Towards catch-up therapy: evaluation of nucleophilic active pharmaceutical ingredients for the treatment of percutaneous VX poisoning, in-vial and in-vitro studies

Article abstract of DOI:10.1016/j.ijpharm.2021.120689

Dermal exposure to low volatility organophosphorus chemical warfare agents (OP CWA) poses a great risk to the exposed person. Due to their lipophilic nature, these compounds rapidly absorb into the skin, leading to the formation of a “dermal reservoir” from which they slowly enter the bloodstream causing prolonged intoxication. Traditionally, strategies to counter the toxicity of such substances consist of chemical decontamination/physical removal of the residual agent from the skin surface (preferably as soon as possible following the exposure) and administration of antidotes in the case of intoxication signs. Hence, these strategies are unable to counter a substantial amount of the agent, which accumulates inthe dermal reservoir. More than a decade ago, the concept of a “catch-up therapy” intended to neutralize the dermal reservoir was suggested. Herein, we describe examples of potential “catch-up therapy” lotions - vehicles designed to deliver small nucleophilic molecules into the skin and potentially decompose the remaining CWA before it reaches the blood stream. Eleven nucleophilic compounds, based on approved drugs, were initially screened. They were then tested in various binary solutions, for their detoxification efficacy and degradation ability towards lipophilic OP CWA models such as dibutylphosphofluoridate and o-nitro-phenyl diphenyl phosphate, as well as the nerve agent VX, by means of kinetic ³¹P NMR and UV–Vis spectroscopy. Of these, the potassium and diethyl ammonium salts of acetohydroxamic acid (AHAK and AHA DEA) in (DMSO/H₂O 1:4) were found to be the most active nucleophiles, hydrolyzing VX in practical time scales (t_1/2 = 5.28 and 6.78 min, respectively). The vehicle solution DMSO/H₂O 1:4 promoted the penetration of substantial amounts of AHA K and AHA DEA through excised pig skin in in-vitro studies, suggesting that such formulations may serve as useful CWA nucleophilic scavengers for both on and within -skin detoxification. These findings may pave the way to a more efficacious treatment against low volatility OP CWA percutaneous poisoning.

Full text of DOI:10.1016/j.ijpharm.2021.120689

International Journal of Pharmaceutics 603 (2021) 120689
Contents lists available at ScienceDirect
International Journal of Pharmaceutics
journal homepage: www.elsevier.com/locate/ijpharm
Towards catch-up therapy: evaluation of nucleophilic active
pharmaceutical ingredients for the treatment of percutaneous VX
poisoning, in-vial and in-vitro studies
,
,*
Victoria Nahuma, Uri Nilib, Eugenia Bloch-Shildermanb, Boris Smolkina *, Nissan Ashkenazia
a Department of Organic Chemistry, IIBR – Israel Institute for Biological Research, P.O. Box 19, Ness Ziona 7410001, Israel
b Department of Pharmacology, IIBR – Israel Institute for Biological Research, P.O. Box 19, Ness Ziona 7410001, Israel
A R T I C L E I N F O
A B S T R A C T
Keywords:
Dermal exposure to low volatility organophosphorus chemical warfare agents (OP CWA) poses a great risk to the
exposed person. Due to their lipophilic nature, these compounds rapidly absorb into the skin, leading to the
formation of a “dermal reservoir” from which they slowly enter the bloodstream causing prolonged intoxication.
Traditionally, strategies to counter the toxicity of such substances consist of chemical decontamination/physical
removal of the residual agent from the skin surface (preferably as soon as possible following the exposure) and
administration of antidotes in the case of intoxication signs. Hence, these strategies are unable to counter a
substantial amount of the agent, which accumulates inthe dermal reservoir. More than a decade ago, the concept
of a “catch-up therapy” intended to neutralize the dermal reservoir was suggested. Herein, we describe examples
of potential “catch-up therapy” lotions - vehicles designed to deliver small nucleophilic molecules into the skin
and potentially decompose the remaining CWA before it reaches the blood stream. Eleven nucleophilic com-
pounds, based on approved drugs, were initially screened. They were then tested in various binary solutions, for
their detoxification efficacy and degradation ability towards lipophilic OP CWA models such as dibutylphos-
phofluoridate and o-nitro-phenyl diphenyl phosphate, as well as the nerve agent VX, by means of kinetic 31P
NMR and UV–Vis spectroscopy. Of these, the potassium and diethyl ammonium salts of acetohydroxamic acid
(AHAK and AHA DEA) in (DMSO/H2O 1:4) were found to be the most active nucleophiles, hydrolyzing VX in
practical time scales (t1/2 = 5.28 and 6.78 min, respectively). The vehicle solution DMSO/H2O 1:4 promoted the
penetration of substantial amounts of AHA K and AHA DEA through excised pig skin in in-vitro studies, sug-
gesting that such formulations may serve as useful CWA nucleophilic scavengers for both on and within -skin
detoxification. These findings may pave the way to a more efficacious treatment against low volatility OP CWA
percutaneous poisoning.
“Catch-up therapy”
VX
Percutaneous poisoning
Dermal depot
Nucleophilic scavengers
Hydroxamic acids
Dermal lotion
1. Introduction
OP CWA largely vary in their physicochemical properties (e.g.
volatility, solubility, chemical stability, etc.). This variability is
Organophosphorus (OP) chemical warfare agents (CWA) are a group
of highly toxic substances which exert their activity by irreversibly
inhibiting acetylcholine esterase (AChE), the enzyme that hydrolyses
acetylcholine (ACh) in nerve synapses. Inhibition of AChE leads to ACh
accumulation which results in a cholinergic crisis (Costanzi et al., 2018;
Thiermann et al., 2016). Although banned by international treaties (i.e.,
Organization for the Prohibition of Chemical Weapons- OPCW), the
deployment of OP CWA against civilians is well documented during the
last decade (Syria 2013 and 2017, Kuala Lumpur, Malaysia 2017, Sal-
isbury, UK 2018, etc.).
expressed in their mode of exposure and toxicokinetics (Munro, 1994;
Thiermann et al., 2016). Exposure to OP CWA may occur by different
routes, primarily inhalation and percutaneous absorption. While volatile
compounds are most likely to be absorbed mainly via the respiratory
system, for low volatility compounds (e.g. the V-type nerve agents VX,
VM, CVX and RVX), dermal uptake is expected to be a dominant route of
entry (Joosen et al., 2017, 2010; Romano et al., 2008; Thiermann et al.,
2016). The route of exposure in turn has a large impact on the tox-
icokinetics of poisoning. Accordingly, following exposure, volatile OP
nerve agents are promptly absorbed through mucous membranes,
* Corresponding authors.
E-mail addresses: boriss@iibr.gov.il (B. Smolkin), nissan.ashkenazi@iibr.gov.il (N. Ashkenazi).
https://doi.org/10.1016/j.ijpharm.2021.120689
Received 25 March 2021; Received in revised form 28 April 2021; Accepted 4 May 2021
Available online 7 May 2021
0378-5173/© 2021 Elsevier B.V. All rights reserved.

V. Nahum et al.
International Journal of Pharmaceutics 603 (2021) 120689
leading to fast (within minutes) onset of cholinergic symptoms.
Conversely, low volatility lipophilic OP nerve agents absorbed into the
skin, form a subcutaneous depot from which they distribute slowly into
the body (Chilcott et al., 2005a, 2005b; Joosen et al., 2017, 2013, 2010).
Thus, systemic signs of poisoning may appear only after a time lag which
may take hours (Dalton et al., 2006; Joosen et al., 2010; Thiermann
et al., 2016). Furthermore, high levels of the agent may persist in the
blood circulation for hours (Joosen et al., 2010), due to continuous
diffusion of the poison from the depot into the blood stream, combined
with its chemical stability towards hydrolysis at physiological pH (Yang,
1999).
Table 1
Molecular structure, name and application of candidate compounds.
Compound
number
Structure
Name
Application
1
2
3
4
Methyl paraben
Preservative, anti-
fungal agent
Butyl paraben
Triethanolamine
Aciclovir
Preservative, anti-
bacterial/fungal
agent
emulsifier
The aforementioned toxicological characteristics of low volatility
nerve agents require special attention when devising appropriate
decontamination and treatment strategies against them. Traditionally,
strategies to counter the toxicity of nerve agents consist of two main
tactics; decontamination of the contaminated skin surface, preferably by
an active decontamination lotion such as RSDL® (Reactive Skin
Decontamination Lotion) or AlldecontMED as soon as possible (these
lotions were found to be effective if used within 15 min following
exposure (Joosen et al., 2017; Thors et al., 2017a), and injection of
antidotes in the case of intoxication signs (Hamilton et al., 2004;
Thiermann et al., 2016). However, in a real-life scenario, toxicological
signs following dermal exposure to persistent nerve agents, which will
most likely be the trigger to treat, may be long delayed. Accordingly, by
the time treatment will commence, a substantial amount of the agent
will most likely already be absorbed into the skin, forming a dermal
reservoir (Joosen et al., 2013; Wetherell et al., 2008). In such cases, late
decontamination of the skin surface will probably be only partially
effective (Chilcott et al., 2005a; Joosen et al., 2013; Wetherell et al.,
2008). Furthermore, at this stage antidotal treatment will likely achieve
only temporary relief, due to reoccurrence of symptoms following
continuous penetration of VX from the dermal reservoir, necessitating
repeated administration of antidotes (Bloch-Shilderman et al., 2019;
Joosen et al., 2013, 2010). Thus, the aforementioned strategies are ex-
pected to be most effective only if the formation of the dermal depot is
prevented (Chilcott et al., 2005a). For this purpose, new complementary
strategies need to be considered.
