Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://cellmotility.genebee.msu.ru/html/articles/alieva95.pdf Äàòà èçìåíåíèÿ: Wed May 15 17:11:56 2002 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 19:59:36 2012 Êîäèðîâêà:

Cell Biology International. Vol. 79, No. 2, 1995

103

CENTROSOME BEHAVIOUR AND ORIENTATION OF CENTRIOLES UNDER THE ACTION OF ENERGY TRANSFER INHIBITORS.
I.B. Alieva and I. A. Vorobjev.

A N Belozersky Institute of Physico-Chemicol Biology, Moscow State University, Moscow
119899, Russia.
ABSTRACT
2,4-dinitrophenol, dinitrophenol together with deoxyglucose, sodium azide and ouabain didn't alter cytoplasmic microtubule (MT) network of cultured PK (pig kidney embryo) cells, meanwhile they induced an increase in the average number of pericentriolar satellites and percentage of centrioles with the primary in these cells. Also all drugs studied increase number of MTs attached to and oriented towards the centrosome. Under the action of ouabain the total number of MTs around the centrosome doubled, meanwhile the number of long MTs emanating from the centrosome increased more than 15 times. Under the action of all drugs studied, except sodium azide, the number of maternal centrioles oriented perpendicularly to the substrate surface increased significantly from that in control cells. Keywords: centriole; centrosome; energy transfer inhibitors; microtubules
INTRODUCTION

In tissue culture cells the centrosome plays a major role in maintaining the microtubule (MT) network. MTs emanate from the centrosome and terminate on or close to the plasma membrane (reviewed in Brinkley et al., 1980). The MT system, shown to be highly dynamic (reviewed in Gelfand and Bershadsky, 1991), might be expected to react to treatments directed against different cell organelles. The centrosome being involved in organization of MTs might also be altered by these treatments. Energy transfer inhibitors were shown to affect the MT system in cultured cells (Maro and Bornens, 1982; Maro et al., 1982). Following treatment by metabolic uncouplers such as carbonylcyanide-ptrifluoromethoxyphenylhydrazone; and 2,4dinitrophenol, most of the MTs disassembled and only those, radiating from the centrosome, remained (Maro and Bornens, 1982). This effect was reported to be specific since it had been induced only by uncouplers, but not by other energy transfer inhibitors (Maro and Bornens, 1982). However, the relationship between MT network and fine structure
1065-6995/95/020103+10/$08.00/0

of the centrosome under these treatments are not clear since no electron microscopic data on the centrosome structure has yet been reported. Detailed electron microscopic studies of the centrosome showed that rehabilitation of the MT network during cell spreading resulted in centrosome activation and in some cells also induced non random (preferentially perpendicular) orientation of maternal (active) centrioles (Gudima et al., 1983; 1986a; 1986b). The perpendicular orientation of active centrioles was observed during spreading of normal and slightly transformed fibroblasts (Gudima et al., 1986a; 1986b; Vorobjev and Nadezhdina, 1987). Recently, we developed a method allowing the location and orientation of MTs around the centrosome to be determined(Alieva et al., 1989; 1992). We presume that energy transfer inhibitors affect the centrosome structure as well as the centrosome-associated MTs. The aim of the present study was to test this proposition.
MATERIALS AND METHODS

Tissue culture. PK (pig kidney embryo) cells were
© 1995 Academic Press Ltd


104

!

