Investigation of the regioselectivity for the Staudinger reaction and its application for the synthesis of aminoglycosides with N-1 modification

Article abstract of DOI:10.1021/jo062588j

(Chemical Equation Presented) The criteria for controlling the regioselectivity of Staudinger reduction of azides have been investigated. These findings enable a convenient direct N-1 modification of the perazidoneamine and perazidoribostamycin resulting

Full text of DOI:10.1021/jo062588j

Investigation of the Regioselectivity for the Staudinger Reaction and
Its Application for the Synthesis of Aminoglycosides with N-1
Modification
Jie Li, Fang-I Chiang, Hsiao-Nung Chen, and Cheng-Wei Tom Chang*
Department of Chemistry and Biochemistry, Utah State UniVersity, 0300 Old Main Hill,
Logan, Utah 84322-0300
chang@cc.usu.edu
ReceiVed December 17, 2006
The criteria for controlling the regioselectivity of Staudinger reduction of azides have been investigated.
These findings enable a convenient direct N-1 modification of the perazidoneamine and perazidoribo-
stamycin resulting in the synthesis of aminoglycoside antibiotics with activity against drug-resistant bacteria.
Introduction
cosides has been extensively studied, especially in neomycin,
kanamycin, and neamine. In general, these methods utilize
metal-mediated chelation,5-9 cyclization between the selected
carbamate-protected amino group and its vicinal hydroxyl
group10-12 or enzymatic approaches.13 However, selectivity of
these strategies closely depends on the scaffolds of parent
aminoglycosides. In addition, there is no applicable method for
regioselective differentiation of the azido groups on perazi-
doaminoglycosides.
Ever since the discovery of aminoglycoside antibiotics, their
challenging synthesis has attracted much effort. Pioneered by
Umezawa and others in the 1970s, a vast amount of aminogly-
coside derivatives along with various modification methods have
been created aiming to improve the activity of aminoglycoside
against resistant bacteria.1-3 A wide range of structural modi-
fications have been documented in the literature, among which
attaching functionalities at the N-1 position of the 2-deoxy-
streptamine among kanamycin or neomycin class antibiotics is
one of the most effective methods of reviving the activity against
aminoglycoside-resistant bacteria.4 This strategy has led to the
development of semisynthetic amikacin that has an (S)-4-amino-
2-hydroxybutyryl (AHB) group at N-1 position.
Results and Discussion
Our group has devoted efforts into the synthesis of novel
aminoglycosides using glycodiversification strategy based on
(5) Tsuchiya, T.; Takagi, Y.; Umezawa, S. Tetrahedron Lett. 1979, 51,
4951-4954.
(6) Nagabhushan, T. L.; Cooper, A. B.; Turner, W. N.; Tsai, H.;
McCombie, S.; Mallams, A. K.; Rane, D.; Wright, J. J.; Reichert, P.; Boxler,
D. L.; Weinstein, J. J. Am. Chem. Soc. 1978, 100, 5253-5254.
(7) Roestamadjli, J.; Grapsas, I.; Mobashery, S. J. Am. Chem. Soc. 1995,
117, 11060-11069.
(8) Umezawa, H.; Miyasaka, T.; Iwasawa, H.; Ikeda, D.; Kondo, S. J.
Antibiot. 1981, 34, 1635-1640.
(9) Shitara, T.; Umemura, E.; Tsuchiya, T.; Matsuno, T. Carbohydr. Res.
1995, 276, 75-89.
(10) Kumar, V.; Remers, W. A. J. Org. Chem. 1978, 43, 3327-3331.
(11) Umezawa, S.; Tsuchiya, T.; Torii, T. J. Antibiot. 1982, 35, 58-61.
(12) Sharma, M. N.; Kumar, V.; Remers, W. A. J. Antibiot. 1982, 35,
905-910.
(13) Cappelletti, L. M.; Spagnoli, R.; Spa, P. J. Antibiot. 1983, 36, 328-
330.
In order to implement an AHB group at N-1 position, a
regioselective differentiation of amino groups is essential. The
regioselective manipulation of the amino groups on aminogly-
* To whom correspondence should be addressed. Fax: 435-797-3390.
(1) Umezawa, S.; Tsuchiya, T. In Aminoglycoside Antibiotics; Umezawa,
H., Hooper, I. R., Ed.; Springer-Verlag: New York, 1982; pp 37-110.
(2) Hanessian, S.; Haskell, T. H. In Antibiotics Containing Sugars;
Pigmann, H., Horton, D., Eds.; Academic Press: New York/London, 1970;
pp 139-211.
(3) Haddad, J.; Kotra, L. P.; Mobashery, S. In Glycochemistry Principles,
Synthesis, and Applications; Wang, P. G., Bertozzi, C. R., Eds.; Marcel
Dekker, Inc.: New York/Basel, 2001; pp 307-424.
(4) Hanessian, S.; Kornienko, A.; Swayze, E. E. Tetrahedron 2003, 59,
995-1007.
10.1021/jo062588j CCC: $37.00 © 2007 American Chemical Society
Published on Web 04/28/2007
J. Org. Chem. 2007, 72, 4055-4066
4055

Li et al.
SCHEME 1. Strategy for Regioselective Reduction of Azido
Group
acyl protecting group at O-5, and an anisotropic effect has been
proposed to offer the shielding of H-2′ based on molecular
modeling.17
However, it is still uncertain to us that such an upfield shift
of H-2′ resulting from the anisotropic effect of the O-5 protecting
group can be accountable for the observed regioselectivity. The
upfield shift of H-2′ caused by anisotropic effect is different
from the one caused by inductive effect. The latter can be
correlated to the electron-deficiency of azido group while the
former may not. The difference raises the issue for the role of
the electron-withdrawing group at the O-5 position. Is the
selectivity of the Staudinger reaction controlled solely by an
electron-withdrawing (electronic) effect, or perhaps, does steric
effect have influence as well? Clearly, the electron-withdrawing
effect is important which explains that no reduction from the
N-6′ azido group on several known azidoneamine derivatives
has yielded an identifiable product. On the other hand, the
electron-withdrawing effect from the O-5 acyl group seems to
be too far away from N-1 azido group to have a significant
contribution to the observed regioselectivity.
SCHEME 2. Reported Regioselective Reduction of Azido
Group18
To unravel the role of the O-5 protecting group, four
tetraazidoneamine derivatives, 2,28 3, 4,21 and 517 were studied
(Scheme 3). Compound 3 has four acetyl groups at O-5, O-6,
O-3′, and O-4′ positions. If an electron-withdrawing effect is
the only factor for the selectivity of Staudinger reaction, the
N-2′ azido group should be the most reactive one since it is the
closest to the anomeric center and should be the most electron-
deficient azide. If the N-1 azido group is the most reactive one,
the primary role for the acyl group at O-5 should be the provider
of steric hindrance that prevents the reduction of the N-2′ azido
group. Interestingly, H-2′ appears to have the lower chemical
shift (δ ) 3.36 ppm) as compared to H-1 and H-3 of 3. This
discrepancy suggests that the upfield shift of H-2′ is a result of
anisotropic effect from the O-5 protecting group and cannot be
interpreted as the electron deficiency for the N-2′ azido group.
When compound 3 was subjected to the treatment of one-
pot Staudinger reduction/Boc-protection following the reported
condition,17,19 three components were obtained (Scheme 4). The
reaction mixture was purified as a two-component mixture in
49% yield and compound 8 (8% yield). Compound 8 was
tetraazidoneamine scaffold.14,15 With the advantages of better
solubility and accessibility for more structural variations, per-
azidoaminoglycosides have been the recent focus for generating
new aminoglycoside derivatives. We have been actively engag-
ing in the studies for the regioselective reduction of azido group
via a Staudinger reaction. It has been suggested that the electron-
deficient azides can be reduced more rapidly than the electron-
rich azides.16 We have developed a strategy that utilizes the
electron-donating protecting groups at O-3′ and O-4′ positions
(ring I) and the electron-withdrawing protecting groups at O-5
and O-6 positions (ring II) of tetraazidoneamine to enable a
regioselective Staudinger reduction of azido group at N-1
position (Scheme 1).17
It has also been reported by Wong and co-workers that the
regioselectivity of the azide reduction can be achieved for
several perazido compounds.18 The observed regioselectivity was
interpreted with the electronic factors, and the chemical shifts
of ipso-protons of azides were used to correlate to the predicted
site of azide reduction. In general, the higher the chemical shift,
the higher the reactivity toward the Staudinger reduction
(Scheme 2).
1
determined to be N-2′ Boc-protected adduct by H-1H COSY.
Further derivatization of the two-component mixture (6 and 7)
via deacetylation, diol-protection, and acetylation furnished
compounds 11 and 12, which were separated by flash column
chromatography. Unfortunately, the regioselectivity of com-
pounds 11 and 12 cannot be identified due to the complex
overlapping of their proton signals. Deprotection of the Boc
group on compounds 11 and 12 followed by coupling with the
AHB derivative, 13,20 led to the synthesis of neamine deriva-
tives, 14 and 15 with AHB at N-1 and N-3, respectively.
However, the signals for H-1 and H-3 necessary for confirming
the regioselectivity were still indiscernible. Finally, acetylation
at O-5 and O-6 created sufficient difference in chemical shifts
allowing the determination of regioselectivity. Compound 16
was verified to have an AHB group at N-1 while 17 was
determined to have an AHB group at N-3.
We are intrigued by the above-reported results for the distinct
reactivity of N-2′ azido groups. Although the reported chemical
shifts of both ipso-protons can be used to explain the observed
selectivity, the unusual upfield shift of the H-2′ of per-azido
per-benzyl tobramycin needs to be further clarified. We have
observed similar upfield shift for the H-2′ in the presence of
(14) Elchert, B.; Li, J.; Wang, J.; Hui, Y.; Rai, R.; Ptak, R.; Ward, P.;
Takemoto, J. Y.; Bensaci, M.; Chang, C.-W. T. J. Org. Chem. 2004, 69,
1513-1523.
(15) Wang, J.; Li, J.; Chen, H.-N.; Chang, H.; Tanifum, C. T.; Liu, H.-
H.; Czyryca, P. G.; Chang, C.-W. T. J. Med. Chem. 2005, 48, 6271-6285.
(16) Bayley, H.; Standring, D. N.; Knowles, J. R. Tetrahedron Lett. 1978,
3633-3634.
(17) Li, J.; Chen, H.-N.; Chang, H.; Wang, J.; Chang, C.-W. T. Org.
Lett. 2005, 7, 3061-3064.
(18) Nyffeler, P. T.; Liang, C.-H.; Koeller, K. M.; Wong, C.-H. J. Am.
Chem. Soc. 2002, 124, 10773-10778.
Compound 2 was designed to have no electron-withdrawing
advantage as a comparison to compound 3. Compound 5 was
(19) Ariza, X.; Vrpi, F.; Viladomat, C.; Vilarrasa, J. Tetrahedron Lett.
1998, 39, 9101-9102.
(20) Moon, M. S.; Jun, S. J.; Lee, S. H.; Cheong, C. S.; Kim, K. S.;
Lee, B. S. Bull. Korean Chem. Soc. 2003, 24, 163-164.
4056 J. Org. Chem., Vol. 72, No. 11, 2007

