Planta (1987)171:30-42
P l a n t a 9 Springer-Verlag 1987
Microtubule cytoskeleton in intact and wounded coenocytic green algae J.W. La Claire II Department of Botany, University of Texas, Austin, TX 78713, USA
Abstract. Microtubule (MT) arrangements were investigated, with immunofluorescence and electron microscopy, in two related species of coenocytic green algae. Intact cells of both Ernodesmis verticillata (Kfitzing) Boergesen and Boergesenia forbesii (Harvey) Feldmann have two morphologically distinct populations of MTs : a highly regular cortical array consisting of a single layer of parallel, longitudinal MTs; and perinuclear MTs radiating from the surface of the envelope of each interphase nucleus. In both algae, mitotic figures lack perinuclear MTs around them. Pre-incubation with taxol does not alter the appearance of these arrays. The cortical and nuclear MTs appear to coexist throughout the nuclear cycle, unlike the condition in most plant cells. At the cut/contracting ends of wounded Ernodesmis cells, cortical MTs exhibit bundling and marked convolution, with some curvature and slight bundling of MTs throughout the cell cortices. In Boergesenia, wound-induced reticulation and separation of the protoplasm into numerous spheres also involves a fasciation of MTs within the attenuating regions of the cytoplasm. Although some cortical MTs are fairly resistant to cold and amiprophos-methyl-induced depolymerization, the perinuclear ones are very labile, depolymerizing in 5 10 rain in the cold. The MT cytoskeleton is not believed to be directly involved in wound-induced motility in these plants because amiprophos-methyl and cold depolymerize most cortical MTs without inhibiting motility. Also, the identical MT distributions in intact cells of these two algae belie the very different patterns of cytoplasmic motility. Although certain roles of Abbreviations." APM-amiprophos-methyl; DIC=differential interference contrast; EGTA = ethylene glycol-bis(fl-aminoethyl ether)-N,N,N',N'-tetraacetic acid; FITC=fluorescein isothiocyanate; MT(s) - microtubule(s); PBS = phosphate-buffered saline
the MT arrays may be ruled out, their exact functions in these plants are not known. Key words: Boergesenia- Chlorophyta - Ernodesmis - Microtubule - Wound healing.
Introduction
The cytoskeleton of plant cells is beginning to emerge as a complex system of microtubules (MTs), microfilaments, and several other structural and regulatory proteins (see Lloyd et al. 1985 for a recent review). Since the first tubulin immuno-localization in plant cells (Franke et al. 1977), use of immunochemistry has made major contributions to the understanding of the plant cytoskeleton. The power of immuno-labelling is that the global distribution of cytoskeletal elements can be visualized in an entire cell or tissue with the light microscope, eliminating the need for tedious serial-section reconstruction to determine the three-dimensional architecture of the cytoskeleton. It should be noted, however, that there are a number of limitations inherent in the technique of immunofluorescence microscopy (e.g. Wang et al. 1982). Studies of plant MTs have led to the concept that the tubulin cytoskeleton in higher plants is a dynamic system that can be composed of up to four distinct arrays of MTs: 1) the interphase cortical array of MTs, 2) the preprophase band of MTs, 3) the mitotic spindle fibers, and 4) the phragmoplast MTs (Lloyd et al. 1985). During studies of wound healing in coenocytic green algae, the potential role of the cytoskeleton in wound-induced motility prompted us to undertake ultrastructural investigations of the MT and
La Claire, J.W., II: Microtubules in coenocytic green algae
31
Fig. 1. Light micrograph of thick section through the apex of Ernodesmis showing the thin protoplasmic layer between the central vacuole (V) and the cell wall (W). x 915; scale bar in gm Fig. 2. Electron micrograph of a transverse section through the peripheral cytoplasmic layer of Ernodesmis, showing evenly-spaced longitudinal MTs in cross section, just beneath the plasma membrane (arrows). x 93,000 ; bar in gm
microfilament systems in these organisms (La Claire 1984). The large size of these coenocytes precludes total reconstruction of the cytoskeleton from serial sections, and therefore localization of tubulin by indirect immunofluorescence has been performed to determine the organization of the MT cytoskeleton in normal cells and during the wound healing in Ernodesmis verticillata (Kfitzing) Boergesen and Boergeseniaforbesii (Harvey) Feldmann, two closely related algae. A technique has been reported for the coenocytic green alga Bryopsis (Menzel and Schliwa 1986 a), that includes fixation and removal of the protoplasm from the cell wall. By opening and flattening the protoplasm on coverslips, large areas of the cytoplasm can be examined in a single plane with immunofluorescence. The two algae possess very different patterns of cytoplasmic motility in response to wounding (e.g. La Claire 1982), but both responses presumably aid the algae in surviving the widespread grazing that occurs in their natural habitats. Upon wounding, the Boergesenia protoplasm reticulates and contrast into hundreds of tiny spherical protoplasts; in Ernodesmis wounding induces longitudinal and centripetal contractions that close the cut end of the cell (La Claire 1982). In an attempt to ascertain any role of MTs in wound-induced motility, effects of cold and the MT poison amiprophos-methyl (APM) on healing and on MT immunofluorescence patterns were investigated in Ernodesmis.
Materials and methods Cultures. Unialgal cultures of Ernodesmis verticillata and Boergesenia forbesii were maintained as described in La Claire (1982). Recently a new line of our Ernodesmis isolate has been obtained with cells twice as large (10 m m long, 1 m m wide on the average) as our normal ones, and therefore more easy to manipulate. The process of wound-healing is identical in every way we can determine, however. Algae are wounded by making a transverse cut with microdissecting scissors at the base of each cell (La Claire 1982). Note: the term " c e l l " is used herein to denote the multinucleate entity containing a single central vacuole and surrounded by a common cell wall. Drug and cold treatments. Amiprophos-methyl (Mobay Chemical Corp., Kansas City, Mo., USA) was prepared as a 1 0 - 2 M stock solution in dimethyl sulfoxide (DMSO). The stock solution was then diluted with growth medium to give final [APM] of 10-50 gM, with the [DMSO] being 0.1-0.5%. Higher concentrations of A P M tended to precipitate in the seawater growth medium. Groups of five to ten Ernodesmis cells were incubated in Petri dishes (15 m m high, 50 m m diameter) containing 10 ml of test solution, for pre-incubations of 1 4 h, to achieve maximal depolymerization of the MTs. The cells were then cut with microdissecting scissors, and allowed to contract/ close for 15 rain (half the time it takes them to heal completely, under normal conditions) prior to chemical fixation for immunofluorescence microscopy. Taxo! (Natural Products Branch, Division of Cancer Treatment, National Cancer Institute, Bethesda, Md., USA; No. NSC 125973) was also dissolved in DMSO, diluted in growth medium, and used at final concentrations of 10 -6 and 10 ~ M (0.01 and 0.1% DMSO). Ernodesmis cells were pre-incubated as above, but for 30 rain, cut, and fixed 60 min later. Thus, cells had a total incubation of 90 rain in taxol, prior to fixation. To determine the degree of cold depolymerization of MTs, groups of five to ten Ernodesmis cells were plunged directly into 10 ml of growth medium pre-chilled on ice, and incubated
32
La Claire, J.W., II : Microtubules in coenocytic green algae for 8 min to 6 h. The cells were then cut and placed in chilled fixative within 3-5 rain. During fixation, the solutions were allowed to warm to room temperature.
Light and electron microscopy. Fixation, embedment and ultramicrotomy were performed as in La Claire (1984). Thick sections (0.1-0.5 ktm) were cut with glass knives, mounted on gelatin-coated microscope slides (Jensen 1962, p. 199) and stained with a modified polychrome stain (Hayat 1975, pp. 270-271). The sections were photographed on Panatomic-X film (Eastman-Kodak, Rochester, N.Y., U S A ; ISO 32), with a Zeiss W L research microscope equipped with a Zeiss OG-590 filter (Carl Zeiss, New York, N.Y., USA). Immunofluorescence. The entire fixation protocols were carried
Fig. 3 a-f. Electr0phoresis and fl-tubulin immunoblotting of Ernodesmis proteins, a Molecular-weight markers with values to left in kDa. Dye front is at the bottom, b Tubulin standard (approx. 55 kDa). c Blot of Ernodesmis proteins stained with India ink. d Immunoblot of tubulin standard, e I m m u n o b l o t of Ernodesmis proteins, f I m m u n o b l o t of Ernodesmis proteins with primary antibody lacking
out at room temperature. Stock fixative carrier is: 0.587 M N a C I + 0 . 1 M K C I + 5 m M MgC12+5 m M ethylene glycolbis(fl-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA) + 5 0 m M piperazine-N,N'-bis-(2-ethanesulfonic acid) (Pipes) (pH 6.8 with concentrated HC1; 1440 mOsm/kg) (which is very similar to our routine electron-microscopy fixation buffer except that the E G T A is substituted for half the MgCI2 and Pipes replaces cacodylate; La Claire 1984). This buffer is used directly for Ernodesmis, but is diluted to 1350 mOsm/kg with distilled water for Boergesenia because of this alga's lower vacuolar tonicity. These values have been empirically determined to provide the best electron-microscopy fixations. The fixative consists of 3.5 ml 4% paraformaldehyde, 0.5 ml 8% glutaraldehyde and 8 ml buffer, to give a final concentration of 1.17% paraformaldehyde and 0.33% glutaraldehyde. Both intact and cut cells were prepared for immunofluorescence. After 30 min in fixative, the cells are rinsed in buffer (with distilled water replacing the aldehydes), then cut longitudinally with microdissecting scissors and the stabilized protoplasm is gently withdrawn from each cell wall. The protoplasm of intact cells adheres somewhat to the cell wall, whereas in cells cut for as little as 2 min prior to fixation the protoplasm is easily removed from the wall with little indication of adhesion. The fixed protoplasts are allowed to settle on poly-L-lysine (Mr 450000)-coated coverslips (Mazia et al. 1975), a technique quite similar to that of Menzel and Schliwa (1986a) for Bryopsis. The coverslips with the protoplasts are then submerged in phosphate-buffered saline (PBS:
Fig. 4a, b. Immunofluorescence control with Ernodesmis. Immunofluorescence (a) and DIC (b) are the same regions in the same focal plane. No antibodies were applied to the cell, showing that no autofluorescence is evident. Cells treated with only secondary antibody and pre-absorption controls also demonstrate no or very little fluorescence, x 725 Figs. 5-8. Immunofluorescence of intact cells of Ernodesmis and Boergesenia. x 725; scale bars in gm Fig. 5. In Ernodesmis, cortical MTs are parallel, straight or slightly wavy, and parallel to the longitudinal axis running from lower left to upper right. Slight bundling is evident and few M T ends are typically apparent Fig. 6. Cortical MTs are not as parallel in Boergesenia, and there is more deviation from the longitudinal axis which runs vertically here Fig. 7. Typical end-on view of Ernodesmis apex showing cortical MTs are irregular, curved and that more M T ends are apparent here than away from the apex (compare Figs. 5, 6) Fig. 8. Perinuclear M T arrays in Boergesenia consist of tangential and radiating MTs from the surface of each interphase nucleus. These MTs are slightly longer and denser than those of Ernodesmis (not shown)
La Claire, J.W., II: Microtubules in coenocytic green algae
33
34 0 . I 3 7 M NaC1, 2 . 7 m M KCI, 1.5mM KH2PO4, 8 r a M NazHPO4, pH 7.3 with 1 N HC1) with 2% cr methylbutyl)phenyl]-co-hydroxypoly(oxy-l,2-ethanediyl) (Triton X-100) for 3-4 h to remove chlorophyll and to promote antibody penetration, and after rinsing with PBS are inverted on 1 ml each of PBS +0.5 mg/ml N a B H 4 for 15 min, to reduce free aldehydes. They are then rinsed again in PBS and each is inverted on a 350-gl drop of approx. 10 gg/ml primary antibody solution consisting of mouse monoclonal to chick brain fl-tubulin (Amersham Corp., Arlington Heights, Ill., USA), diluted 1:750 with PBS +0.5 mg/ml NAN3, for 10-12h in a moist chamber. Rinsing in PBS is followed by inverting the coverslips on 125 gl secondary antibody solution (Amersham sheep anti-mouse immunoglobulin G(IgG), fluorescein isothiocyanate (FITC) labelled, diluted 1 : 25 with PBS) for 2-3 h. After rinsing, each coverslip is mounted on a microscope slide with 10 ~tl mounting medium containing 0.1% p-phenylenediamine to retard fading (Johnson and Araujo 1981), and sealed with nail polish. The immunofluorescence results reported herein are based on observations of more than 250 individual cells. A number of controls were performed to verify antibody specificity and to test method specificity (e.g. Sternberger 1986). Tubulin for pre-absorbing the primary antibody was purified from rabbit brains according to Garland (1978) except that gel filtration was not performed after ion-exchange chromatography. Pre-absortion was carried out with a moiety of i0 gg/ml primary antibody, 150 gg/ml tubulin in PBS, 0.5 mg/ml NaN3 (with a trace of guanosine triphosphate [GTP] and dithiothreitol), and incubated for 75 min at room temperature prior to application on the coverslips. As with all the controls, cells were otherwise treated identically to the normal protocols for indirect immunofluorescence. Specimens were examined with both Nomarski differential interference contrast (DIC) and epifluorescence microscopy, using a Zeiss ICM 405 microscope through a 63X planapochromat lens (numerical aperture-1.4). Images were recorded on Kodak Tri-X film (ISA 400) with 15-s exposures. Fluorescence negatives of tests and their controls were printed in an identical manner. The light source for fluorescence microscopy was a Zeiss HBO 100 mercury burner, and the standard FITC exciter filter .(BP 485/20) was used in conjunction with a band pass 515-565 nm barrier filter (Carl Zeiss) to eliminate any remaining chlorophyll autofluorescence. Another anti-tubulin antibody has been successfully employed in this laboratory with these algae: a rabbit polyclonal anti-soybean tubulin antibody (Cyr et al. 1984) followed by goat anti-rabbit IgG, FITC-labelled antibody (Sigma Chemical Co., St. Louis, Mo., USA).
