Variants of Phosphotriesterase for the Enhanced Detoxification of the Chemical Warfare Agent VR

Article abstract of DOI:10.1021/acs.biochem.5b00629

The V-type organophosphorus nerve agents are among the most hazardous compounds known. Previous efforts to evolve the bacterial enzyme phosphotriesterase (PTE) for the hydrolytic decontamination of VX resulted in the identification of the variant L7ep-3a,

Full text of DOI:10.1021/acs.biochem.5b00629

Subscriber access provided by UNIV OF CAMBRIDGE
Article
Variants of Phosphotriesterase for the Enhanced
Detoxification of the Chemical Warfare Agent VR
Andrew Bigley, Mark F Mabanglo, Steven P. Harvey, and Frank M. Raushel
Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.5b00629 • Publication Date (Web): 14 Aug 2015
Downloaded from http://pubs.acs.org on August 18, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Biochemistry is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 42
Biochemistry
1
2
3
4
5
6
7
Variants of Phosphotriesterase for the Enhanced
Detoxification of the Chemical Warfare Agent VR
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Andrew N. BigleyΨ, Mark F. MabangloΨ, Steven P. HarveyΦ,
and
Frank M. RaushelΨ*
ΨDepartment of Chemistry, Texas A&M University, College Station, Texas 77843
ΦU.S. Army Edgewood Chemical Biological Center, 5183 Blackhawk Road, Aberdeen
Proving Ground, Maryland 21010.
*To whom correspondence may be addressed:
(FMR) telephone: 979-845-3373; fax: 979-845-9452; e-mail: raushel@tamu.edu.
This work was supported by the Defense Threat Reduction Agency (HDTRA1-14-1-0004).
1
ACS Paragon Plus Environment

Biochemistry
Page 2 of 42
1
2
3
4
5
ABBREVIATIONS
6
7
8
9
PTE: phosphotriesterase
DEVX: O,O-diethyl VX
DMVX: O,O-dimethyl VX
DEVR: O,O-diethyl VR
OMVR: O-methyl VR
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
APVR: p-acetophenyl VR
VX: S-(2-(diisopropylamino)ethyl) O-ethyl methylphosphonothioate
VR: S-(2-(diethylamino)ethyl) O-isobutyl methylphosphonothioate
PON1: paraoxonase I from humans
DFPase: diisopropylfluorophosphatase from Loligo vulgaris
OPAA: organophosphorus acid anhydrolase from Alteromonas sp.
DTNB: 5,5'-dithiobis(2-nitrobenzoic acid)
2
ACS Paragon Plus Environment

Page 3 of 42
Biochemistry
1
2
3
4
5
ABSTRACT
6
7
The V-type organophosphorus nerve agents are among the most hazardous
8
9
compounds known. Previous efforts to evolve the bacterial enzyme phosphotriesterase
(PTE) for the hydrolytic decontamination of VX resulted in the identification of the variant
L7ep-3a, which has a kcat value more than two orders of magnitude higher than wild-type
PTE for the hydrolysis of VX. Because of the relatively small size of the O-ethyl,
methylphosphonate center in VX, stereoselectivity is not a major concern. However, the
Russian V-agent, VR, contains a larger O-isobutyl, methylphosphonate center making
stereoselectivity a significant issue since the SP-enantiomer is expected to be significantly
more toxic than the RP-enantiomer. The three-dimensional structure of the L7ep-3a
variant was determined to a resolution of 2.01 Å (PDB id: 4ZST). The active site of the
L7ep-3a mutant has revealed a network of hydrogen bonding interactions between Asp-
301, Tyr-257, Gln-254 and the hydroxide that bridges the two metal ions. A series of new
analogs that mimic VX and VR has helped to identify critical structural features for the
development of new enzyme variants that are further enhanced for the catalytic
detoxification of VR and VX. The best of these mutants has been shown to have a reversed
stereochemical preference for the hydrolysis of VR-chiral center analogs. This mutant
hydrolyzes the two enantiomers of VR 160- and 600-fold faster than wild-type PTE
hydrolyzes the SP-enantiomer of VR.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
ACS Paragon Plus Environment

Biochemistry
Page 4 of 42
1
2
3
4
5
6
7
8
9
The organophosphorus nerve agents are among the most toxic compounds known.
Compounds such as sarin, soman and VX are all chiral methyl phosphonates where the
toxicity of the SP-enantiomer is much greater than the RP-enantiomer.1 Recent events have
dramatically demonstrated the continuing importance of developing rapid and
environmentally compatible methods for the decontamination of these compounds.2 This
situation is particularly true for the V-type nerve agents, where the lethal dose is
approximately 6 mg/person, and these compounds have been shown to persist for long
periods of time.3, 4 The current means of decontamination of chemical warfare agents relies
on treatment with strong base, concentrated bleach, or high temperature incineration.
Medical intervention for nerve agent intoxication relies on compounds such as
atropine, which reduces neurological symptoms, and oximes, which can help to reactivate
the effected neural enzyme, acetylcholine esterase.3 The only currently approved medical
treatment for organophosphate poisoning, which acts directly on the nerve agent, is
injection of the enzyme butyrylcholine esterase, which acts as a stoichiometric scavenger of
organophosphonates.5 The high specificity of enzyme catalyzed reactions also enables
them to be exploited for analytical detection, as has been achieved using acetylcholine
esterase in Disclosure Spray produced by FLIR Systems, which is used by the U.S. military.6
Enzymatic means of environmental decontamination provide significant advantages over
the harsher chemical or incendiary methods.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Several enzymes are known to hydrolyze organophosphate nerve agents, including
human PON1, squid DFPase and the bacterial enzymes OPAA and phosphotriesterase
(PTE).7-12 Among these enzymes only PTE and PON1 are known to hydrolyze the V-type
nerve agents. Wild-type PTE was used as the main ingredient in DEFENZ, which was
4
ACS Paragon Plus Environment

Page 5 of 42
Biochemistry
1
2
3
4
5
6
7
8
9
marketed by Genencor for the decontamination of organophosphate nerve agents.
Unfortunately, testing by the EPA found that while the enzyme formulation did hydrolyze a
substantial portion of the nerve agents tested, the decontamination was not complete in
the prescribed contact time.13 PTE was successfully evolved for the decontamination of
organophosphate insecticides and marketed as Landgauard A900 by CSIRO Ecosystem
Sciences. While not currently licensed for sale in the U.S., this product utilizes variants of
PTE that have been evolved specifically to target organophosphorus insecticides.14 At
protein concentrations of 1 g/100 L of waste water, 99% of the contaminants were
neutralized within 12 hours and the enzyme maintained >70% activity at room
temperature for 6 months. PTE has also been formulated for application in firefighting
foams, and shown to be effective for decontamination of paraoxon on surfaces and in soil.15
Significant advances have been made in developing specific PTE variants for the
decontamination of G-type and VX nerve agents.7, 12 While wild-type PTE has reasonable
activity against the G-type nerve agents (kcat/Km ~105 M-1 s-1), this enzyme preferentially
hydrolyzes the less toxic RP-enantiomers.16 Directed evolution of PTE to specifically target
the G-type nerve agents has led to the identification of the variant H257Y/L303T (YT),
which has proven highly efficient at the hydrolysis of the more toxic SP-enantiomers of
sarin (GB), soman (GD), and cyclosarin (GF) with values of kcat/Km that exceed 106 M-1 s-1.12
Wild-type PTE exhibits little stereoselectivity against the relatively small
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
phosphonate center of VX, but the elevated pKa of the thiol-leaving group provides a
significant challenge for enzyme-catalyzed hydrolysis (kcat/Km ~ 102 M-1 s-1).7 Mutation of
residues contained within the active site of PTE resulted in the isolation of the variant
H245Q/H257F (QF), which exhibited a 100-fold improvement for the hydrolysis of VX,
5
ACS Paragon Plus Environment

Biochemistry
Page 6 of 42
1
2
3
4
5
6
7
8
9
relative to the wild-type enzyme (see Table 1 for identity of variants).7 Additional active
site variations led to the identification of the mutant CVQFL (QF + I106C/F132V/S308L)
with a similar catalytic efficiency for the hydrolysis of VX, but a three-fold improvement in
kcat. The best variant identified to date for the hydrolysis of VX is VRN-VQFL (QF +
F132V/S308L + A80V/K185R/I274N). This mutant combines expression-enhancing
mutations (A80V/K185R/I274N) with additional changes in the active site to achieve a
kcat/Km of 7 x 104 M-1 s-1 for the hydrolysis of the SP-enantiomer of VX.7, 17, 18 Expansion of
the mutation strategy to targeted error-prone PCR, led to the identification of the variant
L7ep3a (CVQFL + H257Y/A270V/L272M/I274N), which has a kcat value enhanced 150-fold
relative to wild-type PTE.7 How these combined mutations, some of which do not fall in the
active site, are able to bring about such a dramatic improvement in catalytic ability is not
clear.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In addition to VX, the V-agents include the Russian (VR) and Chinese versions.
Exemplified by VR, these additional V-agents contain a smaller thiol leaving group, and a
larger ester group attached to the phosphorus center (Scheme 1). Wild-type PTE has
enzymatic activity for the hydrolysis of racemic VR similar to VX, but the larger isobutyl
group attached to the phosphorus center results in a 25-fold preference for the RP-
enantiomer.16, 19 The catalytic activity of wild-type PTE for the hydrolysis of the SP-
enantiomer of VR (kcat/Km = 4.3 M-1 s-1) is significantly lower than for the hydrolysis of VX.
PTE variants, which contain many of the same mutations as the VX-enhanced variants, have
been reported to have substantially improved catalytic activity against SP-VR.20
Unfortunately, the in vivo toxicity of the individual enantiomers of VR has apparently not
been reported. However, the SP-enantiomer of VR has been shown to inactivate human
6
ACS Paragon Plus Environment

Page 7 of 42
Biochemistry
1
2
3
4
5
6
7
8
9
acetylcholine esterase 2.2 x 104 times faster than does the RP-enantiomer.21 We have,
therefore, made the explicit assumption that the SP-enantiomer of VR is significantly more
toxic than is the RP-enantiomer, in accordance with the relative toxicities of the individual
enantiomers of sarin, soman, and VX.1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Currently, there is a lack of three-dimensional structural data that can be used to
explain how the existing set of mutants are able to enhance the rate of hydrolysis of the
phosphorothiolate bond in VX. In an effort to more completely understand the chemical
mechanism for the enhancement of phosphorothiolate bond cleavage, the structure of the
L7ep3a variant was determined. With this new structural information, a series of PTE
variants was created to incorporate changes in the active site of PTE that would more
easily accommodate the O-isobutyl group of VR. To facilitate the further development of
PTE for the hydrolysis of various V-agents, we have designed and synthesized a new series
of analogs. Enhanced variants of PTE with improved activity toward the hydrolysis of the
SP-enantiomer of VR have been identified.
M
ATERIALS and
METHODS
Materials. Growth media and antibiotics were procured from Research Products
Incorporated. DNA polymerase was obtained from Agilent. Other supplies for the
molecular biology experiments were acquired from New England Biolabs. DEVX and N,N-
diisopropylaminoethanethiol were synthesized as previously reported.7, 22 The individual
enantiomers of p-acetophenyl VR (APVR) were synthesized as previously described.16
Samples of VX and VR were Chemical Agent Standard Analytical Reference Material of the
7
ACS Paragon Plus Environment

Biochemistry
Page 8 of 42
1
2
3
4
5
6
7
8
9
highest purity available, and were further purified as described previously.7 Unless
otherwise noted, all other chemicals were purchased from Sigma Aldrich. The
organophosphorus nerve agents used in this investigation are highly toxic and should be
used with the proper safety precautions.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis of Dimethyl VX. Dimethyl VX (DMVX) was made by the reaction of
dimethyl chlorophosphate with N,N-diisopropylaminoethanethiol. N,N-
diisopropylaminoethanethiol (1.5 grams; 9.3 mmol) was added to 100 mL of diethyl ether
and allowed to cool in a dry ice/acetone bath before being purged with N2. To this mixture
was added 7.5 mL (2.0 equivalents) of a 2.5 M solution of n-butyl lithium in hexanes and
the reaction allowed to come to room temperature before re-cooling in a dry ice/acetone
bath. Dimethyl chlorophosphate (2.0 g; 1.5 equivalents) was mixed with 30 mL of diethyl
ether in a separate flask and cooled. The dimethyl chlorophosphate solution was then
added to the thiol solution and the reaction stirred at room temperature for 3 hours. The
reaction was brought to 400 mL with diethyl ether and extracted with water. The product
was extracted into the aqueous phase with 0.5 M HCl. The aqueous phase was neutralized
with sodium bicarbonate and extracted with dichloromethane. The organic phase was
dried over Na2SO4, filtered, and evaporated to dryness. The product was further purified
by silica gel chromatography. The product was dissolved in dichloromethane and eluted
from the column using a 0-5% step gradient of methanol in dichloromethane. Fractions
containing the desired product were combined and the solvent evaporated to provide the
1
product as an oil. Overall isolated yield was 8%. H NMR (300 MHz, CDCl3): 3.84-3.77 (6H,
d, J = 12.6 Hz, OCH3), 3.08-2.62 (6H, m, SCH2CH2N(CH)2), 1.05-0.98 (12H, d, J = 6.9 Hz,
CH(CH3)2. 31P NMR (121.4 MHz, CDCl3): 32.74 ppm.
8
ACS Paragon Plus Environment

Page 9 of 42
Biochemistry
1
2
3
4
5
6
7
8
9
Synthesis of Diethyl VR. Diethyl VR (DEVR) was made by the reaction of diethyl
chlorophosphate with N,N-diethylaminoethanethiol. N,N-diethylaminoethanethiol was
prepared from the hydrochloride salt by dissolving the compound in a saturated NaHCO3
solution and extraction with diethyl ether. The organic phase was dried over Na2SO4 and
evaporated in vacuo at room temperature. The remaining oil was distilled (50 oC) under
high vacuum and recovered as a pure liquid in a dry ice cooled trap. N,N-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
diethylaminoethanethiol (1.1 grams; 8.35 mmol) was added to 100 mL of diethyl ether and
cooled in a dry ice/acetone bath before being purged with N2. To this mixture was added
10 mL (3.0 equivalents) of a 2.5 M solution of n-butyl lithium in hexanes and the reaction
allowed to come to room temperature before re-cooling in a dry ice/acetone bath. Diethyl
chlorophosphate (2.9 g; 2 equivalents) was mixed with 30 mL of diethyl ether in a separate
flask, purged with N2, and then cooled in a dry ice/acetone bath. The cooled diethyl
chlorophosphate solution was added to the thiol solution and the reaction stirred at room
temperature for 3 hours. The reaction was brought to 400 mL with diethyl ether and
extracted with water. The product was extracted into the aqueous phase with 0.5 M HCl.
The aqueous phase was neutralized with sodium bicarbonate and extracted with
dichloromethane. The organic phase was dried over Na2SO4, filtered, and then evaporated,
yielding the product as an oil. Further purification was conducted using silica gel
1
chromatography as described above. Overall yield of the isolated product was 7%. H NMR
(300 MHz, CDCl3): 4.27-4.10 (4H, m, OCH2CH3), 2.99-2.52 (8H, m, SCH2CH2N(CH2)2), 1.43-
1.34 (6H, t, J = 7.5 Hz, OCH2CH3), 1.11-1.01 (6H, t, J = 7.2 Hz, CH2CH3). 31P NMR (121.4 MHz,
CDCl3): 28.50 ppm.
9
ACS Paragon Plus Environment

Biochemistry
Page 10 of 42
1
2
3
4
5
6
Synthesis of O-methyl VR. O-methyl VR (OMVR) was synthesized by the reaction
of methyl isobutyl chlorophosphate with N,N-diethylaminoethanethiol. N,N-
7
8
9
diethylaminoethanethiol was prepared from the hydrochloride salt as described above.
Methyl isobutyl chlorophosphate was prepared by dissolving 750 µL (8.4 mmol) of
isobutanol in 50 mL of diethyl ether. The atmosphere was purged with N2 and then the
mixture was chilled in a dry ice/acetone bath. A total of 3.3 mL (1.0 equivalent) of 2.5 M
butyl lithium in hexanes and 1.5 g (1.0 equivalents) of methyl dichlorophosphate was
added, and the reaction stirred for three hours at room temperature.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In a separate flask, 1.2 g (1.0 equivalents) of N,N-diethylaminoethanethiol was
dissolved in 50 mL of diethyl ether and chilled in a dry ice/acetone bath. A total of 5 mL
(1.5 equivalents) of 2.5 M n-butyl lithium was added and the reaction warmed to room
temperature. The methyl isobutyl chlorophosphate and the thiol solutions were chilled in a
dry ice/acetone bath and combined. The reaction was allowed to proceed at room
temperature for 3 hours. The reaction was then brought to 400 mL with diethyl ether and
washed with water. The product was extracted with 0.5 M HCl, neutralized with sodium
bicarbonate, and then extracted with dichloromethane. The organic phase was dried over
Na2SO4 and evaporated to yield the product as an oil. Overall yield of final product was
1
15%. H NMR (300 MHz, CDCl3): 3.92-3.72 (5H, m, OCH2CH(CH3)2, OCH3), 2.95-2.45 (8H, m,
SCH2CH2N(CH2)2), 2.04-1.88 (1H, OCH2CH(CH3)2), 1.06-0.97 (6H, t, J = 7.0 Hz, NCH2CH3),
0.97-0.90 (6H, d, J = 6.8 Hz, OCH2CH(CH3)2. 31P NMR (121.4 MHz CDCl3): 30.30 ppm.
Mutagenesis, Expression and Enzyme Purification. The gene for PTE was cloned
into the expression vector pET 20b between the NdeI and EcoRI restriction sites as
previously described.16 The new variants of PTE were generated by introducing the
10
ACS Paragon Plus Environment

Page 11 of 42
Biochemistry
1
2
3
4
5
6
7
8
9
mutations Y309F, I106C, I106G, and L308S into the appropriate templates by site directed
mutagenesis using the QuikChange (Agilent) protocol. DNA sequencing at the Gene
Technologies Laboratory at Texas A&M University verified the specific mutations. The
proteins were expressed and purified as previously described.7 Briefly, the variants were
freshly transformed into E. coli BL21 (DE3) cells by electroporation, and single colonies
used to inoculate 5.0 mL cultures of LB medium. After 8 hours of growth at 37 oC, 1.0 mL of
this culture was used to inoculate 1 L cultures of Terrific Broth supplemented with 1.0 mM
CoCl2. The bacterial cultures were grown at 30 oC. IPTG was added to a final concentration
of 1.0 mM after 24 hours and growth continued for 40 hours. The cells were harvested by
centrifugation and stored at -80 oC prior to purification. Cells were resuspended in 100 mL
of purification buffer (50 mM Hepes, pH 8.5, with 100 µM CoCl2) and then lysed by
sonication. The cell debris was cleared by centrifugation, nucleic acids were precipitated
by protamine sulfate (0.45 g in 20 mL purification buffer per liter of culture), and then
removed by centrifugation. The PTE mutants were precipitated with ammonium sulfate
(60% saturation) and recovered by centrifugation. The pellet was resuspended in ~5 mL
of purification buffer, filtered (0.45 µm) and loaded onto a GE Superdex 200 (16/60)
preparatory size exclusion column using a BioRad NCG FPLC system. Fractions with
catalytic activity for the hydrolysis of paraoxon were pooled and then eluted from a 3.0 g
(dry weight) DEAE Sephadex A25 resin that was pre-equilibrated in purification buffer.
Protein purity was verified by SDS-PAGE.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Enzymatic Activity. Catalytic activity with paraoxon was followed by monitoring
the release of p-nitrophenol at 400 nm (ꢀE400 = 17,000 M-1 cm-1) in 250 µL reaction
volumes containing 50 mM Ches, pH 9.0, 100 µM CoCl2, and 0 – 1.0 mM paraoxon. Activity
11
ACS Paragon Plus Environment

Biochemistry
Page 12 of 42
1
2
3
4
5
6
7
8
9
with APVR utilized the same reaction conditions as paraoxon with the release of the leaving
group monitored at 294 nm (ꢀE294 = 7,710 M-1 cm-1). Activity with DEVX, DMVX, DEVR, and
OMVR was measured in 250 µL reactions containing 50 mM Hepes, pH 8.0, 100 µM CoCl2,
0.3 mM DTNB and 0 – 1.0 mM substrate. The release of the thiol leaving group was
followed by inclusion of DTNB in the assay mixture (ꢀE412 = 14,150 M-1 cm-1). All assays
were initiated by the addition of the appropriately diluted enzyme and monitored in a 96-
well format using a Molecular Devices SpectraMax 364 Plus plate reader. Reactions were
monitored for 15 minutes at 30 oC and the linear portion of the time course was used to
calculate the initial rate. Kinetic constants were determined by fitting the data to the
Michaelis-Menten equation.23 When saturation could not be observed, the data were fit to a
linear equation and the slope taken as kcat/Km.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Stereoselective Hydrolysis of Racemic VX and VR. Low initial concentrations
(19 to 160 μM) of racemic VX and VR were hydrolyzed by variants of PTE in a solution
containing 0.1 mM CoCl2, 0.3 mM DTNB, and 50 mM Bis-Tris-propane, pH 8.0. The
reactions were followed to completion and the fraction hydrolyzed plotted as a function of
time. The time courses were fit to equations 1 and 2 where F is the fraction hydrolyzed, a
and b are the magnitudes of the exponential phases, t is time, and k1 and k2 are the rate
constants for each phase.16
F =a
(
1−e−k t
)
1
(1)
(2)
F =a
(
1−e−k t
)
+b
(
1−e−k t
)
1
2
Stereochemical preferences were determined using the previously described
complementation method.7 Briefly, variants with large stereoselective preferences were
placed in a reaction with mutants of PTE of known stereoselective preferences under
12
ACS Paragon Plus Environment

Page 13 of 42
Biochemistry
1
2
3
4
5
6
7
8
9
conditions where each variant alone would exhibit a similar rate for the first exponential
phase. If the variants prefer the same enantiomer, the resulting time course will exhibit
two distinct phases. If the variants have the opposite enantiomers as the preferential
substrate, the time courses will exhibit a single exponential phase.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
pH-Rate Profiles. The pH-rate profiles for wild-type, QF, and L7ep3a were determined
using 100 µM paraoxon as the substrate in a combination buffer system using Mes, Mops,
Hepes, and Ches buffers at 20 mM each. The pH of each stock buffer was adjusted to
desired value using KOH, and the pH was verified after reactions to assure no substantial
change in pH had occurred. Reactions were carried out in 250 µL volumes using 96-well
plates and followed at the pH-independent wavelength of 347 nm (ꢀE347 = 5,176 M-1s-1) as
previously described.24 Data were fit to equation 3 for the ionization of a single proton or
equation 4 for two protons, where y is the enzymatic rate (s-1) and c is the pH-independent
value.
c
log y= log ʢ
log y= log Ƭ
ʣ
(3)
(4)
(pK -pH)
a
1+10
c
2ư
(pK -pH)
a
1+10
X-ray Crystallography. The PTE mutants L7ep-3a and L7ep-3a I106G were
crystallized at 18 oC using the vapor diffusion method. In the crystallization experiments,
1.0 µL of protein (10 mg/mL with 1.0 mM CoCl2) was mixed with 1.0 µL of the precipitant
solution (100 mM sodium cacodylate pH 5.5-7.0, 0.2 M magnesium acetate, 15-30% PEG
8000), and then equilibrated against 500 µL of the same precipitant solution using
Intelliplates. Protein crystals appeared within a week and grew to maximum dimensions
(200 µm x 15 µm x 15 µm) after 21 days. Prior to data collection, the crystals were soaked
13
ACS Paragon Plus Environment

Biochemistry
Page 14 of 42
1
2
3
4
5
6
7
8
9
for 30 seconds in a cryoprotectant solution of the mother liquor containing 30% ethylene
glycol and then frozen in liquid nitrogen. Diffraction data were collected at 120 K on an R-
AXIS IV detector with Cu Kα X-rays produced from a rotating anode generator. X-ray data
reduction and scaling were performed with HKL2000 (25)25. Structures of the PTE
mutants L7ep-3a and L7ep-3aG were determined by molecular replacement using the
coordinates of wild-type PTE (PDB id: 1DPM) as the search model. The structures were
built using COOT and refined with simulated annealing, B-factor randomization, and
coordinate shaking using PHENIX.26, 27 Later stages of refinement were also done in
PHENIX using individual coordinate, anisotropic B-factor, and occupancy optimization. The
PTE mutant structures were refined with Rwork/Rfree values of 13.5-21.5% with excellent
geometry (Table 2).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Computational Docking of High Energy Intermediates. The pentavalent high-
energy reaction intermediate states for the hydrolysis of VX and VR were computationally
docked into the three-dimensional structures of wild-type PTE (PDB id: 1DPM),
H254Q/H257F (PDB id: 2OQL), L7ep-3a (PDB id: 4ZST), and L7ep-3aG (PDB id: 4ZSU). The
high energy intermediate states for the hydrolysis of the RP- and SP-enantiomers were
manually generated as trigonal bipyrimidal structures with the attacking hydroxyl group
protonated and the original phosphoryl oxygen substituent carrying a full negative charge
using ChemBio Ultra 14.0. Computational docking was done using the program AutoDock
Vina.28 The appropriateness of the docked poses was evaluated by the value of the distance
of the attacking hydroxyl group from the α- and β-metal ions, the distance of the
phosphoryl oxygen from the β-metal and the orientation of the side ester substituents into
the large and small group pockets contained in the active site of PTE.
14
ACS Paragon Plus Environment

Page 15 of 42
Biochemistry
1
2
3
4
5
RESULTS
6
7
Three-Dimensional Structure of L7ep-3a. The PTE variant L7ep-3a has the
8
9
highest reported kcat for the hydrolysis of the nerve agent VX.7 In an effort to elucidate the
mechanism by which the activity of this mutant is enhanced, the enzyme was crystallized
and the structure determined to a resolution of 2.01 Å by X-ray diffraction methods (PDB
id: 4ZST). The overall structure of L7ep-3a is very similar to wild-type PTE (Figure 1A).
The core of the (β/α)8-barrel matches very well between the two structures with a Cα
RMSD of 0.64 Å. The only significant change in the backbone structure is apparent in the
conformations of Loop-7 and Loop-8. In this variant, Loop-7, including the Loop-7 α-helix,
is pulled toward the active site (Figure 1B). A portion of Loop-8 is similarly pulled toward
the active site. Loop-8 also participates in the dimer interface, but the cross-interface
interactions are all retained in the L7ep-3a mutant.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The binding site for the substrate in wild-type PTE is divided into the large-group
pocket (His-254, His-257, Leu-271, and Met-317), the leaving-group pocket (Trp-131, Phe-
132, Phe-306, and Tyr-309), and the small-group pocket (Gly-60, Ile-106, Lue-303 and Ser-
308).29 The mutations H257Y and S308L, coupled with the shifting of the Loop-7 α-helix,
have induced significant changes in the substrate binding pockets in the active site of L7ep-
3a. The side-chain of Tyr-309 is repositioned so that the phenolic group now extends into
the active site rather than towards Loop-7, as previously observed in the structure of wild-
type PTE (Figures 1B and 2). The reorientation of Tyr-309, along with the substitution of
a tyrosine for His-257 and the repositioning of Leu-271 into the active site, has compressed
the size of the large-group and leaving-group pockets. The leaving-group pocket is also
constricted by the presence of the S308L mutation, which adds bulk to both the leaving-
15
ACS Paragon Plus Environment

Biochemistry
Page 16 of 42
1
2
3
4
5
6
7
8
9
group and small-group pockets. However, the apparent contraction of the leaving-group
pocket is partially relieved by the F132V mutation. Similarly, the I106C mutation opens
additional space to the small-group pocket.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Generation and Characterization of New PTE Variants. The PTE variants QF,
CVQFL, VRN-VQFL and L7ep-3a were previously shown as having substantially improved
activity for the hydrolysis of VX.7 The stereoselectivity of these mutants for the chiral
center contained in VR was determined using the isolated enantiomers of APVR (Table 3).
With the exception of QF, the VX-optimized variants of PTE prefer to hydrolyze the RP-
enantiomer of the chiral center for VR. The crystal structure of L7ep-3a suggests that the
mutations I106C and S308L alter the size of the small-group pocket. In order for the SP-
enantiomer of VR to productively bind in the active site, the larger isobutyl group must be
positioned in the small-group pocket of PTE. In an effort to maximize the hydrolysis of SP-
VR, a series of new mutants was created from the original CVQFL and VRN-VQFL variants
with residue 106 mutated to isoleucine, cysteine, or glycine in combination with residue
position 308 being mutated to either leucine or serine. In an attempt to achieve the high
kcat seen previously for the hydrolysis of VX, the I106G variant of L7ep3a was also
constructed. The catalytic activities of these variants were first characterized against the
insecticide paraoxon and the V-agent analog DEVX (Table 4), and then the stereochemical
preferences were determined using APVR (Table 3).
Isoleucine at residue position 106 favors hydrolysis of the RP-enantiomer, while
cysteine at position 106 reduces the preference for the RP-enantiomer, and glycine leads to
a preference for the SP-enatiomer. For example, the CVQFL variant has a 2.5-fold
preference for the RP-enantiomer of the VR chiral center, but CVQFL-C106I has a 35-fold
16
ACS Paragon Plus Environment

Page 17 of 42
Biochemistry
1
2
3
4
5
6
7
8
9
preference for the RP-enantiomer. Despite this large difference in stereoselectivity, both of
these variants hydrolyze DEVX with a kcat/Km of ~2 x 104 M-1 s-1. When glycine was
substituted at position 106, the stereochemical preference shifted to the SP-enantiomer of
APVR. L7ep-3a prefers the RP-enantiomer by 1.5-fold, but L7ep-3a I106G prefers the SP-
enantiomer by 6.5-fold. Unfortunately, the I106G mutation reduced the activity against
DEVX by more than an order of magnitude in these variants.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Serine at position 308 results in a much reduced stereoselectivity compared to
leucine at position 308, but at the cost of catalytic efficiency with DEVX. VRN-VQFL prefers
the RP-enantiomer of APVR by 20-fold. The variant VRN-VQFL L308S has a preference of
1.2-fold, but the activity against DEVX was diminished 2-fold. Similarly, CVQFL prefers the
RP-enantiomer of APVR by 2.5-fold, while CVQFL-L308S prefers the RP-enantiomer by 1.4-
fold, but the catalytic activity is reduced ~5-fold for DEVX. The combination of glycine at
residue position 106 with serine at 308 gave the best preference for the SP-enantiomer of
APVR, but also shows the biggest reduction in catalytic activity for DEVX. VRN-VQFL-
I106G/L308S prefers the SP-enantiomer by 14-fold, but activity with DEVX is reduced
nearly 2-orders of magnitude to a third of that seen with the wild-type enzyme.
Catalytic Activity with VX and VR. The most promising new variants were tested
at the Edgewood Chemical Biological Center with racemic VX and VR. The assays were
conducted as the complete hydrolysis of a single low concentration of these agents to
enable the observation of exponential time courses, corresponding to the hydrolysis of
each enantiomer contained within the racemic mixture. The kinetic constants are
presented in Table 5. For enzyme variants with large stereochemical preferences, the
identity of the preferred enantiomer was determined by the ability of the variant to
17
ACS Paragon Plus Environment

Biochemistry
Page 18 of 42
1
2
3
4
5
6
7
8
9
complement the slow phase of a variant of known preference. Wild-type PTE is known to
prefer to hydrolyze the RP-enantiomer of the VR chiral center, while QF is known to prefer
the SP-enantiomer of VX.7, 16. None of the variants tested, with the exception of QF,
exhibited large stereochemical preferences for hydrolysis of the two enantiomers of VX.
Removal of the S308L mutation (VRN-VQFL L308S) resulted in a 2-fold reduction in
catalytic activity for the faster enantiomer of VX and a 5-fold reduction in the rate of
hydrolysis for the slower enantiomer. Introduction of the I106G mutation (VRN-VQFL-
I106G) led to the complete hydrolysis of racemic VX without detectable selectivity, but at a
rate 6-fold slower than VRN-VQFL had for the slower enantiomer.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
With racemic VR, the VRN-VQFL variant exhibited a 20-fold enhancement in the rate
of hydrolysis compared to the wild-type enzyme, but this mutant was found to only
hydrolyze the less toxic RP-enantiomer of VR. Removal of the S308L mutation (VRN-VQFL
L308S) enabled the hydrolysis of both enantiomers of VR, with complete loss of
stereoselectivity. This represents a 64-fold improvement toward the hydrolysis of the toxic
SP-enantiomer compared to wild-type PTE. The introduction of the I106G mutation in the
VRN-VQFL-I106G/L308S variant resulted in a strong preference for the hydrolysis of the
SP-enantiomer relative to the RP-enantiomer. VRN-VQFL-I106G, which has both the I106G
and S308L mutations, has much less stereochemical preference, but the kinetic data for the
hydrolysis of the two enantiomers of APVR indicate that the stereochemical preference is
for the SP-enantiomer. The best variant identified for the hydrolysis of VR was L7ep-3a
I106G, which has a kcat/Km of 2.6 x 103 M-1 s-1 for the faster enantiomer. While the
stereoselectivity was not sufficient to determine the stereochemical preference with VR
directly, the kinetic data with the two enantiomers of APVR has identified the preferred
18
ACS Paragon Plus Environment

Page 19 of 42
Biochemistry
1
2
3
4
5
6
enantiomer as the SP-enatiomer. The L7ep-3a I106G mutant thus has a 620-fold enhanced
rate of hydrolysis of SP-VR relative to wild-type PTE.
7
8
9
Three-Dimensional Structure of L7ep-3a I106G. In an effort to understand the
physical basis for the observed improvement in the catalytic activity of L7ep-3a I106G, the
three-dimensional structure was solved by X-ray crystallography (PDB id: 4ZSU). The core
structure is very similar to wild-type PTE (Cα RMSD = 0.66 Å) and the loop structure
matches that observed with L7ep-3a.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
pH Rate Profiles for PTE. The pH rate profiles for wild-type PTE, QF, and L7ep3a
with cobalt as the active site metal were determined with paraoxon as the substrate
(Figure 3). The wild-type enzyme exhibits the loss of activity by the protonation of a single
group with an observed pKa of 5.9. The QF and L7ep3a variants have pKa values of 4.4 and
4.6, respectively. For both of these variants the pH rate profiles indicate that two groups
are protonated at low pH.
Evaluation of New V-agent Analogs. The majority of research targeting the
catalytic hydrolysis of organophosphorus nerve agents must be done using analogs for
both regulatory and safety reasons. Intrinsic to the use of substrate analogs is the
imperfect representation of catalytic activity with the authentic nerve agent. To address
which structural factors of the VR and VX analogs are most significant, the compounds
DMVX, DEVR and OMVR were synthesized and analyzed as substrates for the optimized
mutants of PTE and compared with the ability of these mutants to hydrolyze the SP-
enantiomers of VR and VX. The catalytic activity using these compounds was determined
with a series of variants for which the hydrolysis of authentic nerve agents was available
(Table 6). Wild-type PTE has a much lower catalytic activity with DMVX (kcat/Km = 1.9 x
19
ACS Paragon Plus Environment

Biochemistry
Page 20 of 42
1
2
3
4
5
6
7
8
9
101 M-1 s-1) than was observed with DEVX (kcat/Km 1.2 x 103 M-1 s-1). The best variant with
DMVX was L7ep-3a, which has a kcat = 28 s-1 and a kcat/Km = 1.3 x 104 M-1 s-1. These values
are 1,300- and 680-fold better, respectively, than wild-type PTE. Compared to DMVR, wild-
type PTE exhibits better catalytic activity with DEVR (kcat/Km = 5.6 x 102 M-1 s-1), but L7ep-
3a also had the best catalytic activity with this analog (kcat/Km = 1.5 x 104 M-1 s-1). With the
exception of wild-type PTE (kcat/Km = 6.8 x 101 M-1s-1), racemic OMVR gave the least activity
for most of the variants tested. The combination of poor solubility of this compound and
high Km values limited analysis to the determination of kcat/Km for most variants. The best
mutant for the hydrolysis of OMVR was L7ep-3a with a kcat/Km = 5.8 x 102 M-1 s-1. However,
given the switch in stereochemical preference, it is highly likely that L7ep-3a I106G has the
best catalytic activity with the RP-enantiomer of OMVR (which corresponds to the same
relative stereochemistry as the SP-enantiomer of VR).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
DISCUSSION
Comparison of Substrate Analogs. The VX analog DEVX was successfully used to
identify PTE variants for the hydrolysis of VX, but it over-estimated the activity of the wild-
type enzyme and failed to detect a 100-fold increase in the catalytic activity with the QF
variant.7 DEVX was initially synthesized to mimic the O-ethyl substituent in VX, but the
bulk of the diethyl phosphorus center is larger than the volume of the methylphosphonate
moiety of VX. DMVX also contains an achiral phosphorus center, but the volume of the
dimethyl center is more representative of authentic VX. The catalytic activity of the PTE
variants with DMVX was surprisingly low, but this analog captures the much lower
catalytic activity of wild-type PTE for the hydrolysis of VX and the substantial increase in
20
ACS Paragon Plus Environment

Page 21 of 42
Biochemistry
1
2
3
4
5
6
7
8
9
activity with the QF variant. The catalytic properties for the hydrolysis of DMVX are also
able to predict the high kcat for the hydrolysis of VX by L7ep-3a. While DEVX was intended
to ensure accommodation of the larger O-ethyl substituent, the data for DMVX suggests that
for mimicking VX, the overall size of the phosphorus center is more important. While the
asymmetry of the phosphonate center is obviously a major contributor to the catalytic
activity using VX as a substrate, the dimethyl center provided a reasonable prediction of
the catalytic activity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The compound DEVR was synthesized to test the significance of diethylamino vs.
diisopropylamino groups contained within the VR and VX leaving groups, respectively. The
smaller leaving group of DEVR resulted in 2-fold less activity compared to DEVX for wild-
type PTE. Similar differences were obtained for most of the other variants. L7ep-3a I106G
was the only variant where the catalytic activity with DEVR was less than 10% of the
catalytic activity with DEVX. The relative activity between variants was similar with either
leaving group but the reduced activity, especially manifested in the value of kcat, suggests
that the interactions of the enzyme with the leaving group may be partially responsible for
aligning the phosphorus center for nucleophilic attack. The smaller leaving group probably
contributes to the lower enzymatic activity observed with VR, but comparison to the
catalytic activity with DEVR also highlights the dominance of the phosphorus center. The
introduction of I106G in L7ep-3a resulted in a 23-fold loss of activity with DEVR, but the
activity with authentic VR was improved.
In an attempt to better reflect the asymmetric phosphorus center of VR, the analog
OMVR was synthesized. The PTE variants exhibited the least activity with OMVR, which
employs a slightly larger but asymmetric phosphorus center and the authentic leaving
21
ACS Paragon Plus Environment

Biochemistry
Page 22 of 42
1
2
3
4
5
6
7
8
9
group of VR. Kinetic constants with racemic OMVR are about an order of magnitude
smaller than is obtained with authentic VR, but the relative activity between variants is
much more consistent than is seen with the other analogs. VRN-VQFL and L7ep-3a both
have substantially better activity than wild-type PTE or QF for both OMVR and authentic
VR. Introduction of I106G or removal of S308L in VRN-VQFL resulted in somewhat
diminished rates for both VR and OMVR.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Active Site Hydrogen Bonding Network. The kcat values for wild-type PTE with
phosphorothiolate substrates are approximately 104-fold lower than that with the best
substrates.7, 30 In the hydrolysis of substrates such as paraoxon there is no need to
protonate the leaving group.31 In the proposed reaction mechanism for wild-type PTE, the
proton from the attacking hydroxide is passed to Asp-301, and, in turn, to His-254 (Figure
4A).24 The proton is then transferred to Asp-233 and on to bulk solvent. The variant QF
has been postulated to have significantly improved catalytic activity with VX in part
because of the disruption of this proton shuttle. In the crystal structure of QF, His-254 of
the wild-type enzyme is a glutamine, which cannot participate in proton shuttling, and Asp-
233 is moved out of hydrogen bonding distance (Figure 4B). It is thought that the
“trapping” of the proton in the active site is useful for the hydrolysis of slow substrates like
VX, where protonation of the leaving group will contribute to improved kcat values. The
crystal structure of L7ep-3a, which has a kcat value for the hydrolysis of VX about an order
of magnitude higher than QF, shows a remarkable rearrangement of the hydrogen bonding
pattern in the active site. In L7ep-3a, Gln-254 hydrogen bonds to Asp-301, while Asp-233
has moved back into hydrogen bonding distance to Gln-254. Tyr-257 is also hydrogen
bonded to Asp-301 (Figure 4C). The significant alteration of the hydrogen bonding
22
ACS Paragon Plus Environment

Page 23 of 42
Biochemistry
1
2
3
4
5
6
7
8
9
network of Asp-301 results in the weakening of the interaction of the bridging hydroxide
with the β-metal. This asymmetrical binding to the binuclear metal center is likely to result
in the bridging hydroxyl being a better nucleophile. In the L7ep-3a I106G mutant, a similar
hydrogen bonding pattern is observed, but Asp-233 is moved out of hydrogen bonding
distance and the displacement of the bridging hydroxyl is not nearly as pronounced.
The asymmetric positioning of the bridging hydroxyl has previously been observed
in the crystal structure of dihydroorotase in the presence of bound dihydroorotate.32
Similarly, it is known that for wild-type PTE, the pKa of the bridging water is dominated by
the interaction with the α-metal.24 In wild-type PTE activated with cobalt, the pKa
determined from the pH-rate profile is 5.9 (Figure 3). The QF variant has a substantially
reduced pKa of 4.4. As has been observed with other variants, which have the putative
proton shuttle disrupted, the slope of the pH-rate profile for the QF variant is +2 indicating
that a second group in the active site is protonated as the pH is reduced.24 Whether this
second group is Asp-301, or possibly one of the other metal ligands, is not clear. The
variant L7ep3a, which shows asymmetric binding of the hydroxyl in the active site, exhibits
the same double protonation as QF but a slightly higher pKa of 4.6.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Docking of Substrates in the Active Site. In an effort to gain insight into the
altered activity of L7ep-3a and L7ep-3a I106G, the substrates VX and VR were docked into
the active site using the program AutoDock Vina. Computational docking was conducted
using the trigonal bipyramidal intermediates formed during the hydrolysis of both
enantiomers of these compounds.33 Productive poses were assessed by the placement of
the attacking hydroxyl group between the two metal ions and the orientation of the
phosphoryl oxygen toward the β-metal.34 For the wild-type PTE and QF, both enantiomers
23
ACS Paragon Plus Environment

Biochemistry
Page 24 of 42
1
2
3
4
5
6
7
8
9
of VX and VR could be reasonably docked into the active site (data not shown). However,
for L7ep-3a and L7ep-3a I106G, only the SP-enantiomer of VR could be reasonably
positioned in the active site. For both of these variants, the constriction of the active site
appears to play an important role in the improved activity, while specific mutations
accommodate the substrates (Figure 5). The mutation F132V provides the extra room to
accommodate the isopropyl amino group of VX. This effect is not as obvious with the
smaller substituent contained within the leaving group in VR. The additional space in the
small-group pocket due to the I106G mutation accommodates the isobutyl group in SP-VR,
while the combination of the H257Y mutation and repositioning of Tyr-309 make binding
of the RP-enantiomer more difficult. The changes in the active site also enable the sulfur
and the nitrogen of the leaving group to potentially hydrogen bond with the side chain
phenol of Tyr-309, suggesting that this interaction may be partially responsible for the
dramatically improved activity. Tyr-309 was previously ruled out as playing a role in
substrate activation with paraoxon.24 The L7ep3a-Y309F mutant shows kinetics virtually
identical to L7ep3a for the hydrolysis of paraoxon, but with DEVX as the substrate kcat/Km
is reduced nearly 3-fold for the L7ep3a-Y309F variant, demonstrating that the interaction
of the phenolic oxygen with the leaving group is helpful for hydrolysis of the V-agents.
The determination of the three-dimensional structure of L7ep-3a has led to a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
greater understanding of the underlying mechanisms by which the hydrolysis of V-agents
is enhanced. The determination of this structure has guided the rational construction of
the new L7ep-3a I106G mutant, which VR and chiral center analog data suggest is
enhanced 620-fold for the hydrolysis of the SP-enantiomer of VR, relative to the wild-type
enzyme. Previous work with the insecticide demeton-S demonstrated the importance of
24
ACS Paragon Plus Environment

Page 25 of 42
Biochemistry
1
2
3
4
5
6
7
8
9
the leaving group on the catalytic activity of PTE, and our results with the new analogs of
VR and VX has further demonstrated that even small changes in the leaving group can have
dramatic effects on the activity of the enzyme.7 The crystal structures of L7ep-3a and L7ep-
3a I106G have provided a physical basis for these observations. The computational
docking results have suggested how the remodeled active site is able to exploit hydrogen
bonding interactions with Tyr-309. The disruption of the proton shuttle, along with new
hydrogen bonds to Asp-301, are apparently able to enhance the attack of the bridging
hydroxide on phosphorothiolate substrates. The initial data with the new racemic analog
OMVR suggests that the combination of an asymmetric phosphorus center and the
authentic leaving group of VR will allow for much more accurate predictions of the activity
against VR and enable the more rapid development of new variants that are more fully
optimized for the hydrolysis of VR.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACKNOWLEDGMENTS
We are grateful to the laboratory of Professor James C. Sacchettini for assistance in
collecting the X-ray crystallographic data.
FUNDING
This work was supported by the Defense Threat Reduction Agency (HDTRA1-14-1-0004).
COMPETING INTERESTS
The authors have no competing interests.
25
ACS Paragon Plus Environment

Biochemistry
Page 26 of 42
1
2
3
4
5
REFERENCES
6
7
8
[1] Benschop, H. P., and De Jong, L. P. A. (1988) Nerve agent stereoisomers: analysis, isolation
and toxicology, Acc. Chem. Res. 21, 368ꢀ374.
9
[2] Rosman, Y., Eisenkraft, A., Milk, N., Shiyovich, A., Ophir, N., Shrot, S., Kreiss, Y., and
Kassirer, M. (2014) Lessons Learned From the Syrian Sarin Attack: Evaluation of a
Clinical Syndrome Through Social MediaLessons Learned From the Syrian Sarin Attack,
Ann. Intern. Med. 160, 644ꢀ648.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[3] Leikin, J. B., Thomas, R. G., Walter, F. G., Klein, R., and Meislin, H. W. (2002) A review of
nerve agent exposure for the critical care physician, Crit. Care Med. 30, 2346ꢀ2354.
[4] Columbus, I., Waysbort, D., Marcovitch, I., Yehezkel, L., and Mizrahi, D. M. (2012) VX
Fate on Common Matrices: Evaporation versus Degradation, Environ. Sci. Technol. 46,
3921ꢀ3927.
[5] Saxena, A., Sun, W., Fedorko, J. M., Koplovitz, I., and Doctor, B. P. (2011) Prophylaxis with
human serum butyrylcholinesterase protects guinea pigs exposed to multiple lethal doses
of soman or VX, Biochem. Pharmacol. 81, 164ꢀ169.
[6] FLIR Systems Threat Detection Technologies: Enzymes, FLIR Systems, Wilsonville, OR.
http://www.flir.com/threatdetection/display/?id=63297 (accessed 8/4/15).
[7] Bigley, A. N., Xu, C., Henderson, T. J., Harvey, S. P., and Raushel, F. M. (2013) Enzymatic
neutralization of the chemical warfare agent VX: evolution of phosphotriesterase for
phosphorothiolate hydrolysis, J. Am. Chem. Soc. 135, 10426ꢀ10432.
[8] Cheng, T., Liu, L., Wang, B., Wu, J., DeFrank, J. J., Anderson, D. M., Rastogi, V. K., and
Hamilton, A. B. (1997) Nucleotide sequence of a gene encoding an organophosphorus
nerve agent degrading enzyme from Alteromonas haloplanktis, J. Ind. Microbiol.
Biotechnol. 18, 49ꢀ55.
[9] Gupta, R. D., Goldsmith, M., Ashani, Y., Simo, Y., Mullokandov, G., Bar, H., BenꢀDavid,
M., Leader, H., Margalit, R., Silman, I., Sussman, J. L., and Tawfik, D. S. (2011)
Directed evolution of hydrolases for prevention of Gꢀtype nerve agent intoxication, Nat.
Chem. Biol. 7, 120ꢀ125.
[10] Kirby, S. D., Norris, J. R., Richard Smith, J., Bahnson, B. J., and Cerasoli, D. M. (2013)
Human paraoxonase double mutants hydrolyze V and G class organophosphorus nerve
agents, Chem. Biol. Interact. 203, 181ꢀ185.
[11] Melzer, M., Chen, J. C., Heidenreich, A., Gab, J., Koller, M., Kehe, K., and Blum, M. M.
(2009) Reversed enantioselectivity of diisopropyl fluorophosphatase against
organophosphorus nerve agents by rational design, J. Am. Chem. Soc. 131, 17226ꢀ17232.
26
ACS Paragon Plus Environment

Page 27 of 42
Biochemistry
1
2
3
4
5
6
[12] Tsai, P. C., Fox, N., Bigley, A. N., Harvey, S. P., Barondeau, D. P., and Raushel, F. M.
(2012) Enzymes for the homeland defense: optimizing phosphotriesterase for the
hydrolysis of organophosphate nerve agents, Biochemistry 51, 6463ꢀ6475.
7
8
9
[13] Oudejans, L. (2013) Evaluation Report EPA 600/Rꢀ12/033: Enzymatic Decontamination of
Chemical Warfare Agents, United States Envirnmental Protection Agency. Washinton
D.C.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[14] Scott, C. (2012) Landguard A900: An enzymeꢀbased remdiant for the detoxification of
organophosphate insecticides in animal dips, CSIRO Ecosystem Science, Canberra,
Australia. https://publications.csiro.au/rpr/download?pid=csiro:EP125235&dsid=DS2
(accessed 8/4/15)
[15] LeJeune, K. E., and Russell, A. J. (1999) Biocatalytic nerve agent detoxification in fire
fighting foams, Biotechnol. Bioeng. 62, 659ꢀ665.
[16] Tsai, P. C., Bigley, A., Li, Y., Ghanem, E., Cadieux, C. L., Kasten, S. A., Reeves, T. E.,
Cerasoli, D. M., and Raushel, F. M. (2010) Stereoselective hydrolysis of
organophosphate nerve agents by the bacterial phosphotriesterase, Biochemistry 49,
7978ꢀ7987.
[17] MeeꢀHie Cho, C., Mulchandani, A., and Chen, W. (2006) Functional analysis of
organophosphorus hydrolase variants with high degradation activity towards
organophosphate pesticides, Protein Eng. Des. Sel. 19, 99ꢀ105.
[18] Roodveldt, C., and Tawfik, D. S. (2005) Directed evolution of phosphotriesterase from
Pseudomonas diminuta for heterologous expression in Escherichia coli results in
stabilization of the metalꢀfree state, Protein Eng. Des. Sel. 18, 51ꢀ58.
[19] Rastogi, V. K., DeFrank, J. J., Cheng, T. C., and Wild, J. R. (1997) Enzymatic hydrolysis of
RussianꢀVX by organophosphorus hydrolase, Biochem. Biophys. Res. Commun. 241,
294ꢀ296.
[20] Cherny, I., Greisen, P., Jr., Ashani, Y., Khare, S. D., Oberdorfer, G., Leader, H., Baker, D.,
and Tawfik, D. S. (2013) Engineering Vꢀtype nerve agents detoxifying enzymes using
computationally focused libraries, ACS Chem. Biol. 8, 2394ꢀ2403.
[21] Reiter, G., Muller, S., Hill, I., Weatherby, K., Thiermann, H., Worek, F., and Mikler, J.
(2015) In vitro and in vivo toxicological studies of V nerve agents: molecular and
stereoselective aspects, Toxicol. Lett. 232, 438ꢀ448.
[22] Amitai, G., Ashani, Y., Grunfeld, Y., Kalir, A., and Cohen, S. (1976) Synthesis and
properties of 2ꢀSꢀ(N,Nꢀdialkylamino)ethyl)thioꢀ1,3,2ꢀdioxaphosphorinane 2ꢀoxide and of
the corresponding quaternary derivatives as potential nontoxic antiglaucoma agents, J.
Med. Chem. 19, 810ꢀ813.
[23] Michaelis, L., and Menten, M. L. . (1913) Die Kinetik der Invertinwirkung, Biochem. Z 49,
333ꢀ369.
27
ACS Paragon Plus Environment

Biochemistry
Page 28 of 42
1
2
3
4
5
[24] Aubert, S. D., Li, Y., and Raushel, F. M. (2004) Mechanism for the hydrolysis of
organophosphates by the bacterial phosphotriesterase, Biochemistry 43, 5707ꢀ5715.
6
7
[25] Otwinowski Z, M. W. ( 1997) Processing of Xꢀray diffraction data collected in oscillation
8
9
mode, Methods Enzymol. 276A, 307–326.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[26] Adams, P. D., Afonine, P. V., Bunkoczi, G., Chen, V. B., Davis, I. W., Echols, N., Headd, J.
J., Hung, L.ꢀW., Kapral, G. J., GrosseꢀKunstleve, R. W., McCoy, A. J., Moriarty, N. W.,
Oeffner, R., Read, R. J., Richardson, D. C., Richardson, J. S., Terwilliger, T. C., and
Zwart, P. H. (2010) PHENIX: a comprehensive Pythonꢀbased system for macromolecular
structure solution, Acta Crystallogr. D Biol. Crystallogr. 66, 213ꢀ221.
[27] Emsley, P., and Cowtan, K. (2004) Coot: modelꢀbuilding tools for molecular graphics, Acta
Crystallogr. D Biol. Crystallogr. 60, 2126ꢀ2132.
[28] Trott, O., and Olson, A. J. (2010) AutoDock Vina: Improving the speed and accuracy of
docking with a new scoring function, efficient optimization, and multithreading, J.
Comput. Chem. 31, 455ꢀ461.
[29] ChenꢀGoodspeed, M., Sogorb, M. A., Wu, F., Hong, S. B., and Raushel, F. M. (2001)
Structural determinants of the substrate and stereochemical specificity of
phosphotriesterase, Biochemistry 40, 1325ꢀ1331.
[30] Caldwell, S. R., Newcomb, J. R., Schlecht, K. A., and Raushel, F. M. (1991) Limits of
diffusion in the hydrolysis of substrates by the phosphotriesterase from Pseudomonas
diminuta, Biochemistry 30, 7438ꢀ7444.
[31] Hong, S. B., and Raushel, F. M. (1996) Metalꢀsubstrate interactions facilitate the catalytic
activity of the bacterial phosphotriesterase, Biochemistry 35, 10904ꢀ10912.
[32] Thoden, J. B., Phillips, G. N., Neal, T. M., Raushel, F. M., and Holden, H. M. (2001)
Molecular Structure of Dihydroorotase:ꢁ A Paradigm for Catalysis through the Use of a
Binuclear Metal Center, Biochemistry 40, 6989ꢀ6997.
[33] Caldwell, S. R., Raushel, F. M., Weiss, P. M., and Cleland, W. W. (1991) Transitionꢀstate
structures for enzymatic and alkaline phosphotriester hydrolysis, Biochemistry 30, 7444ꢀ
7450.
[34] Vanhooke, J. L., Benning, M. M., Raushel, F. M., and Holden, H. M. (1996) Threeꢀ
Dimensional Structure of the ZincꢀContaining Phosphotriesterase with the Bound
Substrate Analog Diethyl 4ꢀMethylbenzylphosphonate, Biochemistry 35, 6020ꢀ6025.
28
ACS Paragon Plus Environment

Page 29 of 42
Biochemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1
29
ACS Paragon Plus Environment