Total synthesis of the hypermodified tRNA nucleoside epoxyqueuosine

Article abstract of DOI:10.1002/ejoc.201300586

Queuosine is one of the most strongly modified nucleosides found in transferRNA (tRNA) of almost all species. Queuosine is found exclusively in the wooble position of the anticodon loop of several tRNA, where it is involved in tRNA^Asp, tRNA

Full text of DOI:10.1002/ejoc.201300586

SHORT COMMUNICATION
DOI: 10.1002/ejoc.201300586
Total Synthesis of the Hypermodified tRNA Nucleoside Epoxyqueuosine
Ines Thoma[a] and Thomas Carell*[a]
Keywords: Synthetic methods / Natural products / Nucleosides / tRNA / Structure elucidation
Queuosine is one of the most strongly modified nucleosides
found in transferRNA (tRNA) of almost all species. Queuos-
ine is found exclusively in the wooble position of the antico-
pendent on this external source. To investigate the interest-
ing biosynthesis of the Q base, we report here the first total
synthesis of the direct biosynthetic precursor oQ. By using
LC–MS in combination with special E. coli growing condi-
tions (no vitamin B12 in the medium) we confirm the struc-
ture of the oQ natural products.
don loop of several tRNA, where it is involved in tRNAAsp
,
tRNAAsn tRNAHis, and tRNATyr. The biosynthesis of queuos-
ine is performed by prokaryotes, and eukaryotes are fully de-
the transformation in these organisms is catalyzed by the
vitamin B12 dependent enzyme QueG. Today, neither the
mechanism nor the structure of this enzyme is known.[2,8]
The biosynthesis of Q by prokaryotes is of immense impor-
tance, because higher organisms such as humans are unable
to produce the base Q. All eukaryotes rely on prokaryotic
biosynthesis. Eukaryotes hydrolyze the glycosidic bond of a
particular guanosine to create an abasic site and to react
this abasic site with the heterocycle quinine. This reaction
is catalyzed by the enzyme tRNA guanine-transglycosylase
(Tgt).[9,10] oQ (1) is consequently an important biosynthesis
precursor, and the vitamin B12 catalyzed conversion of oQ
into Q is of central importance for correct decodation of
the genetic code in all organisms. To elucidate the mecha-
nism and the structure of the interesting vitamin B12 de-
pendent, iron–sulfur-containing reductase[8] that converts
oQ into Q, we developed a synthesis of oQ, which we report
here.
Introduction
The faithful translation of the genetic code at the ribo-
some is of paramount importance for the biosynthesis of
proteins. Incorrect translation can lead to amino acid sub-
stitution or frameshifting events, which may generate mis-
folded or generally unfunctional proteins. Although many
of these proteins are harmless and will be quickly degraded,
some may trigger unwanted cellular processes. To avoid the
energy wasting production of erroneous proteins, the de-
coding process of the genetic code in the ribosome is tightly
controlled, which requires, next to the canonical bases, a
large set of further modified nucleobases that increase the
structural diversity of all RNA components involved in the
decoding process such as transferRNA (tRNA) or ribosom-
alRNA (rRNA). Although the structures of most of the
noncanonical bases involved in codon translation are
known today, their biological function is in many cases not
known.
Particularly interesting in this respect are the biosynthe-
sis proteins involved in synthesizing the noncanonical nu-
cleobase structures. Many of the proteins perform unusual
and not-well-understood chemistry. Epoxyqueuosine (oQ,
1) and queuosine (Q, 2), for example, (Figure 1) are strongly
modified G-derived nucleobases. Q is found specifically in
the wobble position of the tRNAs tRNAAsp, tRNAAsn
tRNAHis, and tRNATyr. oQ was first isolated in 1987 by
Phillipson et al.[1–6] The stereochemistry was deciphered in
2000 by Kinzie et al. who used 13C labeling studies.[7] Re-
cent studies provided evidence that oQ (1) is the biosyn-
thetic precursor of queuosine (Q, 2) in prokaryotes and that
Figure 1. Depiction of epoxyqueuosine (oQ, 1) and queuosine (Q,
2) together with the vitamin B12 dependent biosynthesis pathway
from oQ to Q.
[a] Center for Integrative Protein Science (CiPSM), Department of
Chemistry, Ludwig-Maximilians-Universität, Munich
Butenandtstrasse 5–13, 81377 München, Germany
Fax: +49-89-218077756
Results and Discussion
We recently reported the synthesis of queuosine itself,[11]
and hence, we planned the synthesis of oQ (1) accordingly.
The central step is the reductive amination of formyl deaza-
E-mail: thomas.carell@cup.uni-muenchen.de
Homepage: www.carellgroup.de
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201300586.
Eur. J. Org. Chem. 2013, 4483–4485
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4483

