Cell Surface Display of FunctionalMacromolecule Fusions onEscherichia coli for Developmentof an Autofluorescent Whole-CellBiocatalystC H A O Y A N G , † , ‡ Q I A O Z H A O , |

Z H E N G L I U , † , ‡ Q I Y U N L I , §

C H U A N L I N G Q I A O , * , †

A S H O K M U L C H A N D A N I , * , ⊥ A N DW I L F R E D C H E N ⊥

State Key Laboratory of Integrated Management of Pest Insectsand Rodents, Institute of Zoology, Chinese Academy ofSciences, Beijing 100101, China, Graduate School of theChinese Academy of Sciences, Beijing 100049, China, Instituteof Plant Protection, Jilin Academy of Agricultural Sciences,Gongzhuling 136100, China, Plant Biotechnology Center andDepartment of Plant Cellular and Molecular Biology, OhioState University, Columbus, Ohio 43210, Department ofChemical and Environmental Engineering, University ofCalifornia, Riverside, California 92521

Received February 13, 2008. Revised manuscript receivedJune 1, 2008. Accepted June 4, 2008.

At present, Lpp-OmpA-mediated surface display has openeda new dimension in the development of whole-cell factories. Here,we report the surface display of methyl parathion hydrolase(MPH) and enhanced green fluorescent protein (EGFP) fusions(60 kDa) by employing the Lpp-OmpA chimera as an anchoringmotif. A broad-host-range vector, pLOMG33, coding forLpp-OmpA-MPH-GFP fusion protein was constructed fortargeting the fusion protein onto the surface of Escherichiacoli. The surface localization of fusion protein was demonstratedby Western blot analysis, immunofluorescence microscopy,and a protease accessibility experiment. The surface-exposedfusion protein retains the MPH activity and GFP fluorescence.Anchorage of macromolecule fusions on the outer membraneneither inhibits cell growth nor affects cell viability, as shown bygrowth kinetics of cells and stability of resting cultures. Theengineered E. coli with surface-expressed MPH-GFP has twomajor advantages over the same strain expressing cytosolicMPH-GFP, including 7-fold higher whole-cell activity and 2-foldstronger fluorescence. Moreover, the construct pLOMG33 canpotentially be applied to various bacterial species for enhancingfield use. This is the first report on the presentation of GFPfusions on the cell surface by Lpp-OmpA. Our results suggestthat Lpp-OmpA is a useful tool for the functional display ofmacromolecule passenger proteins on the cell surface.

Introduction

Synthetic organophosphates (OPs) are a group of highly toxicchemicals and exhibit broad-spectrum activity against majoragricultural pests, accounting for ∼38% of total pesticidesused globally (1). These compounds are potent acetylcho-linesterase (AChE) inhibitors. Since AChE is present in allvertebrates, the potential damage caused by OPs to nontargetorganisms is extremely high. Microbial degradation of OPshas received considerable attention, and the bacterial or-ganophosphorus hydrolase (OPH) has been intensivelyresearched (2, 3).

Most microorganisms that produce OPH are Gram-negative bacteria, and their OPH is located within the cells(2, 3). Gram-negative bacteria possess a complex cell envelopestructure that consists of a cytoplasmic membrane, cell wall,and outer membrane. The outer membrane prevents OPsfrom interacting with OPH residing within the cell, reducingthe overall catalytic efficiency (4). However, this permeabilitybarrier could be eliminated by the use of a microorganismdisplaying OPH on the surface (5, 6). The surface-exposedenzymes have free access to OPs, overcoming the rate-limitingstep in the degradation of OPs by natural isolates expressingOPH intracellularly.

Various surface-anchoring motifs that possess the po-tential to cross both the cytoplasmic and outer membraneshave been employed for targeting heterologous proteins ontothe cell surface, such as the lipoprotein-outer membraneprotein A chimera (Lpp-OmpA), ice nucleation protein (INP),and autotransporter (7, 8). The protein to be displayed(passenger protein) can be fused to an anchoring motif(carrier protein) by N-terminal fusion, C-terminal fusion, orsandwich fusion. These characteristics of carrier proteins,passenger proteins, and fusion strategy affect the efficiencyof the surface display of passenger proteins.

The Lpp-OmpA chimera consists of the signal sequenceand the first nine N-terminal amino acids of the major E. colilipoprotein (Lpp) joined to a transmembrane domain (aminoacids 46-159) from outer membrane protein A (OmpA) (9).The Lpp-OmpA-based cell display system was the firstsuccessful approach for displaying full-length heterologousproteins on the surface of Escherichia coli (10) and has beenextensively used for the display of heterologous proteins,such as �-lactamase (10), cellulases (11), the scFv antibody(12), the cellulose-binding domain (13), cyclodextrin glu-canotransferase (14), and the chitin-binding domain (15),on the surface of E. coli.

