Motility of Dimeric Ncd on a Metal-Chelating Surfactant: Evidence That Ncd Is Not Processive

6 pages
56 views
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Description
Motility of Dimeric Ncd on a Metal-Chelating Surfactant: Evidence That Ncd Is Not Processive
Transcript
  Motility of Dimeric Ncd on a Metal-Chelating Surfactant: Evidence That Ncd IsNot Processive † Michael J. deCastro, Chih-Hu Ho, and Russell J. Stewart*  Department of Bioengineering, Uni V  ersity of Utah, Salt Lake City, Utah 84112 Recei V  ed December 10, 1998; Re V  ised Manuscript Recei V  ed February 11, 1999 ABSTRACT : The surface immobilization methods that allowed single-molecule motility experiments withnative kinesin have not worked with the ncd motor protein and other kinesin-related motors. To solvethis problem, a surfactant (Pluronic F108) was chemically modified with the metal-chelating groupnitrilotriacetic acid (NTA) to allow surface immobilization of histidine-tagged microtubule motors. Thechelating surfactant provided a convenient and effective method for immobilization and subsequent motilityexperiments with a dimeric H-tagged ncd protein (H-N195). In experiments with the absorption of H-N195to polystyrene (PS) beads coated with F108-NTA, a monolayer of H-N195 bound in the presence of Ni 2 + , while in the absence of Ni 2 + , the extent of adsorption of H-N195 to PS beads was greatly reduced.In motility experiments with H-N195 immobilized on F108-NTA-coated surfaces, microtubules movedsmoothly and consistently at an average speed of 0.16 ( 0.01  µ m/s in the presence of Ni 2 + , while withoutNi 2 + , no microtubules landed on the F108-NTA-coated surfaces. Investigation of H-N195 motility on theF108-NTA surfaces provided several indications that ncd, unlike kinesin, is not processive. First, a criticalH-N195 surface density for microtubule motility of approximately 250 molecules/   µ m 2 was observed.Second, microtubule landing rates as a function of H-N195 surface density in the presence of MgATPsuggested that several H-N195 molecules must cooperate in microtubule landing. Third, the ATP  K  M  inmotility assays (235  µ M) was substantially higher than the ATP  K  M  of dimeric ncd in solution (23  µ M)[Foster, K. A., Correia, J. J., and Gilbert, S. P. (1998)  J. Biol. Chem. 273 , 35307 - 35318]. The kinesin superfamily is a class of motor proteins thatconvert energy from ATP hydrolysis into movement alongmicrotubules. Since the discovery of the srcinal kinesin ( 1 ),dozens of kinesin-like proteins have been found that areassociated with a wide range of intracellular processes,including vesicle and organelle transport, spindle morpho-genesis, and chromosome movements. Members of thekinesin family are defined by a common motor domain withconserved sequence and structure that contains the ATP andmicrotubule binding sites. Kinesin itself has an N-terminalmotor domain and moves toward the microtubule plus end.Other members of the kinesin family, like ncd, 1 haveC-terminal motor domains and move toward the microtubuleminus end ( 2 ,  3 ).Several key aspects of kinesin’s motility mechanism havebeen relatively well-characterized using in vitro microtubulemotility experiments with surface-immobilized kinesin. Therapid progress was possible, in part, because native kinesincan be nonspecifically adsorbed to glass surfaces in thepresence of carrier proteins, like casein, at very low densitiesof active motors ( 4 ). Low-surface density motility experi-ments demonstrated that native kinesin is highly processive;a single kinesin molecule is able to take hundreds of con-secutive steps before dissociating from its microtubule track.Using nonspecific adsorption procedures, the magnitude of individual translocation events ( 5 ) and the forces generatedby single native kinesin molecules ( 6  - 8  ) were measured inhigh-resolution motility experiments. Other, more specificimmobilization procedures have also been developed forgenetically