A

powbaas©

K

1

ufm(mnhstmceplws

f

l

0d

Available online at www.sciencedirect.com

BioSystems 92 (2008) 69–75

Covalent attachment of actin filaments to Tween 80 coatedpolystyrene beads for cargo transportation

Harsimran Kaur ∗, Tapan Das, Rajesh Kumar, Ram Ajore, Lalit M. BharadwajBiomolecular Electronics and Nanotechnology Division (BEND), Central Scientific Instruments Organization (CSIO), Sector-30C, Chandigarh, India

Received 19 July 2007; received in revised form 3 December 2007; accepted 14 December 2007

bstract

In this manuscript, a new strategy has been reported for circumscribed covalent attachment of barbed and pointed ends of actin filaments toolystyrene beads. A comparative study of attachment of actin filaments to polystyrene beads was performed by blocking functionally active sitesn polystyrene beads with nonionic detergents such as Tween 20, Tween 80 and polyethylene glycol (PEG). Effective blocking of active sitesas obtained with Tween 80 at 0.1% concentration. Attachment of single bundle of actin filament to bead was assessed by rotational motion ofead tailed actin in front and lateral view. Velocity of actin filaments attached to different size of beads in in-vitro motility assay was calculated to

scertain their attachments. Velocity of actin attached to 1.0 and 3.0 �m polystyrene beads was reduced to 3.0–4.0 and 0.0–1.0 �m/s, respectivelys compared to free actin velocity of 4.0–6.0 �m/s. Single point attachment of actin filaments to different size of beads was assessed by decrease inliding velocity. Present study provides insight into the actin–myosin based molecular motor systems for drug delivery and biosensors applications.

2007 Elsevier Ireland Ltd. All rights reserved.

tro m

awipifi

sfiwaaagr

eywords: Molecular motors; Covalent attachment; Cargo transportation; In vi

. Introduction

Bio-molecular motors are well-established nanoscale molec-lar machines present in living systems. These are responsibleor various dynamical processes for transporting singleolecules over small distances to cell movement and growth

Mansson et al., 2005; Korn et al., 1991). Actin–myosin andicrotubule-kinesin systems have been extensively studied for

anoscale transport systems (Vale, 2003). These motor systemsarness the energy of adenosine triphosphate (ATP) hydroly-is to convert chemical energy into linear motion for nanoscaleransportation (Langford, 1995; Schliwa and Woehlke, 2003). To

imic these bio-nanomachines and explore their capability forarrying motes inside the living system, attachment of cargoesither to actin filaments or by coating cargoes with myosin is there-requisite (Mansson et al., 2004; Miyata et al., 1994). Cova-

ent attachment of actin filaments to beads had been reportedhere barbed ends of actin filaments were capped with gel-

olin and the attachment attained was end specific (Suzuki et

∗ Corresponding author. Tel.: +91 172 2656285/7811x482/452;ax: +91 172 2657267.

E-mail addresses: microsimbac@yahoo.co.in (H. Kaur),alitmbharadwaj@hotmail.com (L.M. Bharadwaj).

2

BNbBuw

303-2647/$ – see front matter © 2007 Elsevier Ireland Ltd. All rights reserved.oi:10.1016/j.biosystems.2007.12.003

otility assay; Drug delivery

l., 1996). Similar study of attachment of myosin to cargoesas reported earlier (Kull et al., 1998). Hence, only barbed (pos-

tive) end of actin filament will attach to cargo. Both barbed andointed ends of actin filaments attached to beads may providensights for cargo transportation and mechanical studies of actinlaments.

This experiment has been designed to investigate the circum-cribed covalent attachment of barbed and pointed ends of actinlaments to polystyrene beads. Carboxylated polystyrene beadsere treated with nonionic detergents viz. Tween 20, Tween 80,

nd polyethylene glycol (PEG) to reduce the number of function-lly active sites on beads. We focused on circumscribed covalentttachment of polystyrene beads coated with nonionic deter-ents to actin filaments. This was ascertained by fluorescence,otational motion and load carrying capacity studies.

