Eg

Za

b

a

ARAA

KBGBVC

1

iaadeiwLbtpacrgtmptmp

0d

Colloids and Surfaces B: Biointerfaces 76 (2010) 529–534

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces

journa l homepage: www.e lsev ier .com/ locate /co lsur fb

ffect of unassembled phage protein complexes on the attachment to cellulose ofenetically modified bacteriophages containing cellulose binding modules

huo Lia, Mona Tolbab, Mansel Griffithsb, Theo G.M. van de Vena,∗

Department of Chemistry, Pulp & Paper Research Centre, McGill University, 3420 University Street, Montreal, QC, Canada H3A 2A7Department of Food Sciences, University of Guelph, Guelph, Canada

r t i c l e i n f o

rticle history:eceived 30 October 2009ccepted 16 December 2009vailable online 24 December 2009

a b s t r a c t

One way to make bioactive paper is to incorporate in it bacteriophages, which can capture and deactivatebacteria. The release of newly assembled phages from lysed bacteria can also serve as a measure of thenumber of bacteria that were deactivated. We have studied the deposition of genetically modified phages,containing cellulose binding modules, on model cellulose surfaces by evanescent wave light scattering.

eywords:ioactive paperenetically modified virusesacteriophageirus deposition

It was shown that unassembled protein complexes, released with the phages during the lysis of bacteria,limit the deposition of phages to about one per 10–20 �m2, as there are far more unassembled proteincomplexes than phages. The implication is that a few thousand phages can attach to a cellulose fiber, withmost of the surface area taken up by protein complexes. This is more than sufficient for the production ofbioactive paper. The unassembled protein complexes may be beneficial, as they will prevent non-specific

us, itoper

ellulose binding domain deposition of bacteria. Than expensive purification

. Introduction

Since paper is inexpensive and a potential carrier for antibod-es, enzymes, aptamers or bacteriophages, it is possible to produce

paper based biosensor to detect, capture and deactivate waternd airborne pathogens. This idea of bioactive paper has beeneveloped by SENTINEL (a NSERC Strategic Research Network) andxplored by many researchers in the past few years. One methodnvolves immobilization of bacteriophages onto paper substrates,

hich in turn can be used as water filters or sensors for bacteria.earning about the kinetics of phage interactions with cellulose oracteria at the solid/liquid interface is essential in proving this par-icular bioactive paper concept. However, the impurities from thehage stock can readily adhere to paper as well, due to electrostaticnd dispersion interactions. As impurities larger than the phagesan be removed by simple filtration, the major source of impu-ities comes from unassembled protein complexes during phagerowth or debris of host Escherichia coli bacteria. Understandinghe effect of these partially assembled virus proteins on the attach-

ent of phages to cellulose is important in improving the phage

reparation procedure and in gaining information about the bioac-ive paper’s effectiveness in deactivating bacteria. Therefore, the

ain objective of this study is to prove the concept of bioactiveaper by demonstrating the immobilization of bacteriophages on

∗ Corresponding author.E-mail address: theo.vandeven@mcgill.ca (T.G.M. van de Ven).

927-7765/$ – see front matter © 2010 Elsevier B.V. All rights reserved.oi:10.1016/j.colsurfb.2009.12.014

may not be necessary to separate the protein complexes from phages ination.

© 2010 Elsevier B.V. All rights reserved.

cellulose (a model surface for paper) and to study the competitiveadhesion kinetics of the phages and the unassembled protein com-plexes, using the genetically modified CBM-T4 bacteriophage as amodel system.

A bacteriophage is by definition an obligate intracellular parasitethat multiplies inside bacteria by making use of some or all of thehost biosynthetic machinery, first discovered by Frederick Twortin 1915 and independently by Felix D’Herelle, a French-Canadianmicrobiologist, in 1917 [1,2]. In health related applications, anti-bacterial (or phage) therapy was practiced with sporadic successworldwide in the 1920s and 1930s. However, those early bacterio-phage applications were confounded due to a poor understandingof their biology. Recently, due to the emergence of antibiotic-resistant bacterial strains, interest in searching for alternativetherapies to treat bacterial infections is growing. Meanwhile, thereis a sustained demand for more effective and efficient bacterialrecognition agents in the rapidly developing field of biosensors.

