Structure-activity relationship analyses of fusidic acid derivatives highlight crucial role of the C-21 carboxylic acid moiety to its anti-mycobacterial activity

Article abstract of DOI:10.1016/j.bmc.2020.115530

Fusidic acid (FA) is a potent congener of the fusidane triterpenoid class of antibiotics. Structure-activity relationship (SAR) studies suggest the chemical structure of FA is optimal for its antibacterial activity. SAR studies from our group within the context of a drug repositioning approach in tuberculosis (TB) suggest that, as with its antibacterial activity, the C-21 carboxylic acid group is indispensable for its anti-mycobacterial activity. Further studies have led to the identification of 16-deacetoxy-16β-ethoxyfusidic acid (58), an analog which exhibited comparable activity to FA with an in vitro MIC₉₉ value of 0.8 μM. Preliminary SAR studies around the FA scaffold suggested that the hydrophobic side chain at C-20, like the C-11 OH group, was required for activity. The C-3 OH group, however, can be functionalized to obtain more potent compounds.

Full text of DOI:10.1016/j.bmc.2020.115530

Journal Pre-proofs
Structure-Activity Relationship Analyses of Fusidic Acid Derivatives High-
light Crucial Role of the C-21 Carboxylic Acid Moiety to Its Anti-mycobacte-
rial Activity
Kawaljit Singh, Gurminder Kaur, Petrus Siningu Shanika, Godwin Akpeko
Dziwornu, John Okombo, Kelly Chibale
PII:
S0968-0896(20)30356-4
https://doi.org/10.1016/j.bmc.2020.115530
BMC 115530
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Accepted Date:
12 March 2020
22 April 2020
Please cite this article as: K. Singh, G. Kaur, P. Siningu Shanika, G. Akpeko Dziwornu, J. Okombo, K. Chibale,
Structure-Activity Relationship Analyses of Fusidic Acid Derivatives Highlight Crucial Role of the C-21
Carboxylic Acid Moiety to Its Anti-mycobacterial Activity, Bioorganic & Medicinal Chemistry (2020), doi:
https://doi.org/10.1016/j.bmc.2020.115530
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. 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.
© 2020 Elsevier Ltd. All rights reserved.

Structure-Activity Relationship Analyses of Fusidic Acid Derivatives Highlight
Crucial Role of the C-21 Carboxylic Acid Moiety to Its Anti-mycobacterial Activity
Kawaljit Singh,a,b,c Gurminder Kaur,a,b Petrus Siningu Shanika,a‡ Godwin Akpeko Dziwornu,a John
Okombo,a† Kelly Chibale*a,b,c,d
aDepartment of Chemistry, University of Cape Town, Rondebosch 7701, South Africa
bDrug Discovery and Development Centre (H3D), University of Cape Town, Rondebosch 7701,
South Africa
cSouth African Medical Research Council Drug Discovery and Development Research Unit,
University of Cape Town, Rondebosch 7701, South Africa
dInstitute of Infectious Diseases and Molecular Medicine, University of Cape Town, Rondebosch
7701, South Africa
‡Current address: Department of Chemistry and Biochemistry, University of Namibia, Windhoek,
Namibia
†Current address: Department of Microbiology and Immunology, Columbia University Irving
Medical Center, 1502 Hammer Health Sciences Center, New York, NY 10032, USA
*Corresponding author: Email: kelly.chibale@uct.ac.za; Phone: +27-21-6502553

ABSTRACT
Fusidic acid (FA) is a potent congener of the fusidane triterpenoid class of antibiotics. Structure-
activity relationship (SAR) studies suggest the chemical structure of FA is optimal for its
antibacterial activity. SAR studies from our group within the context of a drug repositioning
approach in tuberculosis (TB) suggest that, as with its antibacterial activity, the C-21 carboxylic
acid group is indispensable for its anti-mycobacterial activity. Further studies have led to the
identification of 16-deacetoxy-16β-ethoxyfusidic acid (58), an analog which exhibited
comparable activity to FA with an in vitro MIC99 value of 0.8 µM. Preliminary SAR studies around
the FA scaffold suggested that the hydrophobic side chain at C-20, like the C-11 OH group, was
required for activity. The C-3 OH group, however, can be functionalized to obtain more potent
compounds.
KEYWORDS
Fusidic acid, anti-mycobacterial activity, cytotoxicity, bioisosteres

