124 Europeon Journol of Cell Biology 73, 124-131 (1997, June\ . @ Wissenschoftliche Verlogsgesellschoft . Siuttgort EJCB

A kinesin-like mechqnoenzyme from the zygomyceteSyncephqlqstrum rqcemosum shores biochemicqlsimilqrities with convenlionql kinesin fromNeurosporq crqssq

Gero Steinbergr)Institute for Cell Biology, Ludwig-Maximilians-University, Miinchen/Germany

Dedicated to Prof. Dr. Rainer Kollmann, Kiel/Germany, on the occasion of his 65th birthday

Received October 17, lS96Acceoted December 16. 1996

Biochemistry - fungi - kinesin - molecular motor -organelle motility - Syncephalastrum racemosum -

zygomycete

The kinesin superfamily consists of mechanoenzymes that convertchemical energy, stored in nucleoside triphosphates, into movementalong microtubules. The founding member of this protein superfamily,the so-called conventional kinesin, was only known from animal sourcesuntil the recent description of kinesin from the filamentous fungusNeurospora crassa (G.Steinberg, M.Schliwa, Mol. Biol. Cell 6,1605-1618 (1995)). To determine whether similar motors with compa-rable features are common in other filamentous fungi, a kinesin froma zygomycete, Syncephalastrum racemosum, was purified. Here, theisolation and characterization of this motorenzyme is described. Thepurified protein consisted of a doublet at 112 kDa and 115 kDa with noaddilional polypeptides. This was consistent with a calculated molecu-far mass of -?,40 kDa and suggests that the motor is a dimer with amore globular shape than conventional kinesin from animal sources.In gliding assays the enzyme moved microtubules at 2,5 to 3.4 pm/sand had a nucleolide specificity similar to the Neurospora kinesinmotor. Peptide antibodies against conserved regions in the head andthe tail domain of conventional kinesins cross-reacted with the Syn-cephalastrum motor. In vitro, the enzyme was able to drive themicrotubule-dependent movement of vesicles isolated from Syncepha-lastrum racemosum, as well as Neurospora crassa, and Aspergillusnidulans. In summary the Syncephalastrum motor has many of theunique features in common with the conventional kinesin from theascomycete Neurospora crassa and probably shares a similar functionin living hyphae.

Abbreviot ions. AMPPNP Adenosine-51 [cr,p-i mido]-tr iphosphote. -DIC Differentiol enhonced controsf. - MTs Microtubules. - NkinNeurosporo kinesin. - PBS Phosphote-buffered sol ine. - SynkinSvnceoho los i rum k ines in .

r) Dr. Gero Steinberg, Institut ftir Genetik, Ludwig-Maximilians-Universitiit, Maria-Ward-Str. 1a, D-80638 Miinchen/Germany.

Introduction

Organelle transport is an important feature of all eukaryoticcells and has fascinated scientists for more than 200 years(reviewed in [a0]). In particular, in elongated axons the molec-ular basis of intracellular movement of vesicles and mito-chondria has been intensively studied (reviewed in [11]). Withthe purification of the microtubule (MT)-dependent mecha-noenzyme kinesin from neuronal tissue [10, 50] and sea urchineggs [41], a potential motor for plus-end directed movementswas discovered. Numerous studies confirmed a major role forkinesin in organelle transport processes in several animal celltypes (reviewed in [8, 42]), presumably in combination withminus-end directed cytoplasmic dynein which was first foundin neurons [36] and was subsequently described for other celltypes (reviewed in [2a]). In addition, recent studies indicate amore complex situation in the axon, as it is now commonlyaccepted that cytoplasmic myosins [6, 13, 29] and severalkinesin-like motors (e.g. [2, 17 , 38f , reviewed in [8, 23]) mayalso take part in the transport of membranous organelles.

Intracellular movement of organelles in fungi is found insingle cells like yeast [26], but the process is more prominentand obvious in the elongated cells of filamentous fungi(reviewed in [20, 3a]). The molecular basis of this process isstill not completely understood. Inhibitor experiments andelectron microscopy studies implicate MTs [1, 18, 21, 46], aswell as actin filaments [22, 27] as the underlying tracks fororganelle transport (reviewed in [19, 20]). Recent geneticresults gave insight into the molecular motors that may beinvolved in intracellular transport in fungi. They establishedan important function for cytoplasmic dynein in nuclearmigration in the ascomycetes Aspergillus nidulans [52], Neu-rospora crassa [39], and Saccharomyces cerevisiae [12]. More-over, it turns out that an actomyosin system might be responsi-ble for the movement of mitochondria in yeast [31, 45]. Inter-estingly, mitochondria and-vesicle movements in Neurosporacrassa are found to depend only on MTs [46], and the Neuro-

,d*{fiiu

EJCB Kinesin from Syncepholoslrum rocemosum 125

spora kinesin is a good candidate for a mechanoenzyme thatmay take part in organelle motility [a7]. This motor enzymefrom N. crassa displayed several unique features, which werenot found for animal kinesins. These include the absence oflight chains, a far more globular shape, unusually fast in vitromotility, and the requirement of ATP for the stabilization ofthe gliding activity. This study was undertaken to address thequestion whether comparable kinesins are present in other fila-mentous fungi, and, if so, whether these motors share some ofthe unusual features with Nkin (Neurospora kinesin). The fila-mentous fungus Syncephalastrum racemosum was chosen as atypical representative of the ancient class of zygomyceteswhich are believed to be an ancestor group for asco- and basi-diomycetes [3]. Here, I describe the purification and charac-terization of a putative mechanoenzyme from this species.MT:motility experiments, along with biochemical and theimmunological data, suggest that this motor represents a "con-ventional" kinesin. with remarkable similarities to the Neuro-spora motor. Probably kinesins of this type exist in all filamen-tous fungi, and they might be responsible for at least some ofthe MT:dependent organelle transport processes in hyphae.

