Cholesterol esterase activity by in vitro selection of RNA against a phosphate transition-state analogue [9]

Full text of DOI:10.1021/ja991848u

10844
J. Am. Chem. Soc. 1999, 121, 10844-10845
Cholesterol Esterase Activity by in Vitro Selection of
RNA against a Phosphate Transition-State Analogue
Sung-Min Chun,† Sunjoo Jeong,‡ Jong-Man Kim,†
Byong-Oh Chong,† Young-Keun Park,§ Hokoon Park,† and
Jaehoon Yu*,†
Life Sciences DiVision, The Korea Institute of Science and
Technology, PO Box 131, Cheongryang, Seoul 130-650, Korea
Department of Molecular Biology
College of Natural Sciences, Dankook UniVersity
Seoul 140-714, Korea
Graduate School for Biotechnology
Korea UniVersity, Seoul 136-701, Korea
ReceiVed June 3, 1999
Figure 1. TS for cholesterol ester (1) hydrolysis and structures of the
TSA analogue, 2 and other substrates.
Creation of efficient catalysts with high specificities is one of
the most challenging goals for chemists. Various attempts have
been made to induce catalytic sites in natural and synthetic
polymers by the use of transition-state analogues (TSA).1
Antibodies have been raised against TSA and used as catalysts
for chemical transformations. The rates of these antibody-
catalyzed transformations are enhanced over background reac-
tions.2 On the basis of the discovery of catalytic antibodies,
attempts have been made to extend the concepts involved in the
antibody catalysis to the development of other biopolymeric
catalysts.3 RNA molecules are also potential catalytic biopolymers
because of their unique conformations, molecular diversity, and
relative ease of generation. However, most of the catalysts in this
family, which have been successfully generated so far4 have been
developed by using direct selection of self-modified RNA.5 While
numerous attempts to select RNA catalysts via TS stabilization
have been made, only a few successful results were reported for
hydrophobic TSA,6,7 owing to the hydrophobic nature of the RNA
pocket.8 These observations suggest that the hydrophobic interac-
tions may be more precise and specific than hydrophilic interac-
tions.9
nogenicity can be solved by use of in vitro selection of RNA, a
process that takes a relatively short period of time. As part of
our continuing interests in the design of TSA-induced catalysts,
we for the first time have succeeded in generating RNA with a
defined cholesterol esterase activity.
The design and synthesis of the TSA 2 were carried out by
the use of known procedures.12 Immobilization of 2 on agarose
was done under mildly basic conditions.13 A 110-mer DNA library
was designed to contain random nucleotides in 70 positions
flanked by defined sequences at both ends for the purpose of PCR
amplification and in vitro transcription.14 The in vitro selection
of RNA was made by the use of modification protocols.15
Enrichment of the RNA was confirmed by means of affinity
chromatography and elution procedures with 32P-labeled RNA.4(c),15
After six cycles of selection, enrichment was achieved as judged
by the fact that 30% of the applied RNA was specifically eluted
by 2. The selected RNA was then cloned and sequenced by
standard protocols.16 Since the affinity elution was performed
under less stringent conditions due to the poor solubility of 2 in
the aqueous phase, no identical sequence was found among 11
cloned RNA molecules.
To measure the binding affinity of 2 to the cloned RNA, a
surface plasmon resonance (SPR) technique was utilized. In the
first process, the affinities between RNAs and 2 were measured
by changing the concentration of RNA in solution with im-
mobilized 2 in a flowcell.17 One clone18 showed the best binding
(KD ) 4.0 × 10-8 M) as compared the original pool (KD ) 1.0 ×
10-5 M). The KD values of other cloned RNAs were similar (KD
We reason that the TSA of cholesterol esterase should be
sufficiently hydrophobic to select specific RNA binders. An
attempt has already been made to generate catalytic antibodies
against a similar hapten as a phosphate ester 2, a TSA in the
cholesterol esterase hydrolysis of the carbonate ester 1 (Figure
1).10 However, this failed, possibly as a result of poor immuno-
genicity.11 The intrinsic difficulties associated with poor immu-
* To whom correspondence should be addressed. Telephone: +82-2-958-
5157. Fax: +82-2-958-5189. E-mail: jhoonyu@kist.re.kr.
† The Korea Institute of Science and Technology.
‡ Dankook University.
§ Korea University.
(10) Galatina, A. I.; Novikava, N. S.; Dekach, L. G.; Kramarenko, N. L.;
Tsyguleva, O. M.; Kuzin, V. F. Mol. Cryst. Liq. Cryst. 1986, 140, 11.
(11) Schultz, P. G., unpublished results.
(12) See Supporting Information.
(13) Yu, J.; Chun, S. M.; Park, Y. K.; Park, H.; Jeong, S. J. Biochem.
Mol. Biol. 1999, 12, 96.
(14) Twenty-four micrograms of 131-mer DNA (Midland Certified Com-
pany, Midland, TX) have 3.3 × 1014 of diversity. Usually, 1 µg was used to
perform PCR for the initial RNA library, affording 1.4 × 1013 of diversity.
The DNA template was as shown below:
(1) Pauling, L. 1946 Chem. Eng. News 1946, 24, 1375.
(2) Schultz, P. G.; Lerner, R. A. Science 1995, 269, 1835. (b) Schultz, P.
G.; Lerner, R. A. Acc. Chem. Res. 1993, 26, 391. (c) Jacobsen, W. H.; Scanlan,
T. S. Annu. ReV. Biophys. Biomol. Struct. 1997, 26, 461.
(3) Breslow, R. Acc. Chem. Res. 1995, 28, 146. (b) Kirby, A. J. Angew.
Chem., Int. Ed. Engl. 1996, 35, 707. (c) Wulff, G. Angew. Chem., Int. Ed.
Engl. 1995, 34, 1812.
(4) Lorsch, J. R.; Szostak, J. W. Acc. Chem. Res. 1996, 29, 103. (b)
Tarasow, T. M.; Tarasow, S. L.; Eaton, B. E. Nature 1997, 389, 54. (c) Zhang,
B.; Cech, T. R. Nature 1997, 390, 96. (d) Unrau, P. J.; Bartel, D. P. Nature
395, 260.
(5) Joyce, G. F.; Orgel, L. E. The RNA World; Cold Spring Harbor
Laboratory Press: Cold Spring Harbor, 1989; p 1. (b) Joyce, G. F. Nature
1989, 338, 217. (c) Ellington, A. D.; Szostak, J. W. Nature 1990, 346, 818.
(d) Tuerk, C.; Gold, L. Science 1990, 249, 505. (e) Breaker, R. R. Chem.
ReV. 1997, 97, 371.
(15) Doudna, J. A.; Cech, T. R.; Sullenger, B. A. Proc. Natl. Acad. Sci.
U.S.A. 1995, 92, 2355. (b) Jayasena, V. K.; Gold, L. ibid. 1997, 94, 10612.
(16) Mezei, L. M.; Storts, D. R. PCR Technology: Current InnoVations;
CRC Press: Boca Raton, FL., 1994; p 21. (b) Clark, J. M. Nucleic Acids Res.
1988, 16, 9677.
(17) Hendrix, M.; Priestley, E. S.; Joyce, G. F.; Wong, C.-H. J. Am. Chem.
Soc. 1997, 119, 3641. (b) Kraus, E.; James, W.; Barclay, A. N. J. Immunol.
1998, 160, 5209.
(6) Prudent, J. R.; Uno, T.; Schultz, P. G. Science 1994, 264, 1924.
(7) Morris, K. N.; Tarasaw, T. M.; Julin, C. M.; Simon, S. L.; Hilvert, D.
Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 13028. (b) Conn, M. M.; Prudent, J.
R.; Schultz, P. G. J. Am. Chem. Soc. 1996, 118, 7012.
(8) Wilson, W. D.; Ratmeyer, L.; Zhao, M., Strekowski, L.; Boykin, D.
Biochemistry 1993, 32, 4098.
(18) The random region of the sequence of this clone is as follows: 5′-
GTGGGGTCGT CTTGGTTAAA CTCCTTGCGC GTCACGAGGT TAGC-
CAGCTT GATACCTCAA GGTGGTGCCT-3′.
(9) Lohman, T. M. CRC Crit. ReV. Biochem. 1986, 19, 191.
10.1021/ja991848u CCC: $18.00 © 1999 American Chemical Society
Published on Web 11/04/1999

Communications to the Editor
J. Am. Chem. Soc., Vol. 121, No. 46, 1999 10845
) 1.1 × 10-7 M) to those of the selected pool. Despite differences
in the sequences of the cloned RNAs, the binding affinities were
similar, suggesting that the binding might be mediated only by
nonspecific stacking of RNA bases. To rule out this possibility,
binding affinities to the cloned RNA by a variety of hydrophobic
molecules were determined using the RNA clone as an im-
mobilized ligand19 in a SPR techniques. Compound 2 and
hydrophobic analogues, such as ergosterol, vitamin D3, â-estradiol,
and p-aminophenyl phosphate in solution were injected to the
RNA-immobilized flowcells. The same KD value was observed
for binding of 2 to the RNA by use of this process. However,
cholesterol and other hydrophobic molecules, as well as p-
aminophenyl phosphate, did not show binding affinity to this RNA
clone. These data suggest that the hydrophobic binding site in
this RNA has quite specific complementarity to 2 and that the
structure which combines both hydrophobic cholesterol and the
relatively hydrophilic p-aminophenyl phosphate moieties are
necessary for the binding of the RNA.