truncated kinesin proteins. These proceduresinclude immobilization of in vivo biotinylated kinesin ( 9 )and chemically biotinylated kinesin ( 10 ) on streptavidin-coated surfaces, immobilization of kinesin fused with glu-tathione  S  -transferase (GST) on surfaces coated with GSTantibodies ( 11 ), and immobilization of kinesin fused withgreen fluorescent protein (GFP) on surfaces coated with GFPantibodies ( 12 ).Unlike kinesin, few details of ncd’s motility mechanismare known beyond the direction and speed of movement.Even the question of whether ncd is processive, like kinesin,has not been fully resolved. The difficulty has been that thenonspecific adsorption in the presence of carrier proteins thatmade motility possible with single native kinesin moleculeshas not worked well with ncd. Other than the initial reportsof ncd minus end-directed motility using the full-length ncdprotein expressed in  Escherichia coli  ( 2 ,  3 ), the only ncdmotility reported has been with GST and GFP fusion proteins( 11 ,  13 - 15 ). Likewise, the motility mechanisms of severalother kinesin-related proteins, some notoriously reluctant to † This work was supported by a grant from the NSF to R.J.S. (CTS-9624907), an NIH postdoctoral fellowship to C.-H.H. (HL09777), anda predoctoral Whitaker Foundation Fellowship to M.J.d.* To whom correspondence should be addressed: Department of Bioengineering, 20S. 2030E., room 506C, University of Utah, Salt LakeCity, UT 84112-9458. Phone: (801) 581-8581. Fax: (801) 581-8966.E-mail: rstewart@ee.utah.edu. 1 Abbreviations: ncd, nonclaret disjunctional; NTA, nitrilotriaceticacid; Pipes, 1,4-piperazinediethanesulfonic acid; IPTG, isopropyl   - D -thiogalactopyranoside; AMPPNP, adenosine 5 ′ -(   - γ -imido)triphos-phate; PPO, polypropylene oxide; PEO, polyethylene oxide. 5076  Biochemistry  1999,  38,  5076 - 5081 10.1021/bi9829175 CCC: $18.00 © 1999 American Chemical SocietyPublished on Web 04/02/1999  move in vitro, have not been investigated in any detail. Auniversally applicable method of immobilizing active motorswith well-defined densities and with minimal structuralmodification is needed to resolve the mechanical details of motility by ncd and other kinesin-related motors. To meetthis need, we developed a specific immobilization procedureusing a metal-chelating Pluronic surfactant that enabled mo-tility experiments with ncd modified only with six terminalhistidine residues.Pluronics are triblock copolymers of two hydrophilicpolyethylene oxide (PEO) chains connected by a hydrophobicpolypropylene oxide (PPO) chain. The PPO chain of Plu-ronics adsorbs onto hydrophobic surfaces, while the hydro-philic PEO chains extend into the aqueous phase, creating aprotein-repellent interface that prevents nonspecific proteinadsorption and denaturation ( 16  ). To modify Pluronics forimmobilization of H-tagged proteins, Pluronic F108 wasderivatized with the metal-chelating group nitrilotriacetic acid(NTA) on the terminal hydroxyl groups of the PEO chainsto form F108-NTA ( 17  ). Proteins with genetically addedterminal histidine residues have a high affinity for chelatedmetal ions ( 18  ). Combining the protein-repellent propertiesof F108 with the high binding stability of immobilized metalaffinity provides a convenient method for surface immobiliz-ing H-tagged recombinant proteins.Here we report that the chelating F108 Pluronic greatlysuppressed nonspecific adsorption of ncd to hydrophobicsurfaces in the absence of surface-blocking proteins andallowed consistent and reproducible surface immobilizationof H-tagged ncd. Using this immobilization procedure witha structurally well-defined dimeric ncd, the surface densitydependence and the ATP dependence of ncd-driven micro-tubule motility were investigated to address the issue of ncdprocessivity. EXPERIMENTAL PROCEDURES Synthesis of F108-NTA.  