The green fluorescent protein (GFP) from the jellyfishAequorea victoria is an ideal marker for monitoring thegenetically engineered microorganisms (GEMs) (16, 17).Fluorescence from GFP does not require additional geneproducts, substrates, or other factors and can be detectednoninvasively using fluorescence microscopy and flowcytometry (18). Moreover, GFP can be expressed as either anN- or a C-terminal fusion and still fluoresce (19). In fieldstudies, GFP has been used as a marker to assess the fate andactivity of specific degrading microorganisms (20). EnhancedGFP (EGFP; GenBank accession no. U57609), which is a red-shifted variant of GFP, assembles the chromophore morerapidly, shows much stronger fluorescence than wild-typeGFP, and fluoresces after exposure to daylight (21).

Recently, an organophosphate degradation gene (calledmpd) was isolated from methyl parathion-degrading Plesi-omonas sp., and its protein product, methyl parathionhydrolase (MPH), showed no homology to OPH (22). Morerecently, we reported the identification of the mpd gene

* Authors to whom correspondence should be addressed. Phone:86-10-64807191 (C.Q.); 951-827-6419 (A.M.).Fax: 86-10-64807099(C.Q.); 951-827-5696 (A.M.). E-mail: qiaocl@ioz.ac.cn (C.Q.); adani@engr.ucr.edu (A.M.).

† Chinese Academy of Sciences.‡ Graduate School of the Chinese Academy of Sciences.§ Jilin Academy of Agricultural Sciences.| Ohio State University.⊥ University of California.

Environ. Sci. Technol. 2008, 42, 6105–6110

10.1021/es800441t CCC: $40.75 2008 American Chemical Society VOL. 42, NO. 16, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 6105

Published on Web 07/15/2008

(GenBank accession no. DQ677027) from chlorpyrifos-degrading Stenotrophomonas sp (23). In this study, we usedthe Lpp-OmpA chimera for the functional display of theMPH-GFP fusion protein on the surface of E. coli. Theengineered E. coli can be applied in the form of a whole-cellbiocatalyst by overcoming the substrate uptake limitationand can be easily monitored by fluorescence for its fate inthe environment.

Materials and MethodsBacterial Strains and Plasmids. All strains, plasmids(4, 21, 23, 24), and primers used in this study are listed inTable 1. E. coli strains bearing plasmids were grown inLuria-Bertani (LB) media (25) supplemented with 50 µg/mLkanamycin or 100 µg/mL ampicillin.

Plasmid Construction. The lpp-ompA fusion gene waspolymerase chain reaction (PCR)-amplified from plasmidpOP131 using primers P1 and P2. The PCR product wasdigested with EcoRI and BamHI and then ligated into similarlydigested pUC19 to generate pLO. The mpd gene was PCR-amplified from plasmid pMDQ using primers P3 and P4. ThePCR product was digested with BamHI and PstI and thenligated into similarly digested pLO to generate pLOM. Thegfp gene was PCR-amplified form plasmid pEGFP-N3 usingprimers P5 and P6. The PCR product was digested with PstIand HindIII and then ligated into similarly digested pLOMto generate pLOMG. The lpp-ompA-mpd-gfp fusion gene wasreleased from pLOMG with EcoRI-HindIII and subclonedinto an E. coli/Pseudomonas shuttle vector, pVLT33, to createpLOMG33. Transformation of the plasmid into E. coli wasdone using the CaCl2 method (25). Expression of the fusionprotein was induced with 0.2 mM isopropyl-�-D-thiogalac-topyranoside (IPTG) for 24 h at 25 °C when cells were grownto an OD600 of 0.5.

To construct a control plasmid for expressing theMPH-GFP fusion protein in the cytoplasm, the mpd-gfpfusion gene was PCR-amplified from plasmid pLOMG usingprimers P6 and P7. The PCR product was digested with EcoRIand HindIII and then ligated into similarly digested pVLT33to generate pMG33.

Cell Fractionation. Cells harboring pLOMG33 wereharvested and resuspended in a 25 mM Tris-HCl buffer (pH8.0). The cells were disrupted by sonication on ice. The crudeextract was centrifuged for 10 min at 10 000 rpm to removecell debris. The cell-free extract was then centrifuged for 1 hat 50 000 rpm to separate the membrane and solublefractions. The supernatant representing the soluble fractionwas retained. For further outer-membrane fractionation, thepellet (total membrane fraction) was resuspended withphosphate-buffered saline (PBS) containing 0.01 mM MgCl2

and 2% Triton X-100 for solubilizing the inner membraneand was incubated for 30 min at room temperature, andthen the outer-membrane fraction was repelleted by ultra-centrifugation (26).