I. Thoma, T. Carell
SHORT COMMUNICATION
Scheme 1. Synthesis of building blocks 3 and 4 for the reductive amination reaction. Reagents and conditions: (a) Pd2(dba)3 (dba =
dibenzylideneacetone), PPh3, Bu3SnH, CO (3 bar), 55 °C, 91%. (b) (CCl3)COCl, pyridine, r.t., 3 h, 79%. (c) 1. AcOH (70%), 40 °C, 20 h;
2. TBSCl, pyridine, 40 °C, 40 h, 77%. (d) m-CPBA, CH2Cl2, 0 °CǞr.t., 48 h, 88%. (e) NaOH (2 m), EtOH, r.t., 13 h, 87%.
guanosine nucleoside 3 with epoxide 4. Nucleoside 3 was
prepared according to published procedures starting from
bromoacetaldehyde diethyl acetal (5) and pyrimidinone 6,
which led, in six steps, to iodonucleoside 7.[11,12] The Pd0-
catalyzed formylation reaction was optimized by working
under CO pressure in an autoclave.
Epoxide 4 was synthesized as depicted in Scheme 1. The
starting material was acetonide-protected cyclopentene de-
Scheme 2. Reductive amination and deprotection to afford oQ (1).
rivative 8, which was prepared in six steps according to pub-
lished procedures.[11] We protected the amine by converting
it into the trichloroacetic acid amide. Next, the acetonide
protecting group was cleaved (70% acetic acid, 40 °C, 20 h)
to give the diol. The hydroxy groups were protected as
bulky tert-butyldimethylsilyl (TBS) ethers (TBSCl, pyridine,
40 °C) to control the stereochemistry of the epoxidation re-
Reagents and conditions: (a) 1. Benzene, 50 °C; 2. NaBH4, MeOH,
0 °C, 73%. (b) 1. NaOH (0.5 m in MeOH), r.t., 24 h; 2. HF·NEt3,
r.t., 23 h, 66%.
Given that oQ was synthesized for the first time, we com-
action and to avoid potential future Payne rearrange- pared the NMR signals of the synthetic compound with the
ments.[13,14] The epoxide was subsequently prepared from data reported for the natural product. Both signal sets were
cyclopentene 9 by using meta-chloroperoxybenzoic acid (m- found to be in excellent agreement. We finally wanted to
CPBA; CH2Cl2, 0 °CǞr.t.). The reaction yielded exclu- unambiguously prove that the synthesized material was
sively cyclopentane 10 with the epoxide oxygen trans to the identical to the natural product, and hence, we performed
diol group as expected. The NOESY spectrum shows, for a direct LC–MS-based comparison. To obtain the natural
example, coupling between the H1 and H2 protons and be- product, we grew E. coli in a minimum amount of 3-(N-
tween the H3 proton and one of the methyl groups from the morpholino)propanesulfonic acid (MOPS) medium lacking
TBS protecting group (Figure S1, Supporting Information). vitamin B12. In the absence of the cofactor, E. coli is unable
Deprotection of the primary amine was performed under to convert oQ into Q, which leads to elevated oQ levels.[15]
basic conditions (2 m NaOH). The synthetic procedure fur- We extracted the tRNA and performed a total enzymatic
nished amine precursor 4 in nine steps with a total yield of digest to obtain the individual nucleosides, as reported pre-
8%.
viously by us.[16] This nucleoside mixture was subsequently
The reductive amination (Scheme 2) was carried out un- analyzed by using LC–MS. Under our conditions, the
der a nitrogen atmosphere. Sodium borohydride was used amounts of Q and oQ in the sample were so low that direct
as the reducing agent. After heating the reaction mixture UV detection was impossible. We therefore performed mass
containing 3 and 4 to 40 °C to maximize imine formation, spectrometry studies by using a filter in the m/z = 426.1580–
we cooled the solution to 0 °C and added NaBH4. The reac- 426.1630 range around the mass of the natural product. The
tion furnished coupled product 11 in 73% yield. Final de- spectrum is depicted in Figure 2. In the expected mass
protection of compound 11 first with NaOH in methanol range, it shows just one signal corresponding to oQ (Fig-
to cleave the base-labile groups followed by treatment with ure 2, b) with a retention time of t = 34.1 min. Setting the
the HF·NEt3 complex to remove the TBS groups provided mass filter around the molecular weight of Q (m/z =
the final compound oQ in an excellent overall yield and 410.1640–410.1730) provided no signal, which proved its
purity after HPLC purification. The total synthesis of oQ absence under minimal medium conditions. When, however,
(1) was achieved in a total of 18 steps with an overall yield the E. coli were grown in full medium, only the formation
of 0.4% [starting from d-(–)-ribose, bromoacetaldehyde di- of Q (2) was detected (Figure 2, a), which again established
ethyl acetal (5) and pyrimidinone 6].
oQ as the direct biosynthesis precursor. We next spiked syn-
4484
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 4483–4485

