Syntheses of silatranyl- and germatranyluridines

Article abstract of DOI:10.1021/ol050133v

(Chemical Equation Presented) Silatranyluridine 1 and germatranyluridine 2 have been prepared in five steps from oxazolinouridine 3 in 27 and 29% yields, respectively. These compounds are novel transition-state analogues (TSAs) for RNA hydrolysis and offe

Full text of DOI:10.1021/ol050133v

ORGANIC
LETTERS
2005
Vol. 7, No. 6
1165-1168
Syntheses of Silatranyl- and
Germatranyluridines
Chad M. Rink,† Matthew C. Mauck,† Irfan Asif,† Michael E. Pitzer,‡ and
Edward E. Fenlon*,‡
Department of Chemistry, XaVier UniVersity, 3800 Victory Parkway,
Cincinnati, Ohio 45207, and Department of Chemistry, Franklin & Marshall College,
P.O. Box 3003, Lancaster, PennsylVania 17604
edward.fenlon@fandm.edu
ABSTRACT
Silatranyluridine 1 and germatranyluridine 2 have been prepared in five steps from oxazolinouridine 3 in 27 and 29% yields, respectively.
These compounds are novel transition-state analogues (TSAs) for RNA hydrolysis and offer a number of advantages over traditional vanadium-
or rhenium-based TSAs. Germatrane 2 is completely stable in D2O at room temperature, and the half-life of silatrane 1 in D2O was found to
be >7 days by 1H NMR spectroscopy.
Transition-state analogues1 (TSAs) have proven to be
exceptionally useful compounds with wide-ranging applica-
tions. For example, TSAs can be used for the analysis of
enzyme mechanisms and conformational changes by X-ray
crystallography2 or NMR spectroscopy.3 They can also be
used to prepare novel catalysts via catalytic antibody4 or
imprinted polymer methodologies.5 Furthermore, information
about the transition state can be used to design TSAs that
are potent enzyme inhibitors.6 However, there are relatively
few examples of TSAs for RNA hydrolysis, in which
nucleophilic attack of the ribose 2′-oxygen atom on the
adjacent phosphorus atom leads to a trigonal bipyramidal
oxyphosphorane transition state, Figure 1a. Existing TSAs
for RNA hydrolysis include ribonucleoside derivatives that
contain a vanadium, technetium, or rhenium atom attached
via the 2′- and 3′-positions of the ribose, Figures 1b and 1c.
These compounds have been utilized in a number of
† Xavier University.
‡ Franklin & Marshall College.
(1) Bernhard, S. A.; Orgel, L. E. Science 1959, 130, 625-626.
(2) Lolis, E.; Petsko, G. A. Annu. ReV. Biochem. 1990, 59, 597-630.
(3) Gu, Z.; Drueckhammer, D. G.; Kurz, L.; Liu, K. Martin, D. P.;
McDermott, A. Biochemistry 1999, 38, 8022-8031.
(4) (a) Mader, M. M.; Bartlett, P. A. Chem. ReV. 1997, 97, 1281-1301.
(b) Schultz, P. G.; Yin, J.; Lerner, R. A. Angew. Chem. 2002, 41, 4427-
4437.
Figure 1. (a) Putative transition state for RNA hydrolysis. (b)
Structure of the vanadate-uridine complex bound in the active site
of RNase A as determined by X-ray crystallography. (c) Janda’s
technetium- and rhenium-based TSAs for RNA hydrolysis.
(5) Wulff, G. Chem. ReV. 2002, 102, 1-27.
(6) Schramm, V. L. Acc. Chem. Res. 2003, 36, 588-596.
10.1021/ol050133v CCC: $30.25
© 2005 American Chemical Society
Published on Web 02/25/2005

