Ribosome-Templated Azide-Alkyne Cycloadditions: Synthesis of Potent Macrolide Antibiotics by in Situ Click Chemistry

Article abstract of DOI:10.1021/jacs.5b13008

Over half of all antibiotics target the bacterial ribosome-nature's complex, 2.5 MDa nanomachine responsible for decoding mRNA and synthesizing proteins. Macrolide antibiotics, exemplified by erythromycin, bind the 50S subunit with nM affinity and inhibit protein synthesis by blocking the passage of nascent oligopeptides. Solithromycin (1), a third-generation semisynthetic macrolide discovered by combinatorial copper-catalyzed click chemistry, was synthesized in situ by incubating either E. coli 70S ribosomes or 50S subunits with macrolide-functionalized azide 2 and 3-ethynylaniline (3) precursors. The ribosome-templated in situ click method was expanded from a binary reaction (i.e., one azide and one alkyne) to a six-component reaction (i.e., azide 2 and five alkynes) and ultimately to a 16-component reaction (i.e., azide 2 and 15 alkynes). The extent of triazole formation correlated with ribosome affinity for the anti (1,4)-regioisomers as revealed by measured K_d values. Computational analysis using the site-identification by ligand competitive saturation (SILCS) approach indicated that the relative affinity of the ligands was associated with the alteration of macrolactone+desosamine-ribosome interactions caused by the different alkynes. Protein synthesis inhibition experiments confirmed the mechanism of action. Evaluation of the minimal inhibitory concentrations (MIC) quantified the potency of the in situ click products and demonstrated the efficacy of this method in the triaging and prioritization of potent antibiotics that target the bacterial ribosome. Cell viability assays in human fibroblasts confirmed 2 and four analogues with therapeutic indices for bactericidal activity over in vitro mammalian cytotoxicity as essentially identical to solithromycin (1).

Full text of DOI:10.1021/jacs.5b13008

Subscriber access provided by RYERSON UNIV
Article
Ribosome-templated azide-alkyne cycloadditions: synthesis
of potent macrolide antibiotics by in situ click chemistry
Ian Glassford, Christiana N. Teijaro, Samer S. Daher, Amy Weil, Meagan C. Small, Shiv
K. Redhu, Dennis J. Colussi, Marlene A. Jacobson, Wayne E. Childers, Bettina Buttaro,
Allen W. Nicholson, Alexander D. MacKerell, Barry S. Cooperman, and Rodrigo B. Andrade
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.5b13008 • Publication Date (Web): 15 Feb 2016
Downloaded from http://pubs.acs.org on February 16, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
Ribosome-templated azide−alkyne cycloadditions: synthesis of potent
macrolide antibiotics by in situ click chemistry
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ian Glassford, 1 Christiana N. Teijaro,1 Samer S. Daher,1 Amy Weil,2 Meagan C. Small,3
Shiv K. Redhu,4 Dennis J. Colussi,6 Marlene A. Jacobson,6 Wayne E. Childers,6
Bettina Buttaro,5 Allen W. Nicholson,4 Alexander D. MacKerell, Jr.,3
Barry S. Cooperman,2 and Rodrigo B. Andrade1,*
1 Department of Chemistry, Temple University, Philadelphia, PA 19122
2 Department of Chemistry, University of Pennsylvania, Philadelphia, PA 19104
3 Department of Pharmaceutical Sciences, School of Pharmacy, University of Maryland,
Baltimore, MD 21201
4 Department of Biology, Temple University, Philadelphia, PA 19122
5 Department of Microbiology and Immunology, Temple University School of Medicine,
Philadelphia, PA 19140
6 Department of Pharmaceutical Sciences, Temple University School of Pharmacy,
Philadelphia, PA 19122
* Corresponding author: tel: +1 215 204 7155; fax: +1 215 204 9851; e-mail:
randrade@temple.edu
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 27
1
2
3
4
5
6
7
8
9
Abstract: Over half of all antibiotics target the bacterial ribosome—Nature’s complex, 2.5 MDa
nanomachine responsible for decoding mRNA and synthesizing proteins. Macrolide antibiotics,
exemplified by erythromycin, bind the 50S subunit with nM affinity and inhibit protein synthesis
by blocking the passage of nascent oligopeptides. Solithromycin (1), a third-generation semi-
synthetic macrolide discovered by combinatorial copper-catalyzed click chemistry, was
synthesized in situ by incubating either E. coli 70S ribosomes or 50S subunits with macrolide-
functionalized azide 2 and 3-ethynylaniline (3) precursors. The ribosome-templated in situ click
method was expanded from a binary reaction (i.e., one azide and one alkyne) to a six-component
reaction (i.e., azide 2 and five alkynes) and ultimately to a sixteen-component reaction (i.e., azide
2 and fifteen alkynes). The extent of triazole formation correlated with ribosome affinity for the
anti (1,4)-regioisomers as revealed by measured Kd values. Computational analysis using the
Site-Identification by Ligand Competitive Saturation (SILCS) approach indicated that the
relative affinity of the ligands was associated with the alteration of
macrolactone+desosamine-ribosome interactions caused by the different alkynes. Protein
synthesis inhibition experiments confirmed the mechanism of action. Evaluation of the minimal
inhibitory concentrations (MIC) quantified the potency of the in situ click products and
demonstrated the efficacy of this method in the triaging and prioritization of potent antibiotics
that target the bacterial ribosome. Cell viability assays in human fibroblasts confirmed 2 and four
analogs with therapeutic indices for bactericidal activity over in vitro mammalian cytotoxicity as
essentially identical to solithromycin (1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Key Words: target-guided synthesis · in situ click chemistry · ribosome · macrolide antibiotics
2
ACS Paragon Plus Environment