. Materials and methods

Actin, myosin of rabbit skeletal muscle, phalloidin–tetramethylrhodamine-

–isothio cyanate (Rh–Ph; fluorescent dye) was procured from Sigma–Aldrich.-hydroxysuccinimide (NHS), EDC [1-ethyl-3-(3-dimethylaminopropyl) car-odiimide HCl] and Tween 80, Tween 20 were procured from Pierce, USA andioBasics Inc., Canada, respectively. All other chemicals used were of molec-lar and analytical grade unless otherwise stated. All the buffers and solutionsere prepared in double distilled deionised (18 M�) autoclaved water.

7 ystem

2

0p15twiwT

2

twrsspwMT3iFgw(

2

smf

bbgswrwsbtfa

3

3

Lt(ao(wmrTnT

Fo

0 H. Kaur et al. / BioS

.1. Polymerization and labeling of actin

G-actin was polymerized in polymerizing buffer (2 mM MgCl2, 50 mM KCl,.25 mM ATP, 0.5 mM DTT, 1 mM EGTA, 0.1 mM CaCl2, 10 mM imidazole,H 7.5) prepared in AB buffer (25 mM imidazole, 25 mM KCl, 1 mM DTT,mM EGTA, 4 mM MgCl2). Twenty-five micrograms of actin was dissolved in0 �l of polymerization buffer and incubated for a period of around 4 h at roomemperature (RT). Five hundred microliters stock solution of Rh–Ph (10 �g/�l)as prepared in methanol. Ten microliters of Rh–Ph (10 �g/�l) was transferred

n a vial and dried with dry nitrogen stream till the pellet was formed. The pelletas dissolved in 2 �l of ethanol, 290 �l AB buffer and 10 �l of F-actin, stepwise.his was incubated for a time period of 12–16 h at 4 ◦C.

.2. Covalent attachment of polystyrene beads to actin filaments

Fifteen microliters of beads (2.6% carboxylated polystyrene beads, diame-er 1 and 3 �m) suspension was placed in a vial of 1.5 ml. Beads were washedith 0.1 M carbonate buffer (0.1 M Na2CO3 and 0.1 M NaHCO3; pH 9.6) and

epeated thrice with 0.2 M MES buffer (pH 5.3) to remove bead suspensionolvent. Beads were resuspended in 42 �l of MES buffer. An aliquot of 125 �lolution [EDC (2%) + NHS (1 mg/ml)] was added dropwise to the bead sus-ension and mixed continuously by shaking at 300 rpm for 3–4 h at RT. Beadsere recollected by centrifugating at 10,000 rpm for 10 min. It was washed withES buffer followed by resuspension of beads in 80 �l of borate buffer (0.2 M).

ween 80 (0.1% in PBS) was added to the suspension and mixed by shaking at00 rpm and was incubated for 3 h at RT. Beads were recollected by centrifugat-ng at 10,000 rpm and were resuspended in 80 �l of borate buffer. Rh–Ph labeled-actin (25 �l) of 0.2 �g/�l was added to the bead suspension and was mixedently by shaking at 80 rpm for 12–14 h at RT in dark. Beads were washed twiceith 0.1% Tween 80 solution and were resuspended in 200 �l of storage buffer

0.01 M phosphate buffer, 1% BSA, 0.1% sodium azide, 5% glycerol).

.3. In vitro motility assay

The microreaction cell was prepared on 0.1% nitrocellulose coated thin glasslide (25 mm × 75 mm × 0.14 mm) (Bagga et al., 2005). Hundred microliters ofyosin (40 �g/ml in AB buffer containing 0.6 M KCl) was infused and incubated

or 60 s. Hundred microliters of AB/BSA solution (0.5 mg/ml of BSA in AB

fimdf

ig. 1. Schematic representation of reaction of blocked and unblocked active sites ofn beads laid down platform for formation of network of beads and actin filaments.

