stress fiber reformation after atp depletion

Cell Motility and the Cytoskeleton 8:118-129 (1987)

Stress Fiber Reformation After ATP Depletion

Peter A. Glascott, Jr., Karen M. McSorley, Balraj Mittal, Jean M. Sanger, and Joseph W. Sanger

Laboratory for Cell Motility Studies, Department of Anatomy, University of Pennsylvania School of Medicine, Philadelphia

Fluorescently labeled heavy meromyosin, alpha-actinin, and vinculin were used to localize actin, alpha-actinin, and vinculin, respectively, in permeabilized and living cells during the process of stress fiber reassembly, which occurred when cells were removed from ATP-depleting medium (20 mM sodium azide and 10 mM 2-deoxyglucose). In 80% of the cells recovering from ATP depletion, small, scattered plaques containing actin, alpha-actinin, and vinculin were replaced by long, thin, periodic fibers within 5 minutes of removal of the inhibitors. These nascent stress fibers grew broader as recovery progressed, until they attained the thickness of stress fibers in control cells. In the other 20% of the cells, the scattered plaques aggregated within 5 minutes of reversal, and almost all the actin, alpha-actinin, and vinculin in the cells became localized in one perinuclear aggregate, with a diameter of approximately 15-25 pm. As recovery progressed, all aggregates resembled rings, with diameters that increased at about 0.5 pm/minute and grew to as large as 70 pm in some giant cells. As the size of the rings increased, fibers radiated outward from them and sometimes spanned the diameter of the rings. The shape of the cells did not change during this time. By 1 hour after reversal, the rings were no longer present and all cells had networks of stress fibers. Indirect immunofluorescence techniques used to localize tubulin and vimentin indicated that microtubules and intermediate filaments were not constitu- ents of the rings, and the rings were not closely apposed to the substrate, judging from reflection contrast optics. The rapid rearrangement of attachment plaques into a perinuclear aggregate that spreads radially in the cytoplasm occurs at the same speed as fibroblast and chromosomal movement, but is unlike other types of intracytoplasmic motility.

Key words: cytoskeleton, actin, alpha-actin, vinuclin, microtubules

INTRODUCTION

Organized arrays of actin filaments and associated proteins, called stress fibers, are present in many non- muscle cells grown in culture and in some cells in situ [Byers et al, 19841. These fibers are composed primarily of actin, myosin, alpha-actinin, and tropomyosin arranged in patterns that suggest a sarcomeric organization of the fibers [Gordon, 1978; Sanger and Sanger, 1980; Sanger et al, 1983a, 1985a, 1986c, 1987bl. Nevertheless, the role of the stress fiber in cells is not known [for review, see Byers et al, 19841. Harris et a1 [I9801 developed an assay that indicated that cells with stress fibers develop substantial tension when grown on a silicone substratum [Stopak and Harris, 19821. Cells that move

0 1987 Alan R. Liss, Inc.

very quickly, however, do not possess stress fibers, and, in fact, cells with numerous stress fibers are less motile than those with only a few stress fibers [Herman et al, 198 1 J . In permeabilized cells, stress fibers are capable of contractions [Isenberg et al, 1976; Kreis and Birchmeier, 19801, and the pronounced tail retraction in motile fibroblasts is thought to result from stress fiber contraction [Chen, 19811. Sanger et a1 [1986a] have recently dem- onstrated that the alpha-actinin spacings in stress fibers

Received December 4, 1986; accepted April 2 , 1987

Address reprint requests to Dr. J.W. Sanger, Department of Anatomy, University of Pennsylvania School of Medicine, Philadelphia, PA 19 104-605 8.

Stress Fiber Reformation 119

change with time in living cells, and they have suggested Proteins that this is due to a sliding of the interdigitating actin and myosin filaments in the sarcomeric units of the stress fibers. While not all cells have stress fibers, all dividing animal cells possess cleavage rings that are structurally similar to stress fibers (Sanger and Sanger, 19801. The assembly of the cleavage ring occurs when the cell’s stress fibers have been disassembled [Sanger, 1975a; Sanger et al, 1987a1, and, conversely, cleavage ring disassembly is accompanied by stress fiber assembly [San- ger, 1975b; Sanger et al, 1987al.