Cell Biology Internationa/, Vol. 19, No. 2, 1995

grown in 199 tissue culture medium supplemented with 10% calf serum and antibiotics (penicillin + streptomycin). Electron microscopy. For the electron microscopic study, cells were prepared as described previously (Vorobjev and Chentsov, 1982). Cells grown on cover slips were fixed with 2.5% glutaraldehyde, postfixed with osmium tetroxide, stained with uranium acetate, dehydrated and embedded in Epon 812 mixture. For a detailed study of MT location, cells were extracted for 15 min at 37° in a MTstabilizing solution consisting of 25 mM phosphate buffer (pH=6.8), 1 mM MgCl2, 1 mM EGTA, 1% (w/v) Triton-X-100 and 4 M glycerol (Gudima et al., 1988) prior to fixation. Fixation, dehydration and embedding were as done as described above. Serial ultrathin (70 nm) and semi-thin (0.25 µm) sections were obtained parallel to the substrate plane using an LKB-III ultramicrotome (LKB, Sweden) and then mounted on single slot grids. Ultrathin sections were examined and photographed in HU-11B or HU-12 electron microscopes (Hitachi, Japan) operating at 75 kV. Semi-thin sections were examined in an Hitachi H-700 electron microscope operating at 175 kV. For the reconstruction of Mts, series of semithin sections containing profiles of both centrioles, one section above the upper centriole, and one section below the lower centriole in each centrosome, were used. All sections were photographed at a standard magnification of 12,000 with the tilt angle at 10° in order to make stereopairs and then printed at a final magnification of 36,000. The space area analysed was 4.6 x 3.6 µ and 1.0-1.75 jam in depth, depending on the position of two centrioles. For the purpose of reconstruction, MT profiles were traced from the prints of subsequent sections on the transparencies. This procedure generates a two-dimentional projection of the MT complex around the centrosome. According to their length, position and orientation of their proximal ends, we divided MTs into five classes as described previously (Alieva et al., 1992). Briefly, Class 1, short associated MTs less than 2 mm in

length whose proximal ends are less than 0.1 µm from centriole surface or pericentriolar satellite; Class 2, long associated MTs, more than 2 µm in length similarly associated with centrioles/satellites; Class 3, short free MTs less than 2 /µm in length with both ends at a distance more than 0.1 µm from centrioles and satellites (the proximal ends of these MTs, however, are oriented toward one of these structures); Class 4, long free MTs, more than 2 /µm in length with both ends at a distance more than 0.1 µm from centrioles and satellites (the proximal ends of these MTs, however, are oriented toward one of these structures); Class 5, free MTs that pass through the space studied but are not oriented toward centrioles and satellites. On the reconstructions presented (Fig. 6) only MTs radiating from the centrosome (referred to as Classes 1-4) are shown, while MTs oriented in any other direction (i.e. passing by the centrosome) are omitted. Immunofluorescence. Immunofluorescent staining of MTs was performed according to the indirect method described elsewhere (Bershadsky and Gelfand, 1981). Briefly, cells grown on coverslips were lysed in MT stabilizing solution, then fixed with 2% glutaraldehyde (Merck) in phosphate buffered saline (PBS) processed with 2% borohydride on PBS and stained with primary anti-a-tubulin monoclonal antibodies (Sigma), then with anti-mouse FITC-conjugated IgG (Sigma). Drug treatment. 2,4-dinitrophenol (DNP) at a final concentration of 800 /µM was added to the culture medium. Cells were fixed 5 and 30 min after introduction of the drug. DNP (800 µM) together with deoxyglucose (DOG) (1 mM) were dissolved in Dulbecco's medium (Adams, 1980). To exhaust endogenous glucose, cells were preincubated in Dulbecco's medium for 2 h, then transferred to the same medium supplemented with DNP+DOG and fixed 10 or 30 min later. In the last experiment, cells incubated in Dulbecco's medium were taken as a control. Sodium azide (final concentration 20 mM) and ouabain (final concentration 1 mM) were introduced directly into the culture medium. Cells were then fixed 30 min after the introduction of these drugs. At the concentration used, DNP led to a loss of


Cell Biology International, Vol. 19, No. 2, 1995

105

rhodamine 123 staining of mitochondria in 2-3 min. The concentration of DOG was chosen to decrease ATP levels approximately 10-fold (cording to the data reported by Spurck et al., 1986). Sodium azide concentration was chosen in the same way (Bershadsky and Gelfand, 1981). Ouabain, at the concentration used, completely inhibited Na,KATPase (Hulser et al., 1974), but was without effect on the rhodamine 123 staining of mitochondria 2 h after being introduced into the culture medium. Orientation of centrioles to the substrate plane was determined according to the method of AlbrechtBuehler and Bushnell (1979). This method allows one to calculate the angle between the centriolar cylinder axis and the substrate plane by the length of projection of centriolar MTs onto the section plane. The histograms were made taking one class equal to 0.03 µm. Each centriole, depending on the length of its projection, was placed in one of the 12 classes of the histogram. Centrioles put into the first class were determined as lying perpendicular to the substrate plane (Vasiliev et al., 1989). Experimental histograms were compared with that for the randomly distributed centrioles (Fig. 3) by the x2 criterion as described previously (Vasiliev et al., 1989).
RESULTS