Page 3 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
INTRODUCTION
7
8
9
Bacterial resistance to antibiotics is a formidable 21st century global public health threat.1-3
If left unaddressed, we risk moving toward a “post-antibiotic” era.4 While resistance is a natural
consequence of antibiotic use (and abuse), the rate at which pathogenic bacteria have evaded
multiple classes of drugs (including those of last resort) has markedly outpaced the rate at which
new drugs have been introduced. Macrolides are among the safest and most effective antibiotic
classes. To date, three generations have been developed with only the lattermost targeting bacterial
resistance.5,6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A
N
B
N
N
H2N
N3
( )4
O
O
Cu(I)-
catalyzed
Click
N
( )4
Me
+
Me
Me
Me
H2N
O
O
N
N
Me
OMe
Me
OMe
O
O
11
11
Me
NMe2
N
Me
Et
Me
Et
3
Me
HO
O
N
O
Me
O
ODes
O
2
Desosamine
(Des)
O
O
O
O
4
F
Me
F
Me
2
1: Solithromcyin
C
!-stacking
A752
U2609
H-bonding
A2059
G748
A2058
C2611
G2057
Figure 1. (A) Retrosynthetic analysis of solithromycin (1) yields azide 2 and aromatic alkyne 3; (B)
side-chain 4 from telithromycin; (C) Rendering of 1 and key 23S rRNA residues (Cate et al., PDB =
3ORB) using VMD.7
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 27
1
2
3
4
5
6
7
8
Solithromyin (1), one of the most potent macrolide antibiotics (Figure 1A), was prepared
from the Cu(I)-catalyzed Huisgen [3+2] dipolar cycloaddition (i.e., “click”) reaction of azide 2 and
3-ethynylaniline (3).8 Inspiration for 1 came from the erythromycin-derived ketolide telithromycin,
which possesses a structurally related pyridyl-imidazole side-chain 4 (Figure 1B).9
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Over half of all known antibiotics, including macrolides, target the bacterial ribosome.10
Macrolides reversibly bind near the peptidyl transferase center of the 50S subunit with low
nanomolar affinity and inhibit protein synthesis by blocking the passage of nascent
oligopeptides.11,12 The structure of the E. coli 70S ribosome-solithromycin (1) complex confirmed
both the location and mode of binding.13 Like other macrolides, 1 interacts with specific 23S rRNA
residues via the macrolactone ring and desosamine sugar; moreover, the biaryl side-chain attached
at N11 engages in π-stacking interactions with the A752-U2609 base pair and H-bonding with
A752 and G748 (Figure 1C). Accordingly, we reasoned these molecular interactions could be
leveraged in the ribosome-templated synthesis of solithromycin (1) from fragments 2 and 3 (Figure
1A); our target-guided synthesis strategy is illustrated in Scheme 1.14
R
I
R
I
R
I
R
I
N3
N3
N3
B
O
S
O
M
E
B
O
S
O
M
E
B
O
S
O
M
E
B
O
S
O
M
E
1
N
2
N
5
N
4
3
Scheme 1. Ribosome-templated in situ click strategy for antibiotic synthesis. Sequential and
proximal binding of azide- and alkyne-bearing fragments (e.g., 2 and 3, respectively) leads to
irreversible anti (1,4)- and/or syn (1,5)-triazole formation by co-localization. The order in which the
4
ACS Paragon Plus Environment

Page 5 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
azide- and alkyne-functionalized fragments bind the target is determined by the individual binding
affinities.
7
8
9
Target-guided in situ click chemistry is predicated on the selective, proximal binding of
azide- and alkyne-bearing fragments, which lowers the activation energy of irreversible 1,2,3-
triazole ligation by co-localization.15 Unlike the copper-catalyzed click reaction that exclusively
provides the anti (1,4)-triazole16 or the ruthenium-catalyzed variant that exclusively provides the
syn (1,5)-triazole,17 the in situ click process selectively provides the regioisomer that establishes
optimal non-covalent interactions with the target (Scheme 1). Accordingly, the resultant
cycloadduct is expected to have greater affinity for the target than the individual fragments.14 In this
regard, in situ click chemistry represents an extension of fragment-based drug design wherein the
target directly participates in the synthesis of its own inhibitor18,19 and has been successfully
employed in the discovery of potent inhibitors for a number of targets, including: acetylcholine
esterase,20,21,22,23 carbonic anhydrase,24 HIV-protease,25 chitinase,26 protein-protein interactions,27
DNA-recognition,28 EthR (a transcriptional regulator in M. tuberculosis),15,29,30 and a toxic RNA,
which was formed in cellulo.31 Moreover, in situ click chemistry has been used extensively to
create antibody-like protein capture agents.32-37
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION
To test our hypothesis that bacterial ribosomes can template the Huisgen reaction, we
synthesized azide 2 using known methods.38 Escherichia coli 70S ribosomes, 50S and 30S
ribosomal subunits were isolated as described.39 After optimizing the concentrations of ribosomes,
azide 2, and commercial 3-ethynylaniline (3) in tris(hydroxymethyl)-aminomethane (Tris) buffer,
we found that 5 µM 70S ribosomes or 50S subunits, 5 µM azide, and 5 mM alkyne at rt for 24−48 h
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 27
1
2
3
4
resulted in the formation of 1 and its syn (1,5)-regioisomer (~2:1 ratio) in 12 ± 4-fold greater
5
6
7
8
amounts than in the absence of 70S ribosome or 50S subunit (Figure 2). Analysis was performed on
an Agilent 6520B Q-TOF LC-MS instrument wherein extracted ion chromatograms were used to
locate and quantify the masses of interest (normalized to highest value).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
100
'!!"
&!"
%!"
$!"
#!"
!"
96
45
44
12
10
6
11
6
8
anti-1
(SOL)
syn-1
6
70S
7
50S
70S +
AZY
30S
BSA
buffer
Figure 2. In situ click experiments with E. coli 70S ribosomes, 50S ribosomal subunits, 70S
ribosomes with inhibitor azithromycin (AZY, 25 µM) and negative controls (30S ribosomal
subunits, BSA, or buffer only). Mass counts (normalized) correspond to the combined anti-1
(solithromycin, SOL) and syn-1 regioisomer ions.
Retention times of both anti (1,4)- and syn (1,5)-regioisomers were confirmed
independently by chemical synthesis via thermal cycloaddition; moreover, solithromycin (1) was
exclusively prepared by Cu(I)-catalysis.40 Several lines of evidence strongly support the
involvement of the large ribosomal subunit in the in situ click reaction: (1) in the absence of 70S
ribosomes or 50S subunits (i.e., only buffer), there was 12 ± 4-fold less product formation, with the
mass counts found corresponding to the thermal cycloaddition background reaction; (2) the 30S
subunits, which do not possess a macrolide-binding site, also displayed mass counts similar to
6
ACS Paragon Plus Environment