s 92 (2008) 69–75

uffer) was infused followed by 100 �l of bead suspension with actin after incu-ation for 60 s. Hundred microliters of AB/BSA, AB/BSA/GOC (0.3 mg/ml oflucose, 0.1 mg/ml of glucose oxidase and 0.018 mg/ml of catalase in AB/BSAolution), and AB/BSA/GOC/ATP (2 mM of ATP in AB/BSA/GOC solution)as infused subsequently and movement of bead tailed actin filaments was

ecorded on Axiovert 200, USA (Bagga et al., 2005). Labeled actin filamentsith attached polystyrene beads were observed in inverted fluorescence micro-

cope. Fluorescence images with resolution of 320 × 240 pixels were recordedy ‘Camatasia Recorder’ at 15 frames per seconds. Irfanview software was usedo extracts the image frames from the recorded motility of actin filaments in aviormat. Images were processed for calculating velocity of individual bead tailedctin filaments by custom built programme in Matlab software.

. Result and discussion

.1. Covalent attachment of beads to actin filaments

G-actin was polymerized to F-actin in polymerization buffer.ength of F-actin filaments as measured were found to be in

he range of 1.5–5 �m. Blocking of functionally active sites–COOH) on polystyrene beads was done to decrease the prob-bility of attachment of number of actin filaments, in absencef which a network of actin filaments and beads will be formedFig. 1). Attachment of single bundle of actin filament to beadill result in directional movement on myosin heads, whereas,ultiple bundles of actin filaments attached to single bead may

esult in multidirectional movement. Nonionic detergents viz.ween 20, Tween 80 and PEG were used to block significantumber of active sites. Effective blocking was achieved withween 80 as we succeeded in preparing single bundle of actin

lament tailed bead (Fig. 2). Fluorescence image of actin fila-ents attached to 3 �m bead is shown in Fig. 2A. Length and

iameter were calculated from superimposed images and wereound to be 3.107 and 3.059 �m, respectively in case of 3 �m

polystyrene beads with single bundle of actin filaments. Unblocked active sites

H. Kaur et al. / BioSystems 92 (2008) 69–75 71

F m ber

boa

attoHoRioibdfwd8p

M(a

AbtThtb

atiob

Fl

ig. 2. Attached actin filaments to Tween 80 coated polystyrene bead. (A–C) 3 �

epresent fluorescent image, grey scale and threshold image.

ead (Fig. 2B and C); 1.810 and 1.064 �m, respectively in casef 1 �m bead (Fig. 2D–F). These results were extracted by imagenalysis.

Other than nonionic detergents BSA, a blocking agent, waslso used to block active sites on polystyrene beads. It was foundo be ineffective as high fluorescence was observed in BSAreated beads (Fig. 3A). High fluorescence underlies attachmentf numerous actin filaments to beads due to unblocked sites.ere fluorescence is correlated with Rh–Ph labeled actin. Flu-rescence observed in Fig. 3A is not due to BSA stained withh–Ph as the same concentration of Rh–Ph was used with PEG

n Fig. 3B and similar results were observed. Clumps of beadsbserved with PEG and BSA are attributed to negligible block-ng of active sites as shown in schematic Fig. 1. Polystyreneeads active site blocking with Tween 20 and Tween 80 showsispersed beads rather than clumped (Fig. 3C and D). Despite theact that beads were dispersed with Tween 20, high fluorescence

as observed suggesting excessive attachment of actin filamentsue to less blocking. More blocking was achieved with Tween0, hence, the numbers of actin filaments attached to beads wererofoundly less as clear from weak fluorescent image (Fig. 3D).

(coa

ig. 3. Polystyrene beads (1 �m) treated with blocking agents. (A) BSA, (B) PEG, (arge number of active sites and better dispersion of beads as compared to BSA and P

ad represent fluorescent image, grey scale and threshold image and (D–F) 1 �m

uch improved dispersion was also observed with Tween 80Fig. 3D). Effective dispersion with Tween 80 is a measure ofctive site blocking on beads.

Tween 80, a nonionic detergent effectively blocks active sites.couple of unblocked active sites were attached with single

undle of actin filaments (Fig. 3D). Weak fluorescence and bet-er dispersion with Tween 80 (Fig. 3D) clearly suggest thatween 80 is an effective blocking agent. Tween 80 treatmentas resulted in blocking of larger number of active sites on beadhus probability of achieving single bundle of actin filaments oneads have increased.

Three different concentrations of Tween 80 viz. 0.05, 0.1,nd 0.5% was set to optimize the percentage concentrationo be used for further experiments. Numerous actin filamentsn bundles were attached to beads and high fluorescence wasbserved while active site (–COOH) blocking of polystyreneeads was attempted at 0.05% concentration of Tween 80

Fig. 4A). Actin filaments in bundles were observed at 0.5%oncentration of Tween 80. This result is attributed to blockingf all the active sites on the bead. Hence, no attachment of beadnd actin was observed (Fig. 4B). Both end of actin filaments

C) Tween 20 and (D) Tween 80. Treatment with Tween 80 shows blocking ofEG.

72 H. Kaur et al. / BioSystems 92 (2008) 69–75

feren

caotT

bbt

isaoa

Fr

Fig. 4. Fluorescence images of polystyrene beads treated with dif

ovalently attached to polystyrene beads were observed atconcentration of 0.1% as shown in Fig. 4C. This suggests

ptimum concentration in the present case is 0.1% to blockhe active sites. This result also suggests that attachment withween 80 is not end specific.

Attachment of single bundle of actin filament to polystyreneead at a concentration of 0.1% of Tween 80 was again affirmedy rotational motion of bead tailed actin. The rotation of beadailed actin was observed in the absence of myosin and ATP

vFfb

ig. 5. Fluorescence images of bead tailed actin filaments with rotational motion. Pootation of the bead whole of actin filaments was visible. The diameter of beads was

t concentration of Tween 80. (A) 0.05%, (B) 0.5% and (C) 0.1%.

n storage buffer as shown in Fig. 5. Front view of bead tailedingle bundle of actin filament is shown in Fig. 5A–D where,ctin filament is seen as a fluorescent dot on bead. Lateral viewf bead tailed single bundle of actin filament has emerged asresult of its rotation in the buffer. Actin filament in lateral

iew has appeared as a small fluorescence streak (Fig. 5E–I).urther rotation of bead has resulted in a view as shown inront view suggesting single bundle of actin is attached toead.

int of attachment of actin filaments to polystyrene bead was observed and with3 �m.

H. Kaur et al. / BioSystems 92 (2008) 69–75 73

le be

3

csnsraenUobpTbn

omwabitmitrwpp

Fp

Fig. 6. Schematic representation of movement of sing

.2. In vitro motility assay

Sliding velocity measurement is another attempt to confirmircumscribe covalent attachment of actin to beads. The rea-on behind sliding velocity measurement is that if we haveot been succeeded in blocking significant number of activeites on the beads, intuitively attachment of actin to bead willemain multiple points. As a result of this a network of beadsnd actin molecules will be formed (Fig. 6). The force gen-rated by hydrolysis of ATP is very less, with this force, it isot possible to move such a huge network of actin and bead.nder these circumstances either the velocity will remain morer less similar for different size beads or resultant velocity wille negligible (Fig. 6). A significant difference in velocity isossible only when single bundle of actin is attached to bead.

o delineate the clear differences in velocities of different sizeeads it is expected actin attached to beads should be limited inumbers.

Faw

ig. 7. Histogram for velocity of bead tailed actin filaments calculated by processingolystyrene bead (B).

ad tailed actin and actin network over myosin heads.

Hence, the velocity measurement was performed for beadsf two different sizes, i.e. 1 and 3 �m so that the difference inagnitude of velocity can be ascertained due to effect of theireight only. This difference in velocities will again ensure the

ttachment of bead with actin filaments. In vitro motility assay ofead tailed actin filaments was performed to calculated veloc-ty of actin filaments sliding over myosin heads. Velocity ofhe motile bead tailed complex in each frame of the motility

ovie was calculated and was plotted in histogram as shownn Figs. 7A and 8A. The average velocity calculated was foundo be 3.29 and 0.933 �m/s for 1 and 3 �m polystyrene beads,espectively. Direction of motile actin filaments attached to beadas tracked by determining the centroid of actin-bead com-lex in each frame. After superimposing consecutive frames theath followed by the actin filaments was obtained as shown in

igs. 7B and 8B. It was found that average velocity of actin fil-ments was reduced about 3.5 times when, the load (bead size)as increased by 27 times by volume.

each frame in Matlab (A). The track of motile actin filaments attached to 1 �m

74 H. Kaur et al. / BioSystems 92 (2008) 69–75

F ssingp

4

mtAoappSa

abT8rPAf(bFwTwaTtw

bbaa

oih

wafawuoSsoavF

5

mbTcs0csoaw

ig. 8. Histogram for velocity of bead tailed actin filaments calculated by proceolystyrene bead (B).

. Discussion

Development of nanorobotic devices based on molecularotors (actin–myosin) is long anticipated. Attachment of actin

o limited cargoes is the pre-requisite to realize such devices.ttachment of actin filaments to beads after capping barbed endsf actin was reported earlier (Suzuki et al., 1996). The cappinggent used, gelsolin, was a costlier one. Similar attachment ofolystyrene beads to actin filaments was also reported whereolylysine was used to cap pointed end of actin (Brown andpudich, 1979). Both the earlier reports are restricted to one endttachment of actin to bead.

In the present study, circumscribed covalent attachment ofctin filaments to carboxylated polystyrene beads was attainedy blocking the significant number of active sites (–COOH) withween 80 at a concentration of 0.1%. The blocking agent (Tween0) used is the cheapest and attachment of actin to beads is notestricted to one end only. Four different agents such as BSA,EG, Tween 20 and Tween 80 were used to block active sites.ctive site blocking with BSA and PEG was found to be inef-

ective as high fluorescence and clumped images were observedFig. 3). High fluorescence and clumps of actin filaments andeads are due to presence of unblocked active sites as shown inig. 1. Weak fluorescence and dispersed images were observedhile blocking of active sites was attempted with Tween 20 andween 80 (Fig. 3B and C). Although bead tailed actin complexesere dispersed with Tween 20 (Fig. 3B), much better dispersion

nd single bundle of actin attached to beads were observed withween 80 (Fig. 3C). Thereafter, a set of Tween 80 concentra-

ions, i.e. 0.05, 0.1, and 0.5% were tried and effective blockingas observed at 0.1% concentration (Fig. 4C).Attachment of single bundle of actin filament to polystyrene

ead at a concentration of 0.1% of Tween 80 was again affirmedy rotational motion of bead tailed actin. Actin attached to beadppeared as fluorescent dot in front view, whereas it appeareds fluorescent streak in its lateral view with the gradual rotation

tmSt

each frame in Matlab (A). The track of motile actin filaments attached to 3 �m

f bead. Such view is only possible when single bundle of actins attached to bead otherwise number of fluorescent streaks mayave appeared.

Sliding velocity of single bundle of actin filaments attachedith polystyrene beads over myosin heads was calculated. The

verage velocity calculated was found to be 3.29 and 0.933 �m/sor 1 and 3 �m polystyrene beads, respectively. It was found thatverage velocity of actin filaments was reduced about 3.5 timeshen, the load (bead size) was increased by 27 times by vol-me. Maximum velocity of single bundle of actin filament onlyn myosin heads was calculated and found to be 4.0–6.0 �m/s.uch a measurable difference in velocity due to change in beadize clearly indicates attachment of beads with single bundlef actin filaments. If there have been multiple attachments ofctin filaments to beads, there would have been no possibleelocity due to formation of actin-bead network as shown inig. 6.

. Conclusion

In the present work, a simple and cheapest method of attach-ent of cargos (beads) to single bundle of actin filaments has

een reported using Tween 80 as active site blocking agent.ween 80 was found to be more effective blocking agent asompared to Tween 20, BSA and PEG. Beads attached to bothides of the (barbed and pointed) ends of actin were observed at.1% concentration of Tween 80. Fluorescent dot and fluores-ent streak observation of front and later view, respectively hasuggested attachment of single bundle of actin with bead. Thebservable difference in velocity measured to bead tailed actinlso support the attachment of single bundle of actin filamentith bead. Both side attachments can be useful for calculating

he force on the actin filaments. It is also useful for calculatingechanical properties like tensile strength of actin filaments.ingle side attached beads can be exploited for cargo transporta-

ion for realizing nanorobotic devices in future.

ystem

A

TMAvsG(

R

B

B

K

K

L

M

M

M

S

H. Kaur et al. / BioS

cknowledgements

This work was supported by Department of Science andechnology (DST), Govt. of India. We are also thankful tor. A.K. Shukla, Ashwani Sharma, Dr. Inderpreet Kaur, Dr.mit Sharma, Deepak Kukkar and Rupinder Kaur for theiraluable guidance and suggestions. One of our authors (Har-imran Kaur) thanks University Grant Commission (UGC),ovt. of India for providing junior Research fellowship

JRF).

eferences

agga, E., Kumari, S., Kumar, R., Kumar, R., Bajpai, R.P., Bharadwaj, L.M.,2005. Covalent immobilization of myosin for in-vitro motility of actin.Pramana 65, 967–972.

rown, S.S., Spudich, J.A., 1979. Nucleation of polar actin fila-

ment assembly by a positively charged surface. J. Cell Biol. 80,499–504.

orn, S.J., Toyoshima, Y.Y., Udeya, T.Q.P., Spudich, J.A., 1991. Assays forActin sliding movement over myosin coated surfaces. Methods Enzymol.196, 399–416.

S

V

s 92 (2008) 69–75 75

ull, F.J., Vale, R.D., Fletterick, R.J., 1998. The case for a common ancestor:kinesin and myosin motor proteins and G proteins. J Muscle Res. Cell Motil.19, 877–886.

angford, G.M., 1995. Actin- and microtubule-dependent organelle motors:interrelationships between the two motility systems. Curr. Opin. Cell Biol.7, 82–88.

ansson, A., Sundberg, M., Bunk, R., Balaz, M., Nicholls, I.A., Omling, P.,Tegenfeldt, J.O., Tagerud, S., Montelius, L., 2005. Actin-based molecularmotors for cargo transportation in nanotechnology—potentials and chal-lenges. IEEE Trans. Adv. Pack. 28, 547–554.

ansson, A., Sundberg, M., Bunk, R., Balaz, M., Rosengren, J.P., Lindahl, J.,Nicholls, I., Omling, P., Tagerud, S., Montelius, L., 2004. Nanotechnologyand actomyosin motility in vitro on different surface chemistries. Biophys.J. 86, 58a.

iyata, H., Hakozaki, H., Yoshikawa, H., Suzuki, N., Kinosita, K., Nishizaka,T., Ishiwata, S., 1994. Stepwise motion of an actin filament over a smallnumber of heavy meromyosin molecules is revealed in an in vitro motilityassay. J. Biochem. 115, 644–647.

chliwa, M., Woehlke, G., 2003. Molecular motors. Nature 422, 759–765.

uzuki, N., Miyata, H., Ishiwata, S., Kinosita, K., 1996. Preparation of bead-

tailed actin filament: estimation of the torque produced by sliding force inan in vitro motility assay. Biophys. J. 70, 401–408.

ale, R.D., 2003. The molecular motor toolbox for intracellular transport. Cell112, 467–480.