Moteriqls qnd methods

ChemicolsUnless otherwise noted, reagents and enzymes were obtained fromSigma (Deisenhofcn/Germany), salts were supplied by Merck (Darm-stadt/Germany), and growth media ingredients by Difco (Augsburg/Germany).

Cells ond culture conditionsSyncephalastrum racemosum, strain CBS 349.35 (provided by Dr.R. Agerer, Munich/Germany) was originally obtained from the Cen-traalburo voor Schimmelcultures, Baarn/The Netherlands. Cultureswere stored on nutdent agar platcs containing SM-medium (16 muKHrPOl, 6 mu K2HPOa, 4 mu MgSOa, 1% glucose. 2o/o bacto-agar,1 % peptone, 1 % yeast extract). For preparation of conidia solutions,agar-plate cultures were grown for 7 to 10 days at room temperatureunder dim light. 5 to 10 ml of sterile water was placed on top of themycelium, and the conidia were harvested with a sterile microscopeslide. The number of conidia in solution was counted in a Neubauerchamber (usually - 107 conidia/ml), then 50 pg/ml kanamycin wasadded. and the stock was stored at 4'C for further use within the fol-lowing two weeks. For microscopic analysis and purification of kinesin1000 ml iiquid medium (3.5 mu Na2HPOa, 3.6 mu KH2POa, 50 mvmaltose, 1.4% peptone, 0.7% yeast extract) was inoculated with9 x 106 conidia, and cells were grown under continuous illuminationand gentle shaking (80-150 rpm) for 40 h at2l 'C. Under these condi-tions the average yield from a 1000 ml liquid culture was 15 g hyphae.

Stoining of the cell wollSynccphalastrum conidia were grown in culture mediumon poly-r-lysine-coated coverslips overnight in a humid chamber. Afterrinsing the coverslip with phosphate-buffered saline (PBS: 8 mrrlNa2HPOa, 14.5 mla KH2PO4, pH 7.2, 0.5 mv MgCl2, 2.7 mv KCl,137 mu NaCl) the sample was fixed with 3% (v/v) formaldehyde,washcd several times with an excess of fresh buffer. The cell wall wasstained with 5 pg/ml calcofluor in PBS for 5 min, followed by 3 rinseswith PBS.

Anolysis of orgonelle morility in vivoTwelve to twcnty hours old Syncephalastrum cells were taken from liq-uid cultures, placed in a flow-through chamber (see above) and ana-lyzed in PBS using video-enhanced DIC microscopy. Thin parts of thebranched hyphae with active organelle transport were selected and,

after flushing in 200 to 400 pg PBS, movements of highly refractilevesicles were monitored. The number of organelles crossing lines inhyphae was counted and used as the 100 % reference value. This wasfollowcd by replacing the PBS with -500 pl of 10 pv nocodazole, beno-myl (Dupon. Muenster/Germany) or cytochalasin D in PBS, and after30 min and additional replacement of the solution organelle move-ments were measured. This experimental procedure, as well as di-methyl sulfoxide (DMSO; Fluka. Neu-Ulm/Germany) alone, had noinfluence on organelle transport as tested by control experiments.

Preporotion of MTsTubulin was purified from fresh pig brain using three cycles of polymer-ization and depolymerization according to the method of Shelanski etal. [44] with modifications as previously described [47]. MTs werepolymerized from pig brain tubulin at a concentration of 2 to 6 mg/mlin the presence of 10 % DMSO and 1 mv GTP for 30 min at 32'C. TheMTS were diluted to 0.5 to 0.7 mg/ml in 100 mu PIPES, pH 6.9, 2 mvrMgCl2, 1 mn ethylene glycol-bis(p-aminoethyl erher)N,N,N',N,-tetraacetic acid (EGTA), 1 mu EDTA, containing 7 prra taxol, brieflyvortexed, and stored at room temperature for 1 to 4 days.

lsolotion of kinesin from S. rocemosum qnddelerminotion of biophysicol propertiesS. racemosum cells were grown for 40 h as described above and groundwitf quartz-sand in the presence of the bulfer AP100 (100 mv PIPES,pH 6.9, 2 mu MgCl2, I mu EGTA, 1 mrra EDTA, 1 mnr dithiothreitol,I mrra, phenylmethylsulfonyl fluoride, 10 pg/ml tosyl-r--arginine methylester, 10 pg/ml soybean trypsin inhibitor, 1 pg/ml aprotinin, 1 pg/mlpepstatin. 1 pglml leupeptin) and thc kinesin was isolated according toa previously described protocol [48]. The rcsulting supernatant afterAIP release of the motor molecules from thc taxol-stabilized pig brainmicrotubules (S5) was tested in gliding assays. The pellct was resus-pended in half of the S-5 volume, centrifuged, and the resulting 56 waspooled with the 55. The motor containing 55/56, which was usually230 pl from 15 g hyphae (1000 ml liquid culture), was stored on ice andstayed active for several days. Additional purification, determinationof the Stokes radius and sedimentation coefficient and the calculationof the molecular weight and axial ratio was done according to previ-ously described procedures [48].

Gel electrophoresis ond ATPose ossoySodium dodccyl sulfate polyacrylamide gel electrophoresis was per-formed according to the method of Laemmli [30] with modificationsaccording to Garfin [14]. Gels were stained with Coomassie Blue or sil-ver nitrate according to Blum et al. [9]. The ATPase activity was mea-sured following the procedure of Itaya and Ui [25] with modificationspreviously described [48].

Anribody prepqrotion ond Weslern blot onolysisCorresponding to zr conserved region in the tail sequence of the Neu-rospora kinesin [47] a 17 amino acid peptide (FLERNLE,QLTQ-VQRQLV) was synthesized by Eurogentec (Seraing/Belgium). Thepeptide was coupled to hemocyanin and two female New Zealandwhite rabbits were immunized with this construct (Eurogentec,Seraing/Belgium). The antibody, named NKC, was purified accordingto Olmsted [35], and subsequent Western blots were done as previ-ously described [47].