None of the drugs used caused any serious changes in the cytoplasmic MT pattern (Fig. 1). Numerous MTs run through the cytoplasm. The overall fine structure of the centrosome also remained similar to the controls. However, certain differences were revealed by statistical analysis of the centrioles orientation and the frequency of appearence of different centrosome components. 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 inclination of the centrioles to the plane of the substrate.

Figure 1. Immunofluorescent staining of MTs in interphase PK cells. A - control; - ouabain (1 mM, 30 min); - sodium azide (20 mM, 30 min). Bar 10 µm.


106

Cell Biology International, Vol. 19, No. 2, 1995

alone or DNP with DOG as well as of ouabain, the percentage of maternal centrioles oriented perpendicularly to the substrate plane increased.
Centrosome and microtubules in the control cells. In untreated PK cells, the centrosome was situated near the nucleus, the distance between active (maternal) and non-active (daughter) centrioles in 23 cells from the 25 studied being less than 1 µm. Microtubules in the centre of the cell were attached to the lateral sides of both centrioles and to the heads of pericentriolar satellites. There were usually 1-2 (sometimes none) satellites on each active centriole. In 1/3 of the cells, primary cilia were observed. Cilia had a short axoneme, which was completely immersed in the cytoplasm and did not emerge from the cell body (Fig. 2a). About 1/4 of the cells possessed striated rootlets, and in 3 cases they ran from both centrioles (Table 1). From active centrioles, striated rootlets ran twice as often than from the non-active ones. The presence of striated rootlets was not correlated with the presence or absence of the primary cilium. The histogram of the distribution of the inclination of active centrioles to the substrate plane was similar to the random distribution histogram (Fig. 3, 4), their differences being statistically insignificant by the x2 test. Only 5 (2 active among them) of the 61 centrioles studied were perpendicular to the substrate plane. The histogram of the distribution of active centrioles was the same both in the control cells and in the cells incubated for 2 h in serum-free Dulbecco's medium (data not shown).

Figure 2. Fine structure of the centrosome in PK cells, a - control cell; b - after DNP treatment (800 µ, 30 min.). A - active; N - nonactive centrioles; - primary cilium; S - pericentriolar satellites. Bar 0.5 µm. 30 min after introduction of the drugs, the mean number of pericentriolar satellites and the frequency of the primary cilium increased compared with control cells (Table 1). Under the effect of DNP

On average, 31.0 + 3.6 MTs radiate from the centrosome in PK cells. The main part of the MTs radiate from the active centriole, only few of them being attached to or oriented towards the non-active centriole (Table 2). Nearly a half of the MTs radiating from the centrosome had their proximal ends free (Table 2). Besides the MTs oriented towards the centrosome, 6.8 + 0.9 MTs passed by the centrosome. 3.0 + 0.6 of these MTs were short, the others being long. Only individual MTs in some cells ran outside the space studied (Table 3).


Cell Biology International, Vol. 19, No. 2, 1995

107

Table 1. Centrosome changes under the action of energy transfer inhibitors. Drugs, time of the treatment (min) Control DNP, 5 DNP, 30 DNP+, DOG , 10 DNP+ DOG, 30 Sodium azide, 30 Ouabain, 30 Mean number of satellites ( + SD) 1.57 + 0.15 1.88 + 0.16 1.75 + 0.16 2.36 + 0.14* 2.42 + 0.09* 2.26 + 0.10* 2.43 + 0.15* Percentage of the cells with primary cilium 29 37 52 37 40 57 48 Percentage of the cells with striated rootlets 25 18 35 24 21 34 26

* significantly different from control cells (p < 0.01).

Table 2. Distribution of microtubules radiating from the centrioles after treatment with different drugs (Mean ± SD). Drugs, time of treatment Control active centriole inactive centriole DNP, 30min active centriole inactive centriole active centriole inactive centriole active centriole inactive centriole active centriole inactive centriole Short MTs with Long MTs with Short MTs with Long MTs with Total number free ends attached end attached end free ends of MTs 11.6 ± 2.5 3.0 ± 0.5 18.1 ± 1.6 3.6 ± 0.5 13.5 ± 2.0 3.8 ± 0.7 19.6 ± 2.4 2.3 ± 0.4 21.7 ± 2.0 3.3 ± 0.6 2.6 ± 0.9 0.0 ± 0.0 3.9 ± 0.4 0.6 ± 0.2 2.7 ± 0.6 0.2 ± 0.1 6.1 ± 0.8 0.6 ± 0.2 8.6 ± 1.1 0.8 ± 0.3 5.4 ± 0.3 0.8 ± 0.4 8.3 ± 1.6 0.8 ± 0.4 7.4 ± 0.7 1.5 ± 0.4 8.9 ± 1.0 1.9 ± 0.4 7.4 ± 0.9 3.8 ± 2.9 7.2 ± 1.6 0.6 ± 0.2 14.7 ± 1.5 0.6 ± 0.2 14.1 ± 0.9 1.2 ± 0.4 20.9 ± 3.0 0.8 ± 0.6 19.7 ± 1.7 1.1 ± 0.4 26.8 ± 5.4 4.4 ± 0.7 45.0 ± 2.8 5.6 ± 0.9 37.7 ± 3.1 6.7 ± 1.2 55.4 ± 5.9 5.6 ± 0.7 57.4 ± 3.5 6.3 ± 0.9

DNP+DO G, 30 min

Sodium azide 3O min

Ouabain, 30 min


108

Cell Biology International, Vol. 19, No. 2, 1995

Table 3. Number of MTs running outside the radius 1.5 /on from the centrosome after treatment with different drugs (Mean + SD). Drugs, time of treatment Control DNP, 30 min. DNP+ DOG, 30 min. Sodium azide, 30 min. Ouabain, 30 min. MTs with attached end 0.4 ± 0.2 0.6 + 0.3 0.7 + 0.3 1.0 + 0.3 1.9 ± 0.6 MTs with free ends 0.0 + 0.0 2.2 + 0.5 3.4 + 0.6 4.3 + 0.7 5.1 ± 0.6 Total number of MTs 0.4 + 0.2 2.7 + 0.5 4.1 + 0.8 5.3 ± 1.0 7.0 + 1.2

Centrosome and microtubules after DNP treatment. 5 min after the introduction of DNP into the culture medium the position of the centrosome towards the nucleus and the mutual position of two centrioles was the same as in the control cells. The mean number of satellites on the active centriole increased and centrioles with 4 and 5 satellites appeared. Primary cilia occurred more often while striated rootlets were rare (Table 1).

cells the distance between two centrioles was more than 1 µm, and in some cases nearly 3 µrn. There were up to 6 satellites on active centrioles (Fig. 2b). In the centrosome of treated cells, striated rootlets and primary cilia were observed more often than in the control cells (Table 1). 30 min after the introduction of DNP the histogram of the distribution of active centrioles was different from that obtained for the short DNP treatment (Fig. 3). The volume of Class 1 increased 1.5 times. The mean angle of inclination of active centrioles towards the substrate plane increased as compared with the 5 min DNP treatment. The histogram of the distribution of nonactive centrioles also differed significantly from that for the control cells (p<0.01), due to a large number of centrioles inclined to the substrate plane at an angle from 54° to 74°. 10 min after the introduction of DNP together with DOG, active centrioles were arranged mainly perpendicularly to the substrate plane (Fig. 4). The histogram of the distribution of non-active centrioles was similar to that for the random distribution. 30 mL_ after the introduction of DNP+DOG, both histograms remained the same (Fig. 4). 50.4 ± 3.1 MTs radiated from the centrosome in cells treated with DNP for 30 min. The total number of MTs increased more than 1.5 times as compared with the normal one. This was mainly due to MTs radiating from the active centriole. Consequently the number of pericentriolar satellites connected MTs increased twice. The number of MTs connected with the centrosome

Figure 3. Histograms of the distribution of length of maternal centrioles projections on the section plane. Control (front row) and DNP-treated cells. The histogram of the distribution of active centrioles differed from that for the control cells (Fig. 3), the difference being statistically insignificant (p>0.5). Also the mean angle between the centriole axis and the substrate plane increased. 30 min after the introduction of DNP in 37 % of the


Cell Biology International, Vol. 19, No. 2, 1995

109

and with both ends free increase proportionally, but some differences took place between MTs from different classes (Table 4), e.g. the 2nd and 3rd classes remained near the control level, while the 4th class increased twice.

all four classes, but it was unequal, the biggest one being in the 4th class (Table 2). The number of MTs radiating from centriolar cylinder of both active and non-active centrioles did not change significantly. Centrosome and microtubules after ouabain treatment. 30 min after the introduction of ouabain, the position of the active and non-active centrioles changed. The distance between them in 15 of the 28

Figure 4. Histograms of the distribution of length of maternal centrioles projections on the section plane. Control (front row) and DNP+DOG-treated cells. 44.4 + 3.1 MTs radiate from the centrosome in DNP/DOG-treated cells. The quantity and distribution of MTs connected with the centrosome did not change significantly from the controls (Table 2). The increase in the total number is only due to the increase in the number of MTs with both ends free, mainly the long ones. Centrosome and microtubules after sodium azide treatment. 30 min after the introduction of sodium azide the position of active and non-active centrioles was not changed as compared with the control cells. The mean number of pericentriolar satellites increased as well as the percentage of cells having primary cilia and striated rootlets (Table 1). Histograms of the distribution of centrioles to the substrate plane (both active and nonactive) were similar to the random one, though the mean angle of inclination of centrioles slightly increased (Fig. 5). 61.0 + 6.1 MTs radiated from the centrosome after 30 min sodium azide treatment. Acceleration occurrred in

Figure 5. Histograms of the distribution of length of maternal centrioles projections on the section plane. Control (front row), sodium azide-treated and ouabaintreated cells. cells studied was more than 1 µm and could reach 4 jam. 2 - 3 satellites were observed on each active centriole. The percentage of cells with primary cilia increased twice while the percentage of cells with striated rootlets remained the same (Table 1). The histogram of the distribution of active centrioles (Fig. 5) was different from that for the control cells the mean angle of inclination of centrioles increased and 14 of the 33 centrioles studied were oriented perpendicularly to the substrate plane (p<0.01). The number of non-active centrioles in the first class of the histogram was unchanged (data not shown). 63.7 + 3.8 MTs radiated from the centrosome after 30 min ouabain treatment (Table 2; Fig. 6). Acceleration occurred in all four classes, but it was unequal being less in the 3rd class. The number of MTs running out of the space studied increased more than 10-fold as compared with normal cells (Table 3).


110

Cell Biology International, Vol. 19, No. 2, 1995

Fig. 6. Reconstruction of profiles of the MTs radiating from the centrosome in control (a) and ouabain-treated (b) cells. DISCUSSION In this paper we describe in detail the fine structure of the centrosome, its relationships with MTs and its reactions to different drugs, according to their mechanisms of action on the living cell. In contrast to the data on uncouplers of oxidative phosphorylation reported previously (Maro and Bornens, 1982; Maro et al., 1982), in our experiments the cytoplasmic MT complex was found not to change significantly. Nevertheless, all the inhibitors studied induced certain changes in the centrosome. These were: increase in the average quantity of pericentriolar satellites and in the frequency of appearance of primary cilia; the nonrandom orientation of centrioles to the substrate plane; and centriole splitting. The perpendicular orientation of active centrioles to the substrate surface has been described previously for PK cells spreading after mitosis (Vorobjev and Chentsov, 1982) and for mouse embryo fibroblasts when seeded on to a coverslip surface (Gudima et al. 1983). In non-synchronized culture of PK cells-orientation of active centrioles to the substrate surface is close to the random one (Vorobjev and Chentsov, 1982). In our experiments the non-random (preferentially perpendicular) orientation of active centrioles appeared under DNP and ouabain treatment. At the same time, the orientation remained close to the random one under sodium azide treatment. Our experiments have shown that depletion of ATP production and alteration of energy transfer in the cells may not result in the perpendicular orientation of active centrioles. On the other hand, such an orientation appears to be due to the specific action of ouabain which depolarizes only the plasma membrane (Kernan, 1970) and does not inhibit mitochondrial membrane potential generation. The latter was shown


Cell Biology International. Vol. 19, No. 2, 1995

111

experimentally; by 2 h after the introduction of ouabain to the culture medium, cell mitochondria were capable of accumulating rhodamine 123. The data obtained enable one to conclude that depolarization of the plasma membrane is enough to induce the perpendicular orientation of active centrioles. The total number of MTs radiating from the centrosome increased under the action of all the drags studied and the maximal increase was observed after 30 min treatment with ouabain. The predominant role in this acceleration is played by pericentriolar satellites since the number of MTs radiating from them increased more than the total one. Among different classes of MT a preferential increase of the 4th class (long MTs with both ends free) was observed. The number of MTs emanating from the centrosome and running out of the analysed pericentrosomal region increased many times after all drug treatments (Table 3). In normal cells all of these long MTs were attached to the electron dense structures in the centrosome. After drug treatments many of these MTs had their proximal end free. Their quantity after ouabain treatment is similar to that found after taxol treatment (Alieva et al., 1992). A quantity of long MTs attached to the centrosome after treatment with different drugs, increased proportionally to the total quantity of MTs. In all cases it remained comparatively small. Under the action of all the drugs used, free MTs increased in number preferentially. Especially high was an increase of long MTs having free proximal ends. It may be concluded that ouabain, uncouplers of oxidative phosphorylation and sodium azide stabilize cytoplasmic MTs rather then promote MT polymerization. According to the data obtained we suggest that during rebuilding of the centrosome under the action of various drugs, at least two different reactions occur, namely an increase in the number of pericentriolar satellites supplemented with an increase in the quantity of MTs radiating from the centrosome (centrosome activation) and the appearence of the non-random orientation of active centrioles towards the substrate surface (centriole turn). The first reaction is less specific as it is observed after different treatments from general inhibition of ATP production (DNP+DOG) to inhibition of one enzyme only (ouabain).

Centrioles are supposed to be the organelles responsible for detecting and locating of ambient signals coming to the cell, particularly to the plasma membrane (Bornens, 1979; Albrecht-Buehler, 1990; 1992). To evaluate the last supposition, we used different energy metabolism inhibitors, which affect the plasma membrane. Among a variety of drugs we chose those depleting membrane potential. Different modes of action of these drugs have only one common issue, which is to decrease the plasma membrane potential. The plasma membrane potential is mainly attributed to the gradients of [K+] and [Na+ ] across the membrane, and is maintained by Na+/K+ pump (Kernan, 1970). It might be lowered by an increase in membrane ion conductivity (Jungle, 1992), or by inhibiting K+/Na+ ATPase (Hulser et al., 1974). Thus uncouplers of oxidative phosphorylation acting as protonophores lower potential on all charged membranes because they increase proton conductivity, which is rather low under normal conditions (Jungle, 1992), while ouabain does the same on the plasma membrane only by inhibiting the K+/Na+ ATPase (Hulser et al., 1974). We suggest that in response to the inhibition of energy metabolism, a cell increases its radial subsystem of MTs emanating from the centrosome. In certain cases it might be accompanied with non-random orientation of the active (maternal) centrioles. The results of our study of the possible role of MTs in the latter event are described in another communication (Alieva and Vorobjev, submitted for publication). In summary, we suggest that from the data obtained that: i) the increase in the quantity of MTs radiating from the centrosome (centrosome activation) is a reaction to the alteration of cell energy metabolism, ii) The non-random (preferentially perpendicular) orientation of centrioles to the substrate surface appears as a result of depolarization of the plasma membrane.