Western Blot Analysis. Samples of whole-cell lysate, thesoluble fraction, and the outer-membrane fraction wereanalyzed on SDS-PAGE with 10% (w/v) acrylamide (25). Afterelectrophoresis, the separated proteins were electroblottedovernight at 40 V to the nitrocellulose membrane (Millipore,Billerica, MA) with a tank transfer system (Bio-Rad, Hercules,CA) containing a transfer buffer (25 mM Tris, 192 mM glycine,10% methanol). After the blocking step, the membrane wasincubated with either rabbit anti-GFP polyclonal antibodies(Molecular Probes, Eugene, OR) or anti-MPH serum at a1:1000 dilution in TBST buffer (20 mM Tris-HCl, pH 7.5, 150mM NaCl, 0.05% Tween-20). Subsequently, the membranewas incubated with alkaline phosphatase-conjugated goatanti-rabbit IgG antibodies (Promega, Madison, WI) at a 1:2000dilution. The membrane was then stained with NBT/BCIPin an alkaline phosphatase buffer (100 mM Tris-HCl, pH 9.0,100 mM NaCl) for visualizing antigen-antibody conjugates.

Immunofluorescence Microscopy. Cells harboring pLOM-G33 and pMG33 were harvested and resuspended (OD600 )0.5) in phosphate-buffered saline (PBS) with 3% bovine serumalbumin. Cells were then incubated with either GFP or MPHantisera at a 1:1000 dilution for 2 h at 30 °C. After washingwith PBS, the cells were resuspended in PBS with goat anti-rabbit IgG conjugated with rhodamine (Invitrogen, Mukilteo,WA) at a 1:500 dilution and were incubated for 1 h at 30 °C.Prior to microscopic observation, cells were washed five times

TABLE 1. Strains, Plasmids, and Primers Used in This Study

strain, plasmid or primer description source or literature

E. coli Strains

XL1-Blue recA1 endA1 gyrA96 thi-1 hsdR17(rK- mK

+) supE44 relA1 lac(F’proAB lacIq∆M15 Tn10 [Tetr])

Stratagene

DH5R supE44 ∆lacU169(�80 lacZ∆M15) recA1 endA1 hsdR17(rK-

mK+) thi-1 gyrA relA1 F- ∆(lacZYA-argF)

Tiangen

PlasmidspOP131 gene source of Lpp-OmpA fusion 4pMDQ source of mpd gene 23pEGFP-N3 source of gfp gene 21pUC19 cloning vector for construction of fusion genes TaKaRapVLT33 E. coli/Pseudomonas shuttle vector, oriT, RSF1010, oriV, lacIq,

tac promoter, Kmr24

pLOMG33 pVLT33 derivative, surface expression vector containing alpp-ompA-mpd-gfp fusion gene

this study

pMG33 pVLT33 derivative, control plasmid for expressing MPH-GFPfusion protein in the cytoplasm

this study

Primers (5′f3′)P1a GAATTCAGGAAACAATGAAAGCTACTAAACTGGTA this studyP2 GGATCCGTTGTCCGGACGAGTGCCGAT this studyP3 GGATCCATGGCCGCACCGCAGGTG this studyP4 CTGCAGCTTGGGGTTGACGACCG this studyP5 CTGCAGATGGTGAGCAAGGGC this studyP6 AAGCTTACTTGTACAGCTCGTCCA this studyP7 GAATTCAGGAAACAATGGCCGCACCGCAGGTG this study

a The restriction sites are underlined.

6106 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 16, 2008

with PBS and mounted on poly-L-lysine-coated microscopicslides. Photographs were taken using a fluorescence mi-croscopy (Nikon) equipped with FITC and Rhodamine filters.

Assay for Whole-Cell MPH Activity. Cells harboringpLOMG33 and pMG33 were suspended in a 100 mMphosphate buffer (pH 7.4) and diluted to an OD600 of 1.0.MPH activity assay mixtures (1 mL, 3% methanol) contained50 µg/mL methyl parathion (added from a 10 mg/mLmethanol stock solution), 960 µL of a 100 mM phosphatebuffer (pH 7.4), and 10 µL of cells. Changes in absorbance(405 nm) were measured for 3 min at 30 °C using a BeckmanDU800 spectrophotometer. Activities were expressed as units(1 µmol of p-nitrophenol formed per minute) per OD600 wholecells (ε405 ) 17 700 M-1 cm-1 for p-nitrophenol).

Measurement of Whole-Cell Fluorescence. Cells harbor-ing pLOMG33 and pMG33 were suspended in a PBS buffer(pH 7.5) and diluted to an OD600 of 1.0, and the similarlydiluted cells harboring pVLT33 were used as backgroundreferences. The GFP fluorescence intensity was determinedusing a fluorescence spectrophotometer (F-4500, HITACHI,Japan) with a bandwidth of 5 nm, an excitation wavelengthof 488 nm, and an emission wavelength of 510 nm.

Protease Accessibility Assay. Cells harboring pLOMG33and pMG33 were harvested, suspended in a PBS buffer, andadjusted to an OD600 of 10. Pronase (4 units/mg; Sigma, St.Louis, MO) was added to a final concentration of 2 mg/mL.Cell suspensions were incubated at 37 °C for 3 h. Subse-quently, Pronase-treated and untreated cells were assayedfor GFP fluorescence and MPH activity as described above.

Stability Study of Resting Cultures. Cells harboringpLOMG33 were grown in 50 mL of LB medium supplementedwith 0.2 mM IPTG and 50 µg/mL kanamycin for 2 days,washed twice with 50 mL of a 150 mM NaCl solution,resuspended in 5 mL of a 100 mM phosphate buffer (pH 7.4),and incubated in a shaker at 25 °C. Over a 2-week duration,0.1 mL of samples were removed each day. Samples werecentrifuged and resuspended in 0.1 mL of a 100 mMphosphate buffer (pH 7.4). MPH activity assays were con-ducted as described above.

ResultsConfirmation of Surface Localization and Functionality ofMPH-GFP Fusion Protein. To target MPH-GFP onto the

surface of E. coli, the Lpp-OmpA chimera comprised of alocalization domain (Lpp) and a transmembrane domain(OmpA) was employed as an anchoring motif. A schematicdiagram of the construction of the Lpp-OmpA-MPH-GFPfusion protein is shown in Figure 1. The fusion gene wassubcloned into a broad-host-range vector, pVLT33, togenerate pLOMG33. Expression of the fusion protein wasunder the control of a tightly regulated tac promoter.

Western blot was performed to verify the synthesis of thefusion protein with either GFP or MPH antisera. A specificband corresponding to the 76 kDa fusion protein was detectedin whole-cell lysates from the cells carrying pLOMG33, whichmatches well with the molecular mass estimated from thededuced amino acid sequence of the fusion protein (Figure2A and B, lanes 3 and 2). However, no signal was detectedwith the control cells carrying pVLT33. When the fractionatedfractions of cells harboring pLOMG33 were probed with eitherGFP or MPH antisera, the 76 kDa band was detected in theouter-membrane fraction (Figure 2A and B, lanes 5 and 4).

FIGURE 1. Schematic diagram of the construction of the Lpp-OmpA-MPH-GFP fusion protein. Lpp SP, lipoprotein signal peptide; first 9aa, the first nine amino acids; OmpA (aa 46-159), amino acids 46-159 from outer-membrane protein A. OmpA forms fivemembrane-spanning � strands, making up �-barrel structures, which allow the exposure of C-terminally fused MPH-GFP on the cellsurface.

FIGURE 2. Western blot analysis for subcellular localization ofexpressed Lpp-OmpA-MPH-GFP fusion protein in E. coliXL1-Blue harboring pLOMG33. (A) Western blot analysis ofdifferent cellular fractions with anti-GFP antibody. Lane 1, proteinmarkers; lane 2, negative control (XL1-Blue harboring pVLT33); lane3, whole-cell lysates; lane 4, soluble fraction; lane 5,outer-membrane fraction. (B) Western blot analysis of differentcellular fractions with anti-MPH serum. Lane 1, protein markers;lane 2, whole-cell lysates; lane 3, soluble fraction; lane 4,outer-membrane fraction; lane 5, negative control (XL1-Blueharboring pVLT33).

VOL. 42, NO. 16, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 6107

Protease accessibility experiments were performed to as-certain the surface localization of the fusion protein. GFP isresistant to many common proteases except Pronase, whichis a mixture of broad-specificity proteases (27). Since Pronasecannot penetrate the outer membrane, only those GFP andMPH molecules that are anchored on the outer membranecan be degraded by Pronase. With the Pronase treatment,the fluorescence intensity of cells carrying pLOMG33 de-creased 87%, while cells carrying pMG33 had only a 6% dropin fluorescence intensity. Similarly, the MPH activity for cells(pLOMG33) decreased 81%, while cells expressing MPHintracellularly (pMG33) had only a 4% drop in activity. Afterthe treatment of cells (pLOMG33) with Pronase, the frac-tionated outer membrane was probed with either GFP orMPH antisera; however, no signal corresponding to the fusionprotein was detected.

Immunolabeling with specific antibodies or antisera is auseful tool to detect surface-exposed proteins. The surfacelocalization of the MPH-GFP fusion protein in E. coli wasdetermined by immunofluorescence microscopy. Cells wereprobed with a primary anti-GFP antibody and then fluo-rescently stained with a rhodamine-labeled IgG antibody.Since the anti-GFP antibody cannot diffuse through the cellmembrane, only those surface-exposed GFP molecules caninteract with the anti-GFP antibody. Under fluorescencemicroscopy, an orange fluorescence was observed on thecells (pLOMG33) with surface-exposed GFP (SupportingInformation, Figure S1). In contrast, the control cells (pMG33)expressing cytosolic GFP were not immunostained. From allof these results, we concluded that the MPH-GFP fusionprotein was indeed displayed functionally on the cell surfaceusing the Lpp-OmpA display system.

