7Si [MH+], 920.441 853;
found, 920.442 200.
sine (8). Reaction of compound 6 (0.510 g, 1 mmol) gave a yellow
1
foam (0.638 g, 78%). H NMR (DMSO-d6) δ 11.15 (s, 1H, NHCO),
8.50 (s, 1H, H8), 8.03 (d, J ) 7.7 Hz, 2H, PhCO), 7.51-7.65 (m,
3H, PhCO), 6.77-7.38 (m, 18H, Ph), 6.05 (d, J ) 4.9 Hz, 1H, 1′-H),
5.61 (d, J ) 5.4 Hz, 1H, OH), 5.29 (d, J ) 5.3 Hz, 1H, OH), 4.78
(m, 1H, 2′-H), 4.33 (m, 1H, 3′-H), 4.10 (m, 1H, 4′-H), 3.70 (s, 3H,
OCH3), 3.69 (s, 3H, OCH3), 3.21-3.34 (m, 2H, 5′-H, 5′′-H), 2.78 (t,
J ) 7.5 Hz, 2H, CH2A), 2.56 (t, J ) 7.5 Hz, 2H, CH2Ph), 1.67 (m,
2H, CH2CH2A), 1.58 (m, 2H, CH2CH2Ph). HRMS(FAB) calcd for
C48H47N5O7 [MH+], 806.355 374; found, 806.355 700.
Procedure for the Silylation of 5′-O-(4,4′-Dimethoxytrityl)-N6-
benzoyl-2-(3-phenylpropyl)adenosine. To a suspension of compound
7 (0.079 g, 0.10 mmol) in dry pyridine (1 mL) was added AgNO3 (0.026
g, 0.15 mmol). After the AgNO3 completely dissolved, a solution of
tert-butyldimethylsilyl chloride (0.017 g, 0.14 mmol) in dry THF (1
mL) was added dropwise. The flask was covered with aluminum foil,
and the reaction was stirred under N2 overnight. The reaction mix-
ture was diluted with CH2Cl2 (2 mL), filtered to remove AgCl, and
washed with 5% NaHCO3 (3 mL). The aqueous layer was extracted
with CH2Cl2 (3 × 3 mL), and the combined CH2Cl2 extractions were
washed with water (2 × 3 mL) and saturated NaCl solution (3 × 3
mL), dried over Na2SO4, and concentrated under reduced pressure. The
resulting sticky residue (∼1 mL) was coevaporated with toluene (2 ×
4 mL) in vacuo to remove residual pyridine. The mixture was separated
by prep TLC (1:1 EtOAc/hexanes) to give the 2′-O-silyl (9, Rf ) 0.44,
0.038 g, 42%) and the 3′-O-silyl (11, Rf ) 0.23, 0.026 g, 28%) products
as white solids.
5′-O-(4,4′-Dimethoxytrityl)-2′-O-(tert-butyldimethylsilyl)-N6-ben-
zoyl-2-(3-phenylpropyl)adenosine (9). 1H NMR (DMSO-d6) δ 11.14
(s, 1H, NHCO), 8.49 (s, 1H, H8), 8.02 (d, J ) 7.2 Hz, 2H, PhCO),
7.50-7.65 (m, 3H, PhCO), 6.78-7.41 (m, 18H, Ph), 6.05 (d, J ) 4.8
Hz, 1H, 1′H), 5.21 (d, J ) 6.0 Hz, 1H, 3′OH), 4.93 (app t, J ) 4.9 Hz,
1H, 2′H), 4.27 (app dd, J ) 10.6, 4.9 Hz, 1H, 3′H), 4.15 (app dd, J )
9.2, 4.6 Hz, 1H, 4′H), 3.703 (s, 3H, OCH3), 3.700 (s, 3H, OCH3), 3.31-
3.40 (m, 2H, 5′-H, 5′′-H, overlapping with broad water peak), 2.75
(m, 2H, CH2A), 2.56 (t, J ) 7.6 Hz, 2H, CH2Ph), 1.95 (m, 2H, CH2),
0.76 (s, 9H, t-Bu), -0.02 (s, 3H, SiMe), -0.11 (s, 3H, SiMe). HRMS-
(FAB) calcd for C53H59N5O7Si [MH+], 906.426 203; found, 906.426 000.
5′-O-(4,4′-Dimethoxytrityl)-3′-O-(tert-butyldimethylsilyl)-N6-ben-
zoyl-2-(3-phenylpropyl)adenosine (11). 1H NMR (CDCl3) δ 8.76 (s,
1H, NHCO), 8.22 (s, 1H, H8), 8.02 (d, J ) 7.1 Hz, 2H, PhCO), 7.50-
7.60 (m, 3H, PhCO), 6.76-7.40 (m, 18H, Ph), 6.06 (d, J ) 4.4 Hz,
1H, 1′H), 4.72 (app dd, J ) 10.1, 5.2 Hz, 1H, 2′H), 4.62 (app t, J )
4.9 Hz, 1H, 3′H), 4.20 (m, 1H, 4′H), 3.77 (s, 6H, OCH3), 3.27-3.53
(m, 3H, 5′-H, 5′′-H and 2′OH), 2.97 (t, J ) 7.7 Hz, 2H, CH2A), 2.69
(t, J ) 7.6 Hz, 2H, CH2Ph), 2.13 (m, 2H, CH2), 0.89 (s, 9H, t-Bu),
0.087 (s, 3H, SiMe), 0.006 (s, 3H, SiMe).
Procedure for the Silylation of 5′-O-(4,4′-Dimethoxytrityl)-N6-
benzoyl-2-(4-phenylbutyl)adenosine. KH in oil was transferred to a
preweighed flask under N2, washed with hexanes (3 × 2 mL), and
dried in vacuo. The flask was filled with N2 and weighed again to
determine the weight of KH (0.063 g, 1.575 mmol). To the flask was
added THF (0.2 mL) and 18-crown-6 (0.060 g, 0.24 mmol). After the
18-crown-6 dissolved, the mixture was cooled to 0 °C, and compound
8 (0.136 g, 0.17 mmol) in THF (1 mL) was added dropwise. After gas
evolution stopped, the mixture was cooled to -78 °C and tert-
butyldimethylsilyl chloride (0.032 g, 0.21 mmol) in THF (0.8 mL) was
added dropwise. The reaction was monitored by TLC (1:1 EtOAc/
hexanes) and was quenched by the addition of water (2 mL) at 0 °C.
The mixture was extracted with CH2Cl2 (6 × 7 mL). The combined
organic layers were dried over Na2SO4 and evaporated to dryness in
vacuo. The mixture was separated by prep TLC (1:1 EtOAc/hexanes)
5′-O-(4,4′-Dimethoxytrityl)-3′-O-(tert-butyldimethylsilyl)-N6-ben-
zoyl-2-(4-phenylbutyl)adenosine (12). 1H NMR (DMSO-d6) δ 11.14
(s, 1H, NHCO), 8.55 (s, 1H, H8), 8.02 (d, J ) 7.3 Hz, 2H, PhCO),
7.51-7.66 (m, 3H, PhCO), 6.75-7.35 (m, 18H, Ph), 5.98 (d, J ) 4.7
Hz, 1H, 1′H), 5.46 (d, J ) 6.2 Hz, 1H, 3′OH), 4.84 (app dd, J ) 10.8,
5.3 Hz, 1H, 2′H), 4.55 (app t, J ) 4.8 Hz, 1H, 3′H), 4.02 (app dd, J )
9.6, 4.7 Hz, 1H, 4′H), 3.69 (s, 3H, OCH3), 3.68 (s, 3H, OCH3), 3.20-
3.40 (m, 2H, 5′-H, 5′′-H), 2.75 (m, 2H, CH2A), 2.53 (m, 2H, CH2Ph),
1.52-1.69 (m, 4H, CH2CH2Ph, CH2CH2A), 0.83 (s, 9H, t-Bu), 0.071
(s, 3H, SiMe), 0.027 (s, 3H, SiMe).
General Procedure for the Synthesis of Phosphoramidites 13 and
14. To a suspension of compound 9 or 10 (1 equiv) in dry THF was
added collidine (8 equiv). After the mixture was cooled on ice,
N-methylimidazole (0.5 equiv) was added, followed by 2-cyanoethyl
N,N-diisopropylchlorophosphoramidite (8 equiv) dropwise. After the
mixture was warmed to room temperature and stirred for 1 h, the
reaction was complete as shown by TLC (10% ether/CH2Cl2). After
the mixture was cooled on ice, collidine (0.2 mL) and MeOH (0.2 mL)
were added to consume the excess phosphorylating reagent and the
solvent was removed in vacuo. The residue was dissolved in CH2Cl2
(8 mL), washed with 5% NaHCO3 (2 × 4 mL), and saturated NaCl (2
× 4 mL). The combined aqueous phases were extracted with CH2Cl2
(4 × 4 mL). The combined CH2Cl2 phases were dried over Na2SO4,
filtered, and concentrated to dryness under reduced pressure. The residue
was purified using prep TLC in 10% ether/CH2Cl2.
5′-O-(4,4′-Dimethoxytrityl)-2′-O-(tert-butyldimethylsilyl)-N6-ben-
zoyl-2-(3-phenylpropyl)adenosine 3′-N,N-Diisopropyl(cyanoethyl)-
phosphoramidite (13). Compound 9 (0.151 g, 0.167 mmol) reacted
to give a white foam (0.116 g, 63%). 31P NMR (DMSO-d6) δ 151.246,
149.901.
5′-O-(4,4′-Dimethoxytrityl)-2′-O-(tert-butyldimethylsilyl)-N6-ben-
zoyl-2-(4-phenylbutyl)adenosine 3′-N,N-Diisopropyl(cyanoethyl)-
phosphoramidite (14). Compound 10 (0.124 mg, 0.135 mmol) reacted
to give a white foam (0.104 g, 69%). 31P NMR (DMSO-d6) δ 151.957,
150.550.
Synthesis and Purification of RNA. RNA sequences were synthe-
sized on a 1 µmol scale with an Applied Biosystems ABI 394 DNA/
RNA synthesizer using standard protocols. Coupling yields of the
modified nucleotides were ∼50-90% determined by colorimetric
quantitation of the trityl fractions. All reaction columns and chemicals
were purchased from Glen Research. RNAs were cleaved and depro-
tected with ethanolic ammonia (3:1 NH4OH/EtOH solution, 500 µL)
at 55 °C for 12 h. After the beads cooled to room temperature, they
were washed with ethanolic ammonia (4 × 250 µL) and the combined
ethanolic ammonia fractions were concentrated to dryness in vacuo.
The TBDMS protecting groups were removed in neat TEA/3HF solution
(250 µL) at room temperature for approximately 12 h, and the reaction
was quenched by water (250 µL). The RNA was precipitated
sequentially with n-butanol and ethanol and was then purified on a
20% denaturing polyacrylamide gel [20% acrylamide, 20:1 mono/
bisacrylamide, 7 M urea in TBE (89 mM Tris-borate, 2 mM EDTA),
15 cm × 40 cm × 0.75 mm, 3 h at 50 W]. The desired band was
9
J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003 2487

A R T I C L E S
Zhao and Baranger
excised from the gel, extracted with TE buffer (10 mM Tris, 1 mM
EDTA pH7.4), dialyzed against TE buffer, concentrated to dryness by
speed-vac, and desalted by ethanol precipitation. The concentration of
the RNA was determined by UV at 260 nm. Correct composition was
confirmed by MALDI mass spectrometry and enzymatic hydrolysis.60
RNA Melting Experiments. CD spectra were recorded on a Jasco
J-810 CD spectrometer using 20 µM RNA in buffer containing 250
mM NaCl, 20 mM sodium cacodylate, pH 6.5, 0.5 mM EDTA, and 1
mM MgCl2. CD melting curves were obtained by heating at a rate of
0.5 °C/min or 1 °C/min using 0.2 cm path length cells, monitored at
211 and 260 nm between 24 and 90 °C. UV melting experiments were
performed on a Beckman DU 650 UV spectrometer using 4 µM RNA
in the same buffer as that used in the CD melting experiments. UV
melting curves were obtained by heating at a rate of 1 °C/min using 1
cm path length cells, monitored at 260 and 280 nm between 30 and 90
°C.
Gel Mobility Shift Assays. The equilibrium binding of stem loop
2 RNA to the U1A protein was monitored by electrophoretic gel
mobility shift assays. 32P-labeled stem loop 2 RNA (0.033 nM) was
incubated with competitor tRNA (1 mg/mL) and varying amounts of
U1A protein for 20 min at room temperature in a buffer containing 10
mM Tris-HCl, pH 7.4, 0.5% Triton X-100, 1 mM EDTA, and 250
mM NaCl. After addition of glycerol to a final concentration of 5%,
the bound and free RNA were separated using an 8% polyacrylamide
gel (80:1 acrylamide/bisacrylamide, 18 cm × 16 cm × 1.5 mm) in
100mM Tris-borate pH 8.3, 1mM EDTA, 0.1% Triton X-100 for 35
min at 350 V. The temperature of the gel was maintained at 25 °C by
a circulating water bath. Gels were visualized on a Molecular Dynamics
Storm 840 phosphorimager. Fraction RNA bound versus protein
concentration was plotted and curves were fitted to the equation:
fraction bound ) 1/(1 + Kd/[P])
Acknowledgment. We thank Dr. Xiaoling Wu at Yale
University for assistance with NMR experiments. Funding was
provided by the NIH (Grant GM-56857). A.M.B. is an Alfred
P. Sloan Research Fellow.
Supporting Information Available: 1H NMR spectra for all
new compounds, mass spectra, denaturing polyacrylamide gels
of A6-3CPh and A6-4CPh RNA, and the HPLC elution profile
of the enzymatic hydrolysis reaction of A6-4CPh RNA. This
material is available free of charge via the Internet at
http://pubs.acs.org.
JA021267W
9
2488 J. AM. CHEM. SOC. VOL. 125, NO. 9, 2003