Biosensors and Bioelectronics 23 (2007) 248–252
SPR biosensor for the detection of L. monocytogenes
using phage-displayed antibody?
Viswaprakash Nanduria, Arun K. Bhuniaa, Shu-I Tub, George C. Paolib, Jeffrey D. Brewsterb,∗
aCenter For Food Safety and Engineering, Department of Food Science, Purdue University, 745 Agriculture Mall Dr., West Lafayette, IN 47907, USA
bMicrobial Biophysics and Residue Chemistry, Eastern Regional Research Center, USDA-ARS, 600 E. Mermaid Lane, Wyndmoor, PA 19038, USA
Received 28 December 2006; received in revised form 29 March 2007; accepted 5 April 2007
Available online 19 April 2007
Whole cells of Listeria monocytogenes were detected with a compact, surface plasmon resonance (SPR) sensor using a phage-displayed scFv
antibody to the virulence factor actin polymerization protein (ActA) for biorecognition. Phage Lm P4:A8, expressing the scFv antibody fused
to the pIII surface protein was immobilized to the sensor surface through physical adsorption. A locally constructed fluidics system was used to
deliver solutions to the compact, two-channel SPREETATMsensor. Specificity of the sensor was tested using common food-borne bacteria and
a control phage, M13K07 lacking the scFv fusion on its coat protein. The detection limit for L. monocytogenes whole cells was estimated to be
2×106cfu/ml. The sensor was also used to determine the dissociation constant (Kd) for the interaction of phage-displayed scFv and soluble ActA
in solution as 4.5nM.
© 2007 Elsevier B.V. All rights reserved.
Keywords: Listeria monocytogenes; SPR; Phage display
Food-borne diseases cause an estimated 76 million illnesses,
accounting for 325,000 hospitalizations and more than 5000
deaths in the United States each year (Mead et al., 1999).
Currently, there are more than 250 known food-borne dis-
viruses, bacteria and fungi. Conventional culture methods for
detecting food-borne pathogens entail a minimum of 3–7 days,
of biosensors for rapid detection of food-borne pathogens. The
need for field-capable biosensors for the rapid, onsite early
detection of pathogens is essential to reduce the risk of con-
in humans is of particular interest because it is able to survive
purpose of providing specific information and does not imply recommendation
or endorsement by the U.S. Department of Agriculture.
∗Corresponding author. Tel.: +1 215 233 6447; fax: +1 215 233 6559.
E-mail address: email@example.com (J.D. Brewster).
and grow at low temperatures and because the mortality rate for
borne pathogens. L. monocytogenes is a Gram-positive aerobic
rod-shaped bacterium, which has been responsible for several
well-documented food poisoning outbreaks (Donnelly, 2002).
Immunocompromised patients, such as pregnant women, new-
be immobilized on a sensing platform. An ideal probe should be
able to achieve sensitive and specific detection of the target ana-
stresses, such as changes in temperature and pH. Antibodies,
both monoclonal and polyclonal, are critical elements of many
pathogen biosensors. Selective, high affinity antibodies pro-
High cost of production, difficulty in reproducing stocks with
similar target specificities, loss of functionality when exposed
to extremes of the environment, and failure of in vivo systems to
produce antibodies selective for important pathogens are some
sor applications. To overcome these limitations, phage display
0956-5663/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
V. Nanduri et al. / Biosensors and Bioelectronics 23 (2007) 248–252
technology has been used to develop single-chain antibodies
for some pathogens (Benhar et al., 2001; Paoli et al., 2004).
The filamentous bacteriophages used as probes in this study are
bacterial viruses that possess a surface displayed single-chain
antibody (scFv). The scFv is expressed as a fusion to the minor
phage surface. The phage probe used in this study, Lm P4:A8,
was selected from a naive scFv phage display library for its spe-
cific interaction with L. monocytogenes (Paoli et al., 2004). The
ActA (Paoli and Brewster, 2007), an L. monocytogenes protein
SPR has become a well-established tool for the characteriza-
recently for the detection of pathogens (Lathrop et al., 2003;
Leonard et al., 2004; Oh et al., 2005; Geng and Bhunia, 2006)
and pathogen products, including Escherichia coli enterotoxin
2006); E.coli O157:H7 (Taylor et al., 2005, 2006); Salmonella
(Bokken et al., 2003) and biological warfare agents (Naimushin
tion of target species.
