Proc. Nati Acad. Sci. USAVol. 79, pp. 1898-1902, March 1982Cell
Biology
Visualization of microtubules in interphase and mitotic plant
cellsof Haemanthus endosperm with the immuno-gold staining
method
(plant immuno-gold stains/immunocytochemistry)
J. DE MEY*, A. M. LAMBERTt, A. S. BAJERt, M. MOEREMANS*, AND M.
DE BRABANDER**Department of Life Sciences, Laboratory of Oncology,
Janssen Pharmaceutica Research Laboratories, B-2340 Beerse,
Belgium; tInstitut de Botanique, Universit6Louis Pasteur, 28 rue
Goethe, 67083 Strasbourg-Cedex, France; and tDepartment of Biology,
University of Oregon, Eugene, Oregon 97403
Communicated by Daniel Mazia, December 10, 1981
ABSTRACT A procedure is presented for the immunocyto-chemical
visualization of microtubules in interphase and mitoticcells of
Haemanthu8 endosperm. It includes preservation of mi-crotubules
(MTs) with glutaraldehyde and uses colloidal gold,coated with
secondary antibodies, in a novel indirect-light micro-scopic
technique: the immuno-gold staining method. This
immu-nocytochemical stain allows us to follow the changes in
distributionof MTs during mitosis with greater precision and
specificity thanallowed by other light microscopic techniques. Many
aspects ofMT arrangements, as reported from ultrastructural
studies, arecorroborated and extended. This demonstrates the
reliability ofthe technique. In addition, a number of significant
observationswere made. These concern (i) the presence of a network
of MTsin interphase cells, (ii) the transformation of this network
into aspindle-like cage of MTs (the clear zone) surrounding the
nucleusduring prophase, (iii) the drastic rearrangement of MT
distribu-tion during prometaphase, (iv) new evidence for the
formation ofaster-like arrays. of polar MTs during anaphase, and
(v) the de-velopment of the phragmoplast.
Plant endosperm cells have been useful for studies on mitosisand
cytokinesis. Much attention has been given to changes inthe
arrangement of microtubules (MTs) during spindle forma-tion,
kinetochore orientation and metaphase alignment, ana-phase
movement, and phragmoplast and cell plate formation(1-4). This
material is also of great potential value for the ex-perimental
study of the organized assembly/disassembly ofMTs in living cells.
However, there is a need for a specific andreliable way of
visualizing the MT distribution. Existing lightmicroscopic methods
either do not permit studies of MT ar-rangements (phase contrast)
or are limited by the inherent prop-erties of polarized light (the
Nomarski system and polarizingmicroscope). Because of the large
number of cells in one prep-aration at different stages of
division, Haemanthus endospermlends itself to an immunocytochemical
approach in which MTsare visualized with tubulin antibodies. In
studies on plant MTs(5-7), including one on endosperm (8), with
techniques effec-tive with animal cells (9, 10), it has been shown
that antibodiesagainst mammalian tubulin crossreact with plant
tubulin. Re-cently (7), brief incubation in cell wall-digesting
enzymes ofmeristematic, wall-enclosed root cells, fixed in
paraformalde-hyde, has allowed retention of both antigenicity and
three-di-mensional properties in interphase cells, mitotic spindle,
andphragmoplast. Thus, it appears that plant cells have much
po-tential for immunocytochemistry.We present here a procedure for
the immunocytochemical
visualization of MTs in interphase and mitotic endosperm
cells
ofHaemanthus. It includes glutaraldehyde fixation of MTs anduses
colloidal gold, coated with secondary antibodies, in a
noveltechnique: the immuno-gold staining (IGS) method (11, 12).This
method is combined with toluidine blue staining of thechromosomes
(13). Preliminary results corroborate and extendmany aspects of MT
arrangements as reported in studies withthe electron microscope and
light microscope. In addition, anumber of significant observations
concerning interphase cells,development of the spindle, formation
of aster-like arrays ofMTs during anaphase-telophase, and formation
of the phrag-moplast are reported.
MATERIALS AND METHODSFixation and Permeabilization. Endosperm
preparations of
Heemanthus katherinae Bak were made as described (14). Thecells
were fixed by the following procedure. The preparationswere treated
for 2-5 min at room temperature in 0.1-0.2%Triton X-100/0.5%
glutaraldehyde/0.1 M Pi buffer, pH 6.9,and for a further 10 min in
1% glutaraldehyde in Pi buffer. Afterwashing in 0.1 M Pi buffer,
the cells were permeabilized in 0.5%Triton X-100 in P1 buffer for
10 min. After brief rinsing in 0.1M P1 buffer, the preparations
were treated with freshly madeNaBH4 (0.5 mg/ml in 0.1 M P1 buffer)
for 10 min at room tem-perature. During this stage, the layer of
gelatine-agar coveringthe cells peeled off. After rinsing three
times in 10 mM Tris-buffered saline (pH 7.6), preparations were
processed forimmunocytochemistry.
Immunocytochemistry. The IGS method for the light micro-scopic
visualization ofMTs is described in detail elsewhere (11,12).
Briefly, the cells were stained as follows: (i) incubation with5%
normal goat serum in Tris-buffered saline, 20 min; (ii) in-cubation
with affinity-purified rabbit antibody (5 ,ug/ml) tohighly purified
dog brain tubulin in 1% normal goat serum (inTris-buffered saline),
overnight at room temperature; (iii) wash-ing in 0.1% bovine serum
albumin buffer (20 mM Tris-bufferedsaline, pH 8.2/0.1% bovine serum
albumin), three times for 10min each; (iv) GAR G20 [colloidal gold
(18- to 20-nm diameter= G-20) coated with affinity-purified goat
antibodies to rabbitimmunoglobulin (designated GAR)] in 1% bovine
serum albu-min buffer, 2 hr at room temperature; (v) three washings
in 0.1%bovine serum albumin buffer, 10 min each; (vi) fixation in
1%glutaraldehyde (in 0.1 M Pi buffer), 10 min; (vii) washing
indistilled water and staining with 0.01% toluidine blue in H20(pH
5.7), 2 min when desired; (viii) washing in distilled water,2 min;
and (ix) dehydration through a graded ethanol series andmounting.
Preparations were observed under bright-fieldtransmitted light
microscopy (Leitz, Orthoplan) with a X100
Abbreviations: MTs, microtubules; IGS, immuno-gold staining.
1898
The publication costs ofthis article were defrayed in part by
page chargepayment. This article must therefore be hereby marked
"advertise-ment" in accordance with 18 U. S. C. §1734 solely to
indicate this fact.
Dow
nloa
ded
by g
uest
on
June
24,
202
1
Proc. Natl. Acad. Sci. USA 79 (1982) 1899
Plan-Apo objective, NA 1.32. Micrographs were taken on
Agfa-chrome 50 S color transparencies. Black and white
picturespresented in this paper were duplicated from this
material.
Preparation of Colloidal Gold Coated with Secondary An-tibodies.
The procedure (to be published elsewhere) is basedon modifications
of the methods of Geoghegan et al. (15) andHorisberger et al. (16).
Briefly, goat antibodies to rabbit im-munoglobulin G were purified
from antiserum (GIBCO) by an-tigen-affinity chromatography.
Colloidal gold (mean diameter, 18-20 nm) was prepared withsodium
citrate as the reducing agent (17). The minimal amountof protein
(in 2 mM borax buffer, pH 9.0) needed for stabili-zation of the
gold sol at pH 9.0 was determined from a concen-tration-variable
isotherm (15, 16).The appropriate amount of protein (1 mg at 1
mg/ml) was
added to the gold sol (100 ml). After 2 min of stirring, 8%
bovineserum albumin (Sigma) fraction V in borax buffer (pH 9.0)
wasadded to a final concentration of 1%. Unadsorbed protein
waseliminated by three alternate centrifugations at 14,000 X g
for60 min at 40C. Resuspensions were in 1% bovine serum
albuminbuffer.The final pool was diluted 1:10 (vol/vol) with 1%
bovine
serum albumin buffer/20 mM NaN3 and stored at 40C. This iscalled
GAR G20.
RESULTS
The IGS method used with specific antibodies to tubulin
stainsMTs red. This color is produced by the specific
accumulationof granules of colloidal gold on the immunoreactive
sites of theMTs, as shown in animal cells (12). This yields
high-resolutionlight microscopic pictures. The use of toluidine
blue (13) per-mits one to follow in detail interrelationships
between chro-mosomes and MTs. In some cases (e.g., analysis of
kinetochoreMT attachment), it was advantageous to have very light
tolui-dine blue staining (Fig. 1). It should be stressed that the
blackand white micrographs included here do not give all the
infor-mation that can be obtained by careful focusing through
themicroscope or the additional contrast of different colors.
Thinsections of cells fixed according to the procedures used for
im-munostaining showed that MTs are well preserved (Fig. 2e).
Interphase. Interphase cells contain very elaborate anddense
arrays ofcytoplasmic MTs (Fig. la). They form a networkthat seems
to radiate from the periphery of the nucleus towardsthe cell
membrane. Even small cytoplasmic fragments withoutnuclei contain
MTs (see also ref. 18). Focal centers, close to thenuclear envelope
toward which MTs converge, were occasion-ally found.
Mitosis. At the onset of prophase (Fig. lb), arrays of
MTsaccumulate along the edges of the disc-shaped nucleus and toa
lesser extent on its upper and lower surface. They form theclear
zone (1) that is visible with phase-contrast or Nomarskioptics. The
IGS staining allowed us to follow interrelationshipsbetween clear
zone organization and the interphase cytoplasmicMT network. The
overall density of cytoplasmic MTs dimin-ishes during the progress
of prophase, concurrent with gradualchanges in their distribution;
many of them intermingle withthe MTs accumulating around the
nucleus. The cytoplasmicMTs show a progressively more radial
distribution and a ten-dency for increased bundling.
Later in prophase, more MTs appear around the nucleus,while the
number of cytoplasmic MTs decreases (Fig. lc). Inareas where many
MTs converge, poorly defined pole-likestructures (sometimes three
to five) are formed, from whichMTs radiate into the cytoplasm (Fig.
1c). At late prophase (Fig.id), two to three similar areas become
well-defined and are
usually located on opposite sides of the nucleus. By that timeof
prophase, cytoplasmic MTs have usually disappeared, andthe nucleus
is completely surrounded by a dense and uniformcage of MTs, the
majority of which are oriented parallel to theaxes between the
poles. These poles often have the form of dis-tinct tips due to MT
convergence.
After the breakdown of the nuclear envelope, MT distribu-tion
changes drastically (Fig. 1 e andf). Distinct bundles of MTsare
formed. A number of these bundles end at kinetochores(chromosomal
fibers) (Fig. le). Intermingling between differentbundles can be
observed and results in a complex spatial dis-tribution. Mantle MTs
are still visible at the surface ofthe spin-dle (Fig. if). Often,
two to three kinetochorefibers convergeto one region (subpole).
This subpole has a different aspect thanthe poles seen at late
prophase (Fig. 1d). MT ends splay andform a meshwork lacking any
clear orientation. In general, inprometaphase and metaphase, the
number and appearance ofpolar areas show variations depending upon
mechanical con-ditions within the cell (e. g., kinetochore
distribution and degreeof flattening) (Fig. lg). At the beginning
of anaphase, this polarorganization is retained. Chromosomal fibers
often remain dis-tinguishable. In the interzone, MTs are arranged
in discretebundles of varying thickness. These bundles are
separated byareas virtually devoid of MTs. Single interzonal fibers
often in-termingle with two sister kinetochore fibers in opposite
half-spindles, as suggested by previous electron microscopic
studies(19). As shown in Fig. 1 h-i, from midanaphase on, the
numberofMTs at the equatorial region ofthe interzone decreases.
Con-current with kinetochore fiber shortening, individual
fibersbecome less distinguishable because ofa sharp increase of
stain-ing density in the two half-spindles. This fibrillar staining
pen-etrates into the interzone between the chromosome arms
(Fig.li). Analysis ofnumerous cells suggest that this increased
stain-ing corresponds to new MTs originating from the polar
region.During later anaphase there is also a tendency for the
formationof more defined poles made up of many converging MTs
(Fig.li). The sets of MTs originating from the two poles are not
con-nected to each other. In cells in which the bundles
ofcontinuousMTs are strongly diminished, they are separated from
eachother by a zone almost depleted ofMTs in the equatorial
region(Fig. li). Later in anaphase, MTs clearly originating from
thepoles (polar fibers) radiate into the cytoplasm, forming an
ar-rangement resembling the aster in animal cells. This is well
seenin the cell illustrated in Fig. 2 a and c.
Phragmoplast formation overlaps with these later stages
ofanaphase. In Fig. 2, phragmoplast development can be fol-lowed.
Because ofvisualization ofthe complete MT distributionin the intact
cell, the present results show that the phragmoplastdevelops in a
more complex way than has been described pre-viously in electron
microscopic studies. It is composed of MTsgrowing from the polar
region, remnants of continuous MTs,and MTs which possibly arise at
the cell plate. During this pro-cess (Fig. 2 a and b), the whole
interzone becomes filled withfine bundles of closely packed MTs
that are arranged perpen-dicular to the forming cell plate. Some of
these MTs intersectthe cell plate. The sharp gap at the equator
seen in IGS is dueto the lack of positive staining of the cell
plate and adjacent re-gion (Fig. 2 a, b, and d).
At the end of mitosis, the chromosomes form telophasic nu-clei
surrounded by nuclear envelopes (Fig. 2b). Aster-like ar-rays
reorganize and the interphase cytoplasmic MT networkdevelops around
the nucleus.
DISCUSSIONThe present work introduces an immunocytochemical
proce-dure for light microscopic studies of chromosome and MT
ar-
Cell Biology: De Mey et al.
Dow
nloa
ded
by g
uest
on
June
24,
202
1
1902 Cell Biology: De Mey et al.
thin sections of freeze-substituted materials (21). The role of
thecytoplasmic network of MTs in endosperm is unknown andneeds
further investigation.The accumulation of MTs in the vicinity of
the nucleus is re-
sponsible for the light microscopic appearance ofthe clear
zone.This region is not a well-defined entity as suggested by
lightmicroscopic analysis (1, 22). The present data support the
notionthat the nuclear envelope or the nucleus, or both, is
instru-mental in MT nucleation and arrangement (23). They do
notallow us, however, to speculate about the mechanisms in-volved.
Further studies utilizing IGS may elucidate some ofthese
problems.The question can be raised as to what extent the display
of
MTs seen in late prophase in endosperm cells is
functionallyequivalent to the central spindle of some lower
organisms suchas diatoms (24). The prophase spindle (clear zone) in
endospermis formed outside the nuclear envelope and becomes a
highlyorganized structure that undergoes profound MT
rearrange-ments during early prometaphase.
As documented in the results, arrays of MTs appear in
theinterzone and half-spindle after midanaphase. These arrays
fo-cus toward both polar regions where they might originate
(seealso ref. 18). These conclusions are reached because of the
newpossibilities of the immunocytochemical approach. It is
welldocumented (29, 30) that the progress of anaphase is related
toMT disassembly. Present data show that simultaneous
assemblyoccurs during anaphase. They are compatible with the
hypoth-esis, previously deduced from experiments in vivo with MT
in-hibitors (28), that this simultaneous assembly is required for
thepoleward movement of chromosomes.
At later stages ofanaphase and in telophase, MTs radiate intothe
cytoplasm, forming aster-like arrays of polar fibers. Thiscould be
interpreted as showing that in plant endosperm cells,from
midanaphase on, the polar region is functionally equiva-lent to the
centrosphere of astral spindles with one major dif-ference: the
lack of centrioles. Similar astral rays composed ofMTs in anaphase
and early telophase have been reported in roottip cells (25) and
especially in the first division ofthe endospermnucleus of Crepis
capillaris (26, 27). Ostergren pointed out thatthese "asters" grow
in size during anaphase. They reach max-imum development in
telophase. These data drew attention toand questioned the anastral
nature of the higher plant spindle.In a more recent study, similar
astral rays in root tip cells wereidentified as MTs (25). Our
findings on Haemanthus endospermshed new light on this intriguing
problem. The functional im-plications of this structure are not
clear at the moment.
As pointed out above, the aster-like arrays ofpolar MTs
prob-ably represent newly formed MTs originating in the polar
re-gion. However, we cannot exclude without additional experi-ments
the possibility that these MTs form, at least in part,through
elongation ofMT fragments generated by the breakageof preexisting
MTs or the release of tubulin dimers from ki-netochore fibers.
Additional discussion ofthis point is presentedelsewhere (18).
In some cells with delayed phragmoplast formation, the
in-terzonal fibers are often diminshed to such an extent that
thereis no visible connection between the two sister
half-spindles.This is in agreement with electron microscopic
results afterhexylene glycol treatment (31) and is compatible with
data fromanimal cells demonstrating that each half-spindle can
migrateas an autonomous unit (32, 33).
Usually, phragmoplast formation overlaps with the laterstages of
anaphase. Phragmoplast development may be morecomplex than has been
deduced from previous studies (1, 4).It seems that, in the
phragmoplast, there is intermingling ofMTs of different origin:
anaphase polar MTs originating in the
two opposite half-spindles, remnants of continuous MTs,
andadditional MTs possibly nucleated at the equatorial region ina
later stage. Recent data on MT polarity in Haemanthus (34)might
help in understanding this complex structure.
In conclusion, the results of this work illustrate the
potentialof this novel immunocytochemical approach for studies on
chro-mosome and MT arrangements in mitosis and phragmoplast
for-mation in plant endosperm. Further experimental work utiliz-ing
this technique could give new insights into the mechanismsof
spindle formation and function.
This research was supported by grants from the Instituut ter
bevor-dering van het Wetenschappelijk Onderzoek in Nijverheid en
Land-bouw (I.W.O.N.L.), Brussels, and by National Institutes of
HealthGrant GM 26121 (to A.S.B.).
1. Bajer, A. & Mole-Bajer, J. (1972) Int. Rev. Cytol Suppl
3, 1-271.2. Bajer, A. (1968) Chromosome 24, 83-417.3. Hepler, P.
& Jackson, W. (1968)J. Cell Biol 38, 437-446.4. Lambert, A. M.
& Bajer, A. (1972) Chromosoma 39, 101-144.5. Lloyd, C., Slabas,
A., Powell, A., MacDonald, G. & Bradley, R.
(1979) Nature (London) 279, 239-241.6. Powell, A., Lloyd, C.,
Slabas, A. & Cove, D. (1980) Plant Sci.
Lett. 18, 401-404.7. Wick, S. M., Seagull, R. W., Osborn, M.,
Weber, K. & Gun-
ning, E. S. (1981) J. Cell Bowl 89, 685-690.8. Franke, W., Seib,
E., Heith, W., Osborn, M. & Weber, K.
(1977) Cell Biol Int. Rep. 1, 75-83.9. Fuller, G. M., Brinkley,
B. R. & Baughter, J. M. (1975) Science
187, 948-950.10. Weber, K., Pollack, R. & Biehring, T.
(1975) Proc. Natl Acad.
Sci. USA 72, 459-463.11. De Mey, J., Moeremans, M., Geuens, G.,
Nuydens, R., Van
Belle, H. & De Brabander, M. (1980) Eur. J. Cell Biol 22,
297a.12. De Mey, J., Moeremans, M., Geuens, G., Nuydens, R. &
De
Brabander, M. (1981) Cell Biol. Int. Rep. 5, 889-899.13. De
Brabander, M., Geuens, G., De Mey, J. & Joniau, M. (1979)
Biol Cellulaire 34, 213-226.14. Mol6-Bajer, J. & Bajer, A.
(1967) La Cellule 67, 257-265.15. Geoghegan, W. D. & Ackerman,
G. A. (1977)J. Histochem. Cy-
tochem. 25, 1187-1200.16. Horisberger, M., Rosset, J. &
Bauer, H. (1975) Experientia 31,
1147-1149.17. Frens, G. (1973) Nat. Phys. Sci. 241, 20-22.18.
Bajer, A. & Mol-Bajer, J. (1982) Cold Spring Harbor Symp.
Quant. Biol 46, 263-283.19. Jensen, C. & Bajer, A. (1973)
Chromosome 44, 73-89.20. Gunning, B. E. S., Hardham, A. R. &
Hughes, J. E. (1978)
Planta 143, 161-180.21. Howard, R. & Aist, J. (1970)J.
Ultrastruct. Res. 66, 224-234.22. Inou6, S. & Bajer, A. (1961)
Chromosoma 5, 48-63.23. Lambert, A. M. (1980) Chromosoma 76,
295-308.24. Pickett-Heaps, J., McDonalds, K. & Tippit, D.
(1975) Proto-
plasmt 86, 205-242.25. Deysson, G. (1975) in Microtubules and
Microtubule Inhibitors,
eds. Borgers, M. & De Brabander, M.
(Elsevier/Nortl-Holland,Amsterdam), pp. 427-451.
26. Ostergren, G. (1954) Proc. 8th Congr. Int. of Bot. Paris,
15-16.27. Ostergren, G., Koopnans, A. & Reitalu, J. (1953) Bot.
Not. 4,
417-419.28. Lambert, A. M. & Benner-Schmit, A. C. (1980) J.
Cell Biol 87,
234a.29. Inoue, S. & Sato, H. (1967)J. Gen. Physiol 50,
259-292-.30. Inoue, S., Fuseler, J., Salmon, E. D. & Ellis, G.
W. (1975) Bio-
phys. J. 15, 725-744.31. Bajer, A. & Mole-Bajer, J. (1979)
in Cell Motility: Molecules and
Organization, eds. Hatano S., Ishikawa, H. & Sato, H.
(Univer-sity of Tokyo Press, Tokyo), pp. 569-592.
32. Bajer, A., De Brabander, M., Mole-Bajer, J., De Mey, J.,
Pau-laitis, S. & Geuens, G. (1980) in Microtubules and
MicrotubuleInhibitors, eds. De Brabander, M. & De Mey, J.
(Elsevier/North-Holland, Amsterdam), pp. 399-425.
33. Mazia, D., Paweletz, N., Sluder, G. & Finze, E. (1981)
Proc.NatL Acad. Sci. USA 78, 377-381.
34. Euteneuer, U. & McIntosh, J. R. (1981) Proc. Nati Acad.
Sci.USA 78, 372-376.
Proc. Natl. Acad. Sci. USA 79 (1982)
Dow
nloa
ded
by g
uest
on
June
24,
202
1