Combined chemical-enzymatic assembly of aminoglycoside derivatives with N-1-AHB side chain

Article abstract of DOI:10.1002/adsc.200800229

A series of unprotected pseudo-disaccharides and pseudo-trisaccharides of 2-deoxystreptamine-containing aminoglycosides have been selectively acylated at the N-1 position with the valuable (S)-4-amino-2-hydroxybutanoyl (AHB) pharmacophore by using the recombinant BtrH and BtrG enzymes from butirosin biosynthesis in combination with a synthetic acyl donor. The process was optimized by performing two enzymatic steps in a sequential manner without purification of the intermediate product.

Full text of DOI:10.1002/adsc.200800229

DEDICATED CLUSTER
COMMUNICATIONS
DOI: 10.1002/adsc.200800229
Combined Chemical-Enzymatic Assembly of Aminoglycoside
Derivatives with N-1-AHB Side Chain
Igor Nudelman,a,d Lilach Chen,a,d Nicholas M. Llewellyn,b El-Habib Sahraoui,b
Marina Cherniavsky,a Jonathan B. Spencer,b,c,* and Timor Baasova,*
a
The Edith and Joseph Fischer Enzyme Inhibitors Laboratory, Schulich Faculty of Chemistry, Technion-Israel Institute of
Technology, Haifa 32000, Israel
Fax : (+972)4-829-5703; phone: (+972)-4-829-2590; e-mail: chtimor@tx.technion.ac.il
Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge CB21EW, United Kingdom
Dr. J. B. Spencer died on April 6, 2008; any correspondence to Cambridge may be directed to Dr. Finian J. Leeper,
b
c
E-mail: FJL1@cam.ac.uk
Equal contributions
d
Received: April 15, 2008; Published online: June 20, 2008
Dedicated to the 60th birthday of Professor Chi-Huey Wong.
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: A series of unprotected pseudo-disacchar-
ides and pseudo-trisaccharides of 2-deoxystrepta-
mine-containing aminoglycosides have been selec-
tively acylated at the N-1 position with the valuable
(S)-4-amino-2-hydroxybutanoyl (AHB) pharmaco-
phore by using the recombinant BtrH and BtrG en-
zymes from butirosin biosynthesis in combination
with a synthetic acyl donor. The process was opti-
mized by performing two enzymatic steps in a se-
quential manner without purification of the inter-
mediate product.
few decades.[4] Among these methods, a selective N-
acylation of the 2-DOS ring (ring II) at the N-1 posi-
tion with the (S)-4-amino-2-hydroxybutanoyl (AHB)
group, turned out to be one of the most effective
modifications of the natural aminoglycosides, restor-
ing the activity of both the kanamycin and neomycin
family of antibiotics against selected resistant bacte-
ria.[5] Earlier investigations in this direction yielded
semisynthetic drugs such as amikacin (the N1-AHB-
derivative of kanamycin A)[5a] and arbekacin (the N1-
AHB-derivative of 3’,4’-dideoxykanamycin B)[5d] that
are largely unaffected by a number of common resist-
ance enzymes and were introduced in clinical practice
in the 1970s and 1990s.