Mofility ossoysMT gliding activrty was measured as described in Steinberg andSchliwa [47]. To guarantee that the observations were comparable withthose made for the Neurospora kinesin, care was taken to comply withall criteria previously described for this assay [a7]. As the kincsin fromSyncephalastrum was unstable after gel filtration, most mcasurementswere done using a flow-through chamber or density gradient material.This had no influence on the motilitv behavior of thc motor enzvmes.Syncephalastrum kinesin iiiowect a

-high sensitivity to the absence of

ATP. Therefore, the motor was stabilized with 0.05 mrra ATP. and the

126 G. Steinberg ffii..::, EJCE

influence of this nucleotide on motility assays was determined as neg-ligible in several control experiments.

To determine the polarity of movement, Chlamydomonas axonemeswere isolated according to Witman [51], salt extracted [28] and placedon poly-t-Jysine coverslips. One pl of carboxylated latex beads (Poly-sciences, Warrington, PA/USA) in AP10 (composition like AP100 but10 mu Pipes) was mixed with 2 pl 5 mg/ml casein and 2 pl motor-containing supernatant (S5) or column-purified peak fraction. After15 min on ice, 1 pl of this mixture was added to the axonemes in thepresence of 2 mv MgATP Single beads were grasped with a laser trapand placed on the axonemes, and after release from the trap theymoved along the axonemes towards the plus-end of the polar structure.

For in vitro organelle motility experiments, an unspecified mem-brane fraction was prepared. Cells were ground with quartz-sand in thepresence of AP100 and protease inhibitors as described above, andcentrifuged for 30 min at 100009. The upper layer of the pellet wasresuspended in protease-containing buffer. The membranes werestripped with 600 mv KCI and sedimented through a 15 % sucrosecushion of equal volume. This solution contained organelles of differ-ent sizes, with a major portion of mitochondria as confirmed by rhod-amine 123 staining. As described for the bead assay, a mixture of kine-sin solution, casein and organelles was incubated on ice for 15 min andadded to stabilized MTS, and organelle movements were monitoredusing video-enhanced DIC light microscopy [5]. Vertebrate organelleswere taken from the first pellet of the tubulin isolation procedure (pigvesicles) or were isolated from rat liver. Statistical analysis was doneusing two-tail t-tests within the program PRISM (GraphPad, SanDiego, CA/USA).

Results

Tronsport of unclqssified vesicles in the hyphoeis MT bosedSyncephalastrum racemosum grows as elongated, highlybranched hyphae of variable diameter, ranging from - 5 to12 pm. The cells are surrounded by a cell wall and are notdivided up by septae (Fig. 1), as is typical for zygomycetefungi [3]. As a result, hyphae are filled with numerous nucleithat are distributed over the length ofthe cells, as seen both byDAPI staining and by contrast-enhanced video-microscopy(data not shown). The cytoplasm contains a continuously mov-ing system of tubes and vacuoles and long MTs, which areoriented along the long axis of the hypha (data not shown).Mitochondria could not be identified with certainty and there-fore were not investigated. Small refractile vesicles of un-known function perform saltatory movement at rates of 2 to2.5 pm/s. Interestingly, unlike Neurospora crassa, the fre-

Fig. 2. Influence of cytoskeleton affecting inhibitors on the organelletransport in S. racemosum. The number of refractile vesicles that crossa line in 30 s after a 30 min treatment with 10 prra of benomyl, nocoda-zole or cytochalasin D was counted and compared to the transport rateof organelles in the same cell under PBS treatment. Cytochalasin Dhas only a minor influence on the organelle transport, but benomyldramatically reduces the transport rate. Nocodazole did not inhibitorganelle motility (values are given as mean t s.e.m., n :7 cells).

quency of intracellular transport in S. racemosum decreasedrapidly, while the specimen was observed under light micros-copy. This could be immediately restored when fresh liquid,culture medium, PBS, or degassed water was flushed into thechamber. The frequency of transport and the movementsdecreased again within 1 to 2 min after the flow ended. 100 puGd3t, a reagent that affects mechanosensitive cation channelsin fungi [15, 53] had no effect on this phenomenon. The subse-quent investigation of organelle transport rates took therequirement for a continuous flow of buffer into account.

To clarify the role of the cytoskeleton in the movement ofrefractile vesicles, the cells were treated with inhibitors thataffect the MT:cytoskeleton (nocodazole. benomyl) or F-actin(cytochalisin D). Placed in PBS, the hyphae lived for over t hwithout dramatic change of their internal organization (e.g.

J-.

*o

o<t)(nL

(u

(!

o

oac(g

benomy nocodazole cytoD

Fig. 1. Morphology of S. racemosum. Calcolluor staining of abranched hypha grown on a poly-r-Jysine-coated coverslip. As is typ-

ical for zygomycetes, the fungus grows coenocytic from a single co-nidium (arrow). - Bar 50 pm.