Page 7 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
background; (3) the presence of ribosomal inhibitor azithromycin (AZY, 25 µM), which competes
for the binding site with azide 2, blocks 70S ribosome-dependent product formation; (4) replacing
ribosomes with bovine serum albumin (BSA), a standard negative control used to rule out non-
specific binding, resulted in mass counts similar to those of the background cycloaddition; and
finally, (5) the regioisomer ratio was ~1:1 in all control reactions (i.e., 30S, BSA, and buffer alone)
and in the inhibition experiment with AZY, whereas in the presence of 70S ribosomes or 50S
ribosomal subunits, the product ratio was 2:1 favoring 1. Such selectivity is a hallmark of the
orientational (i.e., regioselective) nature of target-guided in situ click chemistry.29
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Having established the utility of in situ click chemistry in binary experiments (i.e., one
azide, one alkyne) for the synthesis of solithromycin (1), we selected a small library of structurally
diverse alkynes for competition experiments (Table 1A). The library of fifteen alkynes (Table 1A)
contained both aromatic (e.g., 3, 5−14) and non-aromatic (e.g., 15−18) functionalities, including 3-
ethynylaniline (3) used in the synthesis of solithromycin (1). Aromatic alkynes were selected based
on the potential to engage in π-stacking interactions with the 23S rRNA A752-U2609 Watson-Crick
base-pair, and to allow assessment of the impact of a hydrogen bonding network established
between the aniline in 1 and A752 (PDB 3ORB).13 The non-aromatic alkynes included structural
motifs that could bind rRNA via hydrogen bond donors (e.g., 15−17), acceptors (e.g., 15−18), or by
forming electrostatic interactions (i.e., salt bridges) between the protonated amine in 32, derived
from morpholine 18, and negatively-charged phosphates. As shown in Figure 2, triazoles from the
in situ click reaction with azide 2 and the alkynes (Table 1A) can yield anti (1,4)- and/or syn (1,5)-
regioisomers, depending on the positioning of alkyne fragments that make optimal interactions with
rRNA (Table 1B, represented as ‘R’).
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 27
1
2
3
4
5
6
7
8
9
Experiments were then undertaken to determine the ribosome binding affinity of azide 2 and
the anti (1,4)-triazoles (Table 1B). To this end, anti-triazoles 1, 19−32 were prepared by Cu(I)-
catalysis as template-guided synthesis only provides analytically detectable quantities.40 Products
derived from the in situ click method are anticipated to possess greater target affinity, compared to
the individual fragments, due to the additivity of binding energies (Scheme 1),14 such that triazoles
formed in the greatest amounts (i.e., highest mass counts) should possess higher affinity. To
quantify binding affinity, dissociation constants (Kd) of the anti-regioisomers of triazoles 1, 19−32
and azide 2 for 70S E. coli ribosomes were measured by an established fluorescence polarization
competition assay using BODIPY-functionalized erythromycin.11 The results showed an eight-fold
range of anti-triazole affinities, wherein 1, 19, 21−25, 27−28, and 32 bound more strongly than
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
azide fragment 2 while anti-triazoles 20, 26, 29−31 were weaker binders than 2 (Table 2, Figure
(A)
NC
R
H2N
3: R = CCH
R
HO
5: R = CCH
R
CF3
6: R = CCH
R
HF2C
8: R = CCH
R
7: R = CCH
1: R = Triazole
19: R = Triazole
20: R = Triazole
21: R = Triazole
22: R = Triazole
Me2N
MeO
F
S
N
R
R
R
R
R
9: R = CCH
23: R = Triazole
10: R = CCH
24: R = Triazole
11: R = CCH
25: R = Triazole
12: R = CCH
26: R = Triazole
13: R = CCH
27: R = Triazole
O
O
R
HO
Et Et
R
N
N
R
HO
OH
O
HO
N
R
HO
R
OH
14: R = CCH
28: R = Triazole
15: R = CCH
29: R = Triazole
16: R = CCH
30: R = Triazole
17: R = CCH
31: R = Triazole
18: R = CCH
32: R = Triazole
R
4
(B)
N
N
N
5
N
R
N
1
N
1
O
O
( )4
( )4
Me
Me
Me
Me
O
O
N
N
Me
OMe
Me
OMe
O
O
Me
Et
Me
Et
Me
Me
O
ODes
O
ODes
O
O
O
O
F
Me
F
Me
anti (1,4)
syn (1,5)
isomer
isomer
S1).
8
ACS Paragon Plus Environment

Page 9 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
Table 1. Structures of (A) alkyne fragments in the library and (B) regioisomeric anti (1,4)- and syn
(1,5)-triazoles derived from in situ click experiments (R = Fragment).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In order to rationalize the relatively narrow range of measured Kd values despite significant
differences in the structures of the alkynes, we performed a computational analysis to understand
the relative alkyne contributions to the binding affinities using the Site-Identification by Ligand
Competitive Saturation (SILCS) approach.41-44 SILCS maps the free energy affinity pattern of
macromolecules onto a grid and may be used to quantitatively estimate relative binding affinities of
ligands, yielding ligand grid free energies (LGFE). Details of the SILCS calculation and LGFE
analysis are presented in the supporting information. Notably, the LGFE scores represent an atom-
based free energy approximation allowing estimation of the contributions of different regions of
molecules to binding affinity. Presented in Table 2 along with the experimentally-determined Kd
and ΔG values (ΔG=−RTlnKd) are the calculated LGFE scores for anti-triazoles 1, 19−32, and azide
2, including (1) the total LGFE score of each compound; (2) the LGFE contributions of the
respective side-chains; and, (3) the LGFE contributions of macrolactone and desosamine
(Macro+Des) components. Analysis of the ability of the three LGFE metrics to predict the relative
order of binding was then undertaken by calculating the predictive indices (PIs), which measures
how well molecular modeling calculations track the ordering of the experimental binding values
(see supporting information for details). The index varies from −1 (wrong prediction) to 0 (random)
to +1 (perfect prediction).45
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 27
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Side-
Total
Macro+Des
LGFE
chain
ΔG
(kcal/mol)
LGFE
(kcal/
mol)
Cmpd
Kd (nM)
LGFE
(kcal/
mol)
(kcal/
mol)
SOL (1)
32
0.6 ± 0.1
0.8 ± 0.1
1.1 ± 0.1
1.1 ± 0.1
1.3 ± 0.2
1.4 ± 0.2
1.5 ± 0.1
1.6 ± 0.2
1.7 ± 0.2
1.8 ± 0.2
2.1 ± 0.4
2.4 ± 0.2
2.5 ± 0.3
2.5 ± 0.2
3.5 ± 0.5
5.0 ± 0.4
N/A
−12.54
−12.41
−12.24
−12.22
−12.13
−12.06
−12.02
−12.00
−11.94
−11.91
−11.82
−11.74
−11.73
−11.72
−11.53
−11.32
N/A
−49.67
−51.83
−48.72
−48.70
−52.16
−50.83
−47.80
−50.82
−52.61
−47.93
−38.85
−49.49
−48.96
−49.18
−52.86
−52.90
−0.22
−16.20
−17.09
−14.01
−14.87
−17.19
−15.50
−14.27
−16.23
−17.81
−13.64
−5.04
−5.15
−3.91
−15.39
−18.55
−18.10
−0.14
−35.38
−35.30
−35.33
−35.57
−34.27
−34.86
−35.83
−35.24
−35.08
−34.70
−35.85
−34.80
−35.37
−34.93
−34.91
−34.81
0.37
19
25
23
24
28
21
22
27
Azide 2
31
30
26
20
29
PI
Table 2. Rank-ordering of anti-triazoles 1, 19−32 and azide fragment 2 by dissociation constants
(Kd) for 70S E. coli ribosomes determined by fluorescence polarization, along with experimental
ΔG values (kcal/mol) and calculated normalized Ligand Grid Free Energies (LGFEs, kcal/mol)
from SILCS. LGFEs are calculated for the total molecule, side-chain, and
macrolactone+desosamine (Macro+Des) components. The side-chain is defined as the four-carbon
alkyl linker and functionalized triazole extending from N11. Predictive indices (PIs) are calculated
for each type of LGFE.45
10
ACS Paragon Plus Environment

Page 11 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
As shown in Table 2, the total and side-chain LGFE scores were not predictive whereas the
Macro+Des LGFE yielded a satisfactory level of predictability. While somewhat unexpected, these
results suggest that the binding of the ligands is dominated by the macrolactone and desosamine
moieties, which are common to all of the triazole compounds. This hypothesis is consistent with the
similarities in Kd values, with the relative binding affinities being associated with the ability of the
unique side-chains to alter the interactions of the macrocycle and desosamine moieties with rRNA,
rather than the side-chains directly interacting with rRNA themselves. Further support is found in
the relative binding of known ketolides in which the addition of a side-chain did not markedly
increase efficacy. For example, clarithromycin—the precursor to telithromycin (4) and
solithromycin (1)—has a Kd value of 1.7 nM despite the absence of a side-chain (see supporting
information).11,13 The significance and utility of the side-chains in congeners 1 and 4 is
demonstrated by bacterial ribosomes that acquire resistance from either mutation or modification,
which render first-generation antibiotics erythromycin and clarithromycin ineffective due to
markedly decreased binding affinity.13,46,47
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Detailed analysis revealed several important structure-activity relationships within the
library. Specifically, meta- or 3-substituted aromatic and/or heteroaromatic groups with the ability
to engage in hydrogen bonding provided the best boost in affinity relative to the azide alone (e.g., 1,
19, 21−22, 25). In contrast, the 3-substituted trifloromethylphenyl, and 2-fluorophenyl triazoles, 20
and 26, respectively, with no capacity for hydrogen bonding, failed to enhance affinity. In addition,
the nonaromatic triazoles 29, 30, and 31 all showed decreased binding as compared to 2, indicating
that moieties that participate primarily in hydrogen bond interactions but cannot participate in π-
stacking do not stabilize macrocycle+desosamine−rRNA interactions. These results suggest that the
ability of the side-chain to participate in both π-stacking and hydrogen bonding leads to
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 27
1
2
3
4
5
6
7
8
9
stabilization of macrolactone+desosamine−rRNA interactions. We attribute the relatively high
binding activity of the nonaromatic morpholine-containing triazole 32, which bound only slightly
less tightly than solithromycin (1, SOL), to the presence of a basic amine that can interact
electrostatically with rRNA. Lastly, the five-membered heteroaromatics 27−28 showed increased
binding and thus represent an interesting, novel class to explore.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Guided by the Kd values, two in situ click experiments were designed wherein azide 2 was
incubated with five different alkynes in the presence of 50S E. coli ribosomal subunits to test
whether the target could differentiate between triazoles with Kd values lower than azide 2 and those
with higher Kd values. The first experiment included 3-ethynylaniline (3), which is the precursor to
solithromycin (1), along with 5, 10, 15, and 16 (2 mM each; 10 mM total), 10 µM azide 2, and 10
µM 50S E. coli ribosomal subunits at rt for 48 h. Azide and ribosomal subunit concentrations were
doubled relative to the binary experiment to ensure sufficient product formation under competitive
reaction conditions. The results (Figure 3) show that 1 provides the greatest combined mass counts,
with the anti-regioisomer (solithromycin, 1) being preferred over syn-1. Phenol-functionalized
triazole 19, which possessed a low Kd for the anti-regioisomer, was also formed in significant
amounts. This fact establishes the importance of aromatic fragments with the capacity for hydrogen
bonding with rRNA at the meta-position, again drawing an analogy to 1. Triazole formation from
glucosyl alkyne 15 resulted in small amounts of both syn- and anti-29. Aliphatic compound 30 was
not formed in significant amounts, which we attribute to the absence of π-stacking interactions.
Interestingly, triazole 24 was formed in the lowest amount, even though it is capable of π-stacking
and has a Kd lower than that of azide 2. We posit this phenomenon is most likely due to competitive
product inhibition arising from 1 and 19, which are two of the tightest binders in the library.14,21
12
ACS Paragon Plus Environment