The motivation in exploring the effect of the selec-
tive introduction of the AHB side chain at the N-1
position of different aminoglycosides was largely trig-
gered by the discovery of butirosin.[6] This aminogly-
coside, firstisolaetd from Bacillus circulans, bears the
AHB side chain at N-1 of its 2-DOS moiety, and ex-
Keywords: aminoglycosides; aminoglycosides toxici-
ty; biosynthesis of butirosin; chemoenzymatic syn-
thesis; resistance to aminoglycosides; stop codon
readthrough
2-Deoxystreptamine (2-DOS) aminoglycosides are hibits improved antibiotic activity compared to that of
highly potent, broad-spectrum antibiotics that exert its parent molecule ribostamycin against several ami-
their bactericidal activity by selectively binding to the noglycoside-resistant bacteria. Furthermore, it is of
decoding aminoacyl site (A-site) of the bacterial 16S note that butirosin is also about two times less toxic
rRNA, which leads to the disruption of protein syn- than ribostamycin (LD50 values in mgkgÀ1 of butiro-
thesis by interfering with translational fidelity and sin=520, ribostamycin=260).[7] Importantly, introduc-
translocation.[1] The rapid spread of antibiotic resist- tion of the N-1-AHB moiety to several other amino-
ance towards this family of antibiotics[2] and their rel- glycosides has also been shown to be effective in low-
ative toxicity[3] to mammals are critical problems that ering the acute toxicity of the resulting derivative rel-
largely limit the intensive clinical use of these drugs. ative to that of the parent molecule [neamine LD50 =
To overcome these problems, a wide variety of amino- 125 vs. N-1-AHB-neamine LD50 =260; dibekacin
glycoside derivatives – along with various modifica- LD50 =71 vs. arbekacin (N-1-AHB-dibekacin) LD50 =
tion methods – have been documented during the last 118].[8] Numerous structural and biochemical studies
1682
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1682 – 1688

DEDICATED CLUSTER
COMMUNICATIONS
Combined Chemical-Enzymatic Assembly of Aminoglycoside Derivatives
have revealed the impact of the AHB group on the markedly higher stop codon readthrough activity in
improved antibacterial activity of aminoglycosides. X- cultured mammalian cells[13] and lower toxicity[14]
ray analyses of the bacterial A site in a complex with compared to those of gentamicin (the only aminogly-
amikacin[9] and some other N-1-AHB-modified ami- cosides tested to date in patients) and paromomycin.
noglycosides[10] have demonstrated that the AHB
Encouraged by the observed data on NB30, we
group can increase binding affinity of aminoglycosides sought to further improve both its readthrough activi-
to the A site through making additional direct con- ty and toxicity. For this purpose, we decided to ex-
tacts. In parallel, detailed kinetic analysis of amikacin plore the effect of an AHB group at the N-1 position
and of other synthetic derivatives with several resist- of the pseudo-trisaccharides 5–10. In parallel to this
ance determinant enzymes have revealed that the direction of research, we also strived towards the
AHB side chain can perturb molecular recognition by design of new derivatives of aminoglycosides with im-
resistance enzymes.[11]
proved antibiotic performance. For this purpose, we
Interestingly, in the last several years, aminoglyco- designed the new pseudo-disaccharide scaffold 4 (also
sides have come into the focus of numerous investiga- named NB23)[15] and its pseudo-trisaccharide deriva-
tions that rely on their rather unique ability to induce tives 11–14 (Figure 1). These structures contain a 3’,4’-
mammalian ribosomes to readthrough disease-causing methylidene moiety as a ꢁdefenseꢂ against the action
premature stop codon mutations and generate full- of
various
aminoglycoside
phosphotransferase
length functional proteins in several genetic disor- [APH(3’)] and aminoglycoside nucleotidyltransferase
ders.[12] However, while these studies provided com- [ANT(4’)] resistance enzymes. We reasoned that the
pelling proof of the concept, the high toxicity of clini- installation of an AHB group at the N-1 positions of
cal aminoglycosides, along with their reduced read- 11–14 may further increase the affinity of the result-
through activity at subtoxic doses, largely limits the ing derivatives to the bacterial A site and, conse-
therapeutic utility of the existing aminoglycosides to quently, improve their antibacterial activity and toxic-
treat genetic diseases. Recently, we have reported a ity.
new series of aminoglycoside derivatives that were
Since the regioselective attachment of the AHB
designed for better readthrough activity perfor- moiety on the aminoglycoside structure frequently re-
mance.[13] In this study, the pseudo-disaccharide pa- quires long protection schemes[5] and the efficiency of
romamine 1 was selected as a common scaffold, to a particular strategy is generally dependent on the
which either ribose (compounds 5, 7, and 9) or 5-ami- structure of the parent aminoglycoside, we wanted to
noribose (compounds 6, 8, and 10) were attached at probe a shorter enzymatic approach as an alternative
the 3’, 5, and 6 positions (Figure 1). One of these method. In this context, we have recently reported[16]
structures, compound 8 (also named NB30), showed that the biosynthesis of butirosin from ribostamycin
Figure 1. Sets of 2-DOS-containing pseudo-disaccharides (compounds 1–4) and pseudo-trisaccharides (compounds 5–14) that
were tested for the tolerance toward the enzymatic addition of AHB moiety.
Adv. Synth. Catal. 2008, 350, 1682 – 1688
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1683

DEDICATED CLUSTER
COMMUNICATIONS
Igor Nudelman etal.
