JOURNAL OF STRUCTURAL BIOLOGY 113, 217-224 (1994)

Centrosome Behavior under the Action of a Mitochondrial

Uncoupler and the Effect of Disruption of Cytoskeleton Elements

on the Uncoupler-lnduced Alterations

I. B. ALIEVA AND I. A. VOROBJEV1

A. N. Belozersky Institute of Physico-Chemical Biology, Moscow State University, Moscow 119899, Russia

Received October 3, 1994, and in revised form February 6, 1995

Carbonyl cyanide p-(trifluoromethoxy)phenylhydra-zone (FCCP) induced in pig kidney embryo cells a loss of rhodamine 123 staining of mitochondria in 2-3 min. Within 5 min after FCCP inoculation of cells prestained with rhodamine 123, the diffuse staining of the cytoplasm was absent. FCCP did not induce changes in the cyto-plasmic microtubule complex, but induced nonrandom (preferentially perpendicular to the substrate surface) orientation of maternal centrioles. Nonrandom orientation of maternal centrioles occurred 10 min after treatment and remained for 2 hr. At 30 min after introduction of the drug, FCCP treatment increased the mean number of pericentriolar satellites on maternal centrioles and the frequency of primary cilia. The percentage of centrioles perpendicular to the substrate induced by FCCP treat-ment was slightly increased by disruption of microtu-bules and slightly diminished by disruption of microfila-ments. In both cases centrioles were oriented signify-cantly differently from random (P < 0.01). These results suggest that microtubules are neither involved in the sig-naling pathway from plasma membrane to the centriole, nor do they anchor the centrioles perpendicular to the substrate, as proposed by Albrecht-Buehler and Bushnell (Experimental Cell Research 120, 1979). 1994 Academic Press, Inc.

INTRODUCTION Centrosomes play a key role in maintaining the

microtubule system within cultured cells. Although microtubule behavior has been described in detail, almost nothing is known about the intrinsic properties of an essential component of the centrosome, the centrioles.

Centrioles were first described more than a cen-tury ago (Boveri, 1887; van Beneden and Neyt, 1887). The remarkable fine structure of the centri-oles seen by electron microscopy does not per se help us to understand their functions. Unlike other cen-

1 To whom correspondence should be addressed. Fax: (095) 939-31-81.

trosome components, centrioles exist throughout the cell cycle, undergoing changes during their replication and maturation (Kuriyama and Borisy, 1981; Vorobjev and Chentsov, 1982; Rieder and Borisy, 1982). The unique structure of centrioles has led to several speculations on their role in regulation of cell behavior (Bornens, 1979; Albrecht-Buehler, 1990,1992), but these ideas have little experimental support.

Their behavior in division, which involves their precise segregation to daughter cells through their association with the mitotic poles, leaves the basic proposition that centrioles are reserve basal bodies allowing both offspring to develop cilia if so required at some later stage in their subsequent life history (Pickett-Heaps, 1969; Wheatley, 1982).

An important step toward a better understanding of the role of the centriole in a cell would be to con-centrate on some of its special features or properties in relation to cell behavior. It was shown that, in certain cases, active (maternal) centrioles can be ori-ented predominantly perpendicular to the substrate (Vorobjev and Chentsov, 1982; Gudima et al., 1983a). Studies of the centrosome in mouse embryo and 3T3 fibroblasts provided evidence for the corre-lation of the nonrandom orientation of active centrioles with an increase in the number of microtubules radiating from the centrosome (Gudima et al., 1983a,b, 1986).

By using stereological analysis, it has been possible to obtain quantitative data on the number of centrosome-associated microtubules in tissue culture cells. It was shown that a variety of drugs, including 2,4-dinitrophenol (DNP), DNP with deoxy-glucose, sodium azide, and ouabain, can "activate" the centrosome, causing an increase in the number of microtubules associated with and radiating from it (Alieva et al., 1992). Under the action of these drugs, with the exception of sodium azide, the orientation of active centrioles to the substrate surface was nonrandom, i.e., it was preferentially perpen-

1047-8477/94 $6.00 Copyright © 1994 by Academic Press, Inc.

All rights of reproduction in any form reserved.