experiments. For example, the vanadate-uridine complex
has been shown to be an inhibitor7,8 of ribonuclease A
(RNase A), and numerous X-ray crystal structures of the
enzyme-TSA complex have been determined.2,9 Further-
more, a similar vanadate-RNA complex was recently used
to determine the TSA structure of the hairpin ribozyme by
X-ray crystallography.10 It is interesting to note that a similar
approach with the hammerhead ribozyme11 has not worked
despite repeated attempts;12 thus, vanadium TSAs may be
limited in scope. Meanwhile, Janda and co-workers have
shown that the rhenium compounds (Figure 1c) can inhibit
ribonuclease U2 and be used to elicit antibodies capable of
catalyzing the cleavage of the phosphodiester bond in uridine
3′-(p-nitrophenyl phosphate).13
experiments) and exist in a distorted square planar geometry
rather than in a true trigonal bipyramid.13
The limitations inherent with existing RNA hydrolysis
TSAs led us to consider atranyl-nucleosides18 as alternatives.
Atranes are well-characterized compounds in which triethanol
amine, or a derivative thereof, complexes a central trigonal
bipyramidal atom.19,20 Specifically, we envisioned silatran-
yluridine 1 and germatranyluridine 2 as novel TSAs.
The syntheses of targets 1 and 2 first required the
preparation of a key triethanol amine derivative, uridine-
triol 6, Scheme 1. Thus, uridine was converted to oxazolin-
Scheme 1
However, despite these successes, or maybe because of
them, more and better RNA hydrolysis TSAs are needed.
This is due to the unfortunate fact that the vanadium and
rhenium complexes have significant limitations. For example,
it is well established that oxovanadium alkoxides are
especially unstable in aqueous solution and rapidly undergo
ligand exchange.14 In fact, vanadate tends to form dimers
and trimers even at millimolar concentrations, complicating
the structural analysis of the vanadate-nucleosides and their
precise binding constants with RNase.8 Thus, the vanadium
TSA compounds used in the X-ray diffraction studies of
RNase A2,9 and the hairpin ribozyme10 were prepared in situ.
In fact, 1:1 complexes such as the one shown in Figure 1b
have never been directly detected in solution15 and may, in
fact, only exist in enzyme active sites. The exact nature of
the products formed from vanadate and a nucleoside in
aqueous solution (i.e., outside an enzyme active site) has
been the subject of considerable debate.16 Detailed NMR and
X-ray crystallographic studies15,17 have led to the understand-
ing that the major product is a 2:2 complex containing two
trigonal bipyramidal vanadium atoms and two nucleoside
ligands. On the other hand, the major limitations of the
rhenium complexes (Figure 1c) are that they have relatively
poor water solubility (a DMSO cosolvent is required in some
ouridine 3 using McGee’s intramolecular cyclization method.21
The uracil N-3 was then methylated, either by deprotonation/
alkylation (NaH/dimethyl sulfate) or, more conveniently, by
heating
with
dimethylformamide-dimethyl-
acetal. Methylation of N-3 was necessary to prevent reactivity
