&EPA
EPA/600/R-07/132 I November 2007 I www.epa.gov/ord
United States
Environmental Protection
Agency
Catalytic Enzyme-Based Methods
for Water Treatment and Water
Distribution System
Decontamination
FINAL REPORT
Office of Research and Development
National Homeland Security Research Center
-------
Blank
-------
EPA/600/R-07/132
November 2007
Catalytic Enzyme-Based Methods for Water Treatment
and Water Distribution System Decontamination
Final Report
Prepared by
Joseph J. DeFrank
Research and Technology Directorate
Edgewood Chemical Biological Center
Aberdeen Proving Ground, MD 21010-5424
and
Nona J. Fry
Gregory J. Pellar
Science Applications International Corporation
Abingdon, MD21009
Prepared under
U.S. EPA Contract No. ECBC-TR-489
Submitted to
Paul Randall
Prepared for
National Homeland Security Research Center
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
Blank
-------
DISCLAIMER
The work described in this report was authorized under the "Catalytic
Enzyme-Based Methods for Water Treatment and Water Distribution System
Decontamination" project funded by the U.S. Environmental Protection Agency. This
work was started in May 2004 and completed in September 2005.
The use of either trade or manufacturers' names in this report does not
constitute an official endorsement of any commercial products. This report may not be
cited for purposes of advertisement.
This report has been peer and administratively reviewed and has been
approved for publication as an EPA document. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use of a specific
product.
This report was also released by the Edgewood Chemical Biological
Center, U.S. Army Research, Development and Engineering Command, as ECBC-TR-
489.
-------
Blank
-------
CONTENTS
1. INTRODUCTION 8
2. QUALITY ASSURANCE PROJECT PLAN 8
3. IMMOBILIZED ENZYME DECONTAMINATION OF DRINKING WATER 9
3.1 Phase I Preliminary Studies 9
3.1.1 Immobilization/encapsulation 9
3.2 System ization 11
3.3 Bench-Scale Testing 15
3.3.1 Statistical analysis of the 2-liter system 18
3.4 Conclusions 22
4. STABILIZED ENZYME DECONTAMINATION OF DRINKING WATER 24
4.1 Phase I preliminary studies 24
4.1.1 Stabilization 24
4.2 Systemization 26
4.3 Bench-Scale Testing 26
4.3.1 Statistical analysis of the 2-liter system 30
4.4 Conclusions 35
5. GENERAL CONCLUSIONS 36
6. ABBREVIATIONS AND ACRONYMS 37
-------
FIGURES
1. Specific Activity of Free and Immobilized OPAA in Tap Water 10
2. Specific Activity of Free and Immobilized OPH in Tap Water 11
3. Comparison of Paraoxon Loss With Different Tubing 12
4. Reverse Circulating Filter Loop System 13
5. Five Day Paraoxon Catalysis by OPH-Agarose 14
6. Comparison of OPH-Agarose and Untreated Paraoxon in Tap Water 14
7. Paraoxon Catalysis in Catalytic and Control 2-Liter Filter Loop Systems 15
8. Total umoles pNP Produced by the OPAA-Agarose, OPH-Agarose, and 16
Their BSA-Agarose Controls in the 2-Liter Systems
9. OPAA-Agarose and BSA-Agarose 2-Liter Systems After Five-Day 17
Treatment With Paraoxon in Tap Water
10. pH Profile of the Catalytic and Control Filter Loops 18
11. Specific Activity of PEGylated OPH in Tap Water 25
12. Specific Activity of PEGylated OPAA in Tap Water 25
13. Bench-Scale System Set-up 26
14. Paraoxon Catalysis by OPH 2-kDa 27
15. OPH-2kDa Paraoxon Hydrolysis Measured in ECBC Tap Water 27
16. Paraoxon Catalysis in Catalytic and Control 2-Liter Bench-Scale Systems 28
17. Total umoles pNP produced by the OPH-2kDa, OPAA-2kDa, and Their
BSA-2kDa Controls in the 2-Liter Systems 29
18. Photograph of Bench Study 29
19 pH Profile of the 2-Liter Reactors During Paraoxon Hydrolysis 30
-------
TABLES
1. T Test Analysis of the OPH/BSA-Agarose Bench-Scale Experimental Data 19
2. T Test Analysis of the OPAA/BSA-Agarose Bench-Scale Experimental Data.... 21
3. T Test Analysis of the OPH/BSA 2-kDa Bench-Scale Experimental Data 31
4. T Test Analysis of the OPAA/BSA 2-kDa Bench-Scale Experimental Data 33
-------
CATALYTIC ENZYME-BASED METHODS FOR WATER TREATMENT
AND WATER DISTRIBUTION SYSTEM DECONTAMINATION
FINAL REPORT
1. INTRODUCTION
Drinking water distribution systems supplying large population centers
must be considered serious potential targets for terrorists. Contamination of distribution
system equipment would result from adherence of contaminants to biofilms, tubercles,
and other corrosion products lining the pipes, or from permeation of the pipe material
itself. Because of their nontoxic, noncorrosive, and environmentally benign properties,
enzymes may provide an ideal method for the treatment of agents, pesticides, or other
chemical contaminants in drinking water systems, as well as for the decontamination of
pipes and other equipment with contaminant residue. In addition, enzymes have been
demonstrated to function in foams, sprays, lotions, detergents, and other vehicles that
can be used in flowing water or on material surfaces.
Many special requirements need to be considered in the application of
enzymes to contaminated drinking water systems. Because of the large volumes of
water contained in water distribution and treatment systems, a decontaminant will need
to be active for a much longer time than in military operations. Since drinking water
flows very quickly in pipes, methods are needed to ensure that the enzymes maintain
sufficient contact with the contaminated water or materials.
The goal of this project is to identify, develop, and evaluate at least one
enzyme-based method for treating flowing contaminated water and one enzyme-based
method for decontaminating drinking water pipes.
2. QUALITY ASSURANCE PROJECT PLAN (QAPP)
The QAPP (E4) for this project was compiled jointly by the Edgewood
Chemical Biological Center (ECBC) and Neptune Associates, Inc. personnel. This
comprehensive document (EPA QAPP No: WS3.4.d.10) provided quality assurance
guidance for both Phase I (baseline) and Phase II (bench) studies, although it was only
applicable to the Phase II operations according to the work plan. This document
covered the responsibilities of the personnel involved, quality standards expected for
the project, implementation of these standards, explanations of the technologies and
procedures involved, and the procedures for the statistical analysis. The initial approved
QAPP (5/11/2005) was revised once to reflect corrections needed in the initial
document and to modify some of the experimental procedures that were updated after
the original submission in March, 2005. The final corrected QAPP was approved
7/21/05 and received by ECBC personnel 8/4/05. The approved quality procedures
were implemented for the Phase II bench study.