217

218 ALIEVA AND VOROBJEV

dicular to the plane of the substrate. The data sug-gest that this orientation is independent of the ATP level in the cell which sharply decreases under all these treatments (Spurck et al., 1986; Bershadsky and Gelfand, 1981), but could be associated with de-polarization of the plasma membrane (Alieva and Vorobjev, 1995). Depolarization of the plasma mem-brane might be achieved by a variety of treatments. For example, ouabain is a very specific agent, but its effect depends on the cell type (Hulser et al., 1974). Protonophores like DNP, СССР, or carbonyl cyanide p-(trifluoromethoxy)phenylhydrazone (FCCP) are acting less specificly affecting all charged mem-branes, but more directly increasing membrane con-ductivity (Nicholls, 1982). Among a variety of drugs inducing loss of the membrane potential FCCP is characterized as most effective (Nicholls, 1982).

The mechanism involved in turning centrioles re-mains unknown. Since active centrioles in inter-phase cells always have microtubules associated with them, the perpendicular orientation of centri-oles to the substrate is thought to be due to their anchorage in the cytoplasm with the aid of long mi-crotubules running parallel to the surface (Albrecht-Buehler and Bushnell, 1979). This supposition ap-pears to be supported by our evidence that all the drugs inducing nonrandom orientation of the centri-oles also increase the number of microtubules radi-ating from the centrosome (Alieva et al., 1992; Alieva and Vorobjev, 1995).

Microfilaments were also shown to participate in centriole movement in the cytoplasm (Sherline and Mascardo, 1982; Euteneuer and Schliwa, 1985). Re-cently centractin, an actin-like protein found in the centrosome, was thought to be involved in the asso-ciation between centrioles and the microfilament network (Clark and Meyer, 1992). Therefore, micro-filaments may also be involved in the process which leads to the perpendicular orientation of centrioles. To gain insight into the mechanisms regulating the behavior of centrioles, we have investigated whether microtubules and microfilaments do affect the in-duced perpendicular orientation of centrioles. In ex-periments with pig kidney embryo (PK) cells, FCCP induced nonrandom (preferentially perpendicular) orientation of maternal (active) centrioles to the substrate independently of drugs known to depoly-merize microtubules or microfilaments.

MATERIALS AND METHODS

Tissue culture and drug treatment. PK cells were grown on M199 tissue culture medium supplemented with 10% bovine se-rum and antibiotics (penicillin + streptomycin). Rhodamine staining of cultured cells was performed as described elsewhere (Alieva and Vorobjev, 1989). FCCP (20 mM) was added to the culture medium. Cells were fixed for electron microscopy 5,10, 30 min and 2 hr after the introduction of the drug. In the second experimental series, cells were treated with 10 µg/ml nocodazole for 1 hr, or with 10 µg/ml cytochalasin В for 1 h, prior to incu-

bation with 20 mM FCCP for 30 min. After that time, cells were fixed for electron microscopy.

Immunofluorescence. Immunofluorescent staining of the mi-crotubules was performed according to the standard indirect method (Bershadsky and Gelfand, 1981). Cells grown on the cov-erslips were lysed in the microtubule stabilizing medium, fixed with a 0.5% glutaraldehyde on phosphate-buffered saline (PBS) processed with borohydrid, and stained with primary anti-a-tubulin monoclonal antibodies (Sigma) and then with anti-mouse FITC-conjugated IgG (Sigma). Photographs were taken on an Op-ton Photomicroscope using Planapochromate 63/1.4 and an FITC filter set (No. 09).

Electron microscopy. For electron microscopy, cells were pre-pared as described elsewhere (Vorobjev and Chentsov, 1982). Cells grown on the coverslips were fixed with 2.5% glutaralde-hyde in phosphate-buffered saline (pH 7.2), postfixed with 1% osmium tetroxide on the same buffer, stained with uranyl acetate, dehydrated, and embedded in Epon 812 mixture. Serial silver sections were obtained on an LKB-III ultramicrotome (LKB, Swe-den). By using the tilted head of the ultramicrotome the angle of sectioning was adjusted parallel to the substrate. Actually, the section plane was inclined to the substrate at an angle less than 1.5°. Sections were mounted on single slot grids coated with Formvar film, stained with lead citrate, and then examined and photographed in HU-11B and HU-12 electron microscopes (Hita-chi, Japan) operating at 75 kV. For the measurements of the centriole, microtubule negatives were taken at a magnification of x20 000.