La Claire, J.W., II : Microtubules in coenocytic green algae
Immunoblotting. A crude sample of Ernodesmis soluble proteins was prepared by grinding 3.5 g of fresh materiaI in 8 mt extraction buffer: 1 mM dithiothreitol, 1 mM EGTA, 2 mM guanosine 5'-triphosphate (lithium salt), 1 m M MgClz, 5 gM leupeptin hemisulfate, 5 gM pepstatin in 0.1 M Pipes, pH 6.9 with I N HC1. The slurry was ultracentrifuged at 29000rpm (SW41Ti rotor; Beckman Instruments, Palo Alto, Cal., USA) for 2 h at 4 ~ C. After carefully removing the lipid layer on top, the supernatant was collected and concentrated to 400 [xl (approx. 25-fold concentration) in a Centricon-30 device (Amicon Corp., Danvers, Mass., USA) at 6500 rpm (SS34 rotor; Sorvall Instruments, Wilmington, Del., USA). The concentrate was mixed 3 : 1 with sample treatment buffer (0.25 M 2-amino2-hydroxymethyl)-l,3-propanediol (Tris), 8% sodium dodecyl sulfate, 20% 2-mercaptoethanol, 0.1% bromophenol blue), placed in boiling water for 5 min, and stored at - 2 0 ~ C. Samples of extraction buffer, tubulin (30 gg/ml), and molecularweight standards (Bio-Rad Lab., Richmond, Cal., USA) were separately boiled with treatment buffer prior to electrophoresis. Immunoblotting was carried out to ensure that the antibody was recognizing specifically tubulin in the plant samples. Polyacrylamide gel electrophoresis (according to Laemmli 1970) was modified for 0.8-mm-thick "'mini-gels" (Idea Scientific Co., Corvallis, Ore., USA) composed of 10% polyacrylamide running gels and 4% stacking gels, both with 2.7% crosslinker. Tubulin (5 gl/lane), crude Ernodesmis protein and molecular-weight standards (10 gl/lane) were loaded and electrophoresis was carried out at 20 mA constant current for 90 min (95 V start to 245 V end). Gels were removed and proteins immediately transferred to nitrocellulose sheets (Schleicher and Schuell, Keene, N.H., USA) at 30 mA constant current (3 V to 5 V) in a semi-dry, rapid transfer apparatus (JKA-Biotech, Bronshoj, Denmark) according to Kyhse-Andersen (1984). After protein transfer, the nitrocellulose was incubated in PBS and cut vertically into strips one lane wide for further processing. The strips were incubated in blocking medium (3% bovine serum albumin +0.1% NaN3 in PBS, pH 7.3 with 1 N NaOH) for 60 min, rinsed again in PBS, and soaked in blocking solution containing primary antibody diluted 1:1000 (approx. 7.5 gg/ml) for 16 h with continuous gentle agitation. After rinsing with PBS, the strips were incubated in secondary antibody (Amersham sheep anti-mouse IgG, peroxidase labelled, diluted 1 : 1000 in PBS) for 5 h with agitation. Further PBS rinses were followed by color development for peroxidase localization (Towbin et al. 1979). For comparison with the immunoblots, nitrocellulose strips were also stained with Coomassie blue or with India ink (Hancock and Tsang 1983). An immunoblot control was performed in which the prima-
Figs. 9-12. Immunofluorescence images of cortical MTs in contracting Ernodesmis (Figs. 9-11) and Boergesenia (Fig. 12) cells. x 725 ; bars in Ixm Fig. 9. Collage showing increased convolution and bundling (note brightness) of MT files as they approach the cut end (just below bottom), in an Ernodesmis cell fixed 5 min after wounding. Bright, out-of-focus spots are nuclei Fig. 10. Very convoluted MTs occur at edge of the mouth of the cut in a contracting cell of Ernodesmis fixed 5 min after wounding. Note that there is increased fluorescence in the cytoplasm at the edge of the cut, toward the bottom of the figure Fig. 11. In face view, at the healed end of an Ernodesmis cell fixed 30 min after wounding, extremely convoluted MT bundles are apparent below the plasma membrane Fig, 12. Cortical MTs in a contracting Boergesenia cell fixed 60 min after wounding, where the cytoplasm is beginning to reticulate and break into protoplasts. A bundling of MTs occurs at the peripheries of developing holes, and thinner strands of MTs are bundled into thicker ones
La Claire, J.W., II: Microtubules in coenocytic green algae
35
36 ry antibody was omitted from the procedure, to test for endogenous peroxidase activity, and for non-specific binding of the secondary antibody.
Results
Ultrastructure Each intact cell of Ernodesmis and Boergesenia consists of a thin layer of protoplasm between the large central vacuole and the cell wall (Fig. 1). Beneath the plasma membrane lies a very thin layer of cortical cytoplasm, below which the majority of plastids, nuclei, and other organelles are located. Close inspection of the cortical cytoplasm shows fairly evenly-spaced MTs that are more-or-less parallel to each other and parallel to the longitudinal axis of the cell (Fig. 2). Sometimes small groups of a few MTs have been observed in the cortex also (not shown). The spacing of these MTs ranges from 0.25 to 1.4 gm in Ernodesmis, and upwards of 2-4 lam in Boergesenia. Microtubules are too long to determine their lengths accurately in thin sections. Occasionally MTs have been detected in the internal cytoplasm, girdling the nuclear envelopes (not shown).
Immuno-localization of fl-tubulin Immunoblotting and immunofluorescence controls. In the immunoblots of partially-purified rabbit tubulin, the monoclonal antibody to fl-tubulin binds predominantly to the 55-kDa tubulin band; it also recognizes a band at approx. 40 kDa (possibly a tubulin degradation product) as well as a few highmolecular-weight bands (Fig. 3). Although the
La Claire, J.W., II: Microtubules in coenocytic green algae
blots of Ernodesmis proteins display numerous bands upon staining with India ink, a single band at 55 kDa is recognized by the antibody (Fig. 3). Blotted samples treated with only primary or only secondary antibodies exhibited no evidence of native peroxidase activity nor spurious binding of the secondary antibody. This demonstrates that the antibody is only binding tubulin in the crude protein extracts of Ernodesmis. Immunofluorescence controls demonstrated that there is practically no autofluorescence apparent in the cells following our fixation protocols (Fig. 4). A similar lack of fluorescence was observed in cells treated with normal mouse serum or PBS in lieu of primary-antibody solution, or when the primary antibody is pre-absorbed with rabbit tubulin (not shown).
Immunofluorescence of intact cells. In Ernodesmis, focusing at the level near the plasma membrane demonstrates the presence of nearly parallel, longitudinally-oriented MTs which may be straight or slightly wavy in the middle of the cell but which are always fairly evenly spaced (Fig. 5). In places, brighter regions splay into two or more strands, verifying that a localized bundling of MTs exists, to a greater degree than that indicated by thin sections. Thus it is not possible to trace the exact lengths of MTs, but it appears that many are hundreds of micrometers long. Boergesenia also has predominantly longitudinal MTs in a single focal plane near the cell surface (Fig. 6). However, they usually are not quite as parallel as those in Ernodesmis, and greater scatter/variation from the longitudinal axis is visible in many cells. As the cortical MTs are traced to within 250-500 gm of the cell apex, they are more convoluted in appearance (Fig. 7) and at the very apex there is a varia-
Fig. 13. Cortical MTs of a contracting Boergesenia cell (60 min after wounding) in a region where the cytoplasm is breaking up into discrete protoplasts, the surface of one being in the middle. There is a great degree of M T bundling; some strands are seen with 20 or more MTs converging into them. At the top is a region where strands are thinning, prior to breaking. x 880; here and in Figs. 14-16 scale bars in gm Fig. 14a, b. Immunofluorescence (a) and corresponding DIC (b) images in same focal plane in an Ernodesmis cell treated in 25 pM A P M for 4 h, cut and let contract for 15 min before fixation. Only one cortical M T fragment is apparent in this region (arrow), but there is much diffuse cytoplasmic fluorescence outlining the plastids. Note the clumped nature of the cytoplasm. Although not in the focal plane, perinuclear and spindle MTs have also disappeared, x 725 Fig. 15. Tubulin immunofluorescence of a wounded/healing Ernodesmis cell pretreated with 1 g M taxol for 30 min, cut in the taxol solution and let contract for 60 min before fixation. Taxol treatment may prevent complete closure even within 60 min of wounding. The small opening remaining can be seen in face view, surrounded by very convoluted M T bundles around the rim. Except for out-of-focus regions, diffuse background fluorescence may be less here than in untreated cells, x 725 Fig. 16. Treatment of Ernodesmis with anti-soybean tubulin antibodies. At the level of the lower chloroplast envelopes, fluorescent rings (0.75 gm diameter) are evident as are out-of-focus nuclei and perinuclear arrays, x 725
La Claire, J.W., II: Microtubules in coenocytic green algae
37
38
bly-sized patch where the immunofluorescence patterns are irregular, wavy, and randomly arranged (Fig. 7). In both algae a markedly larger number of MT ends are apparent in the apices than in other regions (compare Figs. 5 and 6, and 7). This pattern of cortical MTs was observed in all intact cells examined, of both species, regardless of size or age. There was no obvious helical pitch to these MTs, nor were MTs ever lacking. A second, spatially-separate array of MTs is evident in the cytoplasm near the tonoplast, where the majority of nuclei occur. These MTs surround, and radiate from, non-dividing nuclei (Fig. 8). Although exact MT numbers were impossible to determine because of the brightness of fluorescence, the perinuclear arrays seem to consist of slightly shorter (up to 20 ~m) and sparser MTs in Ernodesmis than in Boergesenia (up to 30+ gin). In the latter, through-focusing indicates that some MTs traverse from one nucleus to another. Perinuclear MTs are not evident for the most part around mitotic nuclei; most of the fluorescence is localized in spindle MTs and in small, aster-like arrays at the spindle poles (not shown). Although centrioles are known to accompany nuclear division in these algae (Hori and Enomoto 1978), they are not evident with tubulin immunofluorescence, probably because of the extremely bright fluorescence of the spindle and associated MTs.