Page 13 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
HO
R
H2N
R
19
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
O
R
79
77
HO
80
70
60
50
40
30
20
10
0
HO
OH
MeO
Et Et
62
OH
HO
R
29
R
30
24
6
-1
16
Anti
4
0
Syn
5
Mix
1
19
29
30
24
Figure 3. In situ click experiment with azide 2 and alkynes 3, 5, 10, 15, and 16, yielding triazoles 1,
19, 24, 29, and 30, respectively. Mix represents unresolved anti- and syn-isomers. Normalized mass
count percent increases (provided in the bars) are calculated from the ratio of the ribosome-
templated reaction to the background reaction. Results are an average of two experiments. The
remainder of the molecule is abbreviated as ‘R’.
The second, five-alkyne in situ click experiment featured alkynes bearing a range of
functional groups such as alcohol 16, imidazole 14, pyridine 11, nitrile 7, and fluoride 12, selected
to determine how the ribosome-templated reaction would perform in the presence of alkynes that
yield triazoles binding more weakly than 1. The results from the experiment are shown in Figure 4.
Imidazole-functionalized triazole 28, as a mixture of syn- and anti-regioisomers, was detected in the
greatest amount, consistent with the low Kd value for the anti-isomer, followed by 25 then 21. In
contrast, triazoles 26 and 30 were not detected in significant quantities. Taken together, the two
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 27
1
2
3
4
five-alkyne in situ click experiments demonstrate that the ribosome can template the formation of
5
6
tighter-binding triazoles in greater quantity.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
N
R
N
28
N
100
90
80
70
60
50
40
30
20
10
0
R
96
25
F
NC
R
Et Et
HO
21
R
28
26
R
30
19
Anti
20
Syn
9
3
Mix
28
25
21
26
30
Figure 4. In situ click experiment with azide 2 and alkynes 7, 11−12, 14, and 16 yielding triazoles
21, 25−26, 28, and 30, respectively. Mix represents unresolved anti- and syn-isomers. Normalized
mass count percent increases (provided in the bars) are calculated from the ratio of the ribosome-
templated reaction to the background reaction. Results are an average of two experiments. The
remainder of the molecule is abbreviated as ‘R’.
The successful execution of five-alkyne in situ click experiments justified a greater
exploration of chemical space while expanding the scope of the method. To this end, we initiated
experiments with fifteen alkynes, which would yield thirty congeners (Figure 5). These included
alkynes from the two five-alkyne experiments, along with six additional alkynes (i.e., 6, 8−9, 13,
17−18). To ensure complete alkyne solubilization, the concentration of each member was decreased
from 2 mM to 1 mM, and the concentrations of azide 2 and 70S E. coli ribosomes were maintained
at 10 µM each. The fifteen-component alkyne mixture (15 mM total) was sonicated for 1−5 min to
obtain a homogenous solution, prior to the addition of azide 2 and ribosomes, and the reaction
14
ACS Paragon Plus Environment

Page 15 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
mixture was incubated at rt for 48 h. Consistent with the five-alkyne in situ click reactions, we
detected the formation of triazoles with Kd values lower than 2 (i.e., better binders than the azide
fragment) including solithromycin (1), 19, 21−25, 27, and 28 (Figure 5). All of these cycloadducts
were derived from aromatic alkynes, again underscoring the significance of π-stacking interactions
with the A752-U2609 base pair. The only aromatic triazole that was not detected in appreciable
quantity was trifluoromethyl congener 20. However, its Kd value (Table 2) was the second highest
of the library, further illustrating selectivity in the in situ click process. Non-aromatic triazoles 29,
30, and 31 were not detected in significant quantities. In addition, morpholine-functionalized 32
was not detected in significant quantities, despite the fact that it binds ribosomes as well as 1. We
attribute this observation to the basicity of N-propargyl morpholine (18), which, though its pKa is
5.55,48 could be protonated when bound to the ribosome due to electrostatic interactions with
phosphate residues. Such binding could effectively sequester this fragment and preclude coupling
with 2. Competitive product inhibition, which was observed in both five-alkyne competition
experiments (vide supra), may account for the modest formation of triazoles 20, 29−32.
Curiously, the ribosome-templated synthesis of solithromycin (1) gave slightly different
syn/anti ratios for the binary reaction (~2:1) versus the five- and fifteen-alkyne competition
experiments, the latter two of which gave similar syn/anti ratios (~1.24/1). We also observed this
reaction-dependent change in regioisomer ratios with phenol 19 and pyridine 25. The isolation of
nitrile 21 as a ‘mix’ in Figure 4 and subsequent resolution thereof [i.e., ~5:1 (anti/syn)] in Figure 5
is attributed to chromatographic issues and not the reaction per se.49 The modulation of regioisomer
ratios illustrates the complex nature of in situ click competition experiments wherein the alkyne
mixtures, which are in mM concentrations, are likely modifying the architecture of the macrolide-
binding site via direct or allosteric interactions with rRNA. Finally, the marked and reproducible
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 27
1
2
3
4
5
6
7
8
9
ribosome-templated formation of nitrile 21 in the fifteen-alkyne (Figure 5) vis-à-vis the five-alkyne
experiment (Figure 4) is striking. This result is difficult to rationalize in terms of Kd or LGFE and
may arise from the complexity of the reaction mixture. To probe this, we are currently investigating
ten-alkyne mixtures that are more consistent with the prior art.20-25
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
R =
N
NC
R
N
N
O
21
MeO
( )4
Me
Me
O
N
H2
N
R
Me
OMe
R
O
869
1
S
24
Me
Et
900
800
700
600
500
400
300
200
100
0
Me2
N
N
Me
R
HO
R
O
ODes
R
25
19
F
R
27
O
O
N
R
23
F
Me
CF3
R
N
R
20
28
26
HF2
C
R
478
O
O
R
418
HO
22
HO
R
374
31
Et Et
445
HO
OH
396
338
OH
267
HO
R
29
30
271
265
157
158
168
-15
0
0
-25
0
0
100
0
0
0
anti
syn
Mix
1
52
0
0
0
114
0
0
0
0
-59
-54
73
41
0
0
0
0
0
0
0
0
1
19
20
21
22
23
24
25
26
27
28
29
30
31
32
Figure 5. In situ click experiment with azide 2 and alkynes 3, 5−18 yielding triazoles 1
and 19−32, respectively. Mix represents an unresolved mixture of anti- and syn-isomers. Mass
count percent increases (provided in the bars) are calculated from the ratio of the ribosome-
templated reaction to the background reaction. Results are an average of five experiments.
We next assessed the mechanism of action of azide 2 and anti-triazoles 1, 19−32 and
evaluated their antibiotic activities using (1) in vitro protein synthesis assays using a cell-free
system50 and (2) minimum inhibitory concentration (MIC) assays for azide 2, anti-triazoles 1,
and 19−32 (Table 3).51 As the ribosome-templated in situ click process delivers bivalent
inhibitors possessing greater potency than their monovalent components, it is important to
16
ACS Paragon Plus Environment