The synthesis of F108-NTA wascarried out as previously described ( 17  ). Nitrilotriacetic acid(NTA) groups were coupled to the terminal hydroxyl groupsof the PEO chains of Pluronic F108 (BASF, Inc., MountOlive, NJ), creating a metal-chelating Pluronic (F108-NTA). Construction of pRSET-N195.  To construct the pRSET-N195 plasmid, pBS-NCD ( 19 ) was digested with  Afl II,blunted with Klenow polymerase, and then digested with Kpn I. The resulting fragment was ligated into pRSET B(Invitrogen, Inc., Carlsbad, CA), which had been digestedwith  Bam HI, blunted with Klenow polymerase, and thendigested with  Kpn I. The resulting plasmid, pRSET-N195,contains ncd residues K195 - K685 and a six-histidine residuetag at the N-terminus. The molecular mass of H-N195,calculated from the amino acid sequence, is 57.5 kDa. Theprotein expressed by pRSET-N195, except for the histidinetag, is equivalent to the MC1 protein ( 13 ).  Expression and Purification of H-N195.  The recombinantplasmid pRSET-N195 was transformed into  E. coli  strainBL21(DE3) for expression. Single colonies were selectedand grown overnight in LB media with 50  µ g/mL kanamycin.LB cultures (500 mL) with 50  µ g/mL kanamycin were theninoculated at 1:100 with the overnight cultures and grownat 37  ° C until mid-log phase. The temperature was loweredto 22  ° C, and the cells were induced with 0.1 mM isopropyl   - D -thiogalactopyranoside (IPTG) for 10 - 12 h. Cells werepelleted by centrifugation (Beckman JA 10 rotor, 4500 rpmfor 30 min at 4  ° C) and frozen at - 70  ° C. Frozen cell pelletswere resuspended in 5 mL/g lysis buffer [20 mM Hepes (pH7.0), 150 mM NaCl, 4 mM MgSO 4 , and 1 mM EGTA] and100 mg/mL lysozyme and 5 mM phenylmethanesulfonylfluoride (PMSF) and incubated on ice for 30 min. Lysateswere prepared by four cycles of freezing with liquid N 2  andthawing in a 25  ° C water bath. After two freeze - thaw cycles,10  µ g/mL DNase I was added and the lysates were incubatedon ice for 30 min to reduce the viscosity. After four freeze - thaw cycles, the extract was clarified by centrifugation(Beckman JA 17 rotor, 14 500 rpm for 30 min at 4  ° C). Theclarified supernatant was mixed with 2 mL of Ni-NTAagarose for 30 min at 4  ° C. The column was washed with50 mL of lysis buffer and then washed again with 50 mL of lysis buffer containing 60 mM imidazole. H-N195 was elutedwith 300 mM imidazole. The H-N195 was buffer exchangedinto PEM 80 [80 mM Pipes (pH 7.0), 4 mM MgSO 4 , 1 mMEGTA, and 0.1 mM ATP] using a Sephadex G-25 column.The purity of the H-N195 preparation was determined bySDS - PAGE (Figure 2A). The protein was aliquoted, frozenin liquid N 2 , and stored at - 70  ° C. The concentration of thepurified H-N195 was determined using the Bradford methodwith bovine serum albumin as a standard.  Adsorption of H-N195 onto PS Beads.  F108 and F108-NTA (4% w/v) in phosphate buffer (50 mM, pH 7.8) wereadsorbed onto PS beads (0.453  µ m diameter, 1% w/v)overnight with constant end-over-end mixing at room tem-perature. The coated beads were then washed by centrifuga-tion to remove unbound F108 and stored in PEM 80 at 4 ° C. The F108-NTA beads were charged with Ni 2 + byincubating the beads in 50 mM NiSO 4 . The Ni 2 + -chargedbeads were then washed by centrifugation to remove excessNi 2 + and stored in PEM 80 buffer. The truncated H-taggedH-N195 was incubated with PS beads (0.1% w/w) in PEM80 with end-over-end mixing at 4  ° C. The extent of proteinadsorption was determined by pelleting the beads bycentrifugation, measuring the protein concentration in thesupernatant, and subtracting the supernatant concentrationfrom the initial concentration. H-N195 supernatant concen-trations were determined using a Micro BCA protein assaymethod with bovine serum albumin as a standard.  Equilibrium Sedimentation of H-N195.  