Synthesis, antimicrobial activity, attenuation of aminoglycoside resistance in MRSA, and ribosomal A-site binding of pyrene-neomycin conjugates

Article abstract of DOI:10.1016/j.ejmech.2018.11.022

The development of new ligands that have comparable or enhanced therapeutic efficacy relative to current drugs is vital to the health of the global community in the short and long term. One strategy to accomplish this goal is to functionalize sites on cur

Full text of DOI:10.1016/j.ejmech.2018.11.022

Accepted Manuscript
Synthesis, antimicrobial activity, attenuation of aminoglycoside resistance in MRSA,
and ribosomal A-site binding of pyrene-neomycin conjugates
Sandra Story, Michael J. Skriba, Krishnagopal Maiti, Nihar Ranjan, Natalya N.
Degtyareva, Keith D. Green, Verjine Khodaverdian, Adegboyega K. Oyelere, Sylvie
Garneau-Tsodikova, Dev P. Arya
PII:
S0223-5234(18)30980-2
DOI:
https://doi.org/10.1016/j.ejmech.2018.11.022
EJMECH 10879
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 26 July 2018
Revised Date: 7 November 2018
Accepted Date: 8 November 2018
Please cite this article as: S. Story, M.J. Skriba, K. Maiti, Nihar Ranjan, N.N. Degtyareva, K.D.
Green, V. Khodaverdian, A.K. Oyelere, S. Garneau-Tsodikova, D.P. Arya, Synthesis, antimicrobial
activity, attenuation of aminoglycoside resistance in MRSA, and ribosomal A-site binding of pyrene-
neomycin conjugates, European Journal of Medicinal Chemistry (2018), doi: https://doi.org/10.1016/
j.ejmech.2018.11.022.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphics Page for Table of Contents
Synthesis, Antimicrobial Activity, Attenuation of Aminoglycoside
Resistance in MRSA, and Ribosomal A-site Binding of Pyrene-
Neomycin Conjugates
a
b
a
c
b
Michael J. Skriba, Sandra Story, Krishnagopal Maiti, NiharRanjan, Natalya N. Degtyareva, Keith
d
e
e,f
d
D. Green, Verjine Khodaverdian, AdegboyegaK.Oyelere, Sylvie Garneau-Tsodikova, and Dev P.
a,b
Arya,

ACCEPTED MANUSCRIPT
Synthesis, Antimicrobial Activity, Attenuation of Aminoglycoside
Resistance in MRSA, and Ribosomal A-site Binding of Pyrene-Neomycin
Conjugates
a
b
b
c
a
Sandra Story, Michael J. Skriba, Krishnagopal Maiti, Nihar Ranjan, Natalya N. Degtyareva,
d
e
e,f
d
Keith D. Green, Verjine Khodaverdian, Adegboyega K. Oyelere, Sylvie Garneau-Tsodikova,
a,b
and Dev P. Arya,
a
NUBAD, LLC, Greenville, SC, United States 29605.
b
Department of Chemistry, Clemson University, Clemson, SC, United States 29634.
c
National Institute of Pharmaceutical Education and Research (NIPER), Raebareli, India 229010.
d
College of Pharmacy, University of Kentucky, Lexington, KY, United States 40536-0596.
e
School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, GA, United States 30332-0400.
f
Parker H. Petit Institute for Bioengineering and Bioscience, Georgia Institute of Technology, Atlanta, GA, United
States 30332.

ACCEPTED MANUSCRIPT
Abstract
The development of new ligands that have comparable or enhanced therapeutic efficacy relative
to current drugs is vital to the health of the global community in the short and long term. One
strategy to accomplish this goal is to functionalize sites on current antimicrobials to enhance
specificity and affinity while abating resistance mechanisms of infectious organisms. Herein, we
report the synthesis of a series of pyrene-neomycin B (PYR-NEO) conjugates, their binding
affinity to A-site RNA targets, resistance to aminoglycoside-modifying enzymes (AMEs), and
antibacterial activity against a wide variety of bacterial strains of clinical relevance. PYR-NEO
conjugation significantly alters the affinities of NEO, for bacterial A-site targets. The
conjugation of PYR to NEO significantly increased the resistance of NEO to AME modification.
PYR-NEO conjugates exhibited broad-spectrum activity towards Gram-positive bacteria,
including improved activity against NEO-resistant methicillin-resistant Staphylococcus aureus
(MRSA) strains.
Keywords: RNA, Aminoglycoside, MRSA, Neomycin, Ribosome.

ACCEPTED MANUSCRIPT
Introduction
Multidrug-resistant pathogens have been rapidly evolving worldwide over the past few decades,
1
greatly reducing the effectiveness of current drugs. Targeting nucleic acids with small molecules
can bean effective therapeutic strategy for treating bacterial pathogens by disrupting DNA
synthesis, RNA synthesis, and/or interfering with translation. Inhibition of protein synthesis is a
key feature of many antimicrobials. Aminoglycosides are a class of small molecules that bind to
2
nucleic acids (RNA) and have been well established therapeutics since the 1940s.
Aminoglycoside binding occurs through electrostatic, van der Waals, and hydrogen bonding
interactions between the positively-charged amino sugars and the negatively-charged nucleic
acid (RNA) backbone and nucleotide bases. The aminosugars bind tightly to the A-site 16S
ribosomal RNA (rRNA) with high affinity and cause translational errors resulting in ultimate cell
3
-4
death in bacteria. A-form nucleic acid structures, as adopted by the RNA, exhibit deep and
narrow major grooves, which has been shown to be the ideal structural motif for binding of
5
-9
aminoglycosides. Aminosugars can even bind DNA oligonucleotide duplexes such as 5’G C -
4
4
6
7
-
3
’DNA that has A-form characteristics, with a sub-micromolar binding affinities(K ~10 -10 M
a
1
10
),
much higher binding affinity than for predominantly B-form, AT-rich
1
0
sequences. Neomycin B (NEO)is an attractive compound to study nucleic acid binding since it
binds with a higher affinity to a variety of nucleic acid shapes compared to other
aminoglycosides and can be easily modified with different functionalities without loss of the
RNA binding amino groups. This understanding has led to recognition of alternate nucleic acid
1
1
sequence target sites for drug development, including the B-form DNA.
Through
functionalization of the 5"-hydroxy group on ring III of NEO with intercalators, DNA minor
groove binders, and through dimerization of NEO or other comparably functionalized
aminoglycosides, many different binding sites for aminoglycoside-based compounds have been
1
2-15
established.
Functionalization of the 5"-OH of NEO has been shown to lead to higher

ACCEPTED MANUSCRIPT
binding affinities to different aminoglycoside binding sites, decreased resistance to
aminoglycoside-modifying enzymes (AMEs), and improved RNA selectivity and antibacterial
1
6-17
effects.
In bacteria, the aminoglycoside binding-site is the bacterial 16s rRNA A-site. As
bacterial resistance to traditional antibiotics continues to spread, novel aminoglycoside
modifications offer considerable value in identifying effective leads that can circumvent bacterial
resistance mechanisms. Modification of NEO through functionalization of the 5"-OH position
represents one pathway of evading potential antibiotic resistance mechanisms while maintaining
comparable rRNA affinity and specificity of NEO to its target binding site.
Pyrene (PYR) and its derivatives are formed through combustion of organic compounds and are
useful in industry and biochemical applications used to make plastics, pesticides, dyes, and
fluorescent molecular probes. PYR is a relatively weak, non-specific nucleic acid intercalator
that can be easily derivatized. PYR-NEO conjugates can be rapidly synthesized with various
linker lengths through amide and thiourea couplings. Therefore, we postulated that PYR-NEO
conjugates would provide a new class of NEO derivatives that exhibit differential affinities to the
bacterial and mammalian RNA A-sites, compared to NEO, in addition to altered resistance and
antibacterial effects.
In this report, we present the synthesis, rRNA A-site binding characteristics of a series of PYR-
NEO conjugates with varying linker lengths using fluorescein-NEO (F-NEO) displacement, their
activity as AME substrates, and their antibacterial effects against Gram-positive and Gram-
negative pathogens. Five rRNA A-site homologues were screened that included Escherichia coli,
human cytosolic, mitochondrial, and mitochondrial mutant A-site rRNA models. IC50 values
were obtained to determine the efficacy of these conjugates compared to NEO. The conjugates
were also assayed for antibacterial activity against a wide range of bacteria of clinical