Antiviral
5
6
7
Dapsone
Antibiotic
Minoxidil
Salicylamide
Antihypertensive
vasodialator
Analgesic and
antipyretic
8
9
Acetohydroxamic
acid (AHA)
Antibiotic, urease
inhibitor
Salicylhydroxamic
acid (SHA)
Antibiotic, urease
inhibitor
10
11
Caprylhydroxamic
acid (CHA)
Preservative
Bufexamac (BUF)
Anti-
inflammatory
agent
12
2,3-Butanedione
monoxime (DAM)
Active ingredient
in the RSDL®
lotion
Recently, Joosen and coworkers explored the efficacy of delayed
decontamination by RSDL® (90 min following exposure) of hairless
guinea pigs dermally exposed to VX, showing a delay in complete AChE
inhibition. This effect was attributed to the partial neutralization of a VX
depot within the skin following the decontamination, although a single
decontamination was insufficient to prevent morbidity (Joosen et al.,
2017). Similarly, Thors and coworkers showed a decrease in VX pene-
tration rates through human skin following decontamination with
RSDL®, even 120 min following exposure (Thors et al., 2017b). As it is
highly likely that so long after exposure a substantial amount of the
agent had already been absorbed to the skin and formed a depot (Chil-
cott et al., 2005a; Dalton et al., 2006), we may speculate that a small but
sufficient portion of the RSDL® applied might have reached the dermal
depot and neutralized some of the agent (although RSDL® was designed
to neutralize the poison on the skin surface without penetration of its
active ingredient (Connolly et al., 2020)). No explicit reference to such
possibility was made.
and (ii) Intra dermal CWA decomposition.
Notably, OP CWA are all electrophilic by nature, consisting of a
central P(V) atom. Accordingly, the above described “catch-up therapy”
lotions should consist of nucleophilic molecules, able to chemically
deactivate the OP CWA depot. However, delivering strong nucleophiles
through the skin is far from trivial, as such charged molecules may be
too hydrophilic to cross the lipophilic skin barrier (Boroujerdi, 1987).
Thus, careful planning is required; first, in respect to the design of such a
lotion, two primary parameters should be considered: a) the nucleophile
of choice has to be active enough to efficiently neutralize the agent in a
relevant time scale; b) the formulation has to enable the permeation of
the charged nucleophile through skin layers. Obviously, these two re-
quirements may contradict as the greater the nucleophilicity the higher
the dipole moment and the hydrophilicity. Therefore, it can be assumed
that a binary mixture of solvents will be the minimum needed to ensure
both adequate solubility and reactivity of the scavenger, as well as its
penetrability. Second, all components of the lotion should preferably be
clinically approved for use in humans.
In line with the above, more than a decade ago, the concept of a
“catch-up therapy” was suggested by Chilcott et al., who envisioned that
reactive chemicals in an appropriate formulation could be delivered into
the skin, follow the same pathway of the agent, and neutralize the
dermal reservoir (Chilcott et al., 2005b, 2005a). However, to the best of
our knowledge, to date this elegant concept remains at a theoretical
level, with no such medical formulation currently available. Therefore,
we aimed to develop potential “catch-up therapy” lotions intended to
penetrate the skin, in order to chemically deactivate an existing CWA
dermal reservoir before it reaches the blood stream. The possible benefit
of such a lotion in comparison to currently available decontamination
kits would be its dual mode of action: (i) Skin surface decontamination
To address these demands, we scanned pharmaceutical libraries of
approved drugs, in a search for active pharmaceutical ingredients that
could be nucleophilic enough to neutralize OP CWA (1–11, Table 1),
with an efficacy comparable to or higher than that of Diacetyl mon-
oximate (DAM - the active ingredient of RSDL®, 12). As lipophilic model
OP’s (G-type simulants) for initial screening, Dibutylphosphoro-
fluoridate (13) and o-Nitro-phenyl diphenyl phosphate (14) were cho-
sen, since their degradation rates could be measured by kinetic 31P NMR
2

V. Nahum et al.
International Journal of Pharmaceutics 603 (2021) 120689
Fig. 1. Molecular structure of VX and G-type simulants DBPF and ONPDP.
and UV–Vis spectroscopy, respectively. The most active nucleophiles
which degraded the OP models and nerve agent VX (15) at reasonable
rates (Fig. 1), were evaluated for their skin penetration using flow-
through diffusion cells in different formulations. To promote efficient
permeation several techniques were applied, including formation of
organic ion-pairs, use of permeation enhancers and use of concentrated
solutions of the nucleophiles in a H2O/DMSO binary solution. Herein we
report our results, delineating new formulations that can penetrate the
skin depot and hydrolyze VX at a reasonable timescale, which may
accordingly be suitable for use as a “catch-up therapy” against cutaneous
intoxication by OP’s.
2.1.3. O-ethyl-S-2-(N, N-diisopropylaminoethyl)methylphosphonothioate
(VX, 15)
Was synthesized in house (>95% purity).
2.2. NMR spectroscopy
31P {1H} spectra were obtained at 202 MHz at room temperature on
an 11.7 T (500 MHz) Bruker spectrometer (Avance III HD). Chemical
shifts were calibrated to trimethyl phosphate as 0 ppm. The spectra were
recorded using standard parameters of the TopSpin software (version
3.5). Each data point was obtained from 32 scans with a spectral width
of 200 ppm and 4 s recycle delay. All spectra were taken under identical
conditions and integration was done automatically using TopSpin 3.5.2
software and the build in AU multi_integ3.
2. Materials and methods
Caution! OP’s are extremely toxic compound, with special emphasis on
VX. Experiments with these compounds should only be performed by trained
personnel using applicable safety procedures.
2.3. Degradation kinetics
All nucleophiles, accept 3, 5 and 6 which were used as free base,
were converted to sodium, potassium or ammonium salts prior to kinetic
evaluation by reacting them with sodium methoxide, potassium meth-
oxide or the appropriate amine, respectively. Nucleophiles (26 equiv)
were added to DMSO or PG/H2O 1:1 solution (1 mL) and stirred for 1
min until completely dissolved. DBPF (13) or VX (15) (2 µL, 1 equiv)
were carefully added, and the solution was transferred to an NMR tube
for kinetic analysis. Kinetic NMR acquisition started as soon as possible
(without lock and shimming) and continued until full degradation, or
until no further change in reaction conversion was observed. Analysis of
the results was performed by the GraphPad Prism 5 software to deter-
mine half-life times.
2.1. Chemicals
The following active pharmaceutical ingredients, reagents and sol-
vents (HPLC grade): methyl paraben, butyl paraben, triethanolamine,
sodium aciclovir, dapsone, minoxidil, salicylamide, acetohydroxamic
acid (AHA), salicylhydroxamic acid (SHA), caprylhydroxamic acid
(CHA), Bufexamac (BUF), diacetyl monooxime (DAM), diethylamine
(DEA), trimethylamine, ethanolamine, diethanolamine, piperidine,
pyrrolidinoethanol (Epolamine), piperazine, dimethyl sulfoxide
(DMSO), propylene glycol (PG), ethanol (EtOH), isopropanol (IPA),
polyethylene glycol monoethyl ether (mPEG), sodium metoxide and
potassium methoxide were purchased from commercial suppliers and
used without further purification. Reactive Skin Decontamination
Lotion (RSDL®) was obtained from Emergent Protective Products USA
Inc. Deionized water was obtained from a laboratory water purification
system.
2.4. UV–Vis spectroscopy
For kinetic experiments a stock solution of o-nitro-phenyl diphenyl
phosphate (ONPDP) was prepared by dissolving 5 µL (18.85 µmol, 1
equiv) in 1 mL of DMSO. Accordingly, stock solutions for selected nu-
cleophiles were prepared in ethanol with 1:10 ONPDP: nucleophile ratio
(188.53 µmol, 10 equiv). For a typical reaction, 36 µL of ONPDP and the
appropriate nucleophile stock solutions were added to a cuvette con-
taining 2928 µL of buffer/solvent 1:1 solution. The buffer (pH 9.2) was
prepared by mixing 10% (v/v) Na2CO3 (0.1 N) with 90% (v/v) NaHCO3
(0.1 N) solutions. Kinetic measurements were carried out on a Shimadzu
UV-1800 spectrophotometer using a Kinetics built-in function. Analysis
of the results was performed using the GraphPad Prism 5 software to
determine half-life times.
