Design and synthesis of amphiphilic 2-hydroxybenzylphosphonium salts with antimicrobial and antitumor dual action

Article abstract of DOI:10.1016/j.bmcl.2020.127234

Here we report the synthesis and biological evaluation of a series of new 2-hydroxybenzylphosphonium salts (QPS) with antimicrobial and antitumor dual action. The most active compounds exhibit antimicrobial activity at a micromolar level against Gram-positive bacteria Sa (ATCC 209p and clinical isolates), Bc (1–2 μM) and fungi Tm and Ca, and induced no notable hemolysis at MIC. The change in nature of substituents of the same length led to a drastic change of biological activity. Self-assembly behavior of the octadecyl and oleyl derivatives was studied. QPS demonstrated self-assembly within the micromolar range with the formation of nanosized aggregates capable of the solubilizing hydrophobic probe. The synthesized phosphonium salts were tested for cytotoxicity. The most potent salt was active against on M?Hela cell line with IC₅₀ on the level of doxorubicin and good selectivity. According to the cytofluorimetry analysis, the salts induced mitochondria-dependent apoptosis.

Full text of DOI:10.1016/j.bmcl.2020.127234

Journal Pre-proofs
Design and Synthesis of Amphiphilic 2-Hydroxybenzylphosphonium Salts
with Antimicrobial and Antitumor Dual Action
Natalia V. Terekhova, Dmitry A. Tatarinov, Zukhra M. Shaihutdinova,
Tatiana N. Pashirova, Anna P. Lyubina, Alexandra D. Voloshina, Anastasiia
S. Sapunova, Lucia Ya. Zakharova, Vladimir F. Mironov
PII:
S0960-894X(20)30338-3
DOI:
Reference:
https://doi.org/10.1016/j.bmcl.2020.127234
BMCL 127234
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
12 March 2020
23 April 2020
28 April 2020
Please cite this article as: Terekhova, N.V., Tatarinov, D.A., Shaihutdinova, Z.M., Pashirova, T.N., Lyubina,
A.P., Voloshina, A.D., Sapunova, A.S., Ya. Zakharova, L., Mironov, V.F., Design and Synthesis of Amphiphilic
2
-Hydroxybenzylphosphonium Salts with Antimicrobial and Antitumor Dual Action, Bioorganic & Medicinal
Chemistry Letters (2020), doi: https://doi.org/10.1016/j.bmcl.2020.127234
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 Published by Elsevier Ltd.

Design and Synthesis of Amphiphilic 2-Hydroxybenzylphosphonium
Salts with Antimicrobial and Antitumor Dual Action
a
a
b
a
Natalia V. Terekhova , Dmitry A. Tatarinov* , Zukhra M. Shaihutdinova , Tatiana N. Pashirova ,
a
a
a
a
Anna P. Lyubina , Alexandra D. Voloshina , Anastasiia S. Sapunova , Lucia Ya. Zakharova ,
a
Vladimir F. Mironov
aArbuzov Institute of Organic and Physical Chemistry, FRC Kazan Scientific Center of RAS, Arbuzov Str. 8,
Kazan, 420088, Russian Federation
bKazan Federal University, Kremlevskaya str. 18, Kazan 420008, Russian Federation
Corresponding author: Dmitry A. Tatarinov, Email: datint@iopc.ru
ABSTRACT
Here we report the synthesis and biological evaluation of a series of new 2-
hydroxybenzylphosphonium salts (QPS) with antimicrobial and antitumor dual action. The most
active compounds exhibit antimicrobial activity at a micromolar level against Gram-positive bacteria
Sa (АТСС 209p and clinical isolates), Bc (1-2 μM) and fungi Tm and Ca, and induced no notable
hemolysis at MIC. The change in nature of substituents of the same length led to a drastic change of
biological activity. Self-assembly behavior of the octadecyl and oleyl derivatives was studied. QPS
demonstrated self-assembly within the micromolar range with the formation of nanosized aggregates
capable of the solubilizing hydrophobic probe. The synthesized phosphonium salts were tested for
cytotoxicity. The most potent salt was active against on M-Hela cell line with IC on the level of
5
0
doxorubicin and good selectivity. According to the cytofluorimetry analysis, the salts induced
mitochondria-dependent apoptosis.
KEYWORDS
Phosphonium salts, Grignard reagents, antimicrobial activity, cytotoxicity, phosphine oxides,
cyclization, critical micelle concentration, cytofluorimetry
Chemotherapy is one of the most common ways to treat cancer. However, the drugs used cause
numerous side effects weakening immune system and making patients vulnerable to bacterial and
fungal infections. Therefore, the discovery and design of antitumor agents that exhibit both antitumor
and antimicrobial activity are of a high importance.1,2 Especially it is challenging to overcome the
problem of multi-drug resistant (MDR) bacteria. It was noted that development of resistance in
3
microorganisms is complicated to cationic antimicrobial agents. Moreover conjugation of
4
antimicrobial agents with quaternary phosphonium salts (QPSs) can prevent efflux pump-mediated
5
6
resistance or even restore/enhance susceptibility to antifungal compounds. Natural and synthetic
phenols are known for their antitumor and antibacterial activity. Antimicrobial action of phenols is
proposed to be due to several mechanisms that involve alteration or deformation of the cytoplasmic

7
bacterial membrane, inhibition of efflux pumps, and inhibition of membrane ATPases. Thus,
conjugation of (poly)phenol fragment with phosphonium moiety is a promising way to design
8
biologically active compounds that show notable antimicrobial as well as mitochondria-dependent
biological activity8-10, in particular, antitumor action11-13 through the uncoupling of oxidative
phosphorylation in microbial cells.14 Such derivatives have the potential to show dual
antimicrobial/antitumor action and overcome antibiotic resistance. Thus, this work is aimed at the
design of biologically active compounds of dual antitumor/antimicrobial action by combining in a
single molecule phenol and phosphonium moieties (Figure 1).
bacterial membrane disruption
inhibition of efflux pumps
OH
inhibition of membrane ATPases
P+ Lipohilic cation
R1
R3
interaction with the negatively charged bacterial cell wall
R2
apoptosis induction preferable in cancer cells
Fig. 1. Design of compounds with antimicrobial/antitumor dual action.
Here we report the synthesis of a series of phosphonium salts with a functional phenolic group with
the various alkyl chain length and unsaturated substituents, their biological activity evaluation, and
its correlation to self-assembly behavior.
1
5
We applied the recently reported method to synthesize a new group of 2-hydroxyphenyl substituted
phosphonium salts. Starting phosphine oxide 2 was synthesized according to reported procedure16
1
7
with some changing to increase yield. We tried to transform oxaphosphole-2-oxide 1 to 2-chloro-
H-benzo[d][1,2]oxaphosphole-2-oxide 3 which supposed to be more active toward nucleophilic
3
Grignard reagents. Direct reaction with PCl or thionyl chloride appeared to be unsuitable and resulted
5
in either a mixture of different chlorinated products or was extremely time-consuming. Thus, to give
desired compound 3 oxaphosphole-2-oxide 1 underwent sequential ether cleavage through the
1
8
reaction with in situ generated bromotrimethylsilane (TMSBr) and aqueous hydrolysis of formed
silyl ester to give phosphonate 4.1 that was treated then with thionyl chloride to form
chlorophospholene 3 via phospholene 4.2. Compound 3 in the reaction with phenylmagnesium
bromide gave a quantitative conversion to phosphine oxide 2 within half of an hour (Scheme 1). The
phosphine oxide 2 was then transformed to quasiphosphonium compound 5 and the following
reaction with corresponding organomagnesium compounds led to phosphonium salts 6. In the first
go of the reaction with hexylmagnesium bromide proceeds with slight exothermic effect and
hexyldiphenylphosphine oxide 7 was isolated as a byproduct. Compounds 7 may form through the
nucleophilic substitution of the P-C bond.19,20 The byproducts formation was successfully avoided by
performing the reaction with a cooling water bath.

OH
SOCl2
O
P
c
O
c
a, b
O
O
O
P
OH
P
OEt
P
OH
OH
1
4.1
4.2
OH
O
O
g, h
d, e
f
Ph
Ph
O
P+
O
Cl-
Cl
P
Ph
Ph
3
2
5
OH
6
.1 R = C H ;
6.2 R = C H ;
7 15
6
13
R
6.3 R = C H ;
6.4 R = C H ;
8
17
10 21
+
-
P Ph Cl + RP(O)Ph
2
6.5 R = C H ;
6.6 R = C H ;
12
25
14 29
Ph
6.7 R = C H ;
6.8 R = C H ;
16
33
18 37
6.1-6.10
7
6
.9 R = C H ;
6.10 R = C H ;
1
8
35
6 11
Scheme 1. Synthesis of compounds (2) and (3) and (2-hydroxybenzyl)phosphonium chlorides 6. Reagents and conditions:
+
(a) TMSCl, NaBr; (b) H
2
O, – HCl; (c) SOCl
2
(neat), -HCl, -SO
2
, reflux, 10 hr; (d) 2PhMgBr, THF, (e) H
2
O, H ; (f) SOCl
2
,
o
O, H+
DCM, -SO
2
, -HCl (g) RMgX, THF 15-17 C, (h) H
2
In vitro antimicrobial evaluation of all synthesized compounds was performed according to liquid
serial dilutions in the culture growth medium.21,22 Stability of the compounds in biological media was
confirmed by mass-spectral analysis of the extract of phosphonium salt 6.8 from Muller-Hinton media
incubated for 24 hours at 37°C. The recorded spectrum contained only the molecular ion peak of
phosphonium cation.
The series of phosphonium salts containing alkyl substituent at the phosphorus atom were tested
against Gram-positive, Gram-negative bacteria and fungi (Table 1). The salts showed monotonous
increase in antimicrobial activity against Gram-positive strains with grow of alkyl chain length till
derivative bearing 12 carbon atoms with the best MIC at the value 1 μM for 6.4 and 6.5 and abrupt
drop for the 6.8 with 18 carbon atoms. None of the evaluated compounds was active against Gram-
negative strains and compound 6.5 showed fungicidal activity against the fungi Candida albicans and
Trichophyton mentagrophytes var. gypseum at the level of the reference drug ketoconazole.
Table 1. In vitro antimicrobial activity of alkyl(alkenyl)-2-hydroxybenzyldiphenyl phosphonium chlorides*
Compound
Sa
Bc
Tm
Ca
MIC/MBC, μM
MIC/MFC, μM
6
6
6
6
6
.1
.2
.3
.4
.5
18.9±1.7 37.8±3.1 37.8±3.0 151.5±11.8 606.1±46.3
18.3±1.5 18.3±1.6 36.6±2.9 146.5±11.2 586.2±46.1
NAf
NA
606.1±46.3
586.2±46.1
NA
NA
8.8±0.8
8.8±0.7
4.1±0.3
17.7±1.4
1.9±0.1
17.7±1.5
4.1±0.3
283.8±21.4
16.6±1.3
7.8±0.5
NA
283.8±21.4
16.6±1.3
7.8±0.5
NA
1.0±0.08
16.6±1.4
7.8±0.6
16.6±1.4
7.8±0.6
1.0±0.07 2.0±0.16 0.9±0.07
3.8±0.34

