Electron-microscopic and immunochemical analysis of kinetochore microtubules after ultraviolet microbeam irradiation of kinetochores

JULIA A. M. SWEDAK, CYNTHIA LEGGIADRO and ARTHUR FORER

Biology Department, York University, North York, Ontario, Canada M3J 1P3

Summary

We used an ultraviolet microbeam to irradiate kinetochore microtubules are in smaller numbers in kinetochores of chromosomes in crane-fly spermato- the irradiated half-spindle than in the non-irradiated cytes. We used one of two doses, low (0.106 erg fim~2) half-spindle or in non-irradiated cells. Since ir- or high (0.301 erg /«n~2), and then studied the micro- radiation with low doses alters interchromosomal tubules in those spindles using electron microscopy 'signals', but microtubules remain attached to the or immunofluorescence microscopy. After irrad- kinetochore, we argue that low doses of ultraviolet iation with low doses microtubules are present as light damage a signal-related function of kineto- usual, with normal fluorescence and in normal chores without altering the ability of the kineto- numbers. After irradiation with high doses micro- chores to bind microtubules. tubules are no longer associated with the irradiated kinetochore. After irradiation with either dose, non- Key words: kinetochores, ultraviolet microbeam, spindles.

Introduction Louis, MO). The halocarbon oil was replaced with insect Ringer's solution prior to irradiation (Czaban and Forer, 1985). Spermato- After kinetochores in anaphase cells were irradiated with cytes were placed on a 0.35 mm thick quartz coverslip (ESCO low doses of ultraviolet light (UV), using an UV Products Inc., Oak Ridge, NJ), and after irradiation cells were processed for immunofluorescence or electron microscopy as microbeam, there was no loss in birefringence of the described below. spindle fibre associated with the irradiated kinetochore, but all six half-bivalents temporarily stopped moving Microbeam irradiations (Swedak and Forer, 1991). After kinetochores were Ultraviolet microbeam irradiation was as described previously irradiated with high doses of UV the birefringence of the (Swedak and Forer, 1991). Briefly, all irradiations were with a chromosomal spindle fibre disappeared, and all six half- wavelength of 280 nm, with a half-band-width of 3.6 nm. The bivalents permanently stopped moving. We argued that a irradiating spot on the specimen was 1.2 /un in diameter. The 'signalling' function of kinetochores is damaged by the low total energy of UV incident on the specimen, controlled by doses of ultraviolet light, without loss of ability to bind varying the irradiation time (between 10 and 90s), was microtubules. In making this argument we assumed that 0.106erg^m~2±0.05erg;nn~2 (standard deviation) in low-dose 2 2 the presence of spindle fibre birefringence indicates the irradiations (rc=21) and 0.301 erg/Animal preparations (St. Louis, MO), DMSO from Anachemia (Mississauga, Ontario, Spermatocytes from fourth instar larvae were obtained from Canada) and NP40 from BDH Chemicals (Toronto, Ontario).) The Nephrotoma suturalis (Loew) or Nephrotoma abbreviata (Loew), lysis and fixation was done in the well formed by coverslip and using rearing methods and preparation of spermatocytes that slide. The coverslip was then removed from the slide and have been described by Forer (1982). Briefly, spermatocyte smears immersed in phosphate-buffered saline (PBS: 0.12 M NaCl, 6mM under halocarbon oil (series 27) were held in place by a fibrin clot phosphate buffer, pH 6.9) in a Petri dish for 2 min. If the cell was (Fibrinogen: Calbiochem, CA/ Thrombin: Sigma Chem. Co., St fixed with glutaraldehyde the coverslip was removed from the Journal of Cell Science 100, 269-277 (1991) Printed in Great Britain © The Company of Biologists Limited 1991 269 x slide and placed in sodium borohydride (lmgml ) for 20min Results before immersing the coverslip in PBS for 2min. After a quick rinse with Triton X-100 (Sigma Chemical Co.) (0.1 % Triton X-100 We have shown previously (Swedak and Forer, 1991) that in PBS) cells were then processed for indirect immunofluor- birefringent fibres remain after low-dose irradiation of escence with rat monoclonal antibody against tubulin (YL 1/2; Cedarlane, Hornby, Ontario) diluted 1:4000 in PBS; the antibody kinetochores, that spindle fibre birefringence disappears was added for 20min, and after two 5-min rinses in PBS, after high-dose irradiation of kinetochores, and that flourescein-conjugated anti-rat antibodies (dilution, 1:10; Ceder- interchromosomal 'signals' are affected in both cases. lane Labs Ltd., Hornby, Ontario) were added for 20min. The coverslip was then placed in a mixture of mowiol (Polysciences, Analysis using immunofluorescence Inc.) and 0.02 g per 100 ml p-phenylenediamine (PPD; Sigma Chemical Co., St Louis, MO) and the slides were stored at 4°C in We determined whether immunofluorescent staining (of the dark until studied by fluorescence microscopy. tubulin) was altered by irradiating kinetochores of Irradiated cells were relocated in the following way. Prior to autosomes in either metaphase or early anaphase. Cells making preparations, a grid was scratched on the coverslip with a were fixed either immediately following irradiation diamond scribe, and markings were placed on the grid. During the (n=10, low dose; n=ll, high dose) or 5min after lOmin lysis period the cell was located relative to the fixed irradiation (?i=9, low dose; re=ll, high dose). By 'immedi- markings and a map of the area was drawn. After staining, cells ately' we mean less than 2min after irradiation, as were located with respect to the map. determined by analysis of videotapes taken during the Stained cells were viewed with a Nikon inverted Diaphot-TMD course of fixation. The results for cells fixed immediately microscope equipped with epifluorescence (Nikon, 100 x phase- contrast fluor-DL objective lens, oil immersion, NA=1.3). were the same as for cells fixed 5 min after irradiation. Immunofluorescent images from a RCA SIT camera were recorded In cells in which the kinetochore had been irradiated on videotape using an Electrohome HO2 videocasette recorder. with a low dose, the spindle fibre associated with the Photographs were taken as described above. irradiated kinetochore stained with what appeared to be normal intensity (Fig. 1). Thus, these irradiations did not cause the loss of microtubules, though from these data we Electron microscopy cannot rule out that some kinetochore microtubules were Cells to be studied using electron microscopy (EM) were fixed at lost. various times after irradiation using 2 % agar-treated glutaralde- hyde (Nicklas et al. 1982) in the insect Ringer's solution described In cells in which the kinetochore had been irradiated by Czaban and Forer (1985). Coverslips were prepared for electron with a high dose the spindle fibre associated with the microscopy and serially sectioned as described by Hughes et al. irradiated kinetochore did not stain with tubulin anti- (1988). (The irradiated cells were located on coverslips as bodies (Fig. 2), though spindle fibres associated with non- described above.) Sections were observed with a Jeol CX100 irradiated kinetochores stained with normal intensity. electron microscope operated at 120 kV, and micrographs were Thus, irradiation of kinetochores with high doses causes taken on 8 cm x 10 cm film. Electron micrograph serial reconstruc- loss of all microtubules in the associated spindle fibre. tion, as described by Wilson and Forer (1988), involved tracing the chromosomes and spindle microtubules onto acetate sheets from micrographs of each section and stacking the acetates on top of Electron microscopic analysis each other to obtain a representation of the spindle. Irradiation with low doses. We studied two cells

Fig. 1. Low-dose irradiation of a kinetochore in N. suturalis spermatocyte. (A) Cell viewed with polarizing optics prior to irradiation. Site of irradiation is indicated by an arrow. (B) During the irradiation. (C) Following the irradiation. x800. (D) Cell in C viewed with fluorescence optics to reveal anti-tubulin staining. The irradiated area appears normal. X750.

Fig. 2. High-dose irradiation of a kinetochore in a N. suturalis spermatocyte. (A and B) Polarization microscopy. (A) Site of irradiation is indicated by an arrow. (B) Following the irradiation. X200. (C) Cell in B viewed with fluorescence optics to reveal anti-tubulin staining. There is little or no staining of the chromosomal spindle fibres for those 2 chromosomal kinetochores that were irradiated. (In this cell, 2 chromosome kinetochores were irradiated, because they were on top of each other during UV irradiation). X400.

