Reexamination of neomycin B degradation: Efficient preparation of its CD and D rings as protected glycosyl donors

Article abstract of DOI:10.1021/ol026548n

(matrix presented) The degradation of neomycin B was reexamined, and a novel protocol was established to prepare the properly masked neomycin CD ring as a glycol donor in excellent yield. Glycosylation of the CD ring with glycol acceptors provided a facil

Full text of DOI:10.1021/ol026548n

ORGANIC
LETTERS
2002
Vol. 4, No. 20
3455-3458
Reexamination of Neomycin B
Degradation: Efficient Preparation of Its
CD and D Rings as Protected Glycosyl
Donors
,†
Baogen Wu,† Jun Yang, Yun He,* and Eric E. Swayze
Ibis Therapeutics, A DiVision of Isis Pharmaceuticals, Inc., 2292 Faraday AVenue,
Carlsbad, California 92008
yhe@gnf.org
Received July 17, 2002
ABSTRACT
The degradation of neomycin B was reexamined, and a novel protocol was established to prepare the properly masked neomycin CD ring as
a glycol donor in excellent yield. Glycosylation of the CD ring with glycol acceptors provided a facile access to versatile intermediates that
could be utilized to synthesize a variety of novel neomycin B mimetics for RNA recognition.
Recently, there has been an increasing interest in the potential
application of RNAs as targets for small molecular drug
discovery.1 Like proteins, RNAs offer a diversity of three-
dimensional folds to allow the selective binding of effector
molecules.2 Among the drugs obtained from natural sources
that have been shown to work by binding to RNA or RNA/
protein complexes, special attention has been paid to the class
of aminoglycoside antibiotics. These natural aminosugars
have been shown to interact with a number of functional
RNA molecules including group I introns,3 hammerhead
ribozymes,4 and the human hepatitis δ virus ribozyme.5 For
example, neomycin B (Figure 1) induces translational
miscoding, most likely as a result of its interaction with 16S
rRNA in the A site of the ribosome.6 As methods7,8 have
been developed to study small molecule-RNA interactions,
† Current address: Medicinal Chemistry, GNF, 10675 Johns Hopkins
Dr., San Diego, CA 92121.
(1) (a) Tor, Y. Angew. Chem., Int. Ed. 1999, 38, 1579. (b) Hermann, T.
Angew. Chem., Int. Ed. 2000, 39, 1890. (c) Sucheck, S. J.; Wong, C.-H.
Curr. Opin. Chem. Biol. 2000, 4, 678.
(2) (a) Moore, P. B. Annu. ReV. Biochem. 1999, 67, 287. (b) Batey, R.
T.; Rambo, R. P.; Doudna, I. A. Angew. Chem., Int. Ed. 1999, 38, 2326.
(b) Hermann, T.; Patel, D. J. J. Mol. Biol. 1999, 294, 829.
(3) (a) Von Ahsen, U.; Davies, J.; Schreder, R. Nature 1991, 353, 368.
(b) Von Ahsen, U.; Davies, J.; Schroeder, R. J. Mol. Biol. 1992, 226, 935.
Figure 1. Structure of neomycin B.
10.1021/ol026548n CCC: $22.00 © 2002 American Chemical Society
Published on Web 09/07/2002

an increasing number of searches for specific RNA binders
and therapeutic agents have been conducted by rational
design and/or library synthesis.9,10 Most recently, Wong and
co-workers synthesized novel neamine dimers that bind to
16S RNA with high affinity and exhibited significantly
increased antibiotic activity.9a
by multiple byproducts when translated to non-milligram
scales (Scheme 1).
Scheme 1. Degradation of 4 to CD Ring Fragmenta
In a search for new antibacterial agents that target bacterial
ribosomal RNAs, we have initiated a program to synthesize
aminosugar mimetics with simpler structure and improved
biological properties.11 Although studies by Wong and co-
workers have suggested that the CD ring contributes
significantly to the binding affinity (KD ) 0.019 µM for 1
and KD ) 7.8 µM for 2) and specificity (1 is 5-fold more
specific than 2) of neomycin for the target RNA,9e most
previous studies on neomycin B analogues have maintained
the core AB ring structure and eliminated or modified the C
and/or D rings.9b,c,e,10b,e,h To address the relative contributions
to neomycin activity made by the various ring components,
we became interested in the synthesis of neomycin analogues
having the CD ring structure maintained and a variety of
carbohydrate and non-carbohydrate moieties substituted for
the A and/or B rings. Although many studies on neomycin
B and related aminosugars have been reported in the
literature,12 no practical preparation of the neomycin B CD
ring has been documented. We have described the prepara-
tion of the CD fragment from neomycin B with hydrochloric
acid by modifying the literature procedure.13 Unfortunately,
the yield for the CD ring was variable and often accompanied
(4) Stage, T. K.; Hertel, K. J.; Uhlenbeck, O. C. RNA 1995, 1, 95.
(5) (a) Rogers, J.; Chang, A. H.; Von Ahsen, U.; Schroeder, R.; Davies,
J. J. Mol. Biol. 1996, 259, 916. (b) Chia, J.-S.; Wu, H.-L.; Wang, H.-W.;
Chen, D.-S.; Chen, P.-J. J. Biomed. Sci. 1997, 4, 208.
(6) (a) Moazed, D.; Noller, H. F. Nature 1987, 327, 389. (b) Purohit, P.;
Stem, S. Nature 1994, 370, 650. (c) Formy, D.; Recht, M. I.; Blanchard, S.
C.; Puglisi, J. D. Science 1996, 1367.
(7) (a) Griffey, R. H.; Greig, M. J. H.; An, H.; Sasmor, S.; Manalili, S.
J. Am. Chem. Soc. 1999, 121, 474. (b) Griffey, R. H.; Hofstadler, S. A.;
Sannes-Lowery, K. A.; Ecker, S. T.; Crooke, S. T. Proc. Natl. Acad. Sci.
U.S.A. 1999, 96, 10129.
a Reagents and conditions: (a) TfN3 (10 equiv), CuSO4, Et3N,
MeOH, H2O; (b) Ac2O (50 equiv), Py, DMAP (cat.), 82% from 1;
(c) 1.0 N HCl, MeOH-dioxane, reflux; (d) TolSH (1.1 equiv),
BF3‚OEt2 (3.0 equiv), 25 °C, 15 h, 6, 82%; (e) TolSH (2.0 equiv),
BF3‚OEt2 (3.0 equiv), 25 °C, 15 h, 5, 88%; 6, 45%; 8, 33%; 9,
∼5%.
(8) Hendrix, M.; Priestley, E. S.; Joyce, G. F.; Wong, C.-H. J. Am. Chem.
Soc. 1997, 119, 3641.
We report herein a new protocol for the degradation of
neomycin B that allows us to efficiently and reproducibly
access the CD and D ring fragments of neomycin B in
quantity. Furthermore, this protocol conveniently provides
the CD and D rings as thioglycosides with proper protections
for the amino and hydroxy groups in place, making them
amenable to subsequent glycosylations.
(9) (a) Sucheck, S. J.; Wong, A. L.; Koeller, K. M.; Boehr, D. D.; Draker,
K.; Sears, P.; Wright, G. D.; Wong, C.-H. J. Am. Chem. Soc. 2000, 122,
5230. (b) Sucheck, S. J.; Greenberg, W. A.; Tolbert, T. J.; Wong, C.-H.
Angew. Chem., Int. Ed. 2000, 39, 1080. (c) Greenberg, W. A.; Priestley, E.
S.; Sears, P. S.; Alper, P. B.; Rosenbohm, C.; Hendrix, M.; Hung, S. C.;
Wong, C.-H. J. Am. Chem. Soc. 1999, 121, 6527. (d) Wong, C.-H.; Hendrix,
M.; Manning, D. D.; Rosenbohm, C.; Greenberg, W. A. J. Am. Chem. Soc.
1998, 120, 8319. (e) Alper, P. B.; Hendrix, M.; Sears, P.; Wong, C.-H. J.
Am. Chem. Soc. 1998, 120, 1965.
Initially, perazidoperacetylneomycin (4) was chosen as our
starting material for the study of cleavage conditions.
Compound 4 can conveniently be prepared from inexpensive
neomycin B by azido transfer followed by acetylation
(Scheme 1).9c Under acidic conditions (1.0 N HCl-MeOH-
dioxane, Table 1), 4 was decomposed slowly at room
temperature, giving neither the desired CD ring product 5,
AB ring product 6 nor their related de-acylated products
according to LC-MS analysis. Similar results were obtained
by employing sulfuric acid. We then turned our attention to
Lewis acid mediated degradation conditions. Treatment of
4 with 1,3-propanedithiol and SnCl4 in methylene chloride
resulted in slow decomposition of the starting material at
-78 °C. Under similar conditions with different Lewis acids
(10) (a) Luedtke, N. W.; Baker, T. J.; Goodman, M.; Tor, Y. J. Am.
Chem. Soc. 2000, 122, 12035. (b) Kirk, S. R.; Luedtke, N. W.; Tor, Y. J.
Am. Chem. Soc. 2000, 122, 980. (c) Michael, K.; Wang, H.; Tor, Y. Bioorg.
Med. Chem. 1999, 7, 1361. (d) Wang, H.; Tor, Y. Angew. Chem., Int. Ed.
Engl. 1998, 37, 109. (e) Wang, H.; Tor, Y. Bioorg. Med. Chem. Lett. 1997,
7, 1951. (f) Tok, J. B.-H.; Fenker, J. Bioog. Med. Chem. Lett. 2001, 11,
2987. (g) Hamasaki, K.; Ueno, A. Bioog. Med. Chem. Lett. 2001, 11, 591.
(h) Hanessian, S.; Tremblay, M.; Kornienko, A.; Moitessier, N. Tetrahedron
2001, 57, 3255. (i) Hamasaki, K.; Woo, M.-W.; Ueno, A. Tetrahedron Lett.
2000, 41, 8327.
(11) Ding, D.; Hofstadler, S. A.; Swayze, E. E.; Griffey, R. H. Org.
Lett. 2001, 3, 1621.
(12) (a) Umezawa, S.; Tsuchiya, T. In Aminoglycoside Antibiotics;
Humezawa, H., Hooper, I. R., Eds.; Springer-Verlag: Berlin, Heidelberg,
New York, 1982; pp 37-110. (b) See refs 9-11 and references therein.
(13) (a) Ding, Y.; Swayze, E. E.; Hofstadler, S. A.; Griffey, R. H.
Tetrahedron Lett. 2000, 41, 4049. (b) Rinehart, K. L., Jr. The Neomycins
and Related Antibiotics; John Wiley and Sons: New York, 1964; pp 93-
97.
3456
Org. Lett., Vol. 4, No. 20, 2002