6
6
6
6
.6
.7
.8
.9
3.6±0.27 14.8±1.2 3.6±0.32
3.4±0.25 28.2±2.2 14.1±1.3
108±8.3 215±17.6 108±7.9
478±40.2
452±38.9
215±18.3
26.9±2.1
NA
478±36.8
56.6±4.5
NA
NA
NA
478±36.8
56.6±4.5
NA
NA
NA
NA
NA
3.4±0.25
609±58.6
2.4±0.2
6.7±0.4
NA
26.9±2.3
NA
108±8.5
NA
108±7.8
NA
108±8.5
NA
108±7.8
NA
6
.10
Ofloxacin
2.4±0.2
1.5±0.1
4.1±0.3
0.7±0.06
8.5±0.6
1.5±0.1
Ciprofloxacin
Ketoconazolole
0.7±0.05
7.3±0.5
16.6±1.4
7.3±0.5
16.6±1.4
Sa, Staphylococcus aureus; Bс, Bacillus сereus; Tm, Trichophyton mentagrophytes; Ca, Candida аlbicans
Average of three values measured; ± standard deviation (SD); - means non-active
*
Moreover, phosphonium salts 6.1-6.10 were tested and appeared to be active against S.aureus clinical
isolates with resistance both to Amoxicillin and Ciprofloxacin (CI-1) or only to Amoxicillin (CI-2)
(Table 2).
Table 2. Antimicrobial activity of alkyl(alkenyl)-2-hydroxybenzyldiphenyl phosphonium chlorides against antibiotic
resistant S.aureus Clinical Isolates *
Compound
CI-Sa1
CI-Sa2
MIC/MBC, μM
6
6
6
6
6
6
6
6
6
.1
.2
.3
.4
.5
.6
.7
.8
.9
75.6±5.2
36.6±2.9
35.4±2.4
8.2±0.7
2.0±0.12
7.2±0.55
28.2±2.2
-
302.4±21.5 37.8±3.2 151.2±11.3
146.4±10.5 73.2±5.1 73.2±5.3
141.6±9.6
32.8±2.5
8.0±0.62
28.8±2.3
-
17.7±1.4 35.4±2.5
8.2±0.6
4.0±0.3
16.4±1.4
16.0±1.2
7.2±0.57 28.8±2.2
14.1±1.1 112.8±8.8
-
-
-
13.6±0.9
-
217.6±17.4 13.6±1.1 217.6±16.8
6
.10
-
-
-
Ciprofloxacin 377.2±28.7 754.4±60.4
Amoxicillin 94.4±7.8 94.4±8.1
85.6±7.6 85.6±7.9
CI-Sa1and CI-Sa, 2 S.aureus Clinical Isolates
*
Average of three values measured; ± standard deviation (SD); - means non-active
Antimicrobial action of amphiphiles is dependent upon their self-assemble behavior within the rows
of even-numbered alkyl chain substituents.23,24 Thus, considering investigations made for ammonium
biocides25,26 where lead compounds of ammonium biocides characterized by certain interval of values
for CMC and introduction of oleyl substituent instead of the octadecyl led to an improvement in
2
5
bacteriostatic activity we decided to try and also obtain similar derivatives of phosphonium salts in

order to see any relation to self-assembly properties. To the best of our knowledge, they appeared to
be the first example of the phosphonium salt with unsaturated acid-derived substituent linked directly
to the phosphorus atom. Thus, salt 6.9 was tested for its antimicrobial activity (Table 1). Introduction
of unsaturated fragments instead of the saturated alkenyl chain of the same length led to the drastic
increase of the activity in 30-fold scale.
In order to compare the influence of the double bond in the short-chain substituted phosphonium salt
where micelle formation process plays less significant role than in derivatives with the straight chain
of 18 carbon atom, we synthesized phosphonium chloride 6.10 and tested its biological activity.
To our surprise, compound 6.10 bearing hex-5-enyl substituent appeared to be non-active toward
bacterial strains at concentration range below 600 μM. This fact allows us to assume that distanced
from phosphonium atom double bond itself does not lead to better biological activity. Nonetheless,
in comparison with its saturated analogue 6.1, the antimicrobial action dramatically changed that
showed notable influence of the structural changes.
The alteration of the cytoplasmatic membrane is a common way that determines biological action of
many antimicrobial agents. The state of the cell wall can be examined by the hydrophobic dye uptake
2
7
such as crystal violet (CV) from the medium. The CV uptake by S. аureus was 1.8% when not
treated with test compounds and counted as a negative control. CTAB was used as a reference
compound. In Figure 2 it can be seen that at the MIC concentrations (1-2 μM) the test compounds 6.4
and 6.5 almost did not affect the cytoplasmic membrane in comparison to CTAB and the percentage
CV uptake value was approximately two times lower at the same concentrations. In higher
concentrations, there is an enhanced CV uptake up to 24-82.7% for the test compounds. However,
the value is still lower than the same for CTAB (62.4 - 100%). The enhanced penetration power by
the CV dye implies the alteration in S. aureus membrane permeability. Hence, it can be assumed that
in the MIC range the antimicrobial action of the test compounds is not due to alteration in membrane
permeability unlike the surfactants and notable effect on the membrane can be seen only in high
concentration (4-16 µМ).

Fig. 2. CV uptake assay. The percentage of crystalline violet in the supernatant of Staphylococcus aureus after 30-minutes
incubation with compounds 6.4, 6.5 and CTAB. The optical density of the sample with dye, in the absence of cells, was
taken as 100%
The antimicrobial activity difference between compounds 6.8 and 6.9 may be explained by the
difference in self-assembly behavior of these compounds. The values of critical micelle
concentrations (CMC) were determined by tensiometry (Figure 3A), conductometry (ESI, Figure
S37) and spectrophotometry using hydrophobic dye Orange OT solubilization technique (Figure 3B).
Fig. 3. (A) Surface tension isotherms; (B) Orange-OT optical density in phosphonium salt solutions, λmax = 485 nm, L =
1
cm; (C, D) particle size distribution using the number parameters, C: around CMC C6.8 = 0.05, C6.9 = 0.1 mM, D: above
CMC C6.8 = 0.2 mM, C6.9 = 0.4 mM; 25 ° C.
The values of CMCs for 6.8 and 6.9 are presented in Table 3.
Table 3 CMC values determined by tensiometry and conductometry and values of surface excess (Γmax), surface area
m
per surfactant molecule (Amin), standard free energy of micellization per mole of monomer unit (ΔG ) and the standard
free energy of interfacial adsorption at the air/saturated monolayer interface (ΔGad), Solubilization capacity (S) of 6.8
and 6.9.
6
S×103
Compound CMC (mM)
β
pC20
Г
(
max×10
A
min
2
ΔGad
ΔG
m
–
2
-1
-1
mol×m ) (nm ) (kJ×mol ) (kJ×mol )
Tensiometry Conductometry
6
6
.8
.9
0.04
0.1
0.04
0.1
0.2
4.57 4.85
0.34
0.66
-38.7
-49.9
-32.4
-39.1
10.6
19.2
0.57 4.27 2.5

Phosphonium salt 6.8 forms self-assemblies at concentrations 0.04 mM. This CMC value is close to
2
7
that of alkyltriphenylphosphonium bromide (TPPB-n; where n = 18). The CMC of 6.9 is 0.1 mM,
a value 2.5 times higher than for 6.8 and close to CMC of TPPB-16 analog. The highest CMC of the
unsaturated derivative may result from the bend in the hydrocarbon chain caused by cis-double bond,
preventing close packing of hydrocarbon tails upon aggregation. The quantitative characteristics of
the adsorption of phosphonium salts were obtained using tensiometric and conductometric data.29-32
All adsorption characteristics are dependent on the saturation of the hydrocarbon chain (Table 3). The
2
2
Amin value for 6.8 (0.34 nm ) was lower than that for 6.9 with the same alkyl chain length (0.66 nm ).
This result indicates that 6.9 molecules are less closely packed at the interface than 6.8. It should be
noted that phosphonium salts 6.8 and 6.9 are highly effective to reduce water surface tension and
adsorption efficiency (pC20). The pC20 value of 6.8 and 6.9 is higher than that of traditional classical
surfactants sodium dodecyl sulfate33,34 and ammonium salts.35 Solubilization properties of
phosphonium salts 6.8 and 6.9 are also different. The optical density (at λ = 485 nm) increases with
the phosphonium salt concentration (Figure 3B). Solubilization capacity (S) of 6.9 is higher than for
6
.8 (Table 3). This may also result from the difference in self-assembly behavior of 6.8 and 6.9.
Looser packing of aggregates of 6.9 may explain better solubilization properties. Our assumption
about size differences between 6.8 and 6.9 self-assemblies was confirmed by dynamic light scattering
method. Figure 3C shows that the sizes of aggregates are 28±3 nm and 18±2 nm for 6.8 and 6.9,
respectively. The sizes increase with increasing compounds concentrations, and reach 122±1 nm and
5
9±1 nm, respectively (Figure 3D). Zeta-potential increases with increasing phosphonium salt
concentration. Zeta-potential 6.8 and 6.9 self-assemblies above CMC are about +56±3 mV (for 6.8 at
.1 mM) and +71±3 mV (for 6.9 at 0.4mM).
0
24
Commonly, the cationic surfactant self-assemblies should contribute to antimicrobial activity. The
self-assembly behavior of phosphonium salt 6.8 with low CMC value and large particle sizes is
presumably different from the self-assembly behavior of classical surfactant CTAB. As a
consequence, the antimicrobial activity is dramatically reduced. In case 6.9 MIC value is below its
CMC. Thus, it may be hypothesized that the drastic increase (30-fold) of the 6.9 anti-bacterial activity
is due to the binding of individual molecules 6.9 to the bacterial surface.
Compounds 6.1-6.9 were tested for the hemolytic activity. The data presented in Table 4 show that
the hemolysis induced by the compounds with the alkyl chain length up to 10 carbon atoms 6.1-6.4
does not exceed 10% at the highest concentration. For derivatives of 12-18 carbon atoms, the
hemolytic activity of the tested compounds increases significantly in the concentration range from 40
µM to 480 µM. However, at the concentrations below 40 µM, all the studied compounds showed low
hemolysis compared to the reference drug gramicidin S.