Total Synthesis of Epoxyqueuosine
Acknowledgments
The authors thank the Deutsche Forschungsgemeinschaft (DFG)
(grant number CA275 8/4 and DFB749) for financial support. In-
frastructure was provided by the Excellence Cluster ECX114. The
authors thank Daniel Globisch for mass spectrometric studies.
[1] H. Kasai, K. Nakanishi, R. D. Macfarlane, D. F. Torgerson, Z.
Ohashi, J. A. McCloskey, H. J. Gross, S. Nishimura, J. Am.
Chem. Soc. 1976, 98, 5044–5046.
[2] B. Frey, J. McCloskey, W. Kersten, H. Kersten, J. Bacteriol.
1988, 170, 2078–2082.
[3] R. K. Slany, M. Bösl, P. F. Crain, H. Kersten, Biochemistry
1993, 32, 7811–7817.
[4] S. G. Van Lanen, S. D. Kinzie, S. Matthieu, T. Link, J. Culp,
D. Iwata-Reuyl, J. Biol. Chem. 2003, 278, 10491–10499.
[5] F. Harada, S. Nishimur, Biochemistry 1972, 11, 301–308.
[6] D. W. Phillipson, C. G. Edmonds, P. F. Grain, D. L. Smith,
D. R. Davis, J. A. McCloskey, J. Biol. Chem. 1987, 262, 3462–
3471.
Figure 2. (a) HR-mass signal of Q after enzymatic digest of bulk
tRNA extracted from E. coli under full medium conditions.
(b) HR-mass signal of oQ after enzymatic digest of an E. coli ex-
tract obtained under minimal medium conditions. (c) HR-mass sig-
nal of oQ after enzymatic digest under minimal medium conditions
with synthetic oQ spiked into the sample.
thesized oQ (1) into the digest obtained under minimum
medium conditions and again analyzed the nucleoside mix-
ture by LC–MS. In the mass range for oQ, we detected just
one signal (Figure 2, c), which firmly established that the
synthesized compound and the natural product are iden-
tical.
[7] S. Daoud Kinzie, B. Thern, D. Iwata-Reuyl, Org. Lett. 2000, 2,
1307–1310.
[8] Z. D. Miles, R. M. McCarty, G. Molnar, V. Bandarian, Proc.
Natl. Acad. Sci. USA 2011, 108, 7368–7372.
[9] J. P. Reyniers, J. R. Pleasants, B. S. Wostmann, J. R. Katzel,
W. R. Farkas, J. Biol. Chem. 1981, 256, 11591–11594.
[10] R. C. Morris, M. S. Elliott, Mol. Genet. Metab. 2001, 74, 147–
159.
[11] F. Klepper, E.-M. Jahn, V. Hickmann, T. Carell, Angew. Chem.
2007, 119, 2377–2379; Angew. Chem. Int. Ed. 2007, 46, 2325–
2327.
Conclusions
[12] T. Brückl, I. Thoma, A. J. Wagner, P. Knochel, T. Carell, Eur.
J. Org. Chem. 2010, 6517–6519.
[13] G. B. Payne, J. Org. Chem. 1962, 27, 3819–3822.
[14] H. Ovaa, J. D. C. Codée, B. Lastdrager, H. S. Overkleeft, G. A.
van der Marel, J. H. van Boom, Tetrahedron Lett. 1998, 39,
7987–7990.
In conclusion, we report here the first total synthesis of
the hypermodified tRNA nucleoside oQ (1), which is the
direct biosynthetic precursor of Q (2). With the help of the
synthetic material we confirmed the structure of oQ. The
synthetic availability of oQ now paves the way for a detailed
enzymatic analysis of the mechanism employed by the vita-
min B12 dependent enzyme QueG.
[15] F. C. Neidhard, P. L. Bloch, D. F. Smith, J. Bacteriol. 1974,
119, 736–747.
[16] T. Brückl, D. Globisch, M. Wagner, M. Müller, T. Carell, An-
gew. Chem. 2009, 121, 8074–8077; Angew. Chem. Int. Ed. 2009,
48, 7932–7934.
Supporting Information (see footnote on the first page of this arti-
cle): Experimental procedures and spectroscopic data for the com-
pounds prepared.
Received: April 23, 2013
Published Online: June 7, 2013
Eur. J. Org. Chem. 2013, 4483–4485
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4485