,$rvi'

EJCB Kinesin f rom Syncepholoslrum rocemosum 127

94>

67>

43>

30>1 1 6 >

97.4>

66>

1 16>

7.4>

q qS4P3S3DZ @ oo (+)

b

Fig. 3. Isolation of the Syncephalastrum kinesin. - a. A cytoplasmicextract of S. racemosum (S2) was incubated with taxol-stabilized pigbrain MTS. In the absence of ATP the enzyme bound to the polymersand cosedimented with the MTS (P3). The pellet was resuspended infresh buffer, salt washed and sedimented again (P4). The release of thekinesin with MgAIP-containing buffer yields a very pure supernatant(S5), which was further purified by a gel-filtration run (C). - b. Gel

filtration of 200 pl kinesin containing 55. Coomassie stain of a 7o/"polyacrylamide-gel showing fraction 25 to 32 (500 pl) after runningover a superose-6-column. The lI2l115 kDa doublet comigrated withthe MT-gliding activity (plus and plus/minus). - c. Silver stain of thefraction number 29 in a 15 % polyacrylamide gel. Beside the 112 kDaand 115 kDa polypeptides no additional light chains or contaminantsare detected.

size of vacuoles), and still showed apparently unchanged vesi-cle transport. Incubation with 10 [rM nocodazole had no effecton the movements, and although 10 pr,rra cytochalasin Dreduced the transport rate to -70% of the control, thischange was not statistically significant (a:0.05; P:0.192;'Fig.Z). In contrast to nocodazole, 10 pll benomyl signifi-cantly reduced the intracellular transport of refractile vesicleswithin 15 to 30 min (Fig. 2). This effect was in part reversible,but the increase of vesicle movement by the replacement ofbenomyl solution with fresh PBS was accompanied by theappearance of additional vacuoles, and the cytoplasm con-sealed before the cells died.

5. rocemosum cells contoin qn dbundqnt dimericmechonoenzymeTo isolate a potential organelle motor, a previously successfulprotocol for isolating kinesins from filamentous fungi wasapplied [47, 48], which is based on the "fishing" of motormolecules with exogenous MTs in an AlP-depleted extract,followed by an AIP release [10, 50]. The isolation procedureIed to a supernatant containing a double band at 112 and 115kDa (Fig. 3a). Beside these polypeptides, only some exoge-nous pig tubulin and low amounts of contaminating polypep-tides are found in the AIP release (S5). Further purificationby sucrose density gradient (not shown) or gel filtration

Thb. I. Comparison of the biophysical and motility properties of kinesin from S. racemosum, N. crassa and animal sources.

Synkin Nk inu Kinesino

V MT-gl id ingMgATPMgGTPMgCTPMgUTPMgITPAMPPNP i nh ib i t .Stokes radiusDwzoSwzoMW in SDS-PAGEMW nativeaxia l rat io

t u m s ,l % lt % lt % l( % l( % lAMPPNP: ATP( n m )(x 1o-7 cm2s 1)(x 1 013s)(kg mol t )(kg mol-1)A : D

2.1- 3 .410036 -46t 5 - | I

20 -240 - 3

< 1 . 2 - 1 . 57.O4 + 0.433 . 1 3 + 0 . 1 88.37 + 0 .13112/115- 240 (220-262)1 2 . 3

2 .1- 3 .8100509 - 1 9

12 -210

< 0.36.21 + 0.533.34 + O.208.80 + 0 .151 05/1 08-227 (199-26011 1 . 5

0.6 - 0.810078 -1000 - 3 10 - 3 8

2 7 - 5 0< 0.59.64 + 0.87*2.24 + O.21x9.56 + 0.34* (6.5**)124,64*- 379 (294-4941x19.5* (40+* )

Moti l i ty data f rom several exper iments of at least 2 isolat ions, b iophysical data f rom at least 3 isola l jpns. - a Data f rom [48] . - b Nat ive isolatedkinesin f rom several animal sources (summarized in [48]) . - * Folded conformat ion, data f rom [7] . - "* Unfolded conformat ion, data f rom [16] .

128 G. Steinberg EJCg

yielded pure motor (Fig. 3b), which migrated with the glidingactivity as tested in a MT:motility assay. Additional lightchains were never detected in any isolation as confirmed bysilver staining of gels (Fig. 3c). Using the estimated Stokesradius (7.04 + 0.43 nmi n : 3) and the sedimentation coeffi-cient, estimated as 8.37 + 0.13 nm (n : 3), the native molec-ular mass was found to be approximately 240 kDa (range220 262 kDa), with an axial ratio of 12.3 (data are summa-rized inThb. I). These data are consistent with the absence ofadditional polypeptides and suggest that the Syncephalastrummotor is a dimer in its native form.

The gliding activity of the Syncephalastrum motor washighly dependent on the presence of nucleoside triphosphates.In the absence of ATP or ATPyS the motor rapidly lost its MT:gliding activity. Even in the presence of AIR purified gelfiltration or sucrose gradient fractions lost their MT:glidingactivity within several hours on ice. Occasionally, gel filtrationled to protein which was unable to promote MT:gliding, butstill showed a highly variable ATPase activity, ranging from0.088 to 0.207 prmoVmin/mg in the presence of 1 mg/ml MTs,and variable MT activation of 2.5 to 15-fold.

Synkin shqres immunologicql homology wirhconventionol kinesin from NeurosporoThe peptide antibody MMR44.1 (provided by Dr. M. A.McNiven, Rochester, MN/USA) raised against the HIPYR-region of the kinesin head sequence [32] recognized recombi-nant Drosophila kinesin (provided by U. Henningsen from aplasmid of Dr. L. S. B. Goldstein) and showed a strong affinityto both bands of the Il2 kDa/115 kDa doublet of the kinesinfrom S. racemosum (Synkin; Figs. 4a, b). An antibody againsta peptide region of the C-terminal tail of the conventional

abcMMR44.1 MMR44.1 NKC

' I l6>

97.4-

66>

Synkin DKHC Synkin Nkin Synkin DKHC

Fig.4. Immunological characterization of the kinesin from S. race-mosum. - a. A peptidc antibody against a conserved peptide in theglobular domain of the kinesin heavy chain (MMR44.1) recognized theSyncephalastrum kinesin (Synkin) as well as recombinant kinesin fromDrosophila melanogaster (DKHC). - b. Both bands of the 7121175kDa doublet were stained by this antibody. - c. An affinity-purifiedpolyclonal antibody against a sequence region in the C-terminal tail ofkinesin from Neurospora crassa, termed NKC, clearly recognized Neu-rospora kinesin (Nkin) and the kinesin from S. racemosum (Synkin),

but not Drosophila kinesin (DKHC). d. In protein extracts ofS. racemosum (S2) the antibody gave a strong signal at approximatelythe same molecular mass as in 55.