ACCEPTED MANUSCRIPT
importance. We identified an optimal linker-length for PYR-derivatized NEO that invokes
selectivity to E. coli A-site compared to human A-site targets. Most PYR-NEO conjugates were
remarkably resistant to the action of AMEs on the NEO molecule, and subsequently showed
much improved activity against NEO-resistant MRSA strains, as compared to NEO. These
results affirm that functionalization of NEO with readily available and simple compounds is a
viable strategy of developing new antibiotics that have the potential to overcome known
antibiotic resistance mechanisms.
Results and discussion
Synthesis
1
6,
The functionalization of the 5"-OH position on NEO was performed as previously reported,
1
8
and the synthesis of the PYR-NEO conjugates10-16(Figure 1) is summarized in Scheme 1. The
amine groups on rings I, II, and IV are critical for electrostatic binding of NEO to nucleic acids,
so the preferred functionalization site chosen was the 5"-OH group on ring III.Conjugation of
NEO with readily available starting materials, such as PYR activated esters, with simple
derivatization techniques allows for rapid generation of conjugates, and many of these
conjugates were prepared in only a few simple steps requiring mild reaction conditions.

ACCEPTED MANUSCRIPT
Figure 1. Structures of PYR-NEO conjugates10-16 used in this study. Compound purity was verified by RP-HPLC
and HPLC purity profiles are included in the Supporting Information.

ACCEPTED MANUSCRIPT
Scheme 1. Synthesis of PYR-NEO conjugates 10-16. Reaction conditions and yields: (a) 1-3, N-
hydroxysuccinimide, EDC, 1,4-dioxane, 24 h, room temperature, 82-90%. (b) 6, 1,6-hexanediamine, 1,4-dioxane, 12
h, room temperature, 54%. (c) TCDP, DMAP, Py, 6 h, room temperature, 79%. (d) DMAP, Py, 10 h, room
2 2
temperature, 59-73%. (e) TFA, CH Cl , 30 min, room temperature, 81-88%.
Compounds 10-16 were prepared by reacting the appropriate pyrene carboxylic acid compounds
-3 with N-hydroxysuccinimide to yield pyrene succinimidyl esters 4-6 of various linker lengths.
1
This procedure afforded stable PYR intermediates that were purified by ethyl acetate extraction
and washing with deionized water. Compound 16 was prepared in multiple steps using a thiourea
linkage. First, pyrene butyric acid was reacted with N-hydroxysuccinimide to yield the pyrene
succinimide ester, with subsequent nucleophilic attack by 1,6-hexanediamine in large (40 mol)
excess. This intermediate was then reacted with thiocarbonyldipyridone (TCDP) to afford the

ACCEPTED MANUSCRIPT
pyreneisothiocyanate7. Isothiocyanate7 was then reacted with NEO amine 8 to afford the desired
conjugate 16after deprotection of Boc groups with an acid. The linker lengths of all conjugates
are listed in Figure 2in ascending order.
Characteristics of rRNA A-site targets
The A-site targets studied in this report exhibit stem-loop secondary structures. The secondary
structures of the A-sites involved in the study are shown in Figure 2. The A-sites exhibit the
highly conserved G-C, the non-canonical U-U, and A-A-A bulge regions. It is known that
aminoglycoside contacts are made on the A-site from rings I and II in NEO in which ring I
1
9
interacts with the A1408-A1493 in a near orthogonal fashion to rings II, III, and IV. The amino
and hydroxyl groups in these positions make contacts that stabilize the aminoglycoside-rRNA
complex. PYR has shown to be a weak nucleic acid intercalator, alone or when bound to an
2
0-21
adjacent cationic, groove bindingmoiety.
The conjugation of PYR was expected to provide
enhanced binding affinity if the proper atomic linker between the III-ring of NEO and pyrene
was utilized. Thus, it was expected that PYR-NEO would bind in this position with enhanced
affinity to the target provided through intercalation or stacking of pyrene between the base pairs.
E. coli
human cytosolic
mitochondrial
C1494U
A1555G
5
’
3’
C
C
G
C
U
G
5’
3’
C
C
G
C
U
G
5’
3’
C
C
G
C
U
G
5’
3’
C
C
G
C
U
G
5’
3’
C
C
G
C
U
G
G
G
C
G
U
C
G
G
C
G
U
C
G
G
C
G
U
C
G
G
C
G
U
C
G
G
C
G
U
C
~
~
~
~
~
~
~
~
A
A
A
A
A
A
A
A
A
A
1
408
A
1408
G
1492
A
1492
A
1492
A
1492
1492
1557
1557
1557
C
A
C
C
G
U
G
G
C
U
C
C
A
A
G
G
C
C
C
C
~
~
C
A
G
G
C
U
C
C
C
A
G
G
C
C
C
C
C
G
G
G
1
410
1490
1410
1490
1494
1555
1494
1555
1494
1555
U
G
U
G
U
G
U
G
U
G
U
C
U
C
U
C
U
C
U
C
Figure 2. A-site secondary structures used in PYR-NEO study highlighting differences in the bulge regions.

ACCEPTED MANUSCRIPT
Affinity of the PYR-NEO conjugates to rRNAA-site targets
The screening technique used to study PYR-NEO binding to the A-site targets is a fluorescence-
2
2
based F-NEO (Figure 3) displacement assay, developed previously. This technique uses a
fluorescent NEO conjugate, which binds to the rRNA A-site specifically within grooves. The
utility of F-NEO is that the fluorophore maintains pH sensitive fluorescent properties while the
2
3
aminosugar exhibits binding properties similar to non-conjugated NEO. Because of such
properties, both moieties can be exploited using a simple fluorescence-based assay to measure
the A-site binding affinities of molecules in direct comparison to NEO and fluorescein,
respectively. As ligand is added to the bound RNA:F-NEO complex, F-NEO is displaced. F-
NEO exhibits considerable fluorescence changes when bound to its target sequence compared to
unbound F-NEO which is comparable to fluorescein. The fluorescence changes are directly
related to binding or dissociation of F-NEO from the A-site by a competing ligand.
Figure 3. Structure of F-NEO, a molecular reporter of compound binding.
The series of PYR-NEO conjugates were screened against five RNA A-site homologues and the
relative binding affinity was compared to NEO. The data shows some dependence of binding on
the linker-length during conjugate binding (Table 1). Conjugates with linker lengths of 4-5 atoms
bound poorly to all rRNA A-sites compared to NEO. However, as the linker length reached six
atoms, binding affinity increased greatly and equaled the binding affinity of NEO in all A-sites.