of free electrons in a metal that propagates along the interface
between the metal and an adjacent dielectric. For a metal film in
contact with a dielectric, light incident above the critical angle
is ordinarily completely reflected. However, for specific values
of incidence angle, polarization and wavelength, incident pho-
tons are absorbed at the interface, exciting a surface plasmon
and reducing the reflected intensity to a low value. The wave-
strongly on the refractive index (RI) of the dielectric layer, and
can be monitored to detect small changes (∼1×10−6) in RI in
the region near (∼1 wavelength) the interface. SPR biosensors
have receptors immobilized on the metal surface and a fluidics
system for delivering ligand solution to the sensing layer. Bind-
ing of ligand changes the RI (as do changes in temperature or
bulk RI of the liquid) and the time dependence and equilibrium
The SPREETA3TMis a compact (∼2cm×6cm×4cm),
inexpensive (US$ ∼100) and low-power disposable sensor,
which requires only a fluidic system and interface electronics
to provide a complete SPR biosensor platform. Such a platform
have the potential to provide much smaller and less expensive
SPR biosensor platforms than other products currently on the
The SPREETA3TMsensor used in this investigation is sim-
ilar to that described previously (Melendez et al., 1996; Elkind
et al., 1999) except that there are three sensing layers on the
gold surface of the chip as opposed to two in the older models.
SPREETATMsensors have been successfully employed for the
detection of Salmonella enteritidis and E. coli (Waswa et al.,
2005), Salmonella typhimurium (Meeusen, 2001) and Staphylo-
coccus aureus enterotoxin (Naimushin et al., 2002).
In this study, the SPREETA3TMsensor was used with a
displayed scFv with ActA, a protein virulence factor produced
by L. monocytogenes. Whole cells of L. monocytogenes were
also detected using this system. Phage Lm P4:A8 was immo-
bilized through a simple and effective process of physical
adsorption, which did not involve any additional reagents, steps
or complex chemistry (Nanduri et al., 2006).
2. Materials and methods
2.1. Materials and reagents
Deionized water was obtained from a Nanopure (Barnstead,
Dubuque, IA) water treatment system. Bacterial media and
agar were obtained from Difco (Detroit, MI, USA). Dulbecco’s
phosphate-buffered saline solution [DPBS, pH 7.5], bovine
serum albumin (BSA) and Triton X-100 were obtained from
ing 1% Triton X-100 was used for cleaning the sensor surface
prior to use. All solutions were allowed to come to room tem-
perature before use.
2.2. Phage antibody
The phage, Lm P4:A8, expressing the scFv antibody was
isolated previously (Paoli et al., 2004) from the Griffin.1 library
(Griffiths et al., 1994). The phage host strain E. coli TG1 and
the library described above were both a gift of the Centre for
Protein Engineering (Cambridge, UK). Phage Lm P4:A8 was
propagated using E. coli TG1 as described previously (Paoli et
al., 2004). The concentration of phage used in this investigation
was 5×1010transfecting units (TU)/ml.
2.3. Bacterial cultures and media
Culture Collection (Manassas, VA). Lactococcus lactis ATCC
19435 and Enterococcus fecalis SDFC2 were provided by Dr.
George Somkuti (USDA, ERRC, Wyndmoor, PA). L. monocy-
togenes strain DP-L3563 was kindly provided by Dr. Daniel
Portnoy (University of California, Berkley). All Listeria spp.
were grown in Brain–Heart-Infusion (BHI) broth. L. lactis and
E. fecalis were grown in tryptone–yeast extract–lactose (TYL)
and trypticase soy broth (TSB), respectively. Overnight cultures
were prepared by inoculating the respective media with a sin-
gle colony from a plate culture and incubating at 37◦C with
shaking at 250rpm for 16h. Overnight culture cell counts were
estimated by plating appropriate dilutions of the parent culture
V. Nanduri et al. / Biosensors and Bioelectronics 23 (2007) 248–252
Fig. 1. Schematic of the SPR system.
for all further overnight cultures of the same bacteria.
2.4. ActA protein
The ActA protein was purified from L. monocytogenes strain
at a stock concentration of 300?g/ml and stored frozen. From
this, sub-stocks were made and stored at 4◦C. Graded concen-
trations ranging from 45nM to 22pM of ActA in DPBS were
prepared before each experiment from the sub-stock.
2.5. SPR system
The Nomadics SPR3 (Model 1146643) system consisted
of a SPREETA3TMthree-channel sensor (Texas Instruments),
anodized aluminum sensor housing, fluidics block, controller
module and PC-based software as shown in Fig. 1. The working
been described (Naimushin et al., 2002). Briefly, the sensor is
a matchbox-size integrated device containing a light emitting
diode (840nm) with a polarizer, a reflecting mirror, an exposed
gold SPR active surface and a photodiode array that detected
reflected light intensity. Photodiode signals were converted in
software to refractive index units (RIU). A change of 1000RIU
typically corresponds to a surface protein concentration change
of 1ng/mm2(Green et al., 2000; Rich and Myszka, 2000). Data
were observed in real-time using the “Multispr” (version 10.93)
software and stored to disk for offline analysis and plotting
using Origin (Microcal) and Excel (Microsoft). The stock flu-
idics module consisted of a gasket-less polypropylene flow cell
with three flow channels and six press-fit PTFE tubes for fluid
inlet/outlet. The flow cell was pressed against the SPREETA
sensing surface by a pressure plate and thumbscrew. Problems
with leaks and inconsistent flow led us to modify the stock flu-
idics system to include a silicone rubber gasket with two flow
channels. To ensure uniform sealing of the gasket, two addi-
tional thumbscrews were added. With these modifications to the
flow block, the system gave consistent, leak-free flow on the
outer two channels. An Ismatec peristaltic pump (Cole-Parmer
Instrument Co.) was used to pump solutions through the flow
cell at 51?l/min via silicone tubing (0.64mm i.d., 47cm long).
2.6. SPR sensor preparation
Fig. 2. Real-time calibration responses from the two channels of a SPREETA
sensor surface, when both the channels are loaded with five concentrations of
glycerol in DPBS.
of chromium trioxide in water) for 2min as per the manufac-
turer’s guidelines. Following this, the sensor surface was rinsed
thoroughly with copious quantities of deionized water and air-
dried. The initial reference readings in air and water required to
calibrate the sensor were performed without the flow cell, tak-
ing care to shield the sensor from external light. Cleanliness of
the sensor surface was determined by confirming that the mea-
cleanliness of the surface was established, the flow cell with the
ing the thumbscrews of the swinging arm lock, following which
an in situ cleaning of the sensor surface with 0.1M NaOH and
1% Triton X-100 was performed by pumping the cleaning solu-
tion through the entire flow system for approximately 15min.
Prior to each experiment, the uniformity of responses from both
the channels of the flow through system was tested using five
concentrations (1–10%) of glycerol in DPBS. Fig. 2 shows a
typical real-time response obtained from the two channels. Fol-
lowing this, buffer (DPBS) was flushed through the system until
a stable baseline was observed. Unless specifically mentioned,
all the experiments were carried out at room temperature.
2.7. Experimental methods
Materials were introduced to the sensor flow cell by dip-
ping the pump inlet tubing into a vial containing the solution
of interest. Phages were immobilized by passive adsorption
to the cleaned gold active surface using a concentration of
5×1010TU/ml followed by washing with DPBS until a steady
response was reached. BSA (2mg/ml) was passed through the
system for approximately 15min to block the sensor surface
and reduce non-specific binding. Bacteria and ActA solutions
mediate washes with DPBS, and the net changes in RIU were
recorded. Specificity of the sensor for L. monocytogenes was
studied by measuring the response of a number of control bac-
teria. Non-specific binding of ActA and L. monocytogenes was
taining immobilized control phage, M13KO7, lacking the scFv
V. Nanduri et al. / Biosensors and Bioelectronics 23 (2007) 248–252
Fig. 3. A typical real-time response of phage loaded at two different concentra-
tions of 5×1010(gray) and 5×109(black) TU/ml.
3. Results and discussion
3.1. Phage immobilization
bilizing filamentous phage through physical adsorption at a
concentration of 3.2×1011TU/ml for 3h was sufficient to pro-
duce a high, stable signal. In an effort to reduce the amount
of phage and the time required for the preparation of the sen-
sor chip we examined other immobilization conditions. Fig. 3
shows the simultaneous real-time binding of phage at two dif-
ferent concentrations (5×109and 5×1010TU/ml) on the two
channels. Much higher phage immobilization was observed at
the higher concentration and an exposure time of 2h provided
a nearly saturated surface. Based on these and similar results,
selected for all further experiments.
3.2. Detection of L. monocytogenes
Fig. 4 shows the response of the biosensor to various con-
centrations of L. monocytogenes cells in buffer. Binding of
immobilized Lm P4:A8 (black) bearing scFv to L. monocytogenes. Channel 2
contained control phage, M13 K07 (gray) without scFv. The mean responses
obtained from 12 individual experiments repeated under similar conditions are
Fig. 5. Responses obtained for phage immobilized sensor with L. monocy-
togenes (?), L. ivanovii (?), L. seeligeri (?), Entercoccus fecalis (?) and
Lactococcus lactis (♦). Error bars represent one standard deviation about the
mean of six replicate measurements.
L. monocytogenes to immobilized phage displaying scFv was
was low. The curve shown is the sigmoidal fit (Microcal Origin)
of mean values obtained from 12 measurements and the error
bars represent one standard deviation about the mean. Fig. 5
shows the responses obtained for the sensor with L. monocyto-
genes (?); L. ivanovii (?) and L. seligeri (?), E. fecalis (?) and
L. lactis (♦). The mean values obtained from six replicate mea-
surements were sigmoid fit using Microcal Origin software. For
all the tests, responses obtained from L. monocytogenes were
significantly stronger than those obtained for control species.
The SPREETA3TMSPR system was capable of detecting whole
bacterial cells, in contrast to other SPR systems, such as IAsys
(Lathrop et al., 2003). The detection limit was estimated using
a hypothesis-testing approach (Currie, 1968) adapted for a non-
linear calibration curve and non-uniform variance of the signal.
The net signal (scFv phage–control phage) was fit to generate a
false negative and false positive results, the detection limit was
1.28 times the sum of the blank standard deviation (27RIU) and
yielding a value of 2×106cfu/ml.
3.3. Dissociation constant of ActA
The binding of ActA to phage-displayed scFv was examined
over a range of concentrations using adsorbed Lm P4:A8 phage
in the active channel and adsorbed M13K07 phage in the con-
trol channel. The results shown in Fig. 6 are the mean responses
obtained from a set of seven experiments conducted under iden-
tical conditions. Fitting the data to the following equation:
using a non-linear parameter estimation routine (Mathematica)
requires delivering cells to the surface of the sensor. This can
be problematic in samples where the bacteria are tightly bound
252 Download full-text
V. Nanduri et al. / Biosensors and Bioelectronics 23 (2007) 248–252
one standard deviation from the mean of seven replicate measurements.
to food particles or other surfaces (e.g. biofilms). The ability to
detect nanomolar levels of ActA may provide a means to detect
L. monocytogenes indirectly in challenging sample matrices.
This study demonstrates that phage-displayed antibodies
(scFv fusions to pIII) can be used as the biorecognition ele-
ment for specific and sensitive detection of L. monocytogenes
using surface plasmon resonance. Passive adsorption of phage
on clean gold surface of the SPR sensor chip for 2h at
affinity. The biosensor provided an estimated detection limit of
2×106cfu/ml for L. monocytogenes whole cells in buffer. The
biosensor was also used to measure the binding constant for
the interaction between the scFv antibody and ActA in solution
with the Agricultural Research Service of the U.S. Department
of Agriculture project number 1935-42000-035 and the Cen-
ter for Food Safety and Engineering at Purdue University. The
the valuable support, materials and feedback and Greg Winter
from New England Biolabs for graciously providing us with the
Griffin.1 phage library.
Benhar, I., Eshkenazi, I., Neufeld, T., Opatowsky, J., Shaky, S., Rishpon, J.,
2001. Talanta 55, 899–907.
Bokken, G.C., Corbee, A.M., Van Knapen, R.J., Bergwerff, A.A., 2003. FEMS
Microbiol. Lett. 222, 75–82.
Cooper, A.M., 2003. Anal. Bioanal. Chem. 377, 834–842.
Currie, L.A., 1968. Anal. Chem. 40, 586–593.
Donnelly, C.W., 2002. J. Assoc. Official Anal. Chem. 85, 495–500.
Elkind, J.L., Stimpson, D.I., Strong, A.A., Bartholomew, D.U., Melendez, J.L.,
1999. Sens. Actuators B: Chem. 54, 182–190.
Geng, T., Bhunia, A.K., 2006. In: Knopf, G.K., Bassi, A.S. (Eds.), Smart
Biosensor Technology. Taylor and Francis, Boca Raton, FL, pp. 505–
Griffiths, A.D., Williams, S.C., Hartley, O., Tomlinson, I.M., Waterhouse, P.,
EMBO J. 13, 3245–3260.
Green, R.J., Frazier, R.A., Shakesheff, K.M., Davies, M.C., Roberts, C.J.,
Tendler, S.J.B., 2000. Biomaterials 21, 1823–1835.
Hearty, S., Leonard, P., Quinn, J., O’Kennedy, R., 2006. J. Microbiol. Methods
Hof, H., 2003. FEMS Immunol. Med. Microbiol. 35, 199–202.
Johnsson, B., Lofas, S., Lindquist, G., 1991. Anal. Biochem. 198, 268–277.
Lakey, J.H., Raggett, E.M., 1998. Curr. Opin. Struct. Biol. 8, 119–123.
Lathrop, A.A., Jaradat, Z.W., Haley, T., Bhunia, A.K., 2003. J. Immunol. Meth-
ods 281, 119–128.
Leonard, P., Hearty, S., Quinn, J., O’Kennedy, R., 2004. Biosens. Bioelectron.
Mead, P.S., Slutsker, L., Dietz, V., McCaig, L.F., Bresee, J.S., Shapiro, C.,
Griffin, P.M., Tauxe, R.V., 1999. Emerg. Infect. Dis. 5, 607–625.
Meeusen, C.A., 2001. ASAE Annual International Meeting. Sacramento, CA.
Melendez, J., Carr, R., Bartholomew, D.U., Kukanskis, K., Elkind, J., Yee,
S., Furlong, C., Woodbury, R., 1996. Sens. Actuators B: Chem. 35, 212–
Naimushin, A.N., Soelberg, S.D., Nguyen, D.K., Dunlap, L., Bartholomew,
D., Elkind, J., Melendez, J., Furlong, C.E., 2002. Biosens. Bioelectron. 17,
Naimushin, A.N., Spinelli, C.B., Soelberg, S.D., Mann, T., Stevens, R.C., Chi-
nowsky, T., Kauffman, P., Yee, S., Furlong, C.E., 2005. Sens. Actuators B:
Chem. 104, 237–248.
Nanduri, V., Samoylov, A.M., Petrenko, V.A., Vodyanoy, V., Simonian, A.L.,
2004. 206th Meeting of the Electrochemical Society. Honolulu, Hawaii.
Vodyanoy, V., 2006. Biosens. Bioelectron. 22, 986–992.
Oh, B.-K., Lee, W., Chun, B.S., Bae, Y.M., Lee, W.H., Choi, J.-W., 2005.
Colloids. Surf. A: Physicochem. Eng. Aspects 257–258, 369–374.
Paoli, G.C., Chen, C.-Y., Brewster, J.D., 2004. J. Immunol. Methods 289,
Paoli, G.C., Brewster, J.D., 2007. J. Rapid Methods Autom. Microbiol. 15,
Rich, R.L., Myszka, D.G., 2000. Curr. Opin. Biotechnol. 11, 54–61.
Rich, R.L., Myszka, D.G., 2005. J. Mol. Recognit. 18, 431–478.
Soelberg, S.D., Chinowsky, T., Geiss, G., Spinelli, C.B., Stevens, R., Near, N.,
Kauffman, P., Yee, S., Furlong, C.E., 2005. J. Ind. Microbiol. Biotechnol.
Spangler, B.D., Wilkinson, E.A., Murphy, J.T., Tyler, B.J., 2001. Anal. Chim.
Acta 444, 149–161.
Taylor, A.D., Yu, Q., Chen, S., Homola, J., Jiang, S., 2005. Sens. Actuators B:
Chem. 107, 202–208.
Taylor, A.D., Ladd, J., Yu, Q., Chen, S., Homola, J., Jiang, S., 2006. Biosens.
Bioelectron. 22, 752–758.
Waswa, J., Irudayaraj, J., Debroy, C., 2005. Food Sci. Technol. 40, 187–192.
Welch, M.D., Rosenblatt, J., Skoble, J., Portnoy, D.A., Mitchison, T.J., 1998.
Science 281, 105–108.