Langmuir 2009, 25(16), 9615–9618Published on Web 07/16/2009
©2009 American Chemical Society
Orientation Specific Positioning of Organophosphorus Hydrolase on
Solid Interfaces for Biosensor Applications
Tony E. Reeves,§,‡Sheetal Paliwal,†,‡Melinda E. Wales,§James R. Wild,§and
Aleksandr L. Simonian*,†
and§Biochemistry and Biophysics Department, Texas A&M University, 103 Biochemistry Building, 2128
TAMU, College Station, Texas 77843
Received March 2, 2009. Revised Manuscript Received May 31, 2009
Protein immobilization on solid interfaces is a crucial aspect of their successful application in technologies such as
biosensing, purification, separation, decontamination, etc. Although immobilization can improve the long-term and
operational stability of proteins, this is often at the cost of significant losses in the catalytic activity of the tethered
enzyme. Covalent attachment methods take advantage of reactive groups on the amino acid side chains. The
distribution of the solvent exposed side chains on an enzyme’s molecular surface often results in an ensemble of
the attachment mechanism and resulting orientation, access to and from the active site could be restricted. This study
describes a methodology for the design and implementation of an orientation specific attachment of an enzyme to a
surface plasmon resonance sensor surface. The enzyme, organophosphorus hydrolase, was structurally analyzed to
identify surface resides as candidates for modification to optimize active site accessibility and, thus, sensitivity of
detection. A single surface lysine on the active site face of the enzyme dimer was selected for elimination, thus allowing
surface lysine-to-alanine variant retained 80% of the wild-type activity with the neurotoxin substrates, paraoxon and
demeton-S. After immobilization, surfaces bearing the variant were determined to be more active even though the
enzyme coverage on the sensor surface was reduced by 17%.
With the judicious choice of an enzyme with well-defined
kinetics, decontamination and biosensor applications can be
afforded both high specificity and high sensitivity. Organophos-
phorus hydrolase (OPH) is an enzyme capable of degrading a
wide array of organophosphate pesticides and nerve agents,1-7
and this ability makes it ideal for many detection and deconta-
decontamination and detection applications, the simplest being a
mixture. OPH has been successfully used to decontaminate sur-
faces when incorporated in fire fighting foams and latex paints.8,9
This strategy circumvents the problems of using caustic agents on
large areas. Slightly more complicated, but still requiring no
modification of the enzyme, is entrapment. Encapsulation of the
enzyme in murine erythrocytes by hypotonic dialysis has been
shown to be successful at degrading pesticides.10An optical
technique utilizing poly(ethylene glycol) hydrogel encapsulated
OPH was able to detect paraoxon down to 16 nM (0.004 ppm)
bilization of Escherichia coli
has been used to detect paraoxon down to 1?103nM (0.25 ppm)
using a potentiometer biosensor.12
Other slightly more sophisticated methods of immobilizing
poly(thiophene-3-acetic acid) layers allowed for the detection of
paraoxon down to 1.0 nM concentration.13Another approach
used an OPH-cellulose binding domain fusion protein, which
was reported to retain kinetic characteristics similar to the free
enzyme when immobilized.14The studies mentioned above were
electrostatic attractions, or physical containment to immobilize
an enzymatic fraction, whether whole cells or purified enzymes.
Although these approaches offer ease of construction, the use of
trapping or adsorption mechanisms can make the surface subject
to loss of activity through such processes as diffusion or aggrega-
tionoftheprotein.Thenecessityfor tight, essentiallyirreversible,
attachment makes covalent linkage an attractive alternative, and
(E. coli) cells expressing OPH
‡Both authors contributed equally to this work.
(1) Benning, M.M.; Kuo, J.M.; Raushel, F.M.; Holden, H.M. Biochemistry
1994, 33, 15001.
(2) Benning, M.M.; Hong, S.B.; Raushel, F.M.; Holden, H.M. J. Biol. Chem.
2000, 275, 30556.
(3) diSioudi, B.D.; Miller, C.E.; Lai, K.H.; Grimsley, J.K.; Wild, J.R. Chem.-
Biol. Interact. 1999, 120, 211.