H-N195 wasdiluted to a final concentration in PEM 80 containing 0.01mM ATP. The calculated monomer molecular mass forH-N195 based on the amino acid sequence was 57.5 kDa,including the six histidine residues of the H-tag. Sedimenta-tion equilibrium studies were performed in a Beckman XLAanalytical ultracentrifuge using an An-60 Ti rotor. Theconcentration distribution of H-N195 was permitted to cometo equilibrium at three speeds: 12 000, 14 000, and 16 000rpm. Three H-N195 concentrations were analyzed simulta-neously: 1, 0.5, and 0.25 mg/mL. Scans were obtained at280 nm, and four scans were averaged at equilibrium foreach speed. Data analysis was performed using the nonlinearregression program NONLIN ( 20 ), and the goodness of fitwas assessed visually from the residual plots. The Sedimen-tation equilibrium data were best fit to a single ideal speciesmodel with a molecular mass of 108 ( 8 kDa. The calculatedmolecular mass for a dimer was 115 kDa, suggesting thatH-N195 exists as a stable dimer in solution.Ncd Is Not Processive  Biochemistry, Vol. 38, No. 16, 1999  5077   Microtubule Motility and Landing Rate Assays.  Flowchambers (24 mm × 5 mm, 12  µ L volume) were constructedwith a coverslip and a microscope slide separated by twostrips of double-stick tape. The coverslips and microscopeslides were first treated with dimethyldichlorosilane (DDS)to create hydrophobic surfaces. Dry KOH/EtOH-cleanedsurfaces were immersed in 0.05% DDS in trichloroethylene(TCE) for 20 min, washed three times with methanol, washedtwo times with DI water, dried with filtered air, and storeddry. The DDS flow chambers were incubated with 2 mg/ mL F108-NTA and 50 mM Ni 2 + for 5 min and then washedthree times with PEM 80. H-N195 at various concentrationsin PEM 80 with ATP or AMPPNP was introduced into thechamber and incubated for 5 min. Finally, microtubules (0.4  µ M) in PEM 80, 20  µ M taxol, and nucleotide were washedinto the flow chamber. Microtubules on the coverslip surfacewere observed by video-enhanced differential interferencecontrast microscopy. The density of the motor protein onthe DDS coverslip was estimated from the amount of addedH-N195, the chamber volume (12  µ L), and the area of theglass surfaces of the flow chamber. All the H-N195 wasassumed to bind uniformly to the surface of the flowchambers. RESULTS  Immobilization of H-N195 on F108-NTA.  We reportedpreviously that F108-NTA provided a convenient and ef-fective method for immobilizing H-tagged firefly luciferaseon PS beads while retaining high specific activity ( 17  ). Theadsorption of histidine-tagged ncd to PS beads treated withF108 and F108-NTA was similarly examined (Figure 2). Thencd protein used for these experiments, designated H-N195,contains the entire predicted coiled coil of ncd and an amino-terminal histidine tag. Analysis of the molecular mass bysedimentation equilibrium demonstrated that H-N195 isdimeric, as expected, and behaves as a single species (datanot shown). The amount of H-N195 adsorbed onto PS beadswas estimated by subtracting the amount of H-N195 in the F IGURE  1: Immobilization scheme for histidine-tagged motor proteins using NTA-derivatized Pluronic F108. The hydrophobic PPO blockof the Pluronic F108 triblock copolymers interacts with hydrophobic surfaces. The hydrophilic NTA-modified PEO blocks extend intosolution, creating an activity-preserving interface to which histidine-tagged proteins bind through chelated metal ions.F IGURE  2: H-N195    purification and adsorption on polystyrene (PS) beads. (A) Coomassie-stained SDS - PAGE showing the purity of H-N195 prepared as described in Experimental Procedures. (B) Adsorption of H-N195 on unmodified PS beads and Pluronic-treated PSbeads. The dashed line represents a theoretical monolayer of adsorbed N195H on the surface of 0.453  µ m PS beads. The values representthe average of at least three protein adsorption measurements. The amount of protein adsorbed onto F108-NTA-treated PS beads in thepresence of Ni 2 + was close to that of the theoretical monolayer (dashed line). 