The orientation of centrioles to the substrate was determined according to the method of Albrecht-Buehler and Bushnell (1979), which allows one to calculate the angle between the cen-triolar cylinder axis and the substrate by the length of projection of centriolar microtubules in the plane of the section. Histograms of the distribution of centriole projections were made as described elsewhere (Vasiliev et al., 1988). Class value (difference between minimal and maximal projection length) was taken 0.03 µm (i.e., first class, 0-0.03 µm; second, 0.03-0.06 µm, etc.). Taking the section thickness to be about 70 nm, that gives an inclination angle for the centrioles in the first class of the histogram (which are named below "perpendicular to the substrate") of >74°. The difference between treatments was compared by the x* criterion as previously described (Vasiliev et al., 1988).

RESULTS Light Microscopy

AT the concentration used, FCCP induced a loss of rhodamine 123 staining of mitochondria in 2-3 min (Alieva and Vorobjev, 1990). Mitochondria were nei-ther capable of accumulating dye in the presence of FCCP nor did they retain rhodamine 123 after in-troduction of FCCP. In 5 min after inoculation with FCCP of the cells prestained with rhodamine 123, the diffuse staining of the cytoplasm (often observed after the action of uncouplers of oxidative phosphor-ylation; D. B. Zorov, personal communication) was also eliminated, demonstrating complete absence of plasma membrane potential (data not shown).

Figures la and Ib show immunofluorescent stain-ing of microtubules in normal PK cells and in PK cells after nocodazole treatment. Antibodies re-vealed a dense meshwork in normal cells, while 1 hr after nocodazole inoculation of the cells single mi-crotubules remained in some of them. Typically, only weak staining of the centrosome was evident (Fig. 1b).

CENTROSOMES, MICROTUBULES, AND A MITOCHONDRIAL UNCOUPLER 219

FIG. 1. (a) Immunofluorescent staining of microtubules in PK cells using an anti-α-tubulin antibody. Bar, 5 µm. (b) Anti-a-tubulin antibody staining of PK cells after nocodazole treatment (10 µg/ml, 1 hr). Bar, 5 µm.

Electron Microscopy

The ultrastructure of centrosomes in normal and drug-treated cells was studied using a complete series of sections including both centrioles in a cell studied. To describe the centrosome several parameters were used: the presence of a primary cilium; the number of pericentriolar satellites on the active (maternal) centriole; the presence of striated rootlets; the distance between two centrioles; and the orientation of the centrioles to the substrate. The effect of FCCP on the centrosome was studied starting from 5 min after introduction of the drug.

Centrosomes in control cells. The fine structure of the centrosome in normal PK cells has been de-scribed in detail previously (Alieva and Vorobjev, 1989). A total of 1-2 pericentriolar satellites were situated on the active centriole; sometimes they were not present. One-third of the cells possessed a short primary cilium. One-quarter of the cells had striated rootlets; in 3 of the 25 cells studied by com-plete serial sectioning these rootlets ran from both

centrioles (Table I). Only 2 of 31 active centrioles lay perpendicular to the substrate. The histogram of the distribution of active centriole inclination to the substrate was similar to the random distribution histogram (Fig. 2a) and this difference was statisti-cally insignificant according to the χ2 criterion.

Centrosomes after FCCP treatment. At 5 min after the introduction of FCCP into the culture medium (34 cells studied) as well as after more prolonged treatment the position of the centrosome toward the nucleus and the relative of each pair of centrioles did not change compared with control cells. The mean number of satellites on the active centriole was two. Under FCCP treatment the percentage of cells with the primary cilium increased, sometimes considerably. The cilia could be relatively long—up to 1 µm (i.e., relative to the short stubby bases most usually seen in controls). The percentage of cells with striated rootlets remained close to that in the controls (Table I). The mean angle of inclination of active centrioles to the substrate slightly increased. Six of 33 active centrioles studied

TABLE I Centrosome Changes under FCCP, Nocodazole, and Cytochalasin В Together with FCCP Action

Drugs, time of treatment

Number of cells studied

Mean number of satellites (mean ± SD)

Percentage of the cells with primary cilium

Percentage of cells with striated rootlets

Control 25 1.57 ± 0.15 29 25 FCCP, 5 min 34 2.00 ± 0.21* 55 29 FCCP, 10 min 32 2.58 ± 0.10* 26 23 Nocodazole, 60 min + FCCP 30 min

35 1.09 ± 0.16* 35 34

Cytochalasin B, 60 min + FCCP 30 min

38 1.24 ± 0.18 55 38

* Number of satellites significantly different from that for control cells.