Page 17 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
determine the selectivity of the newly formed cycloadducts for bacterial versus mammalian
ribosomes. To this end, we evaluated anti-triazoles 1 (solithromycin), azide 2, and four analogs
from our library using a mammalian cell toxicity assay with human fibroblasts.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For the in vitro translation inhibition studies, all of the compounds were assayed at 1 µM.
Given that the 70S concentration in the cell-free protein synthesis (CFPS) reactions are 1.5 ± 0.2
µM (RTS100 kit, 5PRIME)50 and 1.4 ± 0.1 µM (Expressway Mini kit, Invitrogen, see supporting
information), we would expect these low- to sub- nM affinity compounds to bind 70S
stoichiometrically, negating any differences in affinity and yielding an expected inhibition of
approximately 70 ± 10%. Indeed, all of the compounds, including the azide, inhibited the CFPS
reaction in the range of 48 ± 16% (see supporting information). This inhibition is toward the
lower end of the predicted range, which may be due to inhibitor sequestration by other
components in these heterogeneous lysate-based mixtures, reducing the effective inhibitor
concentration. Differential sequestration between compounds could also explain the variation
observed between the best inhibitor (32, 64 ± 14% inhibition) and the worst inhibitor (29, 32 ±
5% inhibition).
For the MIC assays, we tested solithromycin (1, SOL), azide 2, and anti-triazoles 19−32
against various strains of E. coli, S. pneumoniae, and S. aureus.52 Strains ATCC 29213 (S.
aureus) and ATCC 49619 (S. pneumoniae) served as quality control strains with values for SOL
(1), giving results closely matching published values.51 The results in Table 3 show that
thiophene-functionalized triazole 27 was two-fold more potent than SOL against E. coli DK
pkk3535 strain and A2058G strains and that the relatively high affinity (low Kd) phenol-
functionalized triazole 19 was two-fold more potent than SOL in the S. pneumoniae ATCC wild-
type and E. coli mutant DK A2058G strains.53 Comprehensive structure-activity relationship
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 27
1
2
3
4
5
6
7
8
9
studies of these two analogs could result in the discovery of novel, potent antibiotics. In addition,
three of the poorest-performing compounds against both S. pneumoniae strains (20, 29, 30 shown
in red) include two, 20 and 29, having the highest Kd values. Consistent with this are the
Predictive indices (PIs) of the Kd values in Table 3 for the MIC values, indicating affinity to be a
reasonable predictor of functional activity.45 However, adduct 32 (also shown in red) is less
potent than would be expected from its Kd value, which could indicate an uptake problem. Taken
together, these results indicate satisfactory levels of selectivity in the ribosome-templated in situ
click process. It is important to note that while solithromycin (1) and the library of analogs
prepared herein maintained their efficacy against resistant E. coli and S. pneumoniae strains
(Table 3), this was not the case for all resistant strains tested (see Tables S5 and S6, supporting
information).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
MIC
E. coli
DK
pKK
3535
MIC
S.
pneumo pneumo
ATCC
49619
MIC
S.
MIC
E. coli
DK
ΔG
(kcal/mol)
Cmpd
Kd (nM)
655
mefA
A2058G
27
19
28
1.8
1.1
1.5
0.6
2.1
1.4
2.5
1.1
1.7
1.3
1.6
2.4
2.5
3.5
0.8
5
-11.9
-12.2
-12.0
-12.6
-11.8
-12.1
-11.7
-12.2
-12.0
-12.1
-12.0
-11.7
-11.7
-11.5
-12.4
-11.3
N/A
1
2
2
2
2
2
2
2
2
4
4
4
4
4
8
32
0.46
1
1
2
2
2
2
2
2
4
4
4
4
4
4
8
32
0.48
0.004
0.002
0.032
0.006
0.002
0.005
0.002
0.016
0.004
0.008
0.016
0.016
0.032
0.016
0.063
4
0.5
0.5
4
0.375
0.25
1
0.5
0.5
1
1
1
2
2
SOL (1)
Azide 2
24
26
25
22
23
21
31
30
20
2
4
4
0.44
32
29
PI
N/A
0.34
18
ACS Paragon Plus Environment

Page 19 of 27
Journal of the American Chemical Society
1
2
3
4
Table 3. Evaluation of azide 2, anti-triazoles 1 (SOL) and 19−32 using minimum inhibitory
5
6
7
8
9
concentration (MIC) assays (µg/mL) against E. coli and S. pneumoniae strains. Compounds are
rank-ordered by potency in MIC assays against E. coli then S. pneumoniae strains. MIC values
determined in three independent experiments; translation values in two independent experiments.
Analysis of the data in Table 3 reveals that the poorest-performing compounds (32, 20, and 29,
shown in red) against both strains correlate with the binding data; in fact, 20 and 29 had the highest
Kd values. The polarity of 29 and 32 may be contributing to poor uptake and/or permeability.
Predictive indices (PIs) are calculated for the MICs against each strain with respect to Kd values.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For the cell viability assays, the potential cytotoxicity of solithromycin (1), azide (2), and
analogs 19, 24, 27, and 28 against human dermal fibroblasts (GM05659, Coriell Institute,
Camden, NJ)54 was measured using a commercial luciferase-coupled ATP quantitation assay
(CellTiter-Glo®, Promega).55 The cells were incubated with test compounds at concentrations
ranging from 50 µM to 0.88 nM for 24- and 48-hour time periods (see Figures S3−S5,
supporting information). Significantly, the data showed that, like 1, compounds 24, and 27−28
showed no effect on fibroblasts after 24 or 48 hours up to low micromolar concentrations. The
therapeutic indices for these compounds (i.e., bactericidal activity versus mammalian
cytotoxicity) were essentially the same as that measured for solithromycin (1).
CONCLUSION
We have developed an in situ click chemistry method that employs 70S E. coli ribosomes
and 50S ribosomal subunits as platforms, with the ribosome-templated synthesis of
solithromycin (1) serving as proof-of-concept. The method was applied in five- and fifteen-
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 27
1
2
3
4
5
6
7
8
9
alkyne competition experiments. Consistent with other kinetic, target-guided in situ click
processes, the extent of triazole formation correlated with ribosome binding affinity (see Chart
S1, supporting information). The 50S E. coli ribosomal subunit was also studied using the
computational Site-Identification by Ligand Competitive Saturation (SILCS) approach.
Interestingly, LGFEs associated with the macrolactone and desosamine moieties, rather than the
full triazole structures, were correlated to dissociation constants for the congeners, suggesting
that the side-chain indirectly impacts affinity by altering macrocycle−ribosome interaction.
The inclusion of bacterial ribosomes in the repertoire of targets represents a powerful
drug discovery platform that obviates the onerous need to independently synthesize, characterize,
and evaluate both syn- and anti-triazoles. Significantly, the use of ribosomes possessing known
mechanisms of resistance (e.g., rRNA modification or mutation) can lead to the discovery of
antibiotics that selectively target resistant over wild-type bacterial strains. Protein synthesis
inhibition experiments confirmed the mechanism of action of these congeners. MIC evaluation
of the in situ click products quantified antibiotic activity and firmly established this method as
efficacious in the triaging and prioritization of potent antibiotic candidates targeting the bacterial
ribosome. Finally, we showed that four analogs discovered using ribosome-templated in situ
click chemistry (i.e., 19, 24, 27, 28) displayed similar therapeutic indices as that seen with
solithromycin (1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ASSOCIATED CONTENT
Supporting Information. General and in situ click methods and data, computational methods,
determination of Kd by fluorescence polarization with BODIPY-labeled erythromycin, cell-free
translation inhibition method and data, minimum inhibitory concentration (MIC) method and
20
ACS Paragon Plus Environment

