Submitted April 8, 1998; revised Sept. 7, 1998; accepted Sept. 9, 1998

Improved Sensitivity for High Resolution Situ Hybridization Using Resin Extraction

In of

Methyl Methacry late Embedded Material

Chieko Saito', Makoto Hayashi*, Atsushi Sakai', Makoto Fujie3, Haruko Kuroiwa4 and Tsuneyoshi Kuroiwa'

Department of Biological Sciences, School of Science, University of Tokyo, Hongo 7-3- 1, Tokyo 1 13-0033, 2Department of Biotechnology, Graduate School of Engineering, Osaka University, Osaka 565-0877,

Department of Molecular Biotechnology, Graduate School of Engineering, Hiroshima University, Higashi- Hiroshima 739-8527, and 4Kyoritsu Woman's lunior College, Tokyo 7 07 -8433, lapan

ABSTRACT. An in situ hybridization procedure resulting in both high resolu- tion and sensitivity was establlphad by using the removable methyl methacry- late resin, Technovit 9100. Young bicel- lular pollen of tobacco ( N b t i a n a tabacum L. SR-1) was embedded in Technovit 9100 resin and sectioned. The resin was extracted with (2- methoxyethy1)-acetate followed by in situ hybridization with cRNAprobes to detect cytoplasmic 188/25S rRNA. Sig- nal intensity obtained by this procedure was approximately twlce as great as that obtained by an earlier procedure using Technovit 7100, a glycol methamylate resin that cannot be removed from sec- tions. This improvement in sensitivity made it possible to observe subcellular localization of small amounts of RNA as revealed by visualization of plastid 235 rRNA in a generative cell of PZumbago audculata pollen.

Key words: generative cell, in situ hy- bridization, Nico t iana tabacum, Plum- bag0 aurlculata, pollen, resin extra- ction, Technovit 9100, vegetative cell

Please address correspondence to: Chieko Saito, Department of Biological Sciences, School of Science, University of Tokyo, Hongo 7-3-1, Tokyo 11 3- 0033, Japan, Fax no: + 81 -3-3814-1408, E-mail: chiezo63biol.s.u-tokyo.ac.jp

BlOTECHNlC & HISTOCHEMISTRYQ by Lippincotf Williams L? Wilkins, Inc. 1052-0295/99/40-48 Volume 74, Number 1

n angiosperm pollen development, the I asymmetric mitotic division of the hap- loid microspore (pollen mitosis one, PM I) results in the formation of two dimorphic cells, the vegetative cell and the genera- tive cell (for reviews, see Bedinger 1992, McCormick 1993, Tanaka 1997). The bulk of the pollen grain is the larger vege- tative cell which has a very hard wall (ex- ine) and supports pollen maturation and pollen tube growth. The smaller genera- tive cell divides and produces two sperm cells for double fertilization. The genera- tive/sperm cell is entirely enclosed within the vegetative cell. To study gene expres- sion occurring in a single pollen grain composed of two types of cell requires a method for in situ hybridization that re- sults in both high resolution and high sensitivity. It is not easy, however, to ob- serve expression patterns within a pollen grain because of its thick, hard wall and the small size of the generative/sperm cells.

In recent years, several studies have been undertaken to develop new acrylic embed- ding resins that could be applied to the cut- ting-grinding or thin sectioning techniques (Donath and Breuner 1982, Reynolds and Raghaven 1982, Causton 1984). Samples embedded in such resins can be observed by both light and electron microscopy (Kuroiwa et al. 1991, 1992, Fujie et al.

40

Bio

tech

His

toch

em D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

ufts

Uni

vers

ity o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.

Technovit Section In Situ Hybridization 41

1993, Botti et al. 1997). The thin sectioning technique using glycol methacrylate based embedding materials (commercially avail- able as Technovit 7100) is especially suit- able for the analysis of pollen. Clear images can be obtained by this method, although a pollen grain has a thick, autofluorescent wall (Sodmergen et al. 1995, Saito et al. 1997, 1998a). Furthermore, high resolution in situ hybridization has been achieved by combining the Technovit 7100 section and the immunodetection system using digoxigenin-labeled probes and fluorescein isothiocyanate (FITC) conjugated antibod- ies (Fujie et al. 1995). This resin, however, cannot be removed from the section and probes and antibodies cannot penetrate the section so that antigen-antibody or RNA- RNA hybridization reactions occur only on the surface of the section. Thus, high sensitivity for detection by in situ hybridiza- tion on Technovit 7100 sections is not achieved.

Methyl methacrylate is an acrylic resin specifically formulated to embed large un- decalcified hard tissues such as bones or teeth that are prepared for light mi- croscopy using the cutting-grinding tech- nique (Donath and Breuner 1982). This resin can removed by treating sections with (2-methoxyethyl]-acetate, thereby im- proving the reactivity between probes and target transcripts.

We have developed a method for in situ hybridization on thin sections prepared with methyl methacrylate based resin, commercially available as Technovit 9100, using an immunodetection system with FITC-conjugated antibodies. We confirmed a two-fold increase in sensitiv- ity compared to a previous method using Technovit 7100 resin (Saito et al. 1998a). Furthermore, our new method makes it possible to visualize subcellular localiza- tion of 23s rRNA in nonphotosynthetic plastids. For this purpose, pollen grains of Plumbago auriculata were used be- cause plastids are distributed in one pole of the generative cell just before pollen mitosis two (PM 11) in P. auriculata (Saito et al. 1998b) and P. zeylanica (Sodmer-

gen et al. 1995, Russell et al. 1996) mak- ing it easy to demonstrate specificity of hybridization signals (i. e., signals on plastids only). The significance of detec- tion in 23s rRNA in plastids present in generative cell is discussed in relation to biparental/paternal inheritance of plastids.

MATERIALS AND METHODS

Preparation of Samples in Technovit Resin Anthers were excised from a single flower bud of tobacco (Nicotiana tabacum L. SR- 1; bud length = 22 mm) and Plumbago au- riculata (bud length= 11.0 mm), fixed, and dehydrated in an ethanol series as described previously (Kuroiwa et al. 199 1, Saito et al. 1997). The dehydrated samples were then divided into two groups. One group was embedded in glycol methacry- late-based resin (Technovit 7 100, Kulzer and Co., Wehrheim, Germany) as de- scribed previously (Kuroiwa et al. 1991, Saito et al. 1997), and the other group was embedded in methyl methacrylate-based resin (Technovit 9100, Kulzer). The sam- ples were embedded in Technovit 9100 following manufacturer’s instructions ex- cept that the step removing lipids with xylol was omitted. Briefly, the dehydrated samples were transferred from 100% ethanol through a series of solutions of 50, 75,85, and 100% Technovit 9100 mixture (94% basic solution, 5% softener, and 1% initiator, Kultzer). These procedures were performed at room temperature under at- mospheric conditions on a rotator run at low speed in sample vials capped tightly to prevent volatilization. After soaking the samples overnight in 100% Technovit 9100 mixture on a rotator, the solution was replaced and infiltration was contin- ued for at least an additional 6 hr. The samples were transferred to 1.5 ml airtight sampling tubes, and the resin was solidi- fied by incubation at 45 C overnight. Thin sections (1 pm) of each sample embedded in either Technovit 7100 and Technovit 9 100 were cut with a dry glass knife on an ultramicrotome (MT- lb; RMC EIKO,

Bio

tech

His

toch

em D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

ufts

Uni

vers

ity o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.

42 Biotech nic & H istochem istry

Kawasaki, Japan, or ULTRACUT UCT; Le- ica, Wien, Austria), placed in a drop of dis- tilled water on a coverslip coated with aminopropyltriethoxysilane and glu- taraldehyde (Gottlieb and Glaser 1975) and air dried.

Extraction of Resin from Samples Embedded in Technovit 9100 After attachment to the coated coverslips, sections of the samples embedded in Tech- novit 9100 resin were extracted following manufacturer's instructions with minor modifications (Kultzer). Briefly, the sec- tions attached to coated coverslips were treated with (2-methoxyethyl)-acetate (Wako Pure Chemical Industries, Osaka, Japan) for 5 min, washed in a series of ethanol solutions (100, 90, 80, 70, 50, 30, 10%) and water with five gentle shakes up and down in each, and air dried.

In Situ Hybridization and Microphotometry Sections of the samples embedded in Tech- novit 7100 and Technovit 9100 were attached to the same coverslips, and both sections were treated with (2- methoxyethy1)-acetate in the manner de- scribed above. In situ hybridization was performed as described previously (Fujie et al. 1995) with or without digoxigenin- labeled sense and antisense rRNA probes. Hybridization was allowed to proceed at 45 C for 12 hr. Probes were detected with anti-digoxigenin antibodies (Boehringer Mannheim Biochemicals, Mannheim, Ger- many) and fluorescein isothiocyanate (FITC) conjugated second antibodies (anti- mouse immunoglobulin goat antibody conjugated with FITC, TAGO, Inc., Burlingame, CA). For detection of 23s rFtNA, tertiary antibodies (anti-goat im- munoglobulin pig antibody conjugated with F IE , TAGO) were used. The sample was mounted with 1 p g / d 4', 6-diamidino-2- phenylindole (DAPI) dissolved in TAN buffer

mM 2-mercaptoethanol and 1.2 mM sper- midine) and 1 mg/ml n-propyl gallate dis-

(20 mM 'his-HC1, pH 7.6, 0.5 mM EDTA, 7

solved in 50% glycerol. The sample was ob- served with an epifluorescence microscope (BHS-RFK; Olympus, Tokyo, Japan) (Kuroiwa et al. 1986). The fluorescence in- tensities of the hybridization signals were measured with a video-intensified micro- scope photon counting system (VIMPCS; Hamamatsu Photonics Ltd., Hamamatsu, Japan) connected to an epifluorescence mi- croscope as described previously (Kuroiwa et al. 1986). Photographs were taken at a magmfication of 100. 500, and 1500 X on 35 mm Neopan 400 PRESTO(Fuji Photo Film Ltd., Tokyo, Japan) or Neopan 1600 SUPER PRESTO (Fuji Photo Film Ltd.) film. The photographs were taken under identi- cal conditions to compare the signal inten- sity of the sections in Technovit 7100 and Technovit 9100.

RESULTS

Effects of Treating Technovit 9 100 Sections with (2-Methoxyethy1)-acetate The effects of treatment with (2- methoxyethy1)-acetate on Technovit 9 100 sections (Figs. 1-8) were examined first. N. tabmum anthers containing young bicel- lular pollen were embedded in Technovit 9100 resin, sectioned, and observed by Nomarski/DAPI fluorescence microscopy before (Figs. 1, 3, 4 and 5) and after (Figs. 2, 6, 7 and 8) resin extraction with (2- methoxyethy1)-acetate. Observations at low magnification revealed that the outline of the Technovit 9 100 section disappeared after treatment with (2-methoxyethyl)-ac- etate indicating extraction of the resin (data not shown). In contrast, the outline of the Technovit 7100 section did not dis- appear after the same treatment (data not shown). Sections untreated with (2- methoxyethy1)-acetate were somewhat compressed during sectioning using glass knives, but resin extracted sections re- gained their original shape to a certain ex- tent (compare Figs. 1 and 2, 3-5 and 6-8). Extraction of resin also improved DAPI- fluorescence (Figs. 3 and 6) and Nomarski (Figs. 5 and 8) images. For example, or-

Bio

tech

His

toch

em D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

ufts

Uni

vers

ity o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.

Technovit Section In Situ Hybridization 43

Figs. 1-8. The effect of resin extraction on Technovit 9100 embedded sections. DAPI-epifluorescence (1, 2, 3 and 6) , DAPI- epifluorescence/Nomarski (4 and 7) and Nomarski (5 and 8) micrographs of sections of anther (1 and 2) and pollen grains (3-8) of N. tabacum. Figures 1, 3, 4 and 5 are Technovit 9100 sections before extraction of resin. Figures 2, 6, 7 and 8 areTechnovit 9100 sections after extraction of resin. DAPl images (Figs. 1, 2, 3 and 6) show cell nuclei (black large arrows, vegetative nuclei; black small arows, generative nuclei). Nomarski and DAPl images show amyloplast-like organelles (white arrows) seen as tiny dots in the cytoplasm of the vegetative cell. Pollen walls (white arrowheads) emit strong autofluorescence. Figures 3-5 are one field and 6-8 are another field. Bars represent 50 pm in Figs. 1 and 2, and 10 pm in Figs. 3-8.

ganelle DNA in the DAPI fluorescence im- age, as well as the boundary of various structures in the pollen grains (the pollen wall and amyloplast-like organelle in the vegetative cell) in the Nomarski image, is observed more clearly in sections after treatment of resin extraction than in un- treated ones. These results indicate that the treatment with (2-methoxyethyl)-ac- etate effectively removes the resin from Technovit 9 100 sections without degrad- ing the cellular structure.

Improved Sensitivity of in Situ Hybridization Using Technovit 91 00

We compared the resolution and signal in- tensity of in situ hybridization using sec- tioned material prepared with Technovit 7100 andTechnovit 9100 resin (Figs. 9-16). N. tabacum SR- 1 anthers containing young bicellular pollen embedded in Technovit 7100 (Figs. 9, 11, 13 and 15) and Technovit 9100 (Figs. 10, 12, 14 and 16) were cut at the same thickness, attached to the same

Bio

tech

His

toch

em D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

ufts

Uni

vers

ity o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.

Figs. 9-1 6. Improved sensitivity of in situ hybridization with Technovit 91 00 sections. Epifluorescence micrographs of thin sections of pollen embedded in Technovit 71 00 (9, 11, 13 and 15) and 91 00 (1 0, 12, 14 and 16) resin, attached to coverslips and treated with (2-methoxyethyl)-acetate. Figures 9-1 6 are the same magnification. Figures 9 and 11 are one field and Figs. 10 and 12 are another field. Localization of cytoplasmic 18S/25S rRNA was visualized by in situ hybridization using anti- digoxigenin antibodies and FITC-conjugated secondary antibodies (1 1-1 6). The sections were hybridized with digoxigenin- labeled antisense probes (1 1 and 12), sense probe (1 3 and 14) and no probe (1 5 and 16). DAPl images (9 and 10) show cell nuclei (large black arrows, vegetative cell nuclei; small black arrows, generative cell nuclei) and organelle nucleoid (white arrows) observed as tiny dots in the cytoplasm of vegetative cell. The boundary between the generative cell and the vegeta- tive cell was seen as a black line on the background (1 l and 12, black arrowheads). Bar represents 10 pm.

Bio

tech

His

toch

em D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

ufts

Uni

vers

ity o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.

Technovit Section In Situ Hybridization 45

coverslips, treated with (2-methoxyethy1)- acetate, and processed for in situ hybridiza- tion using sense- and antisense-RNA probes for 18S/25S rRNA. Treatment with (2-methoxyethyl)-acetate did not affect the results of in situ hybridization and DAPI staining with Technovit 7 100 sections (data not shown). On the other hand, it was im- possible to examine the effect of (2- methoxyethy1)-acetate treatment on in situ hybridization with Technovit 9 100 sections because the Technovit 9100 sections with- out treatment with (2-methoxyethyl)-ac- etate fell off of the coverslips during processing for in situ hybridization.

DAPI fluorescence microscopy (Figs. 9 and 10) revealed that the resolution ob- tained with Technovit 7 100 and Technovit 9100 was comparable. In both sections, a highly condensed generative cell nucleus and dispersed vegetative cell nucleus, as well as organelle nucleoids (seen as tiny fluorescent spots), are clearly observed. F I X images (Figs. 11-16) show the re- sults of in situ hybridization. The rRNA signal on the Technovit 9100 section is much stronger than that on the Technovit 7100 section (Figs. 1 1 and 12). When the sections were hybridized with the sense probe (Figs. 13 and 14) or when the probe was omitted (Figs. 15 and 16), the weak fluorescence obtained was regarded as background or autofluorescence.

To estimate the sensitivity of in situ hy- bridization, the signal intensities were quantified using microphotometry (Fig. 17). The net signal intensities were calcu- lated from the difference between the sig- nal intensities obtained with the antisense and sense probes. The results clearly demonstrated that the net intensity of the hybridization signals in the sections of Technovit 9100 after removal of resin was about twice as high as that in Technovit 7100. Thus, an improvement in the sensi- tivity of in situ hybridization was attained.

Localization of 23s rRNA in the Generative Cell of P. auricukfa We applied our method for in situ hybrid- ization to detect plastid 23s rRNA in non- photosynthetic plastids in a pollen grain. In

700

8 600 - L. t 500 2

400

0) 300

L

Q) 0 C

0 cn g 200

$100 3 E

.- c 0

K - Q ) o 71 00 91 00

Fig. 17. Signal intensity of in situ hybridization for cyto- plasmic 25S/18S rRNA in the cytoplasm of samples em- bedded in Technovit 7100 and Technovit 9100. The fluorescence intensity per area (measured areas were 23.4 or 46.8 pm2; camera lens, X 3.3; objective lens, X 100) was measured by VIMPCS. Fluorescence intensities are ex- pressed as relative values, where the value of the sample embedded in Technovit 7100 and hybridized with no probes was arbitrarily set at 100.

the section of a generative cell of P. auricu- lata pollen embedded in Technovit 9100 resin, two types of organelle nucleoids were observed in the DAPIfluorescence image outside the cell nucleus (Figs. 18 and 19). One type of spots emits more intense DAPI fluorescence and the other type of spots emits less intense DAPI fluorescence. These spots are defined as plastid-nu- cleoids and mitochondrial-nucleoids, re- spectively (Saito et al. 1998b). When an antisense probe was hybridized, distinct signals of 23s rRNA were clearly observed on plastids only (Figs. 18 and 20) and no signal was detected on mitochondria (Figs. 18 and 20, white arrowheads). When the sections were hybridized with the sense probe (Fig. 2 l), the observed weak and non- specific fluorescence was regarded as back- ground or autofluorescence.

DISCUSSION

Thin Sectioning of Technovit 91 00 Samples Technovit 9100 tends to compress during sectioning. Furthermore, because Tech- novit 9100 resin is harder and less water permeable than Technovit 7 100, sections could not be extended in a water drop on the coverslips. No significant sample defor-

Bio

tech

His

toch

em D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

ufts

Uni

vers

ity o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.

46 Biotec hnic & H istoc hem istry

Figs. 18-21. Plastid-specific signal of 23s rRNA in generative cell of P. auriculafa visualized by in situ hybridization. Signals were detected with anti-digoxigenin antibodies and FITC-conjugated secondary and tertiary antibodies (20-21 1. Technovit 91 00 sections after treatment of resin extraction were hybridized with digoxigenin-labeled antisense (20) and sense probes (21). DAPl images (18 and 19) show generative cell nuclei (large black arrows), plastid nucleoids (white arrows) and mito- chondria nucleoids (white arrowheads) observed as two types of spots. Figures 18 and 20 are one field, and Figs. 19 and 21 are another field. Figs. 18-21 are the same magnification. Bar represents 5 pm.

mation, however, has been mentioned in previous reports with methyl methacrylate embedded samples (Donath and Breuner 1982, Ikeda et al. 1992, Botti et al. 1997). In this study, we made 1 pm thick sections with a dry M e to compare the results of in situ hybridization using the two kinds of resin because in situ hybridization with Technovit 7 100 ordinarily uses sample sections cut fiom 0.5-1 pm thick (Fujie et al. 1995) with a dry knife. Our method of thin sectioning of Technovit 9100 embed- ded samples might require additional mod- ification to prevent deformation.

was improved. Although a method for in situ hybridization using [%]-labeled probes and a similar type of removable acrylic resin (methyl methacrylate mixed with butyl methacrylate) has been re- ported (Ikeda et al. 1992), the resolution using autoradiography is not high enough to analyze subcellular localization of tran- scripts and requires a long exposure time. In contrast, the method we developed al- lows rapid, high-resolution analysis.

Detection of Plastid 23s rRNA in Generative Cells of P. auriculata Pollen Using an earlier in situ hybridization pro- cedure with Technovit 7100, clear signals of 23s rRNA were obtained on young chloroplasts of the leaf primordium. Dis-

High-Resolution in Situ Hybridization for Analysis of Gene Expression in a Pollen Grain When resin extracted sections of Technovit tinct signals, however, were not observed 9100 were used, the sensitivity of in situ on proplastids of the root apical meristem hybridization detected by fluorescence of Arabidopsis (Fujie et al. 1995). In pres-

Bio

tech

His

toch

em D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

ufts

Uni

vers

ity o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.

Technovit Section In Situ Hybridization 47

ent study, signals of 23s rRNA were first detected in nonphotosynthetic plastids by an improved procedure of in situ hy- bridization with removable Technovit 9100. Signals of 23s rRNA were also de- tected in plastids of generative cells from P. auricuhta using an earlier procedure with Technovit 7100 (data not shown); however, they were dimmer than the sig- nals obtained with Technovit 9 100.

In the generative/sperm cells of P. zey- Zanica, which is closely related to P. auric- ulata, organelle nucleoids emitting strong and weak fluorescence with DAPI have been identified as plastid nucleoids and mitochondrial nucleoids, respectively, by simultaneous observation of fluorescence and electron microscopic images (Sodmer- gen et al. 1995). In P. zeybnica, plastids are distributed to one side of the genera- tive cell just before PM 11. The polarization of the plastids is retained during PM 11, and two highly dimorphic sperm cells, one plastid-rich and the other mitochondria- rich, are generated (Sodmergen et al. 1995, Russell et al. 1996). A similar phenome- non is also observed in P. auriculata (Saito et al. 1998b). In P. zeylanica, the plastid- rich sperm cell is known to fuse preferen- tially with the egg cell (Russell 1985), although the biological significance of the preferential distribution of plastids to sperm cells that fuse with egg cells re- mains unclear.

Some angiosperms exhibit biparental or paternal inheritance of plastids, while the great majority of angiosperms exhibit maternal inheritance of plastids (Hageman and Shroder 1989, Mogensen 1996). Preservation of plastid DNA in the genera- tive/sperm cells of mature pollen grain, as examined by DAPI-fluorescence mi- croscopy, is considered a prerequisite for the biparental/paternal inheritance of plas- tids (Kuroiwa 199 1). Preservation of plastid DNA in generative/sperm cells, however, does not guarantee that the plastids will be inherited biparentally/patemally. For ex- ample, once ribosomes (and the potential of protein synthesis) are lost from the plas- tids, the plastids would not regain their function even if they retained intact DNA

and were introduced into a fertilized egg. In the present study, the presence of 23s rRNA in the plastids of the generative cell from P. auriculQtQ pollen was demonstrated by in situ hybridization. In addition, plas- tids are distributed to one of the sperm cells with their nucleoids retained (Sodmergen et al. 1995, Saito et al. 199813). Since the plas- tid-rich sperm cell selectively fuses with egg cells in P. zeylanica (Russell 1985), these facts strongly suggest that the plastids are inherited biparentally/patemally in Plum- bago, although the pattern of plastid inher- itance in Plumbago has not been confirmed genetically.

ACKNOWLEDGMENTS The authors thank Associate Professor S . Kawano, Dr. Sodmergen, and Dr. T. Takano for technical advice. This research was supported by a grant for Special Re- search on priority Areas (to T. K., project no. 0610 1002) from the Ministry of Educa- tion, Science and Culture of Japan, and by a grant for a Pioneering Research Project in Biotechnology from the Ministry of Agri- culture, Forestry and Fisheries of Japan.

REFERENCES Bedinger, P. 1992. The remarkable biology of pollen.

Plant Cell 4: 879-887. Botti, F., Ragazzoni, E., FWoni, A. and Cocchia, D.

1997. Assessment of the acrylic resin Technovit 7200 VLC for studying the gingival mucosa by light and electron microscopy. Biotech. & His- tochem. 72: 178-184.

Causton, B. E. 1984. The choice of resins for elec- tron immunocytochemistry. In: I m o l a b e U i n g for Electron Microscopy. Polak, J. M. and Varn- dell, I. M., Eds. Elsevier, Amsterdam. pp. 29-36.

D o ~ t h , K. and Breuner G. 1982. A method for the study of undecalcsed bones and teeth with at- tached soft tissues. The Siige-SchilT (sawing and grindingl technique. J. Oral Pathol. 11: 318-326.

Fujie, M., Kuroiwa, H. , Kawano, S. and Kuroiwa, T. 1993. Studies on the behavior of organelles and their nucleoids in the root apical meristem of Arabidopsis thaliana (L.) Col. Planta 189: 149-158.