5078  Biochemistry, Vol. 38, No. 16, 1999  deCastro et al.  supernatant after the beads were pelleted from the totalamount of H-N195 incubated with the PS beads. Theestimated mass of adsorbed H-N195 was compared to thetheoretical mass of a monolayer of H-N195 (Figure 2, dashedline), calculated from the dimensions (70 Å  ×  45 Å  ×  45Å) of the motor domain crystal structure ( 21 ) and the surfacearea of 0.453  µ m diameter PS beads. The model assumedthe dimeric motor was oriented with its heads projectingaway from the bead, as shown in Figure 1, and that the motordomains of the dimer occupied a circular cross-sectional areawith a radius of 45 Å.Untreated PS beads adsorbed 4.33  (  0.32 mg/m 2 of H-N195, significantly greater than that of the predictedmonolayer. The excessive adsorption may have been due totighter packing of randomly oriented, nonspecifically bounddimeric motors, or multiple layers of adsorbed protein. Beadscoated with F108 and F108-NTA in the absence of Ni 2 + ions,on the other hand, adsorbed much less H-N195 (0.09 ( 0.06and 0.38 ( 0.05 mg/m 2 , respectively), which demonstratedthat F108 effectively reduced the level of nonspecific proteinadsorption to PS beads. In the presence of Ni 2 + ions, F108-NTA-coated PS beads adsorbed 2.93  (  0.21 mg/m 2 of H-N195, slightly more than the theoretical monolayer. Thesubstantial increase in the amount of H-N195 adsorbed toF108-NTA-treated PS beads in the presence of Ni 2 + ionsdemonstrated that H-N195 was bound specifically throughthe H-tag in a manner consistent with the assumptions of the oriented monolayer model. Similar experiments havebeen carried out with H-tagged kinesin proteins with similarresults (not shown).  Microtubule Motility on F108-NTA Surfaces.  To use theF108-NTA for standard microtubule motility experiments,glass slides and coverslips were silanized with dichlorodi-methylsilane (DDS) to create hydrophobic surfaces. Flowchambers assembled with the DDS glass were coated withF108-NTA in the presence and absence of Ni 2 + ions, andthen incubated with H-N195. In the presence of Ni 2 + ionsand 1 mM MgATP, microtubules moved smoothly andconsistently with an average speed of 0.16  (  0.01  µ m/s(Figure 3). In the absence of Ni 2 + ions, no microtubules wereobserved to land on the flow chamber surface in the presenceof either MgATP or MgAMPPNP even after incubation of the flow chambers with high concentrations of ncd motor.The F108-NTA, therefore, provided an effective method forsurface immobilizing ncd for in vitro motility experimentsin the absence of carrier proteins and without being fused toGST or GFP.The speed of microtubule movement on the F108-NTA-Ni 2 + surfaces was independent of H-N195 density down toa threshold density of approximately 250 motors/   µ m 2 ; belowthe threshold motor density, no microtubule landing ormotility occurred (Figure 3). In MgAMPPNP, microtubuleslanded on the surface at H-N195 densities of   < 20 motors/   µ m - 2 , which demonstrated the presence of active H-N195on the flow chamber surface. The threshold surface densityfor microtubule motility in MgATP is in sharp contrast toresults from similar experiments with native kinesin ( 4 ) inwhich microtubule landing and motility were observed atsurface densities of  < 10 motors/   µ m 2 , surface densities wheremicrotubules moved on single kinesin motors. The thresholddensity for H-N195 motility was consistent with our pre-vious report that single dimeric ncd motors, in contrast todimeric kinesin, are not capable of sustained microtubulemovement ( 14 ).  H-N195 Density Dependence of the Microtubule Landing Rate.  