Immunofluorescence of contracting cells. Immediately upon wounding an Ernodesmis cell, longitudinal cytoplasmic contractions begin retracting the cut end from the wound site and centripetal contractions at the cut begin to close the wound (La Claire 1982). Tracing the M T files toward the cut/ closing end of the Ernodesmis cells shows a marked increase in convolution and apparent bundling of MTs, even in cells fixed only 5 min after being cut (Figs. 9, 10). Near the mouth of the cut, diffuse cytoplasmic fluorescence is also greater, probably in part as a consequence of the increasing thickness of cytoplasm. Also, there is greater corrugation of the cytoplasmic surface, resulting in fluorescence behind the focal plane. Approximately 30 rain after wounding, longitudinal contractions have resulted in maximal shortening of the protoplasm, and the mouth of the cut is completely closed/healed. Cells fixed when nearly healed show extremely convoluted MT bundles at the closing region, which are quite striking in face view (Fig. 11). There is a high degree of corrugation of the cell surface that polylysine apparently does not smooth out; thus not all the cortical MTs lie in one focal plane. Here again, diffuse cytoplasmic
La Claire, J.W., II: Microtubules in coenocytic green algae
fluorescence is also quite evident in thicker regions. Cells allowed to heal and grow overnight show a similar MT distribution to that of intact cells (e.g., Fig. 5). Wounding does seem to stimulate mitotic divisions in patches of nuclei near the cut end, but otherwise there are no obvious changes in the perinuclear MT arrays during would healing. In Boergesenia the cytoplasm reticulates and breaks up into hundreds of spherical protoplasts within 90 min of wounding (Enomoto and Hirose 1972; La Claire 1982). The changes in MT immuno fluorescence patterns during this process are somewhat different from the situation in Ernodesmis. Thirty minutes after wounding, small holes are evident throughout the cytoplasm in face view, and as would be expected, these regions are devoid of MTs. As the holes grow wider during the wound response, a fasciation of MTs is apparent on the lateral sides of the holes, that develop into thickening MT strands within 60 min of wounding (Fig. 12). Prior to cytoplasmic spheration, highlyfluorescent strands connect thicker regions of protoplasm; into the strands dozens of MTs may converge (Fig. 13). From electron microscopy, these bright strands are known to be cytoplasmic connections up to 1 + pm in diameter, that are packed with MTs. As the reticulum of cytoplasm breaks into separate spherating masses, these strands thin out and are broken, severing connections between the nascent protoplasts. An indication of this is seen in the upper part of Fig. 13, where thinning strands are apparent. Very little convolution of MTs occurs in Boergesenia during the wound response, and no major changes in perinuclear MTs are apparent.
Microtubule depolymerization experiments Amiprophos-methyl treatments. Pre-incubating Ernodesmis cells in up to 5 0 p M APM ( + 0 . 5 % DMSO) for up to 4 h has no effect on the rate and degree of wound-induced contractions: cells close completely and normally in the test solutions within 30 min. There is a clumping/aggregation of organelles evident during the pre-incubation in APM (Fig. 14). If treated cells are wounded and prepared for fl-tubulin immunofluorescence, most cortical MTs are gone within 3-4 h of treatment with A P M ; a few short ones remain (Fig. 14). However, there is abundant diffuse fluorescence in the cytoplasm silhouetting the chloroplasts (Fig. 14), that most likely represents depolymerized tubulin. In contrast, the perinuclear arrays are
La Claire, J.W., II: Microtubules in coenocytic green algae
not apparent around interphase nuclei within 1 h of APM incubation, the only fluorescence evident being that in nuclei with highly condensed chromosomes (probably representing collapsed spindle tubulin). Dimethylsulfoxide (0.5%) alone has no obvious effect on MTs in Ernodesmis. Cold treatments. Up to 6 h on ice does not decrease the cells' healing ability: cells allowed to warm to room temperature for 60 min after treatment heal normally, although there may be minor differences in the rate of healing/contractions. Cells incubated on ice for only 8 min and then cut and immediately prepared for immunofluorescence, show no perinuclear MT arrays, but cortical MTs look normal (not shown). After 6 h on ice most (but not all) cortical MTs are gone, and diffuse cytoplasmic fluorescence is apparent, similar to that in the APMtreated cell in Fig. 14. Other treatments. Pre-incubation of Ernodesmis cells in 10 gM taxol ( + 0 . 1 % DMSO) for 30 min before cutting does not affect the initial patterns of wound-induced contractions, but the centripetal contractions cease prior to complete closure of the wound in all cells examined. For example, cells pre-incubated in taxol, and cut and fixed 60 min later for immunofluorescence may not be quite closed (Fig. 15). There may be slightly less diffuse background fluorescence in the cytoplasm at the cut than is found in untreated cells (except for that resulting from out-of-focus fluorescence) (Fig. 15). If one examines regions slightly away from the cut in taxol-treated cells, the MT patterns are identical to untreated cells (not shown). In addition to confirming both cortical and perinuclear MT immunofiuorescence patterns, use of the anti-soybean polyclonal antibody showed the presence of fine fluorescent rings (approx. 0.75 gm diameter) at the level of the lower chloroplast surfaces (Fig. 16). These have not been seen with other antibodies nor with electron microscoPY. Discussion
The uniform distribution and visual clarity of the cortical MTs in intact cells of Ernodesmis and Boergesenia are both striking and unique. To my knowledge, only the coenocytic green alga Bryopsis (Menzel and Schliwa, 1986a, b) has been shown to have such an extensive array of cortical MTs; but in Bryopsis, these MTs are much more sinuous and greater MT bundling was reported. Strictly longitudinal arrays of cortical MTs also occur in
39
pollen tubes (Derksen et al. 1985) and in some root-hair cells (Newcomb and Bonnett 1965; Lloyd and Wells 1985; Traas et al. 1985), although the latter case is somewhat controversial. Most higher-plant cells are believed to have variouslypitched helical arrays (e.g. Lloyd and Seagull 1985). It appears that the cortical MT arrays in Ernodesmis and Boergesenia are not an artefact of handling or protocols, because the immunofluorescence patterns correspond very well with ultrastructural data, and such images were obtained consistently with different tubulin antibodies. The similarity between the protocols for immunofluorescence fixation and for electron microscopy of these cells, used in this work, provide further confidence; fixation techniques are not a trivial concern in immunofluorescence (see Hepler 1985, for further discussion). Also, it is highly unlikely that E G T A is inducing artificial polymerization of tubulin (Traas et al. 1985); and because taxol-treated cells do not differ markedly from untreated cells, tubulin depolymerization during fixation is probably not a problem. In the present study, correlative ultrastructural data indicate that immunofluorescence is apparently "staining" single MTs in intact ceils, as reported by Osborn etal. (1978). The absolute lengths of individual MTs are not known because of their slight bundling, but it is certain that they are many times longer than typical plant MTs (Lloyd 1984). Well-developed perinuclear arrays of MTs have not been widely reported in plants, unlike the complex arrays in many animal cell types (e.g. Osborn et al. 1978). Notable exceptions are Haemanthus endosperm (De M e y e t al. 1982), moss protonemata (Doonan etal. 1985) and Acetabularia (Woodcock 1971 ; Menzel 1986), and less extensive, transitory perinuclear MTs do occur in some higher plants (e.g. Wick 1985). Aside from being a potential source of spindle tubulin, one function of the perinuclear arrays may be to maintain the even distribution of nuclei seen in these (Fig. 8) and related algal coenocytes (Staves and La Claire 1985). Anchorage of nuclei by MTs has been reported in various plants (see Gunning and Hardham 1982 for review), and interphase nuclei do not translocate in intact Ernodesmis or Boergesenia cells. Interestingly, perinuclear MT arrays were not seen in Bryopsis, where nuclei continuously migrate (Menzel and Schliwa 1986a, b). In some unicellular organisms and in higherplant cells, there are typically four, mutually-exclusive sets of MTs throughout the cell/nuclear cycle (e.g. Clayton 1985). This is clearly not the case
40 in the algae studied here. Firstly, the cortical array persists throughout the nuclear cycle, and secondly, it is distinct and spatially separate from the nuclear/perinuclear MTs. However, the perinuclear arrays do appear to recycle into spindle and aster MTs, as they are not visible during mitosis. No preprophase bands of MTs nor phragmoplasts/ phycoplasts are known in Ernodesmis or Boergesenia, but the unusual mode of cell division (" segregative cell division") has not been studied in detail. Also, one would not expect, a priori, preprophase bands of MTs where cytokinesis is uncoupled from mitosis. Such distinct differences between the coenocytic green algae and typical higher plant cells may reflect their phylogenetic separation, or these differences may represent simple variations deriving from the size and coenocytic habit of these algae. It is tempting to speculate that it might be more efficient for coenocytes to maintain a separate population of MTs (or pools of tubulin) around the nuclei for repeated nuclear divison. Thus, the perinuclear MTs of Ernodesmis and Boergesenia may represent a fifth distinct MT array, even if they are very transitory in the higher plants that have them. It is more difficult to assign a function to the cortical MTs in Ernodesrnis and Boergesenia. As noted below, they are not essential for wound-induced cytoplasmic contractions. These algae do not exhibit any type of cytoplasmic streaming, unlike other coenocytic green algae like Bryopsis, where MTs are directly involved in intracellular movements. It is also unlikely that the cortical MTs of Ernodesmis and Boergesenia participate primarily in maintaining cell shape as was suggested for carrot cells (Lloyd et al. 1980). It is believed that elongating cells do not balloon out because of having predominantly transverse MTs (e.g. Lloyd and Barlow 1982), yet in all stages of development, the cylindrical Ernodesmis cells have solely longitudinally oriented cortical MTs. Also, MT-depolymerizing drugs like APM, and cold, do not alter protoplasmic morphology in Ernodesmis or Boergesenia. Lastly, although it is widely held that cortical MTs orient cellulose deposition in many plants (e.g. Lloyd 1984), such models cannot account for wall formation in these coenocytes. The thick, cellulosic walls of Boergesenia (Mizuta and Wada 1981) and the related Valonia (Itoh and Brown 1984) are polylamellate; yet in the former the cortical MTs are always longitudinal, and in the latter, MT orientation does not correlate with the directionality of microfibrils in the innermost wall layer (Itoh and Brown 1984). It seems that
La Claire, J.W., II : Microtubulesin coenocyticgreen algae different mechanisms control wall deposition in the special case of coenocytic algae. One probable function of the cortical MTs in these algae is to increase structural stability of the thin protoplasmic layer surrounding the large central vacuole. When Ernodesrnis cells were treated with A P M or cold, handling the protoplasm was much more difficult than usual; the protoplasm was brittle and easily torn. A second function might be to maintain an even spatial distribution of peripheral organelles (e.g., chloroplasts) throughout the cytoplasm. Although no direct connections between plastids and cortical MTs were ever seen, clumping of organelles was apparent in APM- (Fig. 14) or cold-treated cells. Also at cell apices (where cortical MTs are less organized), a thicker layer of cytoplasm is often found in which two or more strata of chloroplasts occur. Finally, it is also possible that cortical MTs function in controlling cytokinesis in these coenocytes, a phenomenon that is not well understood in these algae. Despite the similar cortical MT patterns in intact cells of Ernodesrnis and Boergesenia, the changes in MT distribution during their respective wound responses are different from each other. The degree of convolution of MTs in Ernodesmis gives the visual impression of being the result of contraction rather than its cause. I am aware of no other system demonstrating such an extent of MT convolution in vivo, although flagellar beating, for example, demonstrates the flexibility of axonemal MTs. However, bending is typically the result of M T - M T interaction (sliding). In Boergesenia, the strands of cytoplasm containing the MT bundles thin out and eventually break during the wound response, so possibly MT sliding within the bundles plays some role in separation of the nascent protoplasts. But the initial bundling also appears to be a passive phenomenon, resulting from the expanding holes in the cytoplasm. As a specific and potent MT inhibitor in plants (Morejohn and Fosket 1984), APM (and cold) treatment demonstrates unequivocally that most or all cortical (and perinuclear) MTs can be depolymerized with no apparent effect on wound healing. The diffuse cytoplasmic fluorescence in APMor cold-treated Ernodesmis cells probably represents depolymerized tubulin, even though nonpolymerized tubulin has been visualized in very few studies (Albertini and Clarke 1981; Wick and Duniec 1983, 1984; Van Lammeren et al. 1985). Since Ernodesmis and Boergesenia are tropical algae, it was not expected that cortical MTs would resist
La Claire, J.W., II : Microtubules in coenocytic green algae
cold depolymerization. There may be depolymerization of cortical MTs normally occurring at the mouth of the cut during closure: prevention of depolymerization by taxol, a MT-stabilizing agent (Schiff et al. 1979), could inhibit final closure by the physical "stacking u p " of MTs abundant in this region. Finally, the greater lability of perinuclear MTs may underscore molecular differences between the two sets of MTs in these algae. The author is very grateful to: Dr. John A. West (University of California, Berkeley) for the original algal isolates; R.T. Evans and C.F. Smead (Agricultural Chemicals Division, Mobay Corporation, Kansas City, Mo., USA) for the generous gift of APM ; Dr. M. Suffness (Natural Products Branch, Division of Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, Md., USA) for the kind gift of taxol; Dr. R. Cyr (University of Georgia, Athens, USA) for kindly supplying the soybean tubulin antibodies; Dr. Guy A. Thompson, Jr. and Dr. Helen A. Norman (University of Texas) for use of the ultracentrifuge; Dr. Paul A. Richmond (University of the Pacific, Stockton, Cal., USA) for numerous discussions; and Julie M. Palmer (University of Texas) for proofreading the manuscript. Portions of this work were supported by National Science Foundation grant DCB 84-02345.