220 ALIEVA AND VOROBJEV

FIG. 2. (a) Histograms of the distribution of length of projec-

tions of the active centrioles in PK cells after short FCCP treat-ment. From front to back: theoretical histogram of randomly di-rected centrioles (from Vasiliev et al., 1988); control cells; after 5 min FCCP treatment, (b) Histograms of the distribution of length of projections of the active centrioles in PK cells after prolonged FCCP treatment. From front to back: after 10 min FCCP treat-ment; after 30 min FCCP treatment; after 2 hr FCCP treatment. (c) Histograms of the distribution of length of projections of the active centrioles in PK cells after selective depolymerization of cytoskeletal elements under the 30-min action of FCCP. From front to back: after 30 min FCCP treatment; after nocodazole treatment (10 µg/ml, 1 hr) then FCCP + nocodazole treatment; after cytochalasin В treatment (10 (µg/ml, 1 h) then FCCP + cytochalasin В treatment.

lay perpendicular to the substrate, but in general the histogram of their distribution was similar to that of controls and not significantly different from random (P > 0.2; Fig. 2a).

After 10 min of FCCP treatment, the mean num-ber of satellites had increased significantly (Table I, P < 0.05), and sometimes active centrioles carrying as many as five satellites were found (Fig. За). The number of microtubules radiating from the cen-trosome obviously increased (Fig. За). The histo-gram of the distribution of active centrioles diifered from the control pattern. A total of 20 of the 30 cen-trioles studied were perpendicular to the substrate (Fig. 2b). The dramatic increase in the mean angle of inclination of active centrioles means that the ma-jority of the centrioles were inclined at an angle of more than 38° (1-3 classes on Fig. 2b), but some still remained parallel to the substrate (last class).

By 30 min after the introduction of FCCP, the mean number of satellites on active centrioles and the percentage of cells with primary cilia had in-creased (Table I). Striated rootlets were seen more frequently, being present on about 1/3 of active and 1/5 nonactive centrioles. The histogram of the distribution of active centrioles was also different from the control (Figs. 2a and 2b; P < 0.01). Half the active centrioles were oriented perpendicularly to the substrate. At this time no centrioles were in-clined at an angle of <21°, i.e., active centrioles which had been parallel to the substrate also turned. A similar distribution of active centrioles persisted 2 hr after the FCCP treatment (Fig. 2b).

Centrosomes after nocodazole and FCCP treat-ment. At 30 min after the introduction of FCCP to cells preincubated for 60 min with nocodazole, the position of centrioles in relation to the nucleus had not changed. The distance between the two centri-oles in each cell was <1 µm. Microtubules in the cell center were mostly absent (Fig. 3b), and occasion-ally one, rarely two, long microtubules were at-tached to the lateral surface of an active centriole or to the head of satellite. One or two satellites of char-acteristic shape were present on the active centriole (Fig. 3b) in most cells, or none at all. One-third of the cells studied possessed primary cilia (Table I). Striated rootlets were observed near 7 active and 6 nonactive centrioles (of the 33 cells studied). Both centrioles had striated rootlets in 1 cell.

The histogram of the distribution of projections of active centrioles after this treatment was strictly different from control (Fig. 2c; P < 0.01). Active cen-trioles were predominantly perpendicular to the substrate—21 of 31 observed were inclined at an an-gle >74°. No centrioles were found inclined to the substrate at angles of <38° (Fig. 2c).

Centrosomes after cytochalasin В and FCCP treat-ment. In the FCCP-treated cells after preincuba-tion with cytochalasin В (38 cells studied) the posi-

CENTROSOMES, MICROTUBULES, AND A MITOCHONDRIAL UNCOUPLER 221

FIG. 3. (a) Ultrastructure of the centrosome in PK cell after 10 min FCCP treatment. S, pericentriolar satellite. Bar, 0.2 µm. (b)

Ultrastructure of the centrosome in PK cells after nocodazole treatment (10 µg/ml, 1 hr) then FCCP + nocodazole treatment, 30 min. S, pericentriolar satellite. Bar, 0.2 µm. (c) Ultrastructure of the centrosome in PK cells after cytochalasin В treatment (10 µg/ml, 1 hr) then FCCP + cytochalasin В treatment, 30 min. S, pericentriolar satellite. Bar, 0.2 µm. (d) Pericentriolar satellites after 30 min FCCP treatment. S, pericentriolar satellite. Bar, 0.25 µm. (e) Pericentriolar satellites after nocodazole treatment (10 µg/ml, 1 hr) then FCCP + nocodazole treatment, 30 min. S, pericentriolar satellite. Bar, 0.25 µm. (f) Pericentriolar satellites after cytochalasin В treatment (10 µg/ml, 1 hr) then FCCP + cytochalasin В treatment, 30 min. S, pericentriolar satellite. Bar, 0.25 µm.

222 ALIEVA AND VOROBJEV

tion of the centrosome was different from that in the controls. In 16 cells the distance between maternal and daughter centrioles was >0.5 µm (Fig. 3c). The average distance was 1-2 µm, and sometimes reached 10 µm. One of the centrioles in each cell (usually the nonactive centriole) was located far away from the nucleus, near the periphery of the cell. In some cells both centrioles were distant from the nucleus. One or two satellites were present on the active centriole, but sometimes none were found. In 3 cells, the active centriole possessed three satellites. In the majority of cells pericentriolar satellites differed from those in normal or FCCP-treated cells. The filamentous basements of satellites were thin, with striations hardly visible (Figs. 2d-2f). More than half of the cells studied possessed primary cilia (Table I). In 11 cells, active centrioles had striated rootlets, and in 5 of these cases, cilia were present. Striated rootlets were found in only 4 cells associated with nonactive centrioles.

The histogram of the distribution of active centri-oles after these treatments changed relative to con-trols. Of the 38 active centrioles studied 12 were ori-ented perpendicularly to the substrate; additionally, the second and the last classes of the histogram were less well represented than those of the control cells (Fig. 2c). According to the χ2 test, the orientation of the active centrioles was nonrandom, i.e., perpendicular centrioles dominated (P < 0.05). It should be mentioned that after cytochalasin В pretreatment, the percentage of active centrioles perpendicular to the substrate was less than after nocodazole pre-treatment (Fig. 2c) or treatment with FCCP alone (Fig. 2b), but for the number of centrioles studied the difference in both cases was insignificant (P > 0.2).

DISCUSSION

Our data on the action of FCCP confirm the previous observations (Alieva et al., 1992; Alieva and Vorobjev, 1995) on the "activation" of the centrosome as a microtubule-organizing center in response to treatment with uncouplers of oxidative phosphorylation. Besides increase in quantity of microtubules radiating from the centrosome, the number of pericentriolar satellites also increased significantly 30 min after introduction of FCCP. It is interesting to note that under the action of nocodazole or cytochalasin В together with FCCP, the number of pericentriolar satellites was slightly less than in control cells and significantly less than under action of FCCP alone.

The phenomenon of perpendicular orientation of centrioles to the substrate in different types of cultured cells has been described elsewhere (Albrecht-Buehler and Bushnell, 1979; Vorobjev and Chentsov, 1982; Gudima et al., 1983a,b). Reevaluation of these studies using the χ2 criterion confirmed

that nonrandom orientation of active centrioles in PK cells and mouse fibroblasts (Vorobjev and Chentsov, 1982; Gudima et al., 1983a,b) was statistically significant (P < 0.05; see Vasiliev et al., 1988). In PK cells, a predominantly perpendicular orientation of active centrioles to the substrate was described for the early G1 period when cells were spreading after mitosis (Vorobjev and Chentsov, 1982). In a previous study, we have shown that the perpendicular orientation of active centrioles to the substrate surface could be induced independently of the cell cycle by depolarization of membranes, particularly of the plasma membrane (Alieva and Vorobjev, 1989). This occurred after different drug treatments, including dinitrophenol, dinitrophenol together with deoxyglucose, the calcium ionophore A23187, and ouabain.

Proceeding from the assumption that the perpendicular orientation of active centrioles to the substrate is a result of, or related to, plasma membrane depolarization (Alieva and Vorobjev, 1989, 1995), we have now studied the dynamics of centriole movement with the help of the most effective uncou-pler, FCCP. The introduction of this drug led to a loss of rhodamine 123 accumulated in mitochondria in 2-3 min. In 5 min no rhodamine was retained within treated cells. Active centrioles were found to turn perpendicularly to the plane of the substrate 10 min after FCCP introduction, i.e., after mitochon-dria and plasma membranes had lost their mem-brane potential. These data suggest that the rotational movement of centrioles is somehow linked with membrane depolarization. This is also supported by the observation that the nonrandom orientation of active centrioles persists for up to 2 hr of FCCP treatment.

In all cases described above, the nonrandom orientation of centrioles was associated with an increased number of microtubules radiating from the centrosome. To determine whether microtubules are directly involved in centriole orientation, as proposed by Albrecht-Buehler and Bushnell (1979), we used nocodazole treatment prior to the introduction of FCCP. In cells pretreated with drugs selectively destroying cytoskeleton elements, FCCP treatment unexpectedly showed that the orientation of active centrioles could not be related to microtubules. Indeed, after depolymerization of microtubules with nocodazole the percentage of active centrioles perpendicular to the substrate actually increased, although the difference was not statistically significant. Thus, our data argue against the supposition of Albrecht-Buehler and Bushnell (1979) that long microtubules attached to the centriole serve as anchors and assist in the orientation of centrioles perpendicular to the substrate.

Cytochalasin В induced numerous changes in the centrosome of PK cells. Centrioles often move apart

CENTROSOMES, MICROTUBULES, AND A MITOCHONDRIAL UNCOUPLER 223

from each other and the satellites on the active cen-triole were altered. Certain centrosome microfila-ment relationships were observed in the studies of the mechanism of centriole splitting (Sherline and Mascardo, 1982; Euteneuer and Schliwa, 1985). In both these works, cytochalasin В was described to prevent centriole splitting induced by epidermal growth factor and tumor promoting agent (TPA). However, in our study centriole separation seemed to be stimulated by this drug. This discrepancy might be due to the different origin of the cells used. Furthermore, it is possible that the centriole separation by a distance of about 2 µm seen in our study is insignificant and was not taken into account in light microscopic studies.

It should be stressed that the nonrandom orientation of centrioles under the action of FCCP was still established in the presence of cytochalasin B. In the absence of a microfilament network, active centrioles seemed to be partially disoriented. However, the number of perpendicular centrioles was significantly higher than that expected from a random distribution. Taking into account the dramatic changes in cell morphology observed (data not shown), it is unlikely that centriolar turning in the presence of cytochalasin В can be due to the remnants of the microfilament system. We suggest that the mechanisms responsible for centriole orientation in the cytoplasm, if microfilaments are not involved, is due to some as yet undefined cytoskeletal system whose nature remains to be determined.

Summing up the data obtained in the present work and other studies (Alieva and Vorobjev, 1989, 1995), we suppose that centriole orientation is induced by the depolarization of plasma membrane, though the nonspecific effect of all the drugs used can't be excluded. Centriole orientation occurs independently of the action of the anti-microtubule and relatively independently of the anti-microfilament drugs. This means that centrioles must be associated with some cytoskeletal structures besides mi-crotubules and microfilaments, whose molecular basis needs to be elucidated. One of the possible components of such system associated with centrioles had been identified recently as tektins (Steffen and Linck, 1992). The relationship between membrane potential and centriole behavior partially supports the idea that centrioles may be involved in the process of the intracellular signaling pathway (Born-ens, 1979) or might operate as a cell sensor (Al-brecht-Buehler, 1992), but, unlike these authors, we conclude that at least some of the centrosome-membrane relations do not include microtubules.

We thank Dr. Denys Wheatley for his encouragement, helpful discussion, and reading of the manuscript. The work was also supported by Grant 93-04-6523 to I.A.V. from the Russian Basic Science Foundation.

