Second generation analogs of rigid 6,7-spiro scaffolds targeting the bacterial ribosome

Article abstract of DOI:10.1016/j.bmcl.2010.10.001

Previous work from our group described the synthesis and biological evaluation of new rigid, 6,6- and 6,7-spiro aminoglycosidic scaffolds targeting the bacterial ribosome. Herein we describe an improved synthetic protocol for their construction, and exten

Full text of DOI:10.1016/j.bmcl.2010.10.001

Bioorganic & Medicinal Chemistry Letters 20 (2010) 7488–7492
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Second generation analogs of rigid 6,7-spiro scaffolds targeting the
bacterial ribosome
Christos I. Stathakis, Ioannis Mavridis, Georgia Kythreoti, Athanasios Papakyriakou, Ioannis A. Katsoulis,
⇑
Thomas Cottin, Panoula Anastasopoulou, Dionisios Vourloumis
Laboratory of Chemical Biology of Natural Products and Designed Molecules, Institute of Physical Chemistry, NCSR ‘DEMOKRITOS’, GR-15310 Ag. Paraskevi Attikis, Athens, Greece
a r t i c l e i n f o
a b s t r a c t
Article history:
Previous work from our group described the synthesis and biological evaluation of new rigid, 6,6- and
6,7-spiro aminoglycosidic scaffolds targeting the bacterial ribosome. Herein we describe an improved
synthetic protocol for their construction, and extend our study by further amino-functionalization of
their 6,7-spiro analogs. The synthetic strategy, preparation and evaluation of some representative exam-
ples are reported.
Received 14 September 2010
Revised 30 September 2010
Accepted 1 October 2010
Available online 8 October 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Antibiotics
Aminoglycosides
Ribosomal RNA
Spiro-compounds
The ribosomal RNA (rRNA) of bacteria is well recognized as the
actual target for many clinically useful antibiotics.1 Aminoglyco-
sides (Fig. 1A) occupy a unique space alongside other RNA-binding
drugs, especially due to their exceptional ability to bind on the
decoding site (A-site) (Fig. 1B) within the 16S subunit of bacterial
ribosome, interfering with the process of protein production.2
Along with specific toxicity issues, aminoglycoside antibiotics suf-
fer from development of resistance, especially due to their exten-
sive use, facts that significantly limit their therapeutic potential.3
Nevertheless, their capacity to bind with high affinity to the bacte-
rial decoding site and several other RNA targets4 renders them a
lead paradigm in RNA molecular recognition and ideal starting
points for the design and synthesis of novel RNA binders.
Previous research efforts from our group revealed a new series of
spiro bicyclic scaffolds (Fig. 1C), which successfully mimicked the
natural products in their ability to specifically bind the ribosomal
decoding center.5 In that work, we hypothesized that the intro-
duced rigidity would ‘lock’ the specific analogs in the desired bioac-
tive conformation, while their decoration with appropriately
positioned amino- and hydroxyl-functionalities would enhance
their binding ability. Thus, 6,6- and 6,7-spiro compounds were de-
signed, and synthesized. Their biological properties were evaluated
in terms of binding on specific RNA subunits as well as in vitro inhi-
bition of protein synthesis in bacterial and eukaryotic systems.5
Stimulated by the promising results of the parental spiro-com-
pounds we anticipated that further functionalization of these
structures could lead to improved biological profiles. Amikacin’s
unique substitution pattern and exceptional antibacterial proper-
ties,6 along with examination of its predicted binding orientation,
directed us to initially concentrate on N1 (DOS numbering,
Fig. 1A). Moreover, our current study would be performed on the
bulkier 6,7-bicyclic analogs, since they could more efficiently occu-
py the binding cavities of both, the bacterial and the eukaryotic
constructs.