2.1.1. Dibutyl-phosphorofuloridate (13)
To a solution of tributylphosphite (4.62 gr, 18.47 mmol, 1.0 equiv) in
THF (50 mL) bromine (2.95 gr. 18.47 mmol, 1 equiv) was added drop-
wise at ꢀ 78 ◦C until brown color persisted and the mixture was left to
stir for 1 h. Thereafter, the mixture was warmed to ꢀ 40 ◦C and cesium
fluoride (8.43 gr, 55.5 mmol, 3 equiv) was added. The reaction was left
to reach rt and stirred for 8 h. After reaction completion the solvent was
evaporated under reduced pressure and the product was distilled (3.9 gr,
99.5% yield). Spectral data was identical to previously reported (Norlin
et al., 2005).
2.5. pKa measurements
2.1.2. 2-nitrophenyl diphenylphosphate (14)
The pKa values of hydroxamic acids and DAM were determined pH
metrically using an Eutech Instruments (PH 700) pH meter. The in-
strument was calibrated with pH 4, 7 and 10 standard buffer solutions
(Reagecon).
To a solution of diphenylphosphinic chloride (4.83 gr, 2.04 mmol,
1.0 equiv) in ether (50 mL) triethylamine (2.06 gr, 2.04 mmol, 1.0
equiv) and 2-nitrophenol (2.84 gr, 2.04 mmol, 1.0 equiv) were added
dropwise and the mixture was left to stir for 18 h. After reaction
completion the organic phase was washed twice with water (50 mL ×2),
dried over anhydrous Na2SO4, filtered and concentrated in vacuo. The
residue was purified by silica gel column chromatography (Hexane to
Hexane/EtOAc 1:5) to give 14 (6.26 gr, 95% yield). Spectral data was
identical to previously reported (Jones et al., 2016).
2.6. Log P measurements
Measurements of the partition coefficient were performed with n-
octanol as the oil phase and water as the aqueous phase, in accordance
3

V. Nahum et al.
International Journal of Pharmaceutics 603 (2021) 120689
with the OECD guidelines (OECD/OCDE, 1995). The two phases were
mixed for 24 h to form saturated phases. Solutions of AHA (8) and DAM
(12) (1 mM) were prepared in the aqueous phase and solutions of salts
were prepared in the aqueous phase adjusted to pH 11. The aqueous
phases were mixed with n-octanol in 10 mL tube at the following
portions:
that covered a linear range of 0.5–120 µg/ml, r2 > 0.998. Permeation
was evaluated by plotting the cumulative amount of permeate passing
per cm2 of skin versus time. In cases in which steady state permeation
was reached, the absorption rates (Jss) were determined as the slopes of
the linear portion of the plot. The lag times (TL) were calculated from
linear extrapolation of the steady state portion of the plot. The perme-
ability coefficients (Kp) were calculated as the ratio between the ab-
sorption rates and the initial concentrations of the permeates in the
applied formulations. Total accumulations (Q8) were calculated as the
cumulative amount of permeate after 8 h. Each experiment was evalu-
ated by four independent replicates. The integrity of the skins was
assessed by examining the permeation results. Skins that were damaged
and exhibited excessive penetration rate (>10,000 µg cm-2hꢀ 1 without
reaching steady state) were removed from the analysis and replaced by
additional experiments. All skin permeation results are presented as the
mean ± standard error of the mean (SEM). Statistical analysis was
performed using a one-way ANOVA followed by either a Dunnett’s or
Bonferroni multiple comparison test to determine differences in pene-
tration rates. Results were considered significant for P < 0.05.
(1) Water/n-octanol 1:1 (5 mL each)
(2) Water/n-octanol 1:2 (aqueous phase 3 mL n-octanol phase 6 mL)
(3) Water/n-octanol 2:1 (aqueous phase 6 mL n-octanol phase 3 mL)
Duplicates of each solution in tubes were closed with a cap and
shaken for 24 h at room temperature. After careful separation using
centrifugation and by a syringe, the concentration at each phase was
evaluated by means of HPLC. The partition coefficient was calculated as
the ratio between the concentrations of the two phases. Three solutions
with different ratios of water/n-octanol were used to ensure that the
results are not erroneous due to saturation of one of the phases with the
tested compounds.
2.7. In vitro skin penetration
3. Results and discussion
Portions of (5 × 5 cm) of excised domestic pig back skin free of fat
(obtained from Lahav CRO, Israel) were carefully cleaned and stored at
ꢀ 80 ◦C. Prior to use the skin portions were thawed at 4 ◦C overnight.
Flow-through diffusion cells were used throughout the study. PBS was
chosen as the receptor fluid since it is commonly used in similar studies
and as it follows the OECD guidelines which state that a test material
should be soluble up to ten times the likely maximum concentration
achievable in the receptor fluid during the experiment (Pendlington,
2008). The solubility of the various salts of 8 and 12 was found to be
orders of magnitude higher than that required by the above guidelines.
During the experiments, the solution in the receptor chamber was
maintained at 32 ◦C (using a HAAKE P5 circulating water bath) and
stirred with Teflon-coated magnetic stirring bars. The skin was mounted
between the donor and receptor compartments with a pinch clamp with
the stratum corneum facing the donor compartment and the dermis
facing the receptor compartment (exposed area 3.46 cm2). All receptor
cells were filed with PBS (pH 7.4) and connected to peristaltic pumps
(Ismatec 2/12). The receptor solution was continuously pumped at a
flow rate of 166 µL/min. Donor compartments were covered with plastic
lids and the system was equilibrated for 20 min. After equilibration 2 mL
of the formulation was added to the donor compartments and the so-
lution was covered with a plastic lid. Receptor solution was continuously
collected into 5 mL tubes by an automatic fraction collector (ISCO
RETRIEVER IV) for 30 min intervals, up to 8 h, and immediately sealed.
The formulation in the donor compartment was left on the skin for 30
min and then removed by washing four times with PBS solution.
3.1. Initial screening of potential nucleophiles
The preliminary step of this work was to identify and evaluate po-
tential compounds, from pharmaceutical origin, for subcutaneous
neutralization of OP CWAs. Thus, we sought after Active Pharmaceutical
Ingredients (APIs) and excipients in the pharmaceutical and cosmetics
industries, that may fulfill the following criteria. 1) Such APIs must be
nucleophilic in order to react with the electrophilic OP’s. They must also
bear a functional group which is deprotonated at pH < 12 in order to
avoid skin damage. Such compounds may be hydroxyls (with emphasis
on phenols which have a lower pKa than aliphatic alcohols), carboxyl-
ates, amines, oximes and hydroxamates as their conjugated base would
serve as a strong nucleophile. 2) The APIs should be of the lowest mo-
lecular weight possible, as chemicals with a molecular weight greater
than ~500 Da do not penetrate the skin efficiently (the ‘rule of 500’) (Bos
and Meinardi, 2000). 3) The APIs should be the least toxic possible. 4)
Simple structured and non-expensive compounds are preferred. Based on
these criteria, we scanned several pharmaceutical databases: Drug Bank
web site (Wishart et al., 2018), Orange Book– FDA web site (“U.S. Food
and Drug Administration Orange Book.,”) and Handbook of Pharma-
ceutical Excipients (Rowe et al., 2013). Consequently, we located a
handful of APIs that meet the threshold requirements 1–3 and may be
used as potential nucleophiles for this purpose. Among those are phenols
(1, 2, 4, 7), amines (3, 5, 6) and hydroxamic acids (8–11) (Table 1).
Beyond the nucleophiles, additional key aspects also need to be
considered in the formulation design: (i) VX (15) has low solubility in
aqueous solutions (Yang et al., 1992). Therefore, the formulation has to
be based (at least partially) on an organic solvent that could solubilize
VX (15) in order to facilitate its hydrolysis. (ii) It is desirable that the
solvent would accelerate nucleophilic substitution reactions. This
should be achieved either by stabilizing the transition state or destabi-
lizing the ground state, as was previously established mechanistically
(Ashkenazi et al., 2010; Hamlin et al., 2018). (iii) It is advisable that the
solvent would serve as a permeation enhancer in order to increase the
permeation of the nucleophile through the stratum corneum (Williams
and Barry, 2004). (iv) The formulation should preferably solubilize high
quantities of the nucleophile, to assure sufficient hydrolytic and
permeation rates.
2.8. Sample analysis
The receptor samples from permeation experiments, collected over
0.5 h intervals, were analyzed using high-performance liquid chroma-
tography (Dionex Ultimate 3000). Aliquots of 10 µL or 40 µL from the
samples were injected into the column (Phenomenex, Gemini 5 µm C18
110 Å). The mobile phase consisting of a mixture containing acetonitrile
and water (1:99 and 50:50, v/v, for AHA and DAM respectively),
adjusted to pH 2 by phosphoric acid, was used as the eluent. The flow
rate was set to 0.7 mL/min and the column was kept at 35 ◦C. Detection
was performed at 210 nm, using a UV detector.
Keeping in mind the high toxicity of VX which requires special safety
measures, and the fact that 11 candidate APIs multiplied by 3–4 po-
tential cations (both organic and inorganic), all multiplied by a few
possible vehicles, would form a matrix of ca. dozens of kinetic experi-
ments, we decided to reduce the list of possible API’s by initially
examining their nucleophilic capabilities towards model OP’s. These
2.9. Data treatment and statistical analysis
Quantification of the nucleophiles was done using calibration curves,
which were generated for each salt by injecting known standard solu-
tions of the salts in PBS. The slopes were calculated from all the curves
4

V. Nahum et al.
International Journal of Pharmaceutics 603 (2021) 120689
Table 2
Table 3
Kinetic rates (t1/2) of 13 hydrolysis by compounds 1–12.*
Kinetic rates for the hydrolysis of 14 by sodium hydroxamates 8–11 and 12
(DAM) in different solvents.
Compound Number
Compound
number
Compound Name
Salt
Solvent
t1/2
Namea Name
t1/2 (min)/Solventsb
form
(min)
1
2
3
Methyl paraben
Butyl paraben
Na
DMSO
DMSO
DMSO
2.64
2.94
N.R.
AHA
DMSO
PG
EtOH
IPA
mPEG
Na
8
SHA
CHA
BUF
DAM
2.6
13.2
3.9
4.3
9.0
16.8
17.2
6.8
31.1
41.5
17.5
18.2
55.9
48.9
66.1
29.2
26.8
55.6
10.5
N.A.
7.7
Triethanolamine
Free
basea
Na
9
10
11
12
4
5
Aciclovir
Dapsone
DMSO
DMSO
102.48
N.R.
10.3
21.3
N.A.
N.A.
Free
basea
Free
basea
Na
a
6
Minoxidil
DMSO
N.R.
All nucleophiles were used as sodium salts.
b
All reactions were performed in a mixture of solvent/buffer 1:1.
7
8
Salicylamide
Acetohydroxamate
(AHA)
DMSO
DMSO
N.R.
Na
<1
2 × Nab
Na
DMSO
<1
<1
<1
<1
Table 4
9
Salicyl hydroxamate
(SHA)
pKa values of lead nucleophiles at DMSO/H2O mixtures.
10
11
Caprylhydroxamate
(CHA)
PG/H2O
1:1
% DMSO (v/v)*
pKa
AHA (8)
CHA (10)
BUF (11)
DAM (12)
Bufexamac (BUF)
Na
PG/H2O
1:1
0
9.55
10
9.6
9.3
9.6
12
*
DAM
Na
DMSO
20
50
80
10.3
11.7
13.7
10.2
11.4
13.4
10
11.4
13.3
11.4
13.6
In the absence of a nucleophile no observable decomposition in DMSO or
PG/H2O 1:1 at rt could be detected by 31P NMR, during the relevant time frame
(60 min.).
*
All pKa measurements performed in DMSO/H2O mixtures.
a
b
Amines were used as free base.
Salicyl hydroxamate (SHA) used as di-sodium salt. N.R-no reaction detected.
(Table 3). Among the tested solutions Buffer/DMSO 1:1 gave the highest
reaction rates and 8 was the most active nucleophile in this solution
hydrolyzing 14 with a half-life time of 2.6 min. In protic solvents (PG,
EtOH and IPA) the most active nucleophiles were 10 and 11, although
with rates lower than in DMSO. In all tested cases 9 and 12 were the least
active nucleophiles. Notably, in the absence of nucleophile (8–12) only a
minor decomposition of 14 could be observed at rt, with rates slower by
15–60 fold. This contribution to the decomposition, which can be
attributed to the basic pH of the buffer, is negligible during the relevant
time frame (minutes). As 9 was clearly found to be the inferior among all
other hydroxamic acids, we omitted it from the following studies.
OP’s (13 and 14) were chosen based on the fact that they are also
sparingly soluble in water (resembling VX) but possess a better leaving
group than VX (F and o-nitrophenol in comparison to di-i-propylami-
noethyl thiol). Thus, those nucleophiles who fail to cleave the model
compounds within sufficient time scale, would certainly be inadequate
for neutralizing VX. First, the selected nucleophiles were reacted with
13, a lipophilic G-type simulant, in DMSO (100 mM) (except for 10–11,
which were insoluble in DMSO and were therefore tested in PG/H2O 1:1
solution) and followed by 31P NMR spectroscopy. Diacetyl monooximate
(DAM, 12), the active ingredient of RSDL® , was used as a reference,
since it is approved for skin decontamination and has been widely tested
as a scavenger (Elsinghorst et al., 2015; Schwartz et al., 2012; Thors
et al., 2017b). Half-life times (t1/2) were calculated from the kinetic data
and are summarized in Table 2. Among the tested compounds,
hydroxamates (8–11) were the most active, rapidly degrading DBPF
with t1/2 < 1 min (due to the intrinsic NMR time scale limitations, more
accurate reaction rates for 8–11 cannot be determined). Phenols were
slightly less active, degrading 13 with a half-life of 2.64 min and 2.94
min for compounds 1 and 2, respectively. The rest of the tested com-
pounds showed very low activity, t1/2 = 102.48 min for sodium aciclovir
(4) or none with amines 3 and 5. As these rates are impractical for our
purposes, especially while taking into account that phosphonothiolates
are less prone to hydrolysis than phosphofluoridates (Marciano et al.,
2012; Yang et al., 1992), we excluded compounds 1–7 from any further
study. These results are in accordance with previous studies examining
hydroxamate ions as potential nucleophiles towards electrophilic cen-
ters such as organophosphorus nerve agents (Medeiros et al., 2012;
Mello et al., 2011; Orth et al., 2011, 2009; Satnami et al., 2010; Silva
et al., 2009; Swidler and Steinberg, 1956; Wong et al., 2019).
3.3. pKa measurements
Since the concentration of the deprotonated hydroxamate ion
Table 5
Log P and kinetic rates of VX hydrolysis by salts of AHA and DAM.
Entry
Nucleophile
Salt form
Conditions
LogP
t 1/2
(min)
1
DAM (12)
AHA (8)*
AHA (8)
AHA (8)
AHA (8)
AHA (8)
AHA (8)
AHA (8)
AHA (8)
AHA (8)
AHA (8)
AHA (8)
K (12a)
mPEG/H2O
9:1
ꢀ 0.635
ꢀ 2.17
ꢀ 2.17
ꢀ 2.17
ꢀ 2.42
ꢀ 2.06
ꢀ 2.06
–
5.71
2
K (8a)
mPEG/H2O
9:1
7.45
3
K (8a)
DMSO/H2O
1:1
5.14
4
K (8a)
DMSO/H2O
1:4
5.28
5
Na (8b)
DMSO/H2O
1:4
8.91
6
Diethylamine (8c)
Diethylamine (8c)
DMSO/H2O
1:4
6.78
7
DMSO/H2O
1:1
16.89
30.51
37.65
42.92
33.86
31.49
8
Triethylamine
(8d)
DMSO/H2O
1:1
3.2. Reaction condition optimization and solvent selection
9
Ethanolamine
(8e)
DMSO/H2O
1:1
–
In order to optimize reaction conditions and differentiate between
reaction rates, we synthesized another simulant, o-nitro-phenyl
diphenyl phosphate (ONPDPP, 14), which carries a nitrophenol group
and can be followed by UV spectroscopy. Hydrolytic activity of the
hydroxamates was tested in 0.2 mM solutions containing carbonate
buffer (pH 11) with five different solvents commonly used for topical
formulations (DMSO, PG, EtOH, IPA, and mPEG), at a 1:1 ratio
10
11
12
Diethanolamine
(8f)
DMSO/H2O
1:1
–
Epolamine (8g)
DMSO/H2O
1:1
–
Piperazine (8h)
DMSO/H2O
1:1
–
*
Log P of AHA acidic form is ꢀ 1.58.
5

V. Nahum et al.
International Journal of Pharmaceutics 603 (2021) 120689
Fig. 2. VX decomposition rates and rate constants (minꢀ 1) using: A) 12a in mPEG/H2O 9:1; B) 8a in mPEG/H2O 9:1; C) 8a in DMSO/ H2O 1:1; D) 8a in DMSO/ H2O
1:4 measured by 31P NMR spectroscopy. Inset, 31P NMR spectra of VX degradation by: A) 12a in mPEG/H2O 9:1; B) 8a in mPEG/H2O 9:1. The Low field signal
corresponds to VX and the high field signal corresponds to EMPA.
directly influences the rate of hydrolysis, the pKa values of nucleophiles
8, 10–12 were determined pH-metrically in three H2O/DMSO mixtures
(20%, 50%, 80% DMSO) in order to find the lowest pH sufficient for full
ionization of the hydroxamates (Table 4). For all hydroxamic acids the
pKa increased with the increase in DMSO fraction and it was in prox-
imity to 10 in 20% DMSO (v/v) and 11.4 in 50% DMSO (v/v). Therefore,
the ionization percent in formulation with pH 12 in 20% DMSO mixture
will be 99% while in 50% DMSO mixture only 80% will be ionized.