ACCEPTED MANUSCRIPT
PYR-NEO conjugate affinity to the A-site reached a maximum with a seven-atom linker length.
As the linker length became greater between PYR and NEO, the affinity of the respective
conjugate again reduced, though the percent reduction was not as great as observed for the
smallest linkers. These results show that an optimum atomic linker length of seven atoms from
the NEO 3" ring to the fused rings of PYR to achieve the highest binding affinity. It should be
noted that the intermolecular interactions also depend on the shape of molecules, and kind of
hetero atoms in the linkers. It is likely that the bioactive conformations are affected by the
hydrophilicity and flexibility of the linker, in addition to the length of the linker atoms.
Table 1.Percent displacement of F-NEO by PYR-NEO conjugates upon addition to A-Site rRNA:F-NEO
complex, relative to NEO.
Compound
linker length)
NEO
E. coli
human
mitochondrial
C1410U
A1490G
(
100 ± 8
33 ± 1
33 ± 5
99 ± 6
101 ± 2
88 ± 4
78 ± 10
64 ± 2
100 ± 6
38 ± 14
39 ± 1
99 ± 14
114 ± 7
91 ± 5
100 ± 3
47 ± 1
47 ± 1
89 ± 2
102 ± 7
88 ± 5
74 ± 7
70 ± 1
100 ± 3
73 ± 4
100 ± 6
22 ± 12
22 ± 7
88 ± 5
102 ± 10
65 ± 3
1
1
1
1
1
1
1
0 (4)
1 (5)
2 (6)
3 (7)
4 (8)
5 (9)
6 (18)
65 ± 54
103 ± 1
103 ± 2
86 ± 10
92 ± 1
66 ± 18
75 ± 2
58 ± 20
58 ± 6
91 ± 1
IC of PYR-NEO conjugates to the A-site
5
0
IC values represent the amount of ligand required to displace 50% of F-NEO from the nucleic
5
0
acid target. The lower the IC , the lower the concentration of ligand required to displace F-NEO,
5
0
which indicates a better binding ligand to the target sequence. IC values were determined by
5
0
titrating the A-site:F-NEO complex with small amounts of ligand until 50% F-NEO was
displaced. Since F-NEO experiences a significant increase in fluorescence intensity as it is
displaced from the A-site, differences in initial and final fluorescence values makes F-NEO a
convenient probe for IC determination. Listed in Table 2 are the IC values for selected A-sites
5
0
50

ACCEPTED MANUSCRIPT
with PYR-NEO conjugates. The highest affinity conjugates are12 and 13, which required the
lowest concentration of ligand to displace 50% of F-NEO. The optimum atomic linker length of
conjugate to displace F-NEO is seven atoms, which mirrors the results obtained from A-site
single point screening. For shorter linker lengths between pyrene and NEO (four and five atoms)
atoms, IC values are the highest in the group. However, as the linker length between PYR and
5
0
NEO reaches seven atoms, the IC values exhibit a 3 to 4-fold reduction.
5
0
Table 2.IC values of PYR-NEO conjugates to rRNA A-sites.
50
Compound
E. coli
human
mitochondrial
C1410U
(
linker length)
1
1
1
1
1
1
1
0 (4)
1 (5)
2 (6)
3 (7)
4 (8)
5 (9)
6 (18)
303 ± 2
310 ± 1
170 ± 2
95 ± 2
191 ± 1
148 ± 4
254 ± 1
375 ± 9
447 ± 18
200 ± 36
97 ± 3
204 ± 2
175 ± 1
252 ± 4
339 ± 6
373 ± 2
465 ± 35
90 ± 2
211 ± 18
168 ± 1
235 ± 4
274 ± 20
332 ± 2
149 ± 11
70 ± 5
147 ± 15
126**
180 ± 1
Specificity comparison of PYR-NEO conjugates to A-site
NEO shows slight specificity to human RNA A-site targets compared to a model E. coli target
RNA. A useful relationship to compare affinity of a ligand to bacterial A-site and human A-site
is the selectivity factor. This value is a ratio of IC values of ligand to human and E. coli target
50
2
5
sequences, multiplied by a factor of 5.75. Therefore, ligand values above 5.75 represent
selectivity to bacterial A-site targets, and values below 5.75 represent selectivity to human A-site
targets. NEO exhibits selectivity to human A-site targets compared to E. coli A-sites, with a
selectivity factor of 5.02. Conjugation of pyrene to the 5"-OH position of NEO, however,
changes selectivity from human A-sites to bacterial A-sites. Selectivity factor values of the PYR-
NEO conjugates are given in Table 3. These values show that shorter linker lengths exhibit
higher selectivity to bacterial A-site targets than PYR-NEO with longer linker lengths. The
human and E. coliA-site targets exhibit a conserved G-C, non-canonical U-U, and C-G base-pair

ACCEPTED MANUSCRIPT
before the A-site bulge. The bulge region in human A-site, however, contains a G1408, whereas
inE. coli A-site is contains an adenine base in this position. The region after the bulge also differs
between E. coli and human RNA. PYR-NEO conjugates with shorter linkers could make
favorable interactions in the E. coli rRNA bulge compared to the bulge region in human A-site
targets. For optimum E.coli A-site specificity, five atoms between NEO III ring and PYR affords
sufficient chain length and flexibility for NEO to fit into the bulge region as well as for pyrene to
intercalate or stack. As the linker length increases between NEO and PYR, the entropic costs of
the longer linker begins to inhibit simultaneous interactions of pyrene and NEO. It is likely that
subsequent pyrene intercalation in conserved regions reduces the selectivity between human and
E. coli A-sites.
Table 3.Selectivity factors of PYR-NEO series to E. coli and the human
cytosolic A-site.
Compound
IC50 (E. coli)
IC50 (human)
Selectivity factor
NEO
87
76
5.02
7.12
8.29
6.76
5.87
6.14
6.80
5.70
1
1
1
1
1
1
1
0 (4)
1 (5)
2 (6)
3 (7)
4 (8)
5 (9)
6 (18)
303
310
170
95
191
148
254
375
447
200
97
204
175
252
It is known that NEO binds to all rRNA A-site targets with high affinity as is the case reported
here, whereas PYR has been shown to be a weak nucleic acid non-specific intercalator. The
affinity of NEO to the rRNA A-site determined via the F-NEO screening test has been reported
2
3
previously, with an IC value range of 68-114 nM. Most of the PYR-NEO conjugates show
5
0
much lower affinity towardsrRNAA-sites than NEO, except for 13 which has a slightly lower
binding affinity compared to NEO. However, all of the PYR-NEO conjugates exhibit slightly
higher selectivity to bacterial A-sites than NEO.

ACCEPTED MANUSCRIPT
Subtle differences in atomic linker-length had a varying impact on the binding affinity and,
selectivity to the E. coli A-site compared to the human A-site target. The optimum linker length
for the highest binding affinity was seven atoms, and this conjugate (13) has an almost
equivalent binding affinity as NEO to multiple A-site targets. However, deviation to either a
lower number linker length or greater number linker length greatly reduced binding affinity to
the A-sites. In conjugates with a linker-length below seven atoms, the tethered PYR and a short
linker, likely prevent optimal NEO binding to the A-site. In conjugates with atomic lengths
above seven atoms, the steric bulk and the entropic cost results in lowering the affinities of the
conjugates to the A-site. For most linkers, conjugation of PYR to NEO slightly increases
specificity of NEO to bacterial A-sites compared to human cytosolic A-site. Given that NEO
binds with exceptionally high affinity to rRNA targets when compared to other aminoglycosides,
reduction of affinity could be a means of achieving higher selectivity, as seen from our results.
Screening PYR-NEO conjugates for antibacterial activity and determination of MIC values
The seven PYR-NEO conjugates10-16 were screened at a single concentration (6.25 µM) against
2
4
several bacterial strains to estimate their antimicrobial properties. Table 4 lists the strains used
in this study and relevant characteristics encompassing Gram-positive and Gram-negative strains
of clinical importance. E. coli ATCC 25922 and S. aureus ATCC 29523 are reference strains
used to determine the activity of antibiotics to sensitive strains. Other strains were selected based
on their resistance or intermediate sensitivity to NEO. Initial screening showed that all seven
compounds exhibited broad spectrum antibacterial activity against Gram-positive strains
including methicillin resistant S. aureus compared to Gram-negative strains (Table S1).

