Novel application of simple molybdates: Catalytic hydrolysis of an organophosphate neurotoxin under mild aqueous conditions

Article abstract of DOI:10.1016/j.ica.2017.05.007

A novel protocol for hydrolyzing phosphonothioate neurotoxins has been developed that uses a readily available, inexpensive and non-toxic molybdate (MoO₄^2?). The target organophosphate is O,S-diethylphenyl phosphonothioate (1), a mod

Full text of DOI:10.1016/j.ica.2017.05.007

Accepted Manuscript
Research paper
“
Novel application of simple molybdates: catalytic hydrolysis of an organo-
phosphate neurotoxin under mild aqueous conditions”
Louis Y. Kuo, Kristina M. Dill, Yusef A. Shari'ati, Emily K. Bright, Theresa
McCormick
PII:
S0020-1693(17)30719-3
http://dx.doi.org/10.1016/j.ica.2017.05.007
ICA 17576
DOI:
Reference:
To appear in:
Inorganica Chimica Acta
Received Date:
Revised Date:
Accepted Date:
19 July 2016
11 April 2017
6 May 2017
Please cite this article as: L.Y. Kuo, K.M. Dill, Y.A. Shari'ati, E.K. Bright, T. McCormick, “Novel application of
simple molybdates: catalytic hydrolysis of an organophosphate neurotoxin under mild aqueous conditions”,
Inorganica Chimica Acta (2017), doi: http://dx.doi.org/10.1016/j.ica.2017.05.007
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

“Novel application of simple molybdates: catalytic hydrolysis of an organophosphate
neurotoxin under mild aqueous conditions”
a
Louis Y. Kuo,* Kristina M. Dill, Yusef A. Shari’ati, Emily K. Bright and Theresa
a
a
a
b
McCormick
a
Department of Chemistry
Lewis & Clark College
Portland, OR 97219
b
Department of Chemistry
Portland State University
Portland, OR 97207
1

Abstract
A novel protocol for hydrolyzing phosphonothioate neurotoxins has been developed that
2
-
4
uses a readily available, inexpensive and non-toxic molybdate (MoO ). The target
organophosphate is O,S-diethylphenyl phosphonothioate (1), a model compound that has
served as an analog of the chemical warfare agent VX. Molybdate-mediated hydrolysis
of 1 proceeds at room temperature (pH ~7.5) and yields a relatively safe phosphonate
product through P-S specific scission. This is the first report of utilizing the molybdate,
2
-
‡
MoO
4
, to degrade an organophosphate neurotoxin with turnover. A ∆S of -81.5 J/
1
8
mol·K (-19.5 cal/mol·K) indicates a bimolecular process, and O-labelling studies show
2
-
no oxygen exchange of 1 throughout the hydrolysis by MoO
4
(aq). We hypothesize
primarily the monomeric molybdate is the active species at pH 7.5 where a S
N
2(P)
mechanism takes place. Density functional theory methods suggest the molybdate
oxoanion serves as the nucleophile to attack phosphonothioate 1 to form a molybdate-
phosphonate anhydride that subsequently hydrolyzes to the starting molybdate and
phenylethyl phosphonate.
2

Introduction
We report the first application of the molybdate dianion (MoO
hydrolyzing an organophosphate neurotoxin. Phosphonothioates (Figure 1) of the form
2
-
4
) [1] for
1
2
3
R P(O)(SR )(OR ) are acetylcholine esterase inhibitors [2] that have applications as
pesticides [3,4,5], and most notoriously as chemical warfare agents [6,7]. These
compounds are extremely dangerous to human health. For example, the nerve agent VX
1
2
3
(
3 2 2 2
R =CH , R = CH CH N(iPr) , R = Et) has an LD50 = 0.015 mg/kg [8], which means
that just two drops would kill an average adult.
Figure 1. Structures of various organophosphorus compounds.
Organophosphate pesticide exposure has attracted increasing apprehension as a public
health concern.[6] Structurally similar to related chemical warfare agents, these
pesticides contain analogous P-O and P-S ligations.[6] For example, the
phosphorothioate pesticide parathion [9] (ArOP(S)(OEt)
2
) undergoes tautomerization
(
Scheme 1)[10] to form a product with the same P-S and P-O ligation as VX [11, 12].
Scheme 1. Tautomerization of parathion
S
O
P
O N
2
O
P
O
R
O N
2
O
S
R
OR
OR
3

Therefore, effective protocols for breaking down phosphonothioates such as VX and
common pesticides have major environmental and societal benefits.
Several efforts towards degrading phosphonothioates with discrete inorganic
complexes have been reported. Lanthanide(III) – catalyzed methanolysis is an effective
tool for degrading phosphonothioates, [13,14,15] and the rapid nature of this protocol has
been capitalized on live nerve agent (i.e. VX) destruction [16]. The fluoride anion also
hydrolyzes VX, forming an intermediate phosphonofluoridate that subsequently
undergoes hydrolysis to yield a phosphonate [17].
The compound O,S-diethyl phenyl phosphonothioate (1) [18] is an analog of the
chemical warfare agent VX; both 1 and VX have P-O and P-S linkages with similar
hydrolytic labilities [19]. In the hydrolytic degradation of phosphonothioates, P-S
scission is the preferred avenue as it yields a less toxic phosphonate. Specifically,
alkaline hydrolysis of VX (and 1) in 0.10 M NaOH proceeds 15-18% of the time through
P-O scission to yield another phosphonothioate (EA 2192) that is as toxic as the starting
VX (Scheme 2) [20].
Scheme 2. Alkaline hydrolysis of VX in 0.10 M NaOH yields both phosphonate (EMPA)
and toxic phosphonothioate (EA 2192).
4