-------
3. IMMOBILIZED ENZYME DECONTAMINATION OF DRINKING WATER
The initial part of this project, a literature survey, was conducted to
examine the types of enzymes that could be used in the decontamination of tap water
as well as the methods for immobilizing and/or stabilizing them. Enzymes were
identified with activity against organophosphorus nerve agents and pesticides, sulfur
mustard and halogenated pesticides, carbamate pesticides, cyanide, biological agents,
toxins, and biofilms. However, because of their more advanced status, the two nerve
agent/pesticide degrading enzymes organophosphorus acid anhydrolase (OPAA) JD6.5
and organophosphorus hydrolase (OPH) were selected for use in the Phase I and
Phase II studies.
3.1 Phase I Preliminary Studies
3.1.1 Immobilization/encapsulation
These studies examined the effect of immobilizing OPH and OPAA on
enzyme activity after exposure to tap water. This was needed to ensure that the
immobilization technology chosen would result in an active enzyme system after five
days, which was the examination period for the subsequent tap water bench studies.
Initial studies used enzyme kinetic rate analysis as the activity benchmark. This
benchmark was examined at Day 0 and after five days' storage in ECBC tap water.
Kinetic rate comparisons were made between the different immobilization techniques to
find the technique that resulted in the highest activity after the five days of storage.
Paraoxon was used as an OPH substrate because p-nitrophenyl Soman hydrolytic
activity is a very poor substrate for OPH. Although OPAA has better catalytic activity
with p-nitrophenyl Soman, this substrate can't be purchased commercially (unlike
paraoxon), and problems encountered with p-nitrophenyl Soman synthesis by a local
chemist precluded its use for the bench studies. As such, paraoxon was also used as
the substrate for OPAA.
Several immobilization methods were examined for both enzymes. These
included covalent attachment of OPH and OPAA to solid supports such as
polyacrylamide, agarose, and controlled-pore glass beads. Encapsulation of the
enzymes in sol-gels was also examined.
The activity results of the covalently-coupled enzymes are shown in
Figures 1 and 2. The specific activity at Day 0 for the free enzyme is set at 100 percent.
For OPH, the best activity after immobilization is seen with the azlactone-
polyacrylamide coupling method. For OPAA, the best coupling method was Amino-link
Plus agarose. The lowest activity was seen with the azlactone polyacrylamide method
for OPAA and with controlled pore glass for OPH. Preservation of the free enzyme
activity level after immobilization was much better for OPH than for OPAA with all
methods. Preservation of the initial post-modification activity level after five days was
best for both enzymes with the azlactone polyacrylamide coupling method.
-------
The sol-gel encapsulation method used in this study was the
polymerization of locust bean gum (LBG) galactomannan with Tetrakis (2-hydroxyethyl)
orthosilicate (THEOS) to form hybrid silica nanocomposites. The LBG/THEOS
encapsulation method retained the enzyme very well and resulted in detectable enzyme
activity (Figures 1 and 2) after the encapsulation and diffusion of the excess ethylene
glycol. In comparison to the other immobilization methods, the activity performance of
sol-gel OPAA ranked second behind OPAA-agarose and the activity of sol-gel OPH
ranked second behind OPH-polyacrylamide over the five-day tap water storage
examination period. Activity retention after sol-gel encapsulation was much poorer for
OPAA than for OPH, presumably because no covalent modification of the enzyme
occurred during the encapsulation to protect the enzyme. OPH was far more stable as a
free enzyme than was OPAA, which probably accounts for its higher activity as an
unmodified enzyme after sol-gel encapsulation. Unfortunately, highly concentrated
preparations of enzyme were necessary for this procedure because they were highly
diluted by the addition of the THEOS and LBG. The only commercially available THEOS
had a purity of 20 percent (v/v), with the balance of the preparation being ethylene
glycol, so this decreased the volume of enzyme that could be added to the system. In
addition, the aqueous solubility of LBG was low, so it was not possible to make a
concentrated solution of this polymer, which further decreased the enzyme addition
volume. Despite these limitations, we were able to encapsulate sufficient enzyme to
compare hydrogel enzyme activity to that of the other immobilization methods.
IUU "
yu -
oU -
° 7n
5* 70 "
03
Q
— on
* °°
>
^ en
t5 -3D
<
< /n
Q_
O
H 30~
ul
S? 20
m -
DDay 0
• Day5
Fr
ee
00
24.
— I
~~l rta n PL
OPAA-
DPAA OPAA-A OPAA-P ^££ OPAA sol
L/r O
000 22.472 4.775 8.708 9.551
438 5.618 3.933 5.337 3.652
Figure 1. Specific activity of free and immobilized OPAA at zero
and five day in tap water. A = agarose; P = polyacrylamide; CPG=
controlled pore glass; and sol = THEOS-LBG sol gel.
10
-------
100
3.2
Figure 2. Specific activity of free and immobilized OPH at zero
and five days in tap water. A = agarose; P = polyacrylamide;
CPG= controlled pore glass; and sol = THEOS-LBG sol-gel.
System ization
The immobilized enzymes were used to filter-decontaminate paraoxon
from tap water. The benchmark for these studies was the amount of paraoxon
hydrolyzed to p-nitrophenol over a five-day treatment period. A small-scale (50-ml)
reservoir loop was used to transition from the initial rate studies to the 2-liter, bench-
scale studies. The mixing reservoir and the enzyme filter were foil-wrapped to protect
the pNP from light. Many unanticipated technical challenges arose while implementing
this transitional system. These required resolution before the bench-scale
decontamination studies could be attempted.
11
-------
8000
-Pharmed-glass -*-all silicon -*- silicon-glass
First, paraoxon adsorbed to the Tygon and silicon tubing, and p-
nitrophenol adsorbed to the fittings. Further, paraoxon and its hydrolysis product, p-
nitrophenol, were used as a nutritional source by the native bacteria in the tap water,
resulting in formation of biofilms in the tubing. This, in turn, adversely affected the
accuracy of the paraoxon and p-nitrophenol measurements. Sterilization of the system
by autoclaving eliminated bacterial degradation of the substrate/product; however,
Tygon tubing did not survive autoclaving well, so its use was discontinued. The fittings
and most of the tubing were replaced with glass capillaries and polypropylene fittings. A
tubing comparison showed that Pharmed™ tubing gave the least paraoxon adsorption
(Figure 3). Silicon gave the highest paraoxon adsorption; over 90 percent was removed
from the system in four days. The geometry of the system was changed (Figure 4) so
that Pharmed™ tubing did not come into contact with the treatment water until after it
had passed through the immobilized enzyme filter (reverse loop). Using these
modifications and a 24-hour residence time (the time for a sample to pass through the
system), >99 percent of the paraoxon (0.1 mM or 27.5 ppm initial) was hydrolyzed to p-
nitrophenol during the five-day treatment period with the OPH-agarose filter compared
to 4 percent for the untreated control (Figures 5 and 6).
12
-------
Pharmed™
Tubing
Peristaltic
Pump
Enzyme/
BSA
Filter
Stir bar
Glass
Tubing
Mixing
Vessel
Insulator
Stir plate
Figure 4. Reverse circulating filter loop system used in the
50- and 2000-ml systems.
13
-------
100
90
80
"g 70
I
2 60
I
c 50
o
X
2 40
ro
Q.
S? 30
20
10
0
0 1000 2000 3000 4000 5000 6000 7000
min
-OPH-agarose
-Control
Figure
the 50-
5. Five-day paraoxon catalysis by OPH-agarose in
ml reverse circulating enzyme filter loop system.
Figure 6. Comparison of OPH-agarose treated (right) and untreated
(left) paraoxon in tap water after five days in the 50-ml
Pharmed™/glass system. The yellow compound is the p-nitrophenolate
ion of p-nitrophenol, one of the paraoxon hydrolysis products.
14
-------
3.3
Bench-Scale Testing
The bench circulating loop system tested the feasibility of tap water
decontamination with an immobilized enzyme filter. All systems tested were sterilized by
autoclaving to prevent anomalous results from bacterial growth. Obviously, this is not
feasible for large-scale application of the technology. However, in actual use it is
anticipated that a disinfectant or biofilm-degrading system/enzyme will also be
incorporated, thus eliminating this problem. The enzyme filter (30-33-ml bed volume)
circulated 2 liters of 0.091-0.096 mM paraoxon (actual, measured by base hydrolysis) in
ECBC tap water with a hydraulic residence time of 24 h at 24°C. The mixing reservoirs
and the test filters were foil-wrapped to protect the pNP from light. Both OPH-agarose
and OPAA-agarose were used in this demonstration. BSA-agarose was run in parallel
with each enzyme filter as the nonenzymatic control under the same operating
conditions. Temperature, pH, and absorbance (A405) were monitored during the five-
day demonstration period according to the schedule.
The 2-liter apparatus was an enlarged version of the 50-ml system. Larger
Pharmed™ tubing and glass capillaries were built into this system to handle the larger
flow rates (1.39 ml/min). The 50-ml system pump (Rainin RP4) was also used in the
2-liter system. The observed temperature of all systems was 24°C.
1000 2000 3000 4000 5000 6000 7000 8000
min
-%OPAA-A
-% BSA-A
% OPH-A
% BSA-A
Figure 7. Paraoxon catalysis in catalytic and control 2-liter filter loop systems.
Error bars are the +/- 95 percent confidence levels.
15
-------
0 '»
200
1000 2000 3000 4000 5000 6000 7000 8000
min
pNP OPH-BSA -B- |jm pNP OPH 3- |jm pNP BSA
0 1000 2000 3000 4000 5000 6000 7000 8000
min
- |jm pNP OPAA-BSA -B- |Jm pNP OPAA -B- |Jm pNP BSA
Figure 8. Total umoles pNP produced by the OPAA-agarose, OPH-agarose, and
their BSA-agarose controls in the 2-liter systems. The enzyme-BSA plots show the
net catalytic umoles produced. Error bars show the +/- 95 percent confidence
limits.
16
-------
Results of the catalytic filter loop paraoxon decontamination systems
showed excellent performance from both immobilized enzymes on agarose (Figure 7).
After the five-day treatment, the catalytic filters hydrolyzed 99.4-99.8 percent of the
paraoxon. This is in contrast to the control filter loop, which showed only 5.9-6.3
percent paraoxon hydrolysis during the same examination period. The net catalytic
paraoxon hydrolysis from OPAA-agarose and OPH-agarose was 92.8 and 93.9 percent,
respectively (Figure 8). p-Nitrophenol production from paraoxon was quite evident in the
2-liter catalytic filter system compared with the control filter system (Figure 9).
The pH of the systems was also divergent (Figure 10). The initial mean pH
of the catalytic systems was 7.57 (enzyme) and 7.60 (BSA). After the five-day
treatment, the final mean pH was 8.11 (enzyme) and 8.47 (BSA). The lower pH of the
catalytic filter systems is from the production of the acidic products of paraoxon
hydrolysis. The rapid accumulation of these products during the first day accounts for
the observed drop in pH during this period for both enzyme filter systems.
17
-------
0 1000 2000 3000 4000 5000 6000 7000 8000
min
•OPAApH-B-BSApH-A-OPH pH O BSA pH
3.3.1
Statistical analysis of the 2-liter system
Triplicate data generated during the bench studies was subjected to the
T test (in Microsoft Excel) to determine whether the absorbances at 405 nm (A405) of
the enzyme filter systems were significantly different from those of the control filter
systems. The T test determines the significant differences between the catalytic and the
control data based upon the chance that random probability could produce the observed
numbers within a predetermined confidence limit. Using the paired two samples for
means analysis (two-tailed), the resulting parameters of t stat, t critical, and P values
were examined for each time point. Our criteria were that the t stat should be > the
t critical value (two-tailed) and that the P value (two-tailed) should be < 0.05, using
95 percent confidence limits. The results are shown in Tables 1 and 2.
18
-------
TABLE 1
T test analysis of the OPH/BSA-agarose bench-scale experimental A405 data
Mean
Variance
Observations
Pearson
Correlation
Hypothesized
Mean Difference
deg. freedom
tStat
P (T<=t) one-tail
t Critical one-tail
P(T<=t) two-tail
t Critical two-tail
Mean
Variance
Observations
Pearson
Correlation
Hypothesized
Mean Difference
deg. freedom
tStat
P(T<=t) one-tail
t Critical one-tail
P(T<=t) two-tail
t Critical two-tail
OPH
0
0.008333
3.33E-07
3
-0.5
0
2
0
0.5
4.302653
1
6.205347
OPH
240
0.233667
3.33E-07
3
-0.5
0
2
381.0512
3.44E-06
4.302653
6.89E-06
6.205347
BSA
0
0.008333
3.33E-07
3
BSA
240
0.013667
3.33E-07
3
OPH
60
0.045333
3.33E-07
3
0.5
0
2
113
3.92E-05
4.302653
7.83E-05
6.205347
OPH
300
0.286333
3.33E-07
3
3.93E-14
0
2
826
7.33E-07
4.302653
1 .47E-06
6.205347
BSA
60
0.007667
3.33E-07
3
BSA
300
0.011
4.51E-36
3
OPH
120
0.108
0
3
#DIV/0!
0
2
65535
#NUM!
4.302653
#NUM!
6.205347
OPH
360
0.340333
3.33E-07
3
0.5
0
2
983
5.17E-07
4.302653
1.03E-06
6.205347
BSA
120
0.01
0
3
BSA
360
0.012667
3.33E-07
3
OPH
180
0.158333
3.33E-07
3
#DIV/0!
0
2
454
2.43E-06
4.302653
4.85E-06
6.205347
OPH
1440
1.015667
3.33E-07
3
-1
0
2
1487
2.26E-07
4.302653
4.52E-07
6.205347
BSA
180
0.007
0
3
BSA
1440
0.024333
3.33E-07
3
19
-------
Table 1 (continued)
Mean
Variance
Observations
Pearson
Correlation
Hypothesized
Mean
Difference
Deg. freedom
tStat
t Critical one-tail
P (T<=t) two-tail
t Critical two-tail
Mean
Variance
Observations
Pearson
Correlation
Hypothesized
Mean
Difference
Deg. freedom
tStat
P(T<=t) one-tail
t Critical one-tail
P (T<=t) two-tail
t Critical two-tail
OPH
1800
1.148333
1 .33E-06
3
0.5
0
2
1938.165
1.33E-07
4.302653
2.66E-07
6.205347
OPH
4680
1 .474667
1 .03E-05
3
-0.77771
0
2
606.2857
1 .36E-06
4.302653
2.72E-06
6.205347
BSA
1800
0.029333
3.33E-07
3
BSA
4680
0.06
0.000001
3
OPH
2880
1.363333
1.33E-06
3
-0.5
0
2
1499.763
2.22E-07
4.302653
4.45E-07
6.205347
OPH
5760
1.474
0.000001
3
0.5
0
2
2431.799
8.46E-08
4.302653
1.69E-07
6.205347
BSA
2880
0.040667
3.33E-07
3
BSA
5760
0.07
1E-06
3
OPH
3240
1 .409667
3.33E-07
3
-0.5
0
2
1548.142
2.09E-07
4.302653
4.17E-07
6.205347
OPH
6120
1.482
7E-06
3
-0.61859
0
2
643.3915
1.21E-06
4.302653
2.42E-06
6.205347
BSA
3240
0.044333
1.33E-06
3
BSA
6120
0.075667
2.33E-06
3
OPH
4320
1.461
3E-06
3
-0.5
0
2
965.8402
5.36E-07
4.302653
1.07E-06
6.205347
OPH
7200
1 .474667
1.33E-06
3
-1
0
2
1390
2.59E-07
4.302653
5.18E-07
6.205347
BSA
4320
0.057667
1.33E-06
3
BSA
7200
0.084667
3.33E-07
3
All results meet the significance criteria, except for the Time 0 sample (P =
1, t stat = 0). This was expected, as both systems were untreated at Time 0, so their
absorbance values should not differ significantly. The 120' values showed no variance
between replicates, so it was not possible to get a P value from this data (can't divide by
zero variance). The t stat, however, was much larger than the t critical (65535 > 6.205),
so these measurements do meet this significance criterion.
OPAA/BSA T test analysis for the A405 data is shown in Table 2:
20
-------
TABLE 2
T test analysis of the OPAA/BSA-agarose bench-scale experimental A405 data
Mean
Variance
Observations
Pearson
Correlation
Hypothesized
Mean Difference
Deg. freedom
tStat
P (T<=t) one-tail
t Critical one-tail
P (T<=t) two-tail
t Critical two-tail
Mean
Variance
Observations
Pearson
Correlation
Hypothesized
Mean Difference
Deg. freedom
tStat
P (T<=t) one-tail
t Critical one-tail
P (T<=t) two-tail
t Critical two-tail
OPAA
0
0.008333
3.33E-07
3
0.5
0
2
-4
0.028595
4.302653
0.057191
6.205347
OPAA
240
0.269333
3.33E-07
3
-0.5
0
2
293.3004
5.81E-06
4.302653
1.16E-05
6.205347
BSA
0
0.009667
3.33E-07
3
BSA
240
0.010667
1.33E-06
3
OPAA
60
0.058333
2.33E-06
3
0.188982
0
2
56.31671
0.000158
4.302653
0.000315
6.205347
OPAA
300
0.324667
2.33E-06
3
-0.65465
0
2
235.25
9.03E-06
4.302653
1.81E-05
6.205347
BSA
60
0.008667
3.33E-07
3
BSA
300
0.011
0.000001
3
OPAA
120
0.128333
3.33E-07
3
0.5
0
2
359
3.88E-06
4.302653
7.76E-06
6.205347
OPAA
360
0.356667
3.33E-07
3
-3.9E-14
0
2
1034
4.68E-07
4.302653
9.35E-07
6.205347
BSA
120
0.008667
3.33E-07
3
BSA
360
0.012
4.51E-36
3
OPAA
180
0.203
1.16E-33
3
0
0
2
584
1 .47E-06
4.302653
2.93E-06
6.205347
OPAA
1440
1 .077667
2.33E-06
3
-0.65465
0
2
786.5
8.08E-07
4.302653
1 .62E-06
6.205347
BSA
180
0.008333
3.33E-07
3
BSA
1440
0.029
1E-06
3
21
-------
Table 2 (continued)
Mean
Variance
Observations
Pearson
Correlation
Hypothesized
Mean Difference
Deg. freedom
tStat
P (T<=t) one-tail
t Critical one-tail
P(T<=t) two-tail
t Critical two-tail
Mean
Variance
Observations
Pearson
Correlation
Hypothesized
Mean Difference
Deg. freedom
tStat
P (T<=t) one-tail
t Critical one-tail
P (T<=t) two-tail
t Critical two-tail
OPAA
1800
1.199
3E-06
3
-1
0
2
874.25
6.54E-07
4.302653
1.31E-06
6.205347
OPAA
4680
1 .508667
2.33E-06
3
0.188982
0
2
1630.917
1.88E-07
4.302653
3.76E-07
6.205347
BSA
1800
0.033333
3.33E-07
3
BSA
4680
0.070333
3.33E-07
3
OPAA
2880
1.393667
3.33E-07
3
-0.5
0
2
2327.876
9.23E-08
4.302653
1.85E-07
6.205347
OPAA
5760
1.525
3E-06
3
-0.5
0
2
1203.422
3.45E-07
4.302653
6.9E-07
6.205347
BSA
2880
0.049667
3.33E-07
3
BSA
5760
0.078667
3.33E-07
3
OPAA
3240
1.43
3E-06
3
-0.80296
0
2
590.8571
1 .43E-06
4.302653
2.86E-06
6.205347
OPAA
6120
1.532333
3.33E-07
3
-0.86603
0
2
1643.39
1 .85E-07
4.302653
3.7E-07
6.205347
BSA
3240
0.051333
6.33E-06
3
BSA
6120
0.083
0.000001
3
OPAA
4320
1.49
0.000001
3
0.866025
0
2
4277
2.73E-08
4.302653
5.47E-08
6.205347
OPAA
7200
1.542333
6.33E-06
3
0.953821
0
2
2504.545
7.97E-08
4.302653
1.59E-07
6.205347
BSA
4320
0.064333
3.33E-07
3
BSA
7200
0.096333
9.33E-06
3
All results meet the significance criteria, except for the Time 0 sample (P =
0.057, t stat = -4). This was expected, as both systems were untreated at Time 0, so
their absorbance values should not differ significantly.
3.4
Conclusions
Preliminary studies showed that the paraoxon (and nerve agent)
hydrolyzing enzymes OPAA and OPH could be successfully immobilized with four
different methods. Three of these were covalent immobilization on solid supports
(agarose, polyacrylamide, and controlled pore glass) and one encapsulated the
enzymes in a hybrid silica nanocomposite (sol-gel). All immobilization reactions resulted
in loss of enzyme activity, but this loss varied with the enzyme type and the
immobilization method. The immobilized enzymes were tested for activity stability
22
-------
before and after five days' tap water storage. The best immobilization method for activity
was with azlactone-polyacrylamide for OPH (paraoxon) and with Amino-link Plus
agarose for OPAA (p-nitrophenyl Soman). The best stability after five days' tap water
storage was with azlactone-polyacrylamide for both enzymes. Although p-nitrophenyl
Soman was the OPAA substrate for these preliminary studies, it was substituted with
paraoxon in the bench studies. This change was prompted by purity problems
associated with the p-nitrophenyl Soman synthesis needed for the 2-liter experiments.
High-purity paraoxon (99%) was purchased commercially for the bench-scale studies.
Systemization experiments with a small 50-ml loop filter system and
paraoxon in tap water showed several problems with the initial apparatus. First, native
tap water bacteria used the paraoxon and p-nitrophenol as nutritional sources, causing
growth (turbidity) in the treatment water, lowered p-nitrophenol levels, and biofilm
formation in the pump tubing. After sterilizing the system, problems were encountered
with paraoxon adsorption to the pump tubing. A study of paraoxon adsorption in
nonfilter loops showed that a combination of glass capillaries and Pharmed™ tubing
gave the least paraoxon adsorption. To further reduce the adsorption of paraoxon to the
tubing, the geometry of the system was changed so that the pump tubing encountered
the treatment water after it exited the filter, not before. If most of the paraoxon is
degraded in the filter to p-nitrophenol, then less paraoxon is available to adsorb to the
tubing after the treatment water exits the filter. A five-day study using this new system
geometry and apparatus gave excellent paraoxon hydrolysis over five days (99.1
percent) vs. the control (4 percent). Paraoxon loss from the filter system was negligible.
Bench-scale experiments with the catalytic filter loops were conducted
with paraoxon in 2 liters of ECBC tap water. The Amino-link Plus agarose coupling
method was used for both enzymes, because the production of the azlactone-
polyacrylamide was discontinued by the manufacturer. This situation caused a delay in
the OPH coupling (backorder followed by reordering different material), putting the
bench demonstration behind schedule by several weeks. The catalytic filter loop
systems used a 30-33-ml coupled enzyme or BSA filter with a 2-liter total tap water
volume system.
OPAA-agarose and OPH-agarose catalytic filter loop systems gave very
similar results in the bench study. Absorbance measurements revealed that both
catalytic systems hydrolyzed >99 percent of the paraoxon (99.4 percent for OPAA; 99.8
percent for OPH), vs. 5.9-6.3 percent hydrolysis for the BSA-agarose control systems.
pH values for the filter loop systems ranged from an initial average of 7.57 (enzyme)
and 7.60 (BSA) to a final average of 8.11 (enzyme) and 8.47 (BSA). Statistical T test
analysis confirmed that all but the Time 0 absorbance measurements for the enzyme
filter-treated water were significantly different from the BSA-filter treated water for both
the OPAA-agarose and OPH-agarose bench studies and could not have arisen by
random chance within 95 percent confidence limits.
23
-------
4. STABILIZED ENZYME DECONTAMINATION OF DRINKING WATER
4.1 Phase I preliminary studies
4.1.1 Stabilization
These studies examined the effect of PEGylating OPH and OPAA on
enzyme activity after drinking water storage. Initial studies used enzyme kinetic rate
analysis as the activity benchmark. This benchmark was examined at Time 0 and after
5 days' storage in ECBC tap water. Kinetic rate comparisons were made between the
different immobilization techniques to find the technique that resulted in the highest
activity after five days of tap water storage. Paraoxon was used as an OPH substrate
because p-nitrophenyl Soman hydrolytic activity was a very poor substrate for OPH.
Although OPAA has better catalytic activity with p-nitrophenyl Soman, this substrate
can't be purchased commercially (unlike paraoxon), and problems encountered with p-
nitrophenyl Soman synthesis by a local chemist precluded its use for the bench studies.
As such, paraoxon was used as the substrate for OPAA as well.
According to the literature survey, the best method for stabilizing an
enzyme was through the covalent attachment of polyethylene glycol (PEG) groups, also
known as PEGylation. PEG groups were attached to the proteins via primary and
secondary amines employing succinimide-activated PEGs, the oldest and best-tested
coupling chemistry, to yield a stable amide linkage. Activated PEGs consisted of either
a succinimdyl a-methypropionate or succinimdyl a-methybutanoate group attached to
the PEG polymer. The optimal size of the polyethylene glycol group for enzyme
stabilization varies from one enzyme to another and must be determined empirically.
Several different PEG sizes of 2, 5, 20, and 30 kDa, as well as a 40 kDa branched chain
polymer, were chosen for testing.
Phase 1 testing of modified OPH indicated that the 2-kDa PEG was
optimal for enzyme stability with 114 percent activity of Day-5 control (Figure 11). Due
to difficulties in obtaining p-nitrophenyl Soman, paraoxon was used as the substrate
for both enzymes. OPAA results were similar to OPH with the 2-kDa PEG retaining
99 percent of the day-5 control activity (Figure 12). Of the five polymer sizes tested, the
2-kDa PEG polymer yielded the best activity for both OPH and OPAA after five days in
tap water.
24
-------
140
120
Figure 11. Specific activity of PEGylated OPH at day 0 and Day 5 in tap water.
140
Figure 12. Specific activity of PEGylated OPAA at Day 0 and Day 5 in tap water.
PEG molecular weights indicated. Error bars are the +/- 95 percent confidence
levels.
25
-------
4.2
System ization
The stabilized enzymes were used to decontaminate paraoxon from tap
water. The benchmark for these studies was the amount of paraoxon hydrolyzed to
p-nitrophenol over a five-day treatment period. A small-scale (50-ml) reservoir was
employed for transition from the initial rate studies to the 2-liter, bench-scale studies.
The same modifications required for the investigation into the immobilized enzymes
(Section 3.2) were employed in this study.
4.3
Bench-Scale Testing
The bench system tested the feasibility of drinking water decontamination
with a stabilized enzyme. Based upon the Phase 1 studies, the 2-kDa PEG polymer was
chosen for the bench-scale experiments. The reactor set-up consisted of a sealed
2-liter vessel with a pH probe and thermometer passing through the lid. An insulated stir
plate was used to drive a stir bar in the bottom of the vessel (Figure 13). Samples were
taken via an access port in the top. During the experiment the vessel was protected
from light with aluminum foil due to possible photosensitivity of p-nitrophenol. (This was
done for all full-scale runs.) The stabilized enzyme circulated with 2 liters of 0.091-0.103
mM paraoxon (actual, measured by base hydrolysis) in ECBC tap water at 24°C. Both
OPH-2kDa and OPAA-2-kDa were used in this demonstration. BSA-2-kDa was run in
parallel with each enzyme as the nonenzymatic control under the same operating
conditions. Temperature, pH, and absorbance (A405nm) were monitored during the five-
day demonstration period according to the schedule.
Thermometer
Tap Water
+ Substrate
+ Stabilized
Enzyme/BSA
Reservoir
Insulator
Stir plate
In the first bench-scale experiment, OPH failed to hydrolyze any paraoxon
after three days. Stabilized OPH-2kDa equal to that used on Day 0 was added directly
to the reactor on Day 3. The enzyme behaved as was initially expected with greater
26
-------
than 90 percent of the paraoxon hydrolyzed in 24 hours (Figure 14). It was suspected
that the filtering process had removed or damaged the initial addition of stabilized
enzyme.
a.
o
Q.
O
OPH-2kDa
BSA-2kDa
10
20 30
Time (hours)
40
50
60
Figure 14. Paraoxon catalysis by OPH 2-kDa after three-day
interval. Error bars are the +/- 95 percent confidence levels.
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
v//
BTP
Fresh Tap Water
Filtered Tap Water
Phase 2 7/11/05 Reactor Water
012345
Time (minutes)
Figure 15. OPH-2kDa paraoxon hydrolysis measured at
405nm assayed in ECBC tap water as indicated (in the key)
27
-------
The experiment was performed again without filtering the stabilized
enzyme. There was no paraoxon hydrolyzed by OPH even after five days' incubation.
The addition of OPH-2kDa after Day 5, equivalent to that used in Day 0, again resulted
in complete hydrolysis of paraoxon in less than 72 hours.
A series of troubleshooting experiments was performed to clarify why the
enzyme initially failed to hydrolyze the substrate in the bench-scale study. As part of the
preparation process, the stabilized enzyme was dialyzed into cold-aged tap water (48-
72 hr) for final storage and testing (Phase 1). Dialyzing into tap water did not affect the
activity when assayed in BTP buffer. When assayed directly in fresh, filtered, or 1-day-
old unbuffered tap water, the enzyme exhibited a rapid loss in activity—greater than
98 percent in the first 30 seconds (Figure 15).
The enzyme was diluted 1:50 and incubated in fresh tap water, 1-day-old
tap water, and 1-month-old tap water (sterile filtered) for 15 minutes at room
temperature. After incubation, enzyme activity was assayed in 50mM BTP pH 8.5,
100mM paraoxon. All three samples retained 100 percent of their preincubation
paraoxon activity. The OPH bench study was repeated for a third and final time
following some recommendations made during a quality audit of the second study. In
addition, a much larger amount of stabilized enzyme was used for this study. The
amount of stabilized enzyme used should have hydrolyzed all the paraoxon in the
reactor in less than ten minutes based on the measured activity immediately prior to
initiation.
1!
N
O
8
ns
i_
ns
a.
2000
4000
6000
8000
OPH-2kDa
OPAA-2kDa
BSA-2kDa
BSA2-kDa
Figure 16. Paraoxon catalysis in catalytic and control 2-liter bench-scale
systems. Error bars are the +/- 95 percent confidence levels.
28
-------
10
8
I
Q.
Q_
tfi
9)
o 4:
T
J. _
2000
4000
min
6000
8000
|jm OPAA-2kDa
|jm pNP OPAA-BSA
jam BSA-2kDa
Figure 17. Total umolespNP produced by the OPH-2kDa, OPAA-2kDa, and
their BSA-2kDa controls in the 2-liter systems. The enzyme-BSA plots show the
net catalytic umoles produced. Error bars show the +/- 95 percent confidence
limits.
Day 0
Day 0
OPH-2kDa
' .
BSA-2kDa OPAA-2kDa
BSA-2kDa
DayS
DayS
OPH-2kDa BSA-2kDa OPAA-2kDa BSA-2kDa
Figure 18. Photograph of bench study on Day 0 and Day 5.
29
-------
Results of the catalytic paraoxon decontamination systems showed little
performance by either stabilized enzyme (Figure 16). After the five-day treatment, the
stabilized OPH hydrolyzed 19.4% ±0.2 percent of the paraoxon compared to 6.8
percent ±2.7 percent in the BSA control. OPAA failed to hydrolyze any paraoxon and
was not statistically different from the BSA control. The net catalytic paraoxon hydrolysis
from OPH-2kDa and OPAA-2kDa was 11.5 percent and 0 percent, respectively (Figure
17). p-Nitrophenol production from paraoxon was barely discernable in the 2-liter
catalytic system compared with the control filter system for OPH and undetectable in the
OPAA system (Figure 18).
Other parameters measured during the bench-scale studies included
temperature and pH. The temperature during the study was relatively stable (23°C
±0.5°C). The initial mean pH of the catalytic systems was 7.93 (enzyme) and 7.85
(BSA). After the five-day treatment, the final mean pH was 8.30 (enzyme) and 8.25
(BSA) (Figure 19).
7.6
0 2000
+ BSA-2kDa
A BSA-2kDa
4000 6000
min
u OPH-2kDa
9 OPAA-2kDa
8000
4.3.1
Statistical analysis of the 2-liter system
Triplicate data generated during the bench studies were subjected to the
T test (in Microsoft Excel) to determine whether the absorbance at 405 nm (A405) of the
enzyme systems was significantly different from those of the control systems. The T test
determines the significant differences between the catalytic and the control data based
upon the chance that random probability could produce the observed numbers within a
predetermined confidence limit. Using the paired two samples for means analysis (two-
tailed), the resulting parameters of t stat, t critical, and P values were examined for each
time point. Our criteria were that the t stat should be > the t critical value (two-tailed) and
30
-------
that the P value (two-tailed) should be < 0.05, using 95 percent confidence limits. The
results are shown in Tables 3 and 4.
TABLE 3
T test analysis of the OPH/BSA-2kDa bench-scale experimental A405 data
Mean
Variance
Observations
Pearson
Correlation
Hypothesized
Mean Difference
deg. freedom
tStat
P (T<=t) one-tail
t Critical one-tail
P (T<=t) two-tail
t Critical two-tail
Mean
Variance
Observations
Pearson
Correlation
Hypothesized
Mean Difference
deg. freedom
tStat
P (T<=t) one-tail
t Critical one-tail
P (T<=t) two-tail
t Critical two-tail
BSA
0
0.012
4.3E-05
3
-0.69686
0
2
0.27154
0.40572
4.30265
0.81144
6.20535
BSA
1800
0.03633
0.00019
3
-0.50683
0
2
8.90091
0.00619
4.30265
0.01239
6.20535
OPH
0
0.01067
6.3E-06
3
OPH
1800
0.115
7E-06
3
BSA
60
0.009
2.1E-05
3
0.8825
0
2
71.5
9.8E-05
4.30265
0.0002
6.20535
BSA
2880
0.02667
9.3E-06
3
-0.26906
0
2
21.5385
0.00107
4.30265
0.00215
6.20535
OPH
60
0.10433
1E-05
3
OPH
2880
0.12
3.7E-05
3
BSA
120
0.01267
2.1E-05
3
#DIV/0!
0
2
34.625
0.00042
4.30265
0.00083
6.20535
BSA
3240
0.04033
0.00019
3
-0.51923
0
2
9.26782
0.00572
4.30265
0.01144
6.20535
OPH
120
0.105
0
3
OPH
3240
0.12567
1.4E-05
3
BSA
180
0.008
1E-06
3
-0.37115
0
2
20.8405
0.00115
4.30265
0.00229
6.20535
BSA
4320
0.03867
2.5E-05
3
0.39736
0
2
32.3749
0.00048
4.30265
0.00095
6.20535
OPH
180
0.11033
6.5E-05
3
OPH
4320
0.127
1E-06
3
31
-------
Table 3 (continued)
Mean
Variance
Observations
Pearson
Correlation
Hypothesized
Mean Difference
deg. freedom
tStat
P (T<=t) one-tail
t Critical one-tail
P (T<=t) two-tail
t Critical two-tail
Mean
Variance
Observations
Pearson
Correlation
Hypothesized
Mean Difference
deg. freedom
tStat
P (T<=t) one-tail
t Critical one-tail
P (T<=t) two-tail
t Critical two-tail
BSA
240
0.01367
3E-05
3
-0.73974
0
2
15.9264
0.00196
4.30265
0.00392
6.20535
BSA
4680
0.03933
3E-05
3
0.60294
0
2
13.1595
0.00286
4.30265
0.00573
6.20535
OPH
240
0.10967
3.2E-05
3
OPH
4680
0.13633
0.00023
3
BSA
300
0.01133
1.4E-05
3
0.99124
0
2
20.5486
0.00118
4.30265
0.00236
6.20535
BSA
5760
0.15033
0.00024
3
0.98533
0
2
14.9148
0.00223
4.30265
0.00447
6.20535
OPH
300
0.11767
0.00016
3
OPH
5760
0.04267
9.3E-06
3
BSA
360
0.02533
0.00062
3
-0.42464
0
2
3.10053
0.04509
4.30265
0.09017
6.20535
BSA
6120
0.057
0.00011
3
-0.25337
0
2
11.1066
0.004
4.30265
0.00801
6.20535
OPH
360
0.13867
0.00236
3
OPH
6120
0.14167
3.4E-05
3
BSA
1440
0.026
0.00015
3
0.99874
0
2
58.2065
0.00015
4.30265
0.0003
6.20535
BSA
7200
0.06233
0.0001
3
0.8345
0
2
21.1247
0.00112
4.30265
0.00223
6.20535
OPH
1440
0.12867
0.00023
3
OPH
7200
0.16167
0.00021
3
(Table 3).
All results meet the significance criteria, except Time 0 and Hour 6
32
-------
TABLE 4
T test analysis of the OPAA/BSA-2kDa bench-scale experimental A405 data
Mean
Variance
Observations
Pearson
Correlation
Hypothesized
Mean Difference
deg. freedom
tStat
P (T<=t) one-tail
t Critical one-tail
P (T<=t) two-tail
t Critical two-tail
Mean
Variance
Observations
Pearson
Correlation
Hypothesized
Mean Difference
deg. freedom
tStat
P (T<=t) one-tail
t Critical one-tail
P (T<=t) two-tail
t Critical two-tail
BSA
0
0.01167
2.3E-06
3
0.94491
0
2
4
0.0286
4.30265
0.05719
6.20535
BSA
1800
0.022
1.8E-35
3
0
0
2
0.37796
0.3709
4.30265
0.7418
6.20535
OPAA
0
0.01033
1 .3E-06
3
OPAA
1800
0.02233
2.3E-06
3
BSA
60
0.01167
3.3E-07
3
-0.5
0
2
0.37796
0.3709
4.30265
0.7418
6.20535
BSA
2880
0.034
1E-06
3
-0.98198
0
2
1.14708
0.18503
4.30265
0.37006
6.20535
OPAA
60
0.01133
1.3E-06
3
OPAA
2880
0.03567
2.3E-06
3
BSA
120
0.01167
3.3E-07
3
0.86603
0
2
4
0.0286
4.30265
0.05719
6.20535
BSA
3240
0.04033
3.3E-07
3
0.86603
0
2
1.88982
0.09968
4.30265
0.19936
6.20535
OPAA
120
0.013
1E-06
3
OPAA
3240
0.042
4E-06
3
BSA
180
0.00933
3.3E-07
3
-0.5
0
2
1.73205
0.1127
4.30265
0.2254
6.20535
BSA
4320
0.04833
3.3E-07
3
9.8E-15
0
2
2
0.09175
4.30265
0.1835
6.20535
OPAA
180
0.01033
3.3E-07
3
OPAA
4320
0.049
7.2E-35
3
33
-------
Table 4 (continued)
Mean
Variance
Observations
Pearson
Correlation
Hypothesized
Mean Difference
deg. freedom
tStat
P (T<=t) one-tail
t Critical one-tail
P (T<=t) two-tail
t Critical two-tail
Mean
Variance
Observations
Pearson
Correlation
Hypothesized
Mean Difference
deg. freedom
tStat
P (T<=t) one-tail
t Critical one-tail
P (T<=t) two-tail
t Critical two-tail
BSA
240
0.00733
1.3E-06
3
0.5
0
2
0
0.5
4.30265
1
6.20535
BSA
4680
0.05167
8.3E-06
3
-0.5
0
2
1 .25724
0.16779
4.30265
0.33559
6.20535
OPAA
240
0.00733
3.3E-07
3
OPAA
4680
0.04933
3.3E-07
3
BSA
300
0.00867
3.3E-07
3
#DIV/0!
0
2
2
0.09175
4.30265
0.1835
6.20535
BSA
5760
0.06
0
3
#DIV/0!
0
2
1 .67726
0.11775
4.30265
0.23549
6.20535
OPAA
300
0.008
0
3
OPAA
5760
0.06533
3E-05
3
BSA
360
0.011
7E-06
3
-0.75593
0
2
0
0.5
4.30265
1
6.20535
BSA
6120
0.065
4E-06
3
-0.86603
0
2
0.22942
0.41994
4.30265
0.83987
6.20535
OPAA
360
0.011
9E-06
3
OPAA
6120
0.06533
3.3E-07
3
BSA
1440
0.023
7E-06
3
-0.65465
0
2
0.1644
0.44226
4.30265
0.88453
6.20535
BSA
7200
0.07467
1.3E-06
3
-0.5
0
2
1.51186
0.13485
4.30265
0.2697
6.20535
OPAA
1440
0.02267
1.3E-06
3
OPAA
7200
0.07333
3.3E-07
3
None of the time points measured for OPAA was statistically different from
the BSA control according to our criteria.
34
-------
4.4 Conclusions
Initial studies demonstrated that the paraoxon (and nerve agent)
hydrolyzing enzymes OPH and OPAA could be PEGylated with a variety of polymer
sizes (2-40 kilodaltons). The stabilized enzymes were tested for activity stability before
and after five days of storage in tap water. Preliminary studies identified the smallest
polyethylene polymer as the optimal size for activity stabilization of both OPH and
OPAA enzymes.
Bench-scale experiments with the catalytic reactors were conducted with
paraoxon in 2 liters of ECBC tap water. The 2kDa succinimide activated PEG polymer
was used to stabilize the enzymes OPH and OPAA. Stabilized OPH hydrolyzed a
statistically significant amount of paraoxon in the bench study, but the vast majority of
this hydrolysis (94.3 percent) occurred during the first hour. The statistical difference
appears to be the result of a brief initial activity between Time 0 and Hour 1. The
hydrolysis rate after the first hour does not appear to be statistically different from the
BSA control, indicating the stabilized enzyme was inactivated during the first hour of
incubation, likely within the first minute due to the limited amount of paraoxon
hydrolyzed relative to the amount of catalytic activity added.
Stabilized OPAA did not hydrolyze a statistically significant amount of
paraoxon in the bench study. OPAA is far slower at hydrolyzing paraoxon than OPH.
Therefore, all available stabilized OPAA was employed for the experiment. The amount
of enzyme added should have hydrolyzed all the paraoxon in the reactor in
approximately 3.2 days based on hydrolysis rates measured at pH 8.5, immediately
prior to initiation of the experiment. The stabilized OPAA was likely inactivated shortly
after addition to the reactor in a manner similar to that of stabilized OPH. Stabilized
OPAA (less than 1/20 of the amount used for the initial bench study) was added post-
experiment in order to determine whether the inactivating material had dissipated in a
manner similar to that of the OPH bench studies. After five additional days, 6.6 percent
of the remaining paraoxon had been hydrolyzed, compared to an additional 2.9 percent
for the BSA, indicating that the stabilized enzyme was indeed active.
Bench-scale troubleshooting experiments indicated that the enzyme was
not directly inactivated by simple exposure to tap water. However, exposure to tap water
in the presence of substrate apparently caused irreversible inactivation. The ability of
the tap water to inactivate enzyme was lost after three or more days, indicating the
transitory nature of the agent(s) responsible. Further research into the mechanism of
enzyme inactivation was outside of the scope of this work.
After the five-day treatment, the stabilized OPH hydrolyzed 19.4 percent
±0.2 percent of the paraoxon compared to 6.8% ±2.7 percent in the BSA control. OPAA
failed to hydrolyze any paraoxon in the bench study. pH values for the system ranged
from an initial average of 7.93 (enzyme) and 7.85 (BSA) to a final average of 8.30
(enzyme) and 8.25 (BSA).
35
-------
5. GENERAL CONCLUSIONS
Both preliminary and bench-scale studies resulted in the successful
immobilization of OPAA and OPH enzymes. The activity appeared to be stable for at
least five days and potentially much longer. As with many other techniques, the support
that gives the best activity and/or stability will be dependent on the particular enzyme
system being utilized. It appears that enzyme immobilization onto solid supports
(enzyme filters) for the decontamination of flowing tap water is a viable technology for
use in civilian or military water distribution systems. To actually put this technology into
use would require significantly different bioreactor systems. By using a much higher
concentration of enzymes in hollow-fiber systems, the speed of reaction should be
greatly increased, thus reducing the required residence time. In addition, this would
protect the enzymes from attack by microorganisms that may be present in the water,
increasing the active life of the unit. The fate of the end products of the enzyme
reactions was not examined in this effort. Numerous products are available that can
remove chemicals from water, but they do not detoxify the chemicals. Combining such
materials with enzymes could provide significant safety advantages. While research
quantities of enzymes are quite expensive, the large-scale industrial production of OPH
and OPAA by Genencor International is rapidly lowering the cost.
For enzyme stabilization, modification of the enzymes using polyethylene
glycol polymers (PEGylation) was successfully demonstrated in the preliminary studies.
However, when scaled up for the bench-scale studies, both of the modified OPAA and
OPH enzymes showed significant inhibition. Since all other conditions were the same as
in the preliminary studies except for the volume of the systems, the cause of this
inhibition is currently unknown. This will require additional research and development to
overcome this limitation. It is also unknown whether other methods of stabilization may
have the same effect.
36
-------
6. ABBREVIATIONS AND ACRONYMS
Cyclosarin GF, 0-Cyclohexyl methylphosphonofluoridate
LBG Locust bean gum
OPAA Organophosphorus acid anhydrolase
OPH Organophosphorus hydrolase
Paraoxon 0,0-Diisopropyl p-nitrophenylphosphate
PEG Polyethylene glycol
pNP para-Nitrophenol
Sarin GB, 0-lsopropyl methylphosphonofluoridate
Soman GD, 0-Pinacolyl methylphosphonofluoridate
Tabun GA, ethyl A/-dimethylphosphoramidocyanidate
THEOS Tetrakis (2-hydroxyethyl) orthosilicate
VX S-Diisopropylaminoethyl methylphosphonothiolate
37
-------
&EPA
United States
Environmental Protection
Agency
Office of Research and Development
National Homeland Security Research Center
Cincinnati, OH 45268
Official Business
Penalty for Private Use
$300
Recycled/Recyclable
Printed with vegetable-based ink on
paper that contains a minimum of
50% post-consumer fiber content
processed chlorine free
PRESORTED STANDARD
POSTAGES FEES PAID
EPA
PERMIT NO. G-35
------- |