(4) Chen-Goodspeed, M.; Sogorb, M.; Raushel, F.M. FASEB J. 1999, 13,
(5) Chen-Goodspeed, M.; Sogorb, M.A.; Wu, F.Y.; Hong, S.B.; Raushel, F.M.
Biochemistry 2001, 40, 1325.
(6) Hong, S.B.; Raushel, F.M. Protein Eng. 2004, 388, 256.
(7) Lai, K.H.; Dave, K.I.; Wild, J.R. J. Biol. Chem. 1994, 269, 16579.
(8) LeJeune, K.E.; Russell, A.J. Biotechnol. Bioeng. 1999, 62, 659.
(9) McDaniel, C.S.; McDaniel, J.; Wales, M.E.; Wild, J.R. Prog. Org. Coat.
2006, 55, 182.
(10) Pei, L.; Omburo, G.; McGuinn, W.D.; Petrikovics, I.; Dave, K.; Raushel,
F.M.; Wild, J.R.; Deloach, J.R.; Way, J.L. Toxicol. Appl. Pharmacol. 1994, 124,
(11) Russell, R.J.; Pishko, M.V.; Simonian, A.L.; Wild, J.R. Anal. Chem. 1999,
(12) Rainina, E.I.; Efremenco, E.N.; Varfolomeyev, S.D.; Simonian, A.L.;
Wild, J.R. Biosens. Bioelectron. 1996, 11, 991.
(13) Constantine, C.A.; Mello, S.V.; Dupont, A.; Cao, X.H.; Santos, D.;
Oliveira, O.N.; Strixino, F.T.; Pereira, E.C.; Cheng, T.C.; Defrank, J.J.; Leblanc,
R.M. J. Am. Chem. Soc. 2003, 125, 1805.
(14) Richins, R.D.; Mulchandani, A.; Chen, W. Biotechnol. Bioeng. 2000, 69,
DOI: 10.1021/la9007526Langmuir 2009, 25(16), 9615–9618
ArticleReeves et al.
one in which retention of biological activity, uniform structure,
and long-term stability can be achieved.
Covalent attachment relies on chemical modification of the
side chains exposed at the enzyme’s surface. Cystamine-glutar-
aldehyde immobilization has been used in an amperometric
detection system,15,16relying on the detection of pH change
resulting from the release of protons when OPH hydrolyses
substrates. OPH has been immobilized in photosensitive poly-
ethylene glycol (PEG) gels, which employs both covalent attach-
ment and physical entrapment.17The reversible inhibition of
OPH has been utilized for the development of an optical sensor
that detects substrate binding and not hydrolysis of target
compounds.18In this system, purified enzyme is immobilized on
a glass surface through glutaraldehyde activation of lysines.
Introduction of a substrate results in the competitive displace-
ment of the porphyrin bound in the active site, and the change in
the porphyrin absorption spectrum was observed.
tions, and all will have advantages and disadvantages that would
preclude the use of a single technique. A recent extensive review
details the use of OPH, as well as other enzymes, in biomaterials
for detection and decontamination of chemical warfare agents.19
To address the issue ofefficiency ofthe immobilized enzyme, this
study presents a method to further refine surface construction by
the orientation specific attachment of OPH. Such attachment
should facilitate active site access as well as reduce the potential
for restriction of movement of important secondary structural
elements near the active site.
2. Experimental Section
2.1. Variant Design. To aid in the selection of residues for
substitution, two methods were employed to calculate solvent
parameter optimized surfaces analysis (POPS) was used to calcu-
late an approximation of area of the first solvation shell for each
lysine residue.20Second, pfis calculations were performed to
determine the degree of burial, specifically of the ζ-N, of each
2.2. Site-Directed Mutagenesis. The pUC19 plasmid con-
taining the wild-type (WT) (pOP419) gene7was used as the
template, and the mutation was made with the QuickChange
site-directed mutagenesis kit (Stratagene, LaJolla, CA). Nonho-
mologous, nonoverlapping primers were synthesized by Inte-
grated DNA Technologies, Inc. (Coralville, IA) and used for
the mutagenesis reactions. The primers used to construct the
variant were 50-AAAGGGGGTCGCGCAGCCTGTGGTC-30,
50-GACCACAGGCTGCGCGACCCCCTTTCA-30. The mu-
tated plasmids were sequenced to ensure the fidelity of the PCR
reactions. The electrophoretic separation and analysis were per-
formed by the Gene Technology Laboratory of Texas A&M
University, the resulting sequence data was analyzed with Vector
NTI software (Stratagene, LaJolla, CA).