RegioselectiVity for the Staudinger Reaction
SCHEME 3. Synthesis of Azidoneamine Derivatives
SCHEME 4. Regioselective Staudinger Reduction of Azidoneamine Derivatives
designed to have an electron-withdrawing effect for N-1 azide
but not for N-2′ azide. Meanwhile, there is no O-5 protecting
group for blocking the access to the N-2′ azido group. Staudinger
reduction of compounds 2 and 5 produced a complex mixture
with no dominant identifiable product. Compound 4 was
designed to have an electron-withdrawing effect for both N-1
and N-2′ azido groups. However, N-2′ azido group is expected
to be more reactive than N-1 since it is closer to the anomeric
center. When compound 4 was treated in the identical condition,
two dominant products were isolated (Scheme 5). Spectroscopic
characterization confirmed that these were N-2′ Boc-protected
amino (18, 21%) and N-2′ amino (19, 27%) neamine derivatives.
By combining the results from the regioselective Staudinger
reduction, the role of the O-5 protecting group and the factors
that govern the regioselective azide-amino transformation can
be established. First: the electronic effect (electron deficiency)
is the dominant factor in controlling the regioselectivity of azide
reduction. However, the steric hindrance may play an auxiliary
but essential role in controlling the regioselectivity. Comparing
the results from the Staudinger reaction of 3 and 4 clearly
J. Org. Chem, Vol. 72, No. 11, 2007 4057

Li et al.
SCHEME 5. Staudinger Reduction of Compound 4
demonstrates that the steric hindrance of the O-5 protecting
group prevents the reduction of the N-2′ azido group. The
electron-withdrawing group at O-6 is pivotal in making the N-1
azido group the most reactive one. In order to achieve the
favored regioselectivity, i.e., reduction of the N-1 azido group,
it is preferable to make ring I electron-rich by employing
electron-donating protecting groups, such as Bn, at O-3′ and
O-4′, and ring II electron-deficient by introducing electron-
withdrawing protecting groups, such as Ac or Bz, at both O-5
and O-6. Second: although the chemical shifts of ipso-protons
can provide guidance in the favored site of azide reduction, their
significance in regioselectivity needs to be carefully interpreted.
The anisotropic effect resulting from the aromatic ring (Bn or
Bz) and/or carbonyl functionality of the O-5 protecting groups
that causes the upfield shift of H-2′ cannot be correlated with
the lower reactivity of N-2′ azido group. Third: even in the
optimal designs (electron-rich ring I and electron-deficient ring
II), minor reduction at sites other than N-1 azido group could
be inevitable, making only modest yields for the desired N-1
reduced product. In the event of lacking dominant electronic
and steric effects, the Staudinger reduction is expected to occur
at multiple sites creating a complex mixture and, hence,
difficulties in isolating identifiable product.
These findings can now be used to explain the observed
selectivity reported by Wong’s and our group. For example,
for 6,3′,4′-tri-O-benzyltetraazidoneamine, there is no protecting
group at O-5 to provide steric hindrance for blocking the
reduction of N-2′ azide. Meanwhile, there is no electron-
withdrawing protecting group at O-6 to favor the reduction at
N-1 azide. Thus, it is conceivable that N-2′ will be the dominant
site of Staudinger reduction. For per-azido per-benzyl tobra-
mycin, the upfield shift of H-2′ is the result of anisotropic effect
from O-5 Bn. The potential reduction of N-2′ azide is blocked
by the presence of O-5 Bn group. The N-3′′ azide is most
reactive (or most electron-deficient), which can be correlated
to the observed chemical shift.
For compound 2, there is no O-6 electron-withdrawing group
to favor the reduction of N-1 azido group despite having an
O-5 Bn for blocking the reduction of N-2′ azide. Thus, random
reduction at multiple sites may occur, creating a complex
mixture and no identifiable product. The lack of steric hindrance
from O-5 protecting group in compound 5 causes the competi-
tive reductions between N-2′ and N-1 azido groups. Therefore,
the desired N-1 reduced product could not be produced
predominantly and then isolated. Finally, with the designs like
compound 3 and those shown in Scheme 1, a regioselective
reduction at N-1 azide can be achieved by a collaboration of
both electron-withdrawing and steric effects.
Synthesis of Butirosin B and Neokacin. After revealing the
criteria regarding the regioselectivity of the Staudinger reaction
of per-azido neamine, we began to explore the possibility of
introducing the AHB group via a regioselective reduction of
the N-1 azido group for structurally more complicated per-azido
neomycin class derivatives. Neomycin class aminoglycosides
contain a characteristic 4,5-disubstituted neamine core (ring II)
(Figure 1).
FIGURE 1. Structures of neomycin class aminoglycosides.
Our study shows that the protecting group at O-5 is essential
in providing steric hindrance for favoring the regioselective
reduction of N-1 azido group. Thus, it is conceivable that the
furanose ring (ring III) at the O-5 of the neomycin class
antibiotics may exert the same effect. Making ring II electron-
deficient by introducing O-6 acyl protecting group is important
as well. Rather than incorporating an electron-donating group,
such as Bn, at ring I, it is easier synthetically to acylate all the
hydroxy groups.
Per-acetylated per-azido neomycin, such as 20,21 contains two
additional N-2′′′ and N-6′′′ azido groups at ring IV (Scheme
6). The N-6′′′ azido group, which is transformed from a primary
amino group that is analogous to the N-6′ azido group at ring
I, should not affect the outcome of selectivity. Since ring IV
1
adapts an C4 conformation, the N-2′′′ azido group resides at
an axial position. Substituent at the axial position is more
electron rich than that of the equatorial position.23 Additionally,
(21) Wu, B. G.; Yang, J.; He, Y.; Swayze, E. E. Org. Lett. 2002, 4,
3455-3458.
(22) (a) Knapp, S.; Shieh, W. C. Tetrahedron Lett. 1991, 32, 3627-
3630. (b) Antkowiak, W. Z.; Krzyzosiak, W. J. J. Chromatogr. 1974, 90,
399-401.
(23) The axial substituent is typically more shielded as compared to the
equatorial one as indicated by the chemical shifts of H-2ax and H-2eq in
this report. Similar results regarding the chemical shifts of methyl group
on methylcyclohexane can be found in the literature as well (Wiberg, K.
B.; Hammer, J. D.; Castejon, H.; Bailey, W. F.; DeLeon, E. L.; Jarret, R.
M. J. Org. Chem. 1999, 64, 2085-2095). However, computational analysis
of the electron density and electrostatic potential for the axial and equatorial
azido groups on cyclohexane reveals only slight difference favoring
equatorial azido group toward nucleophilic agent. Thus, we reason that a
combination of stereoelectronic effect and steric hindrance should contribute
to the lower reactivity of the N-2′′′ azido group.
4058 J. Org. Chem., Vol. 72, No. 11, 2007

RegioselectiVity for the Staudinger Reaction
SCHEME 6. Regioselective Staudinger Reduction of Azidoneomycin
the reactivity of N-2′′′ azido group was expected to be decreased
due to a 1, 3-diaxial interaction exerted by C-4′′′ acetyloxy
group. Thus, we expected the N-2′′′ azido group should be less
reactive than the N-1 azido group. Finally, despite having acetyl
groups at O-3′ and O-4′ making ring I electron-deficient, the
steric hindrance from the furanose ring (ring III) is expected to
hamper the reduction of N-2′ azido group. Although chemical
shifts of the ipso-protons of N-3 vs N-1 indicate that the former
should be more reactive, we were expecting that the reactivity
of N-3 could be decreased by the steric effect of ring I.
After treating 20 in the same condition, a mixture of N-1
and N-3 Boc-protected adducts, 21, was obtained, although the
N-1 Boc-protected adduct appeared to be the dominant product
by COSY analysis. Deprotection of the Boc followed by the
coupling with Cbz-protected AHB, 22,20 furnished a major
product as 23. Characterization of the major product confirms
the attachment of the AHB group at the desired N-1 position.
Global deprotection of the protecting groups furnished the
desired neomycin derivative with AHB at the N-1 position,
which we called neokacin.
Staudinger reduction can be achieved. Incorporation of Cbz-
protected AHB followed by the global deprotection procedures
yielded the ribostamycin adducts with N-1 AHB group, also
known as butirosin B.
Synthesis of Amikacin Derivatives. The newly developed
regioselective Staudinger reaction can also be applied for the
synthesis of kanamycin class aminoglycosides with N-1 AHB
group or amikacin analogues. Although we have previously
reported the synthesis of two amikacin analogues,17 JLN005
and JLN027, the newly developed protocol provides more
convergent and efficient route for the construction of a library
of amikacin analogues. Similar to the previous synthesis,17 the
glycosyl acceptor 14 was regiospecifically and stereoselectively
glycosylated with thiophenylglycosyl donors, 28a-c,15 that we
have prepared. The afforded glycosylated trisaccharides were
often mixed with inseparable impurities. Nevertheless, after the
hydrolysis of all the acyl protecting groups, three members of
the kanamycin derivatives, 29a-c, were obtained as a more
pure form and with improved R/â ratio (Scheme 8). Global
deprotection via Staudinger reduction and hydrogenation pro-
cedures led to the synthesis of three amikacin analogues,
JLN007, JLN033, and JLN040.
Because the commercially available ribostamycin is too
expensive to be used as the starting material, a glycosylation
strategy for the synthesis of per-azido per-O-acetylated ribo-
stamycin has been developed (Scheme 7). Glycosylation of 4
with 2422 provided the ribostamycin-based adduct, 25. Interest-
ingly, the chemical shifts of the ipso-protons of 25 also suggest
that the N-3 azido group should be the most reactive. However,
after treating 25 in the same condition, a regioselective
Confirming the Regioselectivity by COSY Experiment. As
mentioned previously, the regioselectivity of the Staudinger
reaction can be confirmed by 2D COSY experiment. Two
diagnostic spectroscopic signatures can be utilized to unambigu-
ously assign the site of azido reduction/Boc protection: (a) the
ipso-proton (geminal proton) where azido reduction/Boc protec-
J. Org. Chem, Vol. 72, No. 11, 2007 4059