1. INTRODUCTION
Globally, there were 10 million tuberculosis (TB) infections in 2018, from which 1.5 million people
died.1 This public health concern has been exacerbated by the evolution of multidrug-resistant
TB (MDR-TB) as the World Health Organization recently estimated ~484 000 new cases with
resistance to rifampicin – the most effective first-line drug, of which 78% had MDR-TB. Therefore,
the importance of exploring new treatment options for TB cannot be overemphasized. One
potential candidate compound for anti-TB drug development is the triterpenoid antibiotic, fusidic
acid (FA (1); Fig. 1), which is used in the treatment of infections caused by methicillin-resistant
Staphylococcus aureus2,3 and for which empirical evidence regarding safety and tolerability in
humans4 as well as in vitro potency against various strains of Mycobacterium tuberculosis (M.tb)
exists.5–7 However, preclinical anti-TB development of FA is curtailed by pharmacokinetic
limitations in the rodent model of M.tb characterized by rapid clearance leading to poor
exposure.8
25
24
O
H
OH
HO
16
OAc
3
HO
H
Fig. 1: Chemical structure of fusidic acid (FA, 1) showing the substitutions at positions C-3, C-16
and C-21
The antibacterial structure-activity relationship (SAR) studies of FA have revealed that the trans–
syn–trans conformation of the tetracyclic triterpene backbone, the acetyl-oxy group at C-16 and

the carboxylic acid group at C-21 are requisite for activity. Furthermore, the orientation of the
lipophilic side chain and the carboxylic acid group around the Δ17,20 bond, rather than the
double bond, have been demonstrated to be essential for antibacterial activity.9–11
In a drug repositioning endeavor, our group has in the recent past interrogated the SAR of FA
with respect to structural modifications that influence its metabolic profile, pharmacokinetics,
and antiplasmodium activity.12–16 Even more recently, we described a prodrug approach involving
masking the metabolically labile C-3 position through esterification and demonstrated that this
strategy can improve FA concentration and tissue distribution.17 In preliminary studies, we have
previously reported on the anti-mycobacterial activity of C-21 esters and amides and showed
that the C-21 carboxylic acid group appears indispensable for anti-mycobacterial activity.12,13 In
an attempt to expand the SAR and further validate the essentiality of the C-21 carboxylic acid
group to anti-mycobacterial activity, we invoked bioisosteric replacement of the C-21 carboxylic
acid group and other subtle chemical modifications on the FA triterpenoid structure.
In medicinal chemistry, often the design of bioisosteres envisages structural modifications that
are likely to be beneficial, with shape, size, electronic distribution, polarizability, polarity,
lipophilicity, and pKa potentially playing crucial contributing roles in molecular recognition and
mimicry.18 Among the functional groups we explored as bioisosteres of the carboxylic acid group
include amides, sulfonamides, hydrazides, sulfonyl hydrazides, and oxadiazoles.

2. RESULTS AND DISCUSSION
2.1: Chemistry
The syntheses of the analogs 2–32 have previously been described.15,16 Briefly, the amide (3–11)
and sulfonamide (12-16) derivatives were obtained through amide coupling reactions of the
corresponding amines in the presence of the appropriate coupling reagents (Scheme 1). To 1 in
acetonitrile at ambient temperature was added EDCI-HOBt coupling agent, followed by hydrazine
monohydrate to afford 17, from which the acylhydrazones, 19 and 20, and the sulfonyl hydrazide
analogs, 21–26, were obtained. The N-(thiophene-2-carbonyl) fusidic acid hydrazide derivative,
18, was obtained from a reaction of the thiophene-2-carbohydrazide with 1 in the presence of
16
TBTU and DIPEA. From a reaction of 17 with CDI and Et3N in THF, the 2-oxo-1,3,4-oxadiazole
derivative 27 was obtained while its 2-thioxo congener, 28, was obtained from a reaction of CS2
with 17 under ethanol reflux and KOH as a base. The four 3-substituted 1,2,4-oxadiazole
derivatives, 29–32, were obtained in two steps by first, reacting 1 with the respective amidoxime
in the presence of EDCI and HOBt. The intermediate thus obtained was irradiated under
o
microwave conditions at 100 C in the presence of NaOAc and EtOH as a solvent to afford the
target compounds.16