ACCEPTED MANUSCRIPT
Table 4. Characteristics of bacterial strains used in this study.
Strain
Relevant characteristics
Gram-positive
Staphylococcus aureus 25923
Staphylococcus aureus NRS77
Staphylococcus aureus 6538
Staphylococcus aureus NorA
Staphylococcus epidermidis 12384
MRSA 33591
MRSA A960649
MRSA SU-5
MRSA M0602
Streptococcus pyogenes C203
Enterococcus faecalis 29212
Enterococcus faecium BM4105-RF
Bacillus anthracis BA852
Listeria monocytogenes 19115
Mycobacterium smegmatis
Gram-negative
NEO(S)
NEO(S)
NEO(S)
NEO(S), norA, efflux mutant
NEO(S)
NEO(R), multidrug-resistant
NEO(R), multidrug-resistant
NEO(R), multidrug-resistant
NEO(R), multidrug-resistant
NEO(R)
NEO(IM)
NEO(IM)
NEO(S)
NEO(S)
NEO(S)
Escherichia coli 25922
Escherichia coli TolC
Escherichia coli H4H
NEO(S)
NEO(S), tolC, efflux mutant
NEO(R), multidrug-resistant
NEO(S)
NEO(S)
NEO(R)
NEO(S)
NEO(S)
NEO(IM)
NEO(S)
Enterobacter cloacae 13047
Acinetobacter baumannii 19606
Pseudomonas aeruginosa 27853
Salmonella enterica NR-12171
Shigella sonnei NR-519
Proteus mirabilis HM-752
Klebsiella oxytoca HM-625
Klebsiella pneumoniae NR-151410
Yersinia pestis NR-4689, A12
NEO(IM)
NEO(S)
Abbreviations: NEO(R), NEO(IM), and NEO(S) refer to NEO-resistant,
NEO-intermediate sensitivity, and NEO-sensitive, respectively.
We have previously demonstrated the length and flexibility of the linker used to conjugate NEO
2
5-26
to small molecules plays an important role in RNA binding affinity and antibacterial activity.
PYR-NEO conjugates with the shortest linkers that exhibited greater selectivity to the E. coli A-
site model, had the lowest antibacterial activity (Table S1). That these compounds expressed the

ACCEPTED MANUSCRIPT
lowest percent binding to all A-sites as determined by the F-NEO displacement assay (Table 1)is
consistent with their low antibacterial activity. Based on the initial screen for antibacterial
activity, compounds 12, 13, 15, and 16 were chosen to identify MIC values for NEO sensitive
strains that demonstrated 30-100% growth inhibition and for all NEO-resistant strains that
exhibited any growth inhibition (Table 5). The observation that the E. Coli ATCC 25923 NEO-
sensitive strain was not susceptible to the PYR-NEO conjugates while the E. coli TolC efflux
mutant was inhibited by these compounds indicates that active efflux of the PYR-NEO conjugate
occurs with Gram-negative bacteria making them more immune to antimicrobial action (Table
5
). In contrast, this pattern was not observed between S. aureus ATCC 25923 and S. aureus
NorA (efflux mutant), indicating the permeability barrier imparted by the outer membrane of
Gram-negative bacteria is also a factor in uptake and retention of PYR-NEO conjugates.
Interestingly, multidrug-resistant E. coli H4H had 30% growth inhibition by the compounds with
the longest linker lengths 15 (9 atom linker) and 16 (18 atom linker).
Table 5.Minimal inhibitory concentration (MIC) values for NEO and select PYR-NEO conjugates.
Strain
S. aureus 25923
S. aureus NorA
NEO
1.56
1.56
12
13
12.5
6.25-12.5
6.25
15
12.5
6.25-12.5
6.25
16
12.5
12.5
12.5
6.25
6.25
3.13
S.
2384
epidermidis 1.56
1
MRSA 33591
MRSA A960649
MRSA SU-5
MRSA M0602
S. pyogenes C202
400
100
200-400
>400
NA
12.5
12.5
25
25
25
12.5
12.5
>25
12.5
>25
12.5
12.5
25
12.5
12.5
25
25
50
6.25
6.25
>25
6.25
>25*
25
25
50
25-50
NA
3.13
6.25
>25
1.56-3.13
>25*
25
NA
6.25
6.25
>25
6.25
>25
B. anthracis BA852 1.56
M. smegmatis
E. coli 25923
E. coli TolC
E. coli H4H
0.78
3.13
1.56
400
MIC values were determined by the microdilution method according to the Clinical and Laboratory Standards
Institute (CLSI). MIC values (µM) were identified from triplicate assays. *A 30% reduction in growth was
observed for compounds 15 and 16 with E. coli H4H.

ACCEPTED MANUSCRIPT
Though none of the compounds was more active than NEO with all NEO-sensitive bacterial
strains, PYR-NEO compounds 12, 13, 15, and 16 had significantly lower MIC values than NEO
for the 4 MRSA NEO-resistant strains with MIC values 4-16 times lower than that of NEO.
These MRSA strains in addition to methicillin resistance are also resistant to other
aminoglycosides (kanamycin, amikacin, gentamicin, and tobramycin). The observation that
PYR-NEO conjugates are more effective than NEO in killing the MRSA strains used here
indicates the PYR addition, with an appropriate linker, has allowed us to circumvent the myriad
of aminoglycoside resistant mechanisms by these strains. MRSA ATCC A960649 possesses
ant4' (4'-aminoglycoside nucleotidyltransferase) and aac6'-aph2" (6'-aminoglycoside
acetyltransferase-2"-aminoglycoside phosphotransferase) AMEs, which make the strain resistant
to gentamicin and other aminoglycosides.
AME activity on PYR-NEO conjugates
To determine if PYR-NEO conjugates could resist the action of AMEs, we tested them against a
panel of 6 AMEs: AAC(6')-Ie, AAC(3)-IV, AAC(2')-Ic, Eis, APH(2")-Ia, and APH(3')-Ia. In
general all AMEs tested showed reduced reaction rates with conjugates 10-16, with the exception
of 15, which showed increased activity with APH(2")-Ia and Eis (Figure 4). Quite remarkably,
all of the PYR-NEO conjugates were poor substrates of Eis. This particular AAC modifies
2
7
multiple amines on aminoglycosides, making it very difficult to design aminoglycosides that
completely avoid modification by Eis. Our previous work using anthraquinone-NEO
2
5
conjugates showed that even large aromatic modifications cannot always slow this enzyme. All
regiospecific AACs showed a reduction in reaction rate with conjugates10-16 when compared to
the rate of the parent NEO. Previous studies by us and others have shown that certain
modifications work better for different enzymes. For example, dimerization of NEO with various
2
6
linkers does not affect the rate of AAC(2')-Ic. Conversely, the addition of short peptide chains

ACCEPTED MANUSCRIPT
and aromatic groups seems to slow the rates of all the AMEs tested, as previously reported by
1
0
us. This observation doesn’t just apply to NEO, other modified aminoglycosides have also
2
8-30
shown enzyme-selective rate impediments.
0-16 with AMEs brings us closer to finding a way of preventing drugs from being modified by
these enzymes.
The knowledge gained by studying compounds
1
2
50
AAC(2’)-Ic
AAC(3)-IV
AAC(6’)-Ie
Eis
200
APH(2”)-Ia
APH(3’)-Ia
150
100
50
0
NEO
15
10
12
13
14
16
11
Compound
Figure 4. AME modification of PYR-NEO conjugates relative to NEO.
In vitro inhibition of translation
A few of the compounds (10, 13, 15, 16) tested for antibacterial activity were assayed in a
cell-free translation system for prokaryotes (Figure S72). Many of the compounds were
similar or poor inhibitors of protein synthesis at nanomolar concentrations, as compared to the
parent aminoglycoside (NEO), with IC50 values in the range of 212-1200 nM, compared with
NEO (IC50 = 139 nM). For example, compound 10, with the short linker, did not inhibit
translation effectively (IC ~1200 nM), was one of the weakest binders in F-NEO
5
0
displacement assays (IC ~300 nM), and exhibited very weak antibacterial activity. These
5
0
results are consistent with previous reports where some aromatic-NEO conjugates’ binding to
2
5
A-sites correlates well with antibacterial activity and translation inhibition. IC values for
50

ACCEPTED MANUSCRIPT
compounds 10 and 13 correlate well with the moderate to high antibacterial activities observed
against a broad spectrum of bacterial strains. For these compounds, the appropriate linker
allows both PYR and NEO binding to inhibit translation. Compounds 15 and 16, containing
longer linkers, are poor inhibitors of translation, yet show Gram-positive antibacterial
activities comparable to 13. An overall weaker translation inhibition activity by most
conjugates is consistent with their lower antibacterial activity, when compared to NEO against
NEO sensitive strains (Table 5 and Table S2). Moreover, while they are much more active
against the NEO resistant strains due to their ability to evade the enzymatic resistance
pathways, their lower translation inhibition may be reflected in the MIC values observed for
the MRSA strains (6.25-12.5µM).
Docking studies
To get possible modes of interaction of PYR-NEO conjugates, we performed docking studies of
three PYR-NEO compounds (11-13) using AutoDock Vina. AutoDock Vina, an improved
version of AutoDock 4, contains a new scoring function providing much precision to ligand-
receptor interaction prediction among other advantages. Our initial docking experiments were
performed with compound 11which exhibited best selectivity factor towards bacterial A-site
inhibition in comparison to human cytosolic A-site sequence. The docking studies of compound
1
1were performed with an oligonucleotide model of prokaryotic A-site reported previously by
3
1
Puglisi group. The results from the docking studies of compound 11 showed that NEO
occupies nearly the same space as occupied by paromomycin, which is structurally very similar
3
1
to neomycin, when compared with NMR structure of paromomycin-A-site complex. The best
conformer obtained from this experiment had the binding energy of -9.4 kcal/mol while the
lowest energy conformation had the binding energy of -8.0 kcal/mol. Further analysis of docking

ACCEPTED MANUSCRIPT
results with compound 11 revealed that polar and van der Waals interactions are the key driving
factors of compound 11’s interaction with the bacterial A-site where the primary amine on ring
IV of NEO displayed hydrogen bonding interactions with the phosphate backbone of A1492
(
Figure 5). The secondary amine on Ring IV of NEO makes polar contacts with the adjacent
cytosine where the interactions in the groove are further strengthened by polar contacts with the
amine connecting NEO with PYR and G1404. The PYR moiety of the molecule accommodates
itself towards central part of the groove making π- π interactions with the G1404, U1405 and
C1406 bases. In contrast, the docking studies with the human A-site sequence revealed that the
best conformation of compound 11 (binding energy of -9.2 kcal/mol) showed interaction of NEO
in the groove spanning the upper and lower ends of the lower stem quite similar to the structure
of paromomycin bound to eukaryotic A-site as determined using solution NMR previously
(
Figure 5). The core difference in the bacterial and human A-site structure lies in the change of
base at 1408 position (adenine in bacterial and guanine in human A-site). This single base
change in bulge region leads to a reduction in the aminoglycoside affinity to the RNA structure.
Further inspection of the binding site provides insights in the differential binding of compound
1
1. The PYR moiety in compound 11, quite dissimilar to its binding to bacterial A-site RNA
sequence, orients itself away from the major groove in the vicinity of U1405 and C1406 bases
making fewer interactions with the RNA. The presence of the PYR moiety into the vacant space
of the groove, drastically reduces van der Waals contacts.

ACCEPTED MANUSCRIPT
Figure 5. Models of compound 11 in lowest energy conformations when bound to (a) bacterial
and (b) human A-site RNA structures. The RNA bases are colored as: adenine (red), guanine
(
green), cytosine (purple) and uracil (orange) The pyrene moiety in compound 11 is colored
black. The boxed region shows the aminoglycoside binding pocket in both the structures.
We then performed docking experiments with compound 13 which afforded the best inhibitory
concentration (IC ) with the A-site RNA sequence. Here again, the NEO rings make several
5
0
polar interactions centered in the groove while PYR protrudes away from the groove (Figure
S71). Finally, we performed docking experiments with compound 12, which had an intermediate
selectivity factor (6.76). The docking results with 12 also show that polar interactions of the
neomycin moiety and the interactions with the pyrene moiety play important roles in conferring
extra binding of the PYR-NEO conjugates. The lowest energy conformations of both bacterial
and human A-site complexes are shown in the supporting information (Figure S71). Additional
docked poses (second and third lowest energy conformations) of compound 11 showed that both

ACCEPTED MANUSCRIPT
neomycin and pyrene moieties occupy same region of the major groove with variations in the
ring conformations of neomycin or flipping of positions of the two units (Figure S 71C).
Toxicity screen of select pyrene-neomycin conjugates in Caenorhabditis elegans.
1
.2
NT
DPA543
DPA544
NEO
1
0
0
0
0
.8
.6
.4
.2
0
5
0
25
12.5
6.25
3.125
1.56
Concentration (µM)
Figure 6. Pyrene-neomycin conjugates DPA 543 and 544 were evaluated for toxicity at a
concentration range of 50 – 1.56 µM as compared to the neomycin control. Compounds were
tested on L1-stage C. elegans and viability was assessed using the MTT assay where formazan
1
production was measured by absorbance at 575 nm after 24 h incubation with each compound.
NT: no treatment, NEO: neomycin. The results are presented as an average of 9 replicates. Error
bars represent the standard deviation from the mean.
The nematode Caenorhabditis elegans has been used to evaluate the potential toxicity of novel
drugs, toxicity to mitochondria and efficacy in clearing multiple-drug resistant Staphylococcus
aureus (MRSA) and other bacteria. We evaluated the toxicity of two pyrene-neomycin
conjugates in the C. elegans animal model. The pyrene-neomycin conjugates DPA543 and
DPA544 were chosen based on their overall higher binding affinity to the bacterial A site rRNA
model, activity against MRSA and low aminoglycoside-modifying enzyme activity relative to
neomycin.
Little significant difference in toxicity was observed between neomycin and the PYR-NEO
conjugates within the concentration range of 6.25 – 50 µM with a decrease in formazan

ACCEPTED MANUSCRIPT
production (an indicator of the nematode larvae viability) ranging from ~ 50 – 75%. There was
no significant difference in toxicity of NEO and the untreated control as compared to the PYR-
NEO conjugates at the lowest concentrations used (3.13 -1.56 µM). The significant reduction in
larvae viability at the lowest concentration of the PYR-NEO conjugates suggests some
contribution by the PYR moiety itself. Previous studies have shown that polycyclic aromatic
hydrocarbons reduce the activity and reproductive capacity of adult nematodes and kill L1-stage
5
0
larvae at concentrations as low as 2.5 µg/mL.
Conclusion
The global spread of multidrug-resistant bacterial strains drives the imperative to find novel
antimicrobials that combat this serious public health problem. We show that the conjugation of
NEOvia the 5"-OH position with PYR and an appropriate linker length leads to higherspecificity
of the PYR-NEO to E. coli A-site rRNAthan NEO alone and demonstrates the utility of simple
and efficient conjugation of readily available nucleic acid binders.The conjugation of pyrene to
NEO through the 5"-OH position significantly resists action from AMEs on NEO. PYR-NEO
conjugates have broad spectrum antibacterial activity towards Gram-positive strains, but not with
Gram-negativebacteria. Use of drugs that compromise the outer membrane of Gram-
3
2
negativebacteria could sensitize them to antibacterial action of the PYR-NEO conjugates.
Significant growth inhibition was observed with S. aureus B. anthracis M. smegmatis, and L.
,
,
monocytogenes.That some PYR-NEO compounds were more effective than NEO in killing
NEO-resistant S. aureus strains and a multidrug-resistant E. colistrain indicates the action of
AMEs have been successfully evaded.Further work to optimize the linker length and
composition leading to improved translation inhibition and antibacterial effects is underway and
will be reported in due course.

ACCEPTED MANUSCRIPT
Experimental section
General methods.Chemicals for the preparation of PYRintermediates were purchased from
Sigma-Aldrich and used as received. NEO was purchased from MP Biomedicals (Solon, OH) as
sulfate salt. Di-tert-butyl dicarbonate was purchased from Advanced Chem Tech (Louisville,
KY). All solvents were purchased from VWR (Atlanta, GA). Silica gel was purchased from
1
13
Sorbent (Atlanta, GA). H NMR and C NMR were recorded on Bruker Avance (300/500 MHz)
1
spectrometers. Chemical shifts are given in ppm and are referenced to residual solvent peaks ( H
1
3
and C NMR). Mass (MALDI-TOF) spectra were collected using a Bruker Microflex mass
spectrometer. Ultraviolet (UV-Vis) spectra were collected on a Varian (Walnut Creek, CA) Cary
1
00 Bio UV-Vis spectrophotometer equipped with a thermoelectrically controlled 12-cell holder.
HPLC analysis of all PYR-NEO conjugates were performed on a HP1100 series analytical
instrument. All conjugates exhibited a final purity of >95%.
Synthesis. The synthesis of the PYR-NEO conjugates 10-16(Figure 1) is summarized in Scheme
1
. All conjugates were purified using column chromatography and were deprotected with
trifluoroacetic acid and lyophilized with a Labconco freeze dryer system prior to use. NMR
characterization was performed on either a Bruker 500 MHz or a JEOL 500 MHz NMR
spectrometer. All mass spectra were obtained using a Bruker MALDI mass spectrometer. All
UV-Vis data were obtained using a Cary 100 UV-Vis spectrophotometer.
Synthesis of 4-(2,3-dihydropyren-2-yl)-N-(6-isothiocyanatohexyl)butanamide (compound
7
).Pyrene succinimidyl ester intermediates (4-6) were prepared according to literature
3
3
protocol. Into a clean and dry 100 mL round bottom flask, 1,6-hexanediamine (5 g, 43 mmol)
was dissolved in 10 mL of anydrous pyridine. The reaction flask was purged with argon and
covered with aluminum foil. In a separate vial, a solution of pyrene succinimidyl ester 6 (0.200

ACCEPTED MANUSCRIPT
g, 0.519 mmol) in 10 mL of dry pyridine was prepared. The solution containing compound 6 was
added dropwise to the reaction flask over 2 h and subsequently allowed to stir overnight. The
solvent was then dried under reduced pressure. The residue was purified via column
chromatography on a silica gel column (CH Cl :MeOH, 9:1 v/v) to afford product 7a as a waxy
2
2
1
yellow solid (0.246 g, 92%): R = 0.38 in 9:1/CH Cl :MeOH; H NMR (500 MHz, DMSO-d6
)
δ
f
2
2
8
8
2
1
1
1
.38 (d,
J
= 9.2 Hz, 1H), 8.26 (d,
J
= 7.7 Hz, 1H), 8.23-8.17 (m, 2H), 8.16-8.09 (m, 2H), 8.09-
= 7.2 Hz, 2H), 3.09-3.04 (m, 2H),
J = 6.0 Hz, 2H), 1.86 (s, 1H), 1.68 (s, 1H), 1.50-1.35 (m, 3H),
.00 (m, 1H), 7.93 (d,
.22 (t, = 6.7 Hz, 2H), 2.00 (d,
6
.34-1.23 (m, 5H), 1.17 (s, 1H); C NMR (126 MHz, DMSO-d ) δ 171.81, 136.75, 131.06,
J = 7.8 Hz, 1H), 7.83 (s, 1H), 3.31 (d, J
J
13
30.60, 129.47, 128.32, 127.69, 127.63, 127.38, 126.67, 126.31, 125.11, 124.97, 124.41, 124.32,
23.66, 50.70, 38.62, 35.21, 32.44, 30.88, 30.63, 29.41, 28.90, 27.80, 26.98, 26.58, 26.31; MS
+
(
MALDI-TOF) m/z Calcd for C H N O: 387.25 (M+H) , Found: 387.33.
2
6
31
2
Pyrene hexamethylene amine 7a (0.1 g, 0.259 mmol) was dissolved in dry pyridine (20 mL) in a
dry 100 mL round bottom flask and 1,1’-thiocarbonyldi-2(1H)-pyridone (TCDP) (0.24 g, 1.04
mmol) was then added to this solution. The flask was wrapped in aluminum foil and allowed to
stir for 5 h. The solvent was dried under reduced pressure and the reaction mixture was purified
via column chromatography to afford the product (compound 7) as a white solid: R
f
= 0.51 in
= 7.8
= 7.6 Hz, 1H), 7.94
= 4.0 Hz, 1H), 3.90 (d, = 12.4 Hz, 1H),
= 6.5 Hz, 2H), 3.36 – 3.27 (m, 2H), 3.06 (m, 2H), 2.22 (t, = 7.2 Hz, 2H), 2.10-1.96
m, 2H), 1.63-1.58 (m, 2H), 1.48-1.37 (m, 3H), 1.35-1.28 (m, 3H). C NMR (126 MHz, DMSO-
1
2
:1/hexanes:EtOAc; H NMR (500 MHz, DMSO-d6
)
δ
8.38 (d,
= 3.4 Hz, 1H), 8.06 (t,
= 5.5 Hz, 1H), 3.99 (d,
J = 9.3 Hz, 1H), 8.27 (t, J
Hz, 2H), 8.22 (dd,
d, = 7.8 Hz, 1H), 7.82 (t,
.63 (t,
J
= 8.5, 4.3 Hz, 2H), 8.13 (d,
J
J
(
J
J
J
J
3
J
J
1
3
(
d
6
)
δ
172.14, 137.04, 131.35, 130.90, 129.77, 128.62, 128.00, 127.93, 127.68, 126.97, 126.61,
1
25.41, 125.27, 124.71, 124.62, 123.97, 53.07, 45.13, 38.79, 35.48, 32.73, 29.64, 29.48, 28.06,

ACCEPTED MANUSCRIPT
-1
2
6.18, 26.09; IR (neat, cm ) 3090 (N-H), 2973, 2927 (C-H), 2173, 2050 (NCS), 750, 710
+
(aromatic C-H); MS (MALDI-TOF) m/z Calcd for C27
H
28
N
2
OS: 428.19 (M) , Found: 428.29.
General procedure for the synthesis of PYR-NEO conjugates (10-16).The pertinent NEO-amine
1
6, 18, 25
intermediates were synthesized using previously established protocols.
To a solution of
the corresponding -Boc-protected NEO amine in dry pyridine, a solution of pyrene
N
succinimidyl ester was added followed by addition of a catalytic amount of 4-
dimethylaminopyridine (DMAP). The flask was purged with argon and covered in aluminum
foil. The reaction mixture was allowed to stir for 18 h and monitored via TLC, to assess the
formation of the product. Pyridine was then dried under reduced pressure. The crude reaction
mixture was purified by column chromatography using CH Cl :MeOH as eluent to afford the
2
2
desired Boc-protected conjugates as white powders (59-82%). The conjugates were subsequently
dissolved in CH Cl and the amines were de-protected via the addition of trifluoroacetic acid. The
crude product was dissolved in deionized water and extracted with CH Cl . Products were then
freeze-dried to afford the final conjugates as white powders in 81-88% yield.
2
2
2
2
N-Boc-protected compound 15 (Protected DPA 503). (13 mg, 67%); R = 0.55 in
f
1
9
:1/CH Cl :MeOH; H NMR (300 MHz, Acetone-d6
)
δ
8.46 (dd,
= 4.6, 2.6 Hz, 1H), 8.13-8.07 (m, 2H), 8.08-7.99 (m, 1H),
= 7.8, 1.9 Hz, 1H), 6.44-6.15 (m, 3H), 6.10-5.88 (m, 2H), 5.16 (d, = 3.1 Hz, 1H),
.96-4.93 (m, 1H), 4.77 (s, 1H), 4.47-4.18 (m, 4H), 4.18 (s, 2H), 4.09-4.00 (m, 2H), 3.89-3.76
J = 9.3, 1.9 Hz, 1H), 8.29-8.24
2
2
(
m, 1H), 8.29-8.20 (m, 2H), 8.18 (dd, J
7
4
.97 (dd,
J
J
(m, 4H), 3.52-3.36 (m, 5H), 3.34-3.14 (m, 3H), 2.55-2.44 (m, 2H), 2.19-2.08 (m, 2H), 1.44-1.40
1
3
(
m, 55H); C NMR (75 MHz, Acetone-
d
6
)
δ
174.65, 158.23, 157.49, 157.02, 156.75, 156.26,
37.40, 132.37, 131.90, 130.86, 129.59, 128.42, 128.40, 128.17, 127.49, 126.88, 125.84, 125.79,
25.67, 124.42, 111.03, 101.05, 99.89, 87.92, 83.81, 90.97, 79.81, 79.64, 79.21, 79.06, 78.88,
1
1

ACCEPTED MANUSCRIPT
7
2
1
5.71, 73.73, 73.05, 72.02, 71.05, 63.46, 57.11, 53.33, 42.66, 41.03, 35.82, 33.82, 33.32, 28.84,
113 7
8.76, 28.70, 28.64, 28.60, 27.88; MS (MALDI-TOF) m/zCalcd for: C75H N O25SNa:
+
567.803 (M+Na) , Found: 1567.23.
N-Boc-protected compound 10 (Protected DPA 541).(9 mg, 78%); R = 0.58 in
f
1
9
8
J
:1/CH Cl :MeOH; H NMR (300 MHz, Acetone-d6
)
δ
8.42 (d,
J
= 9.1 Hz, 1H), 8.26 (s, 1H),
= 2.2 Hz, 2H), 8.02 (d,
= 9.4 Hz, 1H),
2
2
.23 (d,
J
= 2.8 Hz, 1H), 8.20 (s, 1H), 8.18 (d,
J
= 2.3 Hz, 1H), 8.08 (d, J
= 7.4 Hz, 1H), 7.95 (s, 1H), 7.59 (s, 1H), 6.45 (s, 1H), 6.29 (s, 1H), 6.19 (d,
.07 (d, = 10.4 Hz, 2H), 5.84 (s, 1H), 5.14 -5.10 (m, 2H), 5.00 (d, = 1.8 Hz, 1H), 4.22 (d,
.4 Hz, 4H), 4.02 (s, 2H), 3.90 (d, = 7.4 Hz, 1H), 3.81-3.75 (m, 5H), 3.56 (s, 3H), 3.41 (d,
.8 Hz, 6H), 3.26-3.13 (m, 4H), 2.78 (d, = 7.2 Hz, 5H), 2.39 (t,
H), 1.41-1.37 (m, 55H); C NMR (75 MHz, Acetone-d6 174.65, 158.22, 157.60, 157.48,
J
6
5
9
3
1
1
7
5
J
J
J
J
=
=
J
J
J = 7.2 Hz, 3H), 2.20-2.09 (m,
1
3
)
δ
57.02, 156.26, 137.40, 132.37, 131.90, 130.86, 129.59, 128.42, 128.40, 128.18, 128.17, 127.49,
26.88, 125.84, 125.79, 125.67, 124.42, 110.97, 101.06, 99.88, 87.92, 83.83, 79.81, 79.64, 79.20,
9.06, 78.86, 75.71, 73.74, 73.06, 72.03, 71.04, 68.27, 67.59, 63.41, 57.11, 54.95, 53.32, 51.96,
1.81, 42.60, 41.03, 33.83, 33.32, 28.84, 28.76, 28.70, 28.64, 28.60, 27.88; MS (MALDI-TOF)
+
m/zCalcd for C71
H
105
N
7
O
25Na: 1479.64 (M+Na) , Found: 1479.68.
N-Boc-protected compound 12 (Protected DPA 542).(6 mg, 82%); Rf = 0.60in
1
9
:1/CH Cl :MeOH; H NMR (500 MHz, Acetone-d6
)
δ
8.48 (d,
= 8.5, 4.3 Hz, 2H), 8.16-8.10 (m, 2H), 8.06 (t,
.8 Hz, 1H), 6.44-6.29 (m, 2H), 6.27 (d, = 5.7 Hz, 1H), 6.15 (d,
0.3 Hz, 1H), 5.97-5.89 (m, 1H), 5.18 (d,
.28 (m, 3H), 4.21-4.18 (m, 2H), 4.11-4.03 (m, 2H), 3.91-3.83 (m, 2H), 3.83-3.78 (m, 2H), 3.59-
J = 9.3 Hz, 1H), 8.27 (t, J = 6.9
2
2
Hz, 2H), 8.23 (dd,
J
J
J
= 7.6 Hz, 1H), 7.99 (d,
= 8.0 Hz, 1H), 6.04 (t,
J
J
=
=
7
1
4
J
J
= 3.1 Hz, 1H), 4.98-4.96 (m, 1H), 4.78 (s, 1H), 4.41-