Scheme 1. Biosynthetic pathway for the conversion of ribostamycin to butirosin B.
involves two sequential enzymatic steps (Scheme 1). sion of the incubation time resulted in complete con-
The AHB moiety is first transferred from the acyl car- version of paromamine as well. The 6-substituted
rier protein Btrl to the precursor aminoglycoside ri- pseudo-trisaccharides, however, are acylated with
bostamycin as a g-glutamylated dipeptide by the acyl- moderate (compound 10, ~50%; compound 13,
transferase enzyme BtrH to yield g-l-Glu-butirosin ~39%) to poor (compound 9, 19%; compound 14,
B; the protective g-glutamyl group is then cleaved by ~27%) yields. Similar low conversion of the native
the BtrG enzyme via an intramolecular transamida- 4,6-disubstituted aminoglycosides kanamycin A and
tion mechanism. The application of this method to gentamicin C1A under the same conditions was previ-
the combined chemical-enzymatic production of a va- ously observed.[17] The observed modest activity of
riety of novel N-1-AHB-bearing aminoglycosides was BtrH with 6-substituted pseudo-trisaccharides, wheth-
particularly attractive because the recombinant BtrH er the substitution is a pyranose (kanamycin and gen-
and BtrG enzymes are easily accessible as N-terminal- tamicin) or furanose ring (compounds 9–10, and 13–
ly His6-tagged proteins,[16] and that the native acyl 14), suggests that the 6-OH group of the 2-DOS
donor g-l-Glu-AHB-S-Btrl (15, Scheme 1), which is moiety may play an important role for the proper rec-
rather difficult to produce in large quantities, can be ognition of the aminoglycoside substrate and catalysis
very efficiently replaced by the synthetic N-acetylcys- by BtrH enzyme. Except for the modifications at 6-
teamine
thioester
g-l-Glu-AHB-SNAC
(16, OH, all the modifications that have been tested so
Scheme 1).[17] In addition, preliminary tests of this sys- far, either on the pseudo-disaccharide scaffold (nea-
temꢂs tolerance for alternative aminoglycoside accept- mine 2) of the native substrate ribostamycin or on ri-
ors indicated that, in addition to ribostamycin, related bostamycin itself are well tolerated by BtrH. The col-
native aminoglycosides such as neomycin and paro- lective data indicate that BtrH, in combination with
momycin could also be efficiently converted to the the synthetic acyl donor 16, possesses broad substrate
corresponding N-1-AHB derivatives.[17]
tolerance. Furthermore, all the g-Glu-AHB-deriva-
Pseudo-disaccharides 1–4 along with pseudo-trisac- tives generated by a BtrH-assisted enzymatic step
charides 5–14 were tested as alternative substrates for turned out to be excellent substrates for the BtrG;
purified recombinant BtrH in the presence of 16 as they were quantitatively deglutamylated in the next
an acyl donor. The reactions were easily monitored step with the purified recombinant BtrG to yield the
by TLC and the products were analyzed by LC-ESI- corresponding AHB-derivatives (Table 1). Since the
MS/MS (Table 1 and Supporting Information). Satis- synthetic dipeptide 16 also functions as a good sub-
fyingly, all the 3’- and 5-substituted pseudo-trisacchar- strate of BtrG, the experiments were performed in a
ides (compounds 5–8, and 11–12) were essentially stepwise manner; first incubation with BtrH and 16
completely converted to the corresponding g-l-Glu- followed by incubation with BtrG.
AHB-derivatives after 6 h incubation at 208C. Except
Using this method, a number of synthetic com-
the paromamine 1, which gave only 65% conversion, pounds (4, 8, 11 and 12), were converted to the corre-
all the pseudo-disaccharides NB23 (96%), neamine 2 sponding AHB-derivatives in several milligrams scale,
(90%), and gentamine 3 (92%) were also almostcom-
and the purified products were characterized by
pletely converted under the same conditions. Exten- NMR and MS analyses. The combination of standard
1684
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1682 – 1688

DEDICATED CLUSTER
COMMUNICATIONS
Combined Chemical-Enzymatic Assembly of Aminoglycoside Derivatives
Table 1. Selective introduction of the AHB moiety to a wide variety of aminoglycosides (AG) by stepwise incubation with
BtrH and 16 followed by incubation with BtrG.[a]
Yield [%]
AG-g-l-Glu-AHB
(Incubation 6 h)
Observed mass [M+H]+.
AG-g-l-Glu-AHB
product
AG
AG-AHB
(Incubation 24 h)
AG
AG-AHB
product
RB[b]
1
2
3
4
5
6
7
8
100
65
90
92
96
97
99
99
99
19
52
100
100
39
27
100
100
100
100
100
100
100
100
100
100
99
455.17
324.11
323.19
291.07
335.10
456.14
454.92
456.15
455.33
456.12
455.11
467.01
466.14
467.07
466.09
685.26
554.21
553.20
521.16
565.14
686.20
685.22
686.21
685.24
686.24
685.26
697.18
696.22
697.19
696.23
556.20
425.12
424.12
392.08
436.09
557.16
556.16
557.15
556.18
557.19
556.21
568.12
567.19
568.13
567.16
9
10
11
12
13
14
99
91
96
97
[a]
The reactions were performed in a total volume of 50 mL (with the AG concentration of 1.2 mM) and the percentages of
conversion (% yield) for each step were determined by LC-MS analysis of the reaction mixtures as described in Experi-
mental Section. The observed average mass differences of 230.1 Da between the AG-g-l-Glu-AHB products and AG sub-
strates are consistent with the attachment of g-l-Glu-AHB ([M+H]+ calculated at 230.10) moiety for each aminoglyco-
side substrate tested. The observed average mass differences of 101.1 Da between the AG-AHB products and AG sub-
strates are consistent with the attachment of AHB ([M+H]+ calculated at 101.06) moiety for each aminoglycoside sub-
strate tested.
[b]
The natural substrate of BtrH, ribostamycin (RB), was used as a positive control.
1
1
NMR experiments (including H, 13C, H-13C HMQC shifted to the downfield region (d=3.77–3.79 ppm) in
and HMBC, 2D COSY, and 1D selective TOCSY) al- the product N-1-AHB-4 due to the presence of AHB.
lowed unambiguous assignmentof all hte hydrogen
atoms present in each structure. In addition, for each N-1 regioselectivity by BtrH with the synthetic sub-
product, the chemical shift comparisons for 1-H and strates, the N-1-AHB-8 (compound NB54,
3-H protons of the 2-DOS ring with those in the Scheme 2)[18] was chemically synthesized as a repre-
parentcompound unequivocally secured hte regiose- sentative example from paromamine 1 as a starting
To provide full confirmation of the preservation of
lective attachment of the AHB side chain at the N-1 material and its analytical data were compared to
position; the 1-H resonance in the product is signifi- those of a sample prepared by the combined chemi-
cantly de-shielded relative to that of the same proton cal-enzymatic method. As expected, all the 1D and
in the parent compound due to the acyl substitution 2D NMR data of the enzymatic product, under the
at the N-1 position. As a representative example, same pH and counterion conditions, are identical to
Figure 2 traces A and B illustrate the partial 2D those of the synthetic compound (see Supporting In-
COSY spectra of the chemoenzymatically produced formation). Furthermore, preliminary tests of NB54
N-1-AHB-4 and its parent pseudo-disaccharide 4, re- for in vitro readthrough activity of the TGA stop
spectively. Both 2-Heq and 2-Hax protons (2-DOS codon, demonstrated that it exhibits significantly
ring) correlate to the neighbouring 1-H and 3-H pro- higher stop codon readthrough activity than its parent
tons in both compounds; while the resonance of the NB30 (8) and the natural drug paromomycin
3-H proton is almost unchanged (3.38–3.43 ppm), the (Figure 2 C). Further biological tests of NB54 along
resonance of 1-H in 4 (d=3.13–3.14 ppm) is strongly with other N-1-AHB derivatives prepared in this
Adv. Synth. Catal. 2008, 350, 1682 – 1688
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1685

DEDICATED CLUSTER
COMMUNICATIONS
Igor Nudelman etal.
study are underway and will be reported in due variety of 2-DOS-containing aminoglycosides with the
course.