270 J.A.M. Swedak et al. electron microscopically, both of which were irradiated with low doses and then fixed immediately. (We estimate from the videotape that fixation occurred less than 3 min after the end of the irradiation.) A representative section of cell 1 is illustrated in Fig. 3: qualitatively, the spindle appears normal except that the non-kinetochore micro- tubules in the irradiated half-spindle have decreased in comparison with micrographs of those from non-irradiated crane-fly spermatocytes (Behnke and Forer, 1966; Fuge, 1971, 1984; LaFountain, 1976; Forer and Brinkley, 1977; Wilson and Forer, 1988). Quantitative estimates of kinetochore microtubules in this spindle were obtained from the spindle as reconstructed from 16 serial sections (Fig. 4): 41 microtubules were inserted into the kineto- chore and extended poleward. (We did not track them for more than a few micrometres towards the pole.) Other non-irradiated kinetochores of autosomal half-bivalents in this cell had from 34-79 microtubules directly inserting into the kinetochore and extending polewards (Fig. 5). The second cell (cell 2) was one in which one kinetochore of a sex chromosome had been irradiated with a low dose; the results were similar in all respects to those for the first i cell: the number of microtubules inserting into the kinetochore and extending polewards was 51, while the non-irradiated kinetochores of the sex chromosomes in this cell had from 45 to 70 microtubules directly inserting into the kinetochore and extending polewards (Table 1). Therefore, low doses of UV seem to have little or no effect on the microtubules associated directly with the kineto- chore. Irradiation with high doses. Two cells were irradiated and fixed immediately (within 3 min) after the end of the irradiation. A representative section of one cell (cell 3) is illustrated in Fig. 6. The cell appears normal in structure except that the area associated with the irradiated kinetochore is devoid of microtubules, and there is an apparent reduction in non-kinetochore microtubules in the irradiated half-spindle. Quantitative estimates of kinetochore microtubules in this spindle were obtained by reconstruction from 22 serial sections (Fig. 7). Only four microtubules were found that were inserted into the kinetochore, but these were short pieces (length <1 fan) Fig. 3. Representative electron micrograph of low-dose and none extended polewards (Fig. 8). Other non-ir- irradiation of a kinetochore in an N. abbreviata spermatocyte. radiated kinetochores in this cell had from 36 to 61 Cell 1. the irradiated kinetochore is indicated by an arrow. microtubules that were directly inserted into the kineto- Microtubules in normal numbers are associated with the chore and extended poleward (Table 1). irradiated kinetochore. Bar, 1 /im.

Table 1. Summary of the number of kinetochore microtubules associated with a kinetochore, determined by electron microscopic analysis Total no. Kinetochore kinetochore Chromosome irradiated Dose microtubules Species Cell 1 (anaphase) Autosome Yes Low 41 N. suturalis Autosome No 34-79 N. suturalis (5 autosomes) Cell 2 (anaphase) Sex-chromosome Yes Low 51 N. abbreviata Sex-chromosome No 45-70 N. abbreviata (3 kinetochores) Cell 3 (anaphase) Autosome Yes High 4 (did not N. suturalis extend polewards) Autosome ft) 36-61 N. suturalis (5 autosomes) Cell 4 (anaphase) Autosome Yes High 10 (did not N. suturalis extend polewards) Autosome No 40-63 N. suturalis (5 autosomes)

Microtubules after irradiation of kinetochores 271 j

Fig. 4. Reconstruction of the low-dose kinetochore irradiation of cell 1, also illustrated in Fig. 3. (A-C) The cell that was irradiated. (A) Prior to irradiation. The irradiation site is indicated by an arrow. (B) During irradiation. (C) Following irradiation. (D) 'Pseudo phase-contrast' picture of the cell. X400. (E-G) Reconstruction of the irradiated cell. The irradiated kinetochore is indicated by an arrow. (E) The whole spindle. (F) The non-irradiated half-spindle. (G) The irradiated half-spindle. There is still a bundle of microtubules associated with the irradiated kinetochore. The number of non-kinetochore microtubules has decreased in the irradiated half-spindle. The reconstruction is based on tracings from 16 sections through the central spindle area.

In a second cell (cell 4), 10 microtubules were found non-irradiated kinetochores of autosomal half-bivalents in inserted into the irradiated kinetochore, and these, too, this second cell had from 40 to 63 kinetochore micro- were short pieces that did not extend polewards. Other tubules that were directly inserted into the kinetochore

272 J. A. M. Swedak et al. Fig. 5. Reconstructions of chromosomes from cell 1 in which one kinetochore of one half-bivalent was irradiated with a low dose. (A) The irradiated half-bivalent; 41 kinetochore microtubules. (B) A half-bivalent in the non-irradiated half spindle; 79 kinetochore microtubules. (C) A half-bivalent in the non- irradiated half-spindle; 34 kinetochore microtubules. (D) A half- bivalent in the irradiated half spindle; 63 kinetochore microtubules. (E) A half-bivalent in the irradiated half-spindle; 26 kinetochore microtubules. The reconstruction is based on tracings from 16 sections through the central spindle area.

B

and extended polewards. Therefore, though there are a few Discussion microtubules associated with the irradiated kinetochore, none of these microtubules extends poleward; thus, We have analysed the effects of high- and low-dose UV on irradiation of kinetochores with high doses causes loss of kinetochores using immunochemical and ultrastructural the associated kinetochore microtubules. techniques.

Microtubules after irradiation of kinetochores 273 Fig. 6. (B) Representative electron micrograph of a high-dose irradiation of a kinetochore in N. suturalis spermatocyte. Cell 3: the irradiated kinetochore is indicated by an arrow. Microtubules appear in normal numbers in the spindle except in the area of the irradiated kinetochore. Bar, 1 /an. (A) Magnification of the irradiated area. Bar, 1 fim. (C) Magnification of the non-irradiated half- spindle. Bar, 1 /an.

Our immunofluorescence and EM data show that high- For comparison, it is relevant to note that the number of dose irradiation of the kinetochore causes loss of the microtubules associated with kinetochores irradiated with microtubules associated with the irradiated kinetochore. low doses or associated with non-irradiated kinetochores On the other hand low-dose irradiation has no effect on the appears normal when compared with the number of microtubules associated with the irradiated kinetochore. microtubules inserting into kinetochores in non-ir-

274 J. A. M. Swedak et al. \

u Fig. 7. Reconstruction of the high-dose irradiation. Cell 3. (A) Prior to irradiation. The irradiation site is indicated by an arrow. (B) Following the irradiation. (C) Reconstruction of the irradiated cell. The irradiated kinetochore is indicated by an arrow. Microtubules appear in normal numbers except in the area of the irradiated kinetochore. The reconstruction is based on tracings from 22 sections through the central spindle. radiated crane-fly spermatocytes that are cited in the gence as a measure of microtubule numbers. Because low- literature (Tables 1 and 2). dose irradiations alter 'signals' (Swedak and Forer, 1987, The absence of immunofluorescence staining of micro- 1991) but microtubules remain associated with the tubules or of EM-visible microtubules after high doses kinetochore in almost normal numbers, we argue that the does not seem to be a fixation artefact, because: (1) we used kinetochore function in the 'signal' mechanism is altered different fixation protocols in the two cases; and (2) the without loss of ability to bind microtubules. But are all results match predictions from living cells using birefrin- microtubules present following low-dose irradiation of

Microtubules after irradiation of kinetochores 275 Fig. 8. Reconstructions of chromosomes from cell 3 in which one kinetochore of one half-bivalent was irradiated with a high dose. (A) The irradiated half-bivalent; 4 kinetochore microtubules. (B) A half-bivalent in the non-irradiated half-spindle; 61 kinetochore microtubules. (C) A half-bivalent in the non- irradiated half-spindle; 36 kinetochore microtubules; (D) A half- bivalent in the irradiated half-spindle; 43 kinetochore microtubules. (E) A half-bivalent in the irradiated half-spindle; 51 kinetochore microtubules. The reconstruction is based on tracings from 22 sections through the central spindle area.

kinetochores? The number of microtubules terminating at The putative signal is not likely to be the loss of non- the irradiated kinetochore is in the same range as the kinetochore microtubules in the irradiated half-spindle number of microtubules terminating at non-irradiated because non-kinetochore microtubules are also lost when kinetochores in the same cells (Table 1) and cited in the we irradiate spindle fibres (see Wilson and Forer, 1989), literature (Table 2), but not at the top end of the range. yet these irradiations do not alter signals. The putative Thus, while we cannot unequivocally rule out the signal could conceivably be related to the loss of other possibility that some microtubules are lost - perhaps up to microtubules in the kinetochore microtubule bundle: those 30 % of those present - it seems likely that they are all that do not end at the kinetochore. We do not know if such present. If the signal is not a loss of microtubules that microtubules are lost, but even if they were it is difficult insert into the kinetochore, what could it be? for us to imagine how loss of these microtubules would

276 J. A. M. Swedak et al. Table 2. Kinetochore microtubules in crane fly spermatocytes: summary of published data Total no. of Species and kinetochore cell stage Chromosome microtubules Reference N. suturalis Metaphase Autosomes 30-40 LaFountain (1974) Metaphase Autosomes and 68-93 LaFountain (1976) sex chromosomes Early anaphase Autosomes 70-87 LaFountain (1976) Anaphase Autosomes 20-30 Forer and Brinkley (1977) Anaphase Sex chromosomes 16-24 Forer and Brinkley (1977) N. ferruginea Early anaphase Autosomes 40-50 Fuge (1973, 1974) Anaphase Sex chromosomes 16-60 Fuge (1985) N. abbreviata Anaphase Sex chromosomes 40-70 Swedak and Forer (unpub signal other motors to stop the movements of the other FUGE, H. (1971). Spindelbau, Mikrotubuli-verteilung und Chromosomen struktur wahrend der I. meiotischen Teilung der Spermatocyten von chromosomes. It could be that the signal is sent from the Pales ferriginea. Z. Zellforsch. mikrosc. Anat. 120, 579-599. kinetochore itself. Or it could be that disruption of a FUGE, H. (1973). Microtubule distribution in metaphase and anaphase 'corona' component extending along kinetochore micro- spindles of the spermatocytes of Pales ferruginea. Chromosoma 43, tubules (Reider et al. 1990) alters a signal, or that the 109-143. microtubules act like 'computers' (Hameroff et al. 1986; FUGE, H. (1974). The arrangement of microtubules and the attachment Koruga, 1986) and thereby transmit signals. Whatever the of chromosomes to the spindle during anaphase in tipulid spermatocytes. Chromosoma 45, 245-260. signal is, we know that it acts quickly, because all six half- FUGE, H. (1984). The three dimensional architecture of chromosome bivalents stop moving within seconds; thus if the signal is fibres in the crane fly.I . Syntelic autosomes in meiotic metaphase and flow of material we can rule out any kind of flow that anaphase I. Chromosoma 90, 323—331. cannot traverse tens of micrometres in seconds. FUGE, H. (1985). The three dimensional architecture of chromosome fibres in the crane fly.II . Amphitelic sex univalents in meiotic To test whether microtubules are involved in the anaphase I. Chromosoma 91, 322-328. transmission of the signal we plan to irradiate the spindle HAMEROFF, S. R., SMITH, S. A. AND WATT, R. C. (1986). Automaton fibre of an autosomal bivalent to produce an area of model of dynamic organization in microtubules. In Dynamic Aspects of reduced birefringence on the fibre. Following this ir- Microtubule Biology (ed. D. Soifer), pp. 949-952. The New York radiation we will then irradiate the kinetochore of the Academy of Sciences, New York. HUGHES, K. D., FORER, A., WILSON, P. J. AND LEGGIADRO, C. (1988). same autosome. This double irradiation will test whether Ultraviolet microbeam irradiation of microtubules in vitro: the action the signal requires microtubules that are continuous from spectrum for local depolymerization of marginal band microtubules in kinetochore to pole. vitro matches that for reducing birefringence of chromosomal spindle In summary, a large number of kinetochore micro- fibres in vivo. J. Cell Sci. 91, 469-478. tubules are still associated with the kinetochore following KORUGA, D. L. (1986). Microtubular screw symmetry packing of spheres as a latent bioinformation code. In Dynamic Aspects of Microtubule ultraviolet microbeam irradiation at a low dose. Few Biology (ed. D. Soifer), pp. 953-955. The New York Academy of microtubules, none of which extend polewards, are Sciences, New York. associated with the kinetochore following ultraviolet LAFOUNTAIN, J. R. JR (1974). Birefringence and fine structure of microbeam irradiation at a high dose. Since low-dose spindles in spermatocytes of Nephrotoma suturalis at metaphase of irradiation nonetheless blocks motion of the irradiated first meiotic division. J. Ultrastruct. Res. 46, 268-278. LAFOUNTAIN, J. R. JR (1976). Analysis of birefringence and chromosomes and other chromosomes as well, irradiation ultrastructure of spindles in primary spermatocytes of Nephrotoma with low dose damages a signalling function of the suturalis during anaphase. J. Ultrastruct. Res. 54, 333-346. kinetochore without disrupting microtubule attachment NICKLAS, R. B., KUBAI, D. F. AND HAYS, T. S. (1982). Spindle to the kinetochore. The exact nature of the putative signal microtubules and their mechanical associations after remains obscure. micromanipulation in anaphase. J. Cell Biol. 95, 91-104. REIDER, C. L., ALEXANDER, S. P. AND RUPP, G. (1990). Kinetochores are transported poleward along a single astral microtubule during We thank Dr John Stevens for generous use of his serial section chromosome attachment to the spindle in newt lung cells. J. Cell Biol. facilities. This work was supported by the Natural Sciences and 110, 81-95. Engineering Research Council of Canada and, for much of the SWEDAK, J. A. M. AND FORER, A. (1987). Sex-chromosome anaphase electron microscopy, by grants to Dr John Stevens. movements in crane-fly spermatocytes are co-ordinated: ultraviolet microbeam irradiation of one kinetochore of one sex chromosome blocks the movements of both sex chromosomes. J. Cell Sci. 88, References 441-452. BEHNKE, O. AND FORER, A. (1966). Some aspects of microtubules in SWEDAK, J. A. M. AND FORER, A. (1991). Kinetochore function can be spermatocyte meiosis in a crane fly {Nephrotoma suturalis Loew): altered by UV microbeam irradiation without loss of the associated intranuclear and intrachromosomal microtubules. C.r. Trav. Lab. birefringent spindle fibre. J. Cell Sci. 100, 000-000. Carlsberg 35, 437-455. WILSON, P. J. AND FORER, A. (1988). Ultraviolet microbeam irradiation CZABAN, B. B. AND FORER, A. (1985). The kinetic polarities of spindle of chromosomal spindle fibres shears microtubules and permits study microtubules in vivo, in crane-fly spermatocytes. I. Kinetochore of new free ends in vivo. J. Cell Sci. 91, 455-468. microtubules that re-form after treatment with colcemid. J. Cell Sci. WILSON, P. J. AND FORER, (1989). Acetylated ff-tubulin in spermatogenic 79, 1-37. cells of the crane fly Nephrotoma suturalis: kinetochore microtubules FORER, A. (1982). Crane fly spermatocytes and spermatids: a system for are selectively acetylated. Cell Motil Cytoskel. 14, 237-250. studying cytoskeletal components. Meth. Cell Biol. 25, 227-252. FORER, A. AND BRINKLEY, B. R. (1977). Microtubule distribution in the anaphase spindle of primary spermatocytes of a crane fly (Nephrotoma suturalis). Can. J. Genet. 19, 503-519. (Received 4 March 1991 - Accepted, in revised form, 9 July, 1991)

Microtubules after irradiation of kinetochores 111