Bacteriophages are a prominent candidate for both demands asthey possess many useful features: they are highly specific to par-ticular bacterial species, they are safe to animals and plants, theycan self-reproduce upon infection and they cannot only identifybut also kill the pathogens in a reasonable time period of about30 min. However, phages are not used routinely for the treatment

of infectious diseases, the testing of food quality or the purificationof water supplies. Instead, antibodies and chemical substances areused. The reasons are complex and are different for various applica-tions. For biosensors or biofilters, one difficulty is the quick removalof phages from the targets as it is hard to immobilize them onto the

530 Z. Li et al. / Colloids and Surfaces B: Biointerfaces 76 (2010) 529–534

F ges cap

spi[

(chonihtmfwgbltTaoitp

iicvdsc

tbmmmsmeaoopssdppIcn

ig. 1. The immobilization of CBM-T4 onto cellulose surface. The attachment of pharotein complexes are expected to adsorb as well.

ubstrates, such as cellulose. It has been reported that by simplyouring an undisclosed (unknown) quantity of phages into a drink-

ng well to control cholera has resulted in disastrous consequences3].

One possible solution is using bacteriophages with CBMscellulose- or carbohydrate-binding modules). Most researchershoose to modify the bacteriophage T4 for model studies, as T4as been extensively investigated. We have a large knowledge basef its structure, its mechanism of bacterial pathogenesis and theature of phage–host interactions. A wild-type bacteriophage T4

nfects specifically E. coli. It consists of an icosahedral (20 sided)ead, a contractile tail, six short and six long fibers for attachingo the target, and a base plate that is the nerve center for com-

unicating between the fibers and the tail. It exists as an inactiveorm (virion) until one of its extended fibers comes into contactith the surface of an E. coli cell. It will then attach and inject its

enetic blueprint into the cell and the phage DNA sets the cell’siosynthetic machinery to work to create phage replicas. During

ysis, one E. coli releases about 300 copies of T4 phage and perhapshousands of unassembled protein complexes and DNA molecules.4 is also one of the largest phages with a 120 nm × 86 nm capsidnd a 120 nm × 20 nm tail. As a T4 phage exemplifies the life cyclef many animal viruses, it is a good candidate to be used in stud-es of cellulose–phage interactions. To promote the interaction ofhe virus with cellulose, it is convenient to first display a bindingrotein (a cellulose binding module) on the capsid of T4.

Gervais et al. [4] have reported the expression of a biotin bind-ng domain on the capsid protein of T4. The phages retained theirnfectivity, burst size and latent period comparable to its wild-typeounterpart. These phages can therefore be attached to strepta-idin coated gold surfaces functioning as a biosensor for E. coli. Onerawback of the application is the requirement of streptavidin. Theame method can be used to express other binding domains on theapsid of T4.

In order to develop the concept of using paper as a poten-ial anti-bacterial agent carrier in diagnostic applications or otheracterial detection areas, we also expressed carbohydrate-bindingodules (CBMs) on the capsids of T4 bacteriophages, using a similarethod. This CBM is a family 9 C-terminal module of the ther-ostable Thermotoga maritima xylanase 10A (CBM9-2). It binds

pecifically to amorphous and crystalline cellulose and a range ofono- and disaccharides [5]. With more than 800 copies of CBMs

xpressed on the capsid (equivalent to a surface concentration ofbout 1 CBM/25 nm2) and a relatively high association constant Ka

f (1–3) × 10−5 M−1 [5], there is a very good chance for the phage torient and bind with the cellulose specifically via the head capsidrotein. The tail and fibers of the phage which are free to move, cantill capture the bacteria efficiently [6]. Ideally the cellulose–phageystem will offer a stable and antigen-free probe for detection andeactivation of E. coli (Fig. 1). Due to the complicated structure of

aper, a cellulose film was produced from cellulose triacetate [7],roviding a smooth and uniform platform for particle attachment.

n addition, the unassembled protein complexes are also likely toontain a number of CBMs. As they are smaller in size and larger inumber, they will move and bind faster with the cellulose, conse-

n be detected by EWLS. The cellulose–phage system can detect E. coli. Unassembled

quently blocking the attachment of the CBM-T4s. Some purificationmethods are available, but they are mostly applied to on a smallscale in the lab. The CsCl step gradient method is often used [8].However, it is not cost-effective for large scale production, becauseit requires time-consuming steps and intensive technical support.

Although the T4 bacteriophages are relatively large in size,it is still a challenge to view them with conventional lightmicroscopy. Transmission electron microscopy (TEM) and scan-ning electron microscopy (SEM) are usually employed in orderto observe the structural details of the phages. However, sinceelectron microscopy is done in vacuum, it will not give anyinformation of the kinetics of phage–cellulose interactions inwater.

The evanescent wave light scattering (EWLS) technique allowsus to “see” phages that are linked to the cellulose surface andthe impinging jet system ensures the controlled transportation ofphages to the interface so that we can monitor the adhesion pro-cess in situ. Since after filtering, phages are likely to be the largeststructures released upon lysis of bacteria, they scatter the mostlight and contribute predominantly to the observed light scatter-ing. The detailed theory of the impinging jet and the evanescentwave technique has been worked out [9]. In short, an evanescentwave region is produced by a laser at the prism/solution interface. Asuspension of phages is transferred through a single radial imping-ing jet and deposited onto the cellulose film that is spin coatedon the prism. The phages linked to the cellulose will scatter light inthe evanescent wave region, which has an area of 440 × 590 �m [9],equivalent to the surface area of about one or two cellulose fibers.By collecting the scattered light during the deposition process weare able to obtain quantitative information about the kinetics ofphage deposition.

The impinging jet can be used to fill the deposition cell and tostudy the deposition kinetics. One can either study flow-induceddeposition, or by reducing the flow rate to zero, diffusion kinetics.Here we study diffusion, because it is commonly involved in prepar-ing a bioactive surface. Diffusion kinetics are also highly relevant tomany other applications such as various filtering procedures, pro-tein adsorption and bioparticle immobilization. The non-stationarydiffusion kinetics can be described by [10]:

ns = 2˛(

Dt

�

)1/2nb (1)

where ns is the number of phages attached to the surface per unitarea, ˛ the deposition efficiency, D the diffusion coefficient, t thetime, and nb the concentration of the phages in the bulk. This equa-tion holds with a few assumptions: the particle-wall specific andhydrodynamic interactions are neglected; the initial particle distri-bution everywhere in the suspension is uniform; there is no particledetachment after adsorption; and there are no blocking effects dueto the adsorbed particles. Therefore, this simple square root of time

solution will not be valid for the entire adsorption process. How-ever, for the initial period when the surface coverage is low, Eq. (1)is reasonably accurate, and applies both to the deposition of CBM-T4 phages and CBM-proteins. At later times, the kinetics are sloweddown by blocking, i.e. by adsorption of biomolecules (unassembled

Z. Li et al. / Colloids and Surfaces B: Biointerfaces 76 (2010) 529–534 531

F on gru

pu

2

2

LabpwTA

2

pv3tb6Tp

2

s

ig. 2. TEM images of CBM-T4 phages and unassembled protein complexes on carbnder different magnifications.

rotein complexes, smaller proteins, DNA fragments, etc.), takingp sites which become unavailable for phage deposition.

. Experimental

.1. Materials

Luria Bertani (LB) medium was prepared by dissolving 25 g ofB powder into 1 L of water. Milli-Q ultra pure water was used inll experiments. LB-agar medium for bottom plates was preparedy adding 40 g of LB agar in 1 L of water. LB-agar for top layer wasrepared by adding 5 g of agar into 1 L of LB medium. The � bufferas prepared by diluting 5.8 g NaCl, 2.0 g MgSO4·7H2O, 50 mL 1 M

ris–HCl, 0.1 g gelatin in 1 L of water and adjusted to pH 8 with HCl.ll LB medium, LB agar and � buffer were autoclaved before use.