Attempts to dissolve 10 and 11 or their sodium and potassium salts in
H2O/DMSO or H2O/PG mixtures, in high concentrations (>50 mg/mL),
failed. DMSO is a polar aprotic solvent known to accelerate nucleophilic
comparison to the free acid 8 (logP = ꢀ 1.58, Table 5), as expected.
Therefore, such increased hydrophilicity of the salts may hinder the
permeation through the lipophilic skin. Nevertheless, it was previously
shown that high solubility, formation of organic ion-pairs and the use of
permeation enhancers can compensate for the low permeation due to
high hydrophilicity (Boroujerdi, 1987; Serajuddin, 2007; Williams and
Barry, 2004). Such ideas were successfully implemented in many com-
mercial formulations including diclofenac, a non-steroidal anti-inflam-
matory drug, e.g. Pennsaid® (45.5% DMSO formulation) and Voltaren
emugel® (diethyl amine salt) (Fini et al., 2012, 1999; Fuller and Roth,
2011). In order to test this concept, we prepared a variety of AHA ion-
pairs with aliphatic amines as counter ions (Table 5, Entries 7–13).
The diethyl ammonium salt showed slightly higher affinity to the lipo-
philic phase (logP = -2.06) than the inorganic salts of AHA, providing
some estimate to the relative permeation of the salts through the skin.
reactions, especially those involving α-nucleophiles (Ghosh et al., 2005).
In addition, DMSO is used as a permeation enhancer and approved for
human use (Williams and Barry, 2004). Therefore, due to the low sol-
ubility of CHA and BUF salts, we decided to exclude them from any
further study and continue exploring only DMSO based formulations of
AHA salts.
3.5. VX degradation efficiency
3.4. Log KO/W measurements
Based on the kinetic experiments and the pKa measurements, H2O/
DMSO solutions (0.2 M) were chosen as the solvent for VX degradation
experiments. The degradation of VX by salts of AHA (8a-h) and DAM
(12a) in H2O/DMSO and mPEG/H2O (RSDL® vehicle) solutions was
monitored using 31P NMR spectroscopy. Notably, VX remains intact for
Partition coefficient values of 12, 8 and several of their salts were
measured. The potassium salt of DAM was the least negative (logP =
ꢀ 0.635). The salt forms of AHA (8a-c) have lower logP values in
6

V. Nahum et al.
International Journal of Pharmaceutics 603 (2021) 120689
form EMPA is slower than the initial step of 8 under these conditions
(Fig. 2b) (Bierwisch et al., 2016). The degradation of VX by the potas-
sium salt 8a was as efficient as 12a under all conditions (t1/2
=
5.14–7.45 min, Table 5, entries 1–4 and Fig. 2, A-B), while the sodium
salt 8b was slightly slower (t1/2 = 8.91 min, Table 5, entry 5 and Fig. 3,
B). Organic salts of AHA (8c-h) in DMSO/H2O 1:1 showed slower rates
in comparison to the inorganic salts (Table 5, entries 7–12 and Fig. 4).
The diethylamine salt 8c was the most active among the tested organic
salts (t1/2 = 16.89 min). Interestingly, lowering the amount of DMSO in
the vehicle mixture increased the hydrolysis rate (Table 5, entries 6–7).
This influence can be explained by the fact that decreasing the relative
volume of DMSO in the solvent mixture increases the dielectric constant
of the solution, lowering the pKa value of 8 (see Table 4) and thus
enhancing the formation of the reactive ‘ion pair’ between the hydrox-
amate and the ammonium cation (Ghosh et al., 2005). This phenomenon
is obviously less pronounced for inorganic cations (Table 5, entries 3–4)
as the potassium is practically insoluble in DMSO. Nonetheless, it is clear
that reasonable hydrolysis rates of VX can be obtained with either
inorganic or DEA salts of AHA (Larsson et al., 2021).
Based on the above described kinetic results, together with the
measured partition coefficients (Tables 2, 3 and 5 and Figs. 2–4) and the
pKa values (Table 4), it became clear that among all the nucleophilic
API’s and solutions vehicles we scanned, the most promising candidates,
at least according to their chemical properties, would be AHA salts (8a
and 8c) dissolved in DMSO/H2O 1:4. Therefore, we proceeded to the
next step of measuring the skin permeation with only a handful of for-
mulations, consist of 8a or 8c as the active nucleophiles.
3.6. Skin permeation
The effect of solvent, counter ion and concentration on the pene-
tration of 8 through excised pig skin was studied with three different
solutions (mPEG/H2O 9:1, H2O/DMSO 4:1, H2O) in comparison to
RSDL®, using flow-through diffusion cells. Solutions of 8 or RSDL® (2
mL) were applied onto the pig skin for 30 min and then removed by
washing the skin four times with PBS solution (4 ×2 mL). Receptor
samples were collected each 30 min (8 h experiment) and analyzed by
HPLC. Steady-state absorption rates (Jss), 30 min accumulation (Q0.5),
total accumulation (Q8) and the lag time (TL) were calculated from the
results. Due to differences in the solubility of AHA salts in the tested
solutions and hence the concentrations, permeability coefficients (Kp)
which are the steady state Fluxes (Jss) normalized by concentration,
were calculated and are summarized in Table 6. Steady-states were
achieved in all cases after lag times spanning from 2.35 to 3.78 h (Fig. 5).
Steady-state absorption rates (Jss) for all solutions applied, except for 8a
in mPEG/H2O 9:1, were significantly larger than those for RSDL®. The
penetration of 8a in water and H2O/DMSO 4:1 was significantly larger
than that in mPEG/H2O 9:1 (P < 0.05). The results also show that in
comparison to water, 20% DMSO exhibited a marginally positive effect
on skin permeation rates of 8a, with an enhancement ratio of 1.13, while
90% mPEG exerted an inhibiting effect. Thus, it may be concluded that
AHA salts penetrate the skin better than DAM salts, especially when
dissolved in a highly polar media (i.e⋅H2O/DMSO mixture). This,
coupled with the fact that salts of 8 are by far less toxic than 12 (Ellin
and Henry Wills, 1964; Griffith and Musher, 1975), emphasizes the skin
scavenging potential of the former. We next tested the effect of the
concentration of 8a on its penetration. Three different solutions in H2O/
DMSO 4:1 were studied (217.2 mg/mL, 539.4 mg/mL and 880.9 mg/mL
- Table 6, entries 4–6, respectively). The most concentrated solution
showed the highest rate of absorption and total accumulation (Table 6,
entry 6 and Fig. 5A-B, blue lines). However, no linear correlation be-
tween concentration and rate of absorption/total accumulation was
observed. The more lipophilic 8c was also tested in two concentrations
(348 mg/mL and 548 mg/mL - Table 6, entries 7–8, respectively). The
absorption rate of the concentrated solution was significantly higher
than that of the less concentrated one. Moreover, this solution of 8c
Fig. 3. VX decomposition rates and rate constants (minꢀ 1) using: A) 8a in
DMSO/ H2O 1:4; B) 8b in DMSO/ H2O 1:4; C) 8c in DMSO/ H2O 1:4 measured
by 31P NMR spectroscopy.
many hours while dissolved in these solutions, which have a neutral pH
in the absence of the nucleophile, as may be anticipated based on to the
literature (Yang, 1999). Consequently, the first kinetic data point of t =
0 min. can be set to 100% VX. All reaction profiles exhibited pseudo-first
order kinetics (Figs. 2–4, S1–S30 in the SI). Following the reaction of
12a with VX the only and final degradation product was ethyl methyl-
phosphonic acid (EMPA) (Fig. 2a). Conversely, with 8 as the nucleo-
phile, an acetamidophosphonate intermediate was observed as well,
suggesting that the Lossen rearrangement step of the intermediate to
7

V. Nahum et al.
International Journal of Pharmaceutics 603 (2021) 120689
Fig. 4. VX decomposition rates and rate constants (minꢀ 1) using: A) 8c in DMSO/ H2O 1:1; B) 8d in DMSO/ H2O 1:1 measured by 31P NMR spectroscopy.
Table 6
The permeation parameters of 8 salts and RSDL® in different formulations.
Entry
Nucleophile
Vehicle
Jss (µg cm-2hꢀ 1
)
TL (h)
Q0.5 (µg cmꢀ 2
)
Q8 (µg cmꢀ 2
)
Co (mg mLꢀ 1
)
Kp (106 cm hꢀ 1
)
1
2
3
4
5
6
7
8
12a a
mPEG/H2O 9:1
mPEG/H2O 9:1
H2O
49.59 ± 6.77
122.52 ± 9.10
3.3 ± 0.07
0
234.15 ± 34.7
175.39
115.74b
205.75
217.24
538.39
880.92
347.81
547.73
282.72 ± 38.62
8a
2.35 ± 0.27**
3.78 ± 0.09
3.27 ± 0.21
3.19 ± 0.10
2.87 ± 0.05
3.25 ± 0.31
2.73 ± 0.21
2.89 ± 0.36***
0.31 ± 0.15
1.24 ± 0.23
0.38 ± 0.14
1.15 ± 0.23
2.58 ± 0.83***
1.62 ± 0.39*
695.33 ± 85.41
1058.62 ± 78.59*
1822.78 ± 123.47***
1949.17 ± 242.73***
649.49 ± 150.04
8a
375.05 ± 25.40*
423.39 ± 52.73**
350.32 ± 80.93*
621.65 ± 39.45***
448.20 ± 87.61**
1181.83 ± 126.79***
1581.11 ± 95.21
2036.42 ± 327.54*
1679.32 ± 371.33
3168.70 ± 199.73***
2212.41 ± 510.67*
6253.15 ± 883.93***
8a
H2O/DMSO 4:1
H2O/DMSO 4:1
H2O/DMSO 4:1
H2O/DMSO 4:1
H2O/DMSO 4:1
8a
8a
705.68 ± 44.78
8c
1288.63 ± 251.89**
2157.69 ± 231.48***
8c
Abbreviations: Jss, Steady state flux; TL, lag time; Q0.5, amount permeated after 0.5 h; Q8, amount permeated after 8 h; Co, initial concentration; Kp, permeation co-
efficient. *P < 0.05, **P < 0.01, ***P < 0.001 compared to the corresponding value of RSDL®. Values are presented as mean ± SEM (n = 4).
a
RSDL® contains in addition to DAM K (Dekon 139) 5% of the DAM acidic form.
b
Due to low solubility maximal concentration of AHA K in mPEG/H2O 9:1 were used and not 200 mg mL-1.
Fig. 5. Penetration profile of 8 in different formulations through excised pig skin in comparison to RSDL®. A) Amount penetrated and B) penetration rate following
application of different formulations for 30 min; RSDL® (black), 8a in mPEG/H2O 9:1 (orange), 8a in H2O (gray), 8a (217 mg/mL) in H2O/DMSO 4:1(yellow), 8a
(538 mg/mL) in H2O/DMSO 4:1(red), 8a (881 mg/mL) in H2O/DMSO 4:1(blue), 8c (348 mg/mL) in H2O/DMSO 4:1 (green), 8c (548 mg/mL) in H2O/DMSO 4:1
(purple). Values are presented as mean ± SEM (n = 4).
showed the highest rate of absorption (Jss = 1181.83 ± 126.79 µg cm-
2hꢀ 1), total accumulation (Q8 = 6253.15 ± 883.93 µg cmꢀ 2) and
permeation coefficient (Kp = 2157.69 ± 231.48 cm hꢀ 1 × 10-6) among
all tested formulations (Table 6, entries 1–8 and Fig. 5A-B, purple lines).
It is noteworthy that the lag time (TL, approx. 3 hr) is fairly consistent
regardless of the different formulations. However, a delayed treatment
might require that a substantial amount of scavenger would penetrate
the skin as promptly as possible. In this regard, using a highly concen-
trated and more lipophilic formula may accelerate the nucleophile’s
diffusion and induce VX hydrolysis even before steady-state absorption
is reached (e.g.; high concentration of 8c in H2O/DMSO 4:1).
permeate from the RSDL® lotion (mPEG/H2O 9:1 solution). However,
the permeation rate that we measured (49.59 ± 6.77 µg cm-2hꢀ 1) is
substantially lower than that expected for efficient VX skin depot
neutralization. Indeed, it is possible that RSDL® was designed this way
in order to prevent the permeation of its toxic ingredients, while sus-
taining good hydrolytic activity for external skin decontamination
(Connolly et al., 2020; Wong et al., 2020). Previously, Vallet and co-
workers showed that VX absorption rates through pig and human skin
under finite conditions are 90.5–129 µg cm-2hꢀ 1 and 45.5–57.5 µg cm-
2hꢀ 1, respectively (Vallet et al., 2008). A similar study conducted by
Dalton et. al showed even higher rates of pig and human skin perme-
ation, albeit under infinite VX dose application conditions (207 ± 62 and
Permeation results demonstrated that 12a can, to some extent,
8

V. Nahum et al.
International Journal of Pharmaceutics 603 (2021) 120689
therapy” against percutaneous intoxication by low volatility OP nerve
agents such as VX, as part of the medical treatment against such agents.
These lotions are comprised of acetohydroxamic salts as their active
ingredients and a mixture of solvents (H2O/DMSO) that allow the hy-
drophilic salt to penetrate through the skin as a vehicle. As the hydro-
lysis rates of VX of such lotions are similar to that obtained by the
contemporary active decontamination solution RSDL®, they are ex-
pected to perform a double action of both external and within skin
decontamination. In this regard it is noteworthy that although the
excised pig skin model may serve as a useful tool, especially in
comparative studies of similar lotions of the kind presented in our work,
this in-vitro model also has limitations. Specifically, as the skin pene-
tration occurs only by diffusion and the integrity of the cells after at least
one freeze–thaw cycle is unclear, the data cannot be used to accurately
predict the in-vivo situation, both in animals and humans. Accordingly,
we conducted such in-vivo studies in domestic pigs which were exposed
percutaneously to VX and were treated with lotions of 8a and 8c,
combined with standard antidotal treatment. Indeed, in these studies the
addition of the above “catch-up therapy” lotions substantially improved
the recovery of the exposed animals (manuscript in preparation).
Table 7
Amount of AHA salts absorbed into the skin after 30 min.
Entry
Nucleophile
Vehicle
Co (mg
A0.5 a (mg
% of applied
dose
mLꢀ 1
)
cmꢀ 2
)
1
2
8
3
4
5
6
7
12a
8a
8a
8a
8a
8a
8c
mPEG/
H2O 9:1
mPEG/
H2O 9:1
H2O
175.39
115.74
205.75
217.24
538.39
880.92
347.81
547.73
3.51 ± 1.92
3.46 ± 1.90
37.40 ± 7.46
31.64 ± 8.85
4.49 ± 0.92
5.13 ± 0.89
12.40 ± 0.55
11.44 ± 2.04
8.93 ± 0.84
25.02 ±
4.99
37.63 ±
10.53**
20.25 ±
14.65
H2O/
DMSO 4:1
H2O/
15.99 ±
2.76
DMSO 4:1
H2O/
63.13 ±
2.81***
23.00 ±
4.10
DMSO 4:1
H2O/
DMSO 4:1
H2O/
8c
28.28 ±
2.65*
DMSO 4:1
a
Absorbed amount was calculated by subtracting the amount of 8 in the donor
compartment after 30 min (removed from skin by washing with PBS four times)
from the applied amount. *P < 0.05, **P < 0.01, ***P < 0.001 compared to the
corresponding value of RSDL®.
Funding sources
333 ± 226 µg cm-2hꢀ 1 respectively) (Dalton et al., 2006). In such a case,
considerably higher rates of nucleophile permeation would be required
in order to neutralize a substantial amount of the depot and minimize
intoxication effects. In this regard, the permeation rates of 8a (621.65 ±
39.45 µg cm-2hꢀ 1) and 8c (1181.83 ± 126.79 µg cm-2hꢀ 1) in H2O/DMSO
4:1 were found to be significantly higher than those of 12a and VX.
Hence, based on the abovementioned VX absorption rates, skin appli-
cation of these formulations for 30 min should allow nucleophile
permeation at a rate and accumulated amount sufficient to neutralize
the expected depot.
This work was supported internally by IIBR.
CRediT authorship contribution statement
Victoria Nahum: Conceptualization, Methodology, Validation,
Investigation, Writing - original draft, Writing - review & editing,
Visualization. Uri Nili: Conceptualization, Methodology, Formal anal-
ysis, Writing - original draft, Writing - review & editing. Eugenia Bloch-
Shilderman: Conceptualization, Methodology, Formal analysis, Re-
sources, Writing - original draft, Writing - review & editing. Boris
Smolkin: Conceptualization, Methodology, Validation, Investigation,
Writing - original draft, Writing - review & editing, Visualization, Su-
pervision. Nissan Ashkenazi: Conceptualization, Methodology, Vali-
dation, Investigation, Writing - original draft, Writing - review & editing,
Visualization, Supervision, Project administration.
The permeation of 8a in mPEG/H2O 9:1 was not significantly higher
than that of RSDL®, emphasizing the important role of the vehicle so-
lution. Replacing it with water or H2O/DMSO 4:1 significantly improved
the permeation by an order of magnitude and more. High concentrations
of the permeate and the use of an organic ion pair (as in 8c) resulted in
further improvement.
As mentioned above, in a realistic scenario, a “catch up” therapy
utilizing a scavenger of the type described above will most likely be
employed only in the case of intoxication signs appearance, in
conjunction with an antidotal treatment. In such cases, a considerable
amount of VX will have already accumulated as a dermal depot even
before application of the scavenger. Therefore, another significant
parameter expected to influence the efficacy of the scavenger, is the
amount of scavenger absorbed and accumulated inside the skin (i.e.,
trapped in the dermal component) within a short time window. In this
regard, the amount of 8a, 8c absorbed in the skin 30 min following
application (A0.5) was significantly higher using formulations with high
permeation rates in comparison to RSDL® (Table 7 entries 5, 7).
Furthermore, it is safe to assume that these amounts of nucleophiles
Declaration of Competing Interest
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.
Appendix A. Supplementary material
Supporting information contains kinetic measurements (NMR,
UV–Vis) for the degradation of VX and other OP’s, pKa measurements
and skin permeation experimental data. Supplementary data to this
article can be found online at https://doi.org/10.1016/j.ijpharm.2021.
120689.
would be sufficient to efficiently hydrolyze a finite dose (<10 mg cmꢀ 2
)
of active VX residing in a dermal reservoir.
References
Last, in regards to practicality, it is worth mentioning that Aceto-
hydroxamic acid (8) is of low toxicity (chronic per os daily dose of up to
15 mg/kg/day is approved for clinical use) orphan drug (Lithostat)
approved by the FDA for the oral treatment of urinary infections
(Lithostat [package insert]. San Antonio, TX:Mission Pharmacel; 2015),
and that both DMSO and diethylamine are approved for human use and
are currently in use in different formulations of diclofenac.
Ashkenazi, N., Segall, Y., Chen, R., Sod-Moriah, G., Fattal, E., 2010. The Mechanism of
Nucleophilic Displacements at Phosphorus in Chloro-Substituted
Methylphosphonate Esters: Pꢀ O vs Pꢀ C Bond Cleavage: A DFT Study. J. Org. Chem.
75, 1917–1926. https://doi.org/10.1021/jo9026325.
Bierwisch, A., Koller, M., Worek, F., Kubik, S., 2016. Pathways for the Reactions Between
Neurotoxic Organophosphorus Compounds and Oximes or Hydroxamic Acids. Eur. J.
Org. Chem. 2016, 5831–5838. https://doi.org/10.1002/ejoc.201601053.
Bloch-Shilderman, E., Yacov, G., Cohen, L., Egoz, I., Gutman, H., Gez, R., Rabinovitz, I.,
Nili, U., 2019. Repetitive antidotal treatment is crucial in eliminating eye pathology,
respiratory toxicity and death following whole-body VX vapor exposure in freely
moving rats. Arch. Toxicol. 93, 1365–1384. https://doi.org/10.1007/s00204-019-
02401-0.
4. Conclusion
We described here nucleophilic lotions based on approved drugs and
Boroujerdi, M., 1987. Kinetics of Ion-Pair Absorption. Drug Dev. Ind. Pharm. 13,
181–191. https://doi.org/10.3109/03639048709040163.
vehicles, which may be useful for skin application as a “catch-up
9

V. Nahum et al.
International Journal of Pharmaceutics 603 (2021) 120689
Bos, J.D., Meinardi, M.M., 2000. The 500 Dalton rule for the skin penetration of chemical
compounds and drugs. Exp. Dermatol. 9, 165–169. https://doi.org/10.1034/j.1600-
0625.2000.009003165.x.
Munro, N., 1994. Toxicity of the Organophosphate Chemical Warfare Agents GA, GB, and
VX: Implications for Public Protection. Environ. Health Perspect. 102, 18–37.
https://doi.org/10.1289/ehp.9410218.
Chilcott, R.P., Dalton, C.H., Hill, I., Davison, C.M., Blohm, K.L., Clarkson, E.D.,
Hamilton, M.G., 2005a. In vivo skin absorption and distribution of the nerve agent
VX (O–ethyl–S–[2(diisopropylamino)ethyl] methylphosphonothioate) in the
domestic white pig. Hum. Exp. Toxicol. 24, 347–352. https://doi.org/10.1191/
0960327105ht537oa.
Norlin, R., Juhlin, L., Lind, P., Trogen, L., 2005. α-Haloenamines as Reagents for the
Conversion of Phosphorus Oxyacids to Their Halogenated Analogues. Synthesis
2005, 1765–1770. https://doi.org/10.1055/s-2005-865363.
OECD/OCDE, 1995. Test No. 107: Partition Coefficient (n-octanol/water): Shake Flask
Method, Giudance, OECD Guidelines for the Testing of Chemicals, Section 1. OECD.
https://doi.org/10.1787/9789264069626-en.
Chilcott, R.P., Dalton, C.H., Hill, I., Davison, C.M., Blohm, K.L., Clarkson, E.D.,
Hamilton, M.G., 2005b. Evaluation of a Barrier Cream against the Chemical Warfare
Agent VX using the Domestic White Pig. Basic Clin. Pharmacol. Toxicol. 97, 35–38.
https://doi.org/10.1111/j.1742-7843.2005.pto_97106.x.
Orth, E.S., da Silva, P.L.F., Mello, R.S., Bunton, C.A., Milagre, H.M.S., Eberlin, M.N.,
Fiedler, H.D., Nome, F., 2009. Suicide Nucleophilic Attack: Reactions of
Benzohydroxamate Anion with Bis(2,4-dinitrophenyl) Phosphate. J. Org. Chem. 74,
5011–5016. https://doi.org/10.1021/jo9007354.
Connolly, J.M., Stevenson, R.S., Railer, R.F., Clark, O.E., Whitten, K.A., Lee-Stubbs, R.B.,
Anderson, D.R., 2020. Impairment of wound healing by reactive skin
Orth, E.S., Medeiros, M., Bortolotto, T., Terenzi, H., Kirby, A.J., Nome, F., 2011.
Dephosphorylation Reactions with Deferoxamine, A Potential Chemical Nuclease.
J. Org. Chem. 76, 10345–10348. https://doi.org/10.1021/jo202074y.
Pendlington, R.U., 2008. Principles and Practice of Skin Toxicology. In: Chilcott, R.P.,
Price, S. (Eds.), John Wiley & Sons Ltd., pp. 129–148.
¨
decontamination lotion (RSDL ®) in a Gottingen minipig ® model. Cutan. Ocul.
Toxicol. 39, 143–157. https://doi.org/10.1080/15569527.2020.1751183.
Costanzi, S., Machado, J.-H., Mitchell, M., 2018. Nerve Agents: What They Are, How
They Work, How to Counter Them. ACS Chem. Neurosci. 9, 873–885. https://doi.
org/10.1021/acschemneuro.8b00148.
Romano, J.A., Jr. Lukey, B.J., Salem, H., 2008. Chemical Warfare agents: Chemistry,
Pharmacology, Toxicology, and Therapeutics., 2nd ed. CRC Press Inc.
Rowe, R.C., Sheskey, P.J., Cook, W.G., Quinn, M.E., 2013. Handbook of Pharmaceutical
Excipients –. Pharm. Dev. Technol. 18 https://doi.org/10.3109/
10837450.2012.751408, 544–544.
Dalton, C.H., Hattersley, I.J., Rutter, S.J., Chilcott, R.P., 2006. Absorption of the nerve
agent VX (O-ethyl-S-[2(di-isopropylamino)ethyl] methyl phosphonothioate) through
pig, human and guinea pig skin in vitro. Toxicol. Vitr. 20, 1532–1536. https://doi.
org/10.1016/j.tiv.2006.06.009.
Ellin, R.I., Henry Wills, J., 1964. Oximes Antagonistic to Inhibitors of Cholinesterase.
J. Pharm. Sci. 53, 1143–1150. https://doi.org/10.1002/jps.2600531002.
Elsinghorst, P.W., Worek, F., Koller, M., 2015. Detoxification of organophosphorus
pesticides and nerve agents through RSDL: Efficacy evaluation by 31P NMR
spectroscopy. Toxicol. Lett. 233, 207–213. https://doi.org/10.1016/j.
toxlet.2014.12.004.
Satnami, M.L., Dhritlahre, S., Nagwanshi, R., Karbhal, I., Ghosh, K.K., Nome, F., 2010.
Nucleophilic Attack of Salicylhydroxamate Ion at C═O and P═O Centers in Cationic
Micellar Media. J. Phys. Chem. B 114, 16759–16765. https://doi.org/10.1021/
jp1066549.
Schwartz, M.D., G. Hurst, C., A. Kirk, M., J.D. Reedy, S., H. Braue, E., 2012. Reactive Skin
Decontamination Lotion (RSDL) for the Decontamination of Chemical Warfare Agent
(CWA) Dermal Exposure. Curr. Pharm. Biotechnol. 13, 1971–1979. https://doi.org/
10.2174/138920112802273191.
Fini, A., Bassini, G., Monastero, A., Cavallari, C., Fini, A., Bassini, G., Monastero, A.,
Cavallari, C., 2012. Diclofenac Salts, VIII. Effect of the Counterions on the
Permeation through Porcine Membrane from Aqueous Saturated Solutions.
Pharmaceutics 4, 413–429. https://doi.org/10.3390/pharmaceutics4030413.
Fini, A., Fazio, G., Gonzalez-Rodriguez, M., Cavallari, C., Passerini, N., Rodriguez, L.,
1999. Formation of ion-pairs in aqueous solutions of diclofenac salts. Int. J. Pharm.
187, 163–173. https://doi.org/10.1016/S0378-5173(99)00180-5.
Fuller, P., Roth, S., 2011. Diclofenac sodium topical solution with dimethyl sulfoxide, a viable
alternative to oral nonsteroidal anti-inflammatories in osteoarthritis: review of current
evidence. J. Multidiscip. Healthc. 4, 223. https://doi.org/10.2147/JMDH.S23209.
Ghosh, K.K., Lal Satnami, M., Sinha, D., Vaidya, J., 2005. Solvent effects on the
nucleophilic substitution reactions of p-nitrophenyl acetate with hydroxamate ions.
J. Mol. Liq. 116, 55–60. https://doi.org/10.1016/j.molliq.2004.06.003.
Griffith, D.P., Musher, D.M., 1975. Acetohydroxamic acid. Urology 5, 299–302. https://
doi.org/10.1016/0090-4295(75)90142-9.
Serajuddin, A.T.M., 2007. Salt formation to improve drug solubility. Adv. Drug Deliv.
Rev. 59, 603–616. https://doi.org/10.1016/j.addr.2007.05.010.
Silva, M., Mello, R.S., Farrukh, M.A., Venturini, J., Bunton, C.A., Milagre, H.M.S.,
Eberlin, M.N., Fiedler, H.D., Nome, F., 2009. The Mechanism of Dephosphorylation
of Bis(2,4-dinitrophenyl) Phosphate in Mixed Micelles of Cationic Surfactants and
Lauryl Hydroxamic Acid. J. Org. Chem. 74, 8254–8260. https://doi.org/10.1021/
jo9017638.
Swidler, R., Steinberg, G.M., 1956. The Kinetics of the Reaction of Isopropyl
Methylphosphonofluoridate (Sarin) with Benzohydroxamic Acid 1. J. Am. Chem.
Soc. 78, 3594–3598. https://doi.org/10.1021/ja01596a011.
Thiermann, H., Aurbek, N., Worek, F., 2016. CHAPTER 1. Treatment of Nerve Agent
Poisoning. In: Franz Worek, John Jenner, H.T. (Ed.), pp. 1–42. https://doi.org/
10.1039/9781782628071-00001.
¨
Hamilton, M.G., Hill, I., Conley, J., Sawyer, T.W., Caneva, D.C., Lundy, P.M., 2004.
Clinical aspects of percutaneous poisoning by the chemical warfare agent VX: effects
of application site and decontamination. Mil. Med. 169, 856–862. https://doi.org/
10.7205/milmed.169.11.856.
Thors, L., Koch, M., Wigenstam, E., Koch, B., Hagglund, L., Bucht, A., 2017a. Comparison
of skin decontamination efficacy of commercial decontamination products following
exposure to VX on human skin. Chem. Biol. Interact. 273, 82–89. https://doi.org/
10.1016/j.cbi.2017.06.002.
¨
Hamlin, T.A., Swart, M., Bickelhaupt, F.M., 2018. Nucleophilic Substitution (SN2):
Dependence on Nucleophile, Leaving Group, Central Atom, Substituents, and
Solvent. ChemPhysChem 19, 1315–1330. https://doi.org/10.1002/
cphc.201701363.
Thors, L., Lindberg, S., Johansson, S., Koch, B., Koch, M., Hagglund, L., Bucht, A., 2017b.
RSDL decontamination of human skin contaminated with the nerve agent VX.
Toxicol. Lett. 269, 47–54. https://doi.org/10.1016/j.toxlet.2017.02.001.
U.S. Food and Drug Administration Orange Book. [WWW Document], URL https://www.
accessdata.fda.gov/scripts/cder/ob/index.cfm.
Jones, C.S., Bull, S.D., Williams, J.M.J., 2016. DBN hexafluorophosphate salts as
convenient sulfonylating and phosphonylating agents. Org. Biomol. Chem. 14,
8452–8456. https://doi.org/10.1039/C6OB01854H.
Vallet, V., Cruz, C., Licausi, J., Bazire, A., Lallement, G., Boudry, I., 2008. Percutaneous
penetration and distribution of VX using in vitro pig or human excised skin.
Toxicology 246, 73–82. https://doi.org/10.1016/j.tox.2007.12.027.
Wetherell, J.R., Armstrong, S.J., Read, R.W., Clough, G.F., 2008. VX Penetration
Following Percutaneous Poisoning: A Dermal Microdialysis Study in the Guinea Pig.
Toxicol. Mech. Methods 18, 313–321. https://doi.org/10.1080/
15376510701884944.
Joosen, M.J.A., van den Berg, R.M., de Jong, A.L., van der Schans, M.J., Noort, D.,
Langenberg, J.P., 2017. The impact of skin decontamination on the time window for
effective treatment of percutaneous VX exposure. Chem. Biol. Interact. 267, 48–56.
https://doi.org/10.1016/j.cbi.2016.02.001.
Joosen, M.J.A., van der Schans, M.J., Kuijpers, W.C., van Helden, H.P.M., Noort, D.,
2013. Timing of decontamination and treatment in case of percutaneous VX
poisoning: A mini review. Chem. Biol. Interact. 203, 149–153. https://doi.org/
10.1016/j.cbi.2012.10.002.
Williams, A.C., Barry, B.W., 2004. Penetration enhancers. Adv. Drug Deliv. Rev. 56,
603–618. https://doi.org/10.1016/j.addr.2003.10.025.
Wishart, D.S., Feunang, Y.D., Guo, A.C., Lo, E.J., Marcu, A., Grant, J.R., Sajed, T.,
Johnson, D., Li, C., Sayeeda, Z., Assempour, N., Iynkkaran, I., Liu, Y.,
Maciejewski, A., Gale, N., Wilson, A., Chin, L., Cummings, R., Le, Di, Pon, A.,
Knox, C., Wilson, M., 2018. DrugBank 5.0: a major update to the DrugBank database
for 2018. Nucleic Acids Res. 46, D1074–D1082. https://doi.org/10.1093/nar/
gkx1037.
Joosen, M.J.A., van der Schans, M.J., van Helden, H.P.M., 2010. Percutaneous exposure
to the nerve agent VX: Efficacy of combined atropine, obidoxime and diazepam
treatment. Chem. Biol. Interact. 188, 255–263. https://doi.org/10.1016/j.
cbi.2010.06.010.
¨
¨
Larsson, A., Qvarnstrom, J., Lindberg, S., Wigenstam, E., Oberg, L., Afshin Sander, R.,
Johansson, S., Bucht, A., Thors, L., 2021. In vitro human skin decontamination
efficacy of MOF-808 in decontamination lotion following exposure to the nerve
agent VX. Toxicol. Lett. 339, 32–38. https://doi.org/10.1016/j.toxlet.2020.12.014.
Lithostat [package insert]. San Antonio, TX:Mission Pharmacel; 2015.
Marciano, D., Columbus, I., Elias, S., Goldvaser, M., Shoshanim, O., Ashkenazi, N.,
Zafrani, Y., 2012. Role of the P-F Bond in Fluoride-Promoted Aqueous VX
Hydrolysis: An Experimental and Theoretical Study. J. Org. Chem. 77,
10042–10049. https://doi.org/10.1021/jo301549z.
Wong, P.T., Bhattacharjee, S., Cannon, J., Tang, S., Yang, K., Bowden, S., Varnau, V.,
O’Konek, J.J., Choi, S.K., 2019. Reactivity and mechanism of α-nucleophile scaffolds
as catalytic organophosphate scavengers. Org. Biomol. Chem. 17, 3951–3963.
https://doi.org/10.1039/C9OB00503J.
Wong, P.T., Tang, S., Cannon, J., Yang, K., Harrison, R., Ruge, M., O’Konek, J.J., Choi, S.
K., 2020. Shielded α-Nucleophile Nanoreactor for Topical Decontamination of
Reactive Organophosphate. ACS Appl. Mater. Interfaces 12, 33500–33515. https://
doi.org/10.1021/acsami.0c08946.
Medeiros, M., Orth, E.S., Manfredi, A.M., Pavez, P., Micke, G.A., Kirby, A.J., Nome, F.,
2012. Dephosphorylation Reactions of Mono-, Di-, and Triesters of 2,4-Dinitrophenyl
Phosphate with Deferoxamine and Benzohydroxamic Acid. J. Org. Chem. 77,
10907–10913. https://doi.org/10.1021/jo302374q.
Yang, Y.-C., 1999. Chemical Detoxification of Nerve Agent VX. Acc. Chem. Res. 32,
109–115. https://doi.org/10.1021/ar970154s.
Yang, Y.-C., Baker, J.A., Ward, J.R., 1992. Decontamination of chemical warfare agents.
Chem. Rev. 92, 1729–1743. https://doi.org/10.1021/cr00016a003.
Mello, R.S., Orth, E.S., Loh, W., Fiedler, H.D., Nome, F., 2011. Polymers Containing
Hydroxamate Groups: Nanoreactors for Hydrolysis of Phosphoryl Esters. Langmuir
27, 15112–15119. https://doi.org/10.1021/la203437j.
10