ACCEPTED MANUSCRIPT
3
2
1
1
7
2
.52 (m, 2H), 3.52-3.43 (m, 5H), 3.24 (dd,
J
= 13.5, 6.3 Hz, 2H), 2.53 (t,
J
= 7.2 Hz, 2H), 2.20-
): 171.1,
1
3
.13 (m, 2H), 2.10-2.04 (m, 3H), 1.50-1.40 (m, 54H); C NMR (125 MHz,Acetone-
d
6
δ
57.0, 156.1, 155.1, 131.4, 131.0, 130.5, 129.6, 128.8, 127.6, 127.4, 127.0, 126.0, 125.0, 124.9,
24.9, 124.8, 124.6, 123.9, 110.92, 101.0, 99.9, 82.2, 81.7, 79.0, 78.8, 78.3, 78.0, 77.9, 73.1,
2.6, 72.1, 70.1, 56.0, 54.0, 52.5, 50.1, 41.1, 40.0, 39.2, 38.0, 37.4, 34.1, 31.3, 29.4, 29.3, 29.3,
9.2, 29.1, 29.0, 28.9, 28.7, 28.5, 28.4, 27.9, 27.8, 27.7, 27.6, 27.1, 25.1, 25.1, 22.8; MS
+
(
MALDI-TOF) m/zCalcd for C H N O Na: 1506.737 (M+Na) , Found: 1506.341.
7
3
109
7
25
N-Boc-protected compound 13 (Protected DPA 543).(19 mg, 77%); Rf = 0.52in
1
9
3
1
2
3
2
1
1
7
5
2
:1/CH Cl :MeOH; H NMR (300 MHz, Acetone-d6
)
δ
8.47 (d,
J
= 9.3 Hz, 1H), 8.37-8.29 (m,
2
2
H), 8.27 (d,
J
= 1.4 Hz, 1H), 8.18 (s, 2H), 8.16-8.06 (m, 2H), 7.67 (s, 1H), 6.43 (s, 1H), 6.32 (s,
= 9.5 Hz, 1H), 6.15 (d, = 9.5 Hz, 2H), 5.92 (d, = 10.2 Hz, 1H), 5.25-5.12 (m,
= 3.3 Hz, 1H), 4.56-4.37 (m, 3H), 4.37-4.14 (m, 4H), 4.14-
.99 (m, 1H), 3.98-3.75 (m, 4H), 3.74-3.36 (m, 10H), 3.35-3.02 (m, 4H), 2.79 (t, = 6.9 Hz,
172.01, 158.42, 157.57, 157.02,
H), 6.24 (d,
J
J
J
H), 5.10-5.04 (m, 1H), 4.77 (d,
J
J
13
6
H), 1.65-1.21 (m, 54H); C NMR (126 MHz, Acetone-d ) δ
56.68, 156.27, 132.36, 131.94, 131.53, 130.57, 129.77, 128.59, 128.43, 127.96, 127.01, 126.05,
26.02, 125.92, 125.83, 125.58, 124.91, 110.99, 101.96, 100.11, 87.57, 83.18, 82.69, 81.49,
9.95, 79.81, 79.27, 79.04, 78.88, 75.50, 75.34, 74.12, 73.64, 73.16, 72.54, 71.09, 68.26, 56.97,
4.95, 53.51, 52.06, 42.75, 41.77, 40.98, 40.20, 35.91, 34.95, 34.24, 32.29, 28.93, 28.90, 28.81,
8.78, 28.69, 28.63, 26.12; MS (MALDI-TOF) m/zCalcd for C H N O SNa: 1538.709
7
3
109
7
25
+
(
M+Na) , Found: 1538.921.
N-Boc-protected compound 14 (Protected DPA 544).(13 mg, 59%); R
f
= 0.55 in
1
9
:1/CH Cl :MeOH; H NMR (300 MHz, Acetone-d6
)
δ
8.48 (d,
J
= 9.3 Hz, 1H), 8.29-8.26 (m,
2
2

ACCEPTED MANUSCRIPT
1
1
4
3
1
1
1
7
5
2
H), 8.24 (dd,
H), 7.98 (d,
.78 (s, 1H), 4.39-4.32 (m, 2H), 3.91-3.80 (m, 5H), 3.65-3.58 (m, 3H), 3.56-3.49 (m, 2H), 3.48-
.40 (m, 4H), 3.22 (d, = 7.0 Hz, 3H), 2.97 (s, 2H), 2.52 (t, = 7.2 Hz, 2H), 2.29-2.12 (m, 2H),
.68-1.06 (m, 54H); C NMR (126 MHz, Acetone-d6 172.01, 158.42, 157.57, 157.02, 156.68,
J
= 4.7, 2.6 Hz, 2H), 8.20 (d,
J = 3.6 Hz, 1H), 8.16-8.08 (m, 2H), 8.07-8.02 (m,
J
= 7.8 Hz, 1H), 6.50-6.10 (m, 3H), 6.08-5.86 (m, 2H), 5.17 (s, 1H), 4.96 (s, 1H),
J
J
1
3
)
δ
56.27, 132.36, 131.94, 131.53, 130.57, 129.77, 128.59, 128.43, 127.96, 127.01, 126.05, 126.02,
25.92, 125.83, 125.58, 124.91, 110.99, 101.96, 100.11, 87.57, 83.18, 82.69, 81.49, 79.95, 79.81,
9.27, 79.04, 78.88, 75.50, 75.34, 74.12, 73.64, 73.16, 72.55, 72.54, 71.09, 68.26, 56.97, 54.95,
3.51, 52.06, 51.71, 42.75, 41.77, 40.98, 40.20, 39.45, 35.91, 34.95, 34.24, 32.29, 28.93, 28.90,
111 7
8.81, 28.78, 28.69, 28.63, 26.12; MS (MALDI-TOF) m/z Calcd for C74H N O25SNa: 1553.78
+
(
M+Na) , Found: 1554.23.
N-Boc-protected compound 16 (Protected DPA 547).(13 mg, 72%); Rf = 0.48in
1
9
3
7
5
:1/CH
2
Cl
2
:MeOH; H NMR (300 MHz, Acetone-
d
6
)
δ
8.46 (d,
= 7.6 Hz, 1H), 7.98 (d,
= 9.3 Hz, 1H), 6.10 (d,
.92 (s, 1H), 5.22 (s, 1H), 5.10 (s, 1H), 5.02 (s, 1H), 4.79 (s, 1H), 4.23 (d,
s, 1H), 3.94 (s, 1H), 3.82 (d, = 9.9 Hz, 3H), 3.60 (s, 5H), 3.48-3.36 (m, 4H), 3.25 (d,
J = 9.3 Hz, 1H), 8.32-8.24 (m,
H), 8.21 (s, 1H), 8.13 (d,
J
= 1.6 Hz, 2H), 8.07 (d,
J
J
= 7.8 Hz, 1H),
.35-7.18 (m, 2H), 6.41 (s, 1H), 6.29 (s, 1H), 6.20 (d,
J
J = 9.4 Hz, 2H),
J
= 6.9 Hz, 5H), 4.07
= 6.4
= 7.1 Hz, 2H), 2.27-2.13 (m, 2H), 1.69-1.30 (m, 72H), 0.89 (s, 2H); C NMR
75 MHz, Acetone-d6) δ 174.64, 158.24, 157.58, 157.47, 157.01, 156.25, 137.39, 132.36, 131.89,
(
J
J
1
3
Hz, 5H), 2.37 (t,
J
(
1
1
6
30.85, 129.58, 128.41, 128.39, 128.17, 127.48, 126.87, 125.83, 125.78, 125.66, 124.41, 110.96,
01.05, 99.87, 87.92, 83.83, 79.81, 79.64, 79.20, 79.06, 78.86, 75.71, 73.74, 73.05,72.03, 71.06,
8.27, 67.59, 57.11, 54.94, 53.32, 51.97, 42.72, 42.60, 41.03, 33.83, 33.32, 28.84, 28.76, 28.70,