Page 21 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
data, mammalian cell toxicity assay, synthetic methods including synthesis and full structural
assignment of azide 2, Cu(I)-catalyzed click synthesis and characterization of anti-triazoles 1,
19−32. This material is available free of charge via the Internet at http://pubs.acs.org.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
AUTHOR INFORMATION
Corresponding Author
randrade@temple.edu
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was supported by the NIH (AI080968 and GM070855), the University of Maryland
Computer-Aided Drug Design Center, and Temple University. We thank Prof. Paul Edelstein
(Penn Medicine) for kindly providing us with wild-type and resistant S. pneumoniae strains. We
thank Dr. Charles DeBrosse, Director of the NMR Facility at Temple Chemistry, for kind
assistance with NMR experiments. Finally, we thank Dr. Charles W. Ross III for assistance with
LC/MS experiments and for carefully reading this manuscript.
REFERENCES
(1) Walsh, C. Nat. Rev. Microbiol. 2003, 1, 65.
(2) Levy, S. B.; Marshall, B. Nature Medicine 2004, 10, S122.
(3) Solomon, S. L.; Oliver, K. B. Am. Fam. Physician 2014, 89, 938.
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 27
1
2
3
4
5
6
7
8
9
(4) Bush, K.; Courvalin, P.; Dantas, G.; Davies, J.; Eisenstein, B.; Huovinen, P.; Jacoby, G. A.;
Kishony, R.; Kreiswirth, B. N.; Kutter, E.; Lerner, S. A.; Levy, S.; Lewis, K.; Lomovskaya, O.;
Miller, J. H.; Mobashery, S.; Piddock, L. J. V.; Projan, S.; Thomas, C. M.; Tomasz, A.; Tulkens,
P. M.; Walsh, T. R.; Watson, J. D.; Witkowski, J.; Witte, W.; Wright, G.; Yeh, P.; Zgurskaya, H.
I. Nat. Rev. Microbiol. 2011, 9, 894.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(5) Fox, J. L. Nat. Biotechnol. 2006, 24, 1521.
(6) Wright, P. M.; Seiple, I. B.; Myers, A. G. Angew. Chem. Int. Ed. 2014, 53, 8840.
(7) Humphrey, W.; Dalke, A.; Schulten, K. Journal of Molecular Graphics & Modelling 1996,
14, 33.
(8) Fernandes, P.; Pereira, D.; Jamieson, B.; Keedy, K. Drug Future 2011, 36, 751.
(9) Bryskier, A. Clin. Microbiol. Infect. 2000, 6, 661.
(10) Tenson, T.; Mankin, A. Mol. Microbiol. 2006, 59, 1664.
(11) Yan, K.; Hunt, E.; Berge, J.; May, E.; Copeland, R. A.; Gontarek, R. R. Antimicrob. Agents
Chemother. 2005, 49, 3367.
(12) Spahn, C. M. T.; Prescott, C. D. J. Mol. Med. 1996, 74, 423.
(13) Llano-Sotelo, B.; Dunkle, J.; Klepacki, D.; Zhang, W.; Fernandes, P.; Cate, J. H. D.;
Mankin, A. S. Antimicrob. Agents Chemother. 2010, 54, 4961.
(14) Jencks, W. P. Proc. Natl. Acad. Sci. U.S.A. 1981, 78, 4046.
(15) Mamidyala, S. K.; Finn, M. G. Chem. Soc. Rev. 2010, 39, 1252.
(16) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. Angew. Chem. Int. Ed. 2002,
41, 2596.
(17) Boren, B. C.; Narayan, S.; Rasmussen, L. K.; Zhang, L.; Zhao, H. T.; Lin, Z. Y.; Jia, G. C.;
Fokin, V. V. J. Am. Chem. Soc. 2008, 130, 8923.
22
ACS Paragon Plus Environment

Page 23 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
(18) Rees, D. C.; Congreve, M.; Murray, C. W.; Carr, R. Nature Reviews Drug Discovery 2004,
3, 660.
7
8
9
(19) Scott, D. E.; Coyne, A. G.; Hudson, S. A.; Abell, C. Biochemistry 2012, 51, 4990.
(20) Manetsch, R.; Krasinski, A.; Radic, Z.; Raushel, J.; Taylor, P.; Sharpless, K. B.; Kolb, H. C.
J. Am. Chem. Soc. 2004, 126, 12809.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(21) Lewis, W. G.; Green, L. G.; Grynszpan, F.; Radic, Z.; Carlier, P. R.; Taylor, P.; Finn, M. G.;
Sharpless, K. B. Angew. Chem. Int. Ed. 2002, 41, 1053.
(22) Krasiński, A.; Radić, Z.; Manetsch, R.; Raushel, J.; Taylor, P.; Sharpless, K. B.; Kolb, H. C.
J. Am. Chem. Soc. 2005, 127, 6686.
(23) Grimster, N. P.; Stump, B.; Fotsing, J. R.; Weide, T.; Talley, T. T.; Yamauchi, J. G.;
Nemecz, A.; Kim, C.; Ho, K. Y.; Sharpless, K. B.; Taylor, P.; Fokin, V. V. J. Am. Chem. Soc.
2012, 134, 6732.
(24) Mocharla, V. P.; Colasson, B.; Lee, L. V.; Roper, S.; Sharpless, K. B.; Wong, C. H.; Kolb,
H. C. Angew. Chem. Int. Ed. 2004, 44, 116.
(25) Whiting, M.; Muldoon, J.; Lin, Y. C.; Silverman, S. M.; Lindstrom, W.; Olson, A. J.; Kolb,
H. C.; Finn, M. G.; Sharpless, K. B.; Elder, J. H.; Fokin, V. V. Angew. Chem. Int. Ed. 2006, 45,
1435.
(26) Hirose, T.; Sunazuka, T.; Sugawara, A.; Endo, A.; Iguchi, K.; Yamamoto, T.; Ui, H.;
Shiomi, K.; Watanabe, T.; Sharpless, K. B.; Omura, S. J. Antibiot. 2009, 62, 277.
(27) Namelikonda, N. K.; Manetsch, R. Chem. Commun. 2012, 48, 1526.
(28) Poulin-Kerstien, A. T.; Dervan, P. B. J. Am. Chem. Soc. 2003, 125, 15811.
(29) Sharpless, K. B.; Manetsch, R. Expert Opinion on Drug Discovery 2006, 1, 525.
23
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 27
1
2
3
4
5
6
(30) Willand, N.; Desroses, M.; Toto, P.; Dirié, B.; Lens, Z.; Villeret, V.; Rucktooa, P.; Locht,
C.; Baulard, A.; Deprez, B. ACS Chemical Biology 2010, 5, 1007.
7
8
9
(31) Rzuczek, S. G.; Park, H.; Disney, M. D. Angew. Chem. Int. Ed. 2014, 53, 10956.
(32) Millward, S. W.; Henning, R. K.; Kwong, G. A.; Pitram, S.; Agnew, H. D.; Deyle, K. M.;
Nag, A.; Hein, J.; Lee, S. S.; Lim, J.; Pfeilsticker, J. A.; Sharpless, K. B.; Heath, J. R. J. Am.
Chem. Soc. 2011, 133, 18280.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(33) Agnew, H. D.; Rohde, R. D.; Millward, S. W.; Nag, A.; Yeo, W. S.; Hein, J. E.; Pitram, S.
M.; Tariq, A. A.; Burns, V. M.; Krom, R. J.; Fokin, V. V.; Sharpless, K. B.; Heath, J. R. Angew.
Chem. Int. Ed. 2009, 48, 4944.
(34) Pfeilsticker, J. A.; Umeda, A.; Farrow, B.; Hsueh, C. L.; Deyle, K. M.; Kim, J. T.; Lai, B.
T.; Heath, J. R. Plos One 2013, 8.
(35) Nag, A.; Das, S.; Yu, M. B.; Deyle, K. M.; Millward, S. W.; Heath, J. R. Angew. Chem. Int.
Ed. 2013, 52, 13975.
(36) Farrow, B.; Hong, S. A.; Romero, E. C.; Lai, B.; Coppock, M. B.; Deyle, K. M.; Finch, A.
S.; Stratis-Cullum, D. N.; Agnew, H. D.; Yang, S.; Heath, J. R. Acs Nano 2013, 7, 9452.
(37) Deyle, K. M.; Farrow, B.; Hee, Y. Q.; Work, J.; Wong, M.; Lai, B.; Umeda, A.; Millward,
S. W.; Nag, A.; Das, S.; Heath, J. R. Nature Chemistry 2015, 7, 455.
(38) Liang, C. H.; Yao, S. L.; Chiu, Y. H.; Leung, P. Y.; Robert, N.; Seddon, J.; Sears, P.;
Hwang, C. K.; Ichikawa, Y.; Romero, A. Biorg. Med. Chem. Lett. 2005, 15, 1307.
(39) Grigoriadou, C.; Marzi, S.; Kirillov, S.; Gualerzi, C. O.; Cooperman, B. S. J. Mol. Biol.
2007, 373, 562.
(40) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem. Int. Ed. 2001, 40, 2004.
24
ACS Paragon Plus Environment

Page 25 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
(41) Raman, E. P.; Yu, W.; Lakkaraju, S. K.; MacKerell, A. D., Jr. J. Chem. Inf. Model. 2013,
53, 3384.
7
8
9
(42) Raman, E. P.; Yu, W.; Guvench, O.; Mackerell, A. D. J. Chem. Inf. Model. 2011, 51, 877.
(43) Guvench, O.; MacKerell Jr, A. D. PLoS Comput. Biol. 2009, 5, e1000435.
(44) Lakkaraju, S. K.; Yu, W.; Raman, E. P.; Hershfeld, A. V.; Fang, L.; Deshpande, D. A.;
MacKerell, A. D., Jr. J. Chem. Inf. Model. 2015, 55, 700.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(45) Pearlman, D. A.; Charifson, P. S. J. Med. Chem. 2001, 44, 3417.
(46) Mankin, A. S. Curr. Opin. Microbiol. 2008, 11, 414.
(47) Dunkle, J. A.; Xiong, L.; Mankin, A. S.; Cate, J. H. D. Proc. Natl. Acad. Sci. U.S.A. 2010,
107, 17152.
(48) Williams, A.; Jencks, W. P. J. Chem. Soc.-Perkin Trans. 2 1974, 1760.
(49) The change in chromatographic separation of syn/anti regioisomers is a result of the reverse-
phase HPLC columns, whose resolution deteriorate with multiple passage of bacterial ribosomes
or ribosomal subunits. We are working to ameliorate this issue.
(50) Rosenblum, G.; Chen, C.; Kaur, J.; Cui, X.; Goldman, Y. E.; Cooperman, B. S. Nucleic
Acids Res. 2012, 40, e88.
(51) Reller, L. B.; Weinstein, M.; Jorgensen, J. H.; Ferraro, M. J. Clin. Infect. Dis. 2009, 49,
1749.
(52) See supporting information for MIC evaluation against a broader panel of wild-type and
resistant strains of E. coli, S. pneumoniae, and S. aureus.
(53) DK/pKK3535 refer to a tolC(−) strain with wild-type (WT) ribosomes. DK/A2058G refer
to a tolC(−) strain with a mixed population of WT and mutant ribosomes (ca. 1:1) wherein
Adenine has mutated to Guanine, which represents a clinically signifcant mechanism of
25
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 27
1
2
3
4
5
6
7
8
9
resistance. TolC is an outermembrane efflux pump that recognizes antibiotics (e.g., macrolides),
thus making these tolC(−) E. coli DK strains useful (and relevant) models of pathogenic gram
positive bacteria. Moreover, the ribosomes employed in this study are derived from E. coli.
(54) https://catalog.coriell.org/0/Sections/Search/Sample_Detail.aspx?Ref=GM05659
(55) https://www.promega.com/products/cell-health-and-metabolism/cell-viability-
assays/celltiter_glo-luminescent-cell-viability-assay/
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
26
ACS Paragon Plus Environment

Page 27 of 27
Journal of the American Chemical Society
1
2
3
4
Graphical Table of Contents:
5
6
7
N
N
N
N
NH2
N
N
8
9
+
O
O
NH2
Me
Me
Me
Me
Me
Me
O
O
N
N
O
O
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
OMe
OMe
Me
Et
NMe2
Me
NMe2
Me
Me
Et
E. coli 50S or 70S ribosome
24-48 h, rt
Ribosome-templated
azide/alkyne
cycloaddition
Me
Me
HO
O
HO
O
O
O
O
O
O
O
O
O
F
Me
F
Me
(in situ Click chemistry)
Solithromycin
27
ACS Paragon Plus Environment