References Albertini, D.F., Clark, J.I. (1981) Visualization of assembled and disassembled microtubule protein by double label fluorescence microscopy. Cell Biol. Int. Rep. 5, 387-397 Clayton, L. (1985) The cytoskeleton and the plant cell cycle. In: The cell division cycle in plants, pp. 113-131, Bryan, J.A., Francis, D., eds. Cambridge University Press, Cambridge, UK Cyr, R., Tochi, L., Fosket, D.E. (1984) Immunological studies on plant tubulins isolated from diverse cell lines. J. Cell Biol. 99, 41a De Mey, J., Lambert, A.M., Bajer, A.S., Moeremans, M., De Brabander, M. (1982) Visualization of microtubules in interphase and mitotic plant cells of Haemanthus endosperm with the immuno-gold staining method. Proc. Natl. Acad. Sci. USA 79, 1898-1902 Derksen, J., Pierson, E.S., Traas, J.A. (1985) Microtubules in vegetative and generative cells of pollen tubes. Eur. J. Cell Biol. 38, 14~148 Doonan, J.H., Cove, D.J., Lloyd, C.W. (1985) Immunof!uorescence microscopy of microtubules in intact cell lineages of the moss, Physcomitrella patens. I. Normal and CIPCtreated tip cells. J. Cell Sci. 75, 131 147 Enomoto, S., Hirose, H. (1972) Culture studies on artificially induced aplanospores and their development in the marine alga Boergesenia forbesii (Harvey) Feldmann (Chlorophyeae, Siphonocladales). Phycologia 11, 119-122 Franke, W.W., Seib, E., Osborn, M., Weber, K., Herth, W., Falk, H. (1977) Tubulin-containingstructures in the anastral mitotic apparatus of endosperm cells of the plant Leucojum aestivum as revealed by immunofluorescence microscopy. Cytobiologie 15, 2448 Garland, D.L. (1978) Kinetics and mechanism of colchicine binding to tubulin: evidence for ligand-induced conformational change. Biochemistry 17, 4266-4272 Gunning, B.E.S., Hardham, A.R. (1982) Microtubules. Annu. Rev. Plant Physiol. 33, 651-698
41 Hancock, K., Tsang, V.C.W. (1983) India ink staining of proteins on nitrocellulose paper. Anal. Biochem. 133, 157-162 Hayat, M.A. (1975) Positive staining for electron microscopy. Van Nostrand Reinhold Co., New York Hepler, P.K. (1985) The plant cytoskeleton. In: Botanical microscopy 1985, pp. 233-262, Robards, A.W., ed. Oxford University Press, Oxford, UK Hori, T., Enomoto, S. (1978) Electron microscope observations on the nuclear division in Valonia ventricosa (Chlorophyceae, Siphonocladales). Phycologia 17, 133-142 Itoh, T., Brown, R.M., Jr. (1984) The assembly of cellulose microfibrils in Valonia macrophysa Kfitz. Planta 160, 372-381 Jensen, W.A. (1962) Botanical histochemistry. Principles and practice. W.H. Freeman and Co., San Francisco Johnson, G.D., Araujo, G.M.N. (1981) A simple method of reducing the fading of immunofluorescence during microscopy. J. Immunol. Meth. 43, 349-350 Kyhse-Andersen, J. (1984) Electroblotting of multiple gels: a simple apparatus without buffer tank for rapid transfer of proteins from polyacrylamide to nitrocellulose. J. Biochem. Biophys. Meth. 10, 203-209 La Claire, J.W., II. (1982) Cytomorphological aspects of wound healing in selected Siphonocladales (Chlorophyceae). J. Phycol. 18, 379-382 La Claire, J.W., II. (1984) Cell motility during wound healing in giant algal cells: contraction in detergent-permeabilized cell models ofErnodesmis. Eur. J. Cell Biol. 33, 180-189 Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685 Lloyd, C.W. (1984) Toward a dynamic helical model for the influence of microtubules on wall patterns in plants. Int. Rev. Cytol. 86, 1-51 Lloyd, C.W., Bartow, P.W. (1982) The co-ordination of celt division and elongation: the role of the cytoskeleton. In: The cytoskeleton in plant growth and development, pp. 203-228, Lloyd, C.W., ed. Academic Press, London Lloyd, C.W., Clayton, L., Dawson, P.J., Doonan, J.H., Hulme, J.S., Roberts, I.N., Wells, B. (1985) The cytoskeleton underlying side walls and cross walls in plants: molecules and macromolecular assemblies. J. Cell Sci. Suppl. 2, 143-155 Lloyd, C.W., Seagull, R.W. (1985) A new spring for plant cell biology: microtubules as dynamic helices. Trends Biochem. Sci. 10, 476-478 Lloyd, C.W., Slabas, A.R., Powell, A.J., Lowe, S.B. (1980) Microtubules, protoplasts and plant cell shape. An immunofluorescent study. Planta 147, 500-506 Lloyd, C.W., Wells, B. (1985) Microtubules are at the tips of root hairs and form helical patterns corresponding to inner wall fibrils. J. Cell Sci. 75, 225-238 Mazia, D., Schatten, G., Sale, W. (1975) Adhesion of cells to surfaces coated with polylysine. Applications to electron microscopy. J. Cell Biol. 66, 198-200 Menzel, D. (1986) Visualization of cytoskeletal changes through the life cycle in Acetabularia. Protoplasma 134, 30-42 Menzel, D., Schliwa, M. (1986a) Motility in the siphonous green alga Bryopsis. I. Spatial organization of the cytoskeleton and organelle movements. Eur. J. Cell Biol. 40, 275-285 Menzel, D., Schliwa, M. (1986b) Motility in the siphonous green alga Bryopsis. II. Chloroplast movement requires organized arrays of both microtubules and actin filaments. Eur. J. Cell Biol. 40, 286-295 Mizuta, S., Wada, S. (1981) Microfibrillar structure of growing cell wall in a coenocytic green alga, Boergesenia forbesii. Bot. Mag. Tokyo 94, 343-353
42 Morejohn, L.C., Fosket, D.E. (1984)Inhibition of plant microtubule polymerization in vitro by the phosphoric amide herbicide amiprophos-methyl. Science 224, 874-876 Newcomb, E.H., Bonnett, H.T., Jr. (1965) Cytoplasmic microtubule and wall microfibril orientation in root hairs of radish. J. Cell Biol. 27, 575 589 Osborn, M., Webster, R.E., Weber, K. (1978) Individual microtubules viewed by immunofluorescenceand electron microscopy in the same PtK2 cell. J. Cell Biol. 77, R27-R34 Schiff, P.B., Fant, J., Horwitz, S.B. (1979) Promotion of microtubule assembly in vitro by taxol. Nature 277, 665-667 Staves, M.P., La Claire, J.W., II. (1985) Nuclear synchrony in Valonia macrophysa (Chlorophyta) : light microscopy and flow cytometry. J. Phycol. 21, 68-71 Sternberger, L.A. (1986) Immunocytochemistry, 3rd edn. John Wiley & Sons, New York Towbin, H., Staehelin, T., Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76, 43504354 Traas, J.A., Braat, P., Emons, A.M.C., Meekes, H., Derksen, J. (1985) Microtubules in root hairs. J. Cell Sci. 76, 303-320 Van Lammeren, A.A.M., Keijzer, C.J., Willemse, M.T.M., Kieft, H. (1985) Structure and function of the microtubular
La Claire, J.W., II: Microtubules in coenocytic green algae cytoskeleton during pollen development in Gasteria verrucosa (Mill.) H. Duval. Planta 165, 1-11 Wang, K., Feramisco, J.R., Ash, J.F. (1982) Fluorescent localization of contractile proteins in tissue culture cells. Meth. Enzymol. 85, 514-562 Wick, S.M. (1985) Immunofluorescence microscopy of tubulin and microtubule arrays in plant cells. III. Transition between mitotic/cytokinetic and interphase microtubule arrays. Cell Biol. Int. Rep. 9, 357-371 Wick, S.M., Duniec, J. (1983) Immunofluorescence microscopy of tubulin and microtubule arrays in plant cells. I. Preprophase band development and concomitant appearance of nuclear envelope-associated tubulin. J. Cell Biol. 97, 235-243 Wick, S.M., Duniec, J. (1984) Immunofluorescence microscopy of tubulin and microtubule arrays in plant cells. II. Transition between the pre-prophase band and the mitotic spindle. Protoplasma 122, 45-55 Woodcock, C.L. (1971) The anchoring of nuclei by cytoplasmic microtubules in Acetabularia. J. Cell Sci. 8, 611-621
Received 6 October; accepted 11 December 1986
P l a n t a 9 Springer-Verlag 1987
Microtubule cytoskeleton in intact and wounded coenocytic green algae J.W. La Claire II Department of Botany, University of Texas, Austin, TX 78713, USA
Abstract. Microtubule (MT) arrangements were investigated, with immunofluorescence and electron microscopy, in two related species of coenocytic green algae. Intact cells of both Ernodesmis verticillata (Kfitzing) Boergesen and Boergesenia forbesii (Harvey) Feldmann have two morphologically distinct populations of MTs : a highly regular cortical array consisting of a single layer of parallel, longitudinal MTs; and perinuclear MTs radiating from the surface of the envelope of each interphase nucleus. In both algae, mitotic figures lack perinuclear MTs around them. Pre-incubation with taxol does not alter the appearance of these arrays. The cortical and nuclear MTs appear to coexist throughout the nuclear cycle, unlike the condition in most plant cells. At the cut/contracting ends of wounded Ernodesmis cells, cortical MTs exhibit bundling and marked convolution, with some curvature and slight bundling of MTs throughout the cell cortices. In Boergesenia, wound-induced reticulation and separation of the protoplasm into numerous spheres also involves a fasciation of MTs within the attenuating regions of the cytoplasm. Although some cortical MTs are fairly resistant to cold and amiprophos-methyl-induced depolymerization, the perinuclear ones are very labile, depolymerizing in 5 10 rain in the cold. The MT cytoskeleton is not believed to be directly involved in wound-induced motility in these plants because amiprophos-methyl and cold depolymerize most cortical MTs without inhibiting motility. Also, the identical MT distributions in intact cells of these two algae belie the very different patterns of cytoplasmic motility. Although certain roles of Abbreviations." APM-amiprophos-methyl; DIC=differential interference contrast; EGTA = ethylene glycol-bis(fl-aminoethyl ether)-N,N,N',N'-tetraacetic acid; FITC=fluorescein isothiocyanate; MT(s) - microtubule(s); PBS = phosphate-buffered saline
the MT arrays may be ruled out, their exact functions in these plants are not known. Key words: Boergesenia- Chlorophyta - Ernodesmis - Microtubule - Wound healing.
Introduction
The cytoskeleton of plant cells is beginning to emerge as a complex system of microtubules (MTs), microfilaments, and several other structural and regulatory proteins (see Lloyd et al. 1985 for a recent review). Since the first tubulin immuno-localization in plant cells (Franke et al. 1977), use of immunochemistry has made major contributions to the understanding of the plant cytoskeleton. The power of immuno-labelling is that the global distribution of cytoskeletal elements can be visualized in an entire cell or tissue with the light microscope, eliminating the need for tedious serial-section reconstruction to determine the three-dimensional architecture of the cytoskeleton. It should be noted, however, that there are a number of limitations inherent in the technique of immunofluorescence microscopy (e.g. Wang et al. 1982). Studies of plant MTs have led to the concept that the tubulin cytoskeleton in higher plants is a dynamic system that can be composed of up to four distinct arrays of MTs: 1) the interphase cortical array of MTs, 2) the preprophase band of MTs, 3) the mitotic spindle fibers, and 4) the phragmoplast MTs (Lloyd et al. 1985). During studies of wound healing in coenocytic green algae, the potential role of the cytoskeleton in wound-induced motility prompted us to undertake ultrastructural investigations of the MT and
La Claire, J.W., II: Microtubules in coenocytic green algae
31
Fig. 1. Light micrograph of thick section through the apex of Ernodesmis showing the thin protoplasmic layer between the central vacuole (V) and the cell wall (W). x 915; scale bar in gm Fig. 2. Electron micrograph of a transverse section through the peripheral cytoplasmic layer of Ernodesmis, showing evenly-spaced longitudinal MTs in cross section, just beneath the plasma membrane (arrows). x 93,000 ; bar in gm
microfilament systems in these organisms (La Claire 1984). The large size of these coenocytes precludes total reconstruction of the cytoskeleton from serial sections, and therefore localization of tubulin by indirect immunofluorescence has been performed to determine the organization of the MT cytoskeleton in normal cells and during the wound healing in Ernodesmis verticillata (Kfitzing) Boergesen and Boergeseniaforbesii (Harvey) Feldmann, two closely related algae. A technique has been reported for the coenocytic green alga Bryopsis (Menzel and Schliwa 1986 a), that includes fixation and removal of the protoplasm from the cell wall. By opening and flattening the protoplasm on coverslips, large areas of the cytoplasm can be examined in a single plane with immunofluorescence. The two algae possess very different patterns of cytoplasmic motility in response to wounding (e.g. La Claire 1982), but both responses presumably aid the algae in surviving the widespread grazing that occurs in their natural habitats. Upon wounding, the Boergesenia protoplasm reticulates and contrast into hundreds of tiny spherical protoplasts; in Ernodesmis wounding induces longitudinal and centripetal contractions that close the cut end of the cell (La Claire 1982). In an attempt to ascertain any role of MTs in wound-induced motility, effects of cold and the MT poison amiprophos-methyl (APM) on healing and on MT immunofluorescence patterns were investigated in Ernodesmis.
Materials and methods Cultures. Unialgal cultures of Ernodesmis verticillata and Boergesenia forbesii were maintained as described in La Claire (1982). Recently a new line of our Ernodesmis isolate has been obtained with cells twice as large (10 m m long, 1 m m wide on the average) as our normal ones, and therefore more easy to manipulate. The process of wound-healing is identical in every way we can determine, however. Algae are wounded by making a transverse cut with microdissecting scissors at the base of each cell (La Claire 1982). Note: the term " c e l l " is used herein to denote the multinucleate entity containing a single central vacuole and surrounded by a common cell wall. Drug and cold treatments. Amiprophos-methyl (Mobay Chemical Corp., Kansas City, Mo., USA) was prepared as a 1 0 - 2 M stock solution in dimethyl sulfoxide (DMSO). The stock solution was then diluted with growth medium to give final [APM] of 10-50 gM, with the [DMSO] being 0.1-0.5%. Higher concentrations of A P M tended to precipitate in the seawater growth medium. Groups of five to ten Ernodesmis cells were incubated in Petri dishes (15 m m high, 50 m m diameter) containing 10 ml of test solution, for pre-incubations of 1 4 h, to achieve maximal depolymerization of the MTs. The cells were then cut with microdissecting scissors, and allowed to contract/ close for 15 rain (half the time it takes them to heal completely, under normal conditions) prior to chemical fixation for immunofluorescence microscopy. Taxo! (Natural Products Branch, Division of Cancer Treatment, National Cancer Institute, Bethesda, Md., USA; No. NSC 125973) was also dissolved in DMSO, diluted in growth medium, and used at final concentrations of 10 -6 and 10 ~ M (0.01 and 0.1% DMSO). Ernodesmis cells were pre-incubated as above, but for 30 rain, cut, and fixed 60 min later. Thus, cells had a total incubation of 90 rain in taxol, prior to fixation. To determine the degree of cold depolymerization of MTs, groups of five to ten Ernodesmis cells were plunged directly into 10 ml of growth medium pre-chilled on ice, and incubated
32
La Claire, J.W., II : Microtubules in coenocytic green algae for 8 min to 6 h. The cells were then cut and placed in chilled fixative within 3-5 rain. During fixation, the solutions were allowed to warm to room temperature.
Light and electron microscopy. Fixation, embedment and ultramicrotomy were performed as in La Claire (1984). Thick sections (0.1-0.5 ktm) were cut with glass knives, mounted on gelatin-coated microscope slides (Jensen 1962, p. 199) and stained with a modified polychrome stain (Hayat 1975, pp. 270-271). The sections were photographed on Panatomic-X film (Eastman-Kodak, Rochester, N.Y., U S A ; ISO 32), with a Zeiss W L research microscope equipped with a Zeiss OG-590 filter (Carl Zeiss, New York, N.Y., USA). Immunofluorescence. The entire fixation protocols were carried
Fig. 3 a-f. Electr0phoresis and fl-tubulin immunoblotting of Ernodesmis proteins, a Molecular-weight markers with values to left in kDa. Dye front is at the bottom, b Tubulin standard (approx. 55 kDa). c Blot of Ernodesmis proteins stained with India ink. d Immunoblot of tubulin standard, e I m m u n o b l o t of Ernodesmis proteins, f I m m u n o b l o t of Ernodesmis proteins with primary antibody lacking
out at room temperature. Stock fixative carrier is: 0.587 M N a C I + 0 . 1 M K C I + 5 m M MgC12+5 m M ethylene glycolbis(fl-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA) + 5 0 m M piperazine-N,N'-bis-(2-ethanesulfonic acid) (Pipes) (pH 6.8 with concentrated HC1; 1440 mOsm/kg) (which is very similar to our routine electron-microscopy fixation buffer except that the E G T A is substituted for half the MgCI2 and Pipes replaces cacodylate; La Claire 1984). This buffer is used directly for Ernodesmis, but is diluted to 1350 mOsm/kg with distilled water for Boergesenia because of this alga's lower vacuolar tonicity. These values have been empirically determined to provide the best electron-microscopy fixations. The fixative consists of 3.5 ml 4% paraformaldehyde, 0.5 ml 8% glutaraldehyde and 8 ml buffer, to give a final concentration of 1.17% paraformaldehyde and 0.33% glutaraldehyde. Both intact and cut cells were prepared for immunofluorescence. After 30 min in fixative, the cells are rinsed in buffer (with distilled water replacing the aldehydes), then cut longitudinally with microdissecting scissors and the stabilized protoplasm is gently withdrawn from each cell wall. The protoplasm of intact cells adheres somewhat to the cell wall, whereas in cells cut for as little as 2 min prior to fixation the protoplasm is easily removed from the wall with little indication of adhesion. The fixed protoplasts are allowed to settle on poly-L-lysine (Mr 450000)-coated coverslips (Mazia et al. 1975), a technique quite similar to that of Menzel and Schliwa (1986a) for Bryopsis. The coverslips with the protoplasts are then submerged in phosphate-buffered saline (PBS:
Fig. 4a, b. Immunofluorescence control with Ernodesmis. Immunofluorescence (a) and DIC (b) are the same regions in the same focal plane. No antibodies were applied to the cell, showing that no autofluorescence is evident. Cells treated with only secondary antibody and pre-absorption controls also demonstrate no or very little fluorescence, x 725 Figs. 5-8. Immunofluorescence of intact cells of Ernodesmis and Boergesenia. x 725; scale bars in gm Fig. 5. In Ernodesmis, cortical MTs are parallel, straight or slightly wavy, and parallel to the longitudinal axis running from lower left to upper right. Slight bundling is evident and few M T ends are typically apparent Fig. 6. Cortical MTs are not as parallel in Boergesenia, and there is more deviation from the longitudinal axis which runs vertically here Fig. 7. Typical end-on view of Ernodesmis apex showing cortical MTs are irregular, curved and that more M T ends are apparent here than away from the apex (compare Figs. 5, 6) Fig. 8. Perinuclear M T arrays in Boergesenia consist of tangential and radiating MTs from the surface of each interphase nucleus. These MTs are slightly longer and denser than those of Ernodesmis (not shown)
La Claire, J.W., II: Microtubules in coenocytic green algae
33
34 0 . I 3 7 M NaC1, 2 . 7 m M KCI, 1.5mM KH2PO4, 8 r a M NazHPO4, pH 7.3 with 1 N HC1) with 2% cr methylbutyl)phenyl]-co-hydroxypoly(oxy-l,2-ethanediyl) (Triton X-100) for 3-4 h to remove chlorophyll and to promote antibody penetration, and after rinsing with PBS are inverted on 1 ml each of PBS +0.5 mg/ml N a B H 4 for 15 min, to reduce free aldehydes. They are then rinsed again in PBS and each is inverted on a 350-gl drop of approx. 10 gg/ml primary antibody solution consisting of mouse monoclonal to chick brain fl-tubulin (Amersham Corp., Arlington Heights, Ill., USA), diluted 1:750 with PBS +0.5 mg/ml NAN3, for 10-12h in a moist chamber. Rinsing in PBS is followed by inverting the coverslips on 125 gl secondary antibody solution (Amersham sheep anti-mouse immunoglobulin G(IgG), fluorescein isothiocyanate (FITC) labelled, diluted 1 : 25 with PBS) for 2-3 h. After rinsing, each coverslip is mounted on a microscope slide with 10 ~tl mounting medium containing 0.1% p-phenylenediamine to retard fading (Johnson and Araujo 1981), and sealed with nail polish. The immunofluorescence results reported herein are based on observations of more than 250 individual cells. A number of controls were performed to verify antibody specificity and to test method specificity (e.g. Sternberger 1986). Tubulin for pre-absorbing the primary antibody was purified from rabbit brains according to Garland (1978) except that gel filtration was not performed after ion-exchange chromatography. Pre-absortion was carried out with a moiety of i0 gg/ml primary antibody, 150 gg/ml tubulin in PBS, 0.5 mg/ml NaN3 (with a trace of guanosine triphosphate [GTP] and dithiothreitol), and incubated for 75 min at room temperature prior to application on the coverslips. As with all the controls, cells were otherwise treated identically to the normal protocols for indirect immunofluorescence. Specimens were examined with both Nomarski differential interference contrast (DIC) and epifluorescence microscopy, using a Zeiss ICM 405 microscope through a 63X planapochromat lens (numerical aperture-1.4). Images were recorded on Kodak Tri-X film (ISA 400) with 15-s exposures. Fluorescence negatives of tests and their controls were printed in an identical manner. The light source for fluorescence microscopy was a Zeiss HBO 100 mercury burner, and the standard FITC exciter filter .(BP 485/20) was used in conjunction with a band pass 515-565 nm barrier filter (Carl Zeiss) to eliminate any remaining chlorophyll autofluorescence. Another anti-tubulin antibody has been successfully employed in this laboratory with these algae: a rabbit polyclonal anti-soybean tubulin antibody (Cyr et al. 1984) followed by goat anti-rabbit IgG, FITC-labelled antibody (Sigma Chemical Co., St. Louis, Mo., USA).
La Claire, J.W., II : Microtubules in coenocytic green algae
Immunoblotting. A crude sample of Ernodesmis soluble proteins was prepared by grinding 3.5 g of fresh materiaI in 8 mt extraction buffer: 1 mM dithiothreitol, 1 mM EGTA, 2 mM guanosine 5'-triphosphate (lithium salt), 1 m M MgClz, 5 gM leupeptin hemisulfate, 5 gM pepstatin in 0.1 M Pipes, pH 6.9 with I N HC1. The slurry was ultracentrifuged at 29000rpm (SW41Ti rotor; Beckman Instruments, Palo Alto, Cal., USA) for 2 h at 4 ~ C. After carefully removing the lipid layer on top, the supernatant was collected and concentrated to 400 [xl (approx. 25-fold concentration) in a Centricon-30 device (Amicon Corp., Danvers, Mass., USA) at 6500 rpm (SS34 rotor; Sorvall Instruments, Wilmington, Del., USA). The concentrate was mixed 3 : 1 with sample treatment buffer (0.25 M 2-amino2-hydroxymethyl)-l,3-propanediol (Tris), 8% sodium dodecyl sulfate, 20% 2-mercaptoethanol, 0.1% bromophenol blue), placed in boiling water for 5 min, and stored at - 2 0 ~ C. Samples of extraction buffer, tubulin (30 gg/ml), and molecularweight standards (Bio-Rad Lab., Richmond, Cal., USA) were separately boiled with treatment buffer prior to electrophoresis. Immunoblotting was carried out to ensure that the antibody was recognizing specifically tubulin in the plant samples. Polyacrylamide gel electrophoresis (according to Laemmli 1970) was modified for 0.8-mm-thick "'mini-gels" (Idea Scientific Co., Corvallis, Ore., USA) composed of 10% polyacrylamide running gels and 4% stacking gels, both with 2.7% crosslinker. Tubulin (5 gl/lane), crude Ernodesmis protein and molecular-weight standards (10 gl/lane) were loaded and electrophoresis was carried out at 20 mA constant current for 90 min (95 V start to 245 V end). Gels were removed and proteins immediately transferred to nitrocellulose sheets (Schleicher and Schuell, Keene, N.H., USA) at 30 mA constant current (3 V to 5 V) in a semi-dry, rapid transfer apparatus (JKA-Biotech, Bronshoj, Denmark) according to Kyhse-Andersen (1984). After protein transfer, the nitrocellulose was incubated in PBS and cut vertically into strips one lane wide for further processing. The strips were incubated in blocking medium (3% bovine serum albumin +0.1% NaN3 in PBS, pH 7.3 with 1 N NaOH) for 60 min, rinsed again in PBS, and soaked in blocking solution containing primary antibody diluted 1:1000 (approx. 7.5 gg/ml) for 16 h with continuous gentle agitation. After rinsing with PBS, the strips were incubated in secondary antibody (Amersham sheep anti-mouse IgG, peroxidase labelled, diluted 1 : 1000 in PBS) for 5 h with agitation. Further PBS rinses were followed by color development for peroxidase localization (Towbin et al. 1979). For comparison with the immunoblots, nitrocellulose strips were also stained with Coomassie blue or with India ink (Hancock and Tsang 1983). An immunoblot control was performed in which the prima-
Figs. 9-12. Immunofluorescence images of cortical MTs in contracting Ernodesmis (Figs. 9-11) and Boergesenia (Fig. 12) cells. x 725 ; bars in Ixm Fig. 9. Collage showing increased convolution and bundling (note brightness) of MT files as they approach the cut end (just below bottom), in an Ernodesmis cell fixed 5 min after wounding. Bright, out-of-focus spots are nuclei Fig. 10. Very convoluted MTs occur at edge of the mouth of the cut in a contracting cell of Ernodesmis fixed 5 min after wounding. Note that there is increased fluorescence in the cytoplasm at the edge of the cut, toward the bottom of the figure Fig. 11. In face view, at the healed end of an Ernodesmis cell fixed 30 min after wounding, extremely convoluted MT bundles are apparent below the plasma membrane Fig, 12. Cortical MTs in a contracting Boergesenia cell fixed 60 min after wounding, where the cytoplasm is beginning to reticulate and break into protoplasts. A bundling of MTs occurs at the peripheries of developing holes, and thinner strands of MTs are bundled into thicker ones
La Claire, J.W., II: Microtubules in coenocytic green algae
35
36 ry antibody was omitted from the procedure, to test for endogenous peroxidase activity, and for non-specific binding of the secondary antibody.
Results
Ultrastructure Each intact cell of Ernodesmis and Boergesenia consists of a thin layer of protoplasm between the large central vacuole and the cell wall (Fig. 1). Beneath the plasma membrane lies a very thin layer of cortical cytoplasm, below which the majority of plastids, nuclei, and other organelles are located. Close inspection of the cortical cytoplasm shows fairly evenly-spaced MTs that are more-or-less parallel to each other and parallel to the longitudinal axis of the cell (Fig. 2). Sometimes small groups of a few MTs have been observed in the cortex also (not shown). The spacing of these MTs ranges from 0.25 to 1.4 gm in Ernodesmis, and upwards of 2-4 lam in Boergesenia. Microtubules are too long to determine their lengths accurately in thin sections. Occasionally MTs have been detected in the internal cytoplasm, girdling the nuclear envelopes (not shown).
Immuno-localization of fl-tubulin Immunoblotting and immunofluorescence controls. In the immunoblots of partially-purified rabbit tubulin, the monoclonal antibody to fl-tubulin binds predominantly to the 55-kDa tubulin band; it also recognizes a band at approx. 40 kDa (possibly a tubulin degradation product) as well as a few highmolecular-weight bands (Fig. 3). Although the
La Claire, J.W., II: Microtubules in coenocytic green algae
blots of Ernodesmis proteins display numerous bands upon staining with India ink, a single band at 55 kDa is recognized by the antibody (Fig. 3). Blotted samples treated with only primary or only secondary antibodies exhibited no evidence of native peroxidase activity nor spurious binding of the secondary antibody. This demonstrates that the antibody is only binding tubulin in the crude protein extracts of Ernodesmis. Immunofluorescence controls demonstrated that there is practically no autofluorescence apparent in the cells following our fixation protocols (Fig. 4). A similar lack of fluorescence was observed in cells treated with normal mouse serum or PBS in lieu of primary-antibody solution, or when the primary antibody is pre-absorbed with rabbit tubulin (not shown).
Immunofluorescence of intact cells. In Ernodesmis, focusing at the level near the plasma membrane demonstrates the presence of nearly parallel, longitudinally-oriented MTs which may be straight or slightly wavy in the middle of the cell but which are always fairly evenly spaced (Fig. 5). In places, brighter regions splay into two or more strands, verifying that a localized bundling of MTs exists, to a greater degree than that indicated by thin sections. Thus it is not possible to trace the exact lengths of MTs, but it appears that many are hundreds of micrometers long. Boergesenia also has predominantly longitudinal MTs in a single focal plane near the cell surface (Fig. 6). However, they usually are not quite as parallel as those in Ernodesmis, and greater scatter/variation from the longitudinal axis is visible in many cells. As the cortical MTs are traced to within 250-500 gm of the cell apex, they are more convoluted in appearance (Fig. 7) and at the very apex there is a varia-
Fig. 13. Cortical MTs of a contracting Boergesenia cell (60 min after wounding) in a region where the cytoplasm is breaking up into discrete protoplasts, the surface of one being in the middle. There is a great degree of M T bundling; some strands are seen with 20 or more MTs converging into them. At the top is a region where strands are thinning, prior to breaking. x 880; here and in Figs. 14-16 scale bars in gm Fig. 14a, b. Immunofluorescence (a) and corresponding DIC (b) images in same focal plane in an Ernodesmis cell treated in 25 pM A P M for 4 h, cut and let contract for 15 min before fixation. Only one cortical M T fragment is apparent in this region (arrow), but there is much diffuse cytoplasmic fluorescence outlining the plastids. Note the clumped nature of the cytoplasm. Although not in the focal plane, perinuclear and spindle MTs have also disappeared, x 725 Fig. 15. Tubulin immunofluorescence of a wounded/healing Ernodesmis cell pretreated with 1 g M taxol for 30 min, cut in the taxol solution and let contract for 60 min before fixation. Taxol treatment may prevent complete closure even within 60 min of wounding. The small opening remaining can be seen in face view, surrounded by very convoluted M T bundles around the rim. Except for out-of-focus regions, diffuse background fluorescence may be less here than in untreated cells, x 725 Fig. 16. Treatment of Ernodesmis with anti-soybean tubulin antibodies. At the level of the lower chloroplast envelopes, fluorescent rings (0.75 gm diameter) are evident as are out-of-focus nuclei and perinuclear arrays, x 725
La Claire, J.W., II: Microtubules in coenocytic green algae
37
38
bly-sized patch where the immunofluorescence patterns are irregular, wavy, and randomly arranged (Fig. 7). In both algae a markedly larger number of MT ends are apparent in the apices than in other regions (compare Figs. 5 and 6, and 7). This pattern of cortical MTs was observed in all intact cells examined, of both species, regardless of size or age. There was no obvious helical pitch to these MTs, nor were MTs ever lacking. A second, spatially-separate array of MTs is evident in the cytoplasm near the tonoplast, where the majority of nuclei occur. These MTs surround, and radiate from, non-dividing nuclei (Fig. 8). Although exact MT numbers were impossible to determine because of the brightness of fluorescence, the perinuclear arrays seem to consist of slightly shorter (up to 20 ~m) and sparser MTs in Ernodesmis than in Boergesenia (up to 30+ gin). In the latter, through-focusing indicates that some MTs traverse from one nucleus to another. Perinuclear MTs are not evident for the most part around mitotic nuclei; most of the fluorescence is localized in spindle MTs and in small, aster-like arrays at the spindle poles (not shown). Although centrioles are known to accompany nuclear division in these algae (Hori and Enomoto 1978), they are not evident with tubulin immunofluorescence, probably because of the extremely bright fluorescence of the spindle and associated MTs.
Immunofluorescence of contracting cells. Immediately upon wounding an Ernodesmis cell, longitudinal cytoplasmic contractions begin retracting the cut end from the wound site and centripetal contractions at the cut begin to close the wound (La Claire 1982). Tracing the M T files toward the cut/ closing end of the Ernodesmis cells shows a marked increase in convolution and apparent bundling of MTs, even in cells fixed only 5 min after being cut (Figs. 9, 10). Near the mouth of the cut, diffuse cytoplasmic fluorescence is also greater, probably in part as a consequence of the increasing thickness of cytoplasm. Also, there is greater corrugation of the cytoplasmic surface, resulting in fluorescence behind the focal plane. Approximately 30 rain after wounding, longitudinal contractions have resulted in maximal shortening of the protoplasm, and the mouth of the cut is completely closed/healed. Cells fixed when nearly healed show extremely convoluted MT bundles at the closing region, which are quite striking in face view (Fig. 11). There is a high degree of corrugation of the cell surface that polylysine apparently does not smooth out; thus not all the cortical MTs lie in one focal plane. Here again, diffuse cytoplasmic
La Claire, J.W., II: Microtubules in coenocytic green algae
fluorescence is also quite evident in thicker regions. Cells allowed to heal and grow overnight show a similar MT distribution to that of intact cells (e.g., Fig. 5). Wounding does seem to stimulate mitotic divisions in patches of nuclei near the cut end, but otherwise there are no obvious changes in the perinuclear MT arrays during would healing. In Boergesenia the cytoplasm reticulates and breaks up into hundreds of spherical protoplasts within 90 min of wounding (Enomoto and Hirose 1972; La Claire 1982). The changes in MT immuno fluorescence patterns during this process are somewhat different from the situation in Ernodesmis. Thirty minutes after wounding, small holes are evident throughout the cytoplasm in face view, and as would be expected, these regions are devoid of MTs. As the holes grow wider during the wound response, a fasciation of MTs is apparent on the lateral sides of the holes, that develop into thickening MT strands within 60 min of wounding (Fig. 12). Prior to cytoplasmic spheration, highlyfluorescent strands connect thicker regions of protoplasm; into the strands dozens of MTs may converge (Fig. 13). From electron microscopy, these bright strands are known to be cytoplasmic connections up to 1 + pm in diameter, that are packed with MTs. As the reticulum of cytoplasm breaks into separate spherating masses, these strands thin out and are broken, severing connections between the nascent protoplasts. An indication of this is seen in the upper part of Fig. 13, where thinning strands are apparent. Very little convolution of MTs occurs in Boergesenia during the wound response, and no major changes in perinuclear MTs are apparent.
Microtubule depolymerization experiments Amiprophos-methyl treatments. Pre-incubating Ernodesmis cells in up to 5 0 p M APM ( + 0 . 5 % DMSO) for up to 4 h has no effect on the rate and degree of wound-induced contractions: cells close completely and normally in the test solutions within 30 min. There is a clumping/aggregation of organelles evident during the pre-incubation in APM (Fig. 14). If treated cells are wounded and prepared for fl-tubulin immunofluorescence, most cortical MTs are gone within 3-4 h of treatment with A P M ; a few short ones remain (Fig. 14). However, there is abundant diffuse fluorescence in the cytoplasm silhouetting the chloroplasts (Fig. 14), that most likely represents depolymerized tubulin. In contrast, the perinuclear arrays are
La Claire, J.W., II: Microtubules in coenocytic green algae
not apparent around interphase nuclei within 1 h of APM incubation, the only fluorescence evident being that in nuclei with highly condensed chromosomes (probably representing collapsed spindle tubulin). Dimethylsulfoxide (0.5%) alone has no obvious effect on MTs in Ernodesmis. Cold treatments. Up to 6 h on ice does not decrease the cells' healing ability: cells allowed to warm to room temperature for 60 min after treatment heal normally, although there may be minor differences in the rate of healing/contractions. Cells incubated on ice for only 8 min and then cut and immediately prepared for immunofluorescence, show no perinuclear MT arrays, but cortical MTs look normal (not shown). After 6 h on ice most (but not all) cortical MTs are gone, and diffuse cytoplasmic fluorescence is apparent, similar to that in the APMtreated cell in Fig. 14. Other treatments. Pre-incubation of Ernodesmis cells in 10 gM taxol ( + 0 . 1 % DMSO) for 30 min before cutting does not affect the initial patterns of wound-induced contractions, but the centripetal contractions cease prior to complete closure of the wound in all cells examined. For example, cells pre-incubated in taxol, and cut and fixed 60 min later for immunofluorescence may not be quite closed (Fig. 15). There may be slightly less diffuse background fluorescence in the cytoplasm at the cut than is found in untreated cells (except for that resulting from out-of-focus fluorescence) (Fig. 15). If one examines regions slightly away from the cut in taxol-treated cells, the MT patterns are identical to untreated cells (not shown). In addition to confirming both cortical and perinuclear MT immunofiuorescence patterns, use of the anti-soybean polyclonal antibody showed the presence of fine fluorescent rings (approx. 0.75 gm diameter) at the level of the lower chloroplast surfaces (Fig. 16). These have not been seen with other antibodies nor with electron microscoPY. Discussion
The uniform distribution and visual clarity of the cortical MTs in intact cells of Ernodesmis and Boergesenia are both striking and unique. To my knowledge, only the coenocytic green alga Bryopsis (Menzel and Schliwa, 1986a, b) has been shown to have such an extensive array of cortical MTs; but in Bryopsis, these MTs are much more sinuous and greater MT bundling was reported. Strictly longitudinal arrays of cortical MTs also occur in
39
pollen tubes (Derksen et al. 1985) and in some root-hair cells (Newcomb and Bonnett 1965; Lloyd and Wells 1985; Traas et al. 1985), although the latter case is somewhat controversial. Most higher-plant cells are believed to have variouslypitched helical arrays (e.g. Lloyd and Seagull 1985). It appears that the cortical MT arrays in Ernodesmis and Boergesenia are not an artefact of handling or protocols, because the immunofluorescence patterns correspond very well with ultrastructural data, and such images were obtained consistently with different tubulin antibodies. The similarity between the protocols for immunofluorescence fixation and for electron microscopy of these cells, used in this work, provide further confidence; fixation techniques are not a trivial concern in immunofluorescence (see Hepler 1985, for further discussion). Also, it is highly unlikely that E G T A is inducing artificial polymerization of tubulin (Traas et al. 1985); and because taxol-treated cells do not differ markedly from untreated cells, tubulin depolymerization during fixation is probably not a problem. In the present study, correlative ultrastructural data indicate that immunofluorescence is apparently "staining" single MTs in intact ceils, as reported by Osborn etal. (1978). The absolute lengths of individual MTs are not known because of their slight bundling, but it is certain that they are many times longer than typical plant MTs (Lloyd 1984). Well-developed perinuclear arrays of MTs have not been widely reported in plants, unlike the complex arrays in many animal cell types (e.g. Osborn et al. 1978). Notable exceptions are Haemanthus endosperm (De M e y e t al. 1982), moss protonemata (Doonan etal. 1985) and Acetabularia (Woodcock 1971 ; Menzel 1986), and less extensive, transitory perinuclear MTs do occur in some higher plants (e.g. Wick 1985). Aside from being a potential source of spindle tubulin, one function of the perinuclear arrays may be to maintain the even distribution of nuclei seen in these (Fig. 8) and related algal coenocytes (Staves and La Claire 1985). Anchorage of nuclei by MTs has been reported in various plants (see Gunning and Hardham 1982 for review), and interphase nuclei do not translocate in intact Ernodesmis or Boergesenia cells. Interestingly, perinuclear MT arrays were not seen in Bryopsis, where nuclei continuously migrate (Menzel and Schliwa 1986a, b). In some unicellular organisms and in higherplant cells, there are typically four, mutually-exclusive sets of MTs throughout the cell/nuclear cycle (e.g. Clayton 1985). This is clearly not the case
40 in the algae studied here. Firstly, the cortical array persists throughout the nuclear cycle, and secondly, it is distinct and spatially separate from the nuclear/perinuclear MTs. However, the perinuclear arrays do appear to recycle into spindle and aster MTs, as they are not visible during mitosis. No preprophase bands of MTs nor phragmoplasts/ phycoplasts are known in Ernodesmis or Boergesenia, but the unusual mode of cell division (" segregative cell division") has not been studied in detail. Also, one would not expect, a priori, preprophase bands of MTs where cytokinesis is uncoupled from mitosis. Such distinct differences between the coenocytic green algae and typical higher plant cells may reflect their phylogenetic separation, or these differences may represent simple variations deriving from the size and coenocytic habit of these algae. It is tempting to speculate that it might be more efficient for coenocytes to maintain a separate population of MTs (or pools of tubulin) around the nuclei for repeated nuclear divison. Thus, the perinuclear MTs of Ernodesmis and Boergesenia may represent a fifth distinct MT array, even if they are very transitory in the higher plants that have them. It is more difficult to assign a function to the cortical MTs in Ernodesrnis and Boergesenia. As noted below, they are not essential for wound-induced cytoplasmic contractions. These algae do not exhibit any type of cytoplasmic streaming, unlike other coenocytic green algae like Bryopsis, where MTs are directly involved in intracellular movements. It is also unlikely that the cortical MTs of Ernodesmis and Boergesenia participate primarily in maintaining cell shape as was suggested for carrot cells (Lloyd et al. 1980). It is believed that elongating cells do not balloon out because of having predominantly transverse MTs (e.g. Lloyd and Barlow 1982), yet in all stages of development, the cylindrical Ernodesmis cells have solely longitudinally oriented cortical MTs. Also, MT-depolymerizing drugs like APM, and cold, do not alter protoplasmic morphology in Ernodesmis or Boergesenia. Lastly, although it is widely held that cortical MTs orient cellulose deposition in many plants (e.g. Lloyd 1984), such models cannot account for wall formation in these coenocytes. The thick, cellulosic walls of Boergesenia (Mizuta and Wada 1981) and the related Valonia (Itoh and Brown 1984) are polylamellate; yet in the former the cortical MTs are always longitudinal, and in the latter, MT orientation does not correlate with the directionality of microfibrils in the innermost wall layer (Itoh and Brown 1984). It seems that
La Claire, J.W., II : Microtubulesin coenocyticgreen algae different mechanisms control wall deposition in the special case of coenocytic algae. One probable function of the cortical MTs in these algae is to increase structural stability of the thin protoplasmic layer surrounding the large central vacuole. When Ernodesrnis cells were treated with A P M or cold, handling the protoplasm was much more difficult than usual; the protoplasm was brittle and easily torn. A second function might be to maintain an even spatial distribution of peripheral organelles (e.g., chloroplasts) throughout the cytoplasm. Although no direct connections between plastids and cortical MTs were ever seen, clumping of organelles was apparent in APM- (Fig. 14) or cold-treated cells. Also at cell apices (where cortical MTs are less organized), a thicker layer of cytoplasm is often found in which two or more strata of chloroplasts occur. Finally, it is also possible that cortical MTs function in controlling cytokinesis in these coenocytes, a phenomenon that is not well understood in these algae. Despite the similar cortical MT patterns in intact cells of Ernodesrnis and Boergesenia, the changes in MT distribution during their respective wound responses are different from each other. The degree of convolution of MTs in Ernodesmis gives the visual impression of being the result of contraction rather than its cause. I am aware of no other system demonstrating such an extent of MT convolution in vivo, although flagellar beating, for example, demonstrates the flexibility of axonemal MTs. However, bending is typically the result of M T - M T interaction (sliding). In Boergesenia, the strands of cytoplasm containing the MT bundles thin out and eventually break during the wound response, so possibly MT sliding within the bundles plays some role in separation of the nascent protoplasts. But the initial bundling also appears to be a passive phenomenon, resulting from the expanding holes in the cytoplasm. As a specific and potent MT inhibitor in plants (Morejohn and Fosket 1984), APM (and cold) treatment demonstrates unequivocally that most or all cortical (and perinuclear) MTs can be depolymerized with no apparent effect on wound healing. The diffuse cytoplasmic fluorescence in APMor cold-treated Ernodesmis cells probably represents depolymerized tubulin, even though nonpolymerized tubulin has been visualized in very few studies (Albertini and Clarke 1981; Wick and Duniec 1983, 1984; Van Lammeren et al. 1985). Since Ernodesmis and Boergesenia are tropical algae, it was not expected that cortical MTs would resist
La Claire, J.W., II : Microtubules in coenocytic green algae
cold depolymerization. There may be depolymerization of cortical MTs normally occurring at the mouth of the cut during closure: prevention of depolymerization by taxol, a MT-stabilizing agent (Schiff et al. 1979), could inhibit final closure by the physical "stacking u p " of MTs abundant in this region. Finally, the greater lability of perinuclear MTs may underscore molecular differences between the two sets of MTs in these algae. The author is very grateful to: Dr. John A. West (University of California, Berkeley) for the original algal isolates; R.T. Evans and C.F. Smead (Agricultural Chemicals Division, Mobay Corporation, Kansas City, Mo., USA) for the generous gift of APM ; Dr. M. Suffness (Natural Products Branch, Division of Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, Md., USA) for the kind gift of taxol; Dr. R. Cyr (University of Georgia, Athens, USA) for kindly supplying the soybean tubulin antibodies; Dr. Guy A. Thompson, Jr. and Dr. Helen A. Norman (University of Texas) for use of the ultracentrifuge; Dr. Paul A. Richmond (University of the Pacific, Stockton, Cal., USA) for numerous discussions; and Julie M. Palmer (University of Texas) for proofreading the manuscript. Portions of this work were supported by National Science Foundation grant DCB 84-02345.
References Albertini, D.F., Clark, J.I. (1981) Visualization of assembled and disassembled microtubule protein by double label fluorescence microscopy. Cell Biol. Int. Rep. 5, 387-397 Clayton, L. (1985) The cytoskeleton and the plant cell cycle. In: The cell division cycle in plants, pp. 113-131, Bryan, J.A., Francis, D., eds. Cambridge University Press, Cambridge, UK Cyr, R., Tochi, L., Fosket, D.E. (1984) Immunological studies on plant tubulins isolated from diverse cell lines. J. Cell Biol. 99, 41a De Mey, J., Lambert, A.M., Bajer, A.S., Moeremans, M., De Brabander, M. (1982) Visualization of microtubules in interphase and mitotic plant cells of Haemanthus endosperm with the immuno-gold staining method. Proc. Natl. Acad. Sci. USA 79, 1898-1902 Derksen, J., Pierson, E.S., Traas, J.A. (1985) Microtubules in vegetative and generative cells of pollen tubes. Eur. J. Cell Biol. 38, 14~148 Doonan, J.H., Cove, D.J., Lloyd, C.W. (1985) Immunof!uorescence microscopy of microtubules in intact cell lineages of the moss, Physcomitrella patens. I. Normal and CIPCtreated tip cells. J. Cell Sci. 75, 131 147 Enomoto, S., Hirose, H. (1972) Culture studies on artificially induced aplanospores and their development in the marine alga Boergesenia forbesii (Harvey) Feldmann (Chlorophyeae, Siphonocladales). Phycologia 11, 119-122 Franke, W.W., Seib, E., Osborn, M., Weber, K., Herth, W., Falk, H. (1977) Tubulin-containingstructures in the anastral mitotic apparatus of endosperm cells of the plant Leucojum aestivum as revealed by immunofluorescence microscopy. Cytobiologie 15, 2448 Garland, D.L. (1978) Kinetics and mechanism of colchicine binding to tubulin: evidence for ligand-induced conformational change. Biochemistry 17, 4266-4272 Gunning, B.E.S., Hardham, A.R. (1982) Microtubules. Annu. Rev. Plant Physiol. 33, 651-698
41 Hancock, K., Tsang, V.C.W. (1983) India ink staining of proteins on nitrocellulose paper. Anal. Biochem. 133, 157-162 Hayat, M.A. (1975) Positive staining for electron microscopy. Van Nostrand Reinhold Co., New York Hepler, P.K. (1985) The plant cytoskeleton. In: Botanical microscopy 1985, pp. 233-262, Robards, A.W., ed. Oxford University Press, Oxford, UK Hori, T., Enomoto, S. (1978) Electron microscope observations on the nuclear division in Valonia ventricosa (Chlorophyceae, Siphonocladales). Phycologia 17, 133-142 Itoh, T., Brown, R.M., Jr. (1984) The assembly of cellulose microfibrils in Valonia macrophysa Kfitz. Planta 160, 372-381 Jensen, W.A. (1962) Botanical histochemistry. Principles and practice. W.H. Freeman and Co., San Francisco Johnson, G.D., Araujo, G.M.N. (1981) A simple method of reducing the fading of immunofluorescence during microscopy. J. Immunol. Meth. 43, 349-350 Kyhse-Andersen, J. (1984) Electroblotting of multiple gels: a simple apparatus without buffer tank for rapid transfer of proteins from polyacrylamide to nitrocellulose. J. Biochem. Biophys. Meth. 10, 203-209 La Claire, J.W., II. (1982) Cytomorphological aspects of wound healing in selected Siphonocladales (Chlorophyceae). J. Phycol. 18, 379-382 La Claire, J.W., II. (1984) Cell motility during wound healing in giant algal cells: contraction in detergent-permeabilized cell models ofErnodesmis. Eur. J. Cell Biol. 33, 180-189 Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685 Lloyd, C.W. (1984) Toward a dynamic helical model for the influence of microtubules on wall patterns in plants. Int. Rev. Cytol. 86, 1-51 Lloyd, C.W., Bartow, P.W. (1982) The co-ordination of celt division and elongation: the role of the cytoskeleton. In: The cytoskeleton in plant growth and development, pp. 203-228, Lloyd, C.W., ed. Academic Press, London Lloyd, C.W., Clayton, L., Dawson, P.J., Doonan, J.H., Hulme, J.S., Roberts, I.N., Wells, B. (1985) The cytoskeleton underlying side walls and cross walls in plants: molecules and macromolecular assemblies. J. Cell Sci. Suppl. 2, 143-155 Lloyd, C.W., Seagull, R.W. (1985) A new spring for plant cell biology: microtubules as dynamic helices. Trends Biochem. Sci. 10, 476-478 Lloyd, C.W., Slabas, A.R., Powell, A.J., Lowe, S.B. (1980) Microtubules, protoplasts and plant cell shape. An immunofluorescent study. Planta 147, 500-506 Lloyd, C.W., Wells, B. (1985) Microtubules are at the tips of root hairs and form helical patterns corresponding to inner wall fibrils. J. Cell Sci. 75, 225-238 Mazia, D., Schatten, G., Sale, W. (1975) Adhesion of cells to surfaces coated with polylysine. Applications to electron microscopy. J. Cell Biol. 66, 198-200 Menzel, D. (1986) Visualization of cytoskeletal changes through the life cycle in Acetabularia. Protoplasma 134, 30-42 Menzel, D., Schliwa, M. (1986a) Motility in the siphonous green alga Bryopsis. I. Spatial organization of the cytoskeleton and organelle movements. Eur. J. Cell Biol. 40, 275-285 Menzel, D., Schliwa, M. (1986b) Motility in the siphonous green alga Bryopsis. II. Chloroplast movement requires organized arrays of both microtubules and actin filaments. Eur. J. Cell Biol. 40, 286-295 Mizuta, S., Wada, S. (1981) Microfibrillar structure of growing cell wall in a coenocytic green alga, Boergesenia forbesii. Bot. Mag. Tokyo 94, 343-353
42 Morejohn, L.C., Fosket, D.E. (1984)Inhibition of plant microtubule polymerization in vitro by the phosphoric amide herbicide amiprophos-methyl. Science 224, 874-876 Newcomb, E.H., Bonnett, H.T., Jr. (1965) Cytoplasmic microtubule and wall microfibril orientation in root hairs of radish. J. Cell Biol. 27, 575 589 Osborn, M., Webster, R.E., Weber, K. (1978) Individual microtubules viewed by immunofluorescenceand electron microscopy in the same PtK2 cell. J. Cell Biol. 77, R27-R34 Schiff, P.B., Fant, J., Horwitz, S.B. (1979) Promotion of microtubule assembly in vitro by taxol. Nature 277, 665-667 Staves, M.P., La Claire, J.W., II. (1985) Nuclear synchrony in Valonia macrophysa (Chlorophyta) : light microscopy and flow cytometry. J. Phycol. 21, 68-71 Sternberger, L.A. (1986) Immunocytochemistry, 3rd edn. John Wiley & Sons, New York Towbin, H., Staehelin, T., Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76, 43504354 Traas, J.A., Braat, P., Emons, A.M.C., Meekes, H., Derksen, J. (1985) Microtubules in root hairs. J. Cell Sci. 76, 303-320 Van Lammeren, A.A.M., Keijzer, C.J., Willemse, M.T.M., Kieft, H. (1985) Structure and function of the microtubular
La Claire, J.W., II: Microtubules in coenocytic green algae cytoskeleton during pollen development in Gasteria verrucosa (Mill.) H. Duval. Planta 165, 1-11 Wang, K., Feramisco, J.R., Ash, J.F. (1982) Fluorescent localization of contractile proteins in tissue culture cells. Meth. Enzymol. 85, 514-562 Wick, S.M. (1985) Immunofluorescence microscopy of tubulin and microtubule arrays in plant cells. III. Transition between mitotic/cytokinetic and interphase microtubule arrays. Cell Biol. Int. Rep. 9, 357-371 Wick, S.M., Duniec, J. (1983) Immunofluorescence microscopy of tubulin and microtubule arrays in plant cells. I. Preprophase band development and concomitant appearance of nuclear envelope-associated tubulin. J. Cell Biol. 97, 235-243 Wick, S.M., Duniec, J. (1984) Immunofluorescence microscopy of tubulin and microtubule arrays in plant cells. II. Transition between the pre-prophase band and the mitotic spindle. Protoplasma 122, 45-55 Woodcock, C.L. (1971) The anchoring of nuclei by cytoplasmic microtubules in Acetabularia. J. Cell Sci. 8, 611-621
Received 6 October; accepted 11 December 1986