and to ensure the stability of the atranes.22 Removal of the
2′,3′-oxazoline from 4 by cesium carbonate treatment21
provided amino-alcohol 5 in 87% yield. Double alkylation
of the 2′-amino group with ethylene oxide in a sealed tube
provided 76% yield of the desired uridine-triol 6.
(7) Lindquist, R. N.; Lynn, J. L., Jr.; Lienhard, G. E. J. Am. Chem. Soc.
1973, 95, 8762-8768.
(8) Messmore, J. M.; Raines, R. T. J. Am. Chem. Soc. 2000, 122, 9911-
9916.
(9) (a) Borah, B.; Chen, C.; Egan, W.; Miller, M.; Wlodawer, A.; Cohen,
J. S. Biochemistry 1985, 24, 2058-2067. (b) Wladkowski, B. D.; Svensson,
L. A.; Sjolin, L. Ladner, J. E.; Gilliland, G. L. J. Am. Chem. Soc. 1998,
120, 5488-5498.
(10) Rupert, P. B.; Massey, A. P.; Sigurdsson, S. Th.; Ferre-D’Amare,
A. R. Science 2002, 298, 1421-1424.
With triol 6 in hand, preparation of the atrane moiety
proceeded smoothly. Thus, heating 6 with tetraethyl ortho-
(11) For reviews, see: (a) Scott, W. G. Q. ReV. Biophys. 1999, 32, 241-
284. (b) Hammann, C.; Lilley, D. M. J. ChemBioChem 2002, 3, 690-700.
(12) Scott, W. G. University of California at Santa Cruz, Santa Cruz,
CA. Personal communication, 2004.
(13) (a) Chen, Y. C. J.; Janda, K. D. J. Am. Chem. Soc. 1992, 114, 1488-
1489. (b) Wentworth, P., Jr.; Wiemann, T.; Janda, K. D. J. Am. Chem.
Soc. 1996, 118, 12521-12527. (c) Weiner, D. P.; Wiemann, T.; Wolfe, M.
M.; Wentworth, P., Jr.; Janda, K. D. J. Am. Chem. Soc. 1997, 119, 4088-
4089.
(14) Crans, D. C.; Felty, R. A.; Miller, M. M. J. Am. Chem. Soc. 1991,
113, 265-269.
(15) Ray, W. J., Jr.; Crans, D. C.; Zheng, J.; Burgner, J. W., II; Deng,
H.; Mahroof-Tahir, M. J. Am. Chem. Soc. 1995, 117, 6015-6026.
(16) Vanadium Compounds; Tracey, A. S., Crans, D. C., Eds.; ACS
Symposium Series 711; American Chemical Society: Washington, DC,
1998.
(18) (a) Sculimbrene, B. R.; Decanio, R. E.; Peterson, B. W.; Muntel,
E. E.; Fenlon, E. E. Tetrahedron Lett. 2001, 42, 4979-4982. (b) Black, C.
A.; Ucci, J. W.; Vorpagel, J. S.; Mauck, M. C.; Fenlon, E. E. Bioorg. Med.
Chem. Lett. 2002, 12, 3521-3523.
(19) Verkade, J. G. Coord. Chem. ReV. 1994, 137, 233-295.
(20) Silatranes are the prototypical atranes; for a review, see: Voronkov,
M. G.; D’yakov, V. M.; Kirpichenko, S. V.: Silatranes. J. Organomet.
Chem. 1982, 233, 1-147.
(21) McGee, D. P. C.; Vaughn-Settle, A.; Vargeese, C.; Yansheng, Z.
J. Org. Chem. 1996, 61, 781-785.
(22) If unmethylated, N-3 is subsequently alkylated by ethylene oxide
further into the synthesis (see Supporting Information). The N3-methyl is
also believed to provide stability to all of the atranyl-nucleosides (1, 2, 7,
and 8). This is due to the fact the 2′-nitrogen is a decent leaving group (it
has a formal positive charge) and it is well-known that uridines with leaving
groups positioned on the 2′-position are subject to anhydrouridine formation.
See Supporting Information and the following reference: Yung, N. C.; Fox,
J. J. J. Am. Chem. Soc. 1961, 83, 3060-3066.
(17) Angus-Dunne, S. J.; Batchelor, R. J.; Tracey, A. S.; Einstein, F.
W. B. J. Am. Chem. Soc. 1995, 117, 5292-5296.
1166
Org. Lett., Vol. 7, No. 6, 2005