REFERENCES

Albrecht-Buehler, G. (1990) The "iris diaphragm"-principle of centriole and basal body formation. Cell Motil. Cytoskeleton 17, 197-213.

Albrecht-Buehler, G. (1992) Function and formation of centrioles and basal bodies, in Kalnins, V. C. (Ed.), The Centrosome, pp. 69-102, Academic Press, New York/San Diego.

Albrecht-Buehler, G., and Bushnell, A. (1979) The orientation of centrioles in migrating 3T3 cells. Exp. Cell Res. 120, 111-118.

Alieva, I. B., and Vorobjev, I. A. (1989) Centrosome behavior under the action of metabolic inhibitors on tissue culture cells. Biologicheskie Nauki No. 6, 45-49.

Alieva, I. B., and Vorobjev, I. A. (1990) Centrosome reaction on depolarization of the cell membranes in tissue culture cells. Biol. Nauki No. 4, 45-49.

Alieva, I. В., Vaisberg, E. A., Nadezhdina, E. S., and Vorobjev, I. A. (1992) Microtubule and intermediate filament patterns around the centrosome in interphase cells, in Kalnins, V. C. (Ed.), The Centrosome, pp. 103-129, Academic Press, New York/San Diego.

Alieva, I. B., and Vorobjev, I. A. (1995) Centrosome behaviour and orientation of centrioles under the action of energy transfer inhibitors. Cell Biol. Int., in press.

Bershadsky, A. D., and Gelfand, V. I. (1981) ATP-dependent regulation of cytoplasmic microtubule assembly. Proc. Natl. Acad. Sci. USA 78, 3610-3613.

Bornens, M. (1979) Centriole as a gyroscopic oscillator. Biol. Cell. 35, 115-132.

Boveri, T. (1887) Uber die Befruchtung der Eier von Ascaris meg-alocephala. Site. Berlin Ges. Morph. Phys. Munchen. Ill, p. 153.

Clark, S. W., and Meyer, D. I. (1992) Centractin is an actin ho-mologue associated with the centrosome. Nature 359, 247-250.

Euteneuer, U., and Schliwa, M. (1985) Evidence for an involvement of actin in the positioning and motility of centrosomes. J. Cell Biol. 101, 96-103.

Gudima, G. O., Vorobjev, I. A., and Chentsov, Yu. S. (1983a) Cell center behavior during flbroblast spreading. Biol. Nauki No. 7, 45-50.

Gudima, G. O., Vorobjev, I. A., and Chentsov, Yu. S. (1983b) Changes in the structure of the cell center during polarization and spreading of fibroblasts. Tsitologiya 25, 883-888.

Gudima, G. O., Vorobjev, I. A., and Chentsov, Yu. S. (1986) Cell Center Behavior during Fibroblast Spreading. 1. Normal Fibroblasts. Proc. Xlth Int. Cong. Electron Microscopy, Kyoto, pp. 2523-2524.

Hulser, D. F., Ristow, H.-J., Webb, D. J., Pachowsky, H., and Frank, W. (1974) Fibroblastoid and epithelioid cells in tissue culture: Differences in sensitivity to oaubain and in phospho-lipid composition. Biophys. Biochem. Acta 372, 84-99.

Kuriyama, R., and Borisy, G. G. (1981) Centriole cycle in Chinese hamster ovary cells as determined by whole-mount electron microscopy. J. Cell Biol. 91, 814-821.

Nicholls, D. G. (1982) Bioenergetics: An Introduction to the Che-miosmotic Theory. Academic Press, London/New York.

Pickett-Heaps, J. D. (1969) The evolution of the mitotic apparatus: An attempt at comparative ultrastructural cytology in dividing plant cells. Cytobios 1, 257-280.

Rieder, C. L., and Borisy, G. G. (1982) The centrosome cycle in PtKj cells: Asymmetric distribution and structural changes in pericentriolar material. Biol. Cell. 44, 117-132.

Schliwa, M., Pryzwanski, К. В., and Euteneuer, U. (1982) Centrosome splitting in neutrophils: An unusual phenomenon related to cell activation and motility. Cell 31, 705-717.