Table 4. Hemolytic activity of 6.1-6.9
Compound
Hemolysis,%
Concentration (µM)
40
4
80
240
120
60
20
0.2±0.01
0.3±0.02
0
10
5
0
0
0
1
0
0
0
6
6
6
.1
.2
.3
4.8±0.38 2.5±0.19
0.4±0.03
0.4±0.03
2.3±0.2
0.3±0.02
0.3±0.02
0.6±0.04
0.2±0.01
0.3±0.02
0.4±0.02
0
0.2±0.01
0
3.6±0.3
8.5±0.6
1.9±0.1
4.8±0.4
6
6
6
6
6
.4
.5
.6
.7
.8
16.4±1.4 9.6±0.84
65.4±5.1 42.3±3.4
100±8.3 100±8.5
100±7.9 100±8.2
100±8.6 100±8.4
5.9±0.45
21.2±1.6
100±7.9
100±8.0
15.6±1.3
2.8±0.22
12.2±0.9
74±6.1
98±7.7
2.5±0.23
2.3±0.2
2.9±0.24
0.3±0.02
2.5±0.21
1.4±0.11
1.3±0.1
1.8±0.1
0
2.1±0.2
1.2±0.09
0.8±0.07
1.3±0.12
0
0.9±0.07
0
0
0
0
0.9±0.08
0
0.3±0.02
0
0
0.3±0.02
0
6
.9
100±9.1 100±8.5
100±9.2 100±8.9
24.6±1.9
100±8.6
10.2±0.8
99±8.6
4.3±0.33
98±7.9
3.6±0.32
24±1.9
1.9±0.1
1.5±0.12
1.8±0.1
0.5±0.03
0.6±0.04
0
Gramicidin S
Cytotoxicity toward cancer and normal human cell lines was studied in 1-100 μM concentration range
Table 5). Compounds 6.2-6.7 showed cytotoxicity against M-Hela cell line at the level of the
common drug doxorubicin. Compound 6.9 showed activity at the tamoxifen level. Compounds 6.3-
(
6
6
6
.7 and 6.9 were most active against MCF7 and HepG2 cell lines. The IC values of test compounds
50
.5 and 6.6 are comparable to the values for the reference drug doxorubicin, and for compounds 6.3,
.4, 6.7, 6.9 - for tamoxifen. Regarding the PANK-1 cell line, compound 6.3 was the most active one
(
IC - 12.5 μM). The selectivity index (SI) was calculated as the ratio between the IC value on
5
0
5
0
3
6
normal cells and the IC values on cancer cells, where a compound with SI ≤ 2 is generally toxic
5
0
3
7
and with SI ≥ 10 is considered highly selective. The SI values of compounds 6.3-6.7 for M-Hela
were 7–14 times higher than for doxorubicin, and 4–7 times higher than for tamoxifen. Interestingly,
compounds 6.3–6.7 showed the highest selectivity profiles against MCF-7 and Hep G2 cell lines,
without considerable selectivity against PANC-1 except for 6.3 (SI - 3.7).
Table 5. In vitro cytotoxic effects (μM) and selectivity index values (SI) of 6.1 – 6.9 *.
Compound
Cancer cell lines
Normal cell
line
Сhang liver
M-Helaa
MCF-7b
Hep G2c
PANC-1d
e
IC50
SI
1.3
7.6
12.5
13.4
16.4
12.1
14.1
1.0
IC50
SI
IC50
SI
0.78
1.6
1.8
3.7
4.5
6.4
3.6
IC50
SI
0.79
6
6
6
6
6
6
6
6
6
.1
.2
.3
.4
.5
.6
.7
.8
.9
50.0±3.9
5.5±0.4
3.7±0.3
3.4±0.2
2.6±0.1
3.5±0.3
3.2±0.2
68±5.2
86.0±7.3
32.8±2.7
12.5±0.9
13.2±1.1
7.2±0.6
6.6±0.5
13.2±1.2
49±3.6
0.76 83±6.9
82.0±6.7
40.0±3.2
12.5±1.1
26.3 ±2.3
25.3±2.1
25±2.1
25.7±2.2
65.0±4.8
25.2±2.3
3.2±0.3
65.3±5.1
41.6±3.5
46.3±3.8
45.7±3.6
42.8±3.3
42.3±3.4
45.2±3.8
69.0±5.5
56.3±4.6
3.0±0.2
1.3
3.7
3.5
5.9
6.4
3.4
1.4
3.1
1.0
1.8
26.3±2.3
25.7±2.1
12.3±1.2
9.8±0.7
6.6±0.5
12.6±1.1
73±6.1
83±7.4
3.3±0.2
23.4±1.9
1.04
3.7
1.7
1.7
1.7
1.8
1.1
2.2
0.93
0.94
0.67
0.9
28.4±2.2
3.0±0.2
28.0±2.5
2.0
1.0
1.7
18.2±1.4
3.0±0.1
25.0±2.2
Doxorubicin
Tamoxifen
2.0
46.2±3.5
*
Average of of three values measured; ± standard deviation (SD)
aM-Hela - epithelioid cervical carcinoma; MCF-7 – human breast adenocarcinoma cells; Hep G2 - human
b
c
d
e
hepatoblastoma cells; PANC-1 - human pancreatic cancer cell line; Сhang liver - human normal liver cells

The type of death induced by the tested compounds of M-Hela cells was investigated using Annexin
V-FITC staining protocol. As shown in Figure 4, after 24 hours of treatment, the test compound 6.5
significantly induced apoptosis in M-Hela cells. The analysis of apoptotic effects indicated a dose-
dependent nature of early apoptosis. At a concentration 3 μM, similar to IC value, compound 6.5
5
0
caused early and late apoptosis in 39.26% and 9.78% of M-Hela cells, respectively. With an increase
in the concentration of the phosphonium salt to 6μM, the number of cell death due to early apoptosis
increased to 53.31%; due to late apoptosis.
Fig. 4. Apoptosis induction on M-Hela cells incubated with 3 and 6 µM of phosphonium salt 6.5 for 24h. Quantification
of dot plots expressed as percentage of total cells (mean ± SD, n = 3), along with the quantification of the numbers of
cells (% of total) in the early and late stages of apoptosis for the control and treatment groups. The values are presented
as the mean ± SD (n = 3); (∗) P < 0.01 compared to control.
The ability of the studied compounds to cause a decrease in the mitochondrial membrane potential
(
Δψ ) in the cells of M-Hela culture was evaluated by JC-10 dye staining method. The ratio between
m
red (normal ψ ) and green (ψ loss) fluorescence was demonstrated using flow cytometry analysis
m
m
(Figure 5). The intensity of the red fluorescence decreased in a dose-dependent manner with
increasing concentration of the test compound as well as induction of the early apoptosis.
The results obtained suggest that the mechanism of action of the studied compounds may be
associated with the induction of apoptosis, which proceeds along with the mitochondria-dependent
pathway.

Fig. 5. Effects on Δψm by 6.5 in M-Hela cells. Flow cytometry analysis of M-Hela cells treated with 6.5 at indicated
concentrations, along with the quantification of % of cells with red aggregates. The values are presented as percentage
ratio (mean ± SD n = 3): (∗) P < 0.01 compared to control.
The series of alkyl(alkenyl)(2-hydroxybenzyl)diphenylphosphonium salts with antimicrobial and
antitumor dual-action were synthesized. The most active compounds 6.4, 6.5 exhibit antibacterial
activity at MIC 1-2 µM against Gram-positive bacteria Sa and Bc as well as drug-resistant strains,
and at the same time with MIC at the value 7.8-16.6 µM against fungi Tm and Ca. The change in
nature of substituents (saturated/unsaturated C18 and C6 hydrocarbon tail) of the same length led to
a drastic change of biological activity. According to the self-assembly behavior of saturated 6.8 and
unsaturated 6.9 this abrupt change (30-fold) in biological activity might be explained by the fact that
the monomeric form of the salts is responsible for antimicrobial action. Moreover, the toxicity
evaluation based on hemolysis model showed that none induced notable hemolysis at MIC values.
Furthermore, synthesized phosphonium salts were tested for cytotoxicity against normal human liver
cells and four human cancer cell lines to evaluate their antitumor activity. Among the tested
compounds, salt 6.5 was the most potent on M-Hela cell line with IC on the level of Doxorubicin
5
0
and with good SI at the same time. The cell death was induced by a mitochondrial-dependent
apoptosis pathway. Finally, it could be concluded that hydroxybenzylphosphonium salts are
promising dual anticancer/antimicrobial agents with a good antitumor and antimicrobial action and
low toxicity to normal human cells.
The authors have declared no conflict of interest.
Acknowledgments
Authors are grateful to the Assigned Spectral-Analytical Center of FRC Kazan Scientific Center of
RAS for technical assistance in research.
Appendix A. Supplementary data
The following are the Supplementary data to this article:
1
13
Supplementary data. Copies of H NMR, C NMR, biological evaluation methods and physico-
chemical experiments.
REFERENCES
1
. Chrzanowska A, Roszkowski P, Bielenica A, Olejarz W, Stępień K, Struga M. Anticancer and antimicrobial effects of
novel ciprofloxacin fatty acids conjugates. Eur J Med Chem. 2020;185:111810. doi:10.1016/j.ejmech.2019.111810
. Rostom SAF, Faidallah HM, Radwan MF, Badr MH. Bifunctional ethyl 2-amino-4-methylthiazole-5-carboxylate
derivatives: Synthesis and in vitro biological evaluation as antimicrobial and anticancer agents. Eur J Med Chem.
014;76:170-181. doi:10.1016/j.ejmech.2014.02.027
2
2

3
2
4
. Carmona-Ribeiro A, de Melo Carrasco L. Cationic Antimicrobial Polymers and Their Assemblies. Int J Mol Sci.
013;14(5):9906-9946. doi:10.3390/ijms14059906
. Brunel F, Lautard C, di Giorgio C, et al. Antibacterial activities of mono-, di- and tri-substituted triphenylamine-based
phosphonium ionic liquids. Bioorg Med Chem Lett. 2018;28(5):926-929. doi:10.1016/j.bmcl.2018.01.057
. Nazarov PA, Osterman IA, Tokarchuk A V., et al. Mitochondria-targeted antioxidants as highly effective antibiotics.
Sci Rep. 2017;7(1):1394. doi:10.1038/s41598-017-00802-8
. Chang W, Liu J, Zhang M, et al. Efflux pump-mediated resistance to antifungal compounds can be prevented by
conjugation with triphenylphosphonium cation. Nat Commun. 2018;9(1):5102. doi:10.1038/s41467-018-07633-9
5
6
7
1
8
. Lucchini J., Corre J, Cremieux A. Antibacterial activity of phenolic compounds and aromatic alcohols. Res Microbiol.
990;141(4):499-510. doi:10.1016/0923-2508(90)90075-2
. Bittner Fialová S, Kello M, Čoma M, et al. Derivatization of Rosmarinic Acid Enhances its in vitro Antitumor,
Antimicrobial and Antiprotozoal Properties. Molecules. 2019;24(6):1078. doi:10.3390/molecules24061078
. Gazzano E, Lazzarato L, Rolando B, et al. Mitochondrial Delivery of Phenol Substructure Triggers Mitochondrial
Depolarization and Apoptosis of Cancer Cells. Front Pharmacol. 2018;9. doi:10.3389/fphar.2018.00580
0. Rokitskaya TI, Terekhova N V., Khailova LS, et al. Zwitterionic Protonophore Derived from 2-(2-
Hydroxyaryl)alkenylphosphonium as an Uncoupler of Oxidative Phosphorylation. Bioconjug Chem. 2019;30(9):2435-
9
1
2
1
443. doi:10.1021/acs.bioconjchem.9b00516
1. Zielonka J, Joseph J, Sikora A, et al. Mitochondria-Targeted Triphenylphosphonium-Based Compounds: Syntheses,
Mechanisms of Action, and Therapeutic and Diagnostic Applications. Chem Rev. 2017;117(15):10043-10120.
doi:10.1021/acs.chemrev.7b00042
1
2. Sassi N, Mattarei A, Azzolini M, et al. Mitochondria-targeted Resveratrol Derivatives Act as Cytotoxic Pro-oxidants.
Curr Pharm Des. 2014;20(2):172-179. doi:10.2174/13816128113199990034
3. Peredo-Silva L, Fuentes-Retamal S, Sandoval-Acuña C, et al. Derivatives of alkyl gallate triphenylphosphonium
exhibit antitumor activity in a syngeneic murine model of mammary adenocarcinoma. Toxicol Appl Pharmacol.
1
2
1
6
1
017;329:334-346. doi:10.1016/j.taap.2017.06.017
4. Paulsen IT, Brown MH, Skurray RA. Proton-dependent multidrug efflux systems. Microbiol Rev. 1996;60(4):575-
08.
5. Tatarinov DA, Kuznetsov DM, Voloshina AD, et al. Synthesis of 2-(2-hydroxyaryl)alkenylphosphonium salts from
phosphine oxides via ring-closing ring-opening approach and their antimicrobial evaluation. Tetrahedron.
2
1
016;72(51):8493-8501. doi:10.1016/j.tet.2016.11.023
6. Tatarinov DA, Kuznetsov DM, Kostin AA, Mironov VF. 2-Ethoxy-2,3-dihydro[d][1,2]oxaphosphole 2-oxide in the
synthesis of dialkyl(diaryl)(2-hydroxybenzyl)phosphine oxides. Russ
J
Gen Chem. 2016;86(3):530-533.
doi:10.1134/S1070363216030063
1
7. Pérez-Prieto J, Galian RE, Burgos PO, del Carmen Morant Miñana M, Miranda MA, López-Ortiz F. Influence of
Substitution at the Benzylic Position on the Behavior of Stereoisomeric Phosphorus Compounds as Precursors of
Stabilized Carbon-Centered Radicals. Org Lett. 2005;7(18):3869-3872. doi:10.1021/ol051254y
1
8. McKenna CE, Higa MT, Cheung NH, McKenna MC. The facile dealkylation of phosphonic acid dialkyl esters by
bromotrimethylsilane. Tetrahedron Lett. 1977;18(2):155-158. doi:10.1016/S0040-4039(01)92575-4
9. Cardellicchio C, Fracchiolla G, Naso F, Tortorella P, Holody W, Pietrusiewicz KM. A novel carbon leaving group in
1
the reaction of organometallic compounds with phosphine oxides. Tetrahedron Lett. 1999;40(31):5773-5776.
doi:10.1016/S0040-4039(99)01110-7