Whole-Cell Activity and Fluorescence. The whole-cellactivity of E. coli XL1-Blue displaying MPH was 7-fold higherthan that of the same strain expressing cytosolic MPH.Moreover, XL1-Blue displaying GFP exhibited 2-fold strongerfluorescence than the same strain expressing cytosolic GFP(Supporting Information, Figure S2). The activity was notdetected, and fluorescence remained at the original back-ground level prior to induction. The activity and fluorescenceincreased gradually after induction with 0.2 mM IPTG andreached a maximum at 24 h (Supporting Information, FigureS3). As shown in Figure 3A, the expressed fusion protein waslocated in the outer-membrane fraction when IPTG inductionwas done at a concentration of 0.2 mM. In contrast, the fusionprotein produced with induction at higher IPTG concentra-tions (0.5 and 1 mM) was only present in the soluble fraction.Accordingly, whole-cell activity reached a maximum at anIPTG concentration of 0.2 mM (Figure 3B). Further inductionresulted in a gradual decline in the activity. Contrarily, whole-cell fluorescence increased with increasing concentrationsof IPTG (Figure 3C). Induction at 25 °C resulted in a highwhole-cell activity and a significant fluorescence. The activityand fluorescence were not detected with induction at 37 °C,and the cultures with induction at 30 °C exhibited a lowactivity (0.02 U/OD600) and a weak fluorescence (20% ofthat achieved at 25 °C). The fluorescence of the surface-expressed GFP at pH 6 dropped to 30% of that at pH 7.5 andwas almost entirely quenched at a pH below 5. In contrast,the fluorescence of the cytosol-expressed GFP at pH 6maintained 80% of that at pH 7.5.

Stability of Cultures Displaying Fusion Protein. To testwhether surface expression of the fusion protein inhibits cellgrowth, growth kinetics of cells carrying pLOMG33 andpVLT33 were compared. No growth inhibition was observedfor cells expressing the fusion protein. Both cultures reachedthe same final cell density after 48 h of incubation (SupportingInformation, Figure S4). To monitor the stability of suspendedcultures, whole-cell activity was determined periodically overa 2-week period. As shown in Figure 4, whole-cell activity of

XL1-Blue (pLOMG33) remained at essentially the originallevel over the 2-week period.

DiscussionSurface display of the active proteins on living cells hasenormous potential in the synthesis of a wide variety ofvaluable products and in the degradation of numerous toxiccompounds (7, 8). Up to now, however, cellular surfacedisplay was merely restricted to a few proteins due to the

FIGURE 3. (A) Localization of the expressed Lpp-OmpA-MPH-GFP fusion protein in E. coli XL1-Blue at increasingconcentrations of IPTG. The fusion proteins in the soluble fraction(S) and outer-membrane fraction (OM) were probed with anti-GFPantibody. Whole-cell activity (B) and fluorescence (C) of XL1-Blueharboring pLOMG33 are shown under different levels of induction.Data are mean values ( standard deviations from three replicates.

FIGURE 4. Whole-cell MPH activity in suspended E. coli culturesexpressing the MPH-GFP fusion protein on the cell surface. Dataare mean values ( standard deviations from three replicates.

6108 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 16, 2008

limited availability of optimized anchoring motifs to ef-ficiently display proteins having different molecular weightsand molecular characteristics. An anchoring motif, which issuccessfully applied to display some target proteins, oftenfails to display other proteins. In the previous work, we usedan INP-based display system to functionally express MPHon the cell surface (28). The mechanism of INP anchoringto display proteins is still unknown. To date, only a fewanchoring motifs that include Lpp-OmpA from E. coli andINP from Pseudomonas syringae were shown to be capableof displaying GFP on the cell surface (26, 29). This difficultyin targeting GFP onto the cell surface may be attributed tothe distinguishing three-dimensional structure of GFP. It hasbeen reported that GFP is an 11-stranded � barrel threadedby an R helix running up the axis of the cylinder (30).

The early developed Lpp-OmpA system has been in-tensively researched for its translocation mechanism (9, 10).All information that is needed for targeting and insertioninto the outer membrane resides in the signal sequence andthe first nine N-terminal amino acids of Lpp (31, 32). Fusionsto the short Lpp sequence become fatty-acylated, export viathe lipoprotein pathway, and insert into the outer membranebut are not surface-exposed (32). OmpA spans the outermembrane five times, and membrane-spanning � strandsmake up �-barrel structures, which allow the translocationof C-terminally attached passenger proteins across the outermembrane (10, 33).

The Lpp-OmpA system looks promising since it is well-suited as a carrier of relatively large inserts. The largest proteinthat has been successfully displayed with Lpp-OmpA in E.coli so far is a 74 kDa cyclodextrin glucanotransferase (14).Our findings demonstrated that MPH-GFP (60 kDa) fusedto the C-terminus of Lpp-OmpA could be functionallyanchored on the outer membrane of E. coli. The physicalbinding between MPH and the substrate was reinforced byexpressing MPH on the E. coli surface, and it enhances whole-cell catalytic efficiency. Moreover, surface-expressed GFPshows much stronger fluorescence than cytosol-expressedGFP due to the elimination of the barrier effect of the cellmembrane, making it ideal for monitoring the fate of thereleased GEMs in the environment. Furthermore, the strategyof linking GFP to MPH facilitates the online monitoring ofthe expression and localization of MPH. GFP has been usedas a fusion partner for online monitoring and quantifyingprotein production (34). To our knowledge, this is the firstreport on the presentation of GFP fusions on the cell surfaceby Lpp-OmpA. Most importantly, our results highlight thepotential of Lpp-OmpA to be utilized for the functionaldisplay of macromolecule passenger proteins on the cellsurface.

A high transcription rate can block the translocationpathway of a secreted protein, as translocation is generallythe limiting step for a secreted protein (35). The inhibitoryeffects of overexpression on the translocation pathway havebeen well-documented (26, 29). In this study, high doses ofIPTG will induce a high transcription rate; however, the largeamounts of protein thus produced cannot be efficientlytranslocated onto the cell surface. Consequently, whole-cellactivity decreased with increasing concentrations of IPTG.However, fluorescence increased with enhanced expressionlevels because it was minimally affected by the barrier effectof the cell membrane. Our results suggest that inductionwith 0.2 mM IPTG provides an optimal balance betweenwhole-cell activity and fluorescence. In this study, 25 °Cproved to be an optimum induction temperature, suggestingthat a low temperature may be favorable to the correct foldingand proper translocation of proteins (36). The adverse effectsof a high temperature on the surface expression of foreignproteins were reported previously (4, 29).

Anchorage of the macromolecule passenger proteins onthe outer membrane may result in instability of the outermembrane and growth inhibition of the cells (7). In additionto the choice of compatible surface anchors, optimization ofthe expression systems may also waive the metabolic burdenplaced on the cell. In the previous works, expression of theLpp-OmpA fusion proteins was controlled by a lpp-lacpromoter that lacks a Cap site; as such, it was not regulatedby catabolite repression and was weakly constitutively activewithout IPTG induction (10, 11). The promoter leakinessresults in decreased viability of the cells when certain proteinsare expressed. It has been reported that surface expressiondriven by the lpp-lac promoter either inhibits cell growthor reduces cell viability (4, 29). In this study, a low-copy-number plasmid, pLOMG33, containing a tightly regulatedtac promoter was used for surface expression of macromol-ecule fusions in E. coli. The expression of Lpp-OmpA-MPH-GFP is tightly regulated by the tac promoter due tothe presence of the lacIq gene on the plasmid. As a result,cells harboring pLOMG33 did not show detectable MPHactivity and GFP fluorescence prior to induction. Growthinhibition of the cells was also not observed in the XL1-Blue(pLOMG33) cultures. Additionally, the level of expressiondriven by the tac promoter is moderately high (5), while theexpression level is quite low using the lpp-lac promoter (29).Consequently, the use of the tac promoter in this study willproduce more enzyme molecules in a single cell, resultingin improved whole-cell activity and fluorescence.

The current technology is very useful not only for thedetoxification of OPs but also for the rapid detection of OPs.It has been reported that surface-expressed GFP exhibits astronger pH-dependent fluorescence compared to cytosol-expressed GFP (29). Our results showed that the fluorescenceof whole cells displaying MPH-GFP fusion was very sensitiveto extracellular pH changes. Since the hydrolysis of OPs byMPH generates protons, it is possible to develop whole-cellbiosensors for OP detection on the basis of the changes influorescence by utilizing the whole cells displaying MPH-GFP fusion.

Unlike divalent cation-dependent OPH activity, MPHrequires no cofactor for maintaining its activity, suggestingthat MPH-displaying systems may be more suitable for field-scale remediation than previously reported OPH-displayingsystems (4, 37). MPH exhibits high activity for dimethyl OPs(22), while OPH lacks any hydrolytic activity toward numerousdimethyl OPs (38), indicating that the MPH-displaying systemis particularly suitable for the simultaneous degradation anddetection of dimethyl OPs. At present, functional expressionof the OPH-GFP fusion protein on the surface of E. coli hasbeen accomplished by the AIDA-I autotransporter pathway(37). However, surface expression mediated by the AIDA-Iautotransporter requires an ompT (outer membrane proteaseT)-negative host strain, E. coli UT5600 (8, 37). In contrast,surface expression of the Lpp-OmpA fusion proteins hasbeen achieved in various E. coli strains, such as JM105 (4),JM109 (11, 29), XL1-Blue (13), and BL21 (DE3) (15).

E. coli, which is very well-known in regard to its geneticbackground, is still the most commonly used host forrecombinant protein expression (39). However, more effectiveand competitive strains that well-adapt the fluctuatingenvironmental conditions and competition from indigenousmicrobial populations are required for in situ bioremediationof contaminated sites. The broad-host-range vector, pVLT33,used in this study is an RSF1010 derivative and thereforeable to replicate in a wide variety of Gram-negative bacteria(24). At present, the pVLT33-based vectors have beensuccessfully used to express several proteins in various Gram-negative bacteria, such as Moraxella sp (5). and P. putidaJS444 (6, 28). Therefore, the pVLT33-derived surface expres-sion vector, pLOMG33, has enormous potential for func-

VOL. 42, NO. 16, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 6109

tionally displaying MPH-GFP fusion on cell surfaces of theenvironmentally robust bacteria.

AcknowledgmentsThis work was financially supported by the 863 Hi-TechResearch and Development Program of the People’s Republicof China (No. 2007AA06Z335) and the Innovation Programof the Chinese Academy of Sciences (No. KSCX2-YW-G-008).

Supporting Information AvailableFour figures showing additional details of our study. Thisinformation is available free of charge via the Internet athttp://pubs.acs.org.

Literature Cited(1) Singh, B. K.; Walker, A. Microbial degradation of organophos-

phorus compounds. FEMS Microbiol. Rev. 2006, 30, 428–471.(2) Serdar, C. M.; Gibson, D. T. Enzymatic hydrolysis of organo-

phosphates: cloning and expression of a parathion hydrolasegene from Pseudomonas diminuta. Bio/Technology 1985, 3, 567–571.

(3) Mulbry, W. W.; Hams, J. F.; Kearney, P. C.; Nelson, J. O.;McDaniel, C. S.; Wild, J. R. Identification of a plasmid-borneparathion hydrolase gene from Flavobacterium sp. by Southernhybridization with opd from Pseudomonas diminuta. Appl.Environ. Microbiol. 1986, 51, 926–930.

(4) Richins, R. D.; Kaneva, I.; Mulchandani, A.; Chen, W. Biodeg-radation of organophosphorus pesticides by surface-expressedorganophosphorus hydrolase. Nat. Biotechnol. 1997, 15, 984–987.

(5) Shimazu, M.; Mulchandani, A.; Chen, W. Simultaneous deg-radation of organophosphorus pesticides and p-nitrophenolby a genetically engineered Moraxella sp.with surface-expressedorganophosphorus hydrolase. Biotechnol. Bioeng. 2001, 76, 318–324.

(6) Lei, Y.; Mulchandani, A.; Chen, W. Improved degradation oforganophosphorusnerveagentsandp-nitrophenolbyPseudomo-nas putida JS444 with surface-expressed organophosphorushydrolase. Biotechnol. Prog. 2005, 21, 678–681.

(7) Samuelson, P.; Gunneriusson, E.; Nygren, P. A.; Stahl, S. Displayof proteins on bacteria. J. Biotechnol. 2002, 96, 129–154.

(8) Lee, S. Y.; Jong, H. C.; Zhaohui, X. Microbial cell-surface display.Trends Biotechnol. 2003, 21, 45–52.

(9) Earhart, C. F. Use of an Lpp-OmpA fusion vehicle for bacterialsurface display. Methods Enzymol. 2000, 326, 506–516.

(10) Francisco, J. A.; Earhart, C. F.; Georgiou, G. Transport andanchoring of �-lactamase to the external surface of Escherichiacoli. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 2713–2717.

(11) Francisco, J. A.; Stathopoulos, C.; Warren, R. A. J.; Kilburn, D. G.;Georgiou, G. Specific adhesion and hydrolysis of cellulose byintact Escherichia coli expressing surface anchored cellulose orcellulose binding domain. Bio/Technology 1993, 11, 491–495.

(12) Daugherty, P. S.; Chen, G.; Olsen, M. J.; Iverson, B. L.; Georgiou,G. Antibody affinity maturation using bacterial surface display.Protein Eng. 1998, 11, 825–832.

(13) Wang, A. A.; Mulchandani, A.; Chen, W. Whole-cell immobiliza-tion with cell surface-exposed cellulose-binding domain. Bio-technol. Prog. 2001, 17, 407–411.

(14) Wan, H. M.; Chang, B. Y.; Lin, S. C. Anchorage of cyclodextringlucanotransferase on the outer membrane of Escherichia coli.Biotechnol. Bioeng. 2002, 79, 457–464.

(15) Wang, J. Y.; Chao, Y. P. Immobilization of cells with surface-displayed chitin-binding domain. Appl. Environ. Microbiol.2006, 72, 927–931.

(16) Errampalli, D.; Leung, K.; Cassidy, M. B.; Kostrzynska, M.; Blears,M.; Lee, H.; Trevors, J. T. Applications of the green fluorescentprotein as a molecular marker in environmental microorgan-isms. J. Microbiol. Methods 1999, 35, 187–199.

(17) Larrainzar, E.; O’Gara, F.; Morrissey, J. P. Applications ofautofluorescent proteins for in situ studies in microbial ecology.Annu. Rev. Microbiol. 2005, 59, 257–277.

(18) Chalfie, M.; Tu, Y.; Euskirchen, G.; Ward, W. W.; Prasher, D. C.Greenfluorescent protein as a marker for gene expression.Science 1994, 263, 802–805.

(19) March, J. C.; Rao, G.; Bentley, W. E. Biotechnological applications

of green fluorescent protein. Appl. Microbiol. Biotechnol. 2003,62, 303–315.

(20) Elvang, A. M.; Westerberg, K.; Jernberg, C.; Jansson, J. K. Useof green fluorescent protein and luciferase biomarkers tomonitor survival and activity of Arthrobacter chlorophenolicusA6 cells during degradation of 4-chlorophenol in soil. Environ.Microbiol. 2001, 1, 32–42.

(21) Cormack, B. P.; Valdivia, R. H.; Falkow, S. FACS-optimizedmutants of the green fluorescent protein (GFP). Gene 1996,173, 33–38.

(22) Cui, Z. L.; Li, S. P.; Fu, G. P. Isolation of methyl parathion-degrading strain M6 and cloning of the methyl parathionhydrolase gene. Appl. Environ. Microbiol. 2001, 67, 4922–4925.

(23) Yang, C.; Liu, N.; Guo, X.; Qiao, C. Cloning of mpd gene froma chlorpyrifos-degrading bacterium and use of this strain inbioremediation of contaminated soil. FEMS Microbiol. Lett. 2006,265, 118–125.

(24) de Lorenzo, V.; Eltis, L.; Kessler, B.; Timmis, K. N. Analysis ofPseudomonas gene products using lacIq/Ptrp-lac plasmids andtransposons that confer conditional phenotypes. Gene 1993,123, 17–24.

(25) Sambrook, J.; Russel, D. W. Molecular Cloning: A LaboratoryManual, 3rd ed.; Cold Spring Harbor Laboratory Press: ColdSpring Harbor, NY, 2001.

(26) Li, L.; Kang, D. G.; Cha, H. J. Functional display of foreign proteinon surface of Escherichia coli using N-terminal domain of icenucleation protein. Biotechnol. Bioeng. 2004, 85, 214–221.

(27) Bokman, S. H.; Ward, W. W. Renaturation of Aequorea greenfluorescent protein. Biochem. Biophys. Res. Commun. 1981, 101,1372–1380.

(28) Yang, C.; Cai, N.; Dong, M.; Jiang, H.; Li, J.; Qiao, C.; Mulchandani,A.; Chen, W. Surface display of MPH on Pseudomonas putidaJS444 using ice nucleation protein and its application indetoxification of organophosphates. Biotechnol. Bioeng. 2008,99, 30–37.

(29) Shi, H.; Su, W. W. Display of green fluorescent protein onEscherichia coli cell surface. Enzyme Microb. Technol. 2001, 28,25–34.

(30) Yang, F.; Moss, L. G.; Phillips, G. N., Jr. The molecular structureof green fluorescent protein. Nat. Biotechnol. 1996, 14, 1246–1251.

(31) Georgiou, G.; Stephens, D. L.; Stathopoulos, C.; Poetschke, H. L.;Mendenhall, J.; Earhart, C. F. Display of �-lactamase on theEscherichia coli surface: outer membrane phenotypes conferredby Lpp′-OmpA′-�-lactamase fusions. Protein Eng. 1996, 9, 239–247.

(32) Ghrayeb, J.; Inouye, M. Nine amino acid residues at the NH2-terminal of lipoprotein are sufficient for its modification,processing, and localization in the outer membrane of Escheri-chia coli. J. Biol. Chem. 1984, 259, 463–467.

(33) Vogel, H.; Jahnig, F. Models for the structure of outer-membraneproteins of Escherichia coli derived from raman spectroscopyand prediction methods. J. Mol. Biol. 1986, 190, 191–199.

(34) Wu, C. F.; Cha, H. J.; Rao, G.; Valdes, J. J.; Bentley, W. E. A greenfluorescent protein fusion strategy for monitoring the expres-sion, cellular location, and separation of biologically activeorganophosphorus hydrolase. Appl. Microbiol. Biotechnol. 2000,54, 78–83.

(35) Rodrigue, A.; Chanal, A.; Beck, K.; Muller, M.; Wu, L. Co-translocation of a periplasmic enzyme complex by a hitchhikermechanism through the bacterial Tat pathway. J. Biol. Chem.1999, 274, 13223–13228.

(36) Bukau, B. Regulation of the Escherichia coli heat-shock response.Mol. Microbiol. 1993, 9, 671–680.

(37) Li, C.; Zhu, Y.; Benz, I.; Schmidt, M. A.; Chen, W.; Mulchandani,A.; Qiao, C. Presentation of functional organophosphorushydrolase fusions on the surface of Escherichia coli by the AIDA-Iautotransporter pathway. Biotechnol. Bioeng. 2008, 99, 485–490.

(38) Horne, I.; Sutherland, T. D.; Harcourt, R. L.; Russell, R. J.;Oakeshott, J. G. Identification of an opd (organophosphatedegradation) gene in an Agrobacterium isolate. Appl. Environ.Microbiol. 2002, 68, 3371–3376.

(39) Terpe, K. Overview of bacterial expression systems for heter-ologous protein production: from molecular and biochemicalfundamentals to commercial systems. Appl. Microbiol. Bio-technol. 2006, 72, 211–222.

ES800441T

6110 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 16, 2008