The experimental manipulation of cells to cause reversible breakdown of stress fibers provides a system in which to study how stress fibers are assembled and disassembled and thereby may provide insights into how stress fiber organization is controlled during the cell cycle. Stress fiber disassembly can be reversibly induced by reducing the level of ATP in cells with sodium azide and 2-deoxyglucose [Bershadsky et al, 1980; Bershadsky and Gelfand, 1983; Sanger et al, 1983bl. The observations in the present study show that in PtK, cells, stress

Heavy meromyosin (HMM) was prepared from rabbit muscle and labeled with either lissamine rhodamine sulfonyl chloride (LR) (Molecular Probes, Junction City, OR) or fluorescein isothiocyanate (FITC) (Sigma Chemical Co., St. Louis, MO) as previously described [Sanger, 1975~1. Alpha-actinin was purified from fresh or frozen chicken gizzards and labeled with LR as previously described [Sanger et al, 1984a,b]. Vinculin was purified from frozen chicken gizzards as described by Evans et a1 [ 19841 and labeled with LR. NBD-phallacidin was purchased from Molecular Probes (Junction City, OR). Polyclonal tubulin antibody was obtained from Polysciences, Inc. (Warrington, PA). Monoclonal anti- vimentin was obtained from Amersham Corporation (Ar- lington Heights, IL). Secondary antibodies, rhodamine- labeled goat antirabbit IgG, and Texas-Red-labeled rabbit antimouse IgG were obtained from Cooper Biomedi- cal, Inc. (Malvern, PA).

Staining and Microscopy-Actin and Alpha-Actinin

MATERIALS AND METHODS Cell Cultures

PtK2 cells, a rat kangaroo epithelial line (American Tissue Type Collection, Rockville, MD), were grown on coverslips in culture dishes as previously described [San- ger, 1975bl. The culture medium used was Eagle’s minimal medium supplemented with 10% bovine serum, 1.5% glutamine, 1 % antibiotic-antimycotic (5,000 units penicillin, 5,000 p g streptomycin) (all obtained from GIBCO, Grand Island, NY). Eagle’s minimal medium contains 1 mg/ml (5.6 mM) glucose.

Inhibitors To deplete the cells of ATP, 20 mM sodium azide

and 10 mM 2-deoxy-L-glucose were added to the culture medium [Bershadsky et al, 1980; Sanger et al, 1983bl. Cells were then incubated for various times (90, 135, 150, and 180 minutes and 25 hours) in this medium to assure total disassembly of all stress fibers. After the appropriate times, cells were washed several times with control culture medium and placed back in the incubator to permit stress fiber reassembly.

rinsed well with standard salt, treated with paraformaldehyde for 15 minutes, mounted in 25 % glycerol on glass slides, and sealed with clear nail polish (Sanger et al, 1983al. Cells were observed with x 100 or x 6 3 Zeiss planapochromat lenses on an Olympus Vanox microscope equipped for epifluorescence. The patterns of actin localization in permeabilized cells using either HMM-LR or NBD-phallacidin were identical. Reflection-contrast microscopy was achieved with a Zeiss X63 Antiflex objective and a Zeiss Photomicroscope 111. Photographs were taken with Tri-X Pan film (Kodak) and were developed in Acufine (Acufine, Inc., Chicago, IL) for a rating of 1,OOO ASA or with Plus-X Pan film (Kodak) and then developed in HC-110 (Kodak, Rochester, NY).

Tubulin and Vimentin

The relationship of tubulin and vimentin to the rings was determined by staining cells first with HMM as described above. After fixation in paraformaldehyde for 5 minutes the coverslips were rinsed with standard salt solution, washed in 50 mM NH4CI (in a standard salt buffer) for 5 minutes [Sanger et al, 1983a], and then washed with standard salt solution. Next, the treated cells

120 Glascott et al

were labeled with tubulin or vimentin antibody for 30 to 45 minutes at 37”C, washed several times with standard salt solution, and then exposed to a fluorescently labeled secondary antibody for 30 to 45 minutes at 37”C, washed with standard salt solution, and mounted on glass slides.