dNKC

'l---lt ll lt lI I

1# rmltl ,'',. I

I qSe"P I

t l>t -1 |

t lt lt lt l>l Ilr ll'l':l.,'t"''..:"' I

S2 S5

kinesin from Neurospora crassa, termed NKC, recognizedboth polypeptides from Syncephalastrum and the Neurosporacontrol protein, but did not bind to recombinant Drosophilakinesin (Fig. ac) or isolated bovine brain kinesin (data notshown). In crude extracts of S. racemosum (S2, see above),NKC clearly recognized a polypeptide of the same relativemolecular mass as the isolated kinesin (Fig. ad). Unfortu-nately, this antibody could not be used for immunofluores-cence, as it gave a vesicular staining in wild type hyphae ofUstilago maydis, but showed the same staining in null mutantsthat contain no kinesin on Western biots (data not shown).

The Syncephqlqstrum mechqnoenzyme supportsin vitro moli l i tyThe purified enzyme was able to transport taxol-stabilizedMTs polymerized from pig tubulin in vitro. In several experi-ments from more than 10 isolations, it supported MT:gliding ata rate of 2.I to 3.4 um/s. and no difference was found whether55, sucrose density or gel filtration fractions were investi-gated. For in vitro motility the Syncephalastrum motor haddifferent efficiencies for different nucleotides (ATP > GTP>CTP>UTP). This gliding activity could only be inhibitedby a relatively high amount of AMPPNP. At a ratio of 1.2 to1.5 AMPPNP:ATP the MT:gliding activity was - 50 % inhib-ited (all motility data are summarized in Tab. I). The directionof motility was determined using unipolar Chlamydomonasaxonemes in a modified in vitro motility assay [37]. Latexbeads that were coated with low amount of kinesin from den-sity gradient fractions or 55 always moved to the fan-shapedend of Chlamydomonas axonemes with 1.5 to 1.7 pm/s(Figs. 5a-d). As this part of the axoneme represents the endwith a higher polymerization rate (plus-end) of the MTs [4] thekinesin from S. racemosum is a plus-end directed motor.

Kinesin containing 55 and, to a lower extent, purified Syn-cephalastrum kinesin was able to move unidentified vesiclesfrom a crude organelle preparation along MTs in vitro. Afterincubating salt-extracted membranes with motor and casein,organelles of different refractive index and size moved unidi-rectionally along MTs in a saltatory manner at a velocity of 1.7to 1.8 pm/s (Figs. 5e-h). This was done using the 55 fraction,and the same motility occurred when using density gradientmaterial, although to a much lower extent. The motility ofmembranes could easily be monitored in video-enhancedDlC-light microscopy and was clearly different from theBrownian motion of unattached vesicles. In some experimentsthis movement was very prominent, other samples just showedoccasional movements or did not transport vesicles at all,while supporting normal MT:gliding. In several control experi-ments salt-treated or untreated membranes alone were notable to move along MTs in the presence of AT?.

Interestingly, vesicles from Syncephalastrum were moved byNeurospora kinesin and vice versa. In several experimentssalt-treated organelles were transported by the Neurosporamotor at velocities of 1.43 + 0.19 pm/s (mean + s.d.;n : 4).This motility was not observed when untreated membraneswere used. On the other hand, Syncephalastrum kinesin sup-portcd motility of untreated Neurospora organelles (1.30 +0.17 pm/s, n : 20). After incubation of these membranes with500 mu KCI the velocity was significantly higher (2.21 +0.15 prm/s, n : 8; a:0.05; P<0.0001). Comparable resultswere found when using Aspergillus nidulans organelles. Inthese experiments Synbephalastrum kinesin moved smallvesicles before and after salt treatment at 2.25 + 0.73 um/s

dJF"-

EJCB Kinesin from Syncepholostrum rocemosum 129

&&arrr:.A:'lri

&,4Ll, 'r,l

gi:,:'

#

ffie

1

A,:

iW$.', qr"* * ,

*

* k', ,',,**',

tu

. W;li,:l:l::.lllll

*ll,l,q:i:l'

$i'.'*iltt:::&ilnr.

$ttl.:$!l&:r:&,rfi,lr..

&6i:

Fig. 5, In vitro motility performed by the Syncephalastrum kinesin. -

a to d. Latex beads were coated with kinesin, grasped by a laser trap(arrowheads) and placed on a stationary bipolar axoneme isolatedfrom Chlamydomonas flagella. After the trap was switched off, motormolecules on the surface moved the beads to the fan-shaoed end of the

(n : 20) and 3.07 + 0.27 prm/s (n : 7), respectively (signifi-cantly different, o:0.05; P:0.0082). In control experimentsneither rat liver mitochondria, nor freshly prepared vesiclesfrom pig brain were transported by the Syncephalastrum orNeurospora motor. These observations suggest that fungalvesicle motors and vesicles are mutually compatible, but morestudies are needed to clarify this point.

Discussion

The recently described kinesin from the filamentous ascomy-cete Neurospora crassa showed unusual features that were notfound for conventional kinesins from animal sources 147, 481.These properties and the loss of sequence similarity in the rodportion of the Neurospora motor clearly set apart this kinesinfrom other conventional kinesins. However, the assumed

,ffaxoneme that consists of the plus-ends of the MTs (marked with p/us);time in second is referred to by white numbers. - e to h. Movement ofa small vesicle (arrows) treated with S. racemosum kinesin alons ataxol-stabilized M't (arrowheads). White numbers indicate the time inseconds. -Bars3pm.

function as an organelle motor, the relative high abundance,and significant sequence similarities in the head and tail por-tion confirmed Nkin to be a fungal representative of this kine-sin family [47]. This study was undertaken to ask whethercomparable kinesins are present in the ancient group ofzygo-mycetes. The purified kinesin from S. racemosum displayed asurprisingly high similarity with the Neurospora kinesin andconfirmed some of the unusual features that are probably typ-ical for kinesin from filamentous fungi. Moreover, vesicles andmotor enzymes of both species were found to be interchange-able which suggests a common function in living cells.

For purification of the motor enzyme a previously successfulprotocol was applied. The main purification step, the bindingto taxol-stabilized MTs, followed by a ATP-release, is based ona tight binding (rigor state) of kinesin on MTs in the absenceof nucleotides [10, 33, 41, 50]. Applied to S. racemosum thisprocedure yielded a remarkably pure supernatant that con-

llliir:Wl'11l1

130 G. Steinberg EJCg

tained a high amount of a 1l2ltl5 kDa polypeptide doublet.These proteins were identified to be kinesin motors, probablyisoforms of true kinesin from Syncephalastrum, as confirmedby antibodies against the conserved region in the tail of theconventional kinesin from Neurospora and the peptide anti-body MMR44.1 [32]. The further characterization of the Syn-cephalastrum kinesin (Synkin) revealed remarkable similar-ities with the kinesin from Neurospora crassa. The purificationof Synkin yielded no additional polypeptides and the motorseems to be a dimer in its native state. Light chains, which areknown from all "conventional" kinesins (summarized in [8]),were also not found for the N. crassa kinesin. This still doesnot rule out the existence of highly protease sensitive lightchains that were degraded over the course of the purification(discussed in [a8]). Nevertheless, it seems unlikely that suchpolypeptides were lost in several purifications from two differ-ent fungal species. This is further supported by the resultsfrom the characterization of a kinesin from a basidiomycetethat also seems to lack light chains (Steinberg et al., manu-script in preparation). These findings are consistent with theresults from a polymerase chain reaction approach where prim-ers against conserved sequence stretches of kinesin lightchains in combination with RNA from Syncephalastrum andNeurospora did not yield appropriate products for the fungi,but were successfully used in a mouse library (E. Granderath-Kube, unpublished). In addition, probing 52 and 55 fromNeurospora and Syncephalastrum with anti-light-chain anti-bodies as well as with recombinant light chains did not indicatethe existence of real light chains (E Goeyda, unpublished).Although typical light chains might not exist in fungi, thisclearly leaves the opportunity for other cofactors that are neces-sary for a proper function of the motor. The instability of puri-fied S. racemosum kinesin might be due to the loss of suchcofactors rather than be a result of protein denaturation or par-t ia l degradat ion.

The remarkable similarities of both fungal kinesins fromphylogenetically distant classes might be due to a similar over-all organization of these molecules, which is different fromkinesins from animal sources [8, 48]. Whether or not there issome relationship between high in vitro motility, the absenceof light chains, and/or the compact native conformation offungal kinesins is not known. Although animal kinesins moveorganelles in vitro with a velocity of 1 to 2 pnls [8] their MTgliding velocity is 2 to 4 times lower. This might be due to assayconditions (e.g. interaction with the glass surface or an inter-action of the enzymatic active heads with the light chains).Therefore, it seems reasonable to assume that a differentmolecular organization of fungal kinesins might overcomesome of these problems and yield a gliding velocity thatreflects the in vitro motility of organelles.

Nothing is known about the in vivo function and propertiesof the refractile vesicles, and it remains open whether Synkinis involved in the MT:dependent intracellular movementsdescribed in this paper. Like other conventional kinesin ([45,49], summarized in [8]), both fungal kinesins were able tomove fungal organelles in vitro. The kinesin from S. racemo-sum moved vesicles from Neurospora crassa but not verte-brate membranes from two different sources? which argues fora specific interaction with the N. crassa organelles. After saltstripping of the fungal organelles, no difference in velocity wasseen for the homogeneous system, but Synkin accelerated thetransport of Neurospora organelles to near the maximumvelocity observed in in vitro gliding assays. Further experi-

ments addressing the binding of fungal kinesins to organelleswill give insight into the underlying mechanism, which mightinvolve salt-sensitive polypeptide on the membrane that con-trol the in vitro velocity.

In conclusion, the purified kinesins from the zygomyceteS. racemosum and the ascomycete N. crassa share remarkablebiochemical and motility similarities. The existence of highlysimilar motors in distant fungal classes suggests that compara-ble kinesins are probably present in all filamentous fungi, andtheir resemblance might be a result of an adaptation to uniquefunctions or regulations in the filamentous hyphae. From itshigh abundance in S. racemosum and N. crassa and the invitro motility of vesicles one can conclude that kinesin mightbe involved in vectorial transport of abundant vesicles,although kinesins cargo in fungi is still a matter of debate([31a, 43a]; Steinberg et al., manuscript in preparation). Ge-netic and physiological studies should elucidate kinesins'rolein tip growth or secretion processes of filamentous fungi.

Acknowledgements. I am deepty indebted to Dr. Manfred Schliwa atwhose institute this study was undertaken and who supported my workin numerous ways. I am also grateful to Sabine Fuchs for her excellenttechnical help and Dr. Eckhart Kube-Granderath for fruitful discus-sions. I thank Dr. Fatima K. Gyoeva, Dr. Eckhart Kube-Granderathfor sharing unpublished data, Ulrike Henningsen for recombinantkinesin, Harald Felgner fbr help with the laser trap, Dr. Mark A.McNiven for antibodies, and Dr. Rainer Agerer for the strain of S.racemosum. I am grateful to Dr. Steven N. Hird and Dr. CynthiaL. Tioxell for helpful comments on the manuscript. Finally, I wish tothank Dr. J. Richard Mclntosh for covering the cost of figure prepara-tion. - This work was supported by a grant of the DeutscheForschungsgemeinschaft to Dr. Manfred Schliwa.

References

[1] Aist, J. R., C. J. Bayles: Organelle motility within mitotic asters ofthe fungus Nectaria haematococca. Eur. J. Cell Biol. 56, 358-363(1991).

[2] Aizawa, H., Y.Sekine, R.Takemura, Z.Zhang, M.Nangaku,N. Hirokawa: Kinesin family in the murine central nervous system.J. Cell Biol. 1r9,1287-1296 (1992).

[3] Alexopoulos, C. J., C. W. Mims, M. Blackwell: Introductorymycology. 4th ed. John Wiley & Sons. New York, Chichester, Bris-bane, Toronto, Singapore 1996.

[4] Allen, C., G. G. Borisy: Structural polarity and directional growthof microtubules of Chlamydomonas flagella. J. Mol. Biol. 90,381-4020974).[5] Al len, R.D., N.S.Al len, J.L.Tiavis: Video-enhanced contrast ,differential interference contrast (AVEC-DIC). Microscopy: A newmethod capable of analyzing microtubule related movement in thereticulopodial network of Allogromia laticollaris. Cell Motil. 1,291-302 l98t).[6] Bahler M., R. Kroschewski, H.-E. Stoffler, T. Behrmann: Rat myr4 defines a novel subclass of myosin I: Identification, distribution,localization, and mapping of calmodulin-binding sites with differentialcalcium sensitivity. J. Cell Biol. 126,375-389 (1994).

[7] Bloom, G. S., M. C. Wagner, K. K. Pfisteq S. T. Brady: Nativestructure and physical properties of bovine brain kinesin and idcnti-fication of the ATP-binding subunit polypeptide. Biochemistry 27,340e-3416 (1e88).

[8] Bloom, G. S., S. A. Endow: Motor proteins. 1: Kinesins. ProteinProfile L, 1059-1105 (1994).

[9] Blum, H., H. Beicr, H. J. Gross: Improved silver staining of plantproteins, RNA, and DNA in polyacrylamid gels. Electrophoresis 8,93-99 (1987).

ls'"

EJCB Kinesin from Syncepholoslrum rocemosum l3l

[10] Brady, S. T.: A novel brain ATPase with properties expected forthe fast axonal transport motor. Nature 317, 73-75 (1985).

[11] Brady, S. T.: Molecular motors in the nervous system. Neuron 7,s21-s33 (1991).[12] Eshel, D., L. A. lJrrestarazu, S. Vissers, J.-C. Jauniaux, J. C. vanVliet-Reedijk, R. J. Plants, I. R. Gibbons: Cl,toplasmic dynein isrequired for normal nuclear segregation in yeast. Proc. Natl. Acad.Sci. USA 90, 11172-11176 (1993).

[13] Espreafico. E. M., R. E. Cheney, M. Matteoli, A. A. Nascimento,P. V De Camilli, R. E,. Larson, M. S. Mooseker: Primary structure andcellular iocalization of chicken brain myosin-V (p190), an unconven-tional myosin with calmodulin light chains. J. Cell Biol. ll9, 1541.-1557(reez).[14] Garfin, D.E.: One-dimensional gel electrophoresis. MethodsEnzymol. 182, 425-441 (1990).

[15] Garr i l l , A. , S.L.Jackson, R.R.Lew, I .B.Heath: Ion channelactivity and tip grorvth: tip-localized stretch-activated channels gener-ate an essential Ca'* gradient in the oomycete Saprolegnia ferax. Eur.J. Cel l Bio l . 60,358-365 (1993).

[16] Hackney, D.D., J .D.Levi t t , J .Suhan: Kinesin undergoes a 9 Sto 6 S conformational transition. J. Biol. Chem. 267,8696-8701(1992\.

[17] Hal l , D.H., E.M.Hedgecock: Kinesin-re lated gene unc-104 isrequired for axonal transport of synaptic vesicles in C. elegans. Cell 65,837-8'17 (1991).

[18] Heath, I.B.: Evidences against a direct role for cortical actinarrays in saltatory organelle motility in hyphae of the fungus Saproleg-nia ferax. J. Cell Sci. 9L,41-47 (1988).

[19] Heath, I. B.: The roles of actin in tip growth of fungi. Int. Rev.Cytol. 123, 95-127 (1990).

[20] Heath, LB.: The cytoskeleton. In: N.A.R.Gow, G.M.Gadd(eds.): The Growing Fungus. pp.99-134. Chapman and Hall. London1995.

[21] Heath, I .B. , M.C.Heath: Microtubules and organel le move-ments in the rust fungus Uromyces phaseoli var. vignae. Cytobiologie16,393-411 (1978).

[22] Heath,I. B., S. G. W Kaminskyj:The organization of tip-growthrelated organelles and microtubules revealed by quantitative analysisof freeze-substituted oomycete hyphae. J. Cell Sci. 93, 41-52 (1989).

[23] Hirokawa, N.: Organelle transport along microtubules - the roleof KIFs. Trends Cell Biol. 6, 135-141 (1996).

[24] Holzbaul E.L.E, R.B.Val lee: Dyneins: Molecular st ructureand cellular function. Annu. Rev. Cell Biol. 10,339-372 (1994).

[25] Itaya, K., M. Ui: A new micromethod for the colorimetric deter-mination of inorganic phosphate. Clin. Chim. Acta 14, 361-366 (1996).

[26] Jones, H.D., M.Schl iwa, D.G.Dubr in: Video microscopy oforganelle inheritance and motility in budding yeast. Cell Motil. Cyto-skeleton E, f29-142 (1993).

[27] Kaminskyj, S. G. W., K. S. Yoon, I. B. Heath: Cytoskeletal inter-actions with post-mitotic migrating nuclei in the oyster mushroom fun-gus, Pleurotus ostreatus: evidences against a force-generating role ofastral MTs. J. Cell Sci. 94,663 614 (1989).

[28] King, S. M., T. Otter, G. B. Witman: Purification and characteri-zation of Chlamydomonas flagellar dyneins. Methods Enzymol. 134,2e1-306 (1e86).

[29] Kuznetsov, S. A., G. M. Langford, D. G. Weiss: Actin-dependentorganelle movement in squid axoplasm. Nature 356, 722-725 (1992).[30] Laemmli, U. K.: Cleavage of structural proteins during the assem-bly of the head of bacteriophage T4. Nature 227, 680-685 (1970).l3ll Lazzarino. D. A., I. Boldogh, M. G. Smith, J. Rosand, L. A. Pon:Yeast mitochondria contain ATP-sensitive, reversible actin-bindingactivity. Mol. Biol. Cell 5, 807 818 (1994).

[31a] Lehmler, C., G.Steinberg, K.Snetselaar. M.Schliwa, R.Kah-mann, M. Bblker: Identification of a motor protein required for fila-mentous growth in Ustilago maydis. EMBO J., in press.[32] Marks, D. L., J. M. Larkin, M. A. McNiven: Association of kine-sin with the Golgi apparatus in rat hepatocytes. J. Cell Sci. 107,24t7-2426 n994\.

[33] McCaffrey, G., R. D. Vale: Identification of a kinesinlikemicrotubule-based motor protein in Dictyostelium discoideum.EMBO J. 8, 3229-3234 (1989).

[34] McKerrachet L.J., I.B.Heath: Topical review: Cytoplasmicmigration and intracellular organelle movements during tip growth offungal hyphae. Exp. Mycol. 11,79-100 (1987).

[35] Olmsted, J.B.: Analysis of cytoskeletal structures using blot-purified homospecific antibodies. Methods Enzymol. t34, 467-472( 1e86).

[36] Paschal, B. M., R. B. Vallee: Retrograde transport by themicrotubule-associated protein MAP 1C. Nature 330, 181-183 (1987).

[37] Paschal, B.M., R.B.Vallee: Microtubule and axoneme glidingassays for force production by MT motor proteins. Methods Cell Biol.39, 6s-74 (1993\.[38] Pesavento, P. A., R. J. Stewart, L. S. B. Goldstein: Characeriza-tion of the KLP68D kinesinlike protein in Drosophila: Possible rolesin axonal transport. J. Cell Biol. lYI, 7041,-1048 (1994).

[39] Plamann, M., P. F. Minke, J. H. Tinsley, K. S. Bruno: Cytoplasmicdynein and actin related protein Arpl are required for normal nucleardistribution in filamentous fungi. J. Cell Biol. 1Il,139-749 (1994).

[40] Schliwa, M.: Mechanisms of intracellular organelle transport. In:J. W. Shay (ed.): Cell and Muscle Motility. Vol. 5: The Cytoskeleron.pp. I-82. Plenum Press. NewYork 1984.

[41] Scholey, J. M., M. E. Porter, P. M. Grissom, J. R. Mclntosh: Iden-tification of kinesin in sea urchin eggs and evidence for its localizationin the mitotic spindle. Nature 318, 483-486 (1985).

[42] Schroer, T.A., M.P.Sheetz: Functions of microtubule-basedmotors. Annu. Rev. Physiol. 53,629-652 (199I).

[43] Schroer, T. A., B. J. Schnapp, T. S. Reese, M. P Sheetz: The roleof kinesin and other soluble factors in organelle movement along MTS.J. Cell Biol. 107, 1785-1792 (1988).

[43a] Seiler, S., F.E.Nargang, G.Steinberg, M.Schliwa: Kinesin isessential for cell morphogenesis and polarized secretion in Neurosporacrassa. EMBO J., in press.

[44] Shelanski, M. L., F, Gaskin, C. R. Cantor: Microtubule assemblyin the absence of added nucleotides. Proc. Natl. Acad. Sci. USA 70,76s-768 i973\.[45] Simon, V.R., T.C.Swayne, L.A.Pon: Act in-dependent mi to-chondrial motility in mitotic yeast and cell-free systems: Identificationof a motor activity on the mitochondrial surface. J. Cell Biol. 130,34s-3s4 (1ees).[46] Steinberg, G., M. Schliwa: Organelle movement in the wild typeand wall-less fz; sg; os-1 mutants of Neurospora crassa are mediated bycytoplasmic microtubules. J. Cell Sci. 106, 555-564 (1993).

[47] Steinberg, G., M. Schliwa: The Neurospora organelle motor: Adistant relative of conventional kinesin with unconventional proper-ties. Mol. Biol. Cell 6, 1605-1618 (1995).

[,18] Steinberg, G., M.Schliwa: Characterization of the biophysicaland motility properties of kinesin from the fungus Neurospora crassa.J. Biof. Chem. Tll,7516-7521 (1996).

[49] Urrutia, R., M. A. McNiven, J. P. Albanesi, D. B. Murphey,B.Kachar: Purified kinesin promotes vesicle motility and inducesactive sliding between microtubules in vitro. Proc. Natl. Acad. Sci.usA 88, 670t-6705 (1991).[50j Vale, R.D., T.S.Reese, M.PSheetz: Identif ication of a novelforce-generation of a novel force-generating protein, kinesin, involvedin microtubule-based motility. Cell 42,39-50 (1985).

[51] Witman, G. B.: Isolation of Chlamydomonas flagella and flagel-lar axonemcs. Methods Enzymol. 134,280-290 (1986).

[52] Xiang, X. , S.M.Beckwith, R.Morr is : Cytoplasmic dynein isinvolved in nuclear migration in Aspergillus nidulans. Proc. Natl.Acad. Sci. USA ql, 2100-2104 (1994').

[53] Zhou, X.-L., M. A. Stumpf, H. C. Hoch, C. Kung: A mechano-sensitive cation channel in membrane patches and in whole cells of thefungus Uromyces. Science.2-S3 , 1415-1417 (1991).