In conclusion, we have developed an efficientche-
valuable AHB pharmacophore by using the BtrH/
BtrG catalytic system with the synthetic acyl donor
moenzymatic method for the preparation of a wide 16. Since aminoglycosides are polycationic, water-
soluble substances, the addition of the AHB side
chain is done in the final steps of the synthesis and, as
such, the presented method significantly shortens and
simplifies the chemical strategies that necessitate long
protection schemes for this purpose. The observed
broad substrate tolerance of the BtrH enzyme for the
synthetic substances tested so far, along with the easy
accessibility of the recombinant BtrH and BtrG en-
zymes also make this method attractive for high
throughput synthesis of a library of 2-DOS-containing
aminoglycosides to discover valuable hits with potent
biomedical relevance. Further investigations in this di-
rection, along with research to develop a suitable
assay for detailed kinetic and mechanistic analysis of
the BtrH-catalyzed reaction with synthetic substrates
are currently underway.
Experimental Section
General Procedure for the Introduction of the AHB
Side Chain
To the reaction mixture (total volume of 50 mL) containing
HEPES buffer (50 mM, pH 7.9), synthetic acyl donor 16
(5 mM) and an aminoglycoside (1.2 mM) was added the pu-
rified BtrH enzyme (125 mg) and the mixture was incubated
at20 8C for 6 h. The protein was removed by addition of
20 mL chloroform followed by vortexing and centrifugation
(13,000 rpm, 5 min). The clear aqueous layer was taken for
the next enzymatic step without further purification. A
small aliquot(about1 mL) of this solution was taken for LC-
ESI-MS/MS analysis to determine the percentage of conver-
sion of the aminoglycoside substrate to the desired product
(Table 1 and Supporting Information). To the aqueous layer
from the previous step was added the purified BtrG enzyme
(18 mg) and the mixture was incubated at 208C for 24 h. The
protein was removed as above and the aqueous layer was
taken for LC-ESI-MS/MS analysis using an Agilent HP1100
HPLC system coupled to a Thermo-Finnigan LCQ ion-trap
mass spectrometer equipped with an electrospray ionization
(ESI) source.
Samples after enzymatic reactions were separated on a
2.0”250 mm Luna 5m C18(2) column (Phenomenex)Àb1y the
following gradientata flow raet of 0.3 mLmin
and
1
Figure 2. Partial 2D H-1H COSY spectra of the chemoenzy-
The p2Luc plasmid containing a TGA C nonsense mutation
in a polylinker between renilla and firefly luciferase genes[20]
was transcribed and translated using the TNTꢃ Reticulocyte
Lysate Quick Coupled Transcription/Translation System
(PromegaTM). Luciferase activity was determined following
90 min incubation using the Dual Luciferase Reporter
Assay System (PromegaTM) and the stop codon suppression
efficiency was calculated as described previously[20]. The re-
sults are averages of at least three independent experiments.
matically produced N-1-AHB-4 (A) and of its parent com-
pound 4 (B). The dashed lines show correlations between 2-
Hax and 2-Heq protons with 1-H and 3-H protons of the 2-
DOS ring, highlighting strong downfield shift of the 1-H
proton in N-1-AHB-4 versus 1-H proton in the parent com-
pound 4. Panel C shows the levels of in vitro suppression of
*
a TGA C nonsense mutation by compound N-1-AHB-8 ( ),
by compound 8 ( ) and by natural drug paromomycin (&).
*
1686
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1682 – 1688

DEDICATED CLUSTER
COMMUNICATIONS
Combined Chemical-Enzymatic Assembly of Aminoglycoside Derivatives
Scheme 2. Chemical transformation of paromamine 1 into NB54.[18]
column temperature of 408C: 0–20 min 10%–50% B, 20–
21 min 50%–10% B, 21–25 min 10% B [buffer A: 0.1% pen-
tafluoropropionic acid (PFPA) in H2O; buffer B: 0.1%
PFPA in MeCN]. Mass spectra were acquired from 250 to
1000 Da. MS/MS was carried out on target ions with 20%
relative collision energy (helium as collision gas).
Commemoration Commission and to the National Science
Foundation Graduate Research Fellowship Program for fi-
nancial support. This work supported by research grants
from the Israel Science Foundation founded by the Israel
Academy ofSciences and Humanities (grant no. 515/07) and
from the US-Israel Binational Science Foundation (grant no.
2006/301) to T.B. and from the Biotechnology and Biological
Sciences Research Council to J.B.S.
Preparative scale reactions were performed as above but
in a total volume of 10–15 mL, and with the addition of
1.5 mg BtrH and 1.0 mg BtrG. The incubation time for both
enzymatic steps was also extended to 24 h. The aqueous
layer, after removal of BtrG, was loaded onto Dowex 50W
+
(NH4 form) 15”80 mm ion-exchange column. The column
was washed with water (50 mL) followed by elution with
1% NH4OH in water. Fractions containing product [TLC:
CH2Cl2/MeOH/H2O/MeNH2 (33% solution in EtOH),
10:15:6:15] were combined and evaporated to dryness. The
residue was dissolved in water, the pH was adjusted to 3.5
by H2SO4 (0.05M) and lyophilized to afford the sulfate salt
which was used for all the spectral analyses.
The recombinant BtrH and BtrG enzymes were isolated
as homogeneous N-terminally His6-tagged proteins accord-
ing to the previously reported procedures.[16] The synthetic
acyl donor 16 was synthesized as previously described.[17]
The pseudo-disaccharides 1,[13] 2, 3 and 4[15] along with the
pseudo-trisaccharides 5–10[13] and 11–14[15] were prepared by
reported strategies.
References
[1] Q. Vicens, E. Westhof, Biopolymers 2003, 70, 42; S.
Magnet, J. S. Blanchard, Chem. Rev. 2005, 105, 477.
[2] G. D. Wright, A. M. Berghuis, S. Mobashery, Adv. Exp.
Med. Biol. 1998, 456, 27.
[3] E. Selimoglu, Curr. Pharm. Des. 2007, 13, 119.
[4] a) W. A. Greenberg, E. S. Priestley, P. S. Sears, P. B.
Alper, C. Rosenbohm, M. Hendrix, S. C. Hung, C. H.
Wong, J. Am. Chem. Soc. 1999, 121, 6527; b) F. Agnelli,
S. J. Sucheck, K. A. Marby, D. Rabuka, S. L. Yao, P. S.
Sears, F. S. Liang, C. H. Wong, Angew. Chem. 2004,
116, 1588; Angew. Chem. Int. Ed. 2004, 43, 1562; c) J.
Li, C. W. Chang, Anti-Infect. Agents Med. Chem. 2006,
5, 255; d) J. Zhou, G. Wang, L. H. Zhang, X. S. Ye,
Med. Res. Rev. 2007, 27, 279.
SupportingInformation
[5] a) H. Kawaguchi, T. Naito, S. Nakagawa, K. I. Fujisawa,
J. Antibiot. (Tokyo) 1972, 25, 695; b) D. Ikeda, T. Tsu-
chiya, S. Umezawa, H. Umezawa, J. Antibiot. 1972, 25,
741; c) S. Kondo, K. Iinuma, H. Yamamoto, K. Maeda,
H. Umezawa, J. Antibiot. (Tokyo) 1973, 26, 412; d) S.
Kondo, K. Iinuma, H. Yamamoto, Y. Ikeda, K. Maeda,
J. Antibiot. (Tokyo) 1973, 26, 705; e) S. H. Lee, C. S.
Cheong, Tetrahedron Lett. 2001, 42, 4801–4815; f) R.
Rai, H. N. Chen, P. G. Czyryca, J. Li, C. W. Chang, Org.
Lett. 2006, 8, 887; g) J. Li, F. I. Chiang, H. N. Chen,
C. W. Chang, J. Org. Chem. 2007, 72, 4055; h) S. H. E.
Swayze, J. Szychowski, S. S. Adhikari, K. Pachamuthu,
X. Wang, M. T. Migawa, R. H. Griffey, (Isis Pharma-
ceuticals Inc), PCT WO2007064954, 2007.
A supplementary table including Rf values of TLC analysis
and retention time data of LC-MS analysis of all the amino-
glycosides tested (1–14) and of their g-l-Glu-AHB and AHB
products, along with the complete characterization of the in-
termediate structures 17, 18, 19 and of the N-1-AHB products
of 4, 8, 11 and 12 can be found in the Supporting Information.
Acknowledgements
We thank John F. Atkins (University ofUtah) of r providing
the p2Luc plasmid. N.M.L. is grateful to the Marshall Aid
Adv. Synth. Catal. 2008, 350, 1682 – 1688
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1687

DEDICATED CLUSTER
COMMUNICATIONS
Igor Nudelman etal.
[6] J. D. Howells, L. E. Anderson, G. L. Coffey, G. D.
Senos, M. A. Underhill, D. L. Vogler, J. Ehrlich, Anti-
microb. Agents Chemother. 1972, 2, 79.
[17] N. M. Llewellyn, J. B. Spencer, Chem. Commun. 2008,
DOI; 10.1039/B802248H.
[18] For the chemical installation of AHB at the N-1 posi-
tion of NB30, we choose an approach that utilizes
direct regioselective differentiation of the aminoglyco-
side amino groups by metal-mediated chelation[19] as il-
lustrated in Scheme 2. Treatment of paromamine with
[7] K. Fujisawa, T. Hoshiya, H. Kawaguchi, J. Antibiot.
(Tokyo) 1974, 27, 677.
[8] S. Kondo, K. Hotta, J. Infect. Chemother. 1999, 5, 1.
[9] J. Kondo, B. Francois, R. J. Russell, J. B. Murray, E.
Westhof, Biochimie 2006, 88, 1027.
Zn(OAc)2 and Boc2O afforded the selectively N-1-Boc-
G
[10] a) J. Kondo, K. Pachamuthu, B. Francois, J. Szychowski,
S. Hanessian, E. Westhof, ChemMedChem 2007, 2,
1631; b) R. J. Russell, J. B. Murray, G. Lentzen, J.
Haddad, S. Mobashery, J. Am. Chem. Soc. 2003, 125,
3410; c) J. B. Murray, S. O. Meroueh, R. J. Russell, G.
Lentzen, J. Haddad, S. Mobashery, Chem. Biol. 2006,
13, 129.
[11] J. Haddad, L. P. Kotra, B. Llano-Sotelo, C. Kim, E. F.
Azucena, M. Z. Liu, S. B. Vakulenko, C. S. Chow, S.
Mobashery, J. Am. Chem. Soc. 2002, 124, 3229.
[12] a) J. F. Burke, A. E. Mogg, Nucleic Acids Res. 1985, 13,
6265; b) R. J. Kaufman, J. Clin. Invest. 1999, 104, 367;
c) M. Manuvakhova, K. Keeling, D. M. Bedwell, RNA
2000, 6, 1044; d) E. Kerem, Curr. Opin. Pulm. Med.
2004, 10, 547; e) M. Hainrichson, I. Nudelman, T.
Baasov, Org. Biomol. Chem. 2008, 6, 227.
[13] I. Nudelman, A. Rebibo-Sabbah, D. Shallom-Shezifi,
M. Hainrichson, I. Stahl, T. Ben-Yosef, T. Baasov,
Bioorg. Med. Chem. Lett. 2006, 16, 6310.
[14] A. Rebibo-Sabbah, I. Nudelman, Z. M. Ahmed, T.
Baasov, T. Ben-Yosef, Hum. Genet. 2007, 122, 373.
[15] L. Chen, M. Hainrichson, D. Bourdetsky, A. Mor, S.
Yaron, T. Baasov, in preparation/revision.
protected paromamine (32% isolated yield), which
after azidation (TfN3) and regioselective acetylation
(Ac2O at low temperature) gave the acceptor 17. Gly-
cosidation of 17 with the trichloroacetimidiate donor
under Lewis acid conditions furnished the desired pseu-
dotrisaccharide 18. Compound 18 was then subjected
to a sequential two-step procedure for the installation
of the AHB moiety: treatment with TFA to remove
the Boc group and the reaction of the resulted N-1
amine with (S)-2-hydroxy-4-azidobutyric acid (HOBT,
DCC, Et3N) afforded the acyl migration product from
the 6 position to N-1 amine as a main product (75%).
To avoid this limitation, 17 was first converted to the
corresponding 6-hydroxy compound 19 in two succes-
sive steps (78%), which after the same two steps for
the installation of AHB followed by treatment with
methylamine and the Staudinger reaction furnished the
desired product NB54 bearing the AHB moiety at the
N-1 position.
[19] H. A. Kirst, B. A. Truedell, J. E. Toth, Tetrahedron Lett.
1981, 22, 295.
[20] G. Grentzmann, J. A. Ingram, P. J. Kelly, R. F. Geste-
land, J. F. Atkins, Rna 1998, 4, 479.
[16] N. M. Llewellyn, Y. Li, J. B. Spencer, Chem. Biol. 2007,
14, 379.
1688
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1682 – 1688