Cell Motility and the Cytoskeleton 37:199–210 (1997) Mitochondrial Inheritance: Cell Cycle and Actin Cable Dependence of Polarized Mitochondrial Movements in Saccharomyces cerevisiae V.R. Simon, S.L. Karmon, and L.A. Pon* Department of Anatomy and Cell Biology, Columbia University, College of Physicians and Surgeons, New York, New York Asymmetric growth and division of budding yeast requires the vectorial transport of growth components and organelles from mother to daughter cells. Time lapse video microscopy and vital staining were used to study motility events which result in partitioning of mitochondria in dividing yeast. We identified four different stages in the mitochondrial inheritance cycle: (1) mitochondria align along the motherbud axis prior to bud emergence in G1 phase, following polarization of the actin cytoskeleton; (2) during S phase, mitochondria undergo linear, continuous and polarized transfer from mother to bud; (3) during S and G2 phases, inherited mitochondria accumulate in the bud tip. This event occurs concomitant with accumulation of actin patches in this region; and (4) finally, during M phase prior to cytokinesis, mitochondria are released from the bud tip and redistribute throughout the bud. Previous studies showed that yeast mitochondria colocalize with actin cables and that isolated mitochondria contain actin binding and motor activities on their surface. We find that selective destabilization of actin cables in a strain lacking the tropomyosin 1 gene (TPM1) has no significant effect on the velocity of mitochondrial motor activity in vivo or in vitro. However, tpm1D mutants display abnormal mitochondrial distribution and morphology; loss of long distance, directional mitochondrial movement; and delayed transfer of mitochondria from the mother cell to the bud. Thus, cell cycle-linked mitochondrial motility patterns which lead to inheritance are strictly dependent on organized and properly oriented actin cables. Cell Motil. Cytoskeleton 37:199–210, 1997. r 1997 Wiley-Liss, Inc. Key words: tropomyosin; organelle motility; yeast; time lapse fluorescence microscopy INTRODUCTION Cellular organelles are produced by growth and division of preexisting organelles. Therefore, organelle inheritance, or transfer of organelles from mother to daughter cells, is necessary for proliferation [for reviews see Nunnari and Walter, 1996; Warren and Wickner, 1996]. Here, we focused on the mechanisms underlying mitochondrial movement and inheritance during asymmetric cell division in the budding yeast Saccharomyces cerevisiae. Previous studies indicate that mitochondria are one of the first organelles to be transferred into the bud [Stevens, 1977], and that mutations that prevent this transfer result in cell death [McConnell et al., 1990]. r 1997 Wiley-Liss, Inc. Abbreviations: ACT1, gene encoding yeast actin; TPM1, gene encoding tropomyosin 1 protein; Tpm1p, tropomyosin 1 protein; tpm1D, deletion of the TPM1 gene; YPD, rich media for yeast growth consisting of yeast extract, peptone and dextrose. Contract grant sponsor: NRSA; Contract grant number: 5 F31 GM15644; Contract grant sponsor: Summer Undergraduate Research Fellowship from the Department of Biological Sciences at Columbia University; Contract grant sponsor: National Institutes of Health; Contract grant number: RO1 GM45735. *Correspondence to: Liza A. Pon, Department of Anatomy and Cell Biology, Columbia University P&S 12-425, 630 West 168th Street, New York, NY 10032. E-mail: email@example.com Received 3 October 1996; accepted 6 March 1997. 200 Simon et al. Thus, mitochondrial inheritance is an essential process which begins early in the cell division cycle. In vegetative yeast, mitochondria are resolved as elongated organelles which are organized as individual tubular structures and as a branching reticulum [Damsky, 1976; Stevens, 1977]. Fusion of mitochondria has been documented during zygote formation and meiosis in yeast [Miyakawa et al., 1984; Azpiroz and Butow, 1993; Smith et al., 1995] and may contribute to mitochondrial plasticity in vegetative cells. Short-term, time lapse microscopy in vegetative yeast revealed that tubular mitochondrial structures align along the mother-bud axis and undergo polarized movement from mother cells into developing buds during cell division. This inheritance-associated movement is linear and long-distance, and displays an average velocity of 50 nm/sec [Simon et al., 1995]. Upon transfer into the bud, newly inherited mitochondria are observed to move towards the bud tip, where they become immobilized [Simon et al., 1995]. This immobilization event prevents return of mitochondria into the mother cell, effectively increasing the efficiency of mitochondrial inheritance. Mitochondrial movement in vegetative yeast is dependent upon the actin cytoskeleton. The actin cytoskeleton of yeast consists of two structures detected by light and electron microscopy: actin cables and actin patches. Actin cables are bundles of actin filaments that extend along the mother-bud axis, and actin patches are believed to be invaginations in the plasma membrane invested with F-actin. Actin mutations which depolarize and destabilize the actin cables and patches result in mitochondrial aggregation, complete inhibition of mitochondrial movement, and impaired mitochondrial inheritance [Drubin et al., 1993; Lazzarino et al., 1994; Simon et al., 1995]. In addition, light and electron microscopy studies indicate that mitochondria colocalize with actin cables [Drubin et al., 1993; Lazzarino et al., 1994; Mulholland et al., 1994]. Moreover, cell free assays indicate that isolated yeast mitochondria bind to the lateral surface of actin filaments. This actin binding activity is ATPsensitive, reversible, saturable and mediated by a protein or proteins on the mitochondrial surface [Lazzarino et al., 1994]. Finally, an ATP-driven, actin-dependent motor activity has been identified on yeast mitochondrial outer membranes [Simon et al., 1995]. These observations support a model whereby mitochondrial movement from mother cells to buds is mediated by mitochondrial motor driven movement using actin cables as tracks. To test this hypothesis, we examine: (1) the temporal relations between actin and mitochondrial organization during yeast cell division, and (2) the effect of actin cable destabilization on mitochondrial movement and inheritance. Reorganization of the yeast actin cytoskeleton is required for the establishment of cell polarity and vectorial transport of cellular components from mother to daughter cells, prior to and during cell division [Lew and Reid, 1993]. We find that cell cycle-dependent changes in actin cytoskeletal organization correlate with inheritanceassociated mitochondrial motility events. The role of actin cables in mitochondrial movement and inheritance was evaluated using a strain bearing a deletion in the tropomyosin 1 gene (TPM1) [Liu and Bretscher, 1989, 1992; Drees et al., 1995]. Deletion of this gene results in selective destabilization of actin cables, without significantly affecting the abundance or polarization of actin patches. Phenotypes associated with this loss of actin cables are accumulation of late secretory vesicles, defects in polarized secretion, abnormal chitin deposition, altered cell size and shape, low growth rates, and defects in shmoo formation and cell fusion. TPM1 encodes the only yeast actin binding protein found exclusively in actin cables. This, together with the finding that overexpression of Tpm1p partially restores actin cables in a strain carrying a mutant allele of the actin gene (act1-2), indicate that Tpm1p interacts directly with F-actin and is required to stabilize yeast actin cables [Liu and Bretscher, 1989]. These findings support a role for actin cables in polarized secretion, cell growth and mating. Our studies indicate that actin cables are also required for cell cycle linked mitochondrial movements leading to inheritance. MATERIALS AND METHODS Yeast Cell Manipulation The following strains were used for this study: MSY106 (MATa/a, his4-619/his4-619), CUY58 (MATa/a, ade2/ADE2, his3D200/his3D200, leu2-3, 112/ leu2-3, 112, lys2-801/LYS2, trp1-1(am)/TRP1, ura3-52/ ura3-52) and ABY320 (MATa, ade2, his3-D200, leu2-3, 112, ura3-52, tpm1D::LEU2). Yeast cell manipulations were carried out according to Sherman . Cell Growth and Synchronization Yeast were grown to midlog phase in lactate-based liquid media at 30°C [Glick and Pon, 1995]. Cells were collected by centrifugation, resuspended in lactate media and applied to a 5–15% sorbitol gradient. Cells were then separated on the basis of density by centrifugation at 121 3 g for 4 min [Mitchison, 1988]. The top gradient fraction, which contained the largest number of unbudded and small budded cells, was transferred to glucose-based liquid media (YPD) such that the final concentration of cells was 1 3 107 cells/ml. Aliquots of this synchronized culture were removed at different time intervals and used for fixation or time-lapse video microscopy. Actin Cable Dependent Mitochondrial Movement Fixation and Immunofluorescence The fixation and immunofluorescence methods used are variations of the methods described in Pringle et al. . In brief, synchronized cells were fixed at 10-min intervals over 2 hr. To fix samples, paraformaldehyde and potassium phosphate buffer (pH 6.5) were added directly to the growth medium to final concentrations of 3.7% and 100 mM, respectively. This mixture was incubated for 15 min at room temperature (RT) and cells were collected by centrifugation. The resulting pellet was resuspended in a second fixative solution (100 mM potassium phosphate pH 6.5, 2 mM MgCl2 and 5% paraformaldehyde) and incubated for 2 hr at RT. Fixed cells were then washed three times with ‘‘wash solution’’ (20 mM potassium phosphate, pH 7.5, 0.8 M KCl). To remove the yeast cell wall, fixed cells were incubated in wash solution containing 10 mM DTT for 20 min followed by incubation in wash solution containing (0.125 mg/ml) Zymolyase 20,000T (Seikagaku, Ijamsville, MD). Fixed spheroplasts were washed with NS (20 mM Tris-HCl pH 7.6, 0.25 M sucrose, 1 mM EDTA, 1 mM MgCl2, 0.1 mM ZnCl2, 0.1 mM CaCl2, 0.8 mM phenylmethylsulfonylfluoride, 0.05% [v/v] 2-mercaptoethanol) and stored at 4°C in NS supplemented with 0.02% Na-azide. Fixed spheroplasts were applied to poly-lysinecoated coverslips. Immobilized spheroplasts were then washed in phosphate-buffered saline (PBS) and incubated in PBT (1 3 PBS, 0.1% [v/v] Triton X-100, 0.02% [v/v] sodium azide, (albumin) 1% [w/v] bovine serum albumin [BSA]). Mitochondria were visualized using polyclonal antiserum raised against total mitochondrial outer membrane proteins [Smith et al., 1995], and FITC-coupled goat anti-rabbit IgG (Boehringer Mannheim, Indianapolis, IN). The actin cytoskeleton was stained using rhodamine phalloidin (Molecular Probes, Eugene, OR). Stained spheroplasts were mounted on microscope slides using mounting solution (1 µg/ml p-phenylenediamine, 90% [w/v] glycerol, and 13 PBS) containing the DNAbinding dye 48,6-diamidino-2-phenylindole (DAPI) [Williamson and Fennell, 1975] at a final concentration of 1 µg/ml. Coverslips were then sealed onto glass microscope slides with clear nail polish. Slides were stored at 220°C in the dark prior to viewing. Vital Staining A growth chamber was prepared using a modification of methods described by Koning et al. . A 1-cm2 well was prepared on a microscope slide using a double layer of cellophane tape. The well was filled with YPD media containing 3% high melting point agarose (Sigma Chemicals, St. Louis, MO), 1 µg/ml of the membrane potential-sensing dye 3,38-dihexyloxacarbocyanineiodide (DiOC6; Molecular Probes), and oxyrase at a concentration recommended by the manufacturer (Oxy- 201 rase Inc., Mansfield, OH). A coverslip was placed over the filled well, and uniform force was applied to the coverslip to compress the agar into the well. The coverslip and excess agar were removed from the well. An aliquot of synchronized cells was resuspended in YPD media to a concentration of 2 3 108 cells/ml. 6 3 105 cells were spread across the surface of the solidified agar and allowed to sink into the agar. The chamber was then covered with a coverslip and sealed on all sides with clear nail polish. These growth chambers support cell growth at rates similar to that of cell growth in liquid culture for up to 5 hr. The length of the cell division cycle for wild type cells under these conditions is approximately 90 min, and was not significantly affected by addition of oxyrase. Therefore, oxyrase-dependent oxygen depletion has little or no effect on cell growth during the time of our visualization (,150 min). Finally, the DiOC6 staining conditions used result in mitochondria-specific staining without altering morphology or spatial organization of mitochondria. Preparation of Actin and Mitochondria From Yeast Actin was isolated from midlog phase yeast by DNase1 affinity chromatography and multiple rounds of polymerization and depolymerization as described previously [Lazzarino et al., 1994]. Highly purified mitochondria were isolated from D273-10B yeast by differential and isopycnic centrifugation using Nycodenz gradients as previously described [Lazzarino et al., 1994; Glick and Pon, 1995]. In Vitro Motility Assay The in vitro filament sliding assay is a modification of the Kron and Spudich  assay as described in Simon et al. . The experimental flow cell consists of a nitrocellulose-coated (Fullam Inc., Latham, NY) coverslip placed on a microscope slide. Two parallel strips of double-sided tape are used as spacers to create a 35–50 µl flow chamber between the coverslip and the microscope slide. Mitochondria were suspended to a final concentration of 1 mg/ml in complete AB buffer (25 mM imidazole hydrochloride, pH 7.4, 25 mM KCl, 4 mM MgCl2, 1 mM EGTA, 10 mM DTT) containing a protease inhibitors cocktail, 1 mM phenylmethyl-sulfonyl fluoride, 10 µM benzamidine, 1 µg/ml 1,10-phenanthroline, 0.5 µg/ml antipain, 0.5 µg/ml chymostatin, 0.5 µg/ml leupeptin, 0.5 µg/ml pepstatin, and 0.5 µg/ml aprotinin (Sigma Chemical Co.); and oxygen scavengers, 3 mg/ml glucose, 9.3 U/ml glucose oxidase (Sigma Chemical Co.), 18 µg/ml catalase (Calbiochem, La Jolla, CA). DiOC6 was added to a final concentration of 100 ng/ml. Mitochondria (100 µg protein) were perfused into the microscope flow cell. After incubation for 10 min at 4°C, the flow cell was washed with three volumes of complete AB/BSA 202 Simon et al. buffer (complete AB buffer containing 4 mg/ml BSA) and nonspecific protein binding sites were blocked by incubation with the AB/BSA buffer for 10 min at 4°C. Rhodamine-phalloidin-labeled yeast actin filaments (1.5 µg/ml in complete AB buffer) were perfused into the flow cell and incubated with immobilized mitochondria for 10 min at RT. Microfilament sliding and release were observed after addition of AB buffer containing 10 µM ATP and an ATP regeneration system (10 µM creatine phosphate and 250 µg/ml creatine phosphokinase; Sigma Chemical Co.). Fluorescence and Video Microscopy Cells were viewed with a Leitz Dialux microscope (Rockleigh, NJ). Rhodamine-phalloidin-labeled actin structures were viewed with excitation and emission wavelengths of 540–552 nm and 570 nm, respectively. FITC-stained yeast mitochondria and DiOC6-stained cells were viewed with excitation and emission wavelengths of 490–495 nm and 525 nm, respectively. DAPI images were viewed with excitation and emission wavelengths of 340–365 nm and 450–488 nm, respectively. Images of stained mitochondria in vivo and fluorescent mitochondria and filaments in vitro were collected for 1–2 sec at different time intervals using a cooled CCD camera (Star-1, Photometrics, Tucson, AZ). Light output from the 100W mercury arc lamp was controlled using a shutter driver (Uniblitz D122, Vincent Associates, Rochester, NY) and attenuated using neutral density filters (Omega Optical Corporation, Brattleboro, VT). Image enhancement and analysis were performed on a Macintosh Quadra 800 computer (Cupertino, CA) using the public domain program NIH Image 1.55. Images were stored on a magnetic optical disk drive (Peripheral Land Inc., Fremont, CA). Velocity Measurements For analysis of mitochondrial motility as a function of progression through the cell division cycle, time lapse series were obtained at 2.5- to 5-min intervals for periods up to 120 min. These long-term imaging studies reveal the patterns of mitochondrial motility throughout the entire cell cycle. However, since organelles move into and out of the plane of focus during 2.5- to 5-min intervals, these conditions preclude tracking of an individual organelle in consecutive still frames. Therefore, velocities of mitochondrial movement were studied in time lapse series obtained at 20-sec intervals over 10 min of real time. For these movements, velocities were determined by measuring the change in position of the end of tubular mitochondrial structures as a function of time in a time lapse series, and motile mitochondrial ends were defined as structures that stayed within one plane of focus and displayed linear, detectable movement for at least 60 sec of real time. The limit of resolution of our system is 1 pixel/20 sec or approximately 8.0 nm/sec. Mitochondrial movements which resulted in displacement less than 1 pixel, or which persisted for less than three consecutive still frames, were assigned a velocity of 0 nm/sec. The velocities of filament sliding on immobilized mitochondria were determined by measuring the change in position of the tip of each moving actin filament as a function of time in time lapse series. Time lapse series for these velocity measurements were obtained at 20-sec intervals for 5–10 min of real time. Sliding actin filaments were defined as filaments which: (1) colocalized with immobilized mitochondria, (2) moved parallel to their long axis, and (3) displayed detectable, linear motion for 60 sec of real time (three consecutive still frames). For all velocity measurement, NIH Image V1.55 was used to determine the position (x-y coordinates) of moving mitochondria or actin filaments. The linear distances between successive positions of mitochondria or microfilaments were calculated using Microsoft Excel, and instantaneous velocities were averaged to give the mean velocity. RESULTS Dynamics of Mitochondrial Inheritance Time lapse imaging of mitochondrial inheritance in dividing yeast revealed polarized, linear movement of mitochondria from mother cells to developing buds [Simon et al., 1995]. However, the methods used during those early efforts precluded long-term imaging of living cells. As a result, fundamental issues pertaining to the timing and duration of mitochondrial inheritance events were not examined. To address these issues, we developed a growth chamber and vital staining conditions which support normal kinetics of yeast cell division, and allow visualization of inheritance-associated mitochondrial movements for intervals equal to or longer than one full cell division cycle (see Materials and Methods). Analysis of mitochondrial motility patterns as a function of progression through the cell division cycle reveals a sequence of motility events leading to partitioning of mitochondria in mother and daughter cells. Mitochondrial tubules are oriented along the mother-bud axis throughout cell division. This organization persists in mother cells after completion of cell division and cell separation (Fig. 1A). Reorganization of these organelles Actin Cable Dependent Mitochondrial Movement 203 (Fig. 1F– H). This transfer of mitochondria is produced by directed, linear movement of mitochondria from the mother cell to the bud (Table I). Newly inherited mitochondria move to the bud tip and accumulate in this region (Fig. 1G, arrowhead). This accumulation is first detected 30–35 min after bud emergence (Fig. 2D), and can persist up to 55–65 min after bud emergence. When the bud is approximately 70% of the size of the mother cell, all inherited mitochondria are immobilized at the bud tip (Fig. 2E, arrow). At this stage, no further transfer of mitochondria from the mother cell to the bud is observed. Mitochondria are then released from the bud tip (Fig. 2F–H). Upon release, mitochondria redistribute throughout the bud but do not reenter the mother cell. Cell septation occurs within 20 min of mitochondrial release from the bud tip, 90–100 min after bud emergence. Mitochondrial Organization and Movements During Inheritance Correlate With Cell Cycle-Regulated Actin Rearrangements Fig. 1. A–H: Mitochondrial movements leading to inheritance in wild type cells. Synchronized wild type cells (MSY106) were immobilized on a microscope growth chamber, which contained the membrane potential sensing dye DiOC6. Time lapse images of mitochondria in living cells were obtained at 2.5-min intervals over a period of 102 min. The times shown in each still frame correspond to minutes of real time, where t 5 0 min was defined as the time of bud emergence. Mitochondria are resolved as tubular structures (C, open arrow) which reorganize to converge at the site of bud emergence (A, arrow), enter the bud, and accumulate in the bud tip (G, arrowhead). Bar 5 1 µm. occurs upon initiation of the next round of cell division, 10–15 min prior to bud emergence (Fig. 1B). During reorganization, mitochondria converge at a new site on the cell surface, the site of bud emergence (Fig. 1C, arrow). A single mitochondrial tubule enters the bud within 2–5 min after bud emergence (Fig. 1D). Mitochondria continue to move linearly into the bud as it grows In yeast, the actin cytoskeleton undergoes a defined sequence of cell cycle-regulated rearrangements [Adams and Pringle, 1984; Kilmartin and Adams, 1984]. A ring of actin patches forms at the presumptive bud site (Fig. 3A). Shortly thereafter, actin cables align along the long axis of the mother cell. After bud emergence, actin patches remain within the bud, and actin cables persist along the mother bud axis (Fig. 3D). During early stages of bud enlargement, growth is directed towards the apical region of the bud. At this stage, actin patches concentrate in the bud tip (Fig. 3G). Thereafter, bud growth is isotropic and directed to the entire surface of the bud. This apical to isotropic switch is associated with redistribution of actin patches from the bud tip to the entire surface of the bud (Fig. 3J). Finally, redistribution of cortical actin structures to the neck region occurs prior to cytokinesis (Fig. 3M) [Lew and Reed, 1993]. We evaluated the temporal relation between mitochondrial inheritance events and cell cycle-regulated actin rearrangements. To do so, we fixed synchronized yeast cultures at various stages of cell division, and determined the organization of mitochondria, actin and nuclei. We find that accumulation of a ring of actin patches at the presumptive bud site precedes convergence of mitochondria at that site (Fig. 3A–C). Transfer of mitochondria from mother cell to bud begins shortly after bud emergence. At this stage, actin patches are localized exclusively in the bud, and actin cables are aligned along the mother-bud axis where they colocalize with mitochondria (Fig. 3D–F). Immobilization of mitochondria in the 204 Simon et al. TABLE I. The Effect of TPM1 Deletion on Mitochondrial Movement and Inheritance* Strain MSY106 (WT) CUY58 (TPM1) ABY320 (tpm1D ) Motile mitochondria (%) Velocity (nm/sec) Polarized Bud tip Inheritance in movement (%) accumulation (%) small buds (%) 55 54 26 49 6 21 (n 5 200) 54 6 17 (n 5 65) 36 6 12 (n 5 36) 59 55 34 65 56 23 95 90 29 *Mitochondrial movement in wild type cells (CUY58) and in cells bearing a deletion in the TPM1 gene (ABY320) was visualized as for Figure 1. The percentage of motile mitochondria represents the mitochondria that showed a net displacement during the time recorded. Polarized movement refers to the percentage of motile mitochondria that showed a net displacement towards the bud. Measurements on the extent, velocity and degree of polarity of mitochondrial movement were determined using time lapse images recorded at 20-sec intervals. Bud tip accumulation was quantified as the percentage of medium-sized and large buds that contained mitochondria at the bud tip. The efficiency of mitochondrial inheritance was determining by measuring the percentage of small buds (1–1.5 µm) which contain mitochondria. n, no. of mitochondria studied. Fig. 2. A–H: Mitochondrial accumulation at the bud tip contributes to efficient mitochondrial inheritance. Mitochondrial movements in synchronized cells were visualized as described in Figure 1. Time lapse images were obtained at 2.5-min intervals for a total of 95 min. The times shown in each still frame correspond to minutes of real time, where t 5 0 min was defined as the time of bud emergence. Mitochondria are observed to move into the bud (C) and accumulate at the bud tip (D and E, arrows). Release of mitochondria from the bud tip is illustrated in F and G. Subsequent to release, mitochondria repopulate the bud, but do not reenter the mother cell. Bar 5 1 µm. bud tip occurs concomitant with actin patch accumulation in this region (Fig. 3G–I). Release of immobilized mitochondria from the bud tip occurs subsequent to redistribution of actin patches within the bud, and prior to nuclear inheritance (Fig. 3J–L). Finally, depolarization of actin cables and accumulation of actin patches at the site of cell-cell separation occur concomitant with redistribution of mitochondria throughout the bud (Fig. 3M–O). Actin Cable Dependent Mitochondrial Movement Fig. 3. Mitochondrial inheritance events during cell division cycle. Synchronized wild type cells (MSY106) were fixed at different time points (see Materials and Methods). In this triple label experiment, actin structures were visualized using rhodamine phalloidin (A, D, G, J, M); mitochondria were visualized using an antibody 205 specific to the mitochondrial outer membrane and an FITC-coupled secondary antibody (B, E, H, K, N), and mitochondrial and nuclear DNA were stained using the DNA binding dye, DAPI (C, F, I, L, O). Arrows point to examples of mitochondrial colocalization with actin cables. m, mitochondrion, n, nuclei. Bar 5 1µm. 206 Simon et al. Selective Destabilization of Actin Cables Results in Defects in the Mitochondrial Inheritance Cycle To determine the relative contributions of actin cables and actin patches in inheritance-associated mitochondrial movements, we studied mitochondrial motility and morphology in a TPM1 deletion mutant (tpm1D). Actin cables are not detected in tpm1D mutants at any stage of cell division. Nonetheless, the abundance and reorganization of actin patches in the tpm1D mutant are similar to those observed in wild type cells. Actin patches accumulate at the presumptive bud site, within the bud and at the tips of growing buds in the tpm1D mutant (Figs. 4A,D and G). Actin patch redistribution in large buds (Fig. 4J) and accumulation at the site of cell-cell separation are also evident in the tpm1D mutant (Fig. 4M). Thus, cell cycle-dependent polarization of actin patches is not defective in these cells. Defects in mitochondrial morphology, inheritance and motility are also observed in tpm1D cells (Fig. 4, Table I). Mitochondria in tpm1D mutants are resolved as abnormal short tubular structures or aggregates in these cells (Fig. 4). Deletion of the TPM1 gene also results in loss of alignment of mitochondria along the mother-bud axis. Mitochondria are detected in the bud in tpm1D mutants; however, we observe a significant delay in transfer of mitochondria from mother cells to buds. Approximately 90% of small buds in a wild type strain contain mitochondria. In contrast, only 29% of small buds in the tpm1D culture contain mitochondria (Table I). Thus, deletion of TPM1 results in a 3-fold decrease in the efficiency of mitochondrial inheritance during early stages of the cell division cycle. Finally, mitochondria within the bud are aggregated and show no significant accumulation in the bud tip. The velocity of mitochondrial movement in the TPM1 mutant is similar to that in wild type cells (Table I). However, time lapse image analysis of the tpm1D strain reveals fundamental differences in the level of mitochondrial motility and in the pattern of mitochondrial movement. In these studies, motile mitochondria were defined as particles that undergo positional displacement over several minutes of real time. We observe a 2-fold decrease in the percentage of motile mitochondria in the tpm1D mutant compared to isogenic wild type cells (Table I). Moreover, mitochondria in tpm1D cells undergo frequent changes in direction (Fig. 5). This is distinct from the prolonged, linear excursions of mitochondria in wild type cells, which average 2.5 µm within an interval of 1.4 min. As a result, deletion of TPM1 causes a 2-fold decrease in the level of bud-directed mitochondrial movement (Table I). Mitochondrial Motor Activity In Vitro is not Affected by Deletion of TPM1 To investigate whether the mitochondrial motor activity was affected by deletion of TPM1, we compared microfilament sliding activity on mitochondria isolated from tpm1D and wild type cells (see Materials and Methods). Mitochondria isolated from tpm1D mutants maintain a membrane potential, as determined by staining with the membrane potential sensing dye, DiOC6. In addition, mitochondria from tpm1D cells bind rhodaminephalloidin-labeled yeast actin filaments in the absence of ATP (data not shown). Finally, mitochondria isolated from tpm1D cells support ATP-dependent, microfilament sliding with velocities similar to those supported by mitochondria from wild type cells (Table II). These experiments suggest that the defects in mitochondrial organization, movement and inheritance observed in the tpm1D mutant are not due to the loss of mitochondrial motor activity. DISCUSSION Cell division in yeast occurs by asymmetric growth of the developing daughter cell or bud. First, a bud site is selected. Thereafter, reorganization of the yeast cytoskeleton results in establishment of cell polarity and vectorial transport of organelles and other cellular components to the bud. Recent studies document a role for Cdc28 and cyclins in reorganization of the actin cytoskeleton in dividing yeast, and further define the phase during cell division in which each actin rearrangement occurs [Lew and Reed, 1993, 1995]. Polarization of actin patches to the site of bud emergence and alignment of actin cables along the mother-bud axis during G1 phase are triggered by G1 cyclin (Cln1/2)-mediated activation of Cdc28. Subsequent depolarization of the actin patches and cables occurs at the interface of G2 and M phases. This event is dependent upon activation of Cdc28 by mitotic cyclins (Clb1/2). Finally, redistribution of cortical actin structures to the site of cell-cell separation occurs at the end of M phase in response to destruction of mitotic cyclins and Cdc28 inactivation. Using long-term time lapse video microscopy to visualize mitochondrial movements through the entire cell cycle, we define the ‘‘mitochondrial inheritance cycle,’’ a sequence of mitochondrial motility events which partition mitochondria between mother and daughter cells during yeast cell division. These inheritanceassociated mitochondrial motility events correlate temporally with cell cycle-regulated actin reorganization. The hallmarks of the mitochondrial inheritance cycle are: (1) convergence of mitochondria at the site of bud emergence following polarization of the actin cytoskeleton in G1 phase, (2) linear, directed movement of mitochondria from mother cells into developing buds in S phase, (3) accumulation of newly inherited mitochondria and actin patches at the bud tip during S and G2 phases, and (4) release of immobilized mitochondria from the bud tip Actin Cable Dependent Mitochondrial Movement Fig. 4. Mitochondrial inheritance in cells lacking actin cables. Cells bearing a deletion in the tropomyosin 1 gene (ABY320) were synchronized, fixed at different time points and stained as for Figure 3. Visualization of actin (A, D, G, J, M) reveals normal patterns of actin patch localization, and absence of actin cables in the tropomyosin 1 deletion mutant. Mitochondria (B, E, H, K, N) and mitochondrial and 207 nuclear DNA (C, F, I, L, O) localization in the same cells are shown. Small buds devoid of mitochondria are observed (G–I, and J–L) suggesting a delay in mitochondrial inheritance. Mitochondrial morphology and organization are defective in mother cells and buds. m, mitochondrion; n, nuclei; and b, bud. Bar 5 1 µm. 208 Simon et al. Fig. 5. Actin cables are required for linear, long distance polarized mitochondrial movements. Mitochondrial movement in wild type (MSY106; A) and tpm1D cells (ABY320; B) was visualized as for Figure 1. Tracings of the movements of individual mitochondria in DiOC6-stained cells were made by marking the position of the tip of motile organelles during the time in which they remained in the plane of focus. The points denote the position of organelles at 20-sec intervals. Tracings are shown relative to the boundary of the dividing yeast cell. Bar 5 1 µm. TABLE II. Mitochondria Isolated From the Tpm1D Mutant Support Actin Filament Sliding on Their Surface* Strain CUY58 (TPM1) ABY320 (tpm1D ) Velocity of filament sliding (nm/sec) 36 6 14 (n 5 20) 33 6 16 (n 5 20) *To evaluate motor activity, mitochondria were isolated from wild type (CUY58) and tropomyosin mutant (ABY320) strains. The velocity of ATP-dependent mitochondrial motor activity was determined using the microfilament sliding assay (see Materials and Methods). n, no. of filaments measured. concomitant with redistribution of actin patches throughout the bud during M phase (Fig. 6). Previous studies on mitochondrial inheritance in budding yeast support roles for an intermediate filamentlike protein (Mdm1p), a fatty acid desaturase (Mdm2p), and specific mitochondrial outer membrane proteins (Mmm1p and Mdm10p) in transfer of mitochondria from mother cells to buds [McConnell and Yaffe, 1992; Stewart and Yaffe, 1991; Sogo and Yaffe, 1994; Burgess et al., 1994]. Our findings indicate that the mitochondrial inheritance cycle consists of at least four cell cycle-linked mitochondrial motility events. Therefore, it is possible that MDM1, MDM2, MDM10, and MMM1 may exert direct or indirect effects on one or more of these mitochondrial motility events. Findings in various laboratories also implicate the actin cytoskeleton in mitochondrial movement and inheritance [Drubin et al., 1993; Lazzarino et al., 1994; Simon Fig. 6. The mitochondrial inheritance cycle. See text. et al., 1995; Smith et al., 1995]. Indeed, we observe colocalization of tubular mitochondria with actin cables during all stages in the mitochondrial inheritance cycle in which polarized mitochondrial movement occurs. Therefore, we used a tropomyosin 1 deletion mutation to examine the effect of selective destabilization of actin cables on inheritance-associated mitochondrial movement. Deletion of the TPM1 gene causes loss of actin cables without affecting the abundance or cell cycledependent reorganization of actin patches. We find that this mutation also produces severe defects in mitochondrial morphology, organization, motility patterns and delayed inheritance. Mitochondria in the tpm1D mutant are resolved as abnormal short tubular structures or aggregates which do not align along the mother-bud axis, and do not display linear, directional movements. Retention of mitochondria in the bud tip is also compromised in the tpm1D mutant. Finally, we observe a delay in mitochondrial inheritance in tpm1D mutants: deletion of TPM1 results in a 3-fold decrease in the efficiency of mitochondrial inheritance during the S phase of the cell division cycle. Since mitochondria isolated from tpm1D mutants display F-actin binding activity and normal velocities of ATP-dependent microfilament sliding, these defects are not due to the inability of mitochondria to bind filamentous actin. It is possible that the abnormal mitochondrial phenotypes observed are due to an indirect effect of the TPM1 deletion. For example, vesicles which are known to accumulate in tpm1D mutants may block transfer of mitochondria across the bud neck, or interaction of mitochondria with actin structures. We favor an alternative interpretation. It is possible that deletion of TPM1 Actin Cable Dependent Mitochondrial Movement destabilizes actin cables and disrupts the organization of filamentous actin structures. Thus, mitochondria move along the disorganized arrays of actin filaments in a nondirected manner. Loss of directed mitochondrial movement could, in turn, decrease the efficiency of mitochondrial inheritance. This interpretation is supported by the observations that deletion of TPM1 alters the pattern of mitochondrial movement without affecting either the velocity of mitochondrial movement in vivo, or the velocity of mitochondria driven microfilament sliding in vitro. In summary, our findings indicate that a defined sequence of mitochondrial reorganization, mobilization and immobilization events results in high-efficiency, directed transfer of mitochondria from mother cells to developing buds, retention of newly inherited mitochondria within bud tips, and redistribution of mitochondria in the fully formed bud. These events correlate with cell cycle-regulated changes in organization of the actin cytoskeleton, and are severely compromised upon deletion of TPM1 and loss of actin cables. The observed defects are not due to TPM1 deletion-dependent loss of mitochondrial DNA, or loss of the mitochondrial motor activity. Rather, it appears that selective destabilization of actin cables perturbs mitochondrial morphology and the mitochondrial inheritance cycle. This, together with the observations that (1) mitochondria colocalize with polarized actin cables, and (2) mitochondria contain an actindependent motor on their surface [Simon et al., 1995], support the model that mitochondria use polarized actin cables as tracks for alignment prior to bud emergence, and subsequently for directed movement from the mother cell to the developing bud. Future studies will focus on the role of cell cycle regulators in control of mitochondrial movement and inheritance. ACKNOWLEDGMENTS We thank I. Boldogh, M. Smith, T. Swayne and Dr. A.J. Silverman for support and critical comments on the manuscript. We also thank Dr. A. Bretscher for yeast strains. This work was supported by a pre-doctral NRSA (5 F31 GM15644) to V.R.S., a Summer Undergraduate Research Fellowship from the department of Biological Sciences at Columbia University to S.L.K. and a Research Grant (RO1 GM45735) from the National Institutes of Health. REFERENCES Adams, A.E.M., and Pringle, J.R. (1984): Relationship of actin and tubulin distribution to bud growth in wild-type and morphogenetic-mutant Saccharomyces cerevisiae. J. Cell Biol. 98: 934–945. 209 Azpiroz, R., and Butow, R.A. (1993): Patterns of mitochondrial sorting in yeast zygotes. Mol. Biol. Cell. 4:21–36. Burgess, S.M., Delannoy, M., and Jensen, R.E. (1994): MMM1 encodes a mitochondrial outer membrane protein essential for establishing and maintaining the structure of yeast mitochondria. J. Cell Biol. 126:1375–1391. Damsky, C.H. (1976): Environmentally induced changes in mitochondria and endoplasmic reticulum of Saccharomyces carlsbergensis yeast. J. Cell Biol. 71:123–135. Drees, B., C. Brown, B., Barrell, G., and Bretscher A. (1995): Tropomyosin is essential in yeast, yet the TPM1 and TPM2 products perform distinct functions. J. Cell Biol. 128:383–392. Drubin, D.G., Jones, H.D., and Wertman, K.F. (1993): Actin structure and function: Roles in mitochondrial organization and morphogenesis in budding yeast and identification of the phalloidinbinding site. Mol. Biol. Cell. 4:1277–1294. Glick, B.S., and Pon, L.A. (1995): Isolation of highly purified yeast mitochondria. Methods Enzymol. 260:213–223. Kilmartin, J.V., and Adams, A.E.M. (1984): Structural rearrangements of tubulin and actin during the cell cycle of the yeast Saccharomyces cerevisiae. J. Cell Biol. 98:922–933. Koning, A.J., Lum, P.K., Williams, J.M., and Wright, R. (1993): DiOC6 staining reveals organelle structure and dynamics in living yeast cells. Cell Motil. Cytoskel. 25:111–128. Kron, S.J., and J. Spudich. (1986): Fluorescent actin filaments move on myosin fixed to a glass surface. Proc. Natl. Acad. Sci. (U.S.A) 83:6272–6276. Lazzarino, D., Boldogh, I., Smith, M.G., Rosand, J., and Pon, L.A. (1994): Yeast mitochondria contain ATP-sensitive, reversible actin-binding activity. Mol. Biol. Cell. 5:807–818. Lew, D.J., and Reed, S.I. (1993): Morphogenesis in the yeast cell cycle: Regulation by Cdc28 and cyclins. J. Cell Biol. 120:1305–1320. Lew, D.J., and Reed S.I. (1995): A cell cycle checkpoint monitors cell morphogenesis in budding yeast. J. Cell Biol. 129:739–749. Lew, D.J., and Reed, S.I. (1995): Cell cycle control of morphogenesis in budding yeast. Curr. Op. Gen. Dev. 5:17–23. Liu, H., and Bretscher A. (1989): Disruption of the single tropomyosin gene in yeast results in the disappearance of actin cables from the cytoskeleton. Cell 57:233–242. Liu, H., and Bretscher, A. (1992): Characterization of TPM1 disrupted yeast cells indicates an involvement of tropomyosin in directed vesicular transport. J. Cell Biol. 118:285–299. McConnell, S.J., and Yaffe, M.P. (1992): Nuclear mitochondrial inheritance in yeast depends on novel cytoplasmic structures defined by the Mdm1 protein. J. Cell Biol. 118:385–395. McConnell, S.J., Stewart, L.C., Talin A., and Yaffe, M.P. (1990): Temperature-sensitive yeast mutants defective in mitochondrial inheritance. J. Cell Biol. 111:967–976. Mitchison, J.M. (1988): Synchronous cultures and age fractionation. In: Campbell, I., and Duffus, J.H. (eds.) ‘‘Yeast: A Practical Approach.’’ Oxford: IRL Press, pp. 51–63. Miyakawa, I., Aoi, H., Sando, N., and Kuroiwa, T. (1984): Fluorescence microscopic studies of mitochondrial nucleoids during meiosis and sporulation in yeast, Saccharomyces cerevisiae. J. Cell Sci. 66:21. Mulholland, J., Preuss, D., Moon A., Wong A., Drubin D., and Botstein D. (1994): Ultrastructure of the yeast actin cytoskeleton and its association with the plasma membrane. J. Cell Biol. 125:381– 391. Nunnari, J., and Walter, P. (1996): Regulation of organelle biogenesis. Cell 84:389–394. Pringle, J.R., Preston, R.A., Adams, A.E.M., Stearns, T., Drubin, D.G., and Haarer, B.K. (1989): Fluorescence microscopy methods for yeast. Methods Cell Biol. 31:357–435. 210 Simon et al. Sherman, F. (1991): Getting started with yeast. Methods Enzymol. 194:3–21. Simon, V.R., Swayne, T.C., and Pon, L.A. (1995): Actin-dependent mitochondrial motility in mitotic yeast and cell-free systems: Identification of a motor activity on the mitochondrial surface. J. Cell Biol. 130:345–354. Smith, M.G., Simon, V.S., O’Sullivan, H., and Pon L.A. (1995): Organelle-cytoskeletal interactions: Actin mutations inhibit meiosis-dependent mitochondrial rearrangement in the budding yeast, Saccharomyces cerevisiae. Mol. Biol. Cell. 6:1381–1396. Sogo, L.F., and Yaffe, M.P. (1994): Regulation of mitochondrial morphology and inheritance by Mdm10p, a protein of the mitochondrial outer membrane. J. Cell Biol. 126:1361–1373. Stevens, B.J. (1977): Variation in number and volume of the mitochondria in yeast according to growth conditions. A study based on serial sectioning and computer graphics reconstitution. Biologie Cellulaire. 28:37–56. Stewart, L.C., and Yaffe, M.P. (1991): A role for unsaturated fatty acids in mitochondrial movement and inheritance. J. Cell Biol. 115:1249–1257. Warren, G., and Wickner, W. (1996): Organelle inheritance. Cell 84:395–400. Williamson, D.H., and Fennell, D.J. (1975): The use of a fluorescent DNA-binding agent for detecting and separating yeast mitochondrial DNA. Methods Cell Biol. 12:335–351.