.2. Preparation of T4/CBM-T4 bacteriophage stocks

Wild-type T4 bacteriophages and CBM-bacteriophages wereroduced and purified using the same method as described pre-iously [7]. E. coli cells were routinely grown in LB medium at7 ◦C and used as the hosts to propagate bacteriophages. The bac-eriophage stock was prepared by plate lysis and elution. Theacteriophages suspended in the � buffer were centrifuged at000 rpm for 10 min and filtered through a 0.22-�m nylon filter.he final concentration was 3.95 × 1010 PFU/mL, determined by alaque assay.

.3. TEM

The TEM images were obtained using a FEI Tecnai-12 (120 kV)ystem. A volume of 2 �L of CBM-T4 stock solution was adsorbed

ids: (a) and (b): CBM-T4 are indicated by arrows; (c)–(f): CBM-T4 phages observed

on a glow-discharged, carbon-coated grid for 30 s and stained with2% uranyl acetate for 1 min.

2.4. SEM

The SEM images were obtained using a Hitachi S4700 Field Emis-sion system. A volume of 10 �L of the CBM-T4 stock solution wasincubated overnight on a cellulose film. The cellulose film on a sili-con wafer was prepared by spin coating [7]. The incubated film waswashed several times with water and blotted dry with a filter paper.The same volume of 100 �L of CBM-T4 solution was adsorbed ona silicon wafer and air-dried without wash and was used as a con-trol. Both surfaces were coated with gold–palladium for 2 min witha sputter coater before observation.

2.5. Bacteriophage viability test

The viability of the CBM-T4 phages after attachment to the filmwas evaluated by a plaque forming method. A 100 �L CBM-T4 sus-pension was incubated for an hour on a cellulose film that was spincoated on a glass slide. After washing several times with water, thesample surface was blotted dry with a filter paper and placed inthe center of a disposable Petri dish. A volume of 5 mL of top agarwas boiled for 2 min and kept in a water bath at 50 ◦C. A volume of100 �L of host E. coli suspension was pipetted into the molten topagar and vortexed for a few seconds. The mixture was immediately

poured onto the top of the sample surface and left to dry for 5 min.The plates were then covered and incubated in an oven over nightat 37 ◦C. A glass slide was used as a control. It was incubated withthe CBM-T4 phages for the same time and treated with the sameprocedure for the viability test.

532 Z. Li et al. / Colloids and Surfaces B: Biointerfaces 76 (2010) 529–534

Fig. 3. Scanning electron micrographs showing attached CBM-T4 phages and unassembled protein complexes: (a) non-specific deposition on silicon surface (b) Specificdeposition onto cellulose film (white dots likely are unassembled protein complexes containing CBMs).

(a) a g

2

oliT4n

3

cTwptpp

sCwlcawm

Fig. 4. Plaques formed by CBM-T4 on

.6. Evanescent wave light scattering experiments

The evanescent wave light scattering experiments were carriedut in an impinging jet apparatus [9]. The flow was mainly used toaunch a vacuum inside the impinging jet so that the film was heldn position throughout the whole experimental period. The CBM-4 stock solution was diluted with � buffer to a concentration of.08 × 108 PFU/mL. The solution was filtered through a 0.22-�mylon filter before use.

. Results and discussion

Wild-type T4 or CBM-T4 bacteriophages were grown to con-entrations of about 1010 PFU/mL (plaque forming units per mL).he stock concentration of CBM-T4 was 3.95 × 1010 PFU/mL, whichas determined by the plaque assay. The CBM-T4s form smallerlaques on the bacterial lawn than the wild-type T4s. The reduc-ion in the plaque size which is a result of the foreign proteins oreptides expressed on the T4 capsid, is seen for many modified T4hages [11,12].

The shape and size of a CBM-T4 were studied by TEM. Fig. 2hows the images of CBM-T4 under different magnifications. TheBM-T4 capsid has dimensions around (89 ± 3) by (107 ± 6) nm,hich are comparable to those of the wild-type T4, reported in the

iterature [13]. In Fig. 2a and b, besides the phages, which are indi-ated by arrows, there are many dark grey and black regions thatre evidence of unassembled protein complexes. The phage stockas filtered through a 0.22-�m filter which effectively excludedost cell fragments, such as those consisting of cell walls. How-

lass substrate and (b) a cellulose film.

ever, this purification method was not sufficient to remove smallbiomolecules, many of which contain CBMs, DNAs, etc., whose sizesare in the nanometer range.

In order to demonstrate the attachment of CBM-T4 phages ontothe cellulose film, a comparative study was conducted on CBM-T4phages deposited onto a cellulose film and on a silicon surface. Bothsubstrates were incubated with CBM-T4 phages overnight, so thatthe adhesion of phages on the surfaces reached saturation. No washwas done for the silicon surface, because the weak electrostaticsurface/particle interactions, such a wash could remove particlesfrom the surface. The cellulose film was washed several times withwater, as the specific interactions of CBM/cellulose should be strongenough to keep the phages immobilized. Both surfaces were imagedusing SEM (Fig. 3). In Fig. 3a, a few white dots, with diameters rang-ing from 31 to 81 nm, are observed in crystallized salt. Comparingthese to the typical size of T4 bacteriophages [13], they should beCBM-T4 phages and unassembled protein complexes of smallersizes, which were embedded in the salt when the salt was crys-tallized during the drying process. In Fig. 3b, a large number of welldistributed white dots were seen on the cellulose film. Their den-sity is 1.25 × 1014 m−2. As the film has been washed several times,it is reasonable to infer the presence of CBMs in the white dots, sothat the pieces could be readily retained on the film. However, thesizes of these white dots range from 23 to 46 nm in diameter which

is smaller than observed in Fig. 3a. This indicates the absence ofT4 bacteriophage within the observation area and hence the whitedots are probably unassembled protein complexes that may con-tain 30–100 copies of proteins (as determined from their size). Thisexperiment alone tells us little about the attachment of CBM-T4

Z. Li et al. / Colloids and Surfaces B: Biointerfaces 76 (2010) 529–534 533

Fd

bi

stmitpct

wFtdas

ttiptfisitt[(

(afa

case they may not be infective as they do not have free fibers to

TE

ig. 5. The dependence of the intensity of scattered light by the CBM-T4, I on theiffusion time, t.

acteriophages to cellulose as the observation area for SEM imagings small (3.2 �m2).

A viability test was performed to further determine the adhe-ion of CBM-T4 bacteriophages on the cellulose film and also assessheir infectivity after attachment. Fig. 4a shows the control experi-

ent in which very few plaques were formed by the CBM-T4 phagesncubated on the glass slides. Only a few were seen at the edges dueo the blotting effect of the filter paper. In Fig. 4b, a large number oflaques were observed, indicating the presence of CBM-T4 on theellulose film. A number of them were obviously still infective afterhe incubation and many washes.

The kinetics of the CBM-T4 phages attaching onto a cellulose filmere studied by the evanescent wave light scattering technique.

ig. 5 shows the dependence of the intensity of scattered light onhe diffusion time. In the initial 10–15 min, the intensity increasesramatically with the diffusion time. It slowed down at a later stagend reached its plateau around 2.6 h. In comparison, the intensity ofcattered light remained almost constant for the � buffer solution.

Fig. 6 shows the increase in intensity of scattered light as a func-ion of the square root of the diffusion time, t1/2. In the initial stage,he intensity changes linearly with t1/2. From the TEM images, SEMmages and the viability test, we know there are both unassembledrotein complexes and CBM-T4 bacteriophages in the stock solu-ion. Both species can readily diffuse and attach onto the celluloselm in the first 10 min (dark dots) as there are plenty of bindingites available. Therefore, the initial kinetics of CBM-T4 depositions expected to be nearly independent of the concentration of pro-eins in the solution. The competitive deposition of a mixture ofwo species has been described before by van de Ven and Kelemen14]. The initial diffusion kinetics should follow the diffusion Eq.1).

With the increase in diffusion time, the observation area

440 × 590 �m) was covered with more and more CBM-T4 phagesnd proteins. As the protein complexes are smaller in size, they dif-use much faster than the phages and so probably coated a largerrea during the same diffusion time. The diffusion slowed down

able 1stimated number density of unassembled protein complexes and CBM-T4 on the cellulo

Surface density, ns (m−2) Bul

Partially assembled protein 1.25 × 1014 5.1CBM-T4 5.9 × 1010 4.1

Fig. 6. Dependence of the intensity of scattered light by the CBM-T4 on the squareroot of the diffusion time. The circles represent the scattered light intensities in theinitial times while the black dots are the ones at later times. The vertical dotted lineindicates the blocking time, tbl.

after 10 min as the binding sites for CBM-T4 were scarce due tothe blocking effect, mostly from the protein coating. The kineticcurve finally reaches a saturation equilibrium when the surface wascovered with a monolayer of the two bioparticles. The diffusioncoefficient of an average-size partially assembled protein with asize determined from Fig. 3b is 1.4 × 10−11 m2 s−1 and for the CBM-T4 it is around 4.9 × 10−12 m2 s−1, assuming the phage diffuses as a100-nm spherical particle. The blocking time, tbl, required for themto reach a steady state is about 58 min, indicated by the dotted linein the graph. If the deposition efficiency ˛ for both species is 1, wecan calculate from Eq. (1) the number density of the unassembledprotein complexes in the bulk suspension from its surface density(1.25 × 1014 m−2, determined from Fig. 3b) and similarly the surfacedensity of CBM-T4 from its concentration of the bulk suspension(4.08 × 1014 m−3). The results are summarized in Table 1.

The results indicate that, at the saturation, the CBM-T4 phageswere largely outnumbered by the unassembled protein complexesin the solution and also on the cellulose surface. From the results inTable 1, we expect one attached phage in an area of about 17 �m2

and, thus, within the whole light scattering area we should expect amaximum number of about 15,000 CBM-T4 phages. This is consis-tent with our observations of both SEMs in Fig. 3 and TEMs in Fig. 2.We may not be able to see any phage in a SEM observation areaof 3.2 �m2 and in most TEM observation areas of about 12 �m2,typically only 1 phage was observed, surrounded by many proteincomplexes.

The deposition efficiency is assumed to be 1 in our model. Thisis a likely scenario for both CBM-T4s and unassembled proteincomplexes in a buffer solution containing a high concentrationof electrolytes. However, although the head-on orientation ofCBM-T4s is preferred due to many copies of high affinity bindingmodules, the phages can still land on their sides or fibers, in which

contact and bind a host E. coli. Thus the number of infective phagesper unit area should be less than our prediction.

It is likely that during lysis of the bacteria not all phages werefully assembled. The number of unassembled protein complexes

se surface at saturation and in the bulk suspension.

k suspension density, nb (m−3) ns(CBM-T4)ns(protein)

nb(CBM-T4)nb(protein)

× 1017 0.47 × 10−4 0.8 × 10−3

× 1014

5 s B: Bi

pbtnnpwatFtEppb

4

bcttsCopIbaClpfbbi

[

[

[

[microscopy analysis of bacteriophages fKZ and T4, J. Electron Microsc. 50 (2001)

34 Z. Li et al. / Colloids and Surface

er phage in the solution (about 1200, Table 1), seems very large,ecause, assuming about 30 proteins per complex, this is equivalento about 10 unassembled phages per phage (considering the totalumber of proteins per phage (3872 copies [15]). A more likely sce-ario is that many more smaller proteins and DNA fragments wereresent, which were mainly responsible for the blocking and whichere below SEM resolution. These smaller molecules also diffused

nd occupied binding sites on the cellulose surface; without themhe blocking time would be much longer than the one shown inig. 6 (58 min) and the resultant concentration of unassembled pro-ein complexes in the suspension will be much lower according toq. (1). Whether blocking is due to small molecules or unassembledrotein complexes does not affect the estimate of the number ofhages on the surface, as Eq. (1) is expected to hold before blockingecomes extensive.

. Concluding remarks

The SEM images and the viability tests show that the CBM-T4acteriophages in the presence of unassembled protein complexesan still specifically bind to cellulose films and retain their infec-ivity after attachment. With the evanescent wave light scatteringechnique, it is possible to obtain quantitative kinetic informationuch as the blocking time and estimates of the concentrations ofBM-T4 or unassembled protein complexes in the solution andn the surface. With this knowledge, one can work with CBM-hages more efficiently by using the minimum incubation time.

t also allows prediction of the maximum bacterial killing capa-ility of each CBM-T4 sample film, assuming all bacteriophagesre infective. If CBM-T4 bacteriophages with the accompanyingBM containing protein complexes were to be applied to cellu-

ose fibers in paper, one would expect several thousand phages

er fiber, more than sufficient to capture and deactivate E. coli. Inact the deposition of CBM containing protein complexes may beeneficial as the protein coating helps to prevent the non-specificinding of pathogens on the bioactive paper and thereby improve

ts efficiency.

[

[

ointerfaces 76 (2010) 529–534

Acknowledgements

The authors would like to acknowledge the financial supportof SENTINEL, a NSERC Strategic Research Network on “BioactivePaper”.

References

[1] W.C. Summers, Bacteriophage therapy, Annu. Rev. Microbiol. 55 (2001) 437.[2] S.M. Grath, D. Sinderen, Bacteriophage: Genetics and Molecular Biology, Hori-

zon Scientific Press, 2007, p. 3.[3] P.A. Barrow, The use of bacteriophages for treatment and prevention of bacterial

disease in animals and animal models of human infection, J. Chem. Technol.Biotechnol. 76 (2001) 677.

[4] L. Gervais, M. Gel, B. Allain, M. Tolba, L. Brovko, M. Zourb, R. Mandeville, M.Griffiths, S. Evoy, Immobilization of biotinylated bacteriophages on biosensorsurfaces”, Sens. Actuators B 125 (2007) 615.

[5] A.B. Boraston, A.L. Creagh, M.M. Alam, J.M. Kormos, P. Tomme, C.A. Haynes,R.A.J. Warren, D.G. Kilburn, Binding specificity and thermodynamics of a fam-ily 9 carbohydrate-binding module from Thermotoga maritima Xylanase 10A,Biochemistry 40 (2001) 6240.

[6] W. Sun, L. Brovko, M. Griffiths, Use of bioluminescent Salmonella for assess-ing the efficiency of constructed phage-based biosorbent, J. Ind. Microbiol.Biotechnol. 27 (2001) 128.

[7] Z. Li, Deposition of viruses on model cellulose surfaces, M.Sc. thesis, Dept. ofChemistry, McGill University, Montreal, QC, Canada, 2008.

[8] J. Sambrook, D.W. Russell, The Condensed Protocols from Molecular Cloning: ALaboratory Manual, CSHL Press, 2006.

[9] Z. Xia, T.G.M. van de Ven, Adhesion kinetics of phosphatidylcholine liposomesby evanescent wave light scattering, Langmuir 8 (1992) 2938.

10] Z. Adamczyk, Particles at Interfaces: Interaction, Deposition, Structure, Elsevier,2006, p. 474.

11] T. Homyk, A. Rodriguez Jr., J. Weil, Characterization of T4 mutants that partiallysuppress the inability of T4rII to grow in lambda lysogens, Genetics 83 (1976)477.

12] N. Malys, V. Klausa, R. Vaigkunait, E. Gineikiene, Post-transcriptional con-trol of bacteriophage T4 gene 25 expression: mRNA secondary structure thatenhances translational initiation, J. Mol. Biol. 288 (1999) 291.

13] N. Matsko, D. Klinov, A. Manykin, V. Demin, S. Klimenko, Atomic force

417.14] T.G.M. van de Ven, S.J. Kelemen, Characterizing polymers with an impinging

jet, J. Colloid Interface Sci. 181 (1996) 118.15] P.G. Leiman, S. Kanamaru, F. Arisaka, M.G. Rossmann, Structure and morpho-

genesis of bacteriophage T4, Cell. Mol. Life Sci. 60 (2003) 2356.