Li et al.
SCHEME 7. Regioselective Staudinger Reduction of Azidoribostamycin
tion occur or where the AHB group is attached will have a
significant (ca. 0.5 ppm) downfield shift, (b) a COSY signal
between the ipso-proton (geminal proton) and the vicinal AHB-
NH proton. The chemical shifts for several selected compounds
in d-chloroform are summarized in Table 1 using the widely
adapted numbering system for neomycin and kanamycin class
aminoglycosides (Figure 2).
(pSF815) and E. coli (pTZ19U-3) are laboratory resistant strains
using E. coli (TG1) as the host. K. pneumoniae (ATCC 700603)
is a clinical isolate that is resistant to ceftazidime, other
â-lactams, and several aminoglycosides (ANT(2′′)).25 Pseudomo-
nas aeruginosa (ATCC 27853) that expresses several AME
including APH(3′)-IIb that render the bacterium manifests
modest resistance toward aminoglycosides.26 Methicillin-
resistant S. aureus (ATCC 33591) (MRSA) is the leading cause
of bacterial infections.27 Many MRSA strains contain genes
encoded for APH(3′), ANT(4′), and AAC(6′)/APH(2′′), which
render the bacteria resistant to many aminoglycosides.
On the basis of the antibacterial activity of the neomycin class
antibiotics, attaching AHB at N-1 is clearly an effective
modification to revive the antibacterial activity as shown in the
results from synthetic neokacin, butirosin B, and JT005 against
susceptible and resistant E. coli (entries 2, 3, and 4, Table 2).
More significantly, these synthetic aminoglycosides are active
against K. pneumoniae (ATCC 700603), which causes severe
pneumonia and renders multidrug resistance by harboring
extended-spectrum â-lactamases (ESBLs) (entry 5, Table 2).
All the synthesized kanamycin class aminoglycosides with an
N-1 AHB group also manifest prominently against this clinically
isolated K. pneumoniae.
FIGURE 2. Numbering system for neomycin and kanamycin classes
aminoglycosides.
Antibacterial Activity Study. The most common mode
bacterial resistance against aminoglycosides arises from the
overexpression of aminoglycoside-modifying enzymes (AME).
These enzymes are grouped as aminoglycoside phosphotrans-
ferases (APHs), aminoglycoside acetyltransferases (AACs), and
aminoglycoside nucleotidyltransferases (ANTs). Our newly
synthesized aminoglycosides were assayed against a panel of
bacteria, and the minimum inhibitory concentration (MIC) was
determined using amikacin, neomycin, butirosin A, gentamicin,
and kanamycin as the controls (Tables 2 and 3).24 Aminogly-
coside susceptible Escherichia coli (ATCC 25922), Staphylo-
coccus aureus (ATCC 25923), and Klebsiella pneumoniae
(ATCC 13883, resistant to ampicillin, susceptible to aminogly-
cosides) were used as standard reference strains. E. coli
(24) Methods for Dilution Antimicrobial Susceptibility Testing for
Bacteria that Grow Aerobically (approved standard M7-A5) and Perfor-
mance Standards for Antimicrobial Disk Susceptibility Tests (approved
standard M2-A7), National Committee for Clinical Laboratory Standards,
Wayne, PA.
(25) Rasheed, J. K.; Anderson, G. J.; Yigit, H.; Queenan, A. M.;
Domenech-Sanchez, A.; Swenson, J. M.; Biddle, J. W.; Ferraro, M. J.;
Jacoby, G. A.; Tenover, F. C. Antimicrob. Agents Chemother. 2000, 44,
2382-2388.
(26) Hachiler, H.; Santanam, P.; Kayser, F. H. Antimicrob. Agents
Chemother. 1996, 40, 1254-1256.
(27) Ida, T.; Okamoto, R.; Shimauchi, C.; Okubo, T.; Kuga, A.; Inouqe,
M. J. Clin. Microb. 2001, 39, 3115-3121.
4060 J. Org. Chem., Vol. 72, No. 11, 2007

RegioselectiVity for the Staudinger Reaction
SCHEME 8. Synthesis of Kanamycin Class Aminoglycosides with N-1 AHB
P. aeruginosa is an opportunistic human pathogen and the
leading cause of nosocomial infection. With the exception of
neokacin, which shows only modest antibacterial activity, the
attachment of N-1 AHB cannot revive the activity of neomycin
class aminoglycosides against P. aeruginosa (entry 9, Table 2).
A similar observation has been observed previously, and it has
been suggested that attaching an N-1 AHB group may deter
the acetylation at N-6′ (AAC(6′)) for kanamycin class ami-
noglycosides but have no effect against the action of AAC(6′)
for the neomycin class aminoglycosides.28 In contrast, the
kanamycin class aminoglycosides bearing an N-1 AHB group
while still having 3′ and 4′ hydroxyl groups appear to have
regained activity (entry 9, Table 3). However, gentimicin that
contain 3′,4′-deoxygenation but no N-1 AHB group also displays
prominent activity against P. aeruginosa. We have recently
synthesized a kanamycin (gentamicin) analogue, JL040,29 which
differs from JLN040 only at the absence of an N-1 AHB group
(Figure 3). JL040 is inactive against P. aeruginosa. The result
demonstrates the importance of having an AHB group at N-1
for the activity of the kanamycin class aminoglycoside against
P. aeruginosa.
(28) Wright, G. D.; Ladak, P. Antimicrob. Agents Chemother. 1997, 41,
956-960.
(29) Li, J.; Wang, J.; Chang, C.-W. T. Unpublished result.
J. Org. Chem, Vol. 72, No. 11, 2007 4061

Li et al.
TABLE 1. Chemical Shifts of Selected Aminoglycoside Derivatives
structure 16 17
H-1 4.1 3.5
H-2 (eq./ax.) 2.45/1.65 2.44/1.65 2.3/1.40 2.1/1.52 2.41/1.62 2.36/1.63 2.35/1.58 2.39/1.60 2.39/1.5
3
8
4
18
19
25
26
20
3.43
21
3.7
23
3.7
3.6
3.54
3.5
3.5
3.4
3.7
4.09
2.39/1.63 2.3/1.63 2.35/1.56
H-3
H-4
H-5
H-6
3.5
3.67
5.16
4.94
5.21
3.36
5.44
5.03
4.5
3.3/3.4
-
3.5
3.69
5.13
4.94
5.1
4.0
5.09
5.04
4.4
3.3-3.4
-
3.5
3.6
4.0
3.7
3.4
3.4
3.4
3.4
3.6
4.86
5.1
3.3
3.5
3.7
3.51
3.66
3.91
4.78
5.9
3.22
5.44
4.99
4.52
3.2-3.3
5.43
5.08
5.2
3.53
3.7
3.9
4.96
5.95
3.16
5.47
4.99
4.4
3.5
3.57
3.71
3.98
4.91
5.91
3.17
5.46
4.97
4.46
3.2-3.3
5.37
4.9
4.42
4.3
4.28
4.87
3.3
3.44
3.58
4.94
5.27
3.03
5.12
4.96
4.2
3.65
3.93
4.79
5.93
3.16
5.46
4.96
4.45
n.d.
5.37
4.88
n.d.
n.d.
4.1
5.21
4.81
5.18
3.3
5.43
5.02
4.4
5.11
4.95
4.82
3.40
5.24
4.99
4.0
3.65
4.93
5.35
3.68
5.49
5.05
4.34
3.87
4.97
5.9
3.25
5.45
5.01
4.4
H-1′
H-2′
H-3′
H-4′
H-5′
H-6′
H-1′′
H-2′′
H-3′′
H-4′′
H-5′′
H-1′′′
H-2′′′
H-3′′′
H-4′′′
H-5′′′
H-6′′′
3.9
5.17
5.02
4.3
3.1-3.4 3.4/3.11 3.43/3.32 3.2-3.3
3.38/3.28 3.3
3.3
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
5.42
5.1
5.19
4.26
5.35
4.89
4.4
-
-
-
-
-
-
4.2
4.32
-
-
4.44/4.16 4.38/4.15 4.26
-
-
-
-
-
-
-
-
-
-
-
-
-
-
4.87
3.3
5.04
4.7
4.09
3.59/3.28 n.d.
4.87
n.d.
5.02
4.7
-
-
-
-
5.02
4.69
4.1
-
-
-
-
n.d.
-
-
3.58/3.27
a n.d.: not distinguishable due to the overlap of multiple signals.
TABLE 2. Minimum Inhibitory Concentrations for Neomycin Class Aminoglycosidesa
entry
strains
E. colib
neomycin
ribostamycin
butirosin A
neokacin
butirosin B
JT00517
1
2
3
4
5
6
7
8
9
4-8
8
4
4
4-8
2
1
0.5-1
1
2
inactive
4
inactive
8-16
8
4
2
4
2
inactive
2-4
16
8
2
2
2
8
4
2-4
4
4
1-2
inactive
4
inactive
E. coli (TG1)c
E. coli (pSF815)d
E. coli (pTZ19U-3)e
K. pneumoniaeg
K. pneumoniaeh
S. aureusi
4-8
4
inactivef
inactive
4-8
inactive
inactive
2-4
inactive
8
2-4
2-4
inactive
4
inactive
1-2
inactive
S. aureusj
P. aeruginosak
inactive
inactive
a Unit: µg/mL. b Escherichia coli (ATCC 25922). c E. coli (TG1) (aminoglycoside susceptible strain). d E. coli (TG1) (pSF815 plasmid encoded for
(AAC(6′)/APH(2′′)). e E. coli (TG1) (pTZ19U-3 plasmid encoded for APH(3′)-I). f Inactive is defined as MIC g 32 µg/mL. g Klebsiella pneumoniae (ATCC
700603). h K. pneumoniae (ATCC 13883). i Staphylococcus aureus (ATCC 33591) (MRSA). j S. aureus (ATCC 25923). k Pseudomonas aeruginosa (ATCC
27853).
Conclusion
We have completed the investigation on the regioselectivity
of Staudinger reactions of perazidoaminoglycosides. Electron-
withdrawing effect is the dominant factor in controlling the
regioselectivity; however, steric factor can have significant
impact as well. It is the combination of these two factors that
a selective Staudinger reaction can occur preferably at the N-1
azido group of per-azido neamine and neomycin derivatives.
Chemical shift of ipso-proton can provide valuable prediction
regarding the preference of reduction site(s). Nonetheless, such
information cannot be interpreted as the absolute factor in
FIGURE 3. Structures of JL040 and gentamicin.
controlling the regioselectivity. Although in some cases, the
yield is moderate, and it can be attributed to the non-
regioselective reduction of azido group at other positions. In
general, we have developed a novel way of directly introducing
structural modifications at the N-1 of neomycin class aminogly-
cosides. In addition, this method can lead to the synthesis of
kanamycin class aminoglycosides with N-1 modifications.
Finally, the results from the antibacterial studies reveal drastic
difference in the activity between the neomycin and kanamycin
class aminoglycosides among various bacteria.
MRSA is another formidable drug-resistant pathogen, which
harbors multiple AMEs. Without surprise, no newly synthesized
aminoglycosides show activity against MRSA. Interestingly,
gentimicin is active against MRSA while structurally similar
JL040 and JLN040 are inactive. The results suggest that methyl
groups at C-6′, N-3′′, and C-4′′ on the carbohydrate scaffold
may play crucial role(s) in the activity of kanamycin class
aminoglycosides against MRSA.
4062 J. Org. Chem., Vol. 72, No. 11, 2007

RegioselectiVity for the Staudinger Reaction
TABLE 3. Minimum Inhibitory Concentrations for Kanamycin Class Aminoglycosidesa
entry
strains
E. colic
amikacin
kanamycin B
gentamicinb
JLN00517
JLN007
JLN02717
JLN033
JLN040
1
2
3
4
5
6
7
8
9
2-4
1
0.5
0.25
1
1-2
16
2
4-8
4
2
2
8-16
4
2
4-8
4
2
2-4
1
1
0.25-0.5
1-2
1
inactive
1-2
2-4
8-16
4
2
0.5
4
4
inactive
2
4-8
4
4
1
0.5
E. coli (TG1)d
E. coli (pSF815)e
E. coli (pTZ19U-3)g
K. pneumoniaeh
K. pneumoniaei
S. aureusj
inactivef
inactive
inactive
1-2
inactive
1
8-16
1
4
0.5
0.5-1
2
0.5-1
ND
ND
ND
2-4
4-8
2-4
2
2
1-2
inactive
1-2
1-2
inactive
1-2
inactive
inactive
2
4
S. aureusk
P. aeruginosal
0.5-1
a Unit: µg/mL. ND: not determined. b Gentamicin consists of 33% gentamicin C1, 19% gentamicin C1a, and 48% gentamicin C2a and C2. All these
compounds have 3′,4′-dideoxygenation. c Escherichia coli (ATCC 25922). d E. coli (TG1) (aminoglycoside susceptible strain). e E. coli (TG1) (pSF815
plasmid encoded for (AAC(6′)/APH(2′’)). f Inactive is defined as MIC g 32 µg/mL. g E. coli (TG1) (pTZ19U-3 plasmid encoded for APH(3′)-I). h Klebsiella
j
pneumoniae (ATCC 700603). i K. pneumoniae (ATCC 13883). Staphylococcus aureus (ATCC 33591) (MRSA). k S. aureus (ATCC 25923). l Pseudomonas
aeruginosa (ATCC 27853).
2.00 (s, 3H), 1.65 (ddd, J ) 13.1, 12.7, 12.6 Hz, 1H, H-2ax), 1.43
(s, 9H, C(O)(CH3)3); 13C NMR (100 MHz, CDCl3) δ 170.9, 169.8
(2C), 169.6, 155.3, 98.9, 80.3, 77.7, 74.2, 73.7, 71.3, 70.2, 69.4,
58.8, 57.9, 53.0, 51.2, 32.0, 28.3, 20.8 (2C), 20.7 (2C); HRESI
calcd for C25H36N10O12Na [M + Na]+ m/z 691.2412, measure m/z
691.2390.
1-N-tert-Butoxycarbonyl-3,2′,6′-triazido-5,6-O-cyclohexylidene-
neamine (9) and 3-N-tert-Butoxycarbonyl-1,2′,6′-triazido-5,6-O-
cyclohexylideneneamine (10). See ref 17.
Experimental Section
Proton magnetic resonance spectra were recorded using 400 MHz
spectrometers. Chemical shifts were reported as parts per million
(ppm) downfield from tetramethylsilane in δ unit, and coupling
constants were given in cycles per second (Hz). Splitting patterns
were designed as s, singlet; d, doublet; t, triplet; q, quartet; m,
multiplet. 13C NMR spectra were obtained by broad-band WALTZ
decoupling using a 100 MHz spectrometer. All NMR spectra were
recorded at ambient temperature unless otherwise noted. Chemical
reagents and starting materials were purchased and used without
purification unless otherwise noted. Dichloromethane was distilled
over CaH2. Other solvents were used without purification.
3′,4′-Di-O-acetyl-1-N-tert-butoxycarbonyl-3,2′,6′-triazido-5,6-
O-cyclohexylideneneamine (11). Please refer to the procedure for
the preparation of compound 3. From a mixture of 9 and 10 (1.97
g, 3.40 mmol), two components were isolated with gradient column
chromatography (hexane:EtOAc ) 60:40 to 0:100) and character-
ized individually. The major product was compound 11 (1.36 g,
2.05 mmol, 60%). 1H NMR (400 MHz, CDCl3) δ 5.54 (d, J ) 3.5
Hz, 1H, H-1′), 5.45 (dd, J ) 10.6, 9.4 Hz, 1H, H-3′), 4.99 (dd, J
) 9.8, 9.6 Hz, 1H, H-4′), 4.8 (m, 1H, NH), 4.3 (m, 1H, H-5′), 3.74
(dd, J ) 9.4, 9.4 Hz, 1H, H-4), 3.7 (m, 1H, H-1), 3.56 (dd, J )
9.5, 9.3 Hz, 1H, H-5), 3.5 (m, 1H, H-3), 3.2-3.3 (m, 3H), 3.24
(dd, J ) 10.1, 3.5 Hz, 1H, H-2′), 2.6 (m, 1H, H-2eq), 2.05 (s, 3H),
2.03 (s, 3H), 1.5-1.6 (m, 10H), 1.42 (s, 9H, C(O)(CH3)3), 1.36
(ddd, J ) 13.9, 12.8, 12.1 Hz, 1H, H-2ax); 13C NMR (100 MHz,
CDCl3) δ 170.0, 169.9, 155.4, 113.9, 96.7, 80.2, 80.0, 78.2 (2C),
69.9, 69.6, 69.5, 60.8, 60.6, 51.1, 48.9, 36.6, 36.1, 35.3, 28.5, 25.0,
23.8, 20.8; HRESI calcd for C27H40N10O10Na [M + Na]+ m/z
687.2827, measure m/z 687.2806.
3′,4′-Di-O-acetyl-3-N-tert-butoxycarbonyl-1,2′,6′-triazido-5,6-
O-cyclohexylideneneamine (12). Please refer to the procedure for
the preparation of compound 3. From a mixture of 9 and 10 (1.97
g, 3.40 mmol), two components were isolated with gradient column
chromatography (hexane:EtOAc ) 60:40 to 0:100) and character-
ized individually. The minor product was compound 12 (0.47 g,
0.71 mmol, 21%). 1H NMR (400 MHz, CDCl3) δ 5.47 (d, J ) 3.5
Hz, 1H, H-1′), 5.39 (dd, J ) 9.9, 9.9 Hz, 1H), 4.97 (d, J ) 9.8, 9.6
Hz, 1H), 4.7 (m, 1H), 4.2 (m, 1H, H-5′), 3.7-3.8 (m, 2H), 3.7 (m,
1H, H-1), 3.60 (dd, J ) 9.4, 9.3 Hz, 1H), 3.41 (dd, J ) 9.7, 9.7
Hz, 1H), 3.4 (m, 2H), 3.25 (dd, J ) 10.7, 3.6 Hz, 1H, H-2′), 2.4
(m, 1H, H-2eq), 2.08 (s, 3H), 2.03 (s, 3H), 1.6-1.7 (m, 11H), 1.48
(s, 9H, C(O)(CH3)3); 13C NMR (100 MHz, CDCl3) δ 169.9, 169.7,
155.1, 113.7, 97.2, 79.9, 79.4, 78.7, 69.9, 69.8, 69.0, 60.8, 57.7,
51.3, 50.8, 36.6, 36.1, 35.6, 28.5, 25.1, 23.9, 20.8, 20.7; HRESI
calcd for C27H40N10O10Na [M + Na]+ m/z 687.2827, measure m/z
687.2800.
5,6,3′,4′-Tetra-O-acetyl-1,3,2′,6′-tetraazidoneamine (3). To a
solution of 1,3,2′,6′-tetraazidoneamine (7.6 g, 17.83 mmol),17 Et3N
(26.3 mL, 0.19 mol), and DMAP (0.50 g, catalyst) in anhydrous
CH2Cl2 (200 mL) was slowly added Ac2O (17.7 mL, 0.19 mol).
The reaction mixture was stirred at room temperature until the
completion of the reaction (monitored by TLC, Rf ) 0.46, EtOAc:
hexane ) 35: 65). The reaction was then quenched by addition of
saturated NaHCO3 and extracted with EtOAc. The organic solution
was washed with 1 N HCl(aq), water, saturated NaHCO3(aq), brine,
and dried over anhydrous Na2SO4. After removal of the solvent
followed by purification with a gradient column chromatography
(hexane:EtOAc ) 80:20 to 40:60), the product was obtained as
1
white crystals (4.88 g, 8.21 mmol, 46%). H NMR (400 MHz,
CDCl3) δ 5.44 (dd, J ) 10.8, 9.2 Hz, 1H, H-3′), 5.21 (d, J ) 3.8
Hz, 1H, H-1′), 5.16 (dd, J ) 9.7, 9.7 Hz, 1H, H-5), 5.03 (dd, J )
10.2, 9.3 Hz, 1H, H-4′), 4.94 (dd, J ) 10.0, 9.9 Hz, 1H, H-6), 4.5
(m, 1H, H-5′), 3.67 (dd, J ) 9.8, 9.7 Hz, 1H, H-4), 3.7 (m, 1H,
H-1), 3.5 (m, 1H, H-3), 3.3-3.4 (m, 3H), 2.45 (ddd, J ) 13.4, 4.6,
4.5 Hz, 1H, H-2eq), 2.09 (s, 6H), 2.08 (s, 3H), 2.06 (s, 3H), 1.65
(ddd, J ) 13.3, 12.5, 12.5 Hz, 1H, H-2ax); 13C NMR (100 MHz,
CDCl3) δ 170.1, 170.0, 169.9, 169.5, 99.0, 78.9, 74.3, 73.8, 70.0,
69.8, 69.6, 60.8, 58.6, 57.8, 50.9, 31.9, 20.8 (2C), 20.7 (2C); HRESI
calcd for C20H30N13O10 [M + NH4]+ m/z 612.2238, measure m/z
612.2213.
5,6,3′,4′-Tetra-O-acetyl-1-N-tert-butoxycarbonyl-3,2′,6′-triazi-
doneamine (6) and 5,6,3′,4′-Tetra-O-acetyl-3-N-tert-butoxycar-
bonyl-1,2′,6′-triazidoneamine (7). Please refer to the general
procedure for selective Staudinger reaction. Compounds 6 and 7
were inseparable and were obtained with a combined yield of 49%.
5,6,3′,4′-Tetra-O-acetyl-2′-N-tert-butoxycarbonyl-1,3,6′-triazi-
doneamine (8). Please refer to the general procedure for selective
1
Staudinger reaction. Compound 8 was obtained in 8% yield. H
NMR (400 MHz, CDCl3) δ 5.13 (dd, J ) 9.6, 9.1 Hz, 1H, H-5),
5.0-5.1 (m, 2H), 5.04 (dd, J ) 9.8, 9.6 Hz, 1H, H-4′), 4.94 (dd,
J ) 9.9, 9.9 Hz, 1H, H-6), 4.55 (d, J ) 10.3 Hz, 1H, NH), 4.40
(ddd, J ) 9.9, 4.3, 4.1 Hz, 1H, H-5′), 4.00 (ddd, J ) 10.5, 10.4,
3.7 Hz, 1H, H-2′), 3.69 (dd, J ) 9.7, 9.7 Hz, 1H, H-4), 3.6 (m,
1H, H-1), 3.5 (m, 1H, H-3), 3.3-3.4 (m, 2H), 2.44 (ddd, J ) 13.4,
4.6, 4.5 Hz, 1H, H-2eq), 2.07 (s, 3H), 2.04 (s, 3H), 2.01 (s, 3H),
3′,4′-Di-O-acetyl-1-N-[(S)-2-O-benzyl-4-(benzyloxycarbonyl-
amino)butanoyl ]-3,2′,6′-triazidoneamine (14). To a solution of
compound 11 (0.60 g, 0.90 mmol) in dichloromethane were added
99% trifluoroacetic acid (3 mL) and water (0.02 mL), and the
reaction mixture was stirred for 7 h at room temperature. The
reaction mixture was quenched with triethylamine, concentrated
to dryness, and used without purification. A mixture of (S)-2-O-
J. Org. Chem, Vol. 72, No. 11, 2007 4063

Li et al.
benzyl-4-(benzyloxycarbonylamino)butyric acid (0.34 g, 0.99 mmol),
EDC (0.26 g, 1.35 mmol), HOBt (0.18 g, 1.35 mmol), Et3N (1
mL), and the above crude amine in anhydrous DMF (5 mL) was
stirred overnight under nitrogen at room temperature. After comple-
tion of the reaction (monitored by TLC, hexane:EtOAc ) 35:65),
the reaction mixture was concentrated and redissolved in EtOAc.
The organic solution was washed with water and brine and dried
over anhydrous Na2SO4. Removal of the solvent followed by
purification with gradient column chromatography (hexane:EtOAc
) 60:40 to 0:100) afforded the product as white crystals (0.37 g,
0.45 mmol, 51%). 1H NMR (400 MHz, CDCl3) δ 7.3-7.4 (m, 10H),
6.7 (m, 1H), 5.5 (m, 2H), 5.0 (m, 4H), 4.60 (d, J ) 11.5 Hz, 1H),
4.49 (d, J ) 11.9 Hz, 1H), 4.4 (m, 1H, H-5′), 4.03 (dd, J ) 4.6,
4.6 Hz, 1H), 3.8 (m, 1H, H-1), 3.2-3.5 (m, 9H), 2.3 (m, 1H, H-2eq),
2.10 (s, 3H, CH3), 2.07 (s, 3H, CH3), 2.0 (m, 2H), 1.3 (m, 1H,
H-2ax); 13C NMR (100 MHz, CDCl3) δ 172.8, 172.2, 169.0, 156.8,
136.9, 136.8, 129.0, 128.7, 128.3, 128.2, 127.9, 98.5, 81.8, 78.7,
75.0, 73.2, 70.7, 69.7, 69.4, 66.9, 61.4, 59.1, 51.1, 48.7, 37.1, 32.7,
32.2, 29.8, 20.8, 20.8; HRESI calcd for C35H44N11O12 [M + H]+
m/z 810.3170, measure m/z 810.3159.
3′,4′-Di-O-acetyl-3-N-[(S)-2-O-benzyl-4-(benzyloxycarbonyl-
amino)butanoyl ]-1,2′,6′-triazidoneamine (15). Please refer to the
procedure for the preparation of compound 14. 1H NMR (400 MHz,
CDCl3) δ 7.3-7.4 (m, 10H), 6.7 (d, J ) 8.3 Hz, 1H), 5.39 (dd, J
) 10.2, 9.5 Hz, 1H, H-3′), 5.1 (m, 2H), 5.02 (dd, J ) 9.9, 9.7 Hz,
1H, H-4′), 5.0 (d, J ) 2.9 Hz, 1H, H-1′), 4.64 (d, J ) 11.1 Hz,
1H), 4.55 (d, J ) 11.2 Hz, 1H), 4.1 (m, 1H, H-5′), 3.98 (dd, J )
5.5, 5.4 Hz, 1H), 3.9 (m, 1H, H-3), 3.66 (dd, J ) 10.4, 3.5 Hz, 1H,
H-2′), 3.56 (dd, J ) 8.8, 8.8 Hz, 1H), 3.2-3.5 (m, 6H), 3.16 (dd,
J ) 13.5, 4.4 Hz, 1H, H-6′), 2.08 (s, 3H), 2.0-2.1 (m, 3H), 2.01
(s, 3H), 1.45 (ddd, J ) 12.5, 12.4, 12.3 Hz, 1H, H-2ax); 13C NMR
(100 MHz, CDCl3) δ 172.2, 170.1, 169.9, 156.5, 137.0, 136.7,
129.0, 128.7, 128.3, 128.2, 98.3, 85.4, 78.0, 75.9, 75.8, 72.7, 71.7,
69.6, 69.1, 66.9, 62.7, 59.9, 50.7, 48.1, 37.4, 33.2, 31.7, 20.8, 20.8;
HRESI calcd for C35H44N11O12 [M + H]+ m/z 810.3170, measure
m/z 810.3145.
3′,4′,5,6-Tetra-O-acetyl-1-N-[(S)-2-O-benzyl-4-(benzyloxycar-
bonylamino)butanoyl]-3,2′,6′-triazidoneamine (16). Please refer
to the procedure for the preparation of compound 3. 1H NMR (400
MHz, CDCl3) δ 7.3-7.4 (m, 10H), 6.78 (d, J ) 7.9 Hz, 1H), 5.43
(dd, J ) 10.4, 9.5 Hz, 1H, H-3′), 5.21 (dd, J ) 10.5, 9.5 Hz, 1H,
H-5), 5.18 (d, J ) 3.9 Hz, 1H, H-1′), 5.1 (m, 2H), 5.02 (dd, J )
9.8, 9.6 Hz, 1H, H-4′), 4.97 (dd, J ) 5.5, 5.5 Hz, 1H, NH), 4.81
(dd, J ) 10.2, 10.2 Hz, 1H, H-6), 4.59 (d, J ) 11.6 Hz, 1H), 4.47
(d, J ) 12.0 Hz, 1H), 4.4 (m, 1H, H-5′), 4.1 (m, 1H, H-1), 3.9 (m,
1H, H-5′), 3.6 (dd, J ) 9.8, 9.3 Hz, 1H, H-4), 3.5 (m, 1H, H-3),
3.2-3.4 (m, 5H), 2.3 (m, 1H, H-2eq), 2.07 (s, 3H), 2.06 (s, 3H),
2.05 (s, 3H), 1.97 (s, 3H), 1.9 (m, 1H), 1.8 (m, 1H), 1.4 (m, 1H,
H-2ax); 13C NMR (100 MHz, CDCl3) δ 172.4, 171.3, 169.9 (2C),
169.0, 156.5, 136.8, 136.7, 129.0, 128.7, 128.5, 128.2, 98.9, 79.2,
78.3, 73.5, 73.1, 70.0, 69.7, 69.6, 66.8, 60.8, 58.6, 50.9, 47.3, 37.5,
32.8, 32.5, 29.8, 20.8, 20.6; HRESI calcd for C39H48N11O14 [M +
H]+ m/z 894.3382, measure m/z 894.3366.
3′,4′,5,6-Tetra-O-acetyl-3-N-[(S)-2-O-benzyl-4-(benzyloxycar-
bonylamino)butanoyl]-1,2′,6′-triazidoneamine (17). Please refer
to the procedure for the preparation of compound 3. 1H NMR (400
MHz, CDCl3) δ 7.3-7.4 (m, 10H), 6.85 (d, J ) 8.8 Hz, 1H), 5.24
(dd, J ) 10.1, 9.6 Hz, 1H, H-3′), 5.11 (dd, J ) 9.5, 9.1 Hz, 1H,
H-5), 5.0 (m, 3H), 4.99 (dd, J ) 9.8, 9.6 Hz, 1H, H-4′), 4.95 (dd,
J ) 9.5, 9.4 Hz, 1H, H-6), 4.82 (d, J ) 2.5 Hz, 1H, H-1′), 4.64 (d,
J ) 11.6 Hz, 1H), 4.51 (d, J ) 12.1 Hz, 1H), 4.0 (m, 1H, H-3),
3.9-4.0 (m, 2H), 3.7 (dd, J ) 9.0, 8.9 Hz, 1H, H-4), 3.5 (m, 1H,
H-1), 3.3-3.4 (m, 2H), 3.2-3.3 (m, 2H), 3.11 (dd, J ) 13.5, 4.2
Hz, 1H, H-6′), 2.1 (m, 1H, H-2eq), 2.08 (s, 3H), 2.04 (s, 3H), 2.02
(s, 3H), 1.99 (s, 3H), 1.9-2.0 (m, 2H), 1.52 (ddd, J ) 12.4, 12.3,
12.2 Hz, 1H, H-2ax); 13C NMR (100 MHz, CDCl3) δ 172.3, 170.1,
169.8 (2C), 169.7 (2C), 156.5, 137.0, 136.7, 129.0, 128.7, 128.3
(2C), 128.2, 98.1, 80.3, 77.8, 73.6, 72.7, 72.5, 70.6, 69.5, 69.0,
66.8, 61.9, 58.0, 50.6, 48.4, 37.2, 32.5, 31.4, 20.7 (4C); HRESI
calcd for C39H48N11O14 [M + H]+ m/z 894.3382, measure m/z
894.3377.
6,3′,4′-Tri-O-acetyl-2′-N-tert-butoxycarbonyl-1,3,6′-triazido-
neamine (18). Please refer to the general procedure for selective
Staudinger reaction. Compound 18 was obtained with a 21% yield.
1H NMR (400 MHz, CDCl3) δ 5.17 (dd, J ) 10.1, 10.0 Hz, 1H,
H-3′), 5.1 (m, 1H, H-1′), 5.02 (dd, J ) 9.9, 9.6 Hz, 1H, H-4′),
4.86 (dd, J ) 9.8, 9.8 Hz, 1H, H-6), 4.3 (m, 1H, H-5′), 3.8 (m, 1H,
H-2′), 3.4-3.6 (m, 2H), 3.2-3.4 (m, 4H), 2.36 (ddd, J ) 13.2,
4.5, 4.4 Hz, 1H, H-2eq), 2.14 (s, 3H), 2.01 (s, 3H), 2.00 (s, 3H),
1.63 (ddd, J ) 12.9, 12.6, 12.6 Hz, 1H, H-2ax), 1.39 (s, 9H); 13C
NMR (100 MHz, CDCl3) δ 171.3, 169.8, 163.1, 155.6, 98.4, 80.8,
80.2, 76.3, 74.7, 71.4, 69.7, 69.5 (2C), 58.9, 58.7, 51.2, 31.9, 28.4
(3C), 21.0, 20.9, 20.8; HRESI calcd for C23H34N10O11Na [M +
Na]+ m/z 649.2306, measure m/z 649.2285.
6,3′,4′-Tri-O-acetyl-2′-amino-1,3,6′-triazidoneamine (19). Please
refer to the general procedure for selective Staudinger reaction.
Compound 19 was obtained with a 27% yield. 1H NMR (400 MHz,
CDCl3) δ 5.27 (d, J ) 4.1 Hz, 1H, H-1′), 5.12 (dd, J ) 10.2, 9.7
Hz, 1H, H-3′), 4.96 (dd, J ) 9.8, 9.6 Hz, 1H, H-4′), 4.94 (dd, J )
9.9, 9.9 Hz, 1H, H-6), 4.2 (m, 1H, H-5′), 3.58 (dd, J ) 9.6, 9.1
Hz, 1H, H-5), 3.5 (m, 1H, H-1), 3.44 (dd, J ) 9.9, 9.0 Hz, 1H,
H-4), 3.38 (dd, J ) 13.3, 2.7 Hz, 1H, H-6′), 3.3 (m, 1H, H-3),
3.28 (dd, J ) 13.5, 5.3 Hz, 1H, H-6′), 3.03 (dd, J ) 10.5, 3.9 Hz,
1H, H-2′), 2.35 (ddd, J ) 13.2, 4.5, 4.2 Hz, 1H, H-2eq), 2.12 (s,
3H), 2.06 (s, 3H), 2.01 (s, 3H), 1.58 (ddd, J ) 12.8, 12.6, 12.5 Hz,
1H, H-2ax); 13C NMR (100 MHz, CDCl3) δ 171.6, 170.5, 169.8,
101.6, 85.0, 74.7, 74.4, 74.0, 69.6, 69.4, 58.7, 58.3, 54.1, 51.1, 32.1,
21.1, 21.0, 20.8; HRESI calcd for C18H27N10O9 [M + H]+ m/ z
527.1962, measure m/z 527.1972.
1-N-tert-Butoxycarbonyl-hepta-O-acetyl-pentaazidoneomy-
cin (21). Please refer to the general procedure for selective
Staudinger reaction. Compound 21 was obtained with a 45% yield.
1H NMR (400 MHz, CDCl3) δ 5.93 (d, J ) 3.9 Hz, 1H), 5.46 (dd,
J ) 10.6, 9.3 Hz, 1H), 5.37 (d, J ) 2.4 Hz, 1H), 5.02 (dd, J ) 2.9,
2.8 Hz, 1H), 4.96 (dd, J ) 10.2, 9.4 Hz, 1H), 4.88 (dd, J ) 2.8,
2.4 Hz, 1H), 4.87 (d, J ) 1.8 Hz, 1H), 4.79 (dd, J ) 9.5, 9.5 Hz,
1H), 4.70 (dd, J ) 1.8, 1.8 Hz, 1H), 4.5 (m, 1H), 4.45 (dd, J )
11.8, 1.8 Hz, 1H), 4.41 (dd, J ) 6.5, 5.1 Hz, 1H), 4.3 (m, 1H),
4.25 (dd, J ) 11.4, 6.4 Hz, 1H), 4.1 (m, 1H), 3.93 (dd, J ) 9.0,
8.8 Hz, 1H), 3.7 (m, 1H, H-1), 3.65 (dd, J ) 9.6, 8.7 Hz, 1H),
3.59 (dd, J ) 12.9, 9.6, 8.1 Hz, 1H), 3.5 (m, 1H), 3.2-3.3 (m,
4H), 3.16 (dd, J ) 10.7, 3.8 Hz, 1H, H-2′), 2.3 (m, 1H, H-2eq),
2.18 (s, 3H), 2.14 (s, 3H), 2.10 (s, 6H), 2.08 (s, 3H), 2.05 (s, 3H),
2.04 (s, 3H), 1.63 (ddd, J ) 12.9, 12.8, 12.4 Hz, 1H, H-2ax), 1.42
(s, 9H); 13C NMR (100 MHz, CDCl3) δ 171.1, 170.9, 170.3, 170.2,
169.9 (2C), 168.7, 155.2, 107.4, 99.3, 96.6, 82.8, 80.5, 79.5, 77.4,
76.8, 75.9, 75.1, 73.7, 70.2, 69.7, 69.5, 68.9, 65.9, 63.5, 61.0, 59.5,
56.7, 51.3, 50.8, 49.1, 33.2, 28.4, 21.0, 20.9 (2C), 20.8 (3C), 20.7;
MALDI calcd for C42H58Cl2N16O22Na ([M + Na]+) m/z 1161.3809,
measure m/z 1161.3801.
1-N-[(S)-4-(Benzyloxycarbonylamino)-2-hydroxybutanoyl]-
hepta-O-acetyl-pentaazidoneomycin (23). Please refer to the
procedure for the preparation of compound 14. Compound 23 was
obtained with a 28% yield. 1H NMR (400 MHz, CDCl3) δ 7.3 (m,
5H), 7.10 (d, J ) 9.1 Hz, 1H, NH), 5.91 (d, J ) 3.7 Hz, 1H, H-1′),
5.46 (dd, J ) 10.6, 9.3 Hz, 1H, H-3′), 5.37 (d, J ) 2.5 Hz, 1H,
H-1′’), 5.2 (m, 1H, NH), 5.11 (d, J ) 13.0 Hz, 1H), 5.08 (d, J )
12.8 Hz, 1H), 5.02 (dd, J ) 2.7, 2.7 Hz, 1H, H-3′′′), 4.97 (dd, J )
9.8, 9.7 Hz, 1H, H-4′), 4.91 (dd, J ) 5.0, 3.7 Hz, 1H, H-6), 4.90
(dd, J ) 2.5, 2.5 Hz, 1H, H-2′′), 4.87 (d, J ) 1.8 Hz, 1H, H-1′′′),
4.69 (dd, J ) 1.6, 1.6 Hz, 1H, H-4′′′), 4.4-4.5 (m, 3H), 4.2-4.3
(m, 2H), 4.0-4.1 (m, 3H), 3.98 (dd, J ) 8.6, 8.4 Hz, 1H, H-5),
3.71 (dd, J ) 9.4, 8.2 Hz, 1H, H-4), 3.5-3.6 (m, 3H), 3.2-3.3 (m,
3H), 3.27 (dd, J ) 12.9, 4.3 Hz, 1H, H-6′′′), 3.17 (dd, J ) 10.8,
3.8 Hz, 1H, H-2′), 3.1 (m, 1H), 2.35 (ddd, J ) 13.5, 4.4, 4.4 Hz,
1H, H-2eq), 2.16 (s, 3H), 2.13 (s, 3H), 2.10 (s, 3H), 2.08 (s, 6H),
2.06 (s, 3H), 2.05 (s, 3H), 1.5-1.6 (m, 3H); 13C NMR (100 MHz,
4064 J. Org. Chem., Vol. 72, No. 11, 2007

RegioselectiVity for the Staudinger Reaction
CDCl3) δ 173.5, 170.8, 170.3, 170.2, 169.9 (2C), 168.6, 158.4,
136.2, 128.8, 128.6, 128.5, 128.4, 128.2, 107.2, 99.3, 96.6, 82.3,
79.6, 76.7, 75.9, 75.2, 75.1, 73.6, 70.1, 69.6, 69.4, 68.9, 67.4, 65.8,
63.6, 61.0, 59.3, 56.6, 51.3, 50.8, 47.2, 37.1, 35.4, 32.1, 29.8, 21.0,
20.9 (2C), 20.8 (3C), 20.6; HRESI calcd for C49H64N17O24 ([M +
H]+) m/z 1274.4310, measure m/z 1274.4295.
(dd, J ) 12.1, 3.0 Hz, 1H, H-5′′), 4.26 (ddd, J ) 4.3, 3.8, 3.3 Hz,
1H, H-4′′), 4.16 (dd, J ) 12.2, 3.9 Hz, 1H, H-5′′), 3.87 (dd, J )
9.2, 9.1 Hz, 1H, H-5), 3.70 (dd, J ) 9.7, 8.9 Hz, 1H, H-4), 3.5 (m,
1H, H-3), 3.4 (m, 1H, H-1), 3.3 (m, 1H, H-6′), 3.25 (dd, J ) 10.7,
3.8 Hz, 1H, H 2′), 2.39 (ddd, J ) 13.3, 4.6, 4.5 Hz, 1H, H-2eq),
2.23 (s, 3H), 2.13 (s, 3H), 2.12 (s, 3H), 2.10 (s, 3H), 2.06 (s, 6H),
1.60 (ddd, J ) 12.9, 12.8, 12.7 Hz, 1H, H-2ax); 13C NMR (100
MHz, CDCl3) δ 170.8, 170.2, 169.8, 169.7, 169.6, 169.5, 106.5,
96.5, 81.0, 79.8, 76.6, 74.8, 74.3, 71.1, 70.2, 69.4, 69.4, 63.3, 61.1,
59.2, 58.3, 51.1, 31.7, 21.0, 20.9 (2C), 20.6 (2C), 20.6; HRESI
calcd for C29H38N12O16Na ([M + Na]+) m/z 833.2426, measure
m/z 833.2434.
5-O-(2,3,5-Tri-O-acetyl-â-D-ribofuranosyl)-1-N-tert-butoxy-
carbonyl-6,3′,4′-tri-O-acetyl-3,2′,6′-triazidoneamine (26). Please
refer to the general procedure for selective Staudinger reac-
tion. Compound 26 was obtained with a 31% yield. 1H NMR
(400 MHz, CDCl3) δ 5.90 (d, J ) 3.7 Hz, 1H, H-1′), 5.44 (dd, J
) 10.7, 9.3 Hz, 1H, H-3′), 5.43 (d, J ) 3.7 Hz, 1H, H-1′′), 5.20
(dd, J ) 4.9, 4.9 Hz, 1H, H-3′′), 5.08 (dd, J ) 4.6, 3.9 Hz, 1H,
H-2′′), 4.99 (dd, J ) 10.0, 9.5 Hz, 1H, H-4′), 4.78 (dd, J ) 10.1,
10.0 Hz, 1H, H-6), 4.64 (d, J ) 9.1 Hz, 1H), 4.52 (ddd, J ) 10.1,
5.1, 2.9 Hz, 1H, H-5′), 4.38 (dd, J ) 11.4, 3.0 Hz, 1H, H-5′′), 4.2
(m, 1H, H-4′′), 4.15 (dd, J ) 12.1, 3.8 Hz, 1H, H-5′′), 3.91 (dd, J
) 9.1, 9.0 Hz, 1H, H-5), 3.7 (m, 1H, H-1), 3.66 (dd, J ) 9.8, 8.9
Hz, 1H, H-4), 3.51 (ddd, J ) 12.1, 9.8, 4.4 Hz, 1H, H-3), 3.2-3.3
(m, 2H, H-6′), 3.22 (dd, J ) 10.8, 3.8 Hz, 1H, H-2′), 2.39 (ddd,
J ) 13.1, 4.4, 4.3 Hz, 1H, H-2eq), 2.14 (s, 3H), 2.10 (s, 3H),
2.08 (s, 3H), 2.08 (s, 3H), 2.05 (s, 3H), 2.04 (s, 3H),1.5 (m,
1H , H-2ax), 1.42 (s, 9H); 13C NMR (100 MHz, CDCl3) δ 171.1,
170.7, 170.1, 169.9, 169.6 (2C), 155.3, 106.6, 96.5, 81.7, 80.3, 79.4,
75.3, 74.5, 71.0, 70.2, 69.5, 69.3, 63.3, 61.2, 59.4, 51.2, 49.2, 33.2,
29.8, 28.4 (3C), 21.0, 20.8 (3C), 20.6, 20.5; HRESI calcd
for C34H48N10O18Na ([M + Na]+) m/z 907.3045, measure m/z
907.3056.
1-N-[(S)-4-Amino-2-hydroxybutanoyl]neomycin (Neokacin).
Compound 23 (0.20 g, 0.16 mmol) was dissolved in tetrahydrofuran
(3 mL) and water (1 mL), and lithium hydroxide monohydrate (0.06
g, 1.54 mmol) was added. The reaction mixture was stirred at room
temperature till the completion of the reaction (ca. 24 h, monitored
by TLC, Rf ) 0, EtOAc:hexane ) 65:35). The reaction mixture
was filtered through Celite, the solvent was removed, and the crude
product was directly used for next step. To a starting material/
THF solution in a reaction vial equipped with a reflux condenser
were added 0.1 M NaOH(aq) (0.5 mL) and PMe3 (1 M in THF,
5-7 equiv). The reaction mixture was stirred at 50 °C for 2 h. The
product has an Rf of 0 when eluted with EtOAc/MeOH (9/1)
solution, and an Rf of 0.6 when eluted with iPrOH/1 M NH4OAc
(2/1) solution. After completion of the reaction, the solvents were
removed, and the crude benzylated aminoglycoside was added with
catalytic amount of Pd(OH)2/C (20% Degussa type) and 5 mL of
degassed HOAc/H2O (1/3). After being further degassed, the
reaction mixture was stirred at room temperature under atmospheric
H2 pressure. After being stirred for 1 day, the reaction mixture was
filtered through Celite. The residue was washed with water, and
the combined solutions were concentrated. The crude product was
purified with Amberlite CG50(NH4+) eluted with a gradient of
NH4OH solution (0-20%). The final product with Cl- salt can be
prepared with an ion-exchange column packed with Dowex 1X8-
200 (Cl- form) and eluting with water. After collection of the
desired fractions and removal of solvent, the final products (35.2
mg, 0.04 mmol, 26%) are subjected to bioassay directly. The
1
reported final products are characterized by H and 13C NMR at
1
this stage. H NMR (400 MHz, D2O) δ 5.94 (d, J ) 3.8 Hz, 1H),
5-O-(2,3,5-Tri-O-acetyl-â-D-ribofuranosyl)-1-N-[(S)-4-(ben-
zyloxycarbonylamino)-2-hydroxybutanoyl]-6,3′,4′-tri-O-acetyl-
3,2′,6′-triazidoneamine (27). Please refer to the procedure for the
preparation of compound 14. 1H NMR (400 MHz, CDCl3) δ 7.2-
7.3 (m, 5H), 7.10 (d, J ) 8.7 Hz, 1H), 5.90 (d, J ) 3.7 Hz, 1H,
H-1′), 5.45 (dd, J ) 10.6, 9.3 Hz, 1H, H-3′), 5.44 (d, J ) 5.2 Hz,
1H, H-1′′), 5.2 (m, 1H), 5.21 (dd, J ) 4.8, 4.8 Hz, 1H, H-3′′),
5.0-5.1 (m, 3H), 5.0 (dd, J ) 9.8, 9.6 Hz, 1H, H-4′), 4.89 (dd, J
) 9.9, 9.8 Hz, 1H, H-6), 4.51 (ddd, J ) 10.2, 5.2, 2.8 Hz, 1H,
H-5′), 4.42 (dd, J ) 12.1, 3.0 Hz, 1H, H-5′′), 4.2 (m, 1H, H-4′′),
4.15 (dd, J ) 12.1, 3.8 Hz, 1H, H-5′′), 4.0-4.1 (m, 2H), 3.95 (dd,
J ) 8.9, 8.8 Hz, 1H, H-5), 3.70 (dd, J ) 9.7, 8.7 Hz, 1H, H-4),
3.58 (ddd, J ) 12.0, 7.6, 4.4 Hz, 1H, H-3), 3.5 (m, 1H), 3.36 (dd,
J ) 13.5, 2.9 Hz, 1H, H-6′), 3.30 (dd, J ) 13.5, 5.5 Hz, 1H, H-6′),
3.22 (dd, J ) 10.7, 3.7 Hz, 1H, H-2′), 3.1 (m, 1H), 2.35 (ddd, J )
13.5, 4.5, 4.3 Hz, 1H, H-2eq), 2.1 (s, 6H), 2.08 (s, 3H), 2.07 (s,
3H), 2.05 (s, 3H), 2.04 (s, 3H), 2.0 (m, 2H), 1.56 (ddd, J ) 12.7,
12.5, 12.3 Hz, 1H, H-2ax); 13C NMR (100 MHz, CDCl3) δ 173.7,
170.9, 170.7, 170.2, 169.9, 169.6 (2C), 158.3, 136.3, 128.7, 128.5,
128.2, 106.4, 96.6, 81.5, 79.5, 75.0, 74.5, 71.0, 70.2, 69.6, 69.3,
69.0, 67.4, 63.4, 61.1, 59.4, 51.2, 47.4, 37.2, 35.3, 32.5, 29.8, 21.0,
20.8 (2C), 20.7, 20.6, 20.5; HRESI calcd for C41H54N11O20 ([M +
H]+) m/z 1020.3546, measure m/z 1020.3545.
5-O-(â-D-Ribofuranosyl)-1-N-[(S)-4-amino-2-hydroxybutanoyl]-
neamine (Butirosin B). Please refer to the procedure for the
preparation of neokacin. 1H NMR (400 MHz, D2O) δ 5.96 (d, J )
3.9 Hz, 1H), 5.33 (s, 1H), 4.29 (dd, J ) 8.2, 4.0 Hz, 1H), 4.1-4.2
(m, 2H), 3.8-3.9 (m, 7H), 3.67 (dd, J ) 12.6, 5.3 Hz, 1H), 3.59
(dd, J ) 10.2, 8.8 Hz, 1H), 3.45 (dd, J ) 9.8, 9.1 Hz, 1H), 3.3-
3.4 (m, 3H), 3.2 (m, 1H), 3.1 (m, 2H), 2.1 (m, 2H), 1.9 (m, 1H),
1.68 (ddd, J ) 12.6, 12.5, 12.3 Hz, 1H, H-2ax); 13C NMR (100
MHz, D2O) δ 175.8, 110.5, 95.8, 86.0, 82.4, 77.4, 75.6, 74.1, 70.9,
69.8, 69.3, 69.1, 68.6, 60.8, 53.9, 49.2, 49.1, 40.4, 36.8, 31.1 (2C);
HRESI calcd for C21H42N5O12 ([M + H]+) m/z 556.2829, measure
m/z 556.2835.
5.39 (d, J ) 1.6 Hz, 1H), 5.27 (s, 1H), 4.53 (dd, J ) 6.9, 4.9 Hz,
1H), 4.39 (dd, J ) 4.7, 2.0 Hz, 1H), 4.3 (m, 1H), 4.29 (dd, J )
8.1, 3.7 Hz, 1H), 4.20 (dd, J ) 3.1, 3.0 Hz, 1H), 4.1 (m, 1H),
3.8-4.0 (m, 6H), 3.75 (dd, J ) 12.4, 4.9 Hz, 1H), 3.63 (dd, J )
10.2, 8.8 Hz, 1H), 3.5 (m, 2H), 3.45 (dd, J ) 10.2, 9.7 Hz, 1H),
3.3-3.4 (m, 5H), 3.25 (dd, J ) 13.7, 6.9 Hz, 1H), 3.1 (m, 2H), 2.1
(m, 2H), 1.9 (m, 1H), 1.67 (ddd, J ) 12.7, 12.5, 12.4 Hz, 1H,
H-2ax); 13C NMR (100 MHz, D2O) δ 175.8, 110.4, 95.8, 95.5, 86.0,
81.3, 77.7, 75.3, 74.1, 73.8, 71.0, 70.4, 69.8, 69.3, 68.7, 67.9, 67.6,
60.4, 53.9, 51.1, 49.2, 49.1, 40.7, 40.4, 36.9, 31.4, 31.1; HRESI
calcd for C27H54N7O15 ([M + H]+) m/z 716.3677, measure m/z
716.3695.
5-O-(2,3,5-Tri-O-acetyl-â-D-ribofuranosyl)-6,3′,4′-tri-O-acetyl-
1,3,2′,6′-tetraazidoneamine (25). A mixture of compounds 24 (0.32
g, 0.89 mmol) and 4 (0.41 g, 0.74 mmol) and activated 4 Å
molecular sieves was stirred in anhydrous CH2Cl2 (15 mL) at room
temperature for 5 h. N-Iodosuccinimide (0.20 g, 0.89 mmol) was
quickly added into the above solution, and the reaction mixture
was cooled to -60 °C. After the solution was warmed to -50 °C,
trimethylsillyl triflate (20 µL, 0.13 mmol) was added. The solution
was stirred overnight, and TLC of the reaction showed the complete
consumption of glycosyl donor (Rf ) 0.27, monitored by TLC, Rf
) 0.27, hexane:EtOAc ) 65:35). The reaction was quenched by
addition of NaHCO3(s), Na2SO4‚10H2O(s), and Na2S2O3(s) then
filtered through Celite. After removal of solvents, the crude product
was redissolved in EtOAc. The organic solution was washed with
water and brine and dried over anhydrous Na2SO4. Removal of
solvent and purification with a gradient column chromatography
(hexane:EtOAc ) 80:20 to 30:70) afforded the desired product (0.22
1
g, 0.27 mmol, 37%). H NMR (400 MHz, CDCl3) δ 5.90 (d, J )
3.8 Hz, 1H, H-1′), 5.45 (dd, J ) 10.6, 9.2 Hz, 1H, H-3′), 5.42 (d,
J ) 3.7 Hz, 1H, H-1′’), 5.19 (dd, J ) 4.9, 4.6 Hz, 1H, H-3′′), 5.10
(dd, J ) 4.8, 3.8 Hz, 1H, H-2′′), 5.01 (dd, J ) 9.9, 9.4 Hz, 1H,
H-4′), 4.97 (dd, J ) 9.8, 8.7 Hz, 1H, H-6), 4.4 (m, 1H, H-5′), 4.44
J. Org. Chem, Vol. 72, No. 11, 2007 4065

Li et al.
6-O-(3-Azido-2,4,6-tri-O-benzyl-r-D-glucopyranosyl)-1-N-[(S)-
2-O-benzyl-4-(benzyloxycarbonylamino)butanoyl]-3,2′,6′-triazi-
doneamine (29a). Please refer to the general procedure for
glycosylation and hydrolysis. 1H NMR (400 MHz, CDCl3) δ 7.2-
7.3 (m, 25H), 7.05 (d, J ) 5.7 Hz, 1H, NH), 5.60 (d, J ) 3.6 Hz,
1H), 5.29 (dd, J ) 5.7, 5.5 Hz, 1H, NH), 5.1 (m, 2H), 4.99 (d, J
) 12.0 Hz, 1H), 4.74 (d, J ) 10.7 Hz, 1H), 4.68 (d, J ) 13.5 Hz,
1H), 4.66 (d, J ) 12.2 Hz, 1H), 4.57 (d, J ) 12.3 Hz, 1H), 4.55
(d, J ) 12.5 Hz, 1H), 4.5 (m, 1H), 4.45 (d, J ) 12.2 Hz, 1H), 4.40
(d, J ) 10.7 Hz, 1H), 4.2 (m, 1H, H-5′), 3.99 (dd, J ) 9.5, 9.5 Hz,
1H), 3.93 (dd, J ) 6.6, 4.7 Hz, 1H), 3.83 (dd, J ) 10.1, 9.8 Hz,
1H), 3.8 (m, 1H), 3.7 (m, 1H), 3.1-3.6 (m, 14H), 2.6 (m, 1H,
H-2eq), 2.1 (m, 1H), 1.9 (m, 1H), 1.20 (ddd, J ) 13.2, 12.4, 12.3
Hz, 1H, H-2ax); 13C NMR (100 MHz, CDCl3) δ 173.9, 156.6, 137.5,
137.4, 137.2, 136.8, 136.7, 129.0, 128.9, 128.8, 128.6 (2C), 128.5,
128.3 (2C), 128.2, 128.1, 127.9, 127.5, 98.4 (2C), 85.4, 79.7, 78.3,
77.4, 76.3, 75.0, 73.6 (2C), 73.1, 71.8, 71.6 (2C), 71.3, 67.9, 66.9,
65.2, 63.0, 59.2, 51.6, 48.4, 37.4, 32.4, 32.2, 29.8; HRESI calcd
for C58H67N14O14 [M + H]+ m/z 1183.4961, measure m/z 1183.4961.
6-O-(4-O-((R)-3-Azido-2-((R)-3-azido-2-O-hydroxylpropyl)-
propyl)-2,3-di-O-benzyl-6-deoxy-r-D-glucopyranosyl)-1-N-[(S)-
2-O-benzyl-4-(benzyloxycarbonylamino)-butanoyl]-3,2′,6′-tri-
azidoneamine (29b). Please refer to the general procedure for
glycosylation and hydrolysis. 1H NMR (400 MHz, CDCl3) δ 7.2-
7.4 (m, 20H), 7.11 (d, J ) 5.4 Hz, 1H, NH), 5.60 (d, J ) 3.6 Hz,
1H), 5.3 (m, 1H, NH), 5.07 (d, J ) 12.0 Hz, 1H), 4.96 (d, J )
12.1 Hz, 1H), 4.88 (d, J ) 11.2 Hz, 1H), 4.70 (d, J ) 11.3 Hz,
1H), 4.65 (d, J ) 12.3 Hz, 1H), 4.55 (d, J ) 12.3 Hz, 1H), 4.6 (m,
1H), 4.48 (s, 2H), 4.2 (m, 1H), 3.97 (dd, J ) 9.7, 9.5 Hz, 1H),
3.93 (dd, J ) 6.4, 5.1 Hz, 1H), 3.7-3.8 (m, 4H), 3.3-3.6 (m, 15H),
3.1-3.3 (m, 4H), 3.14 (dd, J ) 9.7, 9.0 Hz, 1H), 2.90 (dd, J )
9.2, 9.2 Hz, 1H), 2.56 (ddd, J ) 12.8, 4.1, 3.5 Hz, 1H, H-2eq), 2.1
(m, 1H), 1.9 (m, 1H), 1.3 (m, 1H, H-2ax), 1.22 (d, J ) 6.2 Hz, 3H,
CH3); 13C NMR (100 MHz, CDCl3) δ 174.0, 156.6, 138.7, 138.6,
137.6, 136.6, 129.0, 128.9, 128.7 (2C), 128.6, 128.5 (2C), 128.4,
128.2, 127.9, 127.7, 127.3, 127.0, 98.8, 98.5, 85.0, 84.3, 81.2, 80.1,
79.5, 79.3, 78.2, 76.8, 75.3, 73.8, 73.5, 73.1, 71.9, 71.8, 71.6 (2C),
69.9, 68.4, 66.9, 63.2, 59.1, 53.3, 51.8, 51.6, 48.4, 37.5, 32.5, 29.8,
17.9; HRESI calcd for C57H72N17O16 [M + H]+ m/z 1250.5342,
measure m/z 1250.5346.
6-O-(3-Amino-r-D-glucopyranosyl)-1-N-[(S)-4-amino-2-hy-
droxybutanoyl]neamine (JLN007). Please refer to the procedure
for the preparation of neokacin. 1H NMR (400 MHz, D2O) δ 5.81
(d, J ) 3.8 Hz, 1H), 5.13 (d, J ) 3.7 Hz, 1H), 4.24 (dd, J ) 9.2,
3.7 Hz, 1H), 3.9-4.1 (m, 3H), 3.91 (dd, J ) 10.7, 9.1 Hz, 1H),
3.7-3.9 (m, 6H), 3.64 (dd, J ) 10.1, 10.1 Hz, 1H), 3.44 (dd, J )
9.6, 9.5 Hz, 1H), 3.3-3.4 (m, 4H), 3.24 (dd, J ) 13.6, 7.0 Hz,
1H), 3.1 (m, 2H), 2.1 (m, 2H), 1.9 (m, 1H), 1.74 (ddd, J ) 12.7,
12.6, 12.6 Hz, 1H, H-2ax); 13C NMR (100 MHz, D2O) δ 175.7,
98.2, 96.4, 80.5, 79.6, 74.9, 72.3, 71.0, 69.8, 69.3, 69.2, 68.2, 65.8,
60.0, 55.3, 53.9, 49.0, 48.8, 40.4, 37.2, 31.4, 31.0; HRESI calcd
for C22H45N6O12 [M + H]+ m/z 585.3095, measure m/z 585.3068.
6-O-(4-O-((R)-3-Amino-2-((R)-3-amino-2-hydroxylpropyl)-
propyl)-6-deoxy-r-D-glucopyranosyl)-1-N-[(S)-4-amino-2-hy-
droxybutanoyl]neamine (JLN033). Please refer to the procedure
for the preparation of neokacin. 1H NMR (400 MHz, D2O) δ 5.95
(d, J ) 3.9 Hz, 1H), 5.01 (d, J ) 4.0 Hz, 1H), 4.24 (dd, J ) 9.2,
3.8 Hz, 1H), 3.9-4.1 (m, 5H), 3.7-3.8 (m, 5H), 3.69 (dd, J ) 9.6,
9.5 Hz, 1H), 3.57 (dd, J ) 10.1, 6.1 Hz, 1H), 3.52 (dd, J ) 5.5,
4.2 Hz, 1H), 3.4-3.5 (m, 6H), 3.2-3.3 (m, 2H), 3.0-3.2 (m, 4H),
3.0 (m, 2H), 2.22 (ddd, J ) 12.9, 4.3, 4.2 Hz, 1H, H-2eq), 2.1 (m,
1H), 1.9 (m, 1H), 1.82 (ddd, J ) 12.8, 12.7, 12.6 Hz, 1H, H-2ax),
1.24 (d, J ) 6.3 Hz, 3H, CH3); 13C NMR (100 MHz, D2O) δ 175.6,
99.2, 95.9, 84.1, 80.6, 78.2, 75.4, 74.9, 72.6, 71.9, 71.1, 70.8, 69.8,
69.4, 68.5, 67.5, 66.9, 53.7, 48.9, 48.8, 41.9, 40.9, 40.3, 37.2, 31.0,
30.5, 17.1; HRESI calcd for C28H58N7O14 [M + H]+ m/z 716.4041,
measure m/z 716.4026.
6-O-(3-Amino-r-D-xylopyranosyl)-1-N-[(S)-4-amino-2-hydroxy-
butanoyl]neamine (JLN040). Please refer to the procedure for the
preparation of neokacin. 1H NMR (400 MHz, D2O) δ 5.81 (d, J )
3.8 Hz, 1H), 5.02 (d, J ) 3.7 Hz, 1H), 4.16 (dd, J ) 9.3, 3.6 Hz,
1H), 4.0 (m, 1H), 3.8-3.9 (m, 3H), 3.6-3.7 (m, 4H), 3.5-3.6 (m,
2H), 3.3-3.4 (m, 4H), 3.1-3.2 (m, 2H), 3.1 (m, 2H), 2.11 (ddd, J
) 12.8, 4.2, 4.0 Hz, 1H, H-2eq), 2.0 (m, 1H), 1.8 (m, 1H), 1.70
(ddd, J ) 12.7, 12.7, 12.7 Hz, 1H, H-2ax); 13C NMR (100 MHz,
D2O) δ 175.5, 97.5, 96.1, 79.6, 78.5, 74.5, 70.8, 69.7, 69.2, 68.6,
68.0, 65.7, 61.9, 55.4, 53.7, 48.7, 48.6, 40.3, 37.1, 30.9, 30.8;
HRESI calcd for C21H43N6O11 [M + H]+ m/z 555.2989, measure
m/z 555.2965.
6-O-(3-Azido-2-O-benzyl-r-D-xylopyranosyl)-1-N-[(S)-2-O-
benzyl-4-(benzyloxycarbonylamino)butanoyl]-3,2′,6′-triazido-
neamine (29c). Please refer to the general procedure for glycosy-
lation and hydrolysis. 1H NMR (400 MHz, CDCl3) δ 7.2-7.4 (m,
15H), 7.18 (d, J ) 5.5 Hz, 1H), 5.50 (d, J ) 2.9 Hz, 1H), 5.25
(dd, J ) 5.4, 5.4 Hz, 1H), 5.10 (d, J ) 12.2 Hz, 1H), 5.03 (d, J )
12.1 Hz, 1H), 4.7 (m, 1H), 4.65 (d, J ) 12.1 Hz, 1H), 4.50 (d, J
) 12.2 Hz, 1H), 4.46 (s, 2H), 4.3 (m, 1H), 4.1-4.2 (m, 1H), 3.8-
4.0 (m, 2H), 3.5-3.7 (m, 5H), 3.4-3.5 (m, 4H), 3.2-3.3 (m, 5H),
3.07 (dd, J ) 9.5, 9.3 Hz, 1H), 2.4 (m, 1H, H-2eq), 1.8-2.0 (m,
2H), 1.2 (m, 1H, H-2ax); 13C NMR (100 MHz, CDCl3) δ 174.1,
156.8, 137.2, 136.8, 136.5, 129.0 (2C), 128.9, 128.8, 128.7 (2C),
128.4, 98.8, 97.7, 83.6, 80.7, 78.4, 77.8, 77.4, 74.1, 73.4, 71.6, 68.9,
67.1, 65.2, 63.1, 59.1, 51.5, 48.1, 37.5, 32.4, 32.3, 29.9; HRESI
calcd for C43H53N14O13 [M + H]+ m/z 973.3916, measure m/z
973.3900.
Acknowledgment. We acknowledge National Institutes of
Health (AI053138) and DARPA for financial support. We thank
Dr. Dobrowolski of Utah State University for assistance in NMR
experiments. We thank Prof. Mobashery of Notre Dame
University for providing the pTZ19U-3 and pSF815 plasmids.
We also thank Prof. Vernon Parker and Prof. Alvan Hengge of
Utah State University for providing valuable advice and Prof.
Steve Scheiner of Utah State University for assistance in
computational analysis.
Supporting Information Available: 1H and 13C spectra of the
synthesized compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
JO062588J
4066 J. Org. Chem., Vol. 72, No. 11, 2007