From a retrosynthetic perspective, we envisioned the required
N1 versus N3 differentiation to be the outcome of a selective oxa-
zolidinone formation, as presented in Scheme 1. Specifically, we
would take advantage of the immediate proximity of N-1 to the
free 6-hydroxyl, which under basic conditions would lead to a
nucleophilic attack on the adjacent carbamate group, furnishing
intermediate structure III.7,5 Consequently, oxazolidinone III could
be selectively hydrolyzed under controlled basic conditions to
amine II, which could be further functionalized through reductive
amination or acylation transformations, delivering the desired ana-
logs I. In turn, the required precursor IV would be the outcome of
an olefin metathesis reaction from diene V as presented before.5
Regarding positions 30 and 40 of the seven-membered ring (Scheme
1), our docking calculations indicated that both, hydrogen bonding
interactions or
p-stacking, were probable. Thus, retention of the
olefin or dihydroxylation were equally considered for exploitation
purposes. Additionally, the stereochemistry for the dihydroxyla-
tion reaction was of minor importance at this time, since similar
⇑
Corresponding author. Tel.: +30 210 650 3624; fax: +30 210 651 1766.
E-mail address: vourloumis@chem.demokritos.gr (D. Vourloumis).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.10.001

C. I. Stathakis et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7488–7492
7489
B)
A)
R
HO
O
HO
O
HO
H2N
OH NH2
H2N
O
NH2
O
OH
O
R'O
H2N
OH
OH
A
R=NH2, R'=A : Neomycin B
R=OH, R'=A : Paromomycin
R=NH2, R'=H : Neamine
R=OH, R'=H : Paromamine
NH2
HO
HO
3
4
5
2
1
6
NH2
A-site
OH
NH2
2-Deoxystreptamine (2-DOS)
O
HO
HO
C)
H2N
OH
H
NH2
O
OH
O
N
HO
n=1,2
NH2
HO
HO
NH2
n
O
O
HO
H2N
HO
HO
Amikacin
OH
O
Figure 1. (A) Selected examples of aminoglycosides with antibiotic activity. The common feature, 2-deoxystreptamine (2-DOS) core is also presented; (B) Secondary
structure of the bacterial decoding site (A-site) in 16S rRNA. The four base-changes of the eukaryotic sequence are indicated by arrows. The recognition site for the 2-DOS
moiety of aminoglycosides is boxed; (C) New spiro bicyclic analogs demonstrating good affinities for the decoding site of bacterial ribosome.5
hydrogen bonding potential with the RNA was predicted for the
two isomers.
It is important to emphasize that in our previous efforts we
were forced to exchange the amino-protecting carbamates, re-
quired in this study, to the corresponding acetamides, in an at-
The minor diastereomer was never isolated after the final chro-
matographic purification. Upon cleavage of the protecting ketal un-
der acidic conditions, diol 5 was accessed, which was next utilized
for testing the aforementioned hypothesis for amine-differentia-
tion. Indeed, treatment of diol 5 with NaH in DMF,7 produced the
expected oxazolidinone 6 in very good yield (78%). Subsequently,
treatment of oxazolidinone 6 with 0.5 N LiOH in dioxane could
selectively hydrolyze the more labile trans-fused cyclic urethane,
in comparison to the Cbz-carbamate, providing access to the antic-
ipated diol-amine 7. In parallel, dihydroxylation of 5 to tetrol 8 was
accomplished with excellent stereoselectivity, utilizing the condi-
tions presented in Scheme 2.10 Subsequent exposure of 8 to the
same conditions governing the formation of 6, led to an unex-
pected complicated mixture of products. Consequently, the dihydr-
oxylation reaction would be performed as the penultimate step of
our approach.
With the appropriate precursor in hand, our derivatization plan
was realized by means of two distinct protocols. The first involves a
reductive amination condensation with various aldehydes, giving
rise to more complex secondary amines. For illustration of its po-
tential, the sterically demanding and highly lipophilic aldehyde
10 was selected, which under standard conditions (NaBH3CN, cat.
AcOH in MeOH) furnished the corresponding secondary amine in
tempt to overcome the inherent tendency of
1 to react
intramolecularly producing the corresponding oxazolidinone 2
(Scheme 2).5 Also, the dihydroxylation transformation of the spe-
cific series (6,7-spiro) was shown to be non-selective (3:2 d.r.)
regardless of the use of chiral ligands or low temperatures (0 °C).5
With the intention of exploring the viability of the suggested
strategy, we initiated our study from tertiary alcohol 1 (>8:1
d.r).5 The first critical transformation was, as described before,
the allylation of 1, under neutral conditions, so as to avoid the for-
mation of the undesired oxazolidinone. After numerous unsuccess-
ful trials, involving basic (NaH, DIPEA, LiHMDS, Ag2O) or acidic
conditions (allyltrichloroacetamidate and p-TsOH or TfOH or
Sc(OTf)3), the anticipated allylation to ether 3 was accomplished
in very good yield (88%) utilizing allyl-t-butylcarbonate as the al-
lyl-donor in the presence of catalytic amounts of Pd(PPh3)4.8
Ring closing metathesis (RCM) was performed in almost quan-
titative yield, utilizing Grubbs’ second generation catalyst,9 fur-
nishing the corresponding spiro alkene 4 as a single isomer (C-4).
NHR
OH
H2N
O
NH2
OH
H
N
PGHN
O
PGHN
O
2
5
3
4
1
6
O
1'
3'
O
HO
HO
4'
HO
HO
OH
I
OH
II
OH
III
HO
HO
NHPG
O
PGHN
HO
PGHN
O
NHPG
O
NHPG
OH
PGHN
O
O
O
V
OH
IV
VI
Scheme 1. Strategy for the differentiation of the amino groups on the amino- cyclitol moiety. PG = carbamate group.

7490
C. I. Stathakis et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7488–7492
H
N
NHCbz
basic
CbzHN
HO
NHCbz
O
CbzHN
O
NHCbz
O
O
a)
O
O
conditions
O
O
O
2
1
3
b)
CbzHN
O
NHCbz
O
CbzHN
O
NHCbz
OH
CbzHN
O
NHCbz
OH
c)
d)
HO
OH
OH
O
HO
8
5
4
e)
e)
H
N
H
N
CbzHN
O
NH2
OH
CbzHN
O
CbzHN
O
O
O
O
O
HO
HO
f)
OH
OH
OH
7
9
6
Scheme 2. Reagents and conditions: (a) Allyl-t-butylcarbonate (3.0 equiv), Pd(PPh3)4 (0.1 equiv), THF, 45 °C, 88%; (b) Grubbs 2nd generation (5 mol %), CH2Cl2 (0.01 M), 25 °C,
12 h, 91%; (c) THF/AcOH/H2O, 3:3:1, 50 °C, 8 h, 87%; (d) OsO4 (1.2 equiv, 2.5% wt in t-BuOH), TMEDA (1.2 equiv), CH2Cl2, ꢀ78 °C, 30 min, 69%, single diastereomer; (e) NaH
(4.0 equiv, 60% in mineral oil), DMF, 25 °C, 4 h, 78%; (f) 0.5 N LiOH (5.0 equiv), dioxane, 25 °C, 12 h, 74%. AcOH = acetic acid; DMF = dimethylformamide; THF = tetrahy-
drofuran; TMEDA = N,N,N0,N0-tetramethylethylenediamine; TMSCl = trimethylchloro silane.
H
N
H
H
H2N
O
NH2
OH
CbzHN
O
NH2
CbzHN
O
N
CbzHN
O
N
CbzHN
O
NH2
OH
H2N
O
b)
a), b)
R
R
a)
b)
OH
O
OH
HO
HO
OH
OH
OH
7
OH
17: R=
18: R=
OH
15: R=
16: R=
OH
OH
OH
A
B
A
B
12
7
10
11
O
Scheme 3. Reagents and conditions: a) NaBH3CN (2.5 equiv), AcOH (0.2 equiv),
aldehyde 10 (1.1 equiv), MeOH, 25 °C, 8 h, 61%; b) KOH (40.0 equiv), MeOH/H2O
1:1, 95 °C, 12 h, 78% for 11 and 82% for 12.
NHCbz
A
O
13: R =
R
O
OH
N
O
c)
NHCbz
14: R =
O
61% yield (Scheme 3). After basic hydrolysis of the Cbz-carbamate,
functionalized analog 1111 was isolated for further biological eval-
uation. Similar hydrolysis conditions applied on 7 produced the
parent unsaturated compound 1212 for direct comparison. Both
analogs would serve us to evaluate the extent of lipophilicity re-
quired for retention of a reasonable binding affinity to the charged
RNA target.
B
NH2
NH2
OH
H
H
H2N
O
N
H2N
O
N
NH2
NH2
O
O
HO
HO
HO
OH
OH
OH
OH
HO
19
20
The second protocol incorporates an amide coupling with acti-
vated forms of functionalized carboxylic acids. Following a com-
prehensive screening protocol for identifying the ideal coupling
method for the specific transformation (DCC, DCC/4-DMAP, EDC/
Scheme 4. Reagents and conditions: (a) succinate 13 or 14 (2.1 equiv, 0.1 M in
THF), NaHCO3, dioxane/H2O 3:1, 25 °C, 5 h, 58% for 15, 78% for 16; (b) OsO4
(1.2 equiv, 2.5% wt in t-BuOH), TMEDA (1.2 equiv), CH2Cl2, ꢀ78 °C, 30 min, 63% for
17, 70% for 18; (c) 10% Pd/C, 1,3-cyclohexadiene (50.0 equiv), AcOH (0.1 M), 25 °C,
4 h, 82% for 19, 68% for 20.
t
HOBt, HATU, Bu-mixed anhydrides), it was found that the succi-
nate derivatives of the corresponding carboxylic acids represented
the best coupling partner.7 Two examples were selected, one
based on enhancement of the fluorescence emission of 2-aminopu-
rine (2AP) attached to the flexible adenine A1492 of the model oli-
gonucleotide (Table 1, RNA fluorescence assay).15 The biological
activity of the compounds as functional bacterial-translation inhib-
itors was evaluated in a coupled in vitro transcription–translation
assay (IVT; Table 1, Bacterial IVT).16 Additionally, we investigated
the performance of these compounds towards eukaryotic protein
translation inhibition in vitro (Table 1, EIVT).5 Table 1 incorporates
our results in comparison to the values obtained previously for two
natural aminoglycosides and the parent 6,6- and 6,7-bicyclic
analogs.5
involving the
L-hydroxyaminobuteroyl group (amikacin’s side
chain) and the other a lysine, both introducing an extensive hydro-
gen bonding potential to our analogs. Thus, reactions of the corre-
sponding dicarboximides 13 and 14 with diol-amine 7 were
performed, furnishing amides 15 and 16 in good yields (Scheme
4). Both amides were subjected to the pre-described dihydroxyla-
tion conditions (OsO4, TMEDA),10 producing stereoselectively tet-
raols 17 and 18. Two-dimensional NMR experiments on the fully
deprotected derivatives 19 and 20,13,14 formed after hydrogenoly-
sis of the corresponding Cbz-carbamates, confirmed their stereo-
chemistry as the one depicted in Scheme 4.
In order to examine the specificity of our analogs for the ribo-
somal decoding site, an RNA fluorescence assay was employed,
In an attempt to comprehend and, more importantly, ‘visual-
ize’ our results for further development, we employed docking

C. I. Stathakis et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7488–7492
7491
Table 1
(DOS numbering, Fig. 1) maintaining its position relatively to
neamine, while the rest of the skeleton is rotated counterclock-
wise enabling direct interaction of the ring-oxygen with N6 of
A1408, as presented in Figure 2.
Compound 12 represents, as shown by its biological evaluation,
a noteworthy entry in the specific series. Its unexpected binding
affinity, calculated based on a small fluorescence-intensity range
(see Supplementary data) might be indicative of a low efficacy
binding orientation. The latter is affecting only a minor conforma-
tional change, not sufficient to observe the analogous biological ef-
Biological activities for natural aminoglycosides and spiro derivatives
b,c
b,d
b,e
Entry Compounda
RNA EC50
BIVT IC50
EIVT IC50
1
2
Neomycin B
Neamine
0.007
9.5
0.032
5.0
252
120
H2N
O
NH2
OH
3
4
5
6.5
141
74
11.3
>1000
160
HO
OH
HO
H2N
O
NH2
fect. Apparently, the potential of
p-stacking interactions between
the 30,40-double bond and G1491 cannot compensate for the loss
of the two hydroxyls relatively to the parent compounds (entries
4, 5, 6, Table 1). Introduction of a lipophilic side chain sufficiently
recovers some of its biological activity, as indicated by direct com-
parison between 11 and 12. Major improvement in the binding po-
tential was observed by the introduction of hydrophilic chains, as
exemplified in 19 and 20 (entries 8 and 9). Anticipated interactions
of these compounds with the target RNA are presented in (Fig. 2).
Interestingly, the overall effect of these analogs in the inhibition of
bacterial protein translation does not follow the same trend (en-
tries 4 and 5 vs 7, entry 8 vs 9).
168
OH
HO
OH
HO
H2N
O
NH2
OH
118
130
HO
OH
HO
12
11
19
20
6
7
8
9
(8.4)f
63
6.9
>1000
429
186
98
>1000
>1000
>1000
>1000
12
In conclusion, the development of a new synthetic approach en-
abled the construction of four novel 6,7-spiro bicyclic aminoglyco-
side mimics. Our improved synthetic protocol resolved previous
issues for the introduction of an allyl group on a tertiary hydroxyl
by utilizing Pd-chemistry performed under neutral conditions.
Consequently, the retention of the protecting carbamates for the
amines present enabled the differentiation of N1 for further func-
tionalization, through the formation of a trans-oxazolidinone inter-
mediate. Also, careful selection of the dihydroxylation conditions
delivered our final analogs with complete stereocontrol. Overall,
a significant improvement of the binding affinities for the bacterial
ribosome’s decoding center was achieved, in comparison to the
parent compounds.5 Furthermore, we observed that a direct corre-
lation between binding affinities and in vitro biological potency is
not always possible. Application of our findings to other bicyclic
scaffolds and further mechanistic investigations are currently
underway.
a
All final products were quantitated by NMR spectroscopy, utilizing an internal
standard (DiMethylFuran, DMFu).17
b
Concentrations are reported in
lM.
c
RNA-ligand EC50 values were determined by decoding site RNA fluorescence
assay (average of three replicate experiments per compound, 10%).
d
Bacterial in vitro transcription/translation (BIVT) IC50 values were determined by
coupled transcription-translation assay using E. coli extract (average of three rep-
licate experiments per compound, 15%).
e
Eukaryotic in vitro transcription/translation (EIVT) IC50 values were determined
by the TNT quick coupled transcription/translation assay using reticulocyte lysate
(average of three replicate experiments per compound, 15%).
f
Low efficacy in fluorescence-intensity effect (see Supplementary data).
Acknowledgments
This research is supported by Marie Curie Actions, Excellence
Teams (EXT) Grants, Contract number MEXT-CT-2006-039149.
Supplementary data
Supplementary data associated (detailed description of the bio-
logical methods) with this article can be found, in the online ver-
sion, at doi:10.1016/j.bmcl.2010.10.001.
References and notes
Figure 2. Predicted binding orientation of compound 20 (yellow stick carbon chain)
superimposed with the crystallographic position of neamine (cyan stick carbons)
docked at the bacterial A-site. Specific interactions predicted include: (a) the 40-OH
1. (a) Thomas, J. R.; Hengenrother, P. J. Chem. Rev. 2008, 108, 1171; (b) Knowles, D.
J. C.; Foloppe, N.; Matassova, N. B.; Murchie, A. I. H. Curr. Opin. Pharmacol. 2002,
2, 501; (c) Walsh, C. Antibiotics: Actions, Origins, Resistance; ASM: Washington,
2003.
2. (a) Auerbach, T.; Bashan, A.; Harms, J.; Schluenzen, F.; Zarivach, R.; Bartels, H.;
Agmon, I.; Kessler, M.; Pioletti, M.; Franceschi, F.; Yonath, A. Curr. Drug Targets
Infect. Disord. 2002, 2, 169; (b) Moazed, D.; Noller, H. F. Nature 1987, 327, 389;
(c) Vicens, Q.; Westhof, E. ChemBioChem 2003, 4, 1018; (d) Hermann, T. Angew.
Chem. 2000, 112, 1962
with the phosphate backbone of A1492, (b) the
a-NH2 with either U1406 or the
U1406sU1495 base pair, respectively, (c) the terminal NH2 with the phosphates of
G1405 or U1406 and (d) the amide-H with the C1407sG1494 pair. Nitrogen atoms
are shown in blue and oxygen atoms in red.
calculations as previously described.5 Although the absence of a
co-crystal structure renders this method highly speculative,
important interactions may be identified in the process, directing
future synthetic endeavors. As a general trend, the obtained bind-
ing orientation of the new entities was slightly altered, with N-3
3. Rosen, B. P.; Mobashery, S.; Resolving, In. The Antibiotic Paradox Progress in
Understanding Drug Resistance and Development of New Antibiotics; Plenum:
New York, 1998; (b) Vicens, Q.; Westhof, E. Biopolymers 2003, 70, 42.
4. (a) Schroeder, R.; Waldsich, C.; Wank, H. EMBO J. 2000, 19, 1; (b) Carter, A. P.;
Clemons, W. M.; Brodersen, D. E.; Morgan-Warren, R. J.; Wimberly, B. T.;
Ramakrishnan, V. Nature 2000, 407, 340; (c) Fourmy, D.; Recht, M. I.; Blanchard,
S. C.; Puglisi, J. D. Science 1996, 274, 1367; (d) Tor, Y.; Hermann, T.; Westhof, E.

7492
C. I. Stathakis et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7488–7492
Chem. Biol. 1998, 5, R277; (e) Blount, K. F.; Zhao, F.; Hermann, T.; Tor, Y. J. Am.
Chem. Soc. 2005, 127, 9818.
5. Katsoulis, I. A.; Pyrkotis, C.; Papakyriakou, A.; Kythreoti, G.; Zografos, A. L.;
Mavridis, I.; Nahmias, V. R.; Anastasopoulou, P.; Vourloumis, D. ChemBioChem
2009, 10, 1969.
(br s, 1H),2.20 (br s, 1H), 2.15 (d, J = 14.5 Hz, 2H), 2.07–2.03 (m, 1H), 1.95–1.89
(m, 1H), 1.86–1.81 (m, 2H), 1.53 (br s, 1H); 13C NMR (D2O, 125 MHz) d 131.8,
128.3, 78.4, 73.1, 72.3, 60.5, 48.7, 47.3, 33.0, 23.6, 20.8; YRMS-ESI (m/z) [M+H]+
calcd for C11H20N2O3 228.1474, found 228.1471.
13. Compound 19: Rf = 0.12 in MeOH/CH2Cl2/NH4OH(aq.), 10/30/1 v/v; 1H NMR
(500 MHz, D2O) d 4.51 (d, J = 12.9 Hz, 1H), 4.34 (dd, J = 7.2, 3.5 Hz, 1H), 3.73–
3.91 (m, 5H), 3.41 (dd, J = 9.5, 2.0 Hz, 1H), 3.37 (d, J = 9.4 Hz, 1H), 3.17 (m, 2H),
2.54 (dd appt, J = 13.6 Hz, 1H), 2.15–2.20 (m, 1H), 2.01–2.11 (m, 2H), 1.92–2.00
(m, 1H), 1.75–1.89 (m, 3H); 13C NMR (125 MHz, D2O) d 175.4, 77.7, 77.2, 74.6,
71.6, 71.3, 69.5, 64.8, 55.7, 49.2, 36.5, 30.7, 30.0, 27.4, 26.6. YRMS-ESI (m/z)
[M+Na]+ calcd for C15H29N3NaO7 386.1903, found 386.1900.
6. Edson, R. S.; Terrell, C. L. Mayo Clin. Proc. 1999, 74, 519.
7. Haddad, J.; Kotra, L. P.; Llano-Sotelo, B.; Kim, C.; Azucena, E. F., Jr.; Liu, M.;
Vakulenko, S. B.; Chow, C. S.; Mobashery, S. J. Am. Chem. Soc. 2002, 124, 3229.
8. Haight, A. R.; Stoner, E. J.; Peterson, M. J.; Grover, V. K. J. Org. Chem. 2003, 68,
8092.
9. Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18.
10. Donohoe, T. J.; Moore, P. R.; Waring, M. J. Tetrahedron Lett. 1997, 38, 5027.
11. Compound 11: Rf = 0.3 in 50% MeOH/CH2Cl2; 1H NMR (500 MHz, MeOH-d4) d
5.84–5.94 (m, 1H), 5.73 (d, J = 9.4 Hz, 1H), 5.58 (d, J = 9.4 Hz, 1H), 5.04 (d,
J = 15.8 Hz, 1H), 5.01 (d, J = 11.4 Hz, 1H), 4.38 (d, J = 16.9 Hz, 1H), 4.24 (d,
J = 16.9 Hz, 1H), 4.13 (br s, 1H), 3.84 (br s, 1H), 3.74 (d, J = 14.3 Hz, 1H), 3.67 (br
s, 1H), 3.46 (br s, 1H), 2.59 (d, J = 14.3 Hz, 1H), 2.11–2.38 (m, 3H), 1.94–2.07 (m,
2H), 1.87–1.93 (m, 3H), 0.92 (br s, 6H); 13C NMR (125 MHz, MeOH-d4) d 136.4,
132.2, 129.7, 117.6, 80.0, 73.9, 71.1, 61.9, 56.7, 50.9, 46.6, 34.5, 26.2, 26.0, 25.1,
24.2, 23.0. YRMS-ESI (m/z) [M+Na]+ calcd for C18H32N2NaO3 347.2311, found
347.2313.
14. Compound 20: Rf = 0.15 in MeOH/CH2Cl2/NH4OH(aq.), 10/30/1 v/v; 1H NMR
(500 MHz, D2O) d 4.54 (d, J = 13.0 Hz, 1H), 4.04 (t, J = 6.5 Hz, 1H), 3.85–3.95 (m,
3H), 3.75–3.82 (m, 2H), 3.46 (dd, J = 11.1, 5.0 Hz, 1H), 3.40 (d, J = 10.1 Hz, 1H),
3.04–3.09 (m, 2H), 2.57 (t, J = 13.3 Hz, 1H), 1.76–2.08 (m, 9H), 1.47–1.55 (m,
2H); 13C NMR (125 MHz, D2O) d 169.5, 77.6, 77.1, 74.6, 71.6, 71.5, 64.8, 55.6,
53.2, 49.6, 38.9, 30.2, 30.1, 27.3, 26.6, 26.3, 21.3. YRMS-ESI (m/z) [M+Na]+ calcd
for C17H34N4NaO6 413.2376, found 413.2375.
15. Shandrick, S.; Zhao, Q.; Han, Q.; Ayida, B. K.; Takahashi, M.; Winters, G. C.;
Simonsen, K. B.; Vourloumis, D.; Hermann, T. Angew. Chem. 2004, 116, 3239
16. Simonsen, K. B.; Ayida, B. K.; Vourloumis, D.; Takahashi, M.; Winters, G. C.;
Barluenga, S.; Qamar, S.; Shandrick, S.; Zhao, Q.; Hermann, T. ChemBioChem
2002, 3, 1223.
12. Compound 12: Rf: 0.73 (NH4OH(aq)/MeOH, 1:9 v/v); 1H NMR (D2O; 500 MHz) d
5.74 (br s, 1H), 5.60 (br s, 1H), 4.31 (d, J = 16.4 Hz, 1H), 4.21 (d, J = 16.4 Hz, 1H),
4.04 (br s, 1H), 3.87 (br s, 1H), 3.73 (br s, 1H), 3.55 (br s, 1H), 3.46 (br s, 1H), 3.28
17. Gerritz, S. W.; Sefler, A. M. J. Comb. Chem. 2000, 2, 39.