Microinjection of Alpha-Actinin and Vinculin Into Living Cells

Living PtK, cells, grown on glass coverslips and incubated in Eagle’s medium, were microinjected with alpha-actinin-LR or a mixture of dpha-actinin-FITC and vinculin-LR, as previously reported [Sanger et al, 1984b, 1987a1. These cells were placed in an incubator (37°C) overnight and treated the next morning with 20 mM sodium azide and 10 mM 2-deoxyglucose for 135 minutes. The coverslip of injected cells was mounted on a glass slide and examined with a Silicon Intensified Target (SIT) video camera (Dage MTI, Inc., Michigan City, IN) mounted on an Olympus Vanox microscope. The coverslip was removed, washed with normal medium, then immediately remounted for examination with the SIT camera. Images were recorded on a NEC video recorder and photographs taken from the Panasonic video monitor as previously described [Pochapin et al, 1983; Sanger et al, 1985bl.

Electron Microscopy

Control and ATP-inhibited PtK2 cells were fixed, embedded in an Epon mix, sectioned, and examined in a Phillips 201 electron microscope using methods previously described [Sanger and Sanger, 1980, 19851.

RESULTS

PtK2 cells grown to culture at low density formed islands or clusters of interconnected cells with well-defined stress fibers that usually ran parallel to one another along the long axis of the cell. In some cells groups of stress fibers ran in different directions to one another forming a “mosaic” pattern (Fig. 1). The ends of many stress fibers were often arranged in a radial pattern around the nucleus (Fig. 1). Treatment of the PtK2 cells with 20 mM sodium azide and 10 mM 2-deoxyglucose for 90 minutes induced disassembly of almost all the stress fibers (Figs. 2, 3), with total disassembly of all the fibers occurring after 135 minutes. Actin and alpha-actinin were present in small plaques distributed throughout the cytoplasm of these cells, which remained flat and interconnected and retained their shape despite their lack of stress fibers. The plaques were elongated and were arranged approximately parallel to one another reminiscent of the parallel distribution of the normal stress fibers (Fig. 3). Some of the plaques in inhibitor-treated cells were in a

radial arrangement around the nuclei, as in control cells (cf Figs. 1, 3).

Replacement of the azide-deoxyglucose medium with inhibitor-free medium resulted in the rapid reformation of stress fibers throughout the cytoplasm in approximately 80% of the cells. During this process of stress fiber reformation, no change in cell shape or size was detected. In the initial reversal (5-minute recovery), the reforming stress fibers were very thin but comparable in length to control stress fibers and ran parallel to the long axis of the cell (Fig. 4). With time the fibers became thicker (Fig. 5 ) , until by 1 hour they had attained the thickness of control stress fibers. By 2 hours, the stress fibers in some cells were thicker than control stress fibers, and all had a striated pattern of alpha-actinin localization. After 24 hours of recovery, the stress fibers looked like control stress fibers, with a well-defined periodic distribution of alpha-actinin (Fig. 6).

In up to 20% of treated cells, an unusual pattern of actin and alpha-actinin localization was evident in the initial stages of stress fiber recovery. The contractile proteins in these recovering cells formed a single aggregate per cell that was near the ventral surface of the cell and usually positioned adjacent to the nucleus (Figs. 7- 10). Although some actin and alpha-actinin could be detected at the perimeter of the cells, most of the HMM- LR, NBD-phallacidin, and alpha-actinin-LR staining was

-

Fig. 1. Control PtK, cell permeabilized and stained with HMM-LR to localize actin in the stress fibers. The position of the nucleus is indicated (Nu). X 1,450.

Fig. 2. Cell after 90 minutes in 20 mM sodium azide and 10 mM 2- deoxyglucose. Note the disruption of the stress fibers. Permeabilized cell stained with HMM-LR to localize actin. The position of the nucleus is indicated (Nu). X 1,450.

Fig. 3. Plaques of alpha-actinin are scattered throughout the cytoplasm and around the nucleus (Nu) of a cell that had been exposed to inhibitors for 3 hours. Note the complete disassembly of the stress fibers. Alpha-actinin-LR was used to localize the alpha-actinin in the cells. X 1,450.

Fig. 4. Cell that has been returned to control medium for 5 minutes after a 90-minute exposure to inhibitors has thin stress fibers present. HMM-LR was used to localize actin. x 1,450.

Fig. 5 . Cell that has recovered for 10 minutes after a 90-minute exposure to inhibitors has stress fibers that are thicker and more numerous than those in cells that had recovered for 5 minutes. HMM- LR was used to localize actin. X 1,450.

Fig. 6. Twenty-four hours after reversal from a 3-hour exposure to inhibitors stress fibers appear as they do in control cells. Alpha- actinin-LR reveals the periodic arrangement of alpha-actinin in the stress fibers as well as the brightly stained attachment plaques. X 1,450.

122 Glascott et a1

concentrated in these aggregates. Often these perinuclear aggregates exhibited a less intensely stained core (Figs. 7, 9, lo), giving an overall appearance of a ring. Shortly after reversal, the alpha-actinin localization within the aggregates was punctuate (Fig. 9) or fibrous (Fig. lo), whereas actin was found only in thin fibers (Fig. 7). The percentage of cells that developed these aggregates was increased by prolonged exposure of the cells to the inhibitors. In the first 5 minutes of recovery, 5% of cells treated with inhibitors for 90 minutes and 20% of cells treated for 135 minutes contained aggregates of contractile protein. Moreover, in the longer inhibitory periods, a group of neighboring cells often had one aggregate per cell. In one isolated cluster of 10 interconnected cells, each cell had a perinuclear aggregate. This contrasts with the shorter treatment period in which the majority of cells with aggregates were isolated and scattered throughout the culture dish. In an attempt to increase the percentage of cells in which aggregates formed, cells were exposed to ATP-depleting medium for up to 24 hours with no significant increase in the number of recovering cells with perinuclear aggregates of contractile proteins. There was never more than one aggregate per cell, regardless of the size of the cell or the number of nuclei per cell. The aggregates were present in most of the recovery periods up to 45 minutes and were only rarely seen after 1 hour of recovery. The initial aggregates ranged in diameter from 15 to 25 pm, while the aggregates present at later recovery times ranged from 15 to 70 pm and were ring shaped.

Fibers stained with HMM-LR and alpha-actinin- LRR radiated outward from these larger rings (Fig. 11) and sometimes traversed the rings. Occasionally, the ring was indented by the cell’s nucleus (Fig. 12). In a few exceptionally large cells, rings up to 70 pm in diameter were observed even at 1 hour into the recovery period, with faint fibers extending from one edge of the ring to the other edge and radiating outward from the periphery of the ring as well. Within 24 hours, stress fiber formation in these cells matched that of control cells. During the entire process of coalescence of a cell’s contractile proteins into a single perinuclear aggregate, the subsequent expansion of the aggregate into a ring-shaped structure, and the reformation of stress fibers, the cell remained flat and interconnected in sheets.

Living PtK2 cells microinjected with alpha-actinin- LR and observed the next day with a SIT video camera showed incorporation of alpha-actinin-LR into attachment plaques and in a striated pattern into stress fibers (Fig. 13). After exposure to medium containing sodium azide and 2-deoxyglucose for 135 minutes the cells contained no fluorescent stress fibers, but were filled with scattered plaques containing fluorescent alpha-actinin

control medium and immediately remounted for examination. Figure 15 shows a recovering cell 10 minutes after reversal with an alpha-actinin-LR-containing ring similar to those observed in stained cells (Fig. 9). After 1-5 hours of reversal, stress fibers with a periodic distribution of fluorescent alpha-actinin had reformed in the cells (Fig. 16). The inhibitor-treated cells were more susceptible to damage by the excitatory light used for fluorescence, and, therefore, it was very difficult to follow the process of stress fiber reformation in the same injected cell.

PtK2 cells were also microinjected with a mixture of alpha-actinin-FITC and vinculin-LR. In control cells, the vinculin-LR was localized at the ends of stress fiber (the attachment plaques), and the alpha-actinin-FITC was localized in both attachment plaques and stress fibers. Exposure of these cells to ATP-depleting medium for 135 minutes showed that the scattered plaques composed of alpha-actinin also contained vinculin-LR (Figs. 17, 18).

Living and stained cells were examined under different conditions using reflection contrast microscopy. In control cells, the attachment plaques were visible mainly behind the ruffling edges of the cell and at the tail end of the cell (Fig. 19). In ATP-inhibitory medium, the attachment plaques were in a scattered pattern (Fig. 20), similar to the observed pattern in living cells microinjected with alpha-actinin and vinculin (Figs. 17, 18). During the

Fig. 7. Aggregates of actin stained with HMM-LR in two adjacent cells 5 minutes after recovery from a 90-minute exposure to inhibitors. The centers of the aggregate stain less brightly. Although the nuclci (Nu) are not visible in this fluorescent picture, their position as determined by phase microscopy is indicated. X 1,450.

Fig. 8. A phase micrograph of the same two cells as in Figure 7 illustrates that the actin densities are located near the nuclei. X 1,450.

Fig. 9. Two cells 10 minutes after recovery from a 3-hour exposure to inhibitors. Note the punctate nature of the alpha-actinin localization in the perinuclear aggregates. A permeabilized cell was stained with alpha-actinin-LR. The positions of the nuclei (Nu) as determined by phase microscopy are indicated. X 1,450.

Fig. 10. After 10 minutes of recovery from 3 hours in inhibitory medium, this cell was permeabilized and stained with alpha-actinin- LR. Fine fibers extend into the center of the ring and away from the ring. The position of the nucleus (Nu) is indicated in this picture to show its location with respect to this perinuclear aggreate. X 1,450.

Fig. 11. After a 15-minute recovery period from 135 minutes in inhibitors, radiating actin fibers extend from the ring toward the edge of the cell. The cell was stained with HMM-LR. The position of the nucleus (Nu) is indicated. X 1,450.

Fig. 12. Part of the ring is indented by the nucleus (Nu) in a cell recovering for 15 minutes. The cell had been in inhibitors for 135

(Fig. 14). -Inhibitor-treated C&S were then washed with niinutcs i d was stained with HMM-LR. x 1,450.

124 Glascott et a1

Figs. 13-16. Video images of living PtK2 cells that were microinjected with alpha-actinin-LR. The injected cells were placed back into the incubator overnight and exposed to sodium azide and 2-deoxyglucose the next day. A SIT camera was used to detect fluorescence.

Fig. 13. A control PtK, cell. Note the periodic stress fibers. X2,500.

Fig. 14. Cell 3 hours after being placed into inhibitory medium.

Note the absence of stress fibers and the scattered alpha-actinin plaques. Compare with Figure 3. x 1,050.

Fig. 15. gate of alpha-actinin-LR is present. x 1,OOO.

Fig. 16. Periodic stress fibers are present. x 1,000.

Cell 10 minutes after removing the inhibitor. A ring aggre-

Cell approximately 4.5 hours after removing the inhibitors.

initial recovery period (10 minutes to 1 hour), attachment plaques were difficult to detect in any cells. However, by 1 hour, attachment plaques began to appear at the cell periphery. After 24 hours of reversal, attachment plaques were again observed to be concentrated at the ends of the cell, as in control cells. Cells with fluorescent perinuclear rings of HMM-LR exhibited no complementary ring with reflection contrast optics, and rings could not be detected in living cells with this technique.

In inhibitor-treated and recovering cells, the microtubule network, as determined by indirect immunofluorescence using tubulin antibody, retained its characteristic appearance, radiating outward from a perinuclear organizing center to the cell margins (Figs. 21, 22). This was true in recovering cells with perinuclear rings of contractile proteins (20% of the cells) and in those cells without rings (80% of the cells). Immunofluorescence staining with vimentin antibody also showed that the network of


Figs. 17, 18. A living cell previously coinjected with a mixture of Figs. 19, 20. A control PtK2 cell viewed in reflection contrast alpha-actinin-FITC and vinculin-LR was exposed to a medium con- microscopy illustrating the dark attachment plaques (Fig. 19). In the taining sodium azide and 2-deoxyglucose. Two and a half hours later, presence of sodium azide and 2-deoxyglucose for 2.5 hours, the the alpha-actinin-FITC (Fig. 17) and the vinculin-LR (Fig. 18) were attachment plaques are small and scattered throughout this flat PtK, observed to be colocalized in scattered plaques (arrowheads). Nu, cell (Fig. 20). x 1,500, nucleus. X 1,500.

vimentin-containing intermediate filaments was similarly unaffected by the disassembly and reassembly of stress fibers (data not shown). Control and inhibitor-treated cells were also studied in the transmission electron microscope (TEM). Cells that had been processed for TEM after a 2.5-hour incubation in sodium azide and 2-deoxyglucose revealed the absence of stress fibers, but the presence of scattered, osmophilic, fibrous plaques on the ventral surface of the cell (Fig. 23). These isolated, fibrous bodies were never observed as such in control cells, but they did resemble the attachment plaques at the ends of stress fibers in control cells. The networks of

microtubules (Fig. 23) and intermediate filaments (Fig. 24) in sodium azide-2-deoxyglucose inhibited cells appeared normal.

DISCUSSION

The combination of inhibitors such as sodium azide or sodium cyanide with 2-deoxyglucose reduces cellular levels of ATP to 2% of control values, with a rapid return to normal levels when the inhibitors are removed [Stanley et al, 19801. Complete stress fiber disassembly in PtK, cells depleted of ATP was a slow process, re-

126 Glascott et al

Figs. 21, 22. This PtK2 cell was permeabilized 15 minutes into the recovery period after being exposed to ATP-depleting medium for 2.5 hours. The cell was first allowed to bind HMM-FITC, and then it was stained with tubulin antibodies (and rhodamine-labeled second

antibodies). A ring of actin filaments is present (Fig. 21), but no similar ring of microtubules is observed in the cell (Fig. 22). Nu, nucleus. X 1,600.

quiring almost 2.5 hours. At the end of that time, the attachment plaques were the only remnants of the extensive system of stress fibers that had originally been present in the cells. In contrast, the microtubular and intermediate filament networks were stable in the presence of sodium azide-2-deoxyglucose (Figs. 2 1-24), in agreement with the finding that depressed levels of cellular ATP prevent the depolymerization of microtubules [De Brabander et al, 1981; Bershadsky and Gelfand, 19831.

In contrast to the disassembly process, reassembly of the stress fibers proceeded rapidly in most cells (80%) so that 5-10 minutes after removal from inhibitors, cells contained parallel arrays of long, thin stress fibers. Ten to 30 minutes later the stress fibers had become thicker and regained their normal appearance. We were unable to detect any patterns of actin and alpha-actinin organization that were intermediate between the scattered plaques characteristic of complete stress fiber disassembly and the parallel groups of thin fibers present 5 minutes after reversal. It appeared as though the first thin fibers served as templates onto which additional stress fiber proteins added until the appropriate width was attained. A similar increase in width of myofibrils is seen during myogenesis in the same living cell [Sanger et al, 1986bJ. The ability of alpha-actinin and actin molecules to bind to alpha-actinin molecules [Geiger, 1981; Sanger

et al, 1984a-c] may explain how the initially formed, thin stress fibers could serve as templates for the formation of the thicker stress fibers typical of interphase cells. By monitoring a single living microinjected cell, it should also be possible in future work to determine if the scattered plaques of contractile protein present in treated cells (Figs. 3 , 17) serve as organizing centers for reforming stress fibers.

In the 20% of cells in which actin and alpha-actinin coalesced into radially spreading rings, the centrifugal movement of contractile protein from the initial aggregate toward the cell periphery occurred at a rate of 0.5- 1.0 pm/min, similar to the rate of chromosome movement [for review, see Inoue, 19811 or fibroblast movement [Trinkaus, 19841. Despite the extensive movement of contractile proteins, there was no discernible shape change in the cells, which remained in flat, interconnected sheets. Perhaps extracellular adhesive molecules remain in place, holding the cell’s shape despite the internal cytoskeletal changes [Buck and Horwitz, 19871. There was no indication in reflection contrast optics that the rings were in close contact with the substrate, although the scattered attachment plaques in the inhibitor- treated cells were in close contact with the substrate (Fig. 23). It is not clear why recovery in 20% of cells begins with the disassociation of attachment plaques from the substrate. The subsequent centrifugal movement of the


Fig. 23. Electron micrograph illustrating the presence of a fibrous attachment plaquc (arrow) and a number of microtubules in this PtK, cell treatcd with sodium azide and 2-deoxyglucose for 2.5 hours. The

proximity of the section to the substrate is indicated by the presence of an area of extracellular matrix (s) adjacent to the attachment plaque. X46,OOO.

ring is reminiscent of the centrifugal movement of radial increased by 4 gm over a 90-minute period, indicating stress fibers in spreading cells reported by Soranno and that attachment plaques in normal cells can change their Bell (19821. The experiments in this paper provide addi- position in the cell membrane without losing their stress tional support for the idea that the attachment plaques are fiber attachment. not static structures. Sanger et a1 [1987a] had previously No perinuclear aggregates of cytoskeletal proteins reported that in a living postmitotic cell, the distance were seen as long as the cells were kept in the ATP- between the ends of two newly formed stress fibers depleted medium, even up to 24 hours. It was only upon

128 Glascott et al

Fig. 24. observed in a PtK, cell treated with inhibitors for 2.5 hours. X46,OOO.

Electron micrograph demonstrating a normal wavy pattern of intermediate filaments

removal from inhibitors and recovery from ATP deple- reformation of stress fibers. The capping of receptors tion that the aggregates were seen. Thus, ATP appears to and cytoskeletal proteins seen in lymphocytes has been be necessary for the formation of a perinuclear aggregate shown to be an active process that also requires ATP and of contractile proteins and for the subsequent expansion possibly actin, myosin, and their associated proteins [for of the ring of contractile protein, as well as for the review, see Bourguignon and Bourguignon, 19841, but it


Pochapin, M.B., Sanger, J.M., and Sanger, J.W. (1983): Microinjec- tion of Lucifer yellow CH into sea urchin eggs and embryos. Cell Tissue Res. 234:309-318.

Sanger, J.M., Mittal, B., and Sanger, J.W. (1985a): Structure and assembly of microfilament bundles. In Ishikawa, H., Hatano, S . , and Sato, H. (eds): “Cell Motility: Mechanism and Regu- lation.” Tokyo: University of Tokyo Press, pp 461-476.

Sanger, J.M., Mittal, B., and Sanger, J.W. (1986a): Changes in sarcomeric spacings of stress fibers and polygonal networks in living cells. J. Cell. Biol. 103:203a.

Sanger, J.M., Mittal, B., Pochapin, M., and Sanger, J.W. (1986b): Myofibrillogenesis in living cells microinjected with fluorescently labeled alpha-actinin. J. Cell Biol. 102:2053-2066.

Sanger, J.M., Mittal, B., Pochapin, M., and Sanger, J.W. (1986~): Observations of microfilament bundles in living cells microinjected with fluorescently labeled contractile proteins. J . Cell Sci. Suppl. 5:17-44.

Sanger, J.M., Mittal, B., Pochapin, M.B., and Sanger, J.W. (1987a): Stress fiber and cleavage furrow formation in living cells microinjected with fluorescently labeled alpha-actinin. Cell. Motil. Cytoskel. 7:209-220.

Sanger, J.M., Mittal, B., Wegener, A,, Jockusch, B.M., and Sanger, J.W. (1987b): Differential response of stress fibers and myofibrils to gelsolin. Eur. J. Cell Biol. 44 (in press).

Sanger, J.M., Pochapin, M.B., and Sanger, J.W. (1985b): Midbody sealing after cytokinesis in embryos of the sea urchin, Arbuciu punctulutu. Cell. Tissue Res. 240:287-292.

Sanger, J.M., Sanger, J.W. (1980): Banding and polarity of actin filaments in interphase and cleaving cells. J. Cell Biol. 86568- 575.

Sanger, J. W. (1975a): Changing patterns of actin localization during cell division. Proc. Natl. Acad. Sci. U.S.A. 72: 1913-1916.

Sanger, J.W. (1975b): The presence of actin during chromosomal movement. Proc. Natl. Acad. Sci. U.S.A. 72:2451-2455.

Sanger, J. W. (1975~): Intracellular localization of actin with fluorescently labeled heavy meromyosin. Cell Tissue Res. 161 :432- 444.

Sanger, J.W., Mittal, B., and Sanger, J.M. (1984a): Analysis of myofibrillar structure and assembly using fluorescently labeled contractile proteins. J. Cell Biol. 98:825-833.

Sanger, J.W., Mittal, B., and Sanger, J.M. (1984b): Interaction of fluorescently labeled contractile proteins with the cytoskeleton in cell models. J. Cell Biol. 99:918-928.

Sanger, J.W., Mittal, B., Sanger, J.M. (1984~): Formation of myofibrils in spreading chick cardiac myocytes. Cell Motil. 4:405- 416.

Sanger, J.W., and Sanger, J.M. (1985): Sarcoplasmic reticulum i n the adductor muscles of a Bermuda scallop: Comparison of smooth versus cross-striated portions. Biol. Bull. 168:447- 460.

Sanger, J.W., Sanger, J.M., and Jockusch, B.M. (1983a): Differences in the stress fibers between fibroblasts and epithelial cells. J . Cell Biol. 96:961-969.

Sanger, J.W., Sanger, J.M., and Jockusch, B.M. (1983b): Differen- tial response of three types of actin filament bundles to depletion of cellular ATP levels. Eur. J. Cell Biol. 31: 197-204.

Soranno, T., and Bell, E. (1982): Cytostructural dynamics of spreading and translocating cells. J . Cell. Biol. 95: 127-136.

Stanlg., K . K . , Edwards, M.R., and Luzio, J.P. (1980): Subcellular distribution and movement of 5’ nucleotidase in rat cells. Biochem. J . 186:59-69.

Stopak, D., and Harris, A.K. (1982): Connective tissue morphogen- esis by fibroblast traction. I. Tissue culture observations. Dev. Biol. 90:383-398.

Trinkaus, J.P. (1984): “Cells into Organs: The Forces That Shape the Embryo,” 2nd ed. NJ: Prentice-Hall, Inc., Englewood Cliffs, pp. 187-244.

is not known if ATP is necessary for the dispersal of the capped cytoskeletal proteins that follow internalization of the capped receptors.

ACKNOWLEDGMENTS

This research was supported by grants from the National Institutes of Health (HL-15835 to the Pennsyl- vania Muscle Institute and GM-25653 to J.W.S. and J.M.S.). We are grateful to the National Science Foun- dation for funds (DMB 82-19220) for some of the objec- tives, video cameras, and recorders used in this work.

REFERENCES

Barak, L.S., Yocum, R.R., Nothnagel, E.A., and Webb, W.W. (1980): Fluorescence staining of the actin cytoskeleton in living cells with nitrobenzoxa diazole-phallacidin. Proc. Natl. Acad. Aci. U.S.A. 77:980-984.

Bershadsky, A.D., and Gelfand, V.I. (1983): Role of ATP in the regulation of stability of cytoskeletal structures. Cell Biol. Int. Rep. 7:172-187.

Bershadsky, A.D., Gelfand, V.I.. Svitkina, T.M., and Tint, L.S. (1980): Destruction of microfilament bundles in mouse embryo fibroblasts treated with inhibitors of energy metabolism. Exp. Cell Res. 127:421-429.

Bourguignon, L.Y., and Bourguignon, G.S. (1984): Capping and the cytoskeleton. Int. Rev. Cytol. 87: 195-224.

Buck, C., and Horwitz, A. (1987): Cell surface receptors for extracellular matrix molecules. Annu. Rev. Cell Biol. 3 (in press).

Byers, H.R., White, G.E., and Fujiwara, K. (1984): Organization and function of stress fibers in cells in vitro and in situ. In Shay, J.W. (ed.): “Cell and Muscle Motility.” Vol. 5. New York: Plenum Press, pp. 83-137.

Chen, W.T. (1981): Mechanism of retraction of the trailing edge during fibroblast movement. J. Cell Biol. 90: 187-200.

De Brabander, M., Geuens, G . , Nuydens, R., Willebrords, R. and J. De Mey (1981): Microtubule assembly in living cells after release from nocodazole block: The effects of metabolic inhibitors, taxol and pH. Cell Biol. Int. Rep. 5:913-920.

Evans, R.R., Robson, R.M., and Stromer, M.H. (1984): Properties of smooth muscle vinculin. J. Biol. Chem. 259:3916-3924.

Geiger, B. (1981): The association of rhodamine-labeled alpha-actinin with actin bundles in demembraned cells. Cell Biol. Int. Rep.

Gordon, W.E. (1978): Immunofluorescent and ultrastructural studies of “sarcomeric” units in stress fibers of cultured non-muscle cells. Exp. Cell. Res. 117:253-260.

Harris, A.K., Wild, P., and Stopak, D. (1980): Silicone rubber substrata: A new wrinkle in the study of cell locomotion. Science 208: 177-179.

Herman, I.M., Crisona, N.I., and Pollard, T.D. (1981): Relation between cell activity and distribution of cytoplasmic actin and myosin. J. Cell Biol. 90:8S-91.

Inout, S. (1981): Cell division and the mitotic spindle. J. Cell Biol.

Isenberg, G. , Rdthke, P.C., Hulsmann, N., Franke, W.W., and Wohlfarth-Bottermann, K.E. (1976): Cytoplasmic actomyosin fibrils in tissue culture cells: Direct proof of contractability by visualization of ATP-induced contraction in fibrils isolated by laser microbeam dissection. Cell Tissue Res. 166:427-443.

Kreis, T.E., and Birchmeier, W. (1980): Stress fiber sarcomeres of fibroblasts are contractile. Cell 22:555-561.

5 1627-634.

91:131~-147~.

stress fiber reformation after atp depletion

Documents