The critical surface density required for H-N195motility suggested that multiple H-N195 motors were neces-sary for sustained microtubule movement that can beobserved in a standard motility assay. To estimate the numberof dimers that are necessary for motility, the rate of microtubule binding to H-N195 surfaces was determined overa range of H-N195 surface densities (Figure 4). Themicrotubule landing rate is proportional to the probabilitythat a microtubule will contact simultaneously the thresholdnumber of H-N195 motors. The probability of at least onemotor being within the surface area contacted by themicrotubule, from the Poisson distribution function, is givenby the equation 1 - e -F  /  F 0 , where F is the density of motorsand  F 0  is the area effectively occupied by a single motor,the area over which a motor can bind a microtubule ( 22 ). If several ncd dimers must cooperate for a microtubule to land,then the microtubule landing curve should have a generalpower law dependence on the probability of a single motor F IGURE  3: Microtubule gliding velocity on different surfacedensities of H-N195. Each data point represents the average speedof at least five microtubules. In the presence of Ni 2 + ions, themicrotubule gliding velocity (0.16 ( 0.01  µ m/s) is independent of H-N195 surface density above a critical surface density of ap-proximately 250 motors/   µ m 2 . In the absence of Ni 2 + , no microtu-bules were observed to land on the flow chamber surface even inthe presence of MgAMPPNP and at the highest motor concentra-tions.F IGURE  4: Rates of microtubules landing on a H-N195-coatedsurface in the presence of 1 mM MgATP ( 9 ) and 1 mMMgAMPPNP ( b ). H-N195 molecules were immobilized on F108-NTA-Ni 2 + -coated DDS coverslips. Each data point represents datafrom at least six different assays. In the presence of MgAMPPNP,the landing curve was fit when  n  )  1 ( ( 0.167), while in thepresence of MgATP, the landing curve was fit when  n  )  4( ( 0.746), which implies cooperative binding. Also shown areMgATP data fits when  n  )  3 and 7 (dashed lines). Ncd Is Not Processive  Biochemistry, Vol. 38, No. 16, 1999  5079  interacting with a microtubule, (1  -  e -F  /  F 0 ) n , where  n  is aHill-like coefficient that gives an estimate of the number of motors that cooperate in microtubule landing and motility( 22 ). In the presence of 1 mM MgATP, the microtubulelanding rate was constant at approximately 300 microtubuless - 1 mm - 2 above a surface density of about 1500 molecules/   µ m 2 . At H-N195 surface densities of  < 1500 molecules/   µ m 2 ,the landing rate decreased with the H-N195 surface densityuntil no landing was observed below about 250 molecules/   µ m 2 . The best fit to the landing curve in a log - log plot was4, which implied that four H-N195 dimers cooperate in mi-crotubule landing and stable motility. For comparison, fitswith slopes of 3 and 7 are also shown (Figure 4, dashedlines).The nonhydrolyzable ATP analogue, AMPPNP, inducesa tight binding, rigor-like state in ncd. To validate theexperimental procedure for determining the power lawdependence of microtubule landing rates on the H-N195surface density, the landing rates were determined in thepresence of MgAMPPNP. Because a single dimeric ncdbinds tightly to microtubules in the presence of MgAMPPNP,noncooperative microtubule landing by H-N195 was ex-pected. Consistent with this expectation, in the presence of MgAMPPNP, the log - log microtubule landing rate curvewas best fit when  n  )  1 (Figure 4).  ATP Dependence of H-N195 Motility.  The speed of H-N195-driven microtubule motility was investigated as afunction of ATP concentration at a surface density of 1500motors/   µ m 2 (Figure 5). The ATP dependence of the motilityspeed was fit well by the Michaelis - Menten equation witha  V  max  of 0.189  µ m/s and a  K  M - ATP  of 235  µ M. Forcomparison, the  K  M - ATP  for dimeric ncd in solution isapproximately 23  µ M ( 23 ). DISCUSSION Surface Immobilization and Motility of Dimeric Ncd.  