Scheme 2. Degradation of 10 to CD and D Ring Fragmentsa
Table 1. Degradation of 4 under Various Conditions
conditions
5
6
1.0 N HCl-MeOH-dioxane
HS(CH2)3SH, SnCl4
HS(CH2)3SH, TiCl4
HS(CH2)3SH, BF3‚OEt2
EtSH, SnCl4
EtSH, BF3‚OEt2
TolSH, SnCl4
TolSH, TMSOTf
TolSH, BF3‚OEt2
dec
dec
dec
dec
dec
dec
dec
dec
dec
25%
82%
<10%
65%
25%
30%
70%
75%
90%
such as TiCl4, SnCl4, and BF3‚OEt2 using either dithiol,
ethylthiol, or TolSH gave only AB ring product at -78 °C
with CD ring fragment decomposed. When TMSOTf was
employed, the reaction provided the desired CD ring product
in 25% yield at -30 °C. The best result was obtained by
treating 4 with 3.0 equiv of BF3‚OEt2 and 1.1 equiv of TolSH
in methylene chloride (see the Supporting Information).
Under these conditions, the reaction proceeded smoothly to
give both the AB (5) and CD (6) ring fragments in excellent
yields.
Since the products have very similar Rf values on TLC,
the crude reaction mixture was acetylated to convert the AB
ring fragment 5 to 7. Compounds 6 and 7 could then be
readily separated by silica column chromatography. The
chemical structure of 6 was established by various spectro-
scopic analyses, and the chemical shifts of protons in 6 were
assigned according to its 2D NMR spectrum (see the
Supporting Information). Following this procedure, glycosyl
donor 6 could be prepared in gram quantities. Interestingly,
if 2.0 equiv of TolSH was used, the CD ring fragment 6
began to degrade, and a mixture of 5 (88%), 6 (45%), 8
(33%), and 9 (∼5%) was obtained.
a Reagents and conditions: (a) NaOMe, MeOH; (b) BnBr, NaH,
DMF, 76% from 4; (c) 1.0 M HCl, MeOH, dioxane, reflux, 14 h,
11, 82%; 12, 20%; 13 (R/â ) 1:2), 56%; (d) TolSH (1.2 equiv),
BF3‚OEt2 (3.0 equiv), 0 °C, 15 min, 88%; (e) TolSH (1.5 equiv),
BF3‚OEt2 (3.0 equiv), 25 °C, 10 h, 14 (R/â ) 1:2), 66%; (f) TolSH
(2.0 equiv), BF3‚OEt2 (3.0 equiv), 25 °C, 2 h, 15, 37%; 16, 25%;
(g) TolSH (3.0 equiv), BF3‚OEt2 (5.0 equiv), 25 °C, 4 h, 15, 34%;
16, 23%; (h) TolSH (3.0 equiv), BF3‚OEt2 (3.0 equiv), 25 °C, 4 h,
15, 23%; 16, 18%.
Wong et al.9c mentioned the formation of the CD ring as
a minor byproduct when degrading perazidoperbenzylneo-
mycin 10 under acidic conditions. In our hands, when 10
was refluxed in a HCl solution of MeOH and dioxane, it
gave the AB ring fragment 11 as well as the CD ring
fragment as a mixture of two anomers, which could be
separated via careful silica gel chromatography (12 and 13;
Scheme 2). This is in contrast to the methanolysis of 1 or 4,
which resulted in decomposition of the CD ring fragment.
As expected, prolonged heating of the reaction mixture
resulted in decomposition of CD ring products and gave
lower yields.
identical yield from a mixture of 12 and 13. Furthermore,
15 and 16 could be obtained directly from 10, albeit in
slightly reduced yields.
We then tested the reactivity of donor 6 toward a simple
glycosyl acceptors (Scheme 3). The glycosylation reaction
of 6 with methyl R-hydroxyacetate in the presence of NIS
and a catalytic amount of AgOTf proceeded smoothly to give
product 17 as the only anomer in 76% yield.14 Derivatives
such as 17 may serve as appropriate key intermediates in
the future library synthesis of neomycin mimetics with
modified AB rings.
Treatment of either 12 or 13 with BF3‚OEt2 and 1.1 equiv
of TolSH at 0 °C for 15 min led to the isolation of 14 (R/â
) 1:2) in over 88% yield. We subsequently discovered that
donor 14 could be obtained directly from 10 by using the
same conditions as for the degradation of 4. Thus, treatment
of 10 with BF3‚OEt2 and 1.2 equiv of TolSH led to the
formation of 14 (R/â ) 1:2) in 66% yield in one simple
operation. Use of excess BF3‚OEt2 (5 equiv) reduced the
yield of 14 due to the opening of D ring. When 14 was
allowed to react with BF3‚OEt2 and TolSH at 25 °C, the
thioglycosides 15 and 16 derived from D ring were then
obtained. Similarly, 15 and 16 were obtained in nearly
Org. Lett., Vol. 4, No. 20, 2002
In summary, we have established a simple and efficient
method for the preparation of the CD and D ring derivatives
(14) Lahmann, M.; Oscarson, S. Org. Lett. 2000, 2, 3881.
3457

acceptors and could serve as key intermediates for the
synthesis of neomycin mimetics with modifications on the
A and AB ring moieties. Additionally, the protected neo-
mycin B D ring deriviatives 15 and 16 were obtained in a
single step from the readily available 10. Compounds of this
type are relatively difficult to access otherwise. The use of
Lewis acids in concert with thiol scavengers may also be
applicable to the preparation of other aminosugar derivatives
via the degradation of aminoglycosides. The use of these
intermediates for the synthesis of libraries of neomycin
mimetics will be reported in due course.
Scheme 3. Glycosylation of 6a
a Reagents and conditions: (a) NIS (1.3 equiv), AgOTf (cat.),
HOCH2CO2Me (10 equiv) methylene chloride, 4 Å MS, 0 °C, 2 h,
76%.
of neomycin B. Protected derivatives of neomycin such as
4 and 10 have been prepared on a multigram scale and
extensively utilized as a source of the AB ring fragments by
the Wong group.9c The degradation conditions described
herein now allow access to the CD ring derivative 6 and 14
in gram quantity. These synthons react readily with glycosyl
Supporting Information Available: Experimental pro-
cedures, spectral data for representative compounds (6, 8,
and 12-17). This material is available free of charge via
the Internet at http://pubs.acs.org.
OL026548N
3458
Org. Lett., Vol. 4, No. 20, 2002