R4
1,2,4-oxadiazoles
Sulfonamides
O
N
O
12: R1 = Me
13: R1 = C6H5
14: R1 = 3,4-FC6H3
15: R1 = 4-OMeC6H4
16: R1 = 4-CNC6H4
29: R4 = Me
30: R4 = Et
31: R4 = n-Pr
32: R4 = 4-FC6H4
N
H
H
O
HN S R1
HO
H
HO
H
O
OAc
OAc
H
H
HO
HO
k
O
N
N
d
H
H
X
O
O
R2
OAc
HO
H
HO
R
OAc
H
a or b or c
i or j
e or f
N
HO
1
H
H
OAc
HO
HO
17: R2 = NHNH2
H
H
H
Hydrazides
1,3,4-oxadiazoles
HO
H
Amides
27: X = O
28: X = S
18: R2 = thiophene-2-carbohydrazido
2: R = H
3: R = (CH2)11CH3
4: R = (CH2)13CH3
5: R = (CH2)15CH3
g
h
Sulfonyl hydrazides
Hydrazones
6: R = 2-morpholinoethyl
7: R = CH2CH2OH
8: R = 4-hydroxypiperidinyl
9: R = 3-hydroxyazetidinyl
10: R = CN
Y
X
21: R3 = Me
22: R3 = Et
23: R3 = n-Pr
24: R3 = n-Bu
25: R3 = 4-MeC6H4
26: R3 = 4-ClC6H4
19: X = N, Y = CH
20: X = CH, Y = N
O
O
O
H
H
H
NHN
OAc
HN N S R3
HO
HO
H
O
OAc
11: R =2H-tetrazol-5-yl
H
H
HO
HO
H

o
Scheme 1. Reagents and reaction conditions: (a) PyBOP, HOBt, DIPEA, NH4Cl, DMF, 25 C, 16 h
for 2; (b) RNH2, EDCI, DMAP, DCM, 25 oC, 12 h, for 3 - 9; (c) RNH2, EDCI, HOBt, DIPEA, DCM, 25 oC,
o
24 h for 10 and 11; (d) R1SO2NH2, EDCI, DMAP, DCM, 25 C, 48 h for 12 - 16; (e) EDCI, HOBt,
CH3CN, 25 oC, 3 h, NH2NH2.H2O, 25 oC, 16 h for 17; (f) thiophene-2-carbohydrazide, TBTU, DIPEA,
DMF, 25 oC, 16 h for 18; (g) 17, 3-or 4-pyridinecarboxaldehyde, ethanol, 85 oC, 8h for 19 and 20;
(h) 17, RSO2Cl, pyridine, 25 oC, 1–3 h for 21 - 26; (i) 17, CDI, Et3N, THF, 0 – 25 oC, 16 h for 27; (j)
17, CS2, KOH, EtOH, reflux, 16 h for 28; (k) (1) EDCI, HOBt, CH3CN, 25 °C, 3 h, amidoxime, 80 oC,
12 h, (2) NaOAc, EtOH, MW, 100 oC, 2 h for 29 - 32.
The long chain C-3 aliphatic ester derivatives, 33–35, were synthesized in moderate yields (55 -
65%) by reacting 1 with the corresponding carboxylic acid using T3P as a coupling reagent in the
presence of pyridine (Scheme 2). The aromatic ester congeners (36-42) were obtained in low
yields (<50%) through a Steglich esterification reaction using DCC/DMAP as a coupling reagent in
dry DCM (Scheme 2).19 Substituents on the phenyl ring were selected based on the Craig plot to
investigate both electronic and hydrophobic contributions to anti-mycobacterial activity.
Oxidation of 1 with Jones reagent yielded both mono (3-keto) as well as disubstituted (3,11-
diketo) derivatives. The C-3 oximes, 43–45, were synthesized by irradiating a mixture of 3-
ketofusidic acid and the respective hydroxylamine hydrochloride under microwave irradiation at
80 oC in the presence of NH4OAc. The oximes were, as expected, obtained as a mixture of E- and
Z-isomers.

O
O
O
O
m or n
r, s
t
H
H
H
1
H
OH
OH
HN N S
NHNH2
OAc
HO
H
HO
H
HO
H
O
OAc
O
O
O
H
H
H
p, q
R
O
R2
O
R2
O
46: R2 = Et
47: R2 = Bu
48: R2 = Et
49: R2 = Bu
33: R = C6H13
34: R = C9H19
35: R = C17H35
36: R = 4-FC6H5
37: R = 4-BrC6H5
38: R = 4-NO2C6H5
39: R = 4-CF3C6H5
40: R = 4-SO2MeC6H5
41: R = 4-MeC6H5
42: R = 4-pyridyl
O
p
O
21
H
O
H
OH
OH
HO
H
H
HN N S
HO
O
OAc
v
H
H
N
u
R1
O
H
50
43: R1 = OH
44: R1 = OMe
O
O
H
45: R1 = OCH2C6H5
H
HN N S
HO
H
O
O
OH
O
H
H
HN N S
HO
H
O
OAc
HO
H
51
52
HO
H

Scheme 2. Reagents and reaction conditions: (m) ROOH, T3P (50% w/v solution in DMF), pyridine,
o
o
0 – 25 C, 16 h for 33–35; (n) ROOH, DCC, DMAP, DCM, N2, 25 C, 2-12 h for 36–42; (p) Jones
reagent, acetone, 0 oC, 40 min; (q) R1ONH2.HCl, NH4OAc, EtOH, MW, 80 oC, 20 min for 43-45; (r)
propionoic acid, T3P (50% w/v solution in DMF), pyridine, 0 – 25 oC, 17 h, valeric acid anhydride,
pyridine, 25 oC, 3 h; (s) EDCI, HOBt, CH3CN, 25 oC, 3 h, NH2NH2.H2O, 25 oC, 19-24 h for 46 and 47;
(t) MsCl, pyridine, 25 oC, 1-2 h for 48 and 49; (u) H2/Pd-C, EtOH, 25 oC, 16 h; (v) K2CO3, MeOH, 25
oC, 24 h.
In further SAR studies, we investigated the anti-mycobacterial activities of FA analogs bearing
structural modifications at both C-3 and C-21 positions (46-50). The synthesis of target
compounds commenced by coupling 1 with propionic acid and valeric anhydride, mediated by
T3P, to afford the corresponding aliphatic esters,12 which upon reaction with hydrazine
monohydrate in the presence of EDCl-HOBt as a coupling reagent yielded the hydrazides 46 and
47, respectively (Scheme 2).
Further reaction of compounds 46 and 47 with mesyl chloride at ambient temperature using
pyridine as both base and solvent afforded the sulfonylhydrazinyl derivatives 48 and 49,
respectively. Meanwhile, the synthesis of 50 was accomplished through the Jones oxidation of
21. Compound 51 was obtained by catalytic hydrogenation of the double bond between C-24 and
C-25 of the mesyl hydrazide 21. On the other hand, base-catalyzed hydrolysis of 21 with K2CO3 in
methanol afforded its 16-deacetoxy-16α-hydroxy derivative 52.

O
O
O
O
H
H
OH
OAc
HO
H
HO
H
H
b or c
a
OH
OAc
R
OAc
1
H
H
H
HO
HO
d
R
H
54
53
O
55: R =
56: R =
OAc
O
O
O
O
O
H
H
H
OH
OAc
OH
OH
O
e
HO
H
f
HO
H
HO
OH
H
H
H
AcO
AcO
AcO
H
I
III
II
g
O
O
O
57: R1 = Me
58: R1 = Et
h
i
O
O
O
O
H
H
H
OH
OR1
O
O
HO
HO
H
HO
Br
Br
H
H
H
HO
AcO
AcO
H
H
V
IV

o
o
Scheme 3. Reagents and reaction conditions: (a) O3, PPh3, DCM, -78 C to 25 C, 2 h; (b) Jones
reagent, acetone, 0 oC, 40 min for 55; (c) acetic anhydride, pyridine, MW, 80 oC, 1.5 h for 56; (d)
o
o
acetic anhydride, pyridine, 25 C, 3 h; (e) Sat. aq.NaHCO3, MW, 100 C, 10 min.; (f)
ClCH2OCOC(CH3)3, Et3N, DMF, 25 oC, 16 h; (g) PPh3, CBr4, benzene, 25 oC, 1 h; (h) Bu4N+Br-, MeCN,
25 oC, 65 h; (i) (1) R1OH, Ag2CO3, 25 oC, 16 h, (2) 5N aq. NaOH, 80 oC, 1 h
Cleavage of the Δ24,25 and Δ17,20 bonds of FA by an ozonolysis reaction yielded low quantities
(26-27%) of the aldehydic and keto products, 53 and 54, respectively (Scheme 3).20 The 3,11-
diketo derivative, 55, was obtained by Jones oxidation of FA, and the triacetoxy FA derivative, 56,
was obtained by reacting FA with acetic anhydride in the presence of pyridine under microwave
irradiation at 80 oC.
The synthesis of the C-16 derivatives of FA, 57 and 58, was accomplished in six steps (Scheme
3).21,22 First, the C-3 OH group of FA was protected as the acetoxy group by reacting FA with acetic
anhydride in the presence of pyridine to obtain the C-3 acetoxy derivative (I). Then, intermediate
I underwent hydrolysis of the C-16 acetoxy group in saturated aqueous NaHCO3 solution under
microwave irradiation at 100 oC to afford the intermediate II with inversion of configuration at C-
16 (β-acetoxy to α-OH). The C-21 carboxylic acid group of II was then protected as the
pivaloyloxymethyl ester (III) after a reaction with chloromethyl pivalate under basic conditions.
This was followed by C-16 bromination with tetrabromomethane (CBr4) in the presence of
triphenylphosphine (PPh3) to afford intermediate IV with inversion of configuration (α-OH to β-
bromo). The C-16 β- bromo intermediate IV was converted to its α-bromo epimer V by reacting
with tetrabutylammonium bromide in acetonitrile at ambient temperature. Finally, the
intermediate V was reacted with the respective alcohols (MeOH or EtOH) in the presence of

Ag2CO3 to afford the corresponding C-16 ether derivatives, which upon hydrolysis of the C-3
acetate and C-21 pivalate ester groups with 5 N NaOH solution afforded the target compounds
57 and 58.
Detailed synthetic procedures are described in Supplementary information S3-S30. All target
compounds were purified using column chromatography and fully characterized by analytical and
spectroscopic techniques. The isolated yields, anti-mycobacterial and cytotoxic activities of the
target compounds are reported in Table 1 (compounds 1-50) and Table 2 (compounds 51-58)
below.

2.2: Biology
Table 1: Isolated yields, anti-mycobacterial and cytotoxic activities of compounds 1-50
25
24
R
H
HO
16
OAc
3
R1
H
Yield
(%)
H37Rv,
MIC99, µM
CHO*,
IC50, µM
Compound
R
R1
1 (Fusidic acid)
-
0.6
>194
113.0
>146
>140
>135
OH
OH
OH
OH
OH
COOH
2
3
4
5
31
51
71
73
>125
>125
>125
>125
CONH2
CONH(CH2)11CH3
CONH(CH2)13CH3
CONH(CH2)15CH3
O
O
6
7
67
25
>125
>125
-
-
OH
OH
N
N
H
O
OH
N
H
O
O
8
9
18
18
40
40
-
-
OH
OH
N
OH
N
OH
10
11
53
29
40
40
>185
CONHCN
OH
OH
O
N
NH
N
>171.0
N
H
N
O
O
12
55
80
>169.0
OH
N
H
S
O

O
O
S
13
14
15
38
30
15
>125
20
96.1
81.4
-
OH
OH
OH
N
H
O
F
O
O
O
O
S
N
H
F
O
O
S
31.3
N
H
OMe
CN
O
O
S
16
17
18
30
98
50
125
40
-
OH
OH
OH
N
H
O
112.0
78.8
CONHNH2
O
O
S
40
N
H
N
H
O
19
20
63
63
20
20
-
-
OH
OH
N
N
N
H
O
N
H
N
H
N
O
O
O
21
22
90
45
10
40
92.2
77.3
OH
OH
N S
O
N
H
O
H
N S
N
H
O
O
O
O
O
H
23
24
25
71
40
79.7
49.9
OH
OH
N S
N
H
O
O
H
>125
N S
N
H
O
O
H
25
46
>125
48.2
OH
N S
N
H
O

O
O
H
26
65
>125
53.2
OH
N S
O
Cl
N
H
N
NH
27
28
24
56
80
80
82.0
70.0
OH
OH
O
N
O
NH
O
N
S
29
30
24
23
>125
>125
47.8
71.7
OH
OH
OH
N
N
O
N
O
N
31
32
14
20
80
68.2
77.4
N
O
F
>125
OH
N
O
N
O
33
34
35
62
63
57
5.0
36.2
COOH
COOH
COOH
O
O
O
O
3
O
O
O
>125
>125
-
-
6
14
36
37
6
>125
>125
-
-
COOH
COOH
F
O
O
24
O
O
Br
38
14
>125
-
COOH
NO2

O
O
O
O
39
40
35
39
>125
31.3
-
-
COOH
COOH
CF3
SO2Me
O
O
41
42
12
11
31.3
31.3
-
-
COOH
COOH
O
O
N
43
44
45
23
59
51
2.5
20
40
164
COOH
COOH
COOH
N
N
N
OH
OCH3
OCH2C6H5
-
-
O
46
47
78
58
15.6
62.5
-
-
CONHNH2
CONHNH2
O
O
O
O
O
O
H
48
49
50
44
54
29
7.81
62.5
-
-
N S
O
N
H
O
O
O
O
H
O
N S
N
H
O
O
O
H
62.5
-
O
N S
N
H
O
Rifampicin
Emetine
0.005
0.4
*CHO: Chinese Hamster Ovarian cell line
The in vitro anti-mycobacterial activities of all synthesized compounds was investigated against
the drug-sensitive M.tb H37Rv strain. The MIC99 (minimum concentration required to inhibit the
growth of 99% of the bacterial population) was determined for each compound in the glycerol-

alanine-salt with Tween-80 and iron (GAST/Fe) media. The assay was performed at a maximum
concentration limit of each test compound set at 125 µM (Table 1) and rifampicin was used as
the reference drug (Table 1).
The anti-mycobacterial activity (MIC99) of FA (1) was found to be 0.6 μM. To enable analysis of
their SAR, all compounds with MIC99 ≤10 μM were regarded as active (relative to FA (1), Table 1);
those with MIC99 ranging from 10 – 40 μM were classified as having moderate activity while
compounds with MIC99 values in the range 40 – 125 μM were denoted as poorly active and
compounds exhibiting MIC99 >125 μM being considered inactive.
The C-21 esters of FA have exhibited poor anti-mycobacterial activity12 while the aromatic amides
showed encouraging activity.13 Replacing the C-21 carboxylic acid group with the primary and
secondary aliphatic amide groups (2-7, MIC99 >125 μM) led to a loss of activity. However, the
alicyclic amides 8 and 9 regained some activity with a MIC99 value of 40 μM. Likewise, the
cyanoamide (10) and tetrazoloamide (11) bioisosteres showed moderate activity. The methane
sulfonamide derivative (12) exhibited poor activity while the unsubstituted phenyl derivative (13)
was inactive at the highest concentration tested. However, activity was regained following the
introduction of various substitutions on the phenyl ring (14–16, Table 1). The 3,4-
difluorophenylsulfonamide derivative 14 was found to be the most active compound among
amide and sulfonamide derivatives, displaying moderate anti-mycobacterial activity with a MIC99
value of 20 μM.
The unsubstituted hydrazide analog 17 exhibited moderate activity (MIC99 = 40 μM). The
pyridylhydrazone analogs, 19 and 20, showed a 2-fold improvement in potency (MIC99 = 20 μM),

but the thiophene-2-carbonyl derivative, 18, exhibited comparable activity to 17. Among the
sulfonyl hydrazides (21-26), the aromatic derivatives (25 and 26) were found to be inactive
compared to their aliphatic congeners (21-24), which demonstrated a decrease in activity with
increasing carbon chain length from the methyl to the butyl group (Table 1). The methane
sulfonyl hydrazide derivative, 21, was found to be the most active congener in this series (MIC99
= 10 μM).
Replacing the carboxylic acid group of FA with the oxadiazole bioisosteres was found to be
detrimental to anti-mycobacterial activity (Table 1). The analogs 27 and 28 exhibited comparable
activity (MIC99 = 80 μM). The 1,2,4-oxadiazole congeners were also found to be either poorly
active (31, MIC99 = 80 μM) or inactive (29, 30, and 32, MIC99 >125 μM). In summary, bioisosteric
replacement of the carboxylic acid group of FA resulted in poor anti-mycobacterial activity as
compared to FA except for 21, which showed moderate activity (MIC99 = 10 μM).
FA C-3 n-propyl, n-butyl and silicate esters have earlier been synthesized as potential prodrugs.12
These ester analogs demonstrated anti-mycobacterial activity (MIC99 = 0.2 – 2.5 µM) comparable
to FA (MIC99 = 0.6 µM). However, it was observed in the current work that the potency of the
alkyl esters significantly decreased with increasing carbon chain length, as exemplified by 33–35
(MIC99 = 5 – >125 µM [Table 1]). The aryl ester analogs similarly showed poor activity as earlier
observed in their C-21 counterparts.12 However, the 4-mesylphenyl (40), 4-methylphenyl (41) and
pyridyl congeners (42) were moderately active (MIC99 = 31.3 µM), suggesting that electron-
withdrawing and hydrophobic groups were not tolerated. Meanwhile, functionalization of the C-
3 OH group to the oxime resulted in a gain in potency which decreased with increasing size of

substituents from the unsubstituted oxime (43, MIC99 = 2.5 µM) to the benzyl-substituted oxime
(46, MIC99 = 40 µM). The anti-mycobacterial activities of the two valeric esters, 47 and 49, and
the 3-keto sulfonyl hydrazide derivative, 50, were similar with a MIC99 value of 62.5 µM. However,
the propionic ester, 48, showed about a two-fold increase in potency (MIC99 = 7.8 µM) relative
to its hydrazide congener, 46 (Table 1).
We also undertook preliminary SAR studies, through subtle chemical modifications, to investigate
the contributions of the C-16 acetyl group and the C-20 hydrophobic side chain on the anti-
mycobacterial activity of FA (Scheme 3, Table 2). Briefly, a comparison of the activity of 51 to 21
seems to suggest that saturation of the Δ24,25 bond is well tolerated. However, the
deacetylation of the C-16 acetoxy group reduced the activity by 4-fold, indicating the importance
of the C-16 acetoxy group for anti-mycobacterial activity (52, MIC99 = 40 µM). Oxidation of the
Δ24,25 (53) and Δ17,20 (54) bonds were detrimental to anti-mycobacterial activity, thus
suggesting the importance of the C-17 side chain of FA.
The 3-keto derivative of FA exhibited low micromolar anti-mycobacterial activity (MIC99 = 1.25
µM), while its 3-acetoxy analog showed moderate activity (MIC99 = 20 µM).12 However, in the
current work, the C-3,11 diketo (55) and C-3,11-diacetoxy (56) derivatives showed reduced
activity, indicating the significance of the C-11 OH group (Table 2). However, the 16-deacetoxy-
16β-ethoxy derivative of FA (58, MIC99 = 0.8 µM) exhibited activity comparable to FA, indicating
that modifications at the C-16 position are well tolerated.

Table 2: Isolated yields, anti-mycobacterial and cytotoxic activities of compounds 51-58
Compound
Yield (%)
58
H37Rv, MIC99, µM
CHO, IC50, µM
97.8
51
52
53
54
55
56
57
58
10
40
22
32.0
27
40
26
>125
11.9
52.5
18.7
0.8
45
35
35
29.0
48.1
37
2.3: Cytotoxicity
Representative derivatives were evaluated for cytotoxicity against the Chinese Hamster Ovarian
(CHO-K1) mammalian cell line, and their half minimal inhibitory concentration (IC50) was
determined (Table 1 and Table 2). Emetine was used as a reference drug in this assay. The amide
analogs were generally non-cytotoxic. The cyanoamide, tetrazoloamide, and sulfonamide C-21
carboxylic acid surrogates were also non-cytotoxic except 13 and 14 (IC50 <100 µM). Similarly, all
tested sulfonyl hydrazide and oxadiazole analogs exhibited cytotoxic activity with IC50 values
<100 µM. The most potent analog, the 16-deacetoxy-16β-ethoxyfusidic acid (58) was cytotoxic
at IC50 = 48.1 µM.

3. CONCLUSION
In the current work, the anti-mycobacterial activity of a series of FA C-21 carboxylic acid
bioisosteres was evaluated towards expanding the SAR. None of these derivatives possessed
superior in vitro activity to FA. Meanwhile, preliminary modifications of the C-11 OH, C-16
acetoxy and C-20 hydrophobic side chain groups led to a loss of activity, which seems to suggest
their importance for anti-mycobacterial activity.
AUTHOR INFORMATION
Corresponding Author: *E-mail: Kelly.Chibale@uct.ac.za:
Phone: +27-21-6502553. Fax: +27-21-6505195.
CONFLICT OF INTEREST
The authors declare no competing interests.
ACKNOWLEDGMENTS
The authors are grateful for the support of the University of Cape Town, South African Medical
Research Council (SAMRC) and the South African Research Chairs Initiative (SARChI) of the
Department of Science and Technology administered through the South African National
Research Foundation (K.C.). We thank Ronnett Seldon of the Drug Discovery and Development
Centre (H3D) for the anti-mycobacterial data and Dr. Dale Taylor (H3D) for the cytotoxicity data
generated on these compounds.

REFERENCES
(1)
(2)
(3)
WHO.
Global
Tuberculosis
Report
2019
https://www.who.int/tb/publications/factsheet_global.pdf?ua=1.
Whitby, M. Fusidic Acid in the Treatment of Methicillin-Resistant Staphylococcus Aureus.
Int. J. Antimicrob. Agents 1999, 12, S67–S71.
Sahm, D. F.; Deane, J.; Pillar, C. M.; Fernandes, P. In Vitro Activity of CEM-102 (Fusidic Acid)
against Prevalent Clones and Resistant Phenotypes of Staphylococcus Aureus. Antimicrob.
Agents Chemother. 2013, 57 (9), 4535–4536.
(4)
(5)
Kraus, C. N.; Burnstead, B. W. The Safety Record of Fusidic Acid in Non-US Markets: A Focus
on Skin Infections. Clin. Infect. Dis. 2011, 52 (suppl_7), S527–S537.
Cicek-Saydam, C.; Cavusoglu, C.; Burhanoglu, D.; Hilmioglu, S.; Ozkalay, N.; Bilgic, A. In Vitro
Susceptibility of Mycobacterium Tuberculosis to Fusidic Acid. Clin. Microbiol. Infect. 2001,
7 (12), 700–702.
(6)
(7)
(8)
(9)
Van Caekenberghe, D. Comparative In-Vitro Activities of Ten Fluoroquinolones and Fusidic
Acid against Mycobacterium Spp. J. Antimicrob. Chemother. 1990, 26 (3), 381–386.
Hoffner, S. E.; Olsson-Liljequist, B.; Rydgård, K. J.; Svenson, S. B.; Källenius, G. Susceptibility
of Mycobacteria to Fusidic Acid. Eur. J. Clin. Microbiol. Infect. Dis. 1990, 9 (4), 294–297.
Findon, G.; Miller, T. E.; Rowe, L. C. Pharmacokinetics of Fusidic Acid in Laboratory Animals.
Lab. Anim. Sci. 1991, 41 (5), 462–465.
Godtfredsen, W. O.; von Daehne, W.; Tybring, L.; Vangedal, S. Fusidic Acid Derivatives. I.

Relationship between Structure and Antibacterial Activity. J. Med. Chem. 1966, 9 (1), 15–
22.
(10) von Daehne, W.; Godtfredsen, W. O.; Rasmussen, P. R. Structure-Activity Relationships in
Fusidic Acid-Type Antibiotics. Adv. Appl. Microbiol. 1979, 25, 95–146.
(11) Duvold, T.; Sørensen, M. D.; Björkling, F.; Henriksen, A. S.; Rastrup-Andersen, N. Synthesis
and Conformational Analysis of Fusidic Acid Side Chain Derivatives in Relation to
Antibacterial Activity. J. Med. Chem. 2001, 44 (19), 3125–3131.
(12) Njoroge, M.; Kaur, G.; Espinoza-Moraga, M.; Wasuna, A.; Dziwornu, G. A.; Seldon, R.;
Taylor, D.; Okombo, J.; Warner, D. F.; Chibale, K. Semisynthetic Antimycobacterial C-3
Silicate and C-3/C-21 Ester Derivatives of Fusidic Acid: Pharmacological Evaluation and
Stability Studies in Liver Microsomes, Rat Plasma, and Mycobacterium Tuberculosis
Culture. ACS Infect. Dis. 2019, 5 (9), 1634–1644.
(13) Dziwornu, G. A.; Kamunya, S.; Ntsabo, T.; Chibale, K. Novel Antimycobacterial C-21 Amide
Derivatives of the Antibiotic Fusidic Acid: Synthesis, Pharmacological Evaluation and
Rationalization of Media-Dependent Activity Using Molecular Docking Studies in the
Binding Site of Human Serum Albumin. Medchemcomm 2019, 10 (6), 961–969.
(14) Kaur, G.; Pavadai, E.; Wittlin, S.; Chibale, K. 3D-QSAR Modeling and Synthesis of New
Fusidic Acid Derivatives as Antiplasmodial Agents. J. Chem. Inf. Model. 2018, 58 (8), 1553–
1560.
(15) Espinoza-Moraga, M.; Singh, K.; Njoroge, M.; Kaur, G.; Okombo, J.; De Kock, C.; Smith, P.

J.; Wittlin, S.; Chibale, K. Synthesis and Biological Characterisation of Ester and Amide
Derivatives of Fusidic Acid as Antiplasmodial Agents. Bioorg. Med. Chem. Lett. 2017, 27 (3),
658–661.
(16) Kaur, G.; Singh, K.; Pavadai, E.; Njoroge, M.; Espinoza-Moraga, M.; De Kock, C.; Smith, P. J.;
Wittlin, S.; Chibale, K. Synthesis of Fusidic Acid Bioisosteres as Antiplasmodial Agents and
Molecular Docking Studies in the Binding Site of Elongation Factor-G. Medchemcomm
2015, 6 (11), 2023–2028.
(17) Strydom, N.; Kaur, G.; Dziwornu, G. A.; Okombo, J.; Wiesner, L.; Chibale, K.
Pharmacokinetics and Organ Distribution of C-3 Alkyl Esters as Potential Antimycobacterial
Prodrugs of Fusidic Acid. ACS Infect. Dis. 2020, in press.
(18) Meanwell, N. A. Synopsis of Some Recent Tactical Application of Bioisosteres in Drug
Design. J. Med. Chem. 2011, 54 (8), 2529–2591.
(19) Xu, J.-Q.; Shen, Q.; Li, J.; Hu, L.-H. Dammaranes from Gynostemma Pentaphyllum and
Synthesis of Their Derivatives as Inhibitors of Protein Tyrosine Phosphatase 1B. Bioorg.
Med. Chem. 2010, 18 (11), 3934–3939.
(20) Janssen, G.; Vanderhaeghe, H. Modification of the Side Chain of Fusidie Acid (Ramycin). J.
Med. Chem. 1967, 10 (2), 205–208.
(21) Duvold, T.; Bretting, C. A. S.; Rasmussen, P. A.ꢀ; Bouerat, L.; Thorhauge, J. Novel Fusidic
Acid Derivatives. WO2005007669 A1, 2005.
(22) Von Daehne, W.; Rasmussen, P. R. 16-Ethers Of Fusidic Acid Derivatives. US4060606, 1977.

The authors declare that they have no known competing financial interests or personal
relationships that could have appeared to influence the work reported in this paper.