2
2
2
0. Stankevič M, Pisklak J, Włodarczyk K. Aryl group – a leaving group in arylphosphine oxides. Tetrahedron.
016;72(6):810-824. doi:10.1016/j.tet.2015.12.043
1. National Committee for Clinical Laboratory Standards, Methods for dilution antimicrobial susceptibility. Tests for
bacteria that grow aerobically – sixth edition: approved standard, M7-A5, NCCLS (2000) Wayne, Pa., USA.;
2. National Committee for Clinical Laboratory Standards, Reference method for broth dilution antifungal susceptibility
testing of conidium-forming filamentous fungi: proposed standard, M38-P, NCCLS (1998) Wayne, Pa., USA
3. Lukáč M, Devínsky F, Pisárčik M, Papapetropoulou A, Bukovský M, Horváth B. Novel Phospholium-Type Cationic
2
2
Surfactants: Synthesis, Aggregation Properties and Antimicrobial Activity. J Surfactants Deterg. 2017;20(1):159-171.
doi:10.1007/s11743-016-1908-6
2
4. Zhou C, Wang Y. Structure–activity relationship of cationic surfactants as antimicrobial agents. Curr Opin Colloid
Interface Sci. 2020;45:28-43. doi:10.1016/j.cocis.2019.11.009
5. Birnie CR, Malamud D, Schnaare RL. Antimicrobial Evaluation of N-Alkyl Betaines and N-Alkyl-N,N-
2
Dimethylamine Oxides with Variations in Chain Length. Antimicrob Agents Chemother. 2000;44(9):2514-2517.
doi:10.1128/AAC.44.9.2514-2517.2000
2
6. Devínsky F, Lacko I, Mlynarcík D, Racansky V, Krasnec L. Relationship between critical micelle concentrations and
minimum inhibitory concentrations for some non-aromatic quaternary ammonium salts and amine oxides. Tenside
Deterg. 1985;22:10-15.
2
7. Devi KP, Nisha SA, Sakthivel R, Pandian SK. Eugenol (an essential oil of clove) acts as an antibacterial agent against
Salmonella typhi by disrupting the cellular membrane.
doi:10.1016/j.jep.2010.04.025
J
Ethnopharmacol. 2010;130(1):107-115.
2
8. Zakharova LY, Kaupova GI, Gabdrakhmanov DR, et al. Alkyl triphenylphosphonium surfactants as nucleic acid
carriers: complexation efficacy toward DNA decamers, interaction with lipid bilayers and cytotoxicity studies. Phys Chem
Chem Phys. 2019;21(30):16706-16717. doi:10.1039/C9CP02384D
2
9. Jana PK, Moulik SP. Interaction of bile salts with hexadecyltrimethylammonium bromide and sodium dodecyl sulfate.
J Phys Chem. 1991;95(23):9525-9532. doi:10.1021/j100176a089
0. Rosen MJ, Li F, Morrall SW, Versteeg DJ. The Relationship between the Interfacial Properties of Surfactants and
Their Toxicity to Aquatic Organisms. Environ Sci Technol. 2001;35(5):954-959. doi:10.1021/es0015141
1. Zana R. Dimeric (Gemini) Surfactants: Effect of the Spacer Group on the Association Behavior in Aqueous Solution.
J Colloid Interface Sci. 2002;248(2):203-220. doi:10.1006/jcis.2001.8104
2. Zana R. Critical Micellization Concentration of Surfactants in Aqueous Solution and Free Energy of Micellization.
Langmuir. 1996;12(5):1208-1211. doi:10.1021/la950691q
3. Yoshimura T, Esumi K. Synthesis and surface properties of anionic gemini surfactants with amide groups. J Colloid
Interface Sci. 2004;276(1):231-238. doi:10.1016/j.jcis.2004.03.045
4. Yoshimura T, Esumi K. Physicochemical Properties of Ring-Type Trimeric Surfactants from Cyanuric Chloride.
Langmuir. 2003;19(8):3535-3538. doi:10.1021/la026762k
5. Rosen MJ, Liu L. Surface activity and premicellar aggregation of some novel diquaternary gemini surfactants. J Am
Oil Chem Soc. 1996;73(7):885-890. doi:10.1007/BF02517990
6. Koch A, Tamez P, Pezzuto J, Soejarto D. Evaluation of plants used for antimalarial treatment by the Maasai of Kenya.
J Ethnopharmacol. 2005;101(1-3):95-99. doi:10.1016/j.jep.2005.03.011
3
3
3
3
3
3
3

3
7. Peña-Morán O, Villarreal M, Álvarez-Berber L, Meneses-Acosta A, Rodríguez-López V. Cytotoxicity, Post-
Treatment Recovery, and Selectivity Analysis of Naturally Occurring Podophyllotoxins from Bursera fagaroides var.
fagaroides on Breast Cancer Cell Lines. Molecules. 2016;21(8):1013. doi:10.3390/molecules21081013
Declaration of interests
☒
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.
☐
The authors declare the following financial interests/personal relationships which may be considered as
potential competing interests: