Fluorescence Microscopy of Etched Methacrylate SectionsImproves the Study of Mitosis in Plant CellsJOHN C. HOFFMAN,1* KEVIN C. VAUGHN,1 AND J. MICHAEL MULLINS2

1Southern Weed Science Laboratory, U.S. Department of Agriculture, Agricultural Research Service, Stoneville, Mississippi 387762Department of Biology, Catholic University of America, Washington, D.C.

KEY WORDS anti-mitotic antibody; microtubule; serial sections; spermatogenous cell; oat root

ABSTRACT Etched sections of methacrylate infiltrated plant tissue [Gubler (1989) Cell Biol.Int; Rep., 13:137–145; Baskin et al. (1992) Planta, 187:405–413] offer many advantages over themore traditional squash technique of Wick et al. [(1981) J. Cell Biol. 89:685–690] for immunofluores-cence microscopic investigation of the plant cytoskeleton, especially during mitosis. These advan-tages include: (1) unimpeded access of antibody probes, (2) confocal-like imaging without theexpense of confocal equipment, (3) maintenance of organ architecture as well as intracellularstructure, (4) the ability to independently examine separate focal planes with the same or multipleantibody(s) or other labelling compounds, and (5) the ability to archive unetched sections,polymerized or non-polymerized infiltrated tissue. In this paper examples of staining of variousmicrotubule cytoskeletal and mitotic proteins are shown in a variety of methacrylate embeddedplant tissues. Microsc. Res. Tech. 40:369–376, 1998. r 1998 Wiley-Liss, Inc.

INTRODUCTIONThe investigators of mitosis in animal cells were

armed with a powerful tool when Weber et al. (1975)demonstrated the beauty of the microtubule cytoskel-eton as revealed by its decoration with fluorochrome-tagged anti-tubulin antibodies. The ease with whichmammalian cells in culture could be manipulated foranalysis by fluorescence microscopy assured advances inthe understanding of microtubule dynamics and the identi-fication of microtubule-associated proteins and epitopes,especially those related to mitosis, in mammalian cells.Unfortunately for investigators of plant cells, there is noanalogous elegant technique. The plant cell’s evolutionaryboon, the cellulosic cell wall, is the obstruction thatmakes the plant cell biologist’s task more challenging.

Nevertheless, several procedures can be employed tocircumvent this barrier to antibody penetration andsubsequent microscopy, but each has disadvantages.Wall-less endosperm cells can be treated as mammaliancells grown in vitro (Franke et al, 1977), but these cellsdo not possess pre-prophase microtubule bands asvisualized by electron microscopy in multicellular planttissue and postulated to play a role in determiningorientation of cytokinesis, nor do these form cell platesat telophase (Pickett-Heaps and Northcote, 1966). Pro-toplasts derived from plant cells grown in vitro can besimilarly treated (Lloyd et al., 1979), but these also donot generally form pre-prophase bands and microtu-bule arrays are randomly oriented compared to those ofcells in tissue (van der Valk et al., 1980). By fixing planttissue before processing for immunofluorescence micros-copy, Wick et al. (1981) were able to preserve the microtu-bule configuration within individual cells, presumably, as itexists in vivo. However, squashing of the tissue to dispersecells results in the loss of tissue orientation. Further, fixedspecimens often must be treated with digesting enzymes topermeabilize the cell wall with unknown effects on cellstructural integrity and antigenicity.

In addition, there are protocols that require section-ing fixed tissue infiltrated with embedment followed byimmunofluorescence assay of sections etched free ofembedment: cryosectioning (Hogetsu and Oshima,1986); and embedding in polyethylene glycol (Tiwari etal., 1984), Steedman’s wax (Brown et al., 1989), or inmethacrylate (Gubler, 1989). Baskin et al. (1992) aptlydemonstrated the utility of an improved methacrylatetechnique with an abundance of antibody probes. Wetoo have found that fluorescence microscopy of etchedsections of methacylate-infiltrated tissue offers theplant cell biologist the unique advantage of well-preserved, properly oriented cytoskeletal structure vis-ible in thin focal planes (Hoffman and Vaughn, 1994a,b,1995a,b). We describe the technique we use and presentexamples demonstrating the capability of this tech-nique in studying plant cell mitosis.

MATERIALS AND METHODSPlant Material and Treatment

Spores of the pteridophyte Ceratopteris richardiiwere sown according to Warne et al. (1986) and grownaxenically for 7–10 days under 80 µmol/m2/s PAR.Seeds of onion (Allium cepa L.) and oat (Avena sativa L.‘‘Victory’’) were germinated in the dark for 3 days atroom temperature on moistened filter paper. In someinstances, oat seedlings were exposed during the finalday to 1 µM pronamide (Chem Services, West Chester,PA) prepared from a stock concentration of 0.1 M inacetone.

Transmission Electron MicroscopyTissue from C. richardii was prepared for transmis-

sion electron microscopy utilizing the fixation and

*Correspondence to: Kevin C. Vaughn, USDA, ARS, Southern Weed ScienceLab., P.O. Box 350, Stoneville, MS 38776.

Received 10 February 1995; accepted in revised form 27 April 1995

MICROSCOPY RESEARCH AND TECHNIQUE 40:369–376 (1998)

r 1998 WILEY-LISS, INC.

Fig. 1. Triple-label fluorescence of onion root cells. A single sectionof onion root labelled with monoclonal anti-mitotic antibody C9 (A),anti-tublin antisera (C), and DAPI (E). A different section labelledwith monoclonal anti-mitotic antibody MPM-2 (B), anti-tubulin (D),

and DAPI (F). The mitotic cell in A–C is judged to be in late anaphaseand that in D–F is judged to be in transition from prophase tometaphase. Bar 5 50 µm.

370 J.C. HOFFMAN ET AL.

Fig. 2. Immunofluorescence and transmission electron microscopyof MTOCs of spermatogenous cells of Ceratopteris richardii. Double-label indirect immunofluorescence of the same section of cells (A andB) allows unambiguous C9 anti-centrosome antibody localization (A,arrow) to an MTOC (B, arrow) delineated by rabbit anti-tubulin

antisera. The MTOC corresponds to the blepharoplast (b) seen bytransmission electron microscopy from a separate preparation of cells(C, arrowheads designate MTs). N, nucleus. Bars 5 (A,B) 50 µm; (C)0.5 µm.

embedding protocols described previously (Vaughn etal., 1993).

Fluorescence MicroscopyTissue from C. richardii and excised root tips were

processed essentially as Baskin et al. (1992). Tissuewas fixed for 1–2 hours by immersion in freshly pre-pared 3.7% paraformaldehyde in 50 mM 1,4 pipera-zinedithanesulfonic acid (Pipes) buffer, pH 7.2, with 1mM dithiothreitol (DTT). Fixed tissue was rinsed withtwo 15-minute changes of ice-cold 50 mM Pipes with 1mM DTT. Samples were dehydrated on ice in a gradedethanol series of 10, 20, 40, 60, 80, and 95% in 50 mmPipes and 1 mM DTT for 30 minutes each and, then, atleast three 100% ethanol and 10 mM DTT changes at220°C with one change overnight. Dehydrated sampleswere infiltrated at 220°C for 2 hours with degassed 4:1butyl:methyl methacrylate plus 10 mM DTT and 0.35%benzoin methyl ether at 33, 67, and 100% methacrylatein ethanol plus 10 mM DTT, followed by at least threechanges of 100% methacrylate extending over 24 hours.Methacrylate-infiltrated samples were polymerized at4°C overnight in a UV illuminated chamber (Ladd, Inc.,Burlington, VT). For roots, conical Beem capsules with10 µL of epon polymerized in the tips were employed toensure proper longitudinal orientation. Sections of 0.5µm thickness were cut with a diamond knife on aReichert Ultracut ultramicrotome, transferred on wa-ter doplets, and dried down onto coverslips on a slidewarmer or air-dried at room temperature. Methacry-late was etched from sections by immersion of cover-slips in acetone for 5–10 minutes and sections wererehydrated in a graded ethanol series (100, 75, 50, 25,and 0%) in phosphate or Tris buffered saline. Etchedsections were blocked with 1% (w/v) normal goat serumin buffered saline and reacted for 45–60 minutes withthe following dilutions of primary antibody for indirect

immunofluorescence microscopy as described by Osbornand Weber (1982): 1:200, polyclonal rabbit anti-seaurchin tubulin antiserum (Polysciences, Warrington,PA); 1:200, monoclonal anti-a-tubulin or anti-b-tublinascites (Amersham Corporation, Arlington Heights,IL); 1:50, anti-mitotic monoclonal antibody MPM-2ascites (P. Rao, Baylor Medical College, Houston, TX, orUpstate Biotechnology, Inc., Lake Placid, NY); 1:200,anti-mitotic monoclonal antibody C9 ascites. Secondaryantibody preparations were 1:60 dilutions of fluores-cein- or rhodamine-conjugated goat anti-mouse IgG orIgM or anti-rabbit IgG (Kirkegaard and Perry, Gaithers-berg, MD; Sigma, St. Louis, MO). Dried potato extractfrom a commercially available source was added todilutions of ascites to remove antibodies that bind tostarch. When sections were treated with more than oneprimary antibody, these were reacted simultaneously,as were secondary antibodies. Coverslips were reactedbriefly with 1 µg/mL 48,6-diamidino-2-phenyl indole(DAPI), rinsed, and mounted with Mowiol. Labelledsections were viewed using an Olympus OM-2 micro-scope equipped with epifluorescence optics. Simulta-neous DAPI fluorescence of chromatin and autofluores-cence of oat cell walls was recorded under violetexcitation with a 455-nm dichroic mirror and barrierfilter.

RESULTSSections of methacrylate-embedded plant tissue can

be dried onto coverslips, etched free of embedment, andcan then be processed for immunofluorescence micros-copy in a manner identical to that described for fixedand permeabilized mammalian tissue culture cellsgrown on cover slips (Osborn and Weber, 1982). As anexample, Figure 1 shows mitotic onion root cells triply-labelled with anti-mitotic antibody C9 (Hoffman andMullins, 1990) or MPM-2 (Davis et al., 1983), anti-

Fig. 3. Double-label fluorescence microscopy of mitotic spermatog-enous cells of C. richardii. The synchrony of cells within a singleantheridia allows visualization of many mitotic cells from differentorientations but on one plane of focus. Double-label immunofluores-

cence of a section of an antheridia, showing five cells judged to be inearly anaphase, with anti-tubulin antisera (A) and monoclonal anti-mitotic antibody C9 (B).

372 J.C. HOFFMAN ET AL.

tubulin antisera, and DAPI. Because sectioned mate-rial is used, IgM antibodies like C9 have access tointercellular antigens that are sometimes impaired inwhole cell mounts (Wick, 1993). Also, sectioned tissueresults in a planar image with resolution similar to thatobtained under confocal microscopy, but under stan-dard epifluorescence optics.

The ability to label sectioned tissue with more thanone antibody also allows unambiguous identification ofsubcellular organelles/architecture. In Figure 2, themicrotubule focus revealed by anti-tubulin antiserumin a spermatogenous cell of C. richardii is labelled bythe anti-centrosome antibody C9. Transmission EM of aseparate spermatogenous cell apparently at the samestage of development identifies this microtubule orga-nizing center as the blepharoplast.

Because sections are cut from fixed tissue and, thus,maintain tissue orientation, characteristics of the plant

specimen can be exploited to enhance the informationgained from a single microscopic observation. For ex-ample, developing spermatogenous cells of C. richardiimaintain synchrony within an antheridium (Hoffmanand Vaughn, 1995a). This allows visualization of themitotic apparatus of many cells at the same stage ofmitosis, but from different focal planes through theapparatus (Fig. 3). Another example of tissue exploita-tion is the monitoring of the position of cell walls with adifferent filter set than that used for immunofluores-cence. Because oat cell walls autofluoresce with thefilter set used for DAPI staining, thin sections of oatroots display individual oat cells outlined by their ownfluorescence, augmenting information obtained from label-ling experiments (Fig. 4). Cell walls of other plants can belabelleddirectly allowing similar capabilities (Gubler, 1989).

Perhaps the most useful aspect of the post-embed-ding approach is the ability to treat individual serial

Fig. 4. Double-label fluorescence and cell wall auto-fluorescence ofoat root cells. Cell wall autofluorescence and DAPI labeling of chroma-tin allow visualization of isodiametrically swollen cells with irregularnuclei and cells arrested at pro-metaphase (B) resulting from the lossof cortical and mitotic MTs (A). A cell judged to be in prophase, with

pre-prophase band in cross-section, is designated by arrowheads (Cand D). Sections of an oat root treated with 1 µM pronamide (A and B)and of an untreated oat root (C and D) were double-labelled withmonoclonal anti-b tubulin (A and C) and DAPI (B and D). Bar 5100 µm.

373FLUORESCENCE MICROSCOPY OF PLANT SECTIONS

Fig. 5. Double-label fluorescence of sequentially cut sections of oatroot cells. Four sequentially cut sections of oat roots were labelledwith: (1) monoclonal anti-mitotic antibody MPM-2 (A) and DAPI (B),

(2) monoclonal anti-mitotic antibody C9 (C) and DAPI (D), (3) monoclo-nal anti-a tubulin (E) and DAPI (F), and (4) monoclonal anti-b tubulin(G) and DAPI (H). Bar 5 50 µm, gap 5 2 µm.

374 J.C. HOFFMAN ET AL.

sections with a single probe, or a variety of probes. InFigure 5, four, serial 0.5 µm sections of oat root arepresented in sequence. Each of the sections is labelledwith one of four different antibodies. This allows com-parison of the four antibody probes within a tissuethickness of 2 µm. All four sections are also labelledwith DAPI, demonstrating, along with cell wall fluores-cence, the sequential relationship of the sections andassisting in the proper alignment of the cells of interest.We have used an approach similar to this to visualizepost-translational modifications of tubulin in Ceratop-teris spermatogenous cells (Hoffman and Vaughn,1995b).

DISCUSSIONAs previously discussed and demonstrated by Gubler

(1989) and Baskin et al. (1992), post-methacrylateembedded plant tissue offers advantages over the squashtechnique developed by Wick et al. (1981), which is themost widely employed technique for preparing plantcells for fluorescence microscopy. Both techniques em-ploy fixed tissue to maintain proper intracellular cyto-skeletal architecture; however, examination of sectionsof fixed tissue allows visualization of the architecture ofentire regions of plant organs (e.g., the root zone ofelongation; Baskin et al., 1994), and allows assignmentwith certainty of individual cells to those regions, whichis nearly impossible with the squash technique. Thus,in the root cells of the cap, meristem, quiescent center,vasculature, or cortical tissue, which are expected tohave different cell cycle kinetics (Steeves and Sussex,1972), can be examined with different cell cycle-relatedprobes. Maintenance of tissue orientation also allowsexploitation of characteristics (e.g., synchrony of devel-oping cells and fluorescence of cell walls) of the plantunder investigation.

Another advantage of using etched sections is thatthey allow complete access of antibody probes, includ-ing IgMs, to the cell interior without the necessity ofdigesting or breaching the plant cell wall. This contin-ues to be a problem for squashed tissue (McCurdy et al.,1988; Wick, 1993), raising a question of the meaning ofthe absence of monoclonal antibody binding. In sec-tions, absence of antibody binding cannot be attributedto the effects of residual cell wall, but must be due toeither absence of epitope or alteration of epitope suffi-cient to inhibit antibody recognition. In regard to thelatter, we have found that monoclonal antibodies thatrequire, or work optimally with, aldehyde fixed anti-gens (usually from animal cells), recognize analogousantigens in plant cells even after alcohol and acetonetreatments of tissue necessary for this reversible embed-ding technique. Unfortunately, we have also found thatmonoclonal antibodies that require, or work optimallywith, alcohol fixed antigens, do not work following thistechnique (Hoffman et al., 1994).

A disadvantage to this technique is the lengthyprocessing time (approximately 4 days) needed to pre-pare the embedded tissue samples; however, oncesamples have been infiltrated with methacrylate theycan be stored at 220°C for at least 2 years andpolymerized samples as well as unetched sections canbe stored for months with only slight loss of detail afterprocessing (Hoffman, unpublished data; Baskin et al.,1992). The ability to archive specific tissue specimens

and individual preparations of known quality is impos-sible with the squash technique.

The main disadvantage of this technique is that theuse of sectioned tissue precludes the sorts of three-dimensional images of whole cells characteristic ofmaterial prepared using the squash technique. Yet thisdeficiency is ameliorated by the fact that each sectioncan be examined individually, abolishing the limita-tions that accompany indirect immunofluorescence ofwhole cells with multiple probes (i.e., the requirementof different immunoglobin types or species of origin ofantibodies). Further, because exquisite detail can beobtained from the confocal-like image produced from alabelled section, and because multiple antibody probesand/or other fluorescence labelling compounds can beapplied to serial sections (as thin as 0.3 µm in ourhands), the use of three-dimensional reconstructionfrom multiply labelled serial sections can, quite possi-bly, be employed to generate whole cell images withmultiple parameters of cell structural detail. Thus, it ispossible that in the foreseeable future post-embeddingfluorescence microscopy of methacrylate infiltrated tis-sue will empower the plant cell biologist at least to thesame extent that the technique introduced by Weber etal. (1975) did for animal cell biologists.

ACKNOWLEDGMENTSThanks are extended to Lynn Libous-Bailey for ex-

pert technical assistance and to Dr. P. Rao for theMPM-2 antibody.

REFERENCESBaskin, T.I., Busby, C.H., Fowke, L.C., Sammut, M., and Gubler, F.

(1992) Improvements in immunostaining samples embedded inmethacrylate: Localization of microtubules and other antigensthroughout developing organs in plants of diverse taxa. Planta,187:405–413.

Baskin, T.I., Wilson, J.E., Cork, A., and Williamson, R.E. (1994)Morphology and microtubule organization in Arabidopsis rootsexposed to oryzalin and taxol. Plant Cell Physiol., 35:935–942.

Brown, R.C., Lemmon, B.E., and Mullinax, J.B. (1989) Immunofluores-cent staining of microtubules in plant tissue: Improved embeddingand sectioning techniques using polethylene glycol (PEG) andSteedman’s wax. Bot. Acta., 102:54–61.

Davis, F.M., Tsao, T., Fowler, S., and Rao, P. (1983) Monoclonalantibodies to mitotic cells. Proc. Natl. Acad. Sci. U.S.A., 80:2926–2930.

Franke, W.W., Seib, E., Osborn, M., Weber, K., Herth, W., and Falk, H.(1977) Tubulin-containing structures in the anastral mitotic appara-tus of endosperm cells of the plant Leucojum aestivum as revealedby immunofluorescence microscopy. Cytobiologie, 15:24–48.

Gubler, F. (1989) Immunofluorescence localization of microtubules inplant root tips embedded in butyl-methyl methacrylate. Cell Biol.Int. Rep., 13:137–145.

Hoffman, J.C., and Mullins, J.M. (1990) Nuclear and mitoticallyenhanced epitope. Cell Motil. Cytoskeleton, 16:68–79.

Hoffman, J.C., and Vaughn, K.C. (1994a) Griseofulvin destabilizesplant microtubules. Cytobios, 77:253–258.

Hoffman, J.C., and Vaughn, K.C. (1994b) Mitotic disrupter herbicidesact by a single mechanism but vary in efficacy. Protoplasma,179:16–25.

Hoffman, J.C., and Vaughn, K.C. (1995a) Using the developing sperma-togenous cells of Ceratopteris to unlock the mysteries of the plantcytoskeleton. Int. J. Plant Sci. 156:346–358.

Hoffman, J.C., and Vaughn, K.C. (1995b) Post-translational tubulinmodifications in spermatogenous cells of the pteridophyte Ceratop-teris richardii. Protoplasma, 186:169–182.

Hoffman, J.C., Vaughn, K.C., and Joshi, H.C. (1994) Structural andimmunocytochemical characterization of microtubule organizingcenters in pteridophyte spermatogenous cells. Protoplasma, 179:46–60.

Hogetsu, T., and Oshima, Y. (1986) Immunofluorescence microscopy of

375FLUORESCENCE MICROSCOPY OF PLANT SECTIONS

microtubule arrangement in root cells of Pisum sativum L. varAlaska. Plant Cell Physiol., 27:939–945.

Lloyd, C.W., Slabas, A.R., Powell, A.J., MacDonald, G., and Badley,R.A. (1979) Cytoplasmic microtubules of higher plant cells visual-ized with anti-tubulin antibodies. Nature (Lond.), 279:239–241.

McCurdy, D.W., Sammut, M., and Gunning, B.E.S. (1988) Immunofluo-rescent visualization of arrays of transverse cortical actin microfila-ments in wheat root-tip cells. Protoplasma, 147:204–206.

Osborn, M., and Weber, K. (1982) Immunofluorescence and immunocy-tochemical procedures with affinity purified antibodies: Tubulin.Methods Cell Biol. 24:97–132.

Pickett-Heaps, J.D., and Northcote, D.H. (1966) Organization ofmicrotubules and endoplasmic reticulum during mitosis and cytoki-nesis in wheat meristems. J. Cell Sci., 1:109–120.

Steeves, T.A., and Sussex, I.M. (1972) In: Patterns in Plant Develop-ment. Prentice-Hall, Inc., Englewood Cliffs, NJ, pp. 170–190.

Tiwari, S.C., Wick, S.M., Williamson, R.E., and Gunning, B.E.S. (1984)The cytoskeleton and integration of cellular function in cells ofhigher plants. J. Cell Biol., 99:63s–69s.

van der Valk, P., Rennie, P.J., Connolly, J.A., and Fowke, L.C. (1980)Distribution of cortical microtubules in tobacco protoplasts. Animmunofluorescence microscopic and ultrastructural study. Proto-plasma, 105:27–43.

Vaughn, K.C., Sherman, T.D., and Renzaglia, K.S. (1993) A centrinhomologue is a component of the multilayered structure in bryo-phytes and pteridophytes. Protoplasma, 175:58–66.

Warne, T.R., Warker, G.L., and Hickok, L.G. (1986) A novel method forsurface sterilizing fern spores. Am. Fern J., 76:187–188.

Weber, K., Bibring, T., and Osborn, M. (1975) Specific visualization oftubulin-containing structures in tissue culture cells by immunofluo-rescence. Exp. Cell Res., 95:111–120.

Wick, S.M. (1993) Immunolabelling of antigens in plant cells. MethodsCell Biol., 37:171–201.

Wick, S.M., Seagull, R.W., Osborn, M., Weber, K., and Gunning, B.E.S.(1981) Immunofluorescence microscopy of organized microtubulearrays in structurally stabilized meristematic plant cells. J. CellBiol., 89:685–690.

376 J.C. HOFFMAN ET AL.