The RNA that was selected for the TSA 2 had a 250-fold
greater affinity of binding than original RNA. Thus, we anticipated
that this RNA would significantly reduce the activation energy
for the hydrolysis of carbonate ester 1. All cloned RNA molecules
were incubated with a 0.1 mM solution of the substrate 1 in a
saline buffer at 37 °C and pH 7.5.20 The p-nitrophenyl substrate
was used in this process to elicit prompt and precise observation
of catalytic activity. Two clones including the one displaying
maximum binding to 2 showed catalytic activity. The reaction of
1 catalyzed by the maximum-binding RNA showed saturation
kinetics. Enzymatic constants were determined to be kcat ) 1.3
× 10-3/h and Km ) 29 µM at pH 7.5 by use of a Lineweaver-
Burk plot (Figure 2).21 The background hydrolysis rate was
measured under the same conditions (kuncat ) 1.2 × 10-5/h).22
The rate enhancement (kcat/kuncat) was 110. The catalytic activity
was thoroughly inhibited by the TSA, indicating that the binding
pocket for 2 also is the catalytic site for hydrolysis. No rate
enhancement was observed from the hydrolysis of a similar
carbonate ester 310 or of a small carbonate ester 423 by the
maximum-binding RNA.21,22 Nuclease protection study with the
Figure 2. Lineweaver-Burk plot for cholesterol esterase activity by the
maximum-binding RNA clone.
maximum-binding RNA and TSA 2 yielded a protected fragment
of approximately 30 nucleotides in 10% acryl amide gel, whereas
the RNA without or with nonspecific compounds did not show
any protected band.12 Both studies of substrate specificity and
nuclease protection resulted in the conclusion that hydrolysis of
1 by the maximum-binding RNA is only derived from the specific
hydrophobic pocket for TSA 2.
Although the rate enhancement is not as high as usual for
catalytic antibodies, it appears to be a reasonable value on the
binding energy of 12 kJ/mol measured by the affinity selection
(KD values of clone 8 versus original RNA). If all of this binding
energy were fully converted to TS stabilization, the predicted rate
acceleration (kcat/kuncat) would be 250 at 37 °C.24 The observed
rate enhancement was about half as much as this value. This
suggests that a very significant fraction of binding energy of RNA
to the TSA is reflected in the catalytic activity.
Our results demonstrate the first successful selection by TSA
screening of an RNA that catalyzes a reaction that requires a
nucleophilic or general base-activated nucleophilic substitution
at the carbonyl functionality. Detailed molecular recognition
studies are in progress to elucidate the mechanism of this catalytic
process. In addition to the catalytic activity of ester hydrolysis,
the submicromolar affinity of the selected RNA against a
cholesterol derivative suggests that RNA molecules might possibly
be used as antibody substitutes against haptenic hydrophobic
molecules, such as cholesterol and the related steroid hormones.
(19) Misiura, K.; Durrant, I.; Evans, M. R.; Gait, M. J. Nucleic Acids Res.
1990, 18, 4345. (b) Connolly, B. A. ibid. 1987, 15, 3131. (c) Agrawal, S.;
Christodoulou, C.; Gait, M. J. ibid. 1986, 14, 6227.
(20) Since the RNA was selected at pH 7.5, the catalytic activity was
observed only at that pH. For the initial screening, concentrated solution of
RNA (17 µM) in 100 µM of substrate 1 was incubated at 37 °C for 24 h in
assay buffer (10 mM EPPS, 100 mM NaCl, 0.02% NaN3 at pH 7.5). The
product conversion was measured by injecting 10 µL aliquots of the reaction
mixture onto an HPLC column and monitoring at 315 nm for p-nitrophenol.
A C18 column (Microsorb; 4.6 mm × 15 cm, 5 µm) was used as the stationary
phase and 50% aqueous acetonitrile as the mobile phase.
(21) To a 90 µL of RNA solution (0.67 µM) in assay buffer, 10 µL of 10
× substrate solution (variable from 0.15 mM to 1.0 mM in THF) was added.
The resulting solution (0.6 µM of RNA, 15-100 µM of variable substrate
concentration) was incubated at 37 °C for 3 days. The product conversion
was measured by injecting 10 µL aliquots of the reaction mixture onto an
HPLC column in every 24 h.
Acknowledgment. Financially support for this work was provided
by KIST (2V00363). S.J. is a beneficiary of Academic Research Fund
of Ministry of Education, Korea (GE97-211). We thank Mr. Joon-Young
Jun for help with the SPR studies. This contribution is dedicated to J.Y.’s
late father Professor Byung-Sul Yu on the 10th anniversary of his passing,
on December 15, 1999.
Supporting Information Available: Experimental procedures for
preparation of 2, for selection of RNA, for binding affinity studies, and
for nuclease protection studies are available (PDF). This material is
available free of charge via the Internet at http://pubs.acs.org.
(22) In the absence and presence of original pool RNA (0.6 µM),
background hydrolysis rate of the substrate was measured, affording the same
number.
JA991848U
(23) Iimori, T.; Shibazaki, T.; Ilegami, S. Tetrahedron Lett. 1996, 37, 2267.
(24) kcat/kuncat ) exp(Euncat - Ecat/RT)