Fujie, M., Kuroiwa, H., Kawano, S. and Kuroiwa, T. 1995. Changes in cytoplasmic rRNA density in the root apical meristem observed by high reso- lution in situ hybridization. Cytologia 60: 815-821.

Gottlieb, D. I. and Glaser, L. 1975. A novel assay of neuronal cell adhesion. Biochem. Biophys. Res. Commun. 63: 815-821.

Bio

tech

His

toch

em D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

ufts

Uni

vers

ity o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.

48 B iotechn ic & H istochem istry

Hageman, R. and Scktider, M.-B. 1989. The cyto- logical basis of the plastid inheritance in an- giosperms. Protoplasma 152: 57-64.

Ikeda, T., Nomura, S., Yamaguchi, A. Suda, T. and Yoshiki, S. 1992. In situ hybridization of bone ma- trix proteins in undecalcified adult rat bone sec- tions. J. Histochem. Cytochem. 40: 1079- 1088.

Kuroiwa, T. Miyamura, S., Kawano, S., Hizume, M., Toh-e, A,, Miyakawa, I. and Sando, N. 1986. Cytological characterization of NOR in the biva- lent of Saccharomyces cerevisiae. Exp. Cell. Res. 165: 199-206.

Kuroiwa, T. 199 1. The replication, differentiation, and inheritance of plastids with emphasis on the con- cept of organelle nuclei. Int. Rev. Cytol. 128: 1-62.

Kuroiwa, T., Fujie, M., Mita, T. and Kuroiwa, H. 1991. Application of embedding of samples in Technovit 7100 resin to observations of small amounts of DNA in cellular organelles associ- ated with cytoplasmic inheritance. Appl. Fluo- resc. Technol. 3: 23-25.

Kuroiwa, T., Fujie, M. and Kuroiwa. H. 1992. Stud- ies on the behavior of mitochondrial DNA. Syn- thesis of mitochondrial DNA occurs actively in a specific region just above the quiescent center in the root meristem of Pelargonium wm&. J. Cell Sci. 101: 48-93.

McCormick, S. 1993. Male gametophyte develop- ment.

Morgensen, H. L. 1996. The hows and whys of cyto- plasmic inheritance in seed plants. Am. J. Bot. 83: 38-04.

Reynolds. T. L. and Raghaven, V. 1982. An autora- diographic study of RNA synthesis during mat-

uration and germination of pollen grains of Hyoscyamus niger. Protoplasma 1 1 1 : 177-1 88.

Russell, S. D. 1985. Preferential fertilization in Plumbago: ultrastructural evidence for gamete- level recognition in an angiosperm. Proc. Natl. Acad. Sci. USA 82: 6129-6132.

Russell, S. D., Strout, G. W., Stramski, A. K., Mis- lan, T. W., Thompson, R A. and Schoemann, L. M. 1996. Microgametogenesis in Plumbago zey- hica (Plumbaginaceae) . 1. Descriptive cytology and three-dimensional organization. Am. J. Bot. 83: 1435-1453.

Saito, C., Fujie, M., Sakai, A., Kuroiwa, H. and Kuroiwa, T. 1997. Extent of the condensation of nuclear chromatin and the localization of RNA during pollen development in Nicotiana tabacum

Saito, C., Fujie, M., Sakai, A., Kuroiwa, H. and Kuroiwa, T. 1998a. Detection and quantifkation of rRNA by h@-resolution in situ hybridization in pollen grains. J. Plant Res. 11 1: 45-52.

Saito, C., Nagata, N., Sakai, A., Kuroiwa, H. and Kuroiwa, T. 1998b. Distribution pattern of plas- tids and mitochondria during microgametogen- esis in Plumbago auriculata. Plant Cell Physiol. 39: suppl. s46.

Sodmergen, Chen, G. H, Hu, 2. M., Guo, F. L. and Guan, X. L. 1995. Male gametophyte develop- ment in Plumbago zeylanica cytoplasm localiza- tion and cell determination in the early generative cell. Protoplasma 186: 79-86.

Tanaka, I. 1997. Differentiation of generative and vegetative cells in angiosperm pollen. Sex. Plant Reprod. 10: 1-7.

cyt010gla 62: 121-132.

Bio

tech

His

toch

em D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

ufts

Uni

vers

ity o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.