doublet for the same proton in uridine 9 were monitored,
Scheme 3. This experiment revealed that the half-life of 1
Scheme 2
Scheme 3
in the D2O solution at room temperature is greater than 7
days.26 As expected,23 germatranyluridine 2 proved to be
1
completely stable in water. The H NMR spectrum of 2 in
D2O remained unchanged after 3 weeks at room tempera-
ture.26
In summary, silatranyluridine 1 and germatranyluridine 2
have been prepared in 27 and 29% yields, respectively, in
five steps from oxazolinouridine 3.21 These compounds
represent a new class of TSAs for RNA hydrolysis and
possess several advantages over existing TSAs. Namely,
unlike the vanadate TSAs, 1 and 2 can be isolated and
purified. Furthermore, they have sufficient aqueous solubility
and stability to be employed as probes of RNA hydrolysis
enzymes. For example, insertion of TSAs 1 or 2 at the
cleavage site (replacing C17) of the hammerhead ribozyme
should allow a TSA X-ray crystal structure to be obtained.12,28
Such a structure will aid in the elucidation of this ribozyme’s
catalytic mechanism29 and may address the contested issue
regarding the magnitude of conformational rearrangement
during catalysis.30 Compounds 1 and 2 should also prove to
be useful as TSAs for the generation of new catalysts.4,5,13c
Finally, just as silicon- and germanium-containing amino
acids have been prepared and studied,31 one can view
compounds 1 and 2 as novel examples of silicon- and
gemanium-containing nucleosides.32
silicate provided 55-71% yield of silatrane 7, Scheme 2.
Alternatively, 6 was heated with germanium dioxide23 in
acetonitrile-water to provide 92% yield of germatrane 8.
Detritylation of 7 and 8 was best accomplished by treatment
with montmorillonite K 10 (K-10 clay)24 to produce silatran-
yluridine 1 and germatranyluridine 2 in 60 and 51% yields,
respectively.
The structures of 1 and 2 are supported by NMR
spectroscopy and mass spectrometry. Of particular note are
the 29Si NMR signal of 7 at -90.1 ppm, which is indicative
of silatranes,25 and the germanium isotope pattern in the mass
spectrum of 2.26 One indicator of the successful formation
of an atrane in these systems was found to be the coupling
constant for the 1′- and 2′-protons in the ribose ring, 3JH1′-H2′
.
Thus, in amine 5 and triol 6, the ribose is in an S-type (C2′-
endo) conformation with a relatively large 3JH1′-H2′ (8-9 Hz),
whereas in compounds that contain the atrane moiety (i.e.,
1, 2, 7, and 8), the ribose conformation changes to an N-type
(C3′-endo) conformation with a smaller 3JH1′-H2′ (<2 Hz).27
(Oxazolinouridines 3 and 4 also exist in an N-type confor-
mation.)
Acknowledgment. This paper is dedicated to Prof. Robert
J. Kempton on the occasion of his 60th birthday. This work
was supported by the National Science Foundation (MCB-
In order for these TSAs to be employed as probes of
enzymes or ribozymes, they must have adequate aqueous
solubility and stability. Thus, the aqueous stability of the
1
TSAs was determined by monitoring a D2O solution by H
(28) C17 in the hammerhead ribozyme was previously replaced with
3-methyluridine and found to retain much of its catalytic activity, with a
cleavage rate equal to 43% of the wt rate; see: Baidya, N.; Ammons, G.
E.; Matulic-Adamic, J.; Karpeisky, A. M.; Beigelman, L.; Uhlenbeck, O.
C. RNA 1997, 3, 1135-1142.
NMR spectroscopy.18a In the case of silatranyluridine 1, the
decrease in integration of the doublet for the H6-proton in 1
and the corresponding increase in the integration of the
(29) Takagi, Y.; Ikeda, Y.; Taira, K. Top. Curr. Chem. 2004, 232, 213-
251.
(23) Mironov, V. F.; Gar, T. K.; Khromova, N. Yu.; Frid, O. D. J. Gen.
Chem. USSR (Engl. Transl.) 1986, 56, 566-568.
(30) (a) Wang, S.; Karbstein, K.; Peracchi, A.; Beigelman, L.; Herschlag,
D. Biochemistry 1999, 38, 14363-14378. (b) Murray, J. B.; Scott, W. G.
J. Mol. Biol. 2000, 296, 33-41.
(24) Asakura, J.; Robins, M. J.; Asaka, Y.; Kim, T. H. J. Org. Chem.
1996, 61, 9026-9027.
(31) Merget, M.; Gu¨nther, K.; Bernd, M.; Gu¨nther, E.; Tacke, R. J.
Organomet. Chem. 2001, 628, 183-194 and references therein.
(32) Although there are several examples of silicon-containing nucleo-
sides, there are just two examples of germanium-containing nucleosides or
nucleosidelike compounds, see: (a) Mel’nik, S. I.; Bakhmedova, A. A.;
Nedorezova, T. P.; Iartseva, I. V.; Zhukova, O. S. Bioorg. Khim. 1985, 11,
1248-1252. (b) Lukevics, E.; Ignatovich, L.; Shilina, N.; Kemme, A.;
Sjakste, N. Metal-Based Drugs 1994, 1, 65-72.
(25) (a) Bellama, J. M.; Nies, J. D.; Ben-Zvi, N. Magn. Reson. Chem.
1986, 24, 748-753. (b) Tandura, S. N.; Voronkov, M. G., Alekseev, N. V.
Top. Curr. Chem. 1986, 131, 99-189.
(26) See Supporting Information for more details.
(27) (a) Marino, P. P.; Schwalbe, H.; Griesinger, C. Acc. Chem. Res.
1999, 32, 614-623. (b) Bondensgaard, K. Mollova, E. T.; Pardi, A.
Biochemistry 2002, 41, 11532-11542.
Org. Lett., Vol. 7, No. 6, 2005
1167

0335329). This research was also supported by an award
from Research Corporation (CC4768). We thank Amy
Mettman, Julie Villari, Brooke Keeley, Scott Paviol, Michael
Rheam, and Kevin Schaefer for their contributions to this
work. We thank the reviewers for numerous manuscript
suggestions. Finally, we warmly thank Prof. William G. Scott
for encouragement and helpful discussions.
spectral data (1H and 13C NMR and mass spectra for
compounds 2 and 4-8 and H NMR and mass spectra for
1
1); 1H NMR spectra for compounds 1, 2, and 5-8; 13C NMR
spectrum for compound 4; 29Si NMR spectrum of compound
7; the mass spectral germanium isotope pattern of 2; 1H NMR
spectra for the aqueous stability studies of 1 and 2; and a
discussion of the N-3 methyl group’s role in the stability of
the atranyluridines. This material is available free of charge
via the Internet at http://pubs.acs.org.
Supporting Information Available: Experimental pro-
cedures for the preparation of compounds 1, 2, and 4-8;
OL050133V
1168
Org. Lett., Vol. 7, No. 6, 2005