2.3. Enzyme Purification and Biotinylation. The enzymes
were purified, as previously described.22Purified enzymes were
concentrated to greater than 1 mg/mL for storage at 4 ?C in the
final column buffer (10 mM KPO4, 20 mM KCl, 50 μM CoCl2,
pH 8.3). Protein concentration was determined by using
an extinction coefficient of 58000 M-1cm-1when measuring
absorbance at 280 nm. Purity was verified by SDS-PAGE.
The enzymes WT and K175A (1 mg/mL) were incubated with
equal concentrations of biotin (Pierce, Rockford, IL) in 5%
DMSO, 10 mM KPO4, and pH 8.3 overnight on a shaker at
4 ?C. The unbound biotin was removed by dialysis, performed
(pKa>12) are not considered as candidates for biotinylation.
2.4. Enzyme Activity Assay. The substrates used in this
study were paraoxonand demeton-S(ChemService), and thefree
thiol reporter for the demeton-S assays was 2,2’ dithiodipyridine
(2,2’ TP). Michaelis constants (KM) and the catalytic rates (kcat)
for paraoxon and demeton-S were determined by performing
enzymatic assays with varying concentrations of substrate and
constant enzyme concentrations. The activity of the biotinylated
enzymes was measured in a similar fashion with paraoxon as the
Paraoxon hydrolysis was followed by measuring the appear-
in 20 mM CHES (pH 9.0) at 25 ?C, and initial velocities were
calculated and fit to the Michaelis-Menten equation, allowing
for substrate inhibition eq 1:
Demeton-S hydrolysis was followed by the appearance of the
2,2’ TPanion at 343 nm (ε = 7060 M-1cm-1) in tripart buffer at
pH 8.0,23and initial velocities were calculated and fit to the
Michaelis-Menten eq 2:
2.5. SPREETA Preparation and Sensing Layer Con-
struction. The sensor surface was cleaned with piranha solution
(3:1mixture of sulfuricacidand30% hydrogenperoxide;extreme
and reacts violently with organic matter) followed by rinsing and
establishing a baseline with phosphate-buffered saline (PBS),
neutra-avidin (1 mg/mL) was nonspecifically adsorbed on the
gold surface. Bovine serum albumin (BSA) (1 mg/mL) was used
biotinylated enzymes (WT, K175A -1 mg/mL).
2.6. Calculation of Surface Coverage. The amount of
enzyme covering the surface of the sensor was calculated using
the equations described by eq 3 and 4.24,25The thickness of the
adsorbed layer (ad-layer) is calculated using
where dais the thickness of the ad-layer, ldis the characteristic
decay length of an evanescent wave at 307 nm, neffis the effective
RI of the ad-layer (from SPR signal), nbis the RI of the buffer
(1.333), andnaisthe RI ofthe proteins(1.57).Surfacecoverageis
(15) Wang, J.; Krause, R.; Block, K.; Musameh, M.; Mulchandani, A.;
Schoning, M.J. Biosens. Bioelectron. 2003, 18, 255.
(16) Flounders, A.W.; Singh, A.K.; Volponi, J.V.; Carichner, S.C.; Wally, K.;
Simonian, A.S.; Wild, J.R.; Schoeniger, J.S. Biosens. Bioelectron. 1999, 14, 715.
(17) Andreopoulos, F.M.; Roberts, M.J.; Bentley, M.D.; Harris, J.M.;
Beckman, E.J.; Russell, A.J. Biotechnol. Bioeng. 1999, 65, 579.
(18) White, B.J.; Harmon, H.J. Biosens. Bioelectron. 2005, 20, 1977.
(19) Russell, A.J.; Berberich, J.A.; Drevon, G.E.; Koepsel, R.R. Annu. Rev.
Biomed. Eng. 2003, 5, 1.
(20) Fraternali, F.; Cavallo, L. Nucleic Acids Res. 2002, 30, 2950.
(21) Hebert, E.J.; Giletto, A.; Sevcik, J.; Urbanikova, L.; Wilson, K.S.; Dauter,
Z.; Pace, C.N. Biochemistry 1998, 37, 16192.
(23) diSioudi, B.; Grimsley, J.K.; Lai, K.H.; Wild, J.R. Biochemistry 1999, 38,
(24) Jung, L.S.; Campbell, C.T.; Chinowsky, T.M.; Mar, M.N.; Yee, S.S.
Langmuir 1998, 14, 5636.
D.; Elkind, J.; Melendez, J.; Furlong, C.E. Biosens. Bioelectron. 2002, 17, 573.
Langmuir 2009, 25(16), 9615–9618
Reeves et al.Article
calculated using the thickness and density of the protein:
surface coverage ðg=mm2Þ ¼ thickness ðdaÞ
? density ð∼1:3g=cm3Þsurface coverageðmolecules=mm2Þ
surface coverage ðg=mm2Þ ? avagadro number
2.7. Immobilized Enzyme Activity. Paraoxon, 0.048-
0.462 mM, was circulated across the surface using a flow rate of
100 μL/min for 2 min. Activity was determined by collecting
200 μL of the flow through and measuring the absorbance at
405 nm of the p-nitrophenol product.
3.1. Structural Analysis. InOPH,thereare eight lysines per
site where it serves as a bridging ligand for the binuclear metal
exposed. The remaining lysines are accessible and are the more
probableattachment sites.Parameteroptimized surfacesanalysis
(POPS) of the 1DPM PDB structure (Table 1) shows that K175
has 209.6 square angstroms of solvent accessible surface area
(SASA).20The percent burial of the ζ-N of K175 and K294 was
completely exposed side chain at those two positions.21The ζ-N
of K339 and K77 were calculated to be less than 50% buried.
Taken together, these data suggest that K175 and K294 are the
most accessible, although not necessarily the only, biotinylation
3.2. Enzyme Activity in Solution. The K175A variant
is kinetically similar to WT in solution assays, having a kcatof
107compared to 1 ? 108for the WT. It is more susceptible to
substrate inhibition, with a KIof 6.1 mM versus 17 mM for
the WT. With demeton-S, K175A has a kcatof 2 s-1and KMof
3.1 mM, giving a catalytic efficiency of 558 compared to 870
for the WT, which has a kcatof 35 s-1and a KMof 0.04 mM.
3.3. Surface Construction and Immobilized Activity.
From the real time sensorgram of the surface construction
(Figure 1), the surface coverage of the enzymes was calculated.
K175Acoversthesensorsurfaceat1.8? 1010(9.2? 108mm-2,
2.16 ? 1010( 7.3 ? 108mm-2. This suggests that surface
Table 1. Solvent Accessible Surface Area of OPH Lysine Residues in
Lys residuehydrophobic hydrophilictotal
Figure 1. SPR sensorgram of surface assembly.
Table 2. Kinetic Constants for the Immobilized and Bulk Biotinylated Enzymes
kinetic parametersWTKA WT KA
0.00005 ( 1.5 ? 10-6
0.0434 ( 0.005
4.8 ? 106
0.00003 ( 1.4 ? 10-6
0.0624 ( 0.008
1.7 ? 106
0.00044 ( 1.6 ? 10-4
0.1854 ( 0.094
0.00055 ( 1.6 ? 10-4
0.2278 ( 0.089
Figure 3. Activity of biotinylated WT and K175A on surface.
Figure 2. Michaelis-Menten plot for biotinylated WT and
K175A in solution.
DOI: 10.1021/la9007526Langmuir 2009, 25(16), 9615–9618
ArticleReeves et al.
attachment of the K175A variant resulted in a 17% reduction in
activity per mm2of surface is virtually identical between the
enzymes, with immobilized K175A exhibiting a rate of 5.22 ?
10-15( 2.8 ? 10-16μmol/s/mm2with the WT measured rate at
4.30? 10-15(1.3? 10-16μmol/s/mm2(paraoxon= 0.05mM).
an overall reduction of biotinylation of the enzyme population,
providing an improved catalytic environment for those enzymes
which were attached.
3.4. Comparison of Enzyme Activity on Surface and in
Solution. The activity of the biotinylated WT and K175A was
measured both in solution and on the surface. The kinetic
constants were determined using a Michaelis-Menten plot of
K175A (Figure 2), whereas comparison of the surface attached
enzymes determined the activity of K175A to be 18% higher
than that of the WT enzyme. A calibration graph (Figure 3) for
the activity of the immobilized enzymes against the paraoxon
concentrationwasobtained. A linearmodel wasfit totheK175A
10-16( 6.9 ? 10-14x, R = 0.9996) data. This supports the
hypothesis that surfaces with enhanced catalytic capacity can be
created through the orientation-specified attachment of enzyme,
of the K175A variant.
Biotin forms an amide bond with the ζ-N of solvent exposed
lysine side chains. Ordering the lysine residues, based on solvent
accessible surface areas and the degree of exposure of the ζ-N,
allows anevaluationofthelikelihoodanygiven sidechainwill be
biotinylated. The more solvent accessible surface area (SASA)
and the more exposed the ζ-N, the higher the probability of
biotinylation. Using this approach, there are four lysines with an
SASAofgreater than 50% for the ζ-N. Although two, K165 and
K294, were assessed as fully solvent exposed, it is likely that all
four are candidates for biotinylation and can serve as a site for
The mostexposed surface lysine in OPH is K175, which is also
OPH through this side chain would result in its attachment in an
essentially “face down” manner with the active site oriented
toward the sensor surface, potentially resulting in reduced cata-
lytic rates by occluding the active site. Taken together, the data
presented in this study: (1) the kcatof the biotinylated K175A
enzyme, in solution, is reduced 48% relative to that of the
biotinylated WT; (2) the attachment of K175A to the surface is
surfaces are slightly higher than WT surfaces, support a critical
of the substrate interaction with active site structures indicates
that even small changes in the active site structure and/or
accessibility of the substrate to the active site may dramatically
change the catalytic process.
The activity of K175A is 18% greater than that of the WT
enzyme after immobilization on a surface, when the number of
site toward the bulk solvent. Alternatively, elimination of surface
attachment through residue 175 could allow for a more efficient
enzyme by permitting flexibility in secondary structural elements
near the active site.
In conclusion, the orientation-specific attachment of an en-
zyme, OPH, was studied for biosensor applications. Substitution
ofalysineresidue nearthe active siteofthe enzyme resultedinan
increase in the catalytic efficiency of the enzyme when surface
attached, as determined by SPR. By selectively removing attach-
ment sites, in this case a lysine from the protein surface, the
orientation of the protein molecules on the surface can be
Acknowledgment. This research was supported by NSF 280
under the Grant CTS-0330189, by the NIH CounterACT
Program through the NINDS (award #U01 NS058035-03)
and by the USDA-CSREES under grant 2006-34394-16953.
Additionally, this material was based on work which sup-
ported ALS by the National Science Foundation, while work-
ingat theFoundation. Itscontents aresolelytheresponsibility
of the authors and do not necessarily represent the official
views of the NSF, NIH, or the federal government. We would
like to thank Drs. Martin Scholtz and Nick Pace, Texas
A&M University, for assistance with the pfis calculations
and Shankar Balasubramanian for his help in the SPR experi-
ments and calculations.
Figure 4. (A)SolventexposedlysinesidechainsofOPH.K175isshowningreen,theothersareinred.(B)Theactivesiteisshowninyellowto
show the proximity to K175.