Moreover, in alkaline solution the toxic EA 2192 product is an anion that is further
resistant to hydrolysis by the hydroxide nucleophile. Thus, reagents such as
iodosobenzoate (IBA) [21,22] that promote P-S scission of phosphonothioates are
desirable complexes for degrading organophosphate neurotoxins.
The only molybdenum complexes that hydrolyze phosphonothioates are the
5
metallocenes of the form Cp
2
MoCl
2
(Cp = η -C
5
5
H ) [23]. Their hydrolytic activity is
characterized by P-S specific scission of phosphonothioate 1 under mild conditions (pH
º
, 30 C) [24]. However, the thiophilic Mo(IV) results in strong Cp
IV
7
2
Mo -SEt ligation
Mo(SEt)
MoCl (aq); the
[
25,26,27] with the alkylthiolate leaving group of 1 [24]. The subsequent Cp
2
2
adduct (Scheme 3) is inert and prevents additional hydrolysis of 1 by Cp
2
2
metallocene functions not as a catalyst but merely as a reactant.[28].
2 2
Scheme 3. Hydrolysis of 1 by Cp MoCl (aq).
Therefore, to improve phosphonothioate hydrolysis by molybdenum complexes,
we require an inexpensive and air-stable system that does not have a soft thiophilic center
2
-
[
4
29]. To that end, we report the activity of the molybdate dianion, MoO , towards 1
hydrolysis with the desired P-S specific bond scission [30]. This report delineates the
chemistry of phosphonothioate hydrolysis by the simple molybdate dianion. This work
2
-
was prompted by prior results that showed polyoxomolybdates and MoO
4
hydrolyze
organophosphates [31,32,33] and simple peptides [34]. Early molybdate-promoted
hydrolysis of organic phosphate bonds (i.e. ATP, creatine phosphate, fructuose-6-
5

phosphate, and acetyl phosphate) were done under acidic conditions [35,36,37]. In that
connection, molybdates were also shown to affect several acid phosphatases (i.e. yeast
and tomato) [37,38], and this was even implicated as a possible metabolic role for
molybdenum [39,40,41,42].
Results and Discussion
Our studies of the molybdate-mediated hydrolysis of 1 begin by showing that
2
-
MoO
1) the absence of leaving group coordination to the molybdate dianion (i.e. MoO
functions as a catalyst), (2) structure-activity relationship studies with modified
4
hydrolyzes 1 with the desired P – S bond scission, followed by results showing
2
-
(
4
(aq)
1
phosphonothioates, (3) the outcome of tracking studies with O-labeled 1, and (4)
8
3
1
N
computational results that suggest a S 2(P) pathway. P NMR spectroscopy (Figure 2)
shows formation of only one product with a singlet at 15 ppm consistent with
phosphonate formation. The absence of a phosphonothioate signal (~33 ppm) [18]
3
clearly shows only P – S scission, and this P NMR profile is the main spectral feature
1
throughout this report.
1
8 h
1
4
2 h
h
0
h
1
6

3
1
2-
Figure 2: P NMR spectrum of the hydrolysis of 1 (16 mM) by excess MoO
4
(1.0 M)
O), 30 C) that shows only one phosphonate product
15 ppm), resulting from selective P-S scission. A wider spectral expansion from -20 to
o
in MOPS (1.0 M, pH 7.5 (20% D
2
(
1
3
1
20 ppm shows no additional P NMR signals.
4
In addition, 2.0 M NaCl accomplishes no hydrolysis of 1, and LiClO was also inactive in
degrading 1 under similar pH and temperature conditions. Finally, under these conditions
o
pH 7.5 MOPS buffer, 30 C), 1 is stable for at least a month; this lack of background
(
2
-
auto-hydrolysis indicates MoO
4
is responsible for promoting 1 hydrolysis.
2
-
In addition to the specific P – S scission, a proof-of-concept result shows MoO
4
can function as a catalyst for hydrolyzing 1. Our initial question asks if the Mo(VI)
2
-
center in MoO
Accordingly, the C NMR signal of sodium ethanethiolate upon addition of Na
O (pH 7.5 MOPS buffer with 20% D O) remained unchanged, thus indicating no
interaction between the ethanethiol and MoO
hydrolyzed by Na MoO in H O even when excess phosphonothioate was used.
Specifically, 1 mM of Na MoO successfully hydrolyzes 15 mM of 1 in pH 7.5 at 30 C
with only P-S scission with the same P NMR as Figure 2. Prior results with the
4
would be poisoned by the ethanethiolate leaving group in 1 hydrolysis.
1
3
2
MoO
4
in
H
2
2
2
-
4
. In this connection, we found 1 is
2
4
2
o
2
4
3
1
organometallic molybdenum(IV) metallocene showed no such turnover [24,28]. This is
the first evidence of a molybdenum complex (of any oxidation state) that exhibits
catalytic hydrolysis of a phosphonothioate neurotoxin under mild aqueous conditions.
o
While the hydrolysis is impractically slow at 30 C (6 days for just 15 equivalents of 1 –
3
turnover frequency of 2/day), the P NMR indicates the desired P-S scission. This is
1
evident from the presence of just one product signal (15 ppm), which is consistent with
7

phosphonate production; no evidence of the toxic phosphonothioate product (33 ppm)
[
18] is observed.
1
When the C – enriched phosphonothioate O-ethyl S- methyl phenyl
3
13
2
-
phosphonothioate (16 mM) was hydrolyzed by excess MoO
4
(1.0 M), only free
1
methanethiol (confirmed through authentic addition) was formed, as shown with C
3
NMR (Figure 3). There is no indication of metal coordination, nor methylsulfate
-
formation (i.e. no signal at 38 ppm for CH
3
SO
3
). Only the presence of methyl disulfide
at 21.3 ppm is evident, which comes about from the air oxidation of the released
methanethiol. The evidence taken together, which includes the stoichiometry (1eq
2
13
31
MoO
does not coordinate to any of the hydrolytic products and therefore remains as MoO
aq) throughout the reaction. This allows it to hydrolyze 1 multiple times albeit slowly.
4
:15 eq 1) and the aformentioned C and P NMR results suggest the molybdate
2
-
4
(
1
3
3
CH SH
*
*
*
18 h
*
0
h
1
Figure 3: C NMR spectrum of the hydrolysis of O-ethyl S- methyl phenyl
3
13
1
3
2
phosphonothioate ( C-enriched on the methyl carbon at 11.8 ppm; JPC = 6.5 Hz) with
2
-
o
that shows production of only free methanethiol (5.7 ppm). Hydrolysis (30 C)
MoO
4
1
was done on 16 mM of the C-enriched phosphonothioate (2 µL) in 1 mL MOPS* buffer
3
2
-
(pH 7.5) and 1.0 M MoO . The presence of only methanethiol and the oxidized
4
dimethyl sulfide (**) indicates no molybdenum coordination to the leaving group.
8

2
-
Initial results of 1 hydrolysis by MoO
4
indicate that the reaction is first order in
2
-
MoO
4
(Supporting Information S1). There was no concentration dependency in [1]
2
-
when [MoO
4
] is fixed, but we suspect this may be due to the limited water solubility of
the phosphonothioate, because when [1] > 20 mM the solution was cloudy from droplets
of 1. Therefore, saturation kinetics with varying phosphonothioate concentration is
problematic for attaining any Michaelis-Menten values. Addition of alcohols or THF to
increase 1 solubility is also poses data interpretation issues, as these additives may affect
the speciation and solubility of the molybdates.
Operationally, the optimum pH (6.9 to 7.5) for these experimental conditions
favors the formation of the monomeric molybdate in terms of the quickest kinetics runs.
The speciation of polyoxomolybdates is known to be pH sensitive [44, 45] wherein the
6
-
4-
heptamer Mo
4.0, respectively. Indeed, acidification of MoO
and octamer polyoxomolybdates, in which the speciation was previously modeled
7 8
O24 and octamer Mo O26 exist primarily in the pH range of 5.5 – 6.5 and
2
-
<
4
results in formation of the heptamer
(
Supporting information S2) [32]. Specifically, equilibrium calculations show that the
2
-
monomeric MoO
4
(400 mM) is the main species at pH > 7.5. In this connection, we
2
-
found that the maximal rate of 1 hydrolysis by 400 mM MoO
4
occurred at pH 6.9.
Interestingly, Parac-Vogt and coworkers also found that the hydrolysis of dipeptides by
2
-
MoO
4
reached its maximal rate at pH 6.9 [31], which implicated the active species as
the monomeric molybdate. To that end, the pH value of 7.5 was chosen in our studies to
2
-
ensure that monomeric MoO
4
was the key species for hydrolyzing 1. Any work above
pH 8.5 would be problematic as the phosphonothioate undergoes spontaneous alkaline
hydrolysis. We do not discount the possibility that at this pH an equilibrium exists
9

2
-
between the major monomeric (MoO
4
) species and with any of the aforementioned
(
minor) polymeric molybdates that exist in the acidic range. It should be noted that
2
-
4
below pH 6 the hydrolysis of 1 by MoO (aq) is not detectable which suggests these
polymeric molybdates are poor candidates for phosphonothioate hydrolysis.
We set out to examine possible structure-activity relationships by steric
parameters of the phosphonothioate substrate. A similar investigation [24] was made
2 2
with the thiophilic molybdenum metallocene (Cp MoCl ). In this past study several
phenyl phosphonothioates were made wherein the alkylthiolate was varied. In this work
[
24] the hydrolysis rates of four different phenyl phosphonothioates, (PhP(O)(OEt)(SR);
i
, Et, Pr and Pr) by Cp
R = CH
3
2
MoCl
2
(aq) correlate well with the Taft parameter, (Es) of
the alkylthiolate (SR) leaving group. The Taft parameter, a quantitative steric indicator,
is derived from the hydrolysis rates of alkyl benzoates [46]. In addition, this correlation
may also depend on the acidity of the alkylthiolate leaving group (i.e. pKa of RSH) [47].
We saw that both factors affect the rates of phosphonothioate hydrolysis by the thiophilic
2 2
molybdenum metallocene Cp MoCl . There is a marked decrease in the rates of
phosphonothioate hydrolysis as the size of the leaving group increases and as the acidity
of RSH decreases.
In a similar study, we sought to determine if a similar relationship existed for the
molybdate– mediated hydrolysis of these same phenyl phosphonothioates (Figure 4).
1
0

-1.9
-2.1
-2.3
-2.5
-2.7
-2.9
11
10.8
10.6
10.4
10.2
10
Predicted pK of RSH
a
-1.9
-2.1
-2.3
-2.5
-2.7
-2.9
-0.55
-0.35
-0.15
0.05
Taft Parameter of R
Figure 4: The hydrolysis of various S-alkyl O-ethyl phenyl phosphonothioates (16 mM)
2
-
o
by 1.0 M MoO
4
(30 C, pH 7.5, 1.0 M MOPS) shows that there is a correlation between
was
taken from reference [48]. MEPP, DEPP (1), PEPP and iso-PEPP represent methyl,
the size of the alkylthiolate leaving group and the rate of hydrolysis. Predicted pK
a
ethyl, n-propyl and isopropyl thiolate groups, respectively.
Indeed, there is a strong dependence on the size and acidity of the alkylthiolate leaving
2
-
4
group for the hydrolysis of 1 by MoO (aq). Given the absence of any reaction between
1
1

2
-
4
MoO and ethanethiol, it is unlikely that there would be an interaction between the
Mo(VI) oxide and the sulfur linkage of the alkylthiolate leaving group. Therefore, the
trend in Figure 4 shows that the hydrolytic reaction is sensitive to the leaving group
ability (acidity) of the alkylthiolate. This is further supported with computational
calculations discussed at the end of this report.
Thermodynamic measurements (Table 1) of 1 hydrolysis were undertaken to determine
the molecularity of the rate determining step. The 60 degree temperature range yielded a
‡
clean Arrhenius plot (Supporting information S3) that gave a ∆S of -81.5 J/ mol·K (-19.5
±0.3 cal/mol·K). This large negative value indicates a bimolecular rate determining step,
‡
or one that goes through an ordered transition state. A ∆S of -316 J/ mol·K (-75.4
6
-
7
cal/mol·K) was measured for the Mo O24 – catalyzed hydrolysis of 2-hydroxypropyl-4-
nitrophenyl phosphate [32], a model for RNA. The more positive entropy of activation in
our study is consistent with a smaller molybdenum catalyst and with an organophosphate
substrate with fewer degrees of rotation; the path to the transition state requires less
‡
‡
entropy change. Interestingly, our ∆S and ∆H values are not dramatically different from
the molybdate-promoted hydrolysis of p-nitrophenyl acetate [49], even though that latter
‡
‡
reaction (∆H = 15.8 ±0.1 cal/mol, ∆S = -17.4 ±0.4 cal/mol·K) was (1) done under
different conditions, (2) with a different leaving group, and (3) on a carboxylic acid ester
substrate.
We wanted to track the fate of the terminal P=O oxygen to see if it is lost
2
-
4
in the hydrolysis of 1 by MoO (aq); this would help support a mechanistic hypothesis.
1
8
As such, diethyl phenyl phosphonite was hydrolyzed with 25% enriched O water
followed by tautomerization to the O-ethyl phenyl H-phosphinate (Scheme 4 and
Supporting information S4) [50]. Subsequent sulfur insertion [51] in the presence of
1
2

dicyclohexylamine yielded the ammonium salt of the O-ethyl phenyl
phosphonothioate. Finally, nucleophilic substitution with ethyl iodide [18] resulted in
1
the O label on the terminal P=O bond of 1. This was manifested in two P signals (3:1)
8
31
3
displaced by a 0.03 ppm isotopic shift [52, 53]. In this connection, P NMR was used to
1
1
track the fate of the O label during the hydrolysis of 1.
8
1
8
Scheme 4. Synthesis of O-labelled phosphonothioate 1
1
Hydrolysis of 1 in 25% O-labelled water (not O-labelled 1) with 1.0 M MoO
8
18
2-
4
yields the final phosphonate (14.6 ppm) with a 0.03 ppm displacement (3:1 ratio)
1
indicative of O incorporation into the ethyl phenyl phosphonate product (Supporting
8
1
8
information S5). This underscores the validity of using O-incorporation as a method to
1
track the fate of the terminal oxygen of 1. In this connection, O-labelled 1 was used
8
2
-
31
with MoO
4
(aq) and Figure 5 shows the ratio of the two P signals of the phosphonate
(
15 ppm) was identical to the starting phosphonothioate at 49.7 ppm. If there is
1
8
significant P-O scission of the P= O bond, then there would be a diminution in the
3
upfield P signal. The reverse O – labeling where the oxygen-18 is installed on the
1
18
2
-
molybdate is problematic given the rapid oxygen exchange of MoO
4
with the water
solvent [54].
1
3

1
Figure 5: Hydrolysis (pH 7.5 buffer) of O-labelled 1 (16 mM) by 1.0 M MoO
8
2-
4
shows
no oxygen exchange of the P=O functionality when hydrolyzed by excess molybdate.
1
The 0.03 ppm upfield displacement results from the P= O bond, and the 3:1 ratio of
8
1
integrals represent P= O to P= O, respectively. This ratio remains unchanged from the
6
18
starting phosphonothioate (1 at 49.9 ppm) to the phosphonate product (15 ppm).
This clearly indicates that no oxygen exchange took place for the P=O functionality in the
2
-
presence of the MoO
4
(aq).
In terms of a plausible mechanism for 1 hydrolysis by the molybdate the
+
-
amphoteric character of the Mo=O bond (i.e. [Mo=O ↔ Mo –O ]) may allow it to act as
a nucleophile in catalyzing the hydrolysis of 1. Indeed, DFT calculations on vanadate-
catalyzed hydrolysis of peptides proposed that hydrolysis occurs via the amphoteric V=O
bond [55, 56]. In this scheme the vanadate oxoanion attacks the electropositive carbon to
form a tetrahedral intermediate, which is stabilized through intramolecular hydrogen
bonding. Support for this model comes from prior synthetic reports that used MoO
2 2
Cl
as a catalyst for nucleophilic acyl substitutions in organic solvents [57, 58]. In this work,
+
-
the [Mo=O ↔ Mo –O ] functionality acts as an oxoanion nucleophile that attacks the
electrophilic carbon of an anhydride (Scheme 5).
1
4

+
-
Scheme 5. Mechanism for anhydride transesterification in which the [Mo=O ↔ Mo –O ]
functionality acts as an oxoanion nucleophile. Taken from references 57 and 58
Experimental results confirmed the existence of a molybdate-carboxylate intermediate
t
C Bu)
(
MoOCl
2
(O
2
2
) resulting from the molybdenum oxoanion attacking the electrophilic
carbon of pivalic anhydride. This short-lived molybdocarboxylate anhydride undergoes
2
-
subsequent esterification with an alcohol. Such a route may exist for MoO
hydrolysis of 1. In this scenario (Scheme 6), the oxoanion of the molybdate first attacks
2(P)) the phosphonothioate to form a molybdate-phosphonate anhydride (2). This is
followed by rapid hydrolysis to regenerate the molybdate and phosphonate product.
4
– catalyzed
(S
N
N
Scheme 6. Mechanism for phosphonothioate 1 hydrolysis in which an initial S 2(P) is
followed by rapid hydrolysis of the molybdate-phosphonate anhydride.
3
Molybdate-phosphate anhydrides yield distinctive P NMR signals that have
1
provided diffusion coefficients for molecular weight determinations through DOSY
6
-
[
32,59]. In this prior investigation by Parac-Vogt and coworkers, the heptamer Mo
7 24
O
was found to hydrolyze anionic 4-nitrophenylphosphate (NPP). This hydrolysis
4
-
4-
proceeded via (NPP)
2
Mo
5 2 2
O21 and (NPP) Mo12O36(H O) complexes that were seen as
3
intermediate P signals which were characterized through DOSY. In our studies, no
1
1
5

3
intermediate P signal has been observed in the molybdate-mediated hydrolysis of the
1
neutrally charged phosphonothioate. This nucleophilic molybdate mechanism (Scheme
6
) cannot be ruled out, as it is possible the intermediate phosphonomolybdate (2) is
hydrolytically unstable. Furthermore, there is a distinction between this work and the
aforementioned prior study. In the prior studies the interaction was between a
polyoxomolybdate and an anionically charged phosphate, but this current work is
between a monomeric molybdate and a neutrally charged phosphonothioate. Indeed,
Byers and coworkers [49] found that the molybdate anion is an effective nucleophile for
‡
the hydrolysis of activated carboxylic esters [49]. In addition, our ∆S is close in value to
that reported by Byers for the molybdate-promoted nucleophilic attack on carboxylic
esters. In summary, the S
N
2(P) scenario set in Schemes 6 is consistent with the reaction
order, the absence of oxygen exchange on 1 and thermodynamic experimental results.
2
-
4
The first order dependence in [MoO ] indicates the importance of the oxoanion pathway
that uses the nucleophilic Mo=O bond to attack a neutral substrate. Moreover, the large
negative entropy of activation of -81.5 J/ mol·K (-19.5 cal/mol·K) is also consistent with
2
-
nucleophilic (i.e. bimolecular) attack of MoO
4
on phosphonothioate 1
The hypothesized pathway in Scheme 6 was modeled with density functional
theory methods (water as the solvent) in a two-step process as shown in Figure 6. This
N
commences with the initial S 2(P) attack by the molybdate dianion on 1 to yield the
phosphonomolybdate anhydride intermediate 2. As such, DFT calculations reveal a
transition state that had a reasonable vibration (i.e. one imaginary frequency) connecting
2
-
-
reactants MoO
4
+ 1 with 2 + EtS (first step of Scheme 6). Subsequent hydrolysis of
1
6

anhydride intermediate 2 was best effected with hydroxide attack that yielded
intermediate 3.
N
Figure 6: Reaction coordinate diagram for molybdate S 2(P) attack on 1 that delineates
scheme 6 as determined through DFT calculations modeled with implicit water solvent.
Energies are sum of electronic and thermal enthalpies for the compounds at each step
based on an overall balanced reaction. The lowest combined energy were the final
2
- -
4
products (MoO , SEt and PhP(O)(HO)(OEt) that was set to zero, all other energies are
relative to this. Supporting information has full calculation results from a B3LYP
optimization using a 6-31+G(d)/LanL2DZ split basis set.
Although compound 3 resembles the transition state structure of hydroxide attack,
frequency calculations suggest 3 as an intermediate structure. Various attempts to model
3
as a transition state did not yield an imaginary frequency consistent with hydroxide
nucleophilic attack on 2 in a S 2(P). Moreover, water attack on 2 yielded only
nonsensical intermediate structures for the breakdown of 2.
N
1
7

N
Experimentally, we see the S 2(P) yields exclusively P – S scission, and this is
further supported through DFT calculations that indicate a relatively stronger P – O bond
in phosphonothioate 1. Energy calculations on 1 as a function of P – SEt and P – OEt
bond lengths (Figure 7) show a deeper energy well for the P – OEt linkage. This suggests
the P – O bond is stronger (by 55 kJ/mol) than the P – S bond that results in exclusive P –
S scission upon hydrolysis.
P-S or P-O Bond Distance (Å)
-1279.95
-1279.99
-1280.03
-1280.07
-1280.11
-1280.15
-1280.19
1.3
2.3
3.3
4.3
5.3
4
09 kJ/mole
P-O bond
P-S bond
354 kJ/mol
Figure 7: Molecular potential energy curve for P – SEt and P – OEt bond stretches on
phosphonothioate 1 that shows the P – O bond is stronger than the P – S bond by 55.1
kJ/mol. This is consistent with experimental results that show exclusive P –S scission of
2
-
1
4
upon hydrolysis promoted by MoO .
While the S
phosphonothioate degradation by MoO
discounted. One such route treats the Mo(VI) of MoO
N
2(P) pathway for 1 hydrolysis appears to be the likely route for this
2
-
4
, other pathways cannot be completely
2
-
4
as a Lewis acid that directly
bonds to the oxygen of the P=O functionality of 1. This alternate route is elaborated in
supporting information S6, but DFT calculations were unsuccessful in optimizing any of
1
8

the proposed structures. Any minimized structure had the molybdate completely
displaced (i.e. unbonded) from 1.
Conclusion
We report the first example of catalytic phosphonothioate hydrolysis by the
molybdate dianion. The title phosphonothioate (1) is an analog of neurotoxins such as
the chemical warfare agent VX, and its hydrolysis proceeds with selective (and desired) P
o
S scission under mild (pH 7.5 and 30 C) aqueous conditions with a readily available
–
and safe metal oxide. The high oxidation Mo(VI) precludes metal coordination to the
alkylthiolate leaving group, thus preserving the molybdate species and providing turnover
1
8
in hydrolysis. O labeling studies on 1 indicate no exchange with the P=O functionality
on the phosphonothioate substrate. In addition, there is a strong rate dependency on the
acidity of the phosphonothioate leaving group, and the activation parameters indicate a
2
-
bimolecular process. A possible mechanistic route entails nucleophilic attack by MoO
4
on the phosphonothioate through a S 2(P) process; Lewis acid activation by Mo(VI) on
N
the P=O functionality of 1 for water hydrolysis cannot be ruled out. Future studies will
create other molybdenum(VI) systems that yield faster hydrolytic rates.
EXPERIMENTAL
CAUTION: The phosphonothioates are known neurotoxins and all experimental
work was done with gloves in well-ventilated hoods. In addition, all glassware
exposed to the phosphonothioates was bleached overnight prior to base bath
cleaning to oxidize any of the phosphonothioates. It is highly recommended that
1
9

these phosphonothioates be synthesized and used on a microscale to minimize
o
release and exposure. All phosphonothioates were stored below – 5.0 C and
allowed to warm to RT prior to kinetic runs
Equipment and supplies
1
All NMR spectra ( C, P, and H) were acquired using a Bruker Advance-300
3
31
1
MHz NMR spectrometer at 75.468 MHz, 121.495 MHz, and 300.130 MHz respectively,
2
and pH measurements were done with a Vernier pH probe. The pH readings in 20% D O
were uncorrected. The dichlorophenylphosphine and precursors to the phosphonothioates
were purchased from TCI America (Portland, OR), and the sodium molybdate was
1
8
2
obtained from Mallinckrodt Chemical Works. The O labeled H O and diethylphenyl
phosphonate were purchased from Aldrich (Milwaukee, WI) and used as received. The
MOPS (3-(N-morpholino)propanesulfonic acid) and MOPS salt were purchased from
Sigma. All phosphonothioates in Figure 4 were made according to prior literature
protocols [24].
1
Synthesis of O-labelled phosphonothioate (1)
8
The synthesis started with the anaerobic hydrolysis of diethylphenyl phosphonite
1
8
(
2
2.0 mL, 9.8 mmole) in 1 mL of 25% O labeled H O (56 mmole); the water was purged
with Ar for ~3 min to minimize oxidation of the phosphonite. This hydrolysis is known
to yield the O-ethyl phenyl phosphinate via the tautomerization shown in scheme 4 [51].
After 10 h of stirring at room temperature, the water was evaporated off under vacuum
1 31
leaving behind 1.20 g (7.0 mmole, 71% yield) of an oily residue with H and P NMR
2
0

spectral features consistent with literature reports of O-ethyl phenyl phosphinate [18,43].
1
H NMR (CDCl
3
): δ 1.4 (t, 3H, O-CH
2
-CH
3
), 4.2 (dq, 2H, O-CH2-CH
3
), 7.52 (m, 2H,
1
31
meta), 7.6(t, 1H, para), 7.75 (dd, 2H, ortho), 7.6 (s, 1H, P-H, JPH = 550 Hz). P NMR
1
8
(
3
CDCl ): δ 24.6 (Supporting information S6). Completion of the synthesis of O –
labeled 1 followed the protocol set by DeBruin [18] on a microscale [24]. This began
with sulfur addition to the hydrogen phosphinate (0.80 g, 4.6 mmole) in 10 mL
dicyclohexylamine. A white precipitate was isolated and dried which was then subjected
to nucleophilic displacement with excess iodoethane (1.0 mL, 12.4 mmole) in 5 mL
hexane (18 h). Evaporation of the hexane in vacuo left behind only 180 mg (0.78 mmole)
1
8
1
of the O – labeled phosphonothioate 1: 17% overall yield. H NMR (CDCl
H, O-CH -CH ), 1.4 (t, 3H, S-CH -CH ), 2.8 (q, 2H, S-CH2-CH ), 4.2 (q, 2H, O-CH
CH ), 7.5 (m, 3H, meta and para), 7.9 (d, 2H, ortho). P NMR (CDCl ): δ 44.7.
3
): δ 1.3 (t,
3
2
3
2
3
3
2
-
3
1
3
3
Kinetics of 1 hydrolysis
All kinetic runs, with the exception of the pH dependency (vide infra), were done
in a total volume of 1.0 mL of 1 M MOPS buffer (pH 7.5; 20% D O for deuterium lock)
2
within an NMR tube. Sodium molybdate (0.242 g, 1.00 mmol, 1.00 M) was added and
allowed time to completely dissolve wherein the pH rose to 7.5. The phosphonothioate
was then added to give a molybdate to phosphonothioate ratio of 62:1 (in the case of 1
this was 2.0 µL of 1 (0.016 mmole)). This high ratio guaranteed pseudo-first order
o
31
kinetics that could be completed in less than 18 h at 30 C. P spectra were taken with
2 scans and a delay of 2 seconds between each scan, at 20-minute intervals. Integration
of the peaks of phosphonothioate starting materials and phosphonate products allowed for
3
2
1

tracking of their relative concentrations. The fraction of remaining phosphonothioate
versus time was plotted and fitted to an exponential decay (first order kinetics) in Excel
to 95% completion (Supporting information S7-Figure 4 and S8-Arrhenius plot). It was
found that reducing the delay to 0.1 sec made no difference in the rates of hydrolysis.
Computational Details
All molecules in Figure 6 were built from GaussView [44], and full optimization
on all structures were carried using the DFT model on Gaussian 09 [45] with the B3LYP
functional [46, 47, 48]. The chosen basis sets were 6-31+G(d) [49, 50, 51] basis set for
H, C, S, O atoms and LanL2DZ basis set [52,53] for the Mo atom. A polarizable
continuum water model was used with the CPCM key word [54, 55]. All structure
geometries were first optimized prior to frequency calculations to check for negative
frequencies. The sum of electronic and thermal enthalpies was also taken from this
(
frequency) calculation for use in generating the reaction coordinate diagrams. The
2
-
transition-state structure for the 1 + MoO
4
ꢀ 2 process was generated by optimizing the
structure of the transition state using the Berney algorithm. The frequency calculation
revealed exactly one negative frequency; the vibration connected the two sides of the
reaction. Coordinates for the calculation of all starting, intermediate, transition state and
final compounds are in supporting information 9.
ASSOCIATED CONTENT
Details of the analyses are provided, and this information is available free of charge via
the internet.
2
2

AUTHOR INFORMATION
Corresponding Author
*
Email: kuo@lclark.edu
ACKNOWLEDGEMENT
This work was funded by NSF-RUI award CHE-1413090.
2
3

1
2
3
4
N.N. Greenwood, A. Earnshaw, in: Chemistry of the Elements, Pergamon Press,
Oxford, UK, 1984, pp. 1174-1184.
F.A. Cotton, G. Wilkinson, in: Advanced Inorganic Chemistry, fourth ed., John Wiley
Sons, New York, 1980, pp. 472-482.
&
J.M. Battershill, P.M. Edwards, M.K. Johnson, Food Chem. Toxicol. 42 (2004) 1279-
285.
1
R.A. Larson, E.J. Weber, in: Reaction Mechanisms in Environmental Organic
Chemistry, Lewis Press, Boca Raton, LA, 1994, pp. 122-125.
5
6
7
8
9
1
1
J.J. Gordon, L. Leadbeater, Toxicol. Appl. Pharmacol. 40 (1977) 109-114.
K. Kim, O.G. Tsay, D.A. Atwood, D.G. Churchill, Chem. Rev. 111 (2011) 5345-5403.
Y. Yang, J.A. Baker, J.R. Ward, Chem. Rev. 92 (1992) 1729-1743.
N.B. Munro, K.R. Ambrose, A.P. Watson, Environ. Health Perspect. 102 (1994) 18-38.
M.R. Seger, G.E. Maciel, Environ. Sci. Technol. 40 (2006) 552-558.
0
J. Bielawski, J.E. Casida, J. Agric. Food. Chem. 36 (1988) 610-615.
1
Parathion toxicity comes from the P=S ꢀ P=O oxidation that takes place in insects and
converts the applied parathion (LD50 = 13 mg/kg) to the more toxic paraoxon (LD50
.8 mg/kg) [12].
=
1
T.G. Spiro, W.M. Stigliani, in: Chemistry of the Environment, Prentice Hall, Upper
Saddle River, New Jersey, 1996, pp. 281-282.
1
1
2
3
F. Sperling, H.K. Ewenike, T. Farber, Environ. Res. 5 (1972) 164-171.
S.A. Melnychuk, A.A. Neverov, R.S. Brown, Angew. Chem. Int. Ed. 45 (2006) 1767-
770.
1
1
4
T. Liu, A.A. Neverov, J.S.W. Tsang, R.S. Brown, Org. Biomol. Chem. 3 (2005) 1525-
533.
1
1
1
1
5
6
7
J.S.W. Tsang, A.A. Neverov, R.S. Brown, Org. Biomol. Chem. 2 (2004) 3457-3463.
A.A. Neverov, R.S. Brown, Main Group Chem. 9 (2010) 265-281.
D. Marciano, M. Goldvaser, I. Columbus, Y. Zafrani, J. Org. Chem. 76 (2011) 8549-
2
4

8
553.
1
1
8
9
K.E. DeBruin, D.M. Johnson, J. Am. Chem. Soc. 95 (1973) 7921-7922.
L.Y. Kuo, T.T. Adint, A.E. Akagi, L. Zakharov, Organometallics 27 (2008) 2560-
2
564.
2
2
2
2
2
0
1
2
3
4
Y.C. Yang, Acc. Chem. Res. 32 (1999) 109-115.
H. Morales-Rojas, R.A. Moss, Chem. Rev. 102 (2002) 2497-2521.
R.A. Moss, H. Morales-Rojas, H. Zhang, B.D. Park, Langmuir 15 (1999) 2738- 2744.
R. Cooper, M.L.H. Green, J. Chem. Soc. A (1967) 1155-1160.
L.Y. Kuo, D.C. Baker, A.K. Dortignacq, K.M. Dill, Organometallics 32 (2013) 4759-
4
765.
2
2
2
2
5
6
7
8
M.J. Calhorda, A.R. Dias, J.A. Martino Simoes, C. Teixeira, J. Chem. Soc., Dalton
Trans. 12 (1984) 2659-2663.
M.A.A.F. de C.T. Carrondo, P.M. Matias, G.A. Jeffrey, Acta Crystallogr., Sect. C:
Cryst. Struct. Commun. 40 (1984) 932-934.
A.R. Dias, M.H. Garcia, M.F.M. da Piedade, M.A.A.F. de C. T. Carrondo, J.
Organomet. Chem. 632 (2001) 107-112.
L.Y. Kuo, A.K. Bentley, Y.A. Shari'ati, C.P. Smith, Organometallics 31 (2012) 5294-
5
301.
2
3
9
0
L.Y. Kuo, A.P. Blum, M. Sabat, Inorg. Chem. 44 (2005) 5537-5541.
L.Y. Kuo, Y.A. Shari’ati, Degradation of Phosphate Esters by High Oxidation State
Molybdenum Complexes, U.S. Patent 9024106 B2, May 5, 2015.
3
3
1
2
G. Absillis, R.V. Deun, T.N. Parac-Vogt, Inorg. Chem. 50 (2011) 11552-11560.
G. Absillis, E. Cartuyvels, R.V. Deun, T.N. Parac-Vogt, J. Am. Chem. Soc. 130 (2008)
1
7400-17408.
3
3
3
4
B.-T. An, H.-S. Park, E.-J. Lee, I.-H. Eom, Bull. Korean Chem. Soc. 21 (2000) 905-
08.
9
P.H. Hien, K. Stroobants, T.N. Parac-Vogt, Inorg. Chem. 50 (2011) 12025-12033.
2
5

3
3
3
3
3
4
5
6
7
8
9
0
H. Weil-Malherbe, R.H. Green, Biochem. J. 49 (1951) 286-292.
H. Weil-Malherbe, Biochem. J. 55 (1953) 741-745.
E.P. Steyn-Parvé, Biochim. Biophys. Acta 64 (1962) 13-20.
D. Spencer, Aust. J. Biol. Sci. 7 (1954) 151-160.
D.L. Harris, B.J. Mizock, S.J. Pilkis, J. Biol. Chem. 241 (1966) 707-714.
A. dos P. Lemos, A.L.F. de Sousa, A.A.S. Pinheiro, M. Berrêdo, J.R. Meyer-
Fernandes, Z. Naturforsch., C: Biosci. 57 (2002) 500-505.
4
4
4
1
2
4
L.M. Franks, V. Maggi, A.W. Carbonell, J. Anat., 101 (1967) 777-787.
R. van Reen, Arch. Biochem. Biophys. 53 (1954) 77-84.
J.J. Cruywagen, G. Draaijer, J.B.B. Heyns, E.A. Rohwer, Inorg. Chim. Acta. 331
(
2002) 322-329.
4
4
4
4
5
6
7
8
M.T. Pope, in: Heteropoly and Isopoly Oxometalates, Springer Verlag, Berlin, 1983,
pp. 60-117.
R.W. Taft, in: M.S. Newman (Ed.), Steric Effects in Organic Chemistry, John Wiley &
Sons, New York, 1956, Chapter 13.
T.H. Lowry, K.S. Richardson, in: Mechanism and Theory in Organic Chemistry, third
ed., Harper Collins, New York, 1987, pp. 374-375.
Scifinder Substance Identifier.
https://scifinder.cas.org/scifinder/view/scifinder/scifinderExplore.jsf (accessed June
2
0, 2016).
4
5
9
0
B.R. Wikjord, L.D. Byers, J. Org. Chem. 57 (1992) 6814-6817.
B. Ben-Nissan, D.D. Green, G.S.K. Kannangara, C.S. Chai, A.J. Milev, J. Sol-Gel Sci.
Technol. 21 (2001) 27-37.
5
1
W.-M, Wang, J.-J. Lui, L. Yao, F.-J. Meng, Y.-M. Sun, C.-Q.Zhao, Q. Xu, L.-B. Han,
J. Org. Chem. (2016) 6843-6847
5
5
2
3
M. Cohn, A. Hu, Proc. Natl. Acad. Sci. USA 75 (1978) 200-203.
G. Lowe, B.S.J. Sproat, Chem. Commun. (1978) 565-566.
2
6

5
5
5
4
5
6
H.V. Felten, B. Wernli, H.J. Gamsjäger, Chem. Soc., Dalton Trans. (1988), 496-500.
T. Mihaylov, T.N. Parac-Vogt, K. Pierloot, Inorg. Chem. 51 (2012) 9619-9628.
P.H. Ho, T. Mihaylov, K. Pierloot, T.N. Parac-Vogt, Inorg. Chem. 51 (2012) 8848-
8
859.
5
7
C.-T. Chen, J.-H. Kuo, V.D. Pawar, Y.S. Munot, S.-S. Weng, C.H. Ku, C.-Y. Liu, J.
Org. Chem. 70 (2005) 1188-1197.
5
5
4
8
9
3
R. Sanz, M.R. Pedrosa, Curr. Org. Synth. 6 (2009) 239-263.
C.S. Johnson, Prog. Nucl. Magn. Reson. Spectrosc. 34 (1999) 203-256.
L.Y. Kuo, S.K. Glazier, Inorg. Chem. 51 (2012) 328-335.
4
4
5
GaussView, Version 5, Roy Dennington, Todd A. Keith, and John M. Millam,
Semichem Inc., Shawnee Mission, KS, 2016.
4
Gaussian 09, Revision D.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A.
Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J.
Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J.
Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A.
Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N.
Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A.
Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M.
Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L.
Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg,
S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, and
D. J. Fox, Gaussian, Inc., Wallingford CT, 2013.
46
47
48
A. Becke, J. Chem. Phys., vol. 98, pp. 5648–5652, 1993.
C. Lee, W. Yang, and R. G. Parr, Phys. Rev. B, vol. 37, no.2, pp. 785–789, 1988.
B. Miehlich, A. Savin, H. Stoll, and H. Pruess, Chem. Phys. Lett., vol. 157, no. 3, pp.
2
00–206, 1989.
4
9
W. J. Hehre, R. Ditchfield, L. Radom, and J. A. Pople, J. Am. Chem. Soc., vol. 92, no.
6, pp. 4796–4801, 1970.
1
50
R. Ditchfield, W. J. Hehre, and J. A. Pople, J. Chem. Phys., vol. 54, p. 724, 1971.
2
7

5
1
P. C. Hariharan and J. A. Pople, Theor. Chem. Acc., vol. 28, pp. 213–222, 1973.
P. J. Hay and W. R. Wadt, J. Chem. Phys, vol. 82, pp. 270–83, 1985.
P. J. Hay and W. R. Wadt, J. Chem. Phys., vol. 82, pp. 299–310, 1985.
V. Barone and M. Cossi, J. Phys. Chem. A, 102 (1998) 1995-2001.
5
5
5
5
2
3
4
5
M. Cossi, N. Rega, G. Scalmani, and V. Barone, J. Comp. Chem., 24 (2003) 669-81.
2
8

Highlights:
•
•
•
A convenient molybdate that catalytically hydrolyzes phosphonothioate
neurotoxins under mild aerobic conditions
Phosphonothioate hydrolysis is accomplished with desired P-S bond scission
to yield non-toxic phosphonate product
A proposed mechanism uses the molybdate as an oxoanion-nucleophile to
carry out an S 2(P) on the phosphonothioate.
N
2
9