REFERENCES
Adams, R.L.P. (1980). Cell culture for biochemists. Elsevier/North Holland Biomedical Press, Amsterdam. Albrecht-Buehler, G. and Bushnell, A. (1979). The orientation of centrioles in migrating 3T3


112

Cell Biology International, Vol. 19, No. 2, 1995

cells. Exp. Cell Res. 120: 111-118. Alieva, I.B., Vorobjev, LA. and Chentsov, Yu.S. (1989). Stereoscopic analysis of the microtubule pattern around the centrosome in tissue culture cell. Doklady Academii Nauk SSSR (in Russian) 305: 1232-1234. Alieva, I.B., Nadezhdina, E.S., Vaisberg, E.S. and Vorobjev, I.A. (1992). Microtubule and intermediate filament pattern around the centrosome in interphase cells. In: Kamins.V.I. (ed.) The centrosome. Academic Press. New York, San Diego. Bershadsky, A.D. and Gelfand, V.I. (1981). ATPdependent regulation of cytoplasmic microtubule disassembly. Proc. Nat. Acad. Sci. USA 78: 3610-3613. Bornens, M. (1979). The centriole as a giro-scopic oscillator. Implications for cell organ-ization and some other consequence. Biol. Cell. 35: 115-132. Brinkley, B.R., Fistel, S.H., Marcum, J.M. and Pardue, R.L. (1980). Microtubules in cultured cells: indirect immunofluorescent staining with tubulin antibody. Internal. Rev. Cytol. 63: 59-95. Gelfand, V.I. and Bershadsky, A.V. (1991). Microtubule dynamics: mechanism, regulation, and function. Ann. Rev. Cell Biol. 7: 93-116. Guclima, G.O., Vorobjev, I.A. and Chentsov, Yu.S. (1983). Alteration of the structure of the cell center during polarization and motion of fibroblasts. Tsytologia (in Russian) 25: 883888. Gudima, G.O., Vorobjev, LA. and Chentsov, Yu.S. (1986a). Cell center behavior during fibro-blast spreading. I. Normal fibroblasts. In: Imura, ., Maruse, S. and Suzuki, T. (eds.). Proceeding Xlth International Congress Electron Mic-roscopy. Kyoto. Ill (Biology): 2523-2524. Japanese Society of Electron Microscopy. Tokyo. Gudima, G.O., Vorobjev, LA. and Chentsov, Yu.S. (1986b). Cell center behavior during fibroblast spreading. II. Transformed fibroblasts. In: Imura, ., Maruse, S. Suzuki,

1. (eds.). Proceedings Aim International Congress Electron Microscopy. Kyoto. Ill (Biology): 2525-2526. Japanese Society of Electron Microscopy. Tokyo. Gudima, G.O., Vorobjev, LA. and Chentsov, Yu.S. (1988). Centriolar location during blood cell spreading and motion in vitro: an ultra-structural analysis. J. Cell Sci. 89: 225-241. Hulser, D.F., Ristow, H.-J., Webb, D.J., Pachowsky, H. and Frank, W. (1974). Fibro-blastoid and epithelioid cells in tissue culture: differences in sensititvity to ouabain and in phospholipid composition. Biochim. Biophys. Acta 372: 85-99. Jungle, D. (1992). Nerve and muscle excitation. 3rd ed. Sinauer Associates, Inc. Sunderland MA, USA. Maro, B. and Bornens, M. (1982). Re-organizatic of HeLa cell cytoskeleton induced by an"' uncoupler of oxidative phosphorylation. Nature 295: 334-336. Maro, ., Marty, M.-C. and Bornens, M. (1982). In vivo and in vitro effects of the mito-chondrial uncoupler FCCP on microtubules. The EMBO J . I : 1347-1352. Spurck, T.P., Pickett-Heaps, J.D. and Klymkowsky, M.W. (1986). Metabolic inhibitors and mitosis: II. Effects of dinitro-phenol/deoxyglucose and nocodazole on the microtubule cytoskeleton. Protoplasma 131: 61-72. Vasiliev, N.B., Vorobjev, LA., Leontovich, A.M. and Petrovskaya, M.B. (1988). Analysis of the centriole orientation in tissue culture cells. Tsytologia (in Russian) 30: 1091-1100. Vorobjev, LA. and Chentsov, Yu.S. (1982;r Centrioles in the cell cycle. I. Epithelial cells. J. Cell Biol. 98: 938-949. Vorobjev, LA. and Nadezhdina, E.S. (1987). The centrosome and its role in the organization of microtubules. Internat. Rev. Cytol. 106: 227-293.
(Paper received 30.08.04. Paper accepted 02.12.94).