Themetal-chelating F108-NTA reagent provided a convenientand effective method for immobilizing H-tagged ncd pro-teins. To prepare a stable, activity-preserving surface for ncdmotility experiments, the reagent was simply adsorbed ontosilanized glass slides and coverslips. This method can beeasily applied to submicron polystyrene or glass beads forhigh-resolution motility experiments using instrumentationbased on optical tweezers. Because the Pluronic foundationis oriented on the hydrophobic surface and the motor is boundthrough terminal histidine residues, the mechanical linkagebetween the motor and the surface is defined and consistent.And since the attachment is chemically defined, it may bepossible to adjust the linkage compliance between the motorprotein and surface by using a Pluronic with shorter or longerPEO blocks. In single-molecule, micromechanical experi-ments with optical tweezers, the consistent linkage compli-ance will allow direct comparison between individual proteinmolecules, as well as direct comparisons between variousrecombinant forms of kinesin motors, single-headed deriva-tives of kinesin and ncd for example. We have observedmotility similar to that reported here with several otherH-tagged truncations of both ncd and kinesin (unpublishedobservations). Since the attachment method has worked withseveral H-tagged proteins, it may be generally applicableto other H-tagged kinesins, including those that have notyet moved, like MCAK and XCMK ( 24 ,  25 ), allowing di-rect comparisons of motility data between kinesin familymembers. Processi V  ity of Dimeric Ncd.  Highly processive motilityis a hallmark feature of kinesin. Whether processivity is acommon feature of the kinesin family of motors has not beenfully answered, although evidence that ncd is not asprocessive as kinesin has been previously reported ( 12 ,  14 ).The microtubule landing and motility data we report providedadditional indications that multiple ncd dimers are necessaryfor sustained microtubule motility and therefore ncd is notprocessive in the same way as kinesin. The first indicationwas the critical surface density of H-N195 required formicrotubule motility (Figure 3). At densities below about250 motors/   µ m 2 , microtubules did not land on the surface,although active motors were present on the surface sincemicrotubules were reversibly bound in the presence of MgAMPNP. Single molecules of dimeric ncd are not capableof sustaining observable microtubule movements even withlow concentrations of ATP.Comparing the maximum rate of ncd-driven microtubulemotility in vitro to the maximum rate of microtubule-activated ATP hydrolysis in solution provides a secondindication of low ncd processivity. The maximum motilityspeed of about 160 nm/s corresponds to 20 tubulin subunits/ s, or 20 motor binding sites spaced at 8 nm intervals alonga single microtubule protofilament. The  k  cat  for ATP hy-drolysis by dimeric ncd in solution is 2 s - 1 per motor domain,or 4 s - 1 per ncd dimer ( 23 ). That means that five tubulinsubunits (40 nm) must pass by a stationary dimeric motorfor each ATP that is hydrolyzed by the dimer. In other words,even if ncd is weakly processive in the hand-over-handmanner widely accepted for kinesin, taking a minimum of two concerted steps, then at 4 steps/s the ncd dimer will bestrongly bound to the microtubule for 0.5 s, during whichtime the microtubule was observed to move 80 nm. Ncdeither takes very large steps (40 nm), hydrolyzes ATP muchfaster in the motility assay than in solution, or is not evenweakly processive.The critical surface density for motility implied that severalncd dimers must cooperate during sustained microtubulemovement. To estimate the minimum number of ncd dimersthat cooperate in microtubule movement, we determined theH-N195 surface density dependence of the microtubulelanding rate. If multiple motors are necessary, then at lowsurface densities the landing rate should have a power law F IGURE  5: Microtubule gliding velocity as a function of ATPconcentration. The H-N195 surface density was approximately 1500motors/   µ m 2 . Each data point represents the average speed of at leastfive microtubules. The data were fit (solid line) to the Michaelis - Menten equation with a  V  max  of 0.189 ( 0.002  µ m/s and a  K  M - ATP of 235  (  16  µ M. 5080  Biochemistry, Vol. 38, No. 16, 1999  deCastro et al.
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks