Synthesis and Microbiological Evaluation of Novel Tetracyclic Fluoroquinolones

Article abstract of DOI:10.1002/cmdc.201700426

Conformationally constrained tetracyclic fluoroquinolones (FQs) were synthesized and profiled for their microbiological spectrum. The installation of a seven-membered ring between the pyrrolidine substituents and the C8 position on the FQ core scaffold resulted in a remarkable enhancement of microbiological potency toward both Gram-positive and Gram-negative bacteria. Focused optimization of seven-membered ring composition, stereochemistry, and amine placement led to the discovery of the two lead compounds that were selected for further progression.

Full text of DOI:10.1002/cmdc.201700426

Accepted Article
Title: Synthesis and Microbiological Evaluation of Novel Tetracyclic
Fluoroquinolones
Authors: Allan S Wagman, Ryan Cirz, Glenn McEnroe, James Aggen,
Martin S Linsell, Adam A Goldblum, Sara Lopez, Marcela
Gomez, George Miller, Lloyd J Simons, Thomas R Belliotti,
Christina R Harris, Toni-Jo Poel, Michael J Melnick, Ricky D
Gaston, and Heinz E. Moser
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemMedChem 10.1002/cmdc.201700426
Link to VoR: http://dx.doi.org/10.1002/cmdc.201700426
A Journal of
www.chemmedchem.org

10.1002/cmdc.201700426
ChemMedChem
Synthesis and Microbiological Evaluation of Novel Tetracyclic
Fluoroquinolones
Dr. Allan S. Wagman[a][e], Dr. Ryan Cirz[b], Glenn McEnroe[a][f], Dr. James Aggen[a][g], Martin S.
Linsell[a][h], Adam A. Goldblum[a][h], Sara Lopez[a][f], Marcela Gomez[a][i], Dr. George Miller[a], Lloyd J.
Simons[d], Thomas R. Belliotti[d], Dr. Christina R. Harris[c], Toni-Jo Poel[d], Dr. Michael J. Melnick[d], Dr.
Ricky D. Gaston[d], Dr. Heinz E. Moser*[a][f]
[a] Former or [b]current employees of Achaogen, Inc., 1 Tower Place, South San Francisco, CA 94080
[c] Former or [d]current employees of Kalexsyn, 4502 Campus Drive, Kalamazoo, MI 49008
[e] 3-V Biosciences, 3715 Haven Ave. Suite 220, Menlo Park, CA 94025
[f] Novartis Institute for Biomedical Research, 5300 Chiron Way, Emeryville, CA 94608
[g] Revolution Medicines, 700 Saginaw Drive, Redwood City, CA 94063
[h] Dice Molecules, 220 Penobscot Drive, Redwood City, CA 94063
[i] Emery Pharma, 1000 Atlantic Ave. Suite 110, Alameda, CA 94501
Abstract: Conformationally-constrained tetracyclic fluoroquinolones (FQs) were synthesized and
profiled for their microbiological spectrum. The installation of a 7-membered ring between the
pyrrolidine substituents and the C8-position on the FQ core scaffold resulted in a remarkable
enhancement of microbiological potency on both Gram-positive and Gram-negative bacteria. Focused
optimization of 7-membered ring composition, stereochemistry, and amine placement led to the
discovery of the two lead compounds that were selected for further progression.
The need for novel antibiotics, especially against multi-drug resistant (MDR) Gram-negative
pathogens, is becoming increasingly dire.[1,2] After a golden era of antibacterial drug discovery roughly
half a century ago, the field has been plagued over the past several decades with lack of success [3]
due to commercial challenges, regulatory difficulties with the evaluation and approval process, and
technical challenges identifying novel compounds that are active against Gram-negative pathogens.[4]
In contrast to Gram-positive bacteria, Gram-negative bacteria have a second, outer membrane that
lies beyond the cell wall and acts as an additional barrier, preventing highly hydrophobic molecules
from crossing. In addition, Gram-negative bacteria generally have a larger array of multi-drug efflux
pumps that expel many small molecules from the cell. Therefore target access is significantly more
challenging in Gram-negative bacteria.[5]
To this end, we were seeking to make improvements to the existing FQs by exploring the installation
of structural rigidity and how it might impact cell permeation and efflux-pump avoidance in Gram-
negative pathogens. This class of therapeutics had a long history with the initial member – 1 (nalidixic
acid) – introduced to clinical use in 1964 by Sterling.[6-9] The discovery and market introduction of 2
(norfloxacin, 1983), 3 (ofloxacin, 1985; the racemate of levofloxacin), and 4 (ciprofloxacin, 1986),
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201700426
ChemMedChem
marked a new milestone for the quinolone class of antibiotics. The introduction of a fluorine atom in
position 6 combined with the addition of a positive charge in the R7 substituent resulted in excellent
Gram-negative and good Gram-positive activity. At this time the emerging clinical need for agents
active against S. aureus and methicillin-resistant S. aureus (MRSA) had many research groups focus
exclusively on the improvement of Gram-positive antibacterial activity which had an instrumental
impact on the evolving structure activity relationship (SAR) understanding of FQs (for reviews
see[6,10,11]). Since ciprofloxacin/levofloxacin, a focused effort on optimizing Gram-negative activity of
the FQs has not occurred to the best of our knowledge.
Scheme 1. Reference quinolone antibiotics and first tetracyclic analog 6
O
N
O
O
O
O
O
F
6
7
F
3
OH
OH
OH
1
N
N
N
N
N
8
HN
O
HN
2
1
3
O
N
O
O
N
O
O
O
F
F
F
OH
OH
OH
N
N
N
N
Cl
HN
O
H2N
4
5
6
We hypothesized that the conversion of a standard FQ scaffold such as in 4 (ciprofloxacin) or 5
(clinafloxacin) to the corresponding tetracyclic analogs by connecting the R7 substituent with either the
neighboring R8 or R6 position on the bicyclic scaffold should result in a preorganization of the
saturated R7 ring. By optimizing the R7 substituent and its conformation an improvement in activity of
the resulting FQ analogs may be seen.
To test this hypothesis, we synthesized a few simple, exemplary derivatives that indicated more
promise for the R7-R8 than the R6-R7 linked derivatives (data not shown) and our attention
consequently focused on these scaffolds. Based on the synthesis and examination of a few tetracyclic
examples using either piperazine or pyrrolidine as the R7 group, we found the pyrrolidine derivatives
generally to be more potent (data not shown) and we therefore decided to exclusively focus on the
pyrrolidine family of compounds to better understand the corresponding SAR. An extensive literature
research for 7-8 linked tetracyclic FQs yielded only one literature hit, a patent published in 1987
describing a tetracyclic scaffold linking the R7 piperazine substituent to the 8-position via a 6-
membered morpholine ring.[12] We consequently decided to synthesize first key reference scaffolds
by omitting basic substituents on the pyrrolidine ring and preparing racemic compounds (Table 1) to
gain better understanding on the SAR of the 7-membered ring composition.
The detailed synthesis of these tetracyclic FQ derivatives has been published previously[13] and we
chose to outline only the synthesis of the two identified lead compounds 36 and 40. The preparation
of the key pyrrolidine and quinolone building blocks are shown in Scheme 2. The commercially
available quinolone 45 was de-methylated by treatment with hot aqueous HBr, esterified in ethanol by
the addition of thionyl chloride, and converted to the triflate 46 by treatment with N,N-
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201700426
ChemMedChem
bis(trifluoromethylsulfonyl)aniline. The triflate was converted to the aldehyde 47 by vinyl addition
followed by Lemieux–Johnson oxidative cleavage of the double bond. Both pyrrolidine coupling
partners were synthesized starting from trans-3-hydroxy-L-proline. Esterification with methanol was
followed by Boc-protection of the ring nitrogen, silylation of the 3-hydroxy group and reduction of the
methyl ester to yield alcohol 49. The required two carbon atom extension was achieved by Swern
oxidation to the aldehyde, Wittig one-carbon addition to the corresponding enol ether, hydrolysis to
the aldehyde followed by Seyferth–Gilbert homologation afforded the terminal alkyne employing the
in-situ generated Bestmann-Ohira reagent. The alkyne 50 was further transformed to the bis-Boc
protected aminopyrollidine 51 by replacement of the alcohol by the azido group with inversion of
configuration (Taber and Deker modification of the Mitsunobu reaction),[14] reduction to the
corresponding amine and Boc-protection. Preparation of the second pyrrolidine building block 53
started from the trans-3-hydroxy-L-proline methyl ester. Cbz-protection was followed by hydroxyl
group activation (m-nosylate), azide introduction with inversion of stereochemistry, reduction to the
amine with in-situ Cbz-capture to prevent cyclic urea formation followed reduction of the methyl ester
with LiBH4 and global Cbz-deprotection by hydrogenation, and finally selective trifluoracetylation of the
primary amine.
Scheme 2. Synthesis of key quinolone and pyrrolidine intermediates
a) 48% aq. HBr, 100 °C, 18 h, 86% (plus roughly 10% decarboxylated by-product); b) Anh. EtOH, 20 eq.
SOCl2, 0 °C to RT to reflux (15 min), 75%; c) 2 eq. Hϋnig’s base in THF, 1.05 eq. C6H5N(SO2CF3)2, RT,
24 h, 66% (recrystallized); d) 1.7 eq. Bu3SnCHCH2, 2.5 eq. LiCl, 0.1 eq. Pd(PPh3)2Cl2, dioxane, 80 °C,
15 h, 25%; e) 0.02 eq. OsO4, 1 eq. NMO, dioxane/water 3:1, 40 °C, 20 h, then 1 eq. NaIO4, RT, 6 h,
87%; f) 1.44 eq. AcCl in MeOH, 0 °C to 65 °C, 20 h, 95%; g) 2.5 eq. NaHCO3, 1.5 eq. Boc2O, THF, 0 °C
to RT, 10 h, 97%; h) 1.8 eq. imidazole, 1.35 eq. TBDMSCl, DMF, RT, 4 h, 96%; i) 1.5 eq. LiBH4, THF,
0 °C to RT, 18 h, 96%; j) 2 eq. (COCl)2 in MeCl2, -78 °C, then add 4 eq. DMSO followed by 49, ˂-65 °C,
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201700426
ChemMedChem
20 min, then 4 eq. TEA, 86%; k) 2.5 eq. (Methoxymethyl)triphenylphosphonium chloride, 2.3 eq. tBuOK,
THF, 0 °C to RT, 14 h l) 5% TFA in MeCN, RT, 2 h, 58% (2 steps); m) 1.5 eq. Dimethyl-2-
oxopropylphosphonate, 3 eq. K2CO3, 1.5 eq. 4-methylbenzenesulfonyl azide, MeCN, 0 °C to RT, then
addition of SM in MeOH, RT, 10 h, 86%; n) 1.5 eq. DIAD, 1 eq. Ph3P, MeCl2, 0 °C, then 1.65 eq. DPPA,
RT, 12 h; o) 3 eq. Ph3P in MeCl2, RT, conc. in vacuo, THF/water 20:3, 55 °C, 6 h; p) 1.2 eq. Boc2O, 0.1
eq. DMAP, RT, 16h, 29% (3 steps); q) 2.4 eq. NaHCO3, THF/water 10:1, 1.27 eq. CbzCl, 0 °C to RT, 23
h, 96%; r) 1.5 eq. m-Nitrobenzenesulfonyl chloride, pyridine, 0 °C to RT, 64 h, 91%; s) 4 eq. NaN3, DMF,
50°C, 16 h, 75%; t) 1.5 eq. Ph3P, THF, 0 °C to RT, then 56 eq. water, 55°C, 18 h, 3 eq. NaHCO3, 1.5 eq.
CbzCl, RT, 3 h, 91%; u) 2 eq. LiBH4, THF, 0 °C to RT, 42 h, 85%; v) 10% Pd/C, MeOH, 1 atm H2, RT, 2
h, 1 eq. CF3COOEt, THF/MeOH 4:3, 0 °C to RT, then 16 h, quant.
The preparation of the two final tetracyclic lead compounds is illustrated in Scheme 3.
Palladium/copper catalyzed Sonogashira coupling of triflate 46 with alkyne 51 yielded the linked
intermediate that was hydrogenated to give the cis-alkene 54. Deprotection of the pyrrolidine
nitrogens with TFA and deprotonation of the resulting ammonium ions led to cyclization under mild
conditions. Saponification of the ethylester provided 36. The other lead compound 40 was assembled
by SNAr reaction to link building blocks 47 and 53, reduction of the aldehyde, acid-catalyzed
cyclization of the diol with concomitant TFA deprotection and finally basic hydrolysis of the ethyl ester.
Scheme 3. Synthesis of lead compounds 36 and 40
O
N
O
O
N
O
O
N
O
F
F
F
NBoc
O
OH
F
F
c,d
O
a,b
N
+
BocHN
BocN
TfO
H2N
51
NHBoc
36
46
54
O
N
O
O
N
O
O
N
O
F
F
F
F
OH
NH
O
g,h
O
e,f
+
O
F3C
N
N
OH
NH
O
OHC
O
H2N
NH
HO
HO
40
F3C
53
47
55
a) 0.25 eq Ph3P, 0.1 eq. Pd(PPh3)4, 2 eq. Hϋnig’s base, 0.35 eq. CuI, THF, 55 °C, 12 h, 60%; b) 0.36 eq.
Et3N, 0.44 eq. quinoline, Pd/C, EtOH, 1 atm. H2, 5 h, 47%; c) 7% v/v TFA in MeCl2, RT, 12 h, 5 eq.
Hϋnig’s base, MeCN, RT, 16 h, 89%; d) 1.3 eq. NaOH in MeCN/water 12:1, 60 °C, 3 h, 62%; e) 1.25 eq.
53, NMP, 1.39 eq. Hϋnig’s base, 50 °C, 3 h, 57%; f) 2 eq. Na(MeCO2)3BH, MeCl2, RT, 22 h, 79%; g) 4 %
TFA in MeCN, RT, 70 h, 42%; h) 4 eq. K2CO3, MeOH/water 5:2, 132 h, 77%.
Based on known SAR and understanding the importance of basic amino groups appended to the
pyrrolidine substituent especially for Gram-negative potency, the significant loss of activity by the
morpholine derivative 6 was not surprising (see Table 1). However, ring expansion by the addition of a
methylene group to a saturated 7-membered ring dramatically increased the activity (7,8), indicating
the benefit such an expansion likely has on the relative positioning of the pyrrolidine ring in relation to
the bicyclic quinolone core. Replacement of the oxygen by a methylene (9) seemed to slightly
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201700426
ChemMedChem
negatively affect the activity while the corresponding unsaturated ring system was comparable (10).
Replacing the middle position of the 3-atom linker by nitrogen was not only tolerated (11-14) but also
resulted in excellent wild-type activities in E. coli, P. aeruginosa and A. baumannii especially for the
sp3-hybridized and charged amines (12,13). Compound 14, the acetylated derivative of 12, did lose a
substantial amount of activity, indicating the importance of the charged species for cellular activity.
Sulfur was tolerated at the same position (thioether 15) with insignificant changes on potency with the
exception of P. aeruginosa. Oxidation to the sulfoxide 16 or sulfone 17 resulted in a decrease of
cellular activity with the latter displaying a larger difference. Ketone 18 is comparable to sulfoxide 16
and the activity was improved 4-16 fold by conversion to oxime 20. Addition of a methyl group to the
oxime (21) slightly reduced the potency. The related amide 19, replacing a methylene in 18 by an
amine, almost completely abolished the activity, likely due to conjugation with the aromatic core and
hence different geometry. Avoiding this conjugation as in the isomeric amide 23 fully restored the
activity and resulted in one of the more active compounds. Diol 22 was prepared from the unsaturated
derivative with the notion to increase polarity; however, this modification clearly resulted in a reduction
of activity.
Table 1. MIC values in µg/mL of tetracyclic FQs, cipro- and clinafloxacin
O
O
O
N
O
F
F
OH
OH
N
N
N
Y
N
X
H
O
23
6
22
-
Entry
4
X-Y
-
SA001 SP001 EC001 EC003 PA002 PA006 AB1065
0.52a 0.81b 0.027a 0.018a 0.27a 0.046a 0.13c
0.021a 0.051c 0.009a 0.006a 0.12a 0.009a 0.045a
5
-
6
O
11b
0.06d
0.21b
0.5
-
2d
0.67b
32b
0.63d
0.42b 0.008b
1.3b
0.25d
>64b
40d
21b
>64
17b
1
64b
0.75d
0.67b
1
0.42b
0.06
0.125
0.25
8
>64d
2.5d
0.38d
32
7
CH2O
OCH2
CH2CH2
CHCH
NCH
8
9
1
1
1
10
11
12
13
14
15
16
17
0.10b
0.5b
2
0.83b
0.03
0.25
0.18b
0.03
0.25
2d
0.5
>16
0.25
0.25
64
NHCH2
N(CH3)CH2
N(COCH3)CH2
SCH2
0.25
0.25
32
0.082b
1d
1
2
4
0.125 0.125
16
>64
6.3b
>16d
>16d
8
1b
4d
2
>64
>16b
>16d
>16d
0.20b
0.16d
1d
1.67b
3d
12d
2d
e
SOCH2
-
SO2CH2
4d
16d
>16d
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201700426
ChemMedChem
18
19
COCH2
CONH
1
32
-
4
>64
1
0.25
4
>64
>64
>16
>32
64
2
8
>64
0.125
0.25
1
>64
0.5
2
>64
2
20
21
22
23
C(NOH)CH2
C(NOCH3)CH2
CH(OH)CH(OH)
-
2
0.03
0.06
0.25
<0.06
4
4
4
32
0.5
4
2
8
<0.06
0.5
8
<0.06
2
a) Average from ≥10 measurements with outlier values removed b) average of ≥3 measurements c)
average of ≥6 measurements d) average of two measurements e) mixture of stereoisomers.
Organisms used in this study: SA001 (S. aureus ATCC 29213), SP001 (S. pneumoniae ATCC 49619),
EC001 (E. coli ATCC 25922), EC003 (strain KD65 ∆tolC), PA002 (P. aeruginosa PAO1), PA006
(∆mexAB-oprM, ∆mexCD-oprJ, ∆mexEF-oprN),[15] AB1065 (A. baumannii clinical isolate from 2006,
USA, VA; aminoglycoside susceptible). Antibacterial activities were determined according to the broth
microdilution method recommended by the Clinical and Laboratory Standards Institute.[16]
Next we focused on amino-substituted pyrrolidine derivatives to address the question of position and
stereochemistry. Aminomethyl-substitued pyrrolidines are known to be highly potent substituents as
well, but the increase in pKa of the amine often results in an increased hERG activity that has to be
addressed by additional modifications.[17] Based on the potency, broad spectrum activity, acceptable
safety profiles and abundant literature knowledge, we decided to focus our exploration on the
aminopyrrolidines. From the previous data set we decided to include only a subset of promising 3-
atom linkers with the major focus of carbon, oxygen, and nitrogen analogs in the middle position.
Table 2 summarizes the data of the 4-aminopyrrolidines (the nitrogen-containing linker analogs were
not synthesized for this series) which provided the following conclusions:
i) The stereochemistry in position 2 of the pyrrolidine is essential and preferred in the R-configuration
(as determined in compounds 28-31), irrespective of the linker composition.
ii) The stereochemistry in position 4 of the pyrrolidine has a less dramatic impact on potency but is
consistently preferred in the S-configuration, averaging a 4-8 fold improvement.
iii) From the compounds with optimal stereochemistry (27,31,33,34), the saturated rings with carbon
and oxygen in the middle position (31,33) and the unsaturated carbon linker (27) looked most
promising.
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201700426
ChemMedChem
Table 2. MIC values in µg/mL of tetracyclic FQs: Evaluation of stereochemistry for clinafloxacin
analogs (first series)
O
O
O
O
O
O
O
O
F
F
F
F
OH
OH
OH
OH
H2N
H2N
H2N
H2N
N
N
N
N
N
N
N
N
Y
Y
Y
Y
X
X
X
X
24 (XY = CHCH)
28 (XY = CH2CH2)
25 (XY = CHCH)
29 (XY = CH2CH2)
32 (XY = OCH2)
26 (XY = CHCH)
30 (XY = CH2CH2)
27 (XY = CHCH)
31 (XY = CH2CH2)
33 (XY = OCH2)
34 (XY = CH2O)
Entry
4
SA001
0.52a
16
SP001
0.81b
8
EC001
0.027a
8
EC003
0.018a
8
PA002
0.27a
>64
1
PA006
0.046a
4
AB1065
0.13c
16
24
25
26
27
28
29
30
31
32
33
34
0.25
1
0.25
0.125
0.06
0.5
2
4
2
2
32
0.22b
1
4
≤0.06b
8
≤0.06b
0.013b
16
0.008b
16
0.09d
4
≤0.06b
16
64
16
0.25
8
0.038b
1
0.25
8
0.02b
0.004
0.0005
0.023d
0.25
8
0.006b
0.015
0.008
0.034d
1
0.125
2
0.021b
<0.06
<0.06
0.015
0.5
16
0.12b
32
0.45b
8
0.067b
0.25
<0.06
0.093d
0.5
2
0.5
0.015
0.125d
<0.06
0.5
0.25
0.38
See footnote of Table 1 for specifics
With this SAR understanding in mind, we also evaluated the 3-amino-pyrrolidines 35-44 with only
compounds shown with the preferred stereochemistry in the 2-position of the pyrrolidine ring (the
other diastereoisomers in this series confirmed the same trend as observed for the 4-amino-
pyrrolidines, data not shown). While the S stereochemistry is generally preferred at the 3-position of
the pyrrolidine ring for the derivatives examined, the difference is relatively small and certainly less
pronounced than that demonstrated for the 4-amino pyrrolidines. From the linker composition in the 7-
membered ring, the unsaturated carbon analog 36 and the saturated oxygen containing derivative 40
were the most promising compounds identified in this series with excellent activity against Gram-
positive and Gram-negative bacteria. Comparison of the matched pairs of 4- and 3-aminopyrrolidines
with optimized stereochemistry (Table 2 and 3; 27 vs. 36 and 33 vs. 40) revealed no significant
differences as judged by MIC values of this small bacterial panel.
Finally, we wanted to compare the optimized compounds 36 and 38 containing the 7-membered rings
with their 6- and 8-membered ring analogs 45 - 49. Removal or one methylene group in the bridging
ring (36 vs. 45 and 38 vs. 46) resulted in a greater than hundred-fold loss of antibacterial potency
against most strains examined, presumably due to a more coplanar preorganization of the pyrrolidine
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201700426
ChemMedChem
ring in relation to the quinolone core and as permeability (data not shown) and efflux properties are
comparable to the FQ comparators. Interestingly, replacement of one methylene by sulfur (46 vs. 47)
could restore most of the potency observed for the 7-membered analog 38. This experimental
observation can be rationalized by the longer carbon-sulfur bonds (1.83 vs. 1.54 Å) and the
preference of substituted thiophenols to position the substituent perpendicular to the aromatic plane.
Expansion of the ring size to the 8-membered analogs 48 and 49 reveals a negative impact on
antibacterial potency that is more pronounced for the saturated derivative 49 (roughly 100-fold
decrease of activity for most strains compared to 38) than for the unsaturated analog 48 (roughly 10-
fold decrease of activity for most strains compared to 38).
Table 3. MIC values in µg/mL of tetracyclic FQs: Evaluation of stereochemistry for bridged
clinafloxacin analogs
O
O
O
O
F
F
OH
OH
N
N
N
N
Y
Y
X
X
H2N
H2N
35
37
39
36
38
40
(XY = CH2CHCH)
(XY = CH2CH2CH2)
(XY = CH2OCH2)
(XY = CH2CHCH)
(XY = CH2CH2CH2)
(XY = CH2OCH2)
41 (XY = CH2NCH)
43
42 (XY = CH2NCH)
44
45
46
47
(XY = CH2NHCH2)
(XY = CH2NHCH2)
(XY = CHCH)
(XY = CH2CH2)
(XY = CH2S)
48 (XY = CH2CH2CHCH)
49
(XY = CH2CH2CH2CH2)
Entry
4
SA001
0.52a
0.015
0.007b
0.045d
0.015d
0.03
0.0082b
2
SP001
0.81b
0.03
≤0.015b
0.125
≤0.06b
0.125
≤0.06c
2
EC001
EC003
0.018a
0.004
0.0022b
0.06d
0.004d
0.004
0.002b
0.004
≤0.06
0.125
≤0.06
0.25
PA002
0.27a
0.093d
0.16c
0.38d
0.29b
0.125
0.25c
0.5
PA006
AB1065
0.13c
0.06
≤0.015b
0.38d
0.045d
≤0.06
≤0.03b
1
0.027a
0.015
0.0054b
0.045
0.012d
0.004
0.0057b
0.004
≤0.06
0.03
0.046a
0.002
0.0083
0.023d
0.019d
≤0.06
0.010b
≤0.06
≤0.06
0.125
≤0.06
2
35
36
37
38
39
40
41
42
43
44
45
2
2
0.5
2
4
8
1
4
1
1
≤0.06
1
0.5
2
1
8
16
4
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201700426
ChemMedChem
46
47
48
49
8
0.03
0.125
4
64
0.25
0.5
16
4
0.015
0.06
1
1
>64
0.5
4
4
64
<0.03
0.06
2
0.004
0.03
0.25
0.03
0.25
2
32
See footnote of Table 1 for specifics
In conclusion we have identified multiple novel lead compounds with exquisite potency by bridging the
8-position on the quinolone core with the 2-position of amino-substituted pyrrolidines to form a 7-
membered ring and optimizing the corresponding stereochemistry. Tetracyclic FQs 36 and 40 were
chosen as lead compounds; their structure bound to S. pneumoniae topoisomerase IV has recently be
published[18] and further in-depth microbial and preclinical profiling will be published elsewhere.
Acknowledgements: This work was partially funded by contract number HDTRA-07-C-0005. The
authors would like to thank John Griffin, Phil Patten and Kevin Judice for support and valuable
discussions.
Keywords: Antibiotics, gyrase and topoisomerase, fluoroquinolones, structure-activity relationships,
antimicrobial spectrum
Bibliography:
[1] B. Spellberg, J. Bartlett, R. Wunderink, D. N. Gilbert, Amer. J. Respir. Crit. Care Med. 2015, 191,
135-140.
[2] Antimicrobial resistance interactive database (EARS-net), European Centre for Disease
Prevention and Control: http://ecdc.europa.eu/en/healthtopics/antimicrobial-resistance-and-
consumption/antimicrobial_resistance/database/Pages/database.aspx.
[3] D. J. Payne, M. N. Gwynn, D. J. Holmes, D. L. Pompliano, Nat. Rev. Drug Discovery 2007, 6, 29-
40.
[4] H. W. Boucher, G. H. Talbot, J. S. Bradley, J. E. Edwards Jr., D. Gilbert, L. B. Rice, M. Scheld, B.
Spellberg, J. Bartlett, Clin. Infect. Dis. 2009, 48, 1-12.
[5] L. L. Silver, Bioorg. Med. Chem. 2016, 24, 6379-6389.
[6] G. Bisacchi, J. Med. Chem. 2015, 58, 4874-4882.
[7] D. C. Hooper, Emerging Infect. Dis. 2001, 7, 337-341.
[8] G. A. Jacoby, Clin. Infect. Dis. 2005, 41 (Suppl. 2), S120-S126.
[9] P. C. Appelbaum, P. A. Hunter, International Journal of Antimicrobial Agents 2000, 16, 5-15.
[10] J. M. Domagala, S. E. Hagen in Quinolone Antimicrobial Agents 3rd Edition (Eds.: D. C. Hooper,
E. Rubenstein), ASM Press, Washington DC, 2003, pp. 3-18.
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201700426
ChemMedChem
[11] L. A. Mitscher, Chem. Rev. 2005, 105, 559-592.
[12] S. Jolidon, I. Kompis, R. Locher, E. Weiss, P.-C. Wyss, (F. Hoffmann-La Roche & Co., Basel,
Switzerland), European Patent No. EP0216345A2, 1986.
[13] A. Wagman, H. E. Moser, G. A. McEnroe, J. Aggen, M. S. Linsell, A. A. Goldblum, J. H. Griffin, L.
J. Simons, T. R. Bellotti, C. R. Harris, T.-J. Poel, M. J. Melnick, R. D. Gaston, (Achaogen, Inc.,
South San Francisco, California USA), Int. PCT Pub. No. WO2009/137130 A2, 2009.
[14] D. F. Taber, P. B. Deker, H. M. Fales, T. H. Jones, H. A. Lloyd, J. Org. Chem. 1988, 53, 2968-
2971.
[15] O. Lomovskaya, A. Lee, K. Hoshino, H. Ishida, A. Mistry, M. S. Warren, E. Boyer, S.
Chamberland, V. J. Lee, Antimicrob. Agents Chemother. 1999, 43, 1340-1346.
[16] Performance standards for antimicrobial susceptibility testing; 19th informational supplement.
Clinical and Laboratory Standards Institute, Wayne, PA 2009. pp. M100-S19.
[17] C. Jamieson, E. M. Moir, Z. Rankovic, G. Wishart, J. Med. Chem. 2006, 49, 5029-5046.
[18] I. Laponogov, X.-S. Pan, L. M. Fisher, D. A. Veselkov, M. R. Sanderson, R. T. Cirz, A. Wagman,
H. E. Moser, Open Biol. 2016, 6, DOI: 10.1098/rsob.160157.
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201700426
ChemMedChem
Graphical abstract:
Conformationally constrained tetracyclic
fluoroquinolones (FQs) were synthesized and profiled
for their microbiological spectrum. The installation of a 7-
membered ring between the pyrrolidine substituents and
the C8-position on the FQ core scaffold resulted in a
remarkable enhancement of microbiological potency on
both Gram-positive as well as Gram-negative bacteria.
Focused optimization of 7-membered ring composition,
stereochemistry and amine placement yielded the two
lead compounds 36 and 40.
This article is protected by copyright. All rights reserved.