Incorporation of triphenylphosphonium functionality improves the inhibitory properties of phenothiazine derivatives in Mycobacterium tuberculosis

Article abstract of DOI:10.1016/j.bmc.2014.07.050

Tuberculosis (TB) is a difficult to treat disease caused by the bacterium Mycobacterium tuberculosis. The need for improved therapies is required to kill different M. tuberculosis populations present during infection and to kill drug resistant strains. Protein complexes associated with energy generation, required for the survival of all M. tuberculosis populations, have shown promise as targets for novel therapies (e.g., phenothiazines that target type II NADH dehydrogenase (NDH-2) in the electron transport chain). However, the low efficacy of these compounds and their off-target effects has made the development of phenothiazines as a therapeutic agent for TB limited. This study reports that a series of alkyltriphenylphosphonium (alkylTPP) cations, a known intracellular delivery functionality, improves the localization and effective concentration of phenothiazines at the mycobacterial membrane. AlkylTPP cations were shown to accumulate at biological membranes in a range of bacteria and lipophilicity was revealed as an important feature of the structure-function relationship. Incorporation of the alkylTPP cationic function significantly increased the concentration and potency of a series of phenothiazine derivatives at the mycobacterial membrane (the site of NDH-2), where the lead compound 3a showed inhibition of M. tuberculosis growth at 0.5 μg/mL. Compound 3a was shown to act in a similar manner to that previously published for other active phenothiazines by targeting energetic processes (i.e., NADH oxidation and oxygen consumption), occurring in the mycobacterial membrane. This shows the enormous potential of alkylTPP cations to improve the delivery and therefore efficacy of bioactive agents targeting oxidative phosphorylation in the mycobacterial membrane.

Full text of DOI:10.1016/j.bmc.2014.07.050

Accepted Manuscript
Incorporation of triphenylphosphonium functionality improves the inhibitory
properties of phenothiazine derivatives in Mycobacterium tuberculosis
Elyse A. Dunn, Marina Roxburgh, Lesley Larsen, Robin A.J. Smith, Alexander
D. McLellan, Adam Heikal, Michael P. Murphy, Gregory M. Cook
PII:
S0968-0896(14)00574-4
DOI:
http://dx.doi.org/10.1016/j.bmc.2014.07.050
BMC 11745
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
23 May 2014
21 July 2014
30 July 2014
Please cite this article as: Dunn, E.A., Roxburgh, M., Larsen, L., Smith, R.A.J., McLellan, A.D., Heikal, A., Murphy,
M.P., Cook, G.M., Incorporation of triphenylphosphonium functionality improves the inhibitory properties of
phenothiazine derivatives in Mycobacterium tuberculosis, Bioorganic & Medicinal Chemistry (2014), doi: http://
dx.doi.org/10.1016/j.bmc.2014.07.050
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1
2
3
Incorporation of triphenylphosphonium functionality improves the
inhibitory properties of phenothiazine derivatives in Mycobacterium
tuberculosis
4
5
6
7
Elyse A. Dunn1 , Marina Roxburgh2, Lesley Larsen2, Robin A. J. Smith2,
Alexander D. McLellan1, , Adam Heikal1, Michael P. Murphy3 and Gregory M.
Cook1
8
9
1 Department of Microbiology & Immunology, University of Otago, PO Box 56,
Dunedin 9054, New Zealand
10
11 2 Department of Chemistry, University of Otago, Dunedin 9056, Dunedin
12
9054, New Zealand
13 3 MRC Mitochondrial Biology Unit, Hills Road, Cambridge CB2 0XY, UK
14
15
16
Corresponding author: Gregory M. Cook1, gregory.cook@otago.ac.nz,
+64 (3) 479-7722
17

18
ABSTRACT
19
20
Tuberculosis (TB) is a difficult to treat disease caused by the bacterium
Mycobacterium tuberculosis. The need for improved therapies is required to
21 kill different M. tuberculosis populations present during infection and to kill
22 drug resistant strains. Protein complexes associated with energy generation,
23 required for the survival of all M. tuberculosis populations, have shown
24 promise as targets for novel therapies (e.g. phenothiazines that target type II
25 NADH dehydrogenase (NDH-2) in the electron transport chain). However, the
26 low efficacy of these compounds and their off-target effects has made the
27 development of phenothiazines as a therapeutic agent for TB limited. This
28 study reports that a series of alkyltriphenylphosphonium (alkylTPP) cations, a
29 known intracellular delivery functionality, improves the localization and
30 effective concentration of phenothiazines at the mycobacterial membrane.
31 AlkylTPP cations were shown to accumulate at biological membranes in a
32 range of bacteria and lipophilicity was revealed as an important feature of the
33 structure-function relationship. Incorporation of the alkylTPP cationic function
34 significantly increased the concentration and potency of a series of
35 phenothiazine derivatives at the mycobacterial membrane (the site of NDH-2),
36 where the lead compound 3a showed inhibition of M. tuberculosis growth at
37
0.5 μg/mL. Compound 3a was shown to act in a similar manner to that
38 previously published for other active phenothiazines by targeting energetic
39 processes (i.e. NADH oxidation and oxygen consumption), occurring in the
40 mycobacterial membrane. This shows the enormous potential of alkylTPP
41 cations to improve the delivery and therefore efficacy of bioactive agents
42 targeting oxidative phosphorylation in the mycobacterial membrane.
43
1. INTRODUCTION
44
45
Tuberculosis (TB) is a difficult to treat disease caused by the pathogen
Mycobacterium tuberculosis. Annually, 1.4 million people worldwide die from
46 TB infection, 9.4 million new cases are diagnosed and US $7–8 billion is
47 required for TB control programmes [1]. The first line drugs (isoniazid,
48 rifampicin, pyrazinamide and ethambutol) developed in the 1950s–1960s, are

49 today still used to treat TB infection. However with no new advances in TB
50 therapy, the emergence of multi-drug resistant tuberculosis (MDR-TB) and
51 extensively drug-resistant tuberculosis (XDR-TB) has become increasingly
52 apparent [1]. New drugs are required to reduce the current treatment time of
53
54
TB therapy and improve compliance to therapy regimens.
The generation of cellular energy by oxidative phosphorylation is essential
55 for the survival of replicating and non-replicating populations of M.
56 tuberculosis [2,3]. Protein complexes within the electron transport chain have
57
58
already shown promise as novel drug targets; such as the F1F0 ATPase which
is inhibited by bedaquiline, and the cytochrome bc1 complex which is inhibited
59 by imidazo[1,2-a]pyridine analogues [4,5]. In recent years, much work has
60 been done on the characterization of type II NADH dehydrogenase (NDH-2) in
61 bacteria[6-8] with the first bacterial structure of this enzyme recently
62 published[9]. NDH-2 is a monotopic membrane protein that does not
63 contribute to the formation of the proton electrochemical potential gradient.
64
NDH-2 has been identified as a potential target for the treatment of M.
65 tuberculosis [10,11]. M. tuberculosis contains two copies of NDH-2 (ndh and
66 ndhA) and when ndh is disrupted by transposon mutagenesis, M. tuberculosis
67
is non-viable[12]. Neither NDH-2 copies from M. tuberculosis have
68 mammalian homologues and have already been shown to be targetable by
69 several groups of compounds, namely the phenothiazines [10,11,13], which
70
71
have shown potential as novel therapies for tuberculosis.
Phenothiazines are bactericidal against several populations of M.
72 tuberculosis: namely replicating, non-replicating, hypoxic and nutrient starved
73 bacteria, raising the prospect of shortening the current tuberculosis therapy
74 regime [14-18]. Phenothiazines are toxic to clinical isolates of M. tuberculosis,
75 including MDR-TB and XDR-TB [11,17,19-22]. Chlorpromazine (CPZ) and
76 thioridazine (THR) have been shown to have bactericidal action against MDR-
77 TB that is resistant to all four of the frontline therapies [21]. Trifluoperazine
78 (TPZ) and CPZ inhibit purified NDH-2 (ndh gene) from M. tuberculosis at an
79 IC50 of 12 μM and 10 μM respectively [11,23]. However, the concentrations of
80 these compounds require to kill M. tuberculosis is still higher than inhibitors
81 used in current therapies such as rifampicin and isoniazid (μM) and
82 bedaquiline (nM) [4,11,24,25]. Due to their ability to also target neurological

83 pathways, efforts are required to increase the potency and specificity of
84 phenothiazines for their desired anti-tubercular effects.
85
Alkyltriphenylphosphonium (alkylTPP) cations are lipophilic, positively
86 charged ions known to accumulate within negatively charged compartments in
87 response to the membrane potential and to adsorb at the surface of biological
88 membranes [26,27]. Incorporation of alkylTPP cation function into more
89 complex structures has been shown to localise these conjugates to the
90 mammalian mitochondria for the treatment of a range of animal models of
91 diseases that are caused by mitochondrial dysfunction [28,29]. For example
92 MitoQ, an alkylTPP cation-antioxidant combination, has been used in Phase II
93 clinical trials [30]. The success of alkylTPP cations in facilitating the
94 localisation of bioactive compounds to the mitochondria in response to
95 membrane potential and charge is well documented. We propose that
96 alkylTPP cations can be further exploited to increase the potency of
97 phenothiazine derivatives by localizing them more effectively to NDH-2 in the
98
99
mycobacterial membrane.
In this study we demonstrate that lipophilicity is important for the
accumulation of alkylTPP cations by M. smegmatis, M. tuberculosis,
Staphylococcus aureus and Enterococcus faecalis, but not the gram-negative
organism Escherichia coli. We demonstrate that the conjugation of an
alkylTPP cation to phenothiazine derivatives increases their potency up to
130-fold against M. tuberculosis. The lead alkylTPP cation-phenothiazine
derivative conjugate (3a) was shown to inhibit both oxygen consumption and
NADH oxidation, two essential energetic processes in the growth and survival
of mycobacteria.
100
101
102
103
104
105
106
107
108
109
2. RESULTS and DISCUSSION
110
111
2.1.Novel synthesis of alkylTPP cation-phenothiazine derivative
conjugates.
112
113
114
It has been reported that the three-ring structure of phenothiazines is
essential for their antimycobacterial activity and this feature was maintained in
all analogues synthesised in our investigation [10]. Compounds 3a, 3b and 3c

115
116
117
118
119
120
121
122
123
were prepared from commercially available precursor materials (Scheme 1).
Sodium hydride was reacted with each phenothiazine derivative (2a–2c) to
give a highly conjugated system, which formed the desired intermediates
(2a’–2c’) on addition of 1-bromo-3-chloropropane. Subsequent reaction of
these intermediates with triphenylphosphine resulted in the desired products
(3a–3c). All compounds were analysed with HPLC and mass spectrometry.
Compounds 1e, 1g, 1h, 1i, 1j, 1l and 2d were also successfully synthesized
following previously established protocols [31-36].
124
2.2.The interaction of alkylTPP cations with bacterial membranes.
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
AlkylTPP cation moieties have been used to deliver bioactive agents to
mammalian mitochondria and structure-activity relationships have been
developed to improve their localization to the membrane [28-30]. An example
of this is the alkylTPP-antioxidant combination, MitoQ, where an increase in
accumulation at the mitochondrial membrane and subsequent antioxidant
activity was observed with the more lipophilic analogues[37]. In the current
study, the structure-activity relationship between alkylTPP cation lipophilicity
(Figure 1) and membrane accumulation was investigated in a range of
bacterial species, including M. smegmatis (obligate aerobe, acid fast), M.
tuberculosis (obligate aerobe, acid fast), S. aureus (facultative aerobe, gram-
positive), E. faecalis (obligate fermenter, gram-positive) and E. coli (facultative
anaerobe, gram-negative) (Figure 2). The alkylTPP cations 1a–1l, with
increasing lipophilicities (logP values 3.41–9.36), were tested. Toxicity, as
indicated by a low minimum inhibitory concentration (MIC, μg/ml), was used
as a measure of effective membrane accumulation[38].
In mycobacteria, alkylTPP cation accumulation correlated with an increase
of lipophilicity between logP values of 5.9–9.4 and 5.4–7.9 in M. tuberculosis
and M. smegmatis respectively (Figure 2). This relationship was not sustained
in M. smegmatis at logP values greater than 8, where the MIC values steadily
increased, returning to 64 μg/ml at logP 9.4. This may be attributed to
differences in the structure of the cell wall of these organisms. The mycolic
acid moieties of the mycobacterial cell wall contain cis and trans double

147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
bonds, where an increase in the latter decreases permeability at the inner
most part of the bilayer [39]. M. tuberculosis contains <10% trans bonds, while
M. smegmatis contains 68% [39]. This higher proportion of trans double
bonds in the cell wall of M. smegmatis may hinder the passage of more
lipophilic compounds to the membrane. Alternately, it could be argued that the
accumulation of more hydrophobic compounds in M. smegmatis is hindered
by the presence of a drug efflux mechanism exclusive to M. smegmatis with
the capacity to transport lipophilic cations. However, there is little evidence to
support this theory and contrary to this, it has been reported that M.
tuberculosis expresses the efflux pump Mmr, which confers resistance to
tetraphenylphosphonium bromide [40].
Staphylococcus aureus, a facultative aerobe, also displayed a relationship
between toxicity/accumulation and lipophilicity, between logP values of 5.4–
9.4 (Figure 2). In contrast, Enterococcus faecalis, a fermentative bacterium,
did not show this relationship until much higher logP values, with the MIC
remaining at 128 μg/ml until the logP of the tested compounds reached the
range 6.9–9.4 (Figure 2). These differences can most likely be explained by
the differences in membrane potential between these organisms. Aerobic
organisms such as S. aureus generally have a higher membrane potential
(125–160 mV) in comparison to fermentative bacteria such as E. faecalis
(~100 mV) [41,42]. It is therefore not surprising that accumulation at the E.
faecalis membrane with its lower potential is not seen until higher
lipophilicities, as this is the driving force of alkylTPP cation uptake.
E. coli, a gram-negative bacterium with an outer membrane, showed
little or no accumulation of alkylTPP cations in response to lipophilicity, where
only compounds between logP values of 8.7–9.4 elicited toxicity at greater
than or equal to 128 ug/mL (Figure 2). It may be that alkylTPP cations still
follow this structure-function relationship in E. coli, but at higher logP values in
comparison to other bacteria (128 μg/mL vs. 1–8 μg /mL in M. tuberculosis).
The lack of permeability of alkylTPP cations to the E. coli membrane has been
demonstrated by Hirotia et al [43], where it was shown that E. coli mutants
defective in lipopolysaccharide (LPS) were able to accumulate the alkylTPP
cation membrane probe [3H]methyltriphenylphosphonium iodide to a greater
concentration than wild-type. This suggests that LPS is responsible for the

181
182
183
inability of alkylTPP cations to permeate the gram-negative cell wall at toxic
concentrations. It is
184
185
2.3.Effect of alkylTPP-phenothiazine derivative conjugates against
mycobacteria.
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
We next chose to investigate the use of alkylTPP cations to localize a
series of phenothiazine derivatives to their proposed target (NDH-2) in the
mycobacterial membrane, which is beneath the impermeant cell wall and
outer membrane [44-46].
Compound 1b was chosen as a model localizing system as it was shown
itself to be non-toxic to mycobacteria within the range of our assay (Table 1),
whilst still being able to accumulate at the mycobacterial membrane.
Compound 1b was conjugated to a series of phenothiazine derivatives (2a–
2c) to give three alkylTPP cation-phenothiazine derivative conjugates (3a–3c)
(Figure 1). The alkylTPP cation alone (1b) and phenothiazine derivatives
alone (2a–2c) showed no toxicity to either M. smegmatis or M. tuberculosis
within the range of our assay (Table 1). The known phenothiazine, TPZ,
inhibited the growth of M. smegmatis and M. tuberculosis at MICs of 64 μg/mL
and 8–16 μg/mL respectively in accordance with the published literature [47].
The alkylTPP-phenothiazine derivative conjugates (3a–3c) showed a
dramatic increase in toxicity against both M. smegmatis and M. tuberculosis
(Table 1). Compound 3a had an MIC value of 0.5 μg/mL against M.
tuberculosis; a 130-fold increase in potency in comparison to the
phenothiazine derivative alone and a 16–32-fold increase in potency when
compared with the known phenothiazine, TPZ (Table 1). Compound 3a is
more potent than other published phenothiazines, such as THR (MIC = 8–32
μg/mL) and CPZ (MIC= 12 μg/mL) [48,49].
The addition of the alkylTPP cation (1b) to the phenothiazine derivative (2a)
to give the conjugate 3a, increases its lipophilicity (logP of 3.72–8.39, Figure
1) and charge. To see if the increase in potency seen in 3a–3c was merely a
consequence of being either highly lipid-soluble or charged we deconstructed
their effects by testing the toxicity of a lipophilic (2d, logP 8.94), and of a

213
214
215
216
217
218
219
220
221
charged (2e) derivative of the phenothiazine (2a) against mycobacteria. Both
the lipophilic (2d) and charged (2e) derivatives showed negligible toxicity
against M. tuberculosis with >64 μg/mL and 128 μg/mL being observed
respectively (Table 1). This indicates that the mycobacterial growth inhibition
seen in the presence of the alkylTPP cation-phenothiazine derivative
conjugates is not the result of a non-specific increase in charge or lipophilicity
but rather the presence of the alkylTPP cation greatly improves the ability of
the phenothiazine derivative to accumulate within the mycobacteria and
adsorb to the surface of the membrane.
222
223
2.4. The mode of action of alkylTPP cation-phenothiazine
derivative conjugates in M. smegmatis.
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
Phenothiazines such as TPZ, CPZ and THR, are proposed to inhibit M.
tuberculosis NDH-2 in the electron transport chain, leading to subsequent
bactericidal cell death [10,11,13,50]. The phenothiazine THR has also been
shown to inhibit NADH oxidation activity in M. smegmatis membrane vesicles,
where all NADH oxidation activity is attributed to NDH-2 [15,50]. We
hypothesised that our lead alkylTPP cation-phenothiazine derivative
conjugate, 3a, was also targeting NDH-2 in the mycobacterial membrane.
To test this hypothesis, a series of experiments were conducted in M.
smegmatis whole cells and membrane vesicles. NDH-2 from M. smegmatis
(ndh MSMEG_3621) shares 81% identity (NCBI Blast) with NDH-2 from M.
tuberculosis (ndh Rv1854c). Based on the MIC of 3a against M. smegmatis (4
μg/mL, Table 1), the exact concentration required to inhibit an actively
growing culture was determined to be 4.8 μg/mL per OD600 and this was used
in subsequent experiments. Compound 3a was found to be bacteriostatic
against M. smegmatis at 4.8 μg/mL per OD600 (Figure 3). A study conducted
by Selvakumar et al [51] showed that the phenothiazine derivatives CPZ and
promethazine (PTZ) had MIC values of 10 and 20 μg/mL against M.
tuberculosis respectively, however these agents did not show measurable
bactericidal activity below 50 μg/mL [52]. Similarly, Molnár et al [52], show that
the bacteriostatic or bactericidal action of TPZ is concentration dependent
against M. tuberculosis with 8 μg/mL being bacteriostatic and 16 μg/mL

245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
bactericidal. These observations may explain the bacteriostatic action of 3a
seen in this study, where an increase in concentration may lead to a
bactericidal effect i.e. that bacterial killing by this compound may be
concentration dependent. Alternatively, phenothiazine derivatives may have a
different mechanism of action in M. smegmatis versus M. tuberculosis,
however phenothiazines have been shown to inhibit oxygen consumption
(suggesting inhibition of the electron transport chain) in both M. smegmatis
and M. tuberculosis[50]. The alkylTPP cation (1b) and phenothiazine
derivative (2a) controls did not affect M. smegmatis viability or growth. The
ionophore, CCCP, was shown to be bactericidal, as previously reported[53].
M. smegmatis contains copies of both the proton translocating type I NADH
dehydrogenase (NDH-1; nuoA-nuoN, MSMEG_2050-MSMEG_ 2063) and
NDH-2. In M. smegmatis membranes NADH oxidation occurs exclusively
through NDH-2 as no inhibition of NADH oxidation is seen in the presence of
the NDH-1 specific inhibitor, rotenone (Figure 4). This allowed us to test our
compounds for specific inhibition of NDH-2 meditated NADH oxidation.
Compound 3a was shown to inhibit NADH oxidation in M. smegmatis
262 membranes (Figure 4) at an IC50 of 15.14 μM; 4-fold lower than that of TPZ
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
(60.07 μM). The alkylTPP cation alone (1b), phenothiazine derivative alone
(2a) and DMSO controls had no significant effect on NADH oxidation relative
to the no compound control (two-tailed t-test, p<0.05, 95% confidence).
Complete inhibition of oxygen consumption by 3a was observed at 40
μM, when NADH was used as the electron donor (Figure 5). This was
significantly lower than the concentration of TPZ required to inhibit oxygen
consumption (100–200 μM). The phenothiazine derivative only (2a) and
DMSO controls did not significantly affect oxygen consumption, however the
alkylTPP cation (1b) control showed a significant increase relative to the no
compound control (two-tailed t-test, p<0.05, 95% confidence). This was not
surprising as at high concentrations the addition of the positive alkylTPP
cation (1b) will dissipate the membrane potential. This occurs by movement of
a positive charge into the membrane, where the cell responds to this by
stimulating respiration to re-establish the membrane potential. This has been
observed previously with other compounds such as Ru(CO)3Cl(Glycinate),
which affect ion transport and the proton motive force (PMF) [54].

279
280
281
282
283
284
285
286
287
288
289
As expected the known inhibitor of respiration, potassium cyanide (KCN)[55],
showed complete inhibition of oxygen consumption. The inhibition of both
NADH oxidation and oxygen consumption when NADH is the electron donor
indicates that 3a targets energetic processes in the M. smegmatis electron
transport chain. Our data is in concordance with experiments conducted by
Boshoff et al[50], where the phenothiazine THR (135 μM) was shown to inhibit
NADH oxidation activity in M. smegmatis membranes. Thioridazine also
reduced oxygen consumption in both M. smegmatis and M. tuberculosis
whole cells[50]. From these data, we propose that 3a targets NDH-2 in the
mycobacterial membrane, leading to inhibition of oxygen consumption and
NADH oxidation.
290
2.5.Conclusion
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
This communication shows the ability of alkylTPP cations to accumulate at
biological membranes in a range of bacteria (M. smegmatis, M. tuberculosis,
S. aureus and E. faecalis), where lipophilicity was shown to be an important
feature of the structure-function relationship. The alkylTPP cation-
phenothiazine derivative conjugate, 3a, was shown to act in a similar manner
to previously published phenothiazines (e.g. THR) by targeting energetic
processes in the mycobacterial membrane i.e. NADH oxidation and oxygen
consumption. AlkylTPP cations significantly increased the potency of
phenothiazine derivatives beyond that of any previously published study. We
propose they localise the phenothiazine derivatives to the mycobacterial
membrane, thus increasing their effective concentration at the site of NDH-2.
Phenothiazines have previously been reported to accumulate within M.
tuberculosis infected macrophages at concentrations at least 10-times greater
than in serum (THR, TPZ and CFZ) [17,21,56]. Phenothiazines are reported
to target dopamine and serotonin receptors in the CNS. We propose that the
addition of the alkylTPP cation will improve the specificity of phenothiazines,
as their accumulation at biological membranes is in direct response to
membrane potential. The membrane potential of M. tuberculosis (~160 mV)
[57] is higher than that of neurons in the CNS (resting potential -70 mV) [58].
Based on this we would propose that the alkylTPP cation delivery molecule

311
312
313
314
315
316
and the attached phenothiazine would have preference for uptake in bacterial
cells versus cells such as neurons which house the receptors associated with
off target effects. Together, these findings demonstrate the enormous
potential of alkylTPP cations to improve the localization of phenothiazine
derivatives to the mycobacterial membrane in an in vivo setting, targeting
energetic processes that are essential for the survival of mycobacteria.
317
318
3. METHODS
3.1.Chemicals.
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
All chemicals were obtained from Sigma Aldrich unless stated otherwise.
The structures of the compounds tested in biological assays are displayed in
Figure 1 and were either synthesized as part of the current study or
purchased from Sigma Aldrich. Stock solutions of the alkylTPP cations
(methyltriphenylphosphonium iodide (1a), propyltriphenylphosphonium
bromide
pentyltriphenylphosphonium bromide (1d), heptyltriphenylphosphonium
bromide (1f), dodecyltriphenylphosphonium bromide (1k),
(1b),
butyltriphenylphosphonium
bromide
(1c),
hexadecyltriphenylphosphonium bromide (1l)), phenothiazine derivatives
(phenothiazine (2a), 2-chloro-phenothiazine (2b), 2-trifluoro-phenothiazine
(2c), 10-hexyl-10H-phenothiazine (2d), promazine hydrochloride (2e)) and
alkylTPP cation-phenothiazine derivative conjugates (3-(10H-Phenothiazin-
10-yl)propyl
triphenylphosphonium
chloride
(3a),
3-(2-chloro-10H-
3-(2-
phenothiazin-10-yl)propyl triphenylphosphonium
iodide (3b),
trifluoromethyl-10H-phenothiazin-10-yl)propyl triphenylphosphonium iodide
(3c)) were prepared in dimethyl sulfoxide (DMSO; Thermo Fisher) and TPZ
was prepared in distilled water. All stock solutions were stored at –20°C.
336
337
3.2.Predication of partition coefficient (logP) and statistical
analysis.
338
339
340
341
Partition coefficients were predicted using the virtual computational
chemistry laboratory (VCCLAB) software ALOGPS[59] (Figure 1). Statistical
analysis was conducted in GraphPad Prism 6 (two-tailed t-test, p<0.05, 95%
confidence).

342
3.3.Chemical synthesis.
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
Acetonitrile was dried using the Pure Solv MD-7 solvent purification system
and stored over 4 Å molecular sieves. All other solvents and reagents were
used as received. All aqueous solutions used in work-up procedures were
saturated and all organic extracts were dried with magnesium sulphate,
filtered and concentrated in vacuo on a rotary evaporator. Thin layer
chromatography (TLC) was performed on Merck (Darmstadt, Germany)
aluminium-backed silica gel 60 (0.20 mm) plates. Column chromatography
was performed using silica gel 60 (230–400 mesh).
1H, and 13C NMR spectra were recorded on a Varian 500 MHz
Spectrometer (500 MHz for 1H, and 125 MHz for 13C). All chemical shifts are
reported relative to the residual CHCl3 solvent peaks (1H, 7.26 ppm and 13C,
77.16 ppm) according to the δ scale. Chemical shifts are rounded to 0.01
ppm and coupling constants rounded to 0.1 Hz. Multiplicity abbreviations are
reported by the following conventions: s (singlet), d (doublet), t (triplet), m
(multiplet) and dd (doublet of doublets). High-resolution mass-spectrometry
(HR-MS) spectra were recorded on a Bruker microTOFQ mass spectrometer
using an electrospray ionisation (ESI) source in positive mode. HPLC analysis
was carried out using an Agilent HP1100 at 25°C on a C18 column
(Phenomenex Prodigy ODS(3) 5 μm 100 Å 250 x 3 mm) with a 2 x 4 mm C18
guard column. Peaks were detected at 210 and 254 nm and UV spectra
recorded from 190 to 600 nm. The mobile phase was acetonitrile in water
containing 0.1% (v/v) trifluoroacetic acid: t0=10%, t12.5=100%, t15=100%,
t16=10%, t20=10%. The flow rate was 0.5 mL/min, with an injection volume of
3 μL.
Hexyltriphenylphosphonium bromide (1e), octyltriphenylphosphonium bromide
(1g), nonyltriphenylphosphonium bromide (1h), decyltriphenylphosphonium
bromide
(1i),
undecyltriphenylphosphonium
bromide (1l)
bromide
(1j),
hexadecyltriphenylphosphonium
and
10-dodecyl-10H-
phenothiazine (2d) were synthesized and characterized as reported[31-36].
The novel synthesis of 3a–3c is outlined in Scheme 1.

374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
3-(10H-Phenothiazin-10-yl)propyl triphenylphosphonium chloride (3a).
The intermediate known 10-(3-chloropropyl)-10H-phenothiazine (2a’)[60] was
prepared from phenothiazine (2a; 500 mg, 2.51 mmol) and 1-bromo-3-
chloropropane (300 μL, 3.01 mmol) using previously described methods (yield
576.3 mg, 83.3 %)[61]. A solution of 2a’ (400 mg, 1.525 mmol), sodium iodide
(343 mg, 2.288 mmol) and triphenylphosphine (463 mg, 1.678 mmol) in
anhydrous acetonitrile (30 mL) was heated to 80ºC for 24 hours with stirring.
The NaCl precipitate was removed by filtration and was washed with
acetonitrile. The solvent was removed in vacuo, the solid dissolved in
dichloromethane and washed with saturated brine (5 x 200 mL). The organic
layer was dried over magnesium sulphate and concentrated in vacuo to give a
white solid. This was dissolved in minimal acetonitrile and precipitated in ether
to give 3a (623 mg, 76 %) as a white solid. HPLC 12.39 mins 100% pure.
1
HRMS m/z found: 502.1720 calcd. For C33H29NPS+ 502.1753. H NMR (500
MHz, CDCl3): δ 7.68–7.72 (3H, m, H4”) 7.61–7.66 (6H, m, H2”) 7.53–7.57
(6H, m, H3”) 7.11–7.17 (4H, m, H2, H4, H6, H8) 6.93–6.96 (4H, m, H1, H3,
H7, H9) 4.35 (2H, m, H3’) 3.85 (2H, m, H1’) 2.19 (2H, m, H2’). 13C NMR (125
MHz, CDCl3): δ 145.11 (C9a, C10a) 135.00 (3C, d, JC4P = 3 Hz, C4”) 133.63
(6C, d, JC2P = 10 Hz, C2”) 130.49 (6C, d, JC3P = 12 Hz, C3”) 128.00 (C2, C8)
127.7 (C4, C6) 125.85 (C4a, C5a) 123.18 (C3, C7) 118.18 (3C, d, JCP = 86
Hz, C1”) 116.75 (C1, C9) 46.37 (1C, d, JC3P = 18 Hz, C3’) 20.26 (1C, d, JC2P
3 Hz, C2’) 20.15 (1C, d, JCP = 52 Hz, C1’).
=
3-(2-chloro-10H-phenothiazin-10-yl)propyl
triphenylphosphonium
chloride (3b). The known intermediate 10-(3-chloropropyl)-2-chloro-10H-
phenothiazine (2b’)[61] was prepared from 2-chloro-phenothiazine (2b; 500
mg, 2.14 mmol) and 1-bromo-3-chloropropane (254 μL, 2.57 mmol) using
previously described methods (yield 429.0 mg, 66.7 %)[61]. A solution of 2b’
(429 mg, 1.636 mmol), sodium iodide (368 mg, 2.453 mmol) and
triphenylphosphine (558 mg, 1.799 mmol) in anhydrous acetonitrile (30 mL)
was heated to 80ºC for 24 hours with stirring. The NaCl precipitate was
removed by filtration and washed with acetonitrile. The solvent was removed
in vacuo, the solid dissolved in dichloromethane and washed with saturated
brine (5 x 200 mL). The organic layer was dried over magnesium sulphate

408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
and concentrated in vacuo to give a white solid. This was dissolved in minimal
acetonitrile and precipitated in ether to give 3b (213 mg, 23 %) as a white
solid. HPLC 12.89 mins 100% pure. HRMS m/z found: 536.1323 calcd. For
1
C33H28ClNPS+ 536.1363. H NMR (500 MHz, CDCl3): δ 7.65–7.74 (9H, m,
H4”, H2”) 7.56–7.60 (6H, m, H3”) 7.18 (1H, dt, J = 2, 7 Hz, H8) 7.15 (1H, dd, J
= 2, 8 Hz, H6) 7.04 (1H, d, J = 8 Hz, H4) 7.01 (1H, dd, J = 1, 8 Hz, H9) 6.97
(1H, dt, J = 1, 7 Hz, H7) 6.90 (1H, dd, J = 2, 8 Hz, H3) 6.80 (1H, d, J = 2 Hz,
H1) 4.33 (2H, m, H3’) 3.86 (2H, m, H1’) 2.20 (2H, m, H2’). 13C NMR (125
MHz, CDCl3): δ 146.20 (C10a) 144.64 (C9a) 135.11 (3C, d, JC4P = 3 Hz, C4”)
133.86 (C2) 133.66 (6C, d, JC2P = 10 Hz, C2”) 130.53 (6C, d, JC3P = 13 Hz,
C3”) 128.40 (C8) 128.23 (C4)127.73 (C6) 125.34 (C4a*) 124.53 (C5a*)
123.70 (C7) 123.08 (C3) 118.11 (3C, d, JCP = 86 Hz, C1”) 117.22 (C9) 116.81
(C1) 46.50 (1C, d, JC3P = 18 Hz, C3’) 20.36 (1C, d, JC1P = 53 Hz, C1’) 20.16
(1C, d, JC2P = 2 Hz, C2’). *Assignments may be interchanged.
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
3-(2-trifluoromethyl-10H-phenothiazin-10-yl)propyl
triphenylphosphonium
chloride (3c). The known intermediate 10-(3-
chloropropyl)-2-(trifluoromethyl)-10H-phenothiazine (2c’)[62] was prepared
from 2-trifluro-phenothiazine (2c; 500 mg, 1.87 mmol) and 1-bromo-3-
chloropropane (222 μL, 2.25 mmol) using previously described methods (yield
429.0 mg, 66.7 %)[61]. A solution of 2c’ (463.9 mg, 1.765 mmol), sodium
iodide (397 mg, 2.65 mmol) and triphenylphosphine (668 mg, 1.942 mmol) in
anhydrous acetonitrile (30 mL) was heated to 80ºC for 24 hours with stirring.
The NaCl precipitate was removed by filtration and was washed with
acetonitrile. The solvent was removed in vacuo, the solid dissolved in
dichloromethane and washed with saturated brine (5 x 200 mL). The organic
layer was dried over magnesium sulphate and concentrated in vacuo to give a
white solid. This was dissolved in minimal acetonitrile and precipitated in ether
to give the product as a white solid (54%, 507.8 mg). HPLC 13.06 mins 100%
pure, HRMS m/z found: 570.1621 calc. For C34H28F3NPS+ 570.1627; 1H NMR
(500 MHz, CDCl3): δ 7.66–7.73 (9H, m) 7.55–7.59 (6H, m) 7.19–7.22 (2H, t, J
= 7 Hz) 7.14–7.16 (2H, m) 7.08 (1H, d, J = 8 Hz) 6.99 (1H, t, J = 7 Hz,) 6.95
(1H, s) 4.40 (2H, t, J = 5 Hz) 3.89 (2H, m) 2.19 (2H, m): 13C NMR (125 MHz,

441
442
443
444
445
CDCl3): δ 145.34, 144.49, 135.12 (3C, d, J = 3 Hz), 133.62 (6C, d, J = 10 Hz),
130.95 (q, J = 1 Hz), 130.55 (6C, d, J = 13 Hz), 130.04 (q, J = 33 Hz), 128.68,
127.88, 127.75, 124.00 (q, J = 272 Hz), 124.41, 123.92, 119.71 (q, J = 4 Hz),
118.06 (d, J = 86 Hz) 117.38, 112.72, 46.56 (d, J = 18 Hz), 20.47 (d, J = 53
Hz), 20.05 (d, J = 3 Hz).
446
3.4.Bacterial strains and growth conditions.
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
M. smegmatis mc2155 [63] was routinely grown with agitation (200
rpm) in Lysogeny broth (LB) medium supplemented with 0.05% (w/v)
Tween80 (LBT), or maintained on LBT agar (1.5% w/v) at 37°C. M.
tuberculosis H37Ra [64] was grown with agitation (200 rpm) in 7H9 medium
(BD & Company) supplemented with 0.5% (w/v) glycerol, 10% albumin
dextrose catalase (ADC; Middlebrook) and 0.05% (w/v) Tween80 (Fischer
Scientific) or on 7H11 agar (BD & Company) supplemented with 0.5% (w/v)
glycerol, 10% oleic albumin dextrose (OAD) and 0.05% (w/v) Tween80 at
37°C, 5% CO2. S. aureus BB255 [65] and E. coli DH5α [66] were grown with
agitation (200 rpm) in LB or on LB agar (1.5% w/v) at 37°C. E. faecalis JH2-2
[67] was grown without agitation in Brain heart infusion (BHI) broth (BD &
Company) or on BHI agar (1.5% w/v) at 37°C. Optical density was measured
at 600 nm (OD600) using a Jenway 6300 spectrophotometer. Cultures were
diluted in phosphate-buffered saline (PBS, pH 7.0) to ensure that OD600 was
below 0.5 when measured in cuvettes of 1 cm path length.

462
463
3.5.Minimum inhibitory concentration (MIC) assays against
mycobacteria.
464
465
466
467
468
469
470
471
The toxicity of all compounds (Figure 1) was measured against M.
smegmatis mc2155 and M. tuberculosis H37Ra by serial macrobroth dilution
from 128–0.5 μg/mL in LBT and 7H9 medium respectively. Log phase cultures
of each strain were used to inoculate each dilution to a final OD600 of 0.005.
Cultures were incubated at 37°C, 200 rpm, where the visual presence or
absence of growth was recorded after 24 hours for M. smegmatis and 5 days
for M. tuberculosis. Controls containing no compound and the compound
solvent (DMSO) were also included.
472
473
3.6.Minimum inhibitory concentration (MIC) assays against fast
growing bacterial strains.
474
475
476
477
478
479
480
481
The toxicity of all compounds (Figure 1) was measured against E. coli
DH5α and S. aureus BB255 by serial macrobroth dilution form 128–0.5 μg/mL
in LBT. The visual presence or absence of growth was recorded after 12
hours. The toxicity of the described compounds against E. faecalis JH2-2 was
determined by serial macrobroth dilution in BHI medium supplemented with
0.05% Tween80 with no agitation at 37°C. The visual presence or absence of
growth was recorded after 12 hours. Controls containing no compound and
the compound solvent (DMSO) were also included.
482

483
3.7.Cell viability assays.
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
M. smegmatis cultures at mid log phase (OD600 ~0.5) were challenged with
3a at concentrations around the MIC (1–4 μg/mL), with growth monitored by
OD600 over 24 hours. This was to determine the concentration required to
inhibit the growth of a culture at an OD600 of 1 for further experiments. To
determine if 3a was bacteriostatic or bactericidal against M. smegmatis,
viability studies were conducted over a range of concentrations (1–4 μg/mL)
surrounding the determined MIC. Cultures of M. smegmatis were grown to an
OD600 of approximately 0.5 after which 3a (8.9 μM per OD600), 1b (8.9 μM per
OD600), 2a (8.9 μM per OD600) and the no compound and solvent (DMSO)
controls were added. CCCP (500 μM) [15] was also included as a bactericidal
control. Growth was monitored by OD600 at 0, 3, 9 and 24 hour timepoints.
Cell viability at 0, 9 and 24 hours was determined using colony forming unit
(CFU) counts, where cultures were washed and diluted in sterile PBS
supplemented with 0.05% Tween80 and plated onto LBT in 5 μL aliquots as
per the Miles and Misra protocol [68]. These were incubated at 37°C for 3
days after which the number of viable colonies were counted, with countable
plates defined as those containing 5–50 colonies.

502
3.8.Preparation of M. smegmatis membrane vesicles.
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
M. smegmatis cells were grown to an OD600 of 0.7–1 (log phase) and
harvested by centrifugation (8, 000 × g, 10 minutes, 4°C). Cell pellets were re-
suspended at 4°C in 8 mL per g/cells of lysis buffer A (50 mM MOPS pH 7.5,
1 mM PMSF, 1 mM DTT (Roche Diagnostics), 5 mM MgCl2, 0.05% (w/v)
Tween80, 2.5 mg/mL lysozyme (Roche Diagnostics), 0.1mg/mL DNase 1
(Roche Diagnostics)) using a homogenizer, and incubated for 30 minutes.
The mixture was then passed 4–6 times through a French pressure cell at 20,
000 p.s.i. until the cell lysate was translucent, followed by centrifugation (8,
000 × g, 10 minutes, 4°C) to remove unbroken cells. The membrane fraction
was isolated from the clarified cell lysate by ultracentrifugation (100, 000 × g,
1 hour, 4°C), washed with lysis buffer B (50 mM MOPS pH 7.5, 1 mM PMSF,
1 mM DTT, 5 mM MgCl2, 0.05% (w/v) Tween80), followed by further
ultracentrifugation (100, 000 × g, 1 hour, 4°C). Membranes were re-
suspended in 1–2 mL lysis buffer B by passage through a hypodermic syringe
fitted with a 22 G needle, followed by snap freezing on dry ice (50 μL aliquots)
and storage at –80°C.
519
3.9.NADH oxidation assays.
520
521
522
523
524
525
526
527
528
529
530
531
532
533
The NADH:quinone oxidoreductase activity of M. smegmatis membranes
was monitored spectrophotometrically at 37°C by following the oxidation of
NADH at 340 nm (Cary 50 Probe UV/Vis Spectrophotometer, Varian). M.
smegmatis membranes (50 μg/mL) and buffer (20 mM sodium phosphate, pH
8.0) were added to a 1 cm path length quartz cuvette and incubated for 30
seconds prior to the reaction being initiated by NADH (200 μM; Roche
Diagnostics), in a total assay volume of 1 mL. The extinction coefficient of
6.22 mM-1 cm-1 was used to calculate NADH concentration and NADH
dehydrogenase activity was expressed as μmol NADH/min/mg protein. The
effect of 3a on NADH oxidation was measured over a range of concentrations
(0–200 μM) by addition to the cuvette at the incubation step. Compounds 1b
(100 μM), 2a (100 μM), rotenone (30 μM), TPZ (0–200 μM) and the
compound solvent, DMSO, were included as controls. The IC50 values for 3a
and TPZ were determined in GraphPad Prism 6 by normalizing the data to the

534
535
536
percentage of total NADH oxidation activity (i.e. 0% inhibitor) and log
transforming the data. The non-linear function, log(inhibitor) vs. response, was
then used to calculate the IC50 values.
537
3.10. Oxygen consumption assays.
538
539
540
541
542
543
544
545
546
547
Respiration in M. smegmatis membranes was measured by monitoring the
change in dissolved oxygen tension with a Rank Bros Clark-type oxygen
electrode in a 2 mL reaction mixture at 37°C as previously described by de
Melo et al [69]. The electrode was calibrated to 220 nmol dissolved O2/mL
using air saturated 20 mM sodium phosphate buffer (pH 8.0) and pure
nitrogen gas at 37°C. Oxygen consumption in M. smegmatis membranes
stimulated with 10 mM NADH was measured in the presence of 3a (0–80 μM)
and TPZ (0–200 μM), in addition to 1b (80 μM), 2a (80 μM), no compound,
solvent (DMSO) and potassium cyanide (200 μM) controls.
548
4. ACKNOWLEDGEMENTS
549
550
551
552
553
554
555
This research was funded by the Health Research Council of New Zealand
and Lottery Health. E. Dunn was funded by PhD scholarships from the
University of Otago, The Freemason Charity, The Livestock Improvement
Corporation and The Edith Wheal Trust and. G. M. Cook is supported by a
James Cook Fellowship from the Royal Society of New Zealand. We thank C.
Porteous, A. Dunn and A. Rahman for their technical assistance.
556
5. REFERENCES
557
558
559
560
561
562
563
564
565
566
[1]
[2]
A. Zumla, A. George, V. Sharma, N. Herbert, I. Baroness Masham of,
WHO's 2013 global report on tuberculosis: successes, threats, and
opportunities, Lancet. 382 (2013) 1765–1767. doi:10.1016/S0140-
6736(13)62078-4.
M. Gengenbacher, S.P. Rao, K. Pethe, T. Dick, Nutrient-starved, non-
replicating Mycobacterium tuberculosis requires respiration, ATP
synthase and isocitrate lyase for maintenance of ATP homeostasis
and viability, Microbiology. 156 (2010) 81–87.
doi:10.1099/mic.0.033084-0.
J.G. Hurdle, A.J. O'Neill, I. Chopra, R.E. Lee, Targeting bacterial
[3]

567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
membrane function: an underexploited mechanism for treating
persistent infections, Nat Rev Micro. 9 (2011) 62–75.
doi:10.1038/nrmicro2474.
A. Koul, N. Dendouga, K. Vergauwen, B. Molenberghs, L. Vranckx, R.
Willebrords, et al., Diarylquinolines target subunit c of mycobacterial
ATP synthase, Nat. Chem. Biol. 3 (2007) 323–324.
[4]
[5]
[6]
[7]
doi:10.1038/nchembio884.
K.A. Abrahams, J.A.G. Cox, V.L. Spivey, N.J. Loman, M.J. Pallen, C.
Constantinidou, et al., Identification of novel imidazo[1,2-a]pyridine
inhibitors targeting M. tuberculosis QcrB, PLoS ONE. 7 (2012).
doi:10.1371/journal.pone.0052951.
K. Bjorklof, V. Zickermann, M. Finel, Purification of the 45 kDa,
membrane bound NADH dehydrogenase of Escherichia coli (NDH-2)
and analysis of its interaction with ubiquinone analogues, FEBS
Letters. 467 (2000) 105–110. doi:10.1016/s0014-5793(00)01130-3.
T. Yano, M. Rahimian, K.K. Aneja, N.M. Schechter, H. Rubin, C.P.
Scott, Mycobacterium tuberculosis type II NADH-menaquinone
oxidoreductase catalyzes electron transfer through a two-site ping-
pong mechanism and has two quinone-binding sites, Biochemistry. 53
(2014) 1179–1190. doi:10.1021/bi4013897.
J. Liu, T.A. Krulwich, D.B. Hicks, Purification of two putative type II
NADH dehydrogenases with different substrate specificities from
alkaliphilic Bacillus pseudofirmus OF4, Biochim Biophys Acta. 1777
(2008) 453–461. doi:10.1016/j.bbabio.2008.02.004.
A. Heikal, Y. Nakatani, E. Dunn, M.R. Weimar, C.L. Day, E.N. Baker,
et al., Structure of the bacterial type II NADH dehydrogenase: a
monotopic membrane protein with an essential role in energy
generation, Mol Microbiol. 91 (2014) 950–964.
[8]
[9]
doi:10.1111/mmi.12507.
[10]
[11]
T. Yano, L.S. Li, E. Weinstein, J.S. Teh, H. Rubin, Steady-state
kinetics and inhibitory action of antitubercular phenothiazines on
Mycobacterium tuberculosis type-II NADH-menaquinone
oxidoreductase (NDH-2), J Biol Chem. 281 (2006) 11456–11463.
doi:10.1074/jbc.M508844200.
E.A. Weinstein, T. Yano, L.S. Li, D. Avarbock, A. Avarbock, D. Helm,
et al., Inhibitors of type II NADH:menaquinone oxidoreductase
represent a class of antitubercular drugs, P Natl Acad Sci USA. 102
(2005) 4548–4553. doi:10.1073/pnas.0500469102.
C.M. Sassetti, D.H. Boyd, E.J. Rubin, Genes required for
mycobacterial growth defined by high density mutagenesis, Mol
Microbiol. 48 (2003) 77–84.
A.J. Warman, T.S. Rito, N.E. Fisher, D.M. Moss, N.G. Berry, P.M.
O'Neill, et al., Antitubercular pharmacodynamics of phenothiazines,
Journal of Antimicrobial Chemotherapy. 68 (2013) 869–880.
doi:10.1093/jac/dks483.
[12]
[13]
[14]
A.B. Bate, J.H. Kalin, E.M. Fooksman, E.L. Amorose, C.M. Price, H.M.
Williams, et al., Synthesis and antitubercular activity of quaternized
promazine and promethazine derivatives, Bioorganic & Medicinal
Chemistry Letters. 17 (2007) 1346–1348.
doi:10.1016/j.bmcl.2006.11.091.

617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
[15]
S.P. Rao, S. Alonso, L. Rand, T. Dick, K. Pethe, The protonmotive
force is required for maintaining ATP homeostasis and viability of
hypoxic, nonreplicating Mycobacterium tuberculosis, Proceedings of
the National Academy of Sciences of the United States of America.
105 (2008) 11945–11950. doi:10.1073/pnas.0711697105.
Z. Xie, N. Siddiqi, E.J. Rubin, Differential antibiotic susceptibilities of
starved Mycobacterium tuberculosis isolates, Antimicrobial Agents and
Chemotherapy. 49 (2005) 4778–4780. doi:10.1128/aac.49.11.4778-
4780.2005.
[16]
[17]
[18]
M.J. Advani, I. Siddiqui, P. Sharma, H. Reddy, Activity of
trifluoperazine against replicating, non-replicating and drug resistant
M. tuberculosis, PLoS ONE. 7 (2012) e44245.
doi:10.1371/journal.pone.0044245.
C. Lienhardt, A. Vernon, M.C. Raviglione, New drugs and new
regimens for the treatment of tuberculosis: review of the drug
development pipeline and implications for national programmes,
Current Opinion in Pulmonary Medicine. 16 (2010) 186–193.
doi:10.1097/MCP.0b013e328337580c.
D.V. Gadre, V. Talwar, H.C. Gupta, P.S. Murthy, Effect of
trifluoperazine, a potential drug for tuberculosis with psychotic
disorders, on the growth of clinical isolates of drug resistant
Mycobacterium tuberculosis, Int Clin Psychopharmacol. 13 (1998)
129–131. doi:10.1097/00004850-199805000-00006.
L. Amaral, J.E. Kristiansen, M. Viveiros, J. Atouguia, Activity of
phenothiazines against antibiotic-resistant Mycobacterium
tuberculosis: a review supporting further studies that may elucidate the
potential use of thioridazine as anti-tuberculosis therapy, J Antimicrob
Chemother. 47 (2001) 505–511.
D. Ordway, M. Viveiros, C. Leandro, R. Bettencourt, J. Almeida, M.
Martins, et al., Clinical concentrations of thioridazine kill intracellular
multidrug-resistant Mycobacterium tuberculosis, Antimicrobial Agents
and Chemotherapy. 47 (2003) 917–922. doi:10.1128/aac.47.3.917-
922.2003.
F.J.L.T.D. Veronique Dartois, Antiparasitic and antibacterial drug
discovery: from molecular targets to drug candidates, WILEY-VCH,
2009.
S. Sharma, A. Singh, Phenothiazines as anti-tubercular agents:
mechanistic insights and clinical implications, Expert Opin. Investig.
Drugs. 20 (2011) 1665–1676. doi:10.1517/13543784.2011.628657.
K. Andries, P. Verhasselt, J. Guillemont, H.W.H. Gohlmann, J.M.
Neefs, H. Winkler, et al., A diarylquinoline drug active on the ATP
synthase of Mycobacterium tuberculosis, Science. 307 (2005) 223–
227. doi:10.1126/science.1106753.
A. De Logu, V. Onnis, B. Saddi, C. Congiu, M.L. Schivo, M.T. Cocco,
Activity of a new class of isonicotinoylhydrazones used alone and in
combination with isoniazid, rifampicin, ethambutol, para-aminosalicylic
acid and clofazimine against Mycobacterium tuberculosis, J
Antimicrob Chemother. 49 (2002) 275–282. doi:10.1093/jac/49.2.275.
Q.F. Cheng, D. Lichtstein, P. Russell, J.S. Zigler, Use of a lipophilic
cation to monitor electrical membrane potential in the intact rat lens,
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]

667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
Invest Ophthalmol Vis Sci. 41 (2000) 482–487.
[27]
M.F. Ross, A. Filipovska, R.A.J. Smith, M.J. Gait, M.P. Murphy, Cell-
penetrating peptides do not cross mitochondrial membranes even
when conjugated to a lipophilic cation: evidence against direct
passage through phospholipid bilayers, Biochem. J. 383 (2004) 457–
468. doi:10.1042/bj20041095.
R.A.J. Smith, C.M. Porteous, A.M. Gane, M.P. Murphy, Delivery of
bioactive molecules to mitochondria in vivo, P Natl Acad Sci USA. 100
(2003) 5407–5412. doi:10.1073/pnas.0931245100.
C.M. Porteous, A. Logan, C. Evans, E.C. Ledgerwood, D.K. Menon, F.
Aigbirhio, et al., Rapid uptake of lipophilic triphenylphosphonium
cations by mitochondria in vivo following intravenous injection:
Implications for mitochondria-specific therapies and probes, BBA -
General Subjects. 1800 (2010) 1009–1017.
[28]
[29]
doi:10.1016/j.bbagen.2010.06.001.
[30]
[31]
[32]
R.A.J. Smith, M.P. Murphy, Animal and human studies with the
mitochondria-targeted antioxidant MitoQ, in: Y.H. Wei, C.R. Tzeng,
H.M. Lee (Eds.), Ann.NY Acad.Sci, Wiley-Blackwell, Malden, 2010:
pp. 96–103. doi:10.1111/j.1749-6632.2010.05627.x.
C.F. Hauser, M.A. Raymond, M.L. Miles, T.W. Brooks, G.B. Butler,
Wittig Synthesis .1. Use of Certain Aliphatic Aldehydes as
Intermediates in Synthesis of Olefins, J Org Chem. 28 (1963) 372–&.
doi:10.1021/Jo01037a023.
I. Bidd, D.W. Holdup, M.C. Whiting, Studies on the Synthesis of Linear
Aliphatic-Compounds .3. The Synthesis of Paraffins with Very Long
Chains, J. Chem. Soc., Perkin Trans. 1. (1987) 2455–2463.
doi:10.1039/P19870002455.
G. Ohloff, C. Vial, F. Naf, M. Pawlak, Stereoselective Syntheses of
Isomeric 5, 10-Pentadecadienals, Helv Chim Acta. 60 (1977) 1161–
1174. doi:10.1002/Hlca.19770600404.
[33]
[34]
[35]
A.N. Starratt, Synthesis of Erythro-5,6-Diacetoxyhexadecanoic Acid, a
Novel Fatty-Acid from Mosquito Eggs, Chem Phys Lipids. 16 (1976)
215–218. doi:10.1016/0009-3084(76)90028-1.
R.F. Keyes, W.M. Golebiewski, M. Cushman, Correlation of anti-HIV
potency with lipophilicity in a series of cosalane analogs having normal
alkenyl and phosphodiester chains as cholestane replacements, J.
Med. Chem. 39 (1996) 508–514. doi:10.1021/Jm950666h.
Y.S. Kang, P. Baglioni, H.J.D. Mcmanus, L. Kevan, Alkyl chain-length
effects on the photoionization of N-alkylphenothiazines and sulfonated
alkylphenothiazines in anionic alkyl sulfate and cationic
alkyltrimethylammonium bromide micelles, J Phys Chem-Us. 95
(1991) 7944–7947. doi:10.1021/j100173a071.
[36]
[37]
[38]
J. Asin-Cayuela, A.R.B. Manas, A.M. James, R.A.J. Smith, M.P.
Murphy, Fine-tuning the hydrophobicity of a mitochondria-targeted
antioxidant, FEBS Letters. 571 (2004) 9–16.
doi:10.1016/j.febslet.2004.06.045.
A. Kanazawa, T. Ikeda, T. Endo, Synthesis and antimicrobial activity
of dimethyl-substituted and trimethyl-substituted phosphonium salts
with alkyl chains of various lengths, Antimicrobial Agents and
Chemotherapy. 38 (1994) 945–952. doi:10.1128/aac.38.5.945.

717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
[39]
[40]
P.J. Brennan, H. Nikaido, The Envelope of Mycobacteria, Annu Rev
Biochem. 64 (1995) 29–63.
E. De Rossi, M. Branzoni, R. Cantoni, A. Milano, G. Riccardi, O.
Ciferri, mmr, a Mycobacterium tuberculosis gene conferring resistance
to small cationic dyes and inhibitors, Journal of Bacteriology. 180
(1998) 6068–6071.
W.E.J. Alborn, N.E. Allen, D.A. Preston, Daptomycin disrupts
membrane potential in growing Staphylococcus aureus, Antimicrobial
Agents and Chemotherapy. 35 (1991) 2282–2287.
A. Gálvez, M. Maqueda, M. Martínez-Bueno, E. Valdivia, Permeation
of bacterial cells, permeation of cytoplasmic and artificial membrane
vesicles, and channel formation on lipid bilayers by peptide antibiotic
AS-48, Journal of Bacteriology. 173 (1991) 886–892.
N. Hirota, S. Matsuura, N. Mochizuki, N. Mutoh, Y. Imae, Use of
lipophilic cation-permeable mutants for measurement of
transmembrane electrical potential in metabolizing cells of Escherichia
coli, Journal of Bacteriology. 148 (1981) 399–405.
M. Niederweis, O. Danilchanka, J. Huff, C. Hoffmann, H. Engelhardt,
Mycobacterial outer membranes: in search of proteins, Trends in
Microbiology. 18 (2010) 109–116. doi:10.1016/J.Tim.2009.12.005.
P.J. Brennan, Structure, function, and biogenesis of the cell wall of
Mycobacterium tuberculosis, Tuberculosis. 83 (2003) 91–97.
doi:10.1016/s1472-9792(02)00089-6.
H. Nikaido, Molecular basis of bacterial outer membrane permeability
revisited, Microbiology and Molecular Biology Reviews. 67 (2003) 593.
doi:10.1128/mmbr.67.4.593-656.2003.
D.V. Gadre, V. Talwar, In vitro susceptibility testing of Mycobacterium
tuberculosis strains to trifluoperazine, Journal of Chemotherapy. 11
(1999) 203–206.
J.L. Bourdon, Contribution to the study of the antibiotic properties of
chlorpromazine or 4560 RP, Annales De l'Institut Pasteur. 101 (1961)
876–886.
L. Amaral, J.E. Kristiansen, L.S. Abebe, W. Millett, Inhibition of the
respiration of multi-drug resistant clinical isolates of Mycobacterium
tuberculosis by thioridazine: potential use for initial therapy of freshly
diagnosed tuberculosis, J Antimicrob Chemother. 38 (1996) 1049–
1053.
H.I. Boshoff, T.G. Myers, B.R. Copp, M.R. McNeil, M.A. Wilson, C.E.3.
Barry, The transcriptional responses of Mycobacterium tuberculosis to
inhibitors of metabolism: novel insights into drug mechanisms of
action, J Biol Chem. 279 (2004) 40174–40184.
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
doi:10.1074/jbc.m406796200.
[51]
[52]
N. Selvakumar, V. Kumar, P.V. Krishnamurthy, R. Prabhakar, P.S.
Murthy, In vitro susceptibility of Mycobacterium tuberculosis to
trifluoperazine, Current Science. 73 (1997) 79–80.
J. Molnar, I. Beladi, I. Foldes, Studies on antituberculotic action of
some phenothiazine derivatives in vitro, Zentralblatt Fur Bakteriologie,
Parasitenkunde, Infektionskrankheiten Und Hygiene. Erste Abteilung
Originale. Reihe a: Medizinische Mikrobiologie Und Parasitologie. 239
(1977) 521–526.

767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
[53]
[54]
M. Rao, T.L. Streur, F.E. Aldwell, G.M. Cook, Intracellular pH
regulation by Mycobacterium smegmatis and Mycobacterium bovis
BCG, Microbiol-Uk. 147 (2001) 1017–1024.
J.L. Wilson, H.E. Jesse, B. Hughes, V. Lund, K. Naylor, K.S. Davidge,
et al., Ru (CO) 3Cl (Glycinate)(CORM-3): A carbon monoxide–
releasing molecule with broad-spectrum antimicrobial and
photosensitive activities against respiration and cation transport in
Escherichia coli, Antioxidants & Redox Signaling. 19 (2013) 497–509.
T. Yano, S. Kassovska-Bratinova, J.S. Teh, J. Winkler, K. Sullivan, A.
Isaacs, et al., Reduction of clofazimine by mycobacterial type 2 NADH:
quinone oxidoreductase. A pathway for the generation of bactericidal
levels of reactive oxygen species, J Biol Chem. 286 (2011) 10276–
10287. doi:10.1074/Jbc.M110.200501.
J. van Ingen, The broad-spectrum antimycobacterial activities of
phenothiazines, in vitro: somewhere in all of this there may be
patentable potentials, Recent Pat Antiinfect Drug Discov. 6 (2011)
104–109. doi:10.2174/157489111796064623.
Y. Zhang, H. Zhang, Z.H. Sun, Susceptibility of Mycobacterium
tuberculosis to weak acids, J Antimicrob Chemother. 52 (2003) 56–60.
doi:10.1093/jac/dkg287.
R.I. Herzog, T.R. Cummins, S.G. Waxman, Persistent TTX-resistant
Na+ current affects resting potential and response to depolarization in
simulated spinal sensory neurons, J. Neurophysiol. 86 (2001) 1351–
1364.
I.V. Tetko, J. Gasteiger, R. Todeschini, A. Mauri, D. Livingstone, P.
Ertl, et al., Virtual computational chemistry laboratory--design and
description, J Comput Aided Mol Des. 19 (2005) 453–463.
doi:10.1007/s10822-005-8694-y.
S.O. Winthrop, M.A. Davis, Sulfonium analogs of pharmacologically
active amines. The synthesis of ι-(10-phenothiazinyl)-alkyl-
dialkylsulfonium halides and 2'-(10-phenothiazinecarboxy)-
ethyldialkylsulfonium halides, Journal of the American Chemical ….
(1958).
M.W. Harrold, Y.A. Chang, R.A. Wallace, T. Farooqui, L.J. Wallace, N.
Uretsky, et al., Charged analogues of chlorpromazine as dopamine
antagonists, J. Med. Chem. 30 (1987) 1631–1635.
H.L. Yale, F. Sowinski, 4-{3-[10-(2-trifluoromethyl)-phenothiazinyl]-
propyl}-1-piperazine-ethanol and related compounds. II, J Am Chem
Soc. 82 (1960) 2039–2042. doi:10.1021/ja01493a047.
S.B. Snapper, R.E. Melton, S. Mustafa, T. Kieser, W.R. Jacobs,
Isolation and characterization of efficient plasmid transformation
mutants of Mycobacterium smegmatis, Mol Microbiol. 4 (1990) 1911–
1919.
[55]
[56]
[57]
[58]
[59]
[60]
[61]
[62]
[63]
[64]
[65]
R. Brosch, W.J. Philipp, E. Stavropoulos, M.J. Colston, S.T. Cole, S.V.
Gordon, Genomic analysis reveals variation between Mycobacterium
tuberculosis H37Rv and the attenuated M. tuberculosis H37Ra strain,
Infection and Immunity. 67 (1999) 5768–5774.
T.T. Luong, C.Y. Lee, Improved single-copy integration vectors for
Staphylococcus aureus, Journal of Microbiological Methods. 70 (2007)
186–190. doi:10.1016/j.mimet.2007.04.007.

817
818
819
820
821
822
823
824
825
826
827
828
[66]
[67]
S.G.N. Grant, J. Jessee, F.R. Bloom, D. Hanahan, Differential plasmid
rescue from transgenic mouse dnas into Escherichia coli methylation-
restriction mutants, P Natl Acad Sci USA. 87 (1990) 4645–4649.
A.E. Jacob, S.J. Hobbs, Conjugal transfer of plasmid-borne multiple
antibiotic-resistance in Streptococcus faecalis var. zymogenes,
Journal of Bacteriology. 117 (1974) 360–372.
A.A. Miles, S.S. Misra, J.O. Irwin, The estimation of the bactericidal
power of the blood, The Journal of Hygiene. 38 (1938) 732–749.
A. deMelo, G.M. Cook, R.J. Miles, R.K. Poole, Nisin stimulates oxygen
consumption by Staphylococcus aureus and Escherichia coli, Appl
Environ Microbiol. 62 (1996) 1831–1834.
[68]
[69]

829
LIST OF FIGURES
830
831
832
Figure 1 Structure and predicted partition coefficient (logP) values of
alkyltriphenylphosphonium (alkylTPP) cation and phenothiazine derivatives
used in the current study.
833
834
835
836
837
838
839
840
841
Figure 2 Structure-activity relationship between lipophilicity (logP) of
compounds 1a – 1l and membrane accumulation in a range of bacteria. An
increase in toxicity (minimum inhibitory concentration) correlates to effective
accumulation at the bacterial membrane. This relationship was measured in
M. tuberculosis, M. smegmatis, S. aureus and E. faecalis and E. coli over a
concentration of 0.5–128 μg/mL of each compound. Each point is the average
of three biological replicates, with variance reported as the standard error of
the mean. Note: concentrations plotted at 128 μg/mL are greater than or equal
to this value.
842
843
844
845
846
847
848
Figure 3 Determination of bacteriostatic/bactericidal action of compound 3a
against M. smegmatis. Growth rate of M. smegmatis monitored over 24 hours
in response to 3a and the no compound, 1b, 2a and carbonyl cyanide m-
chlorophenyl hydrazone (CCCP, bactericidal) controls. Viability, measured in
colony forming units, of M. smegmatis in the presence of 3a and the no
compound, 1b, 2a and CCCP controls. Variation between 3 biological
replicates is reported as the standard error of the mean.
849
850
851
852
853
854
855
856
Figure 4 Inhibition of NADH oxidation in M. smegmatis membranes in the
presence of trifluoperazine (TPZ) and 3a. The median inhibition concentration
(IC50) was determined in the presence of 3a and the known mycobacterial
NDH-2 inhibitor, TPZ. No compound, the compound solvent (DMSO), 1b, 2a
and rotenone (an NDH-1 inhibitor), were included as controls. Error bars
represent the standard error of the mean and data points are the mean of
three independent experiments. Statistical analysis was a two-tailed t-test,
p<0.05, 95% confidence.
857
858
859
Figure 5 Inhibition of oxygen consumption in M. smegmatis membranes in the
presence of trifluoperazine (TPZ) and 3a. No compound, the compound
solvent (DMSO), 1b, 2a and KCN (an inhibitor of respiration), were included

860
861
862
as controls. Error bars represent the standard error of the mean and data
points are the mean of three independent experiments. Statistical analysis
was a two-tailed t-test, p<0.05, 95% confidence.
863
864
865
Scheme
1
General
reaction
scheme
for
the
synthesis
of
alkyltriphenylphosphonium (alkylTPP) cation – phenothiazine derivative
conjugates (3a – 3c).
866
867
868

869
870

871
872