免费文献传递   相关文献

Cytochemical Localization of Pectinase Activity in Pollen Mother Cells of Tobacco During Meiotic ProphaseⅠand Its Relation to the Formation of Secondary Plasmodesmata and Cytoplasmic Channels


Pectinase activity was localized at the ultrastructural level in pollen mother cells of tobacco (Nicotiana tabacum L.) during meiotic prophaseⅠto elucidate its role in the biogenesis of secondary plasmodesma (sPD) and cytoplasmic channel (CC). At the leptotene stage the enzyme was mainly present in the cisternae of smooth endoplasmic reticulum (SER) and their derived vesicles, but absent in the Golgi body and Golgi vesicles. Later at the zygotene stage, when sPDs and CCs were actively formed, strong pectinase activity was observed not only in the SER cisternae and their derived vesicles but also in the cell wall, especially in the vicinity of or within both simple and branched plasmodesmata, notably along the middle lamellae, which also characterized the sites of CCs being formed. The presence of exocytotic vesicles containing reaction products suggests that pectinase shares the same excretive pathway as that used by cellulase for its delivery into the wall, i.e. in active form via smooth endoplasmic reticulum (ER) and its derived vesicles by exocytosis. In combination with cellulase, pectinase also promotes the secondary formation of plasmodesmata and CCs by specifically digesting the pectin in middle lamella.


全 文 :Received 22 Jun. 2004 Accepted 30 Sept. 2004
* Author for correspondence. Tel (Fax): +86 (0)931 8912561; E-mail: .
http://www.chineseplantscience.com
Acta Botanica Sinica
植 物 学 报 2004, 46 (12): 1443-1453
Cytochemical Localization of Pectinase Activity in Pollen Mother Cells of
Tobacco During Meiotic ProphaseⅠ and Its Relation to the Formation
of Secondary Plasmodesmata and Cytoplasmic Channels
YU Chun-Hong, GUO Guang-Qin, NIE Xiu-Wan (NIEH Hsiu-Wan), ZHENG Guo-Chang (CHENG Kuo-Chang)*
(Cell Biology Laboratory, Lanzhou University, Lanzhou 730000, China)
Abstract: Pectinase activity was localized at the ultrastructural level in pollen mother cells of tobacco
(Nicotiana tabacum L.) during meiotic prophaseⅠto elucidate its role in the biogenesis of secondary
plasmodesma (sPD) and cytoplasmic channel (CC). At the leptotene stage the enzyme was mainly present
in the cisternae of smooth endoplasmic reticulum (SER) and their derived vesicles, but absent in the Golgi
body and Golgi vesicles. Later at the zygotene stage, when sPDs and CCs were actively formed, strong
pectinase activity was observed not only in the SER cisternae and their derived vesicles but also in the cell
wall, especially in the vicinity of or within both simple and branched plasmodesmata, notably along the
middle lamellae, which also characterized the sites of CCs being formed. The presence of exocytotic
vesicles containing reaction products suggests that pectinase shares the same excretive pathway as that
used by cellulase for its delivery into the wall, i.e. in active form via smooth endoplasmic reticulum (ER) and
its derived vesicles by exocytosis. In combination with cellulase, pectinase also promotes the secondary
formation of plasmodesmata and CCs by specifically digesting the pectin in middle lamella.
Key words: cytochemical localization; secondary plasmodesmata; cytoplasmic channel; pectinase; pollen
mother cells; tobacco
In plants, plasmodesmata are classified as “primary” and
“secondary” according to their origin. The former are formed
during cytokinesis, while the latter generate de novo across
the established cell wall (Jones, 1976; Kollmann et al., 1985;
Kollmann and Glockmann, 1985; 1991; Roberds and Lucas,
1990; Ding et al., 1992; Lucas et al., 1993). Cytoplasmic
channel is another form of intercellular connection much
larger than plasmodesma (PD) (Gates, 1908; Weiling, 1965;
Bisalpufra and Stein, 1966; Heslop-Harrison, 1966; Baquar
and Husain, 1969; Owen and Makaroff, 1995; Wang et al.,
2002). While the mechanism responsible for primary PD
formation is generally clear, those for the secondary PD
(sPD) and the cytoplasmic channel (CC) still remain largely
as enigmas, for which a few hypotheses that lack solid
experimental support have been presented (Jones, 1976;
Kollmann and Glockmann, 1991; Lucas et al., 1993; Ding
and Lucas, 1996; Kragler et al., 1998; Ehler and Kollmann,
2001). In Jones’ model (1976), sPD formation is quite direct
and simple, arising from wall perforation by enzymatic
hydrolysis. However, as Jones (1976) noted: “the mecha-
nism by which a site for secondary perforation of the wall is
selected by the cell(s), and how the wall digesting enzymes
are localized there is unknown.” The model proposed by
Kollmann and Glockmann (1991) involves more complex
processes of wall thinning and removal followed by its
rebuilding through concerting activities of Golgi vesicles
and endoplasmic reticulum (ER) branches, which is espe-
cially suitable to explain the sPD with complex branching
morphology in some cases.
The formation of sPDs and CCs between developing
PMCs has been studied in great detail (Cheng et al., 1987;
Wang et al., 1998). They begin to form at leptotene, peak at
synizesis (especially for CCs), and are sharply occluded at
pachytene (Nie et al., 1984). Early investigations based on
ultrastructural observations and acid phosphatase local-
ization favor Jone’s model and suggest that the hydrolases
needed for this process are likely moved to the wall either
after their release out of ER cisternae into the cytoplasm or
by ER-derived lysosomal-like vesicles via exocytosis (Cheng
et al., 1987). The second picture was directly substantiated
by the subsequent localization of cellulase activity during
sPD and CC formation (Wang et al., 1998). The whole pro-
cess appeared to consist of several successive steps in
time: the enzyme was possibly synthesized on the rough
ER and stored in active form in the smooth endoplasmic
reticulum (SER), which often enlarged in some parts, giving
rise to SER-vesicles that delivered their cellulase content
into the cell wall by exocytosis, leading to the formation of
sPD and CC by perforating the cell wall through hydrolysis,
which also requires the co-action of pectinase to digest the
Acta Botanica Sinica 植物学报 Vol.46 No.12 20041444
middle lamellae (Jones, 1976; Lucas et al., 1993; Wang
et al., 1998). Here we report cytochemical localization of
pectinase activity in pollen mother cells of tobacco
(Nicotiana tabacum) during meiotic prophase Ⅰ and dis-
cuss its relation to the formation of secondary plasmodes-
mata and cytoplasmic channels
1 Materials and Methods
Cytochemical localization of pectinase activity was per-
formed according to the procedure of Allen and Nessler
(1984). Tobacco (Nicotiana tabacum L.) was grown in the
Botanic Garden at Lanzhou University. Anthers were col-
lected and grouped according to their size (Table 1). The
exact developmental stage of each group of anthers was
determined by light microscopic examination. Anthers at
different developmental stages were cut into 0.5-1.0 mm
sized pieces and fixed in Karnovsky’s (1965) fixative (a mix-
ture of 1% glutaraldehyde and 4% paraformaldehyde in
0.05 mol/L phosphate buffer, pH 7.2) at 0 ℃ for 5 h. Speci-
mens were then rinsed with more than 10 changes of phos-
phate buffer and stored in the same buffer overnight at 0
℃. They were incubated at 25 ℃ for 30 min in 0.1 mol/L
sodium acetate buffer (pH 5.0) containing 0.5% pectin and
transferred to hot Benedict’s solution with 1.73% cupric
sulfate, 17.3% sodium citrate and 10% sodium carbonate,
and boiled for 10 min. Control tissues were incubated in
acetate buffer without pectin, or were boiled for 10 min
prior to incubation with pectin. Omission of Benedict’s re-
agent from the reaction mixture was used as further con-
trols to check any artifacts arising from this reagent. After
incubation, tissues were washed several times with dis-
tilled water and post-fixed in 1% OsO4 overnight at 0 ℃,
followed by another wash with distilled water. Subsequent
dehydration, Epon812 embedding, ultrathin sectioning, and
double staining with lead and uranium were carried out as
before (Cheng et al., 1987). The stained sections were ex-
amined with the Phillip-400T transmission electron
microscope. Semi-thin sections (2 µm thick) cut with micro-
tome were stained with Toluidine blue O and observed un-
der OLYMPUS microscope.
2 Results
2.1 Characteristics of pectinase activity reaction
products
In accordance with previous reports (Allen and Nessler,
1984; Wang, 1998), the deposits of pectinase reaction
product (RP) (cuprous oxide) seen in this experiment ap-
peared as small (ca. 20 nm) electron dense crystalline de-
posits in linear form (Figs.1-5, 7-10, 12-14, 16-18 and 23-
28) and could not be detected in the controls (Figs.11, 15).
In the present experiment, pectinase activity and its pat-
tern of localization within endothecium, middle layer, tape-
tum and pollen mother cells (PMCs) changed during devel-
opment (Table 2). The following is a detailed description of
pectinase activity during meiotic prophaseⅠ.
2.2 Dynamic distribution of pectinase activity in pollen
mother cells
At the leptotene stage, a pollen mother cell has a big
central nucleus and rich cytoplasm with abundant cell
organelles, such as mitochondria, ER and Golgi bodies
(Figs.1-6, 9-10). The PMCs at this stage were separated
Table 1 Size (mm) of flower bud and anther of Nicotiana
tabacum at different stages of development
Bud Anther Stage
3.0-3.5 1.3 Leptotene
4.0-4.5 1.5 Zygotene
5.0-6.0 1.6-1.8 Synizesis
6.5-7.0 1.9-2.1 Pachytene
Table 2 Subcellular distribution of pectinase activity within anther wall (endothecium, middle layer and tapetum) and PMCs of
Nicotiana tabacum at different stages of meiotic prophase Ⅰ
Organelle
Anther wall PMCs
Endothecium Middle layer Tapetum Leptotene Zygotene Synizesis Pachytene
ER ++ + ++ + - -
ER vesicle ++ ++ + ++ + - -
Golgi body - - - - - - -
Golgi vesicle - - - - - - -
Mitochondrion - - - - - - -
Nucleus - - - - - - -
Cell wall + ++ + + ++ ± -
Middle lamellae ± ++ ++ + ++ ± -
Plasmodesma ± + ++
CC + ++ ±
-, no activity; ±, weak activity; +, easily detectable activity; ++, high activity; CC, cytoplasmic channel; ER, endoplasmic reticulum; PMCs,
pollen mother cells.
YU Chun-Hong et al.: Cytochemical Localization of Pectinase Activity in Pollen Mother Cells of Tobacco During Meiotic Prophase
Ⅰ and Its Relation to the Formation of Secondary Plasmodesmata and Cytoplasmic Channels 1445
Figs.1-6. Pollen mother cells of tobacco at leptotene stage: pectin and Benedict’s solution staining. 1-5. Pectinase reaction products
(arrows) are present in the cisternae of smooth endoplasmic reticulum and their derived vesicles. The cisternae with pectinase reaction
products (RPs) are often distended and sometimes oriented to the cell wall where a few deposits of RPs (arrow) are present (Fig.5). Note
the absence of reaction product in Golgi body and Golgi vesicles (Figs.2, 3, 6). Bar = 0.5 mm (1-3, 5, 6), 1 mm (Fig.4). Abbreviations: GB,
Golgi body; GV, Golgi vesicle; SER, smooth endoplasmic reticulum.
Acta Botanica Sinica 植物学报 Vol.46 No.12 20041446
Figs.7-10. 7-9. ER-derived vesicles with pectinase reaction products (arrows) are near cell wall (Fig.7), being fused with the
plasmalemma (Fig.8) or are releasing their pectinase into the wall (Fig.9) by exocytosis. 10. Pectinase RPs (arrow) are present in the cell
wall. Abbreviations: CW, cell wall; ER, endoplasmic reticulum; ERV, endoplasmic reticulum-derived vesicle; N, nucleus; RPs, reaction
products. Bar = 0.5 mm (Figs.1-3, 5-10), 1 mm (Fig.4).
Figs.11-16. Pollen mother cells of tobacco at zygotene stage: pectin and Benedict’s solution staining. 11, 15. Negative controls,
showing H- (Fig.11) and Y-shaped (Fig.15) plasmodesmata. No reaction products are observed. 12. Large amount of reaction products
(RPs) is localized in the cell wall. Note the small RP deposit at the site between the two neighboring plasmodesmata (arrows). 13. Strong
pectinase activity (arrow) is concentrated in the middle lamellae between two adjacent plasmodesmata, presumably responsible for H-
shaped plasmodesma formation. 14. Pectinase reaction products (arrow) are present in the two simple plasmodesmata that appear to be
in the process of their formation. 16. A forming Y-shaped plasmodesmata with strong pectinase activity (arrows) and closely associated
smooth endoplasmic reticulum. Abbreviations: N, nucleus; PD, plasmodesmata; RP, reaction product; SER, smooth endoplasmic
reticulum. Bar = 0.33 mm (Figs.11, 15), 0.5 mm (Figs.12-14, 16).

YU Chun-Hong et al.: Cytochemical Localization of Pectinase Activity in Pollen Mother Cells of Tobacco During Meiotic Prophase
Ⅰ and Its Relation to the Formation of Secondary Plasmodesmata and Cytoplasmic Channels 1447
Acta Botanica Sinica 植物学报 Vol.46 No.12 20041448
Figs. 17-22. 17, 18. Tapetal cells: pectin and Benedict’s solution staining, the reaction products (RPs) (arrows) are mainly located in
the middle lamellae at the sites of cytoplasmic channels in their process of formation. 19. Semi-thin section showing cytomixis (long
arrow) between tapetal cells when pollen mother cells are at pachytene stage. Note that the cell wall (short arrow) and nuclei of tapetal
cells (two nuclei within a cell) are different from those of pollen mother cells. 20. Intercellular migration of nuclear materials via
cytoplasmic channel between tapetal cells (arrow). Little RP is found. 21-22. Cytomixis between pollen mother cells at synizesis; little
RP is present. Abbreviations: CC, cytoplasmic channel; CH, chromatin; CW, cell wall; N, nuclei; Nu, nucleolus; PD, plasmodesmata;
PMC, pollen mother cells; RP, reaction product; TC, tapetal cells. Bar = 0.5 mm (Figs.17, 18, 20-22), 2.5 mm (Fig.19).
YU Chun-Hong et al.: Cytochemical Localization of Pectinase Activity in Pollen Mother Cells of Tobacco During Meiotic Prophase
Ⅰ and Its Relation to the Formation of Secondary Plasmodesmata and Cytoplasmic Channels 1449
Figs. 23-28. Cytochemical localization of pectinase activity in the cells of middle layers (Figs. 23-27) and endothecium (Fig.28) of
tobacco anther, which is similar to that observed in pollen mother cells at the lepotene-zygotene stage (Figs.1-10). Abbreviations: CW,
cell wall; ERV, endoplasmic reticulum-derived vesicle; GB, Golgi body; GV, Golgi vesicle; N, nucleus; PD, plasmodesmata; SER, smooth
endoplasmic reticulum; V, vacuole.
by thin cell walls (Fig.10). As reported in the previous study
on cytochemical localization of cellulase (Wang et al., 1998),
some parts of SER cisternae contained strong pectinase
activity and distended (Figs.1-3, 5), which subsequently
gave rise to irregular-shaped ER-vesicles by budding (Figs.
3, 4). It was found that SER cisternae could be directly
oriented to the cell wall (Fig.5), possibly preparing to re-
lease their enzyme content into the cell wall, as shown in
Figs.7-9 via exocytosis. A few deposits of reaction prod-
uct could be seen in the cell wall (Fig.10). We failed to
observe enzyme activity in Golgi bodies and Golgi vesicles
(Figs. 2, 3, 6).
At zygotene stage, ER and ER-vesicles (ERVs) with pec-
tinase reaction products were still rich in their cytoplasm.
The prominent feature of pectinase distribution at this stage
was that strong activity existed in the wall of the pollen
mother cells, especially along the middle lamellae and in the
vicinity of plasmodesmata. RPs were also observed in the
middle lamellae between two adjacent simple plasmodes-
mata (Figs.12, 13), and within simple (Fig.14) and Y-shaped
Acta Botanica Sinica 植物学报 Vol.46 No.12 20041450
plasmodesmata undergoing formation (Fig.16). Again as
expected, such localization patterns of pectinase activity
were absent in the controls (Figs.11, 15).
2.3 Dynamic distribution of pectinase in anther wall
cells
Pectinase reaction products could be observed in the
wall between tapetal cells. Of particular relevance, they were
often present between two very neighboring intercellular
connections (Fig.18), or in the middle lamellae at the form-
ing sites of CC (Figs.17, 18). The amount of reaction depos-
its was greatly reduced in the walls of both tapetal cells (at
pachytene Fig. 19) and PMCs (at synizesis) after cytoplas-
mic channels reached their full sizes, which were usually
filled with migrating cytoplasm or chromatin substances
across cell walls (Figs.19-22).
In the cells of other layers (epidermis, endothecium and
middle layer) of anther wall, a similar distribution pattern of
pectinase reaction products (RPs) to that in leptotene PMCs
was observed (Table 2): pectinase activity was often ob-
served in some distended SER cisternae and their derived
vesicles, from which the enzyme seemed to be released into
the wall by exocytosis, where the RPs were found to be
present in a line along the middle lamellae (Figs. 23-28).
Like PMCs and tapetal cells, CCs were also often observed
between those anther wall cells, although their frequency
is much lower than that observed between PMCs.
3 Discussion
3.1 Pectinase reaction products indicate the presence of
pectinase activity
Ultrastructural localization of pectinase activity was first
developed by Allen and Nessler (1984) through modifica-
tion of the method reported by Bal (1974) for the localiza-
tion of cellulase activity. Its principle is similar to that un-
derlying the cellulase localization (Wang et al., 1998) and
based on the fact that reducing sugars (galacturonic acid)
liberated from the enzymatic hydrolysis of exogenous pec-
tin can react with Benedict’s reagent, reducing cupric salts
to cuprous oxide precipitates that appear as small electron-
dense crystalline deposits in linear form under electron
microscope. Since in our present experiment such deposits
are only produced in the samples incubated in complete
reaction medium at normal reaction conditions, but not in
the negative controls, their presence must indicate pecti-
nase activity. The same conclusion was also obtained by
Li et al. (2004).
3.2 Delivery of pectinase to its action sites
Like cellulose, enzymatic hydrolysis of pectin is also
considered to be a critical step in sPD formation (Ding and
Lucas, 1996). However, the exact cellular mechanisms re-
sponsible for this process remain elusive. Ding and Lucas
(1996) speculated that the localized degradation of the
middle lamella and adjacent cell wall material could be
achieved by several possible mechanisms. First, active en-
zymes may be directly transported, via the primary PD, into
the middle lamella. Secondly, inactive form of enzymes may
be secreted into the wall via vesicle fusion with the plasma
membrane, and triggering molecules may be transported,
via primary PD, into the middle lamella to activate the en-
zymes locally. Thirdly, the inactive enzymes may have been
deposited in the walls by fusing Golgi vesicles during cell
plate formation, and the triggering molecules may enter the
middle lamella through PDs to activate the enzymes when
the modification process begins. The presence of active
wall-degrading enzymes (pectinase in this study and cellu-
lase in our previous report (Wang et al., 1998)) within ER
cisternae and their derived vesicles, and their subsequent
movement toward the wall to secrete these enzymes in ac-
tive form by exocytosis, does not favor the last two of the
above speculated mechanisms; rather, our results suggest
a concept of de novo synthesis and transport of enzymes
in active form during or just before the process of sPD
formation. Although studied in two separate experiments,
the same picture found for the two enzymes for their trans-
port to the action sites in the wall suggests that these two
major wall-degrading enzymes may be co-delivered by the
same cellular apparatus (SER and its derived vesicles), which
made it possible for the cell to coordinate their actions in
wall degradation, i.e. at exactly the same place and time,
which is necessary for sPD and CC formation correctly.
Double labeling of the two enzymes by immuno-cytochemi-
cal method should be used to directly check this possibility.
It remains to be seen if this is also the case for other wall
degrading enzymes such as hemicellulase. The absence of
pectinase and cellulase activity in Golgi bodies and their
vesicles indicated that this organelle played no role in this
process. In addition, as shown in Figs.14, 16, the enzymes
might also be transported, via the forming plasmodesmata,
into the middle lamella. So we proposed two alternative
pathways for delivering pectinase to the middle lamellae
during the development of sPD: the active enzymes may be
either directly transported into the middle lamella through
forming plasmodesma, or diffused into middle lamella after
being released into the cell wall via ER-vesicles/plasma
membrane fusion.
3.3 Possible roles of pectinase in secondary plasmodes-
mata and cytoplasmic channel formation
Parallel changes between the occurrence of pectinase
YU Chun-Hong et al.: Cytochemical Localization of Pectinase Activity in Pollen Mother Cells of Tobacco During Meiotic Prophase
Ⅰ and Its Relation to the Formation of Secondary Plasmodesmata and Cytoplasmic Channels 1451
and the process of sPD and CC formation, as has been
reported previously for cellulase (Wang et al., 1998), indi-
cate that the enzyme plays an important role in this process.
At leptotene stage, most pectinase were localized in the
SER and ER-vesicles but scarce in the cell wall. At zygo-
tene stage, when large number of sPD and CC being formed,
strong pectinase activity could be observed in the vicinity
of or within PDs, as well as in the middle lamellae at the
sites of CC being formed, which was subsequently reduced
greatly at synizesis (PMCs) and pachytene (tapetal cells)
stages when CCs were completely formed to mediate
cytomixis.
In the model proposed by Kollmann and Glockmann
(1991) to explain the de novo formation of sPDs between
heterografts of Vicia faba on Helianthus annuus, they
guessed that cell wall degrading enzymes might be only
involved in the initial stages of wall loosening/thinning
and removal at the site of the graft union. Subsequent en-
trapment of ER cisternae during wall rebuilding by Golgi
vesicles activity resulted in sPD formation, which could
satisfactorily explain the complex branched morphology of
the sPD in their case. In PMCs, however, we failed to ob-
serve such Golgi-mediated wall rebuilding process, indi-
cating that it did not participate in the formation of sPDs
and CC. Rather, both our present and previous studies
(Wang et al., 1998) favor a more direct process of enzy-
matic perforation across the wall for sPD and CC formation,
as proposed by Jone (1976). Based on cytochemical local-
ization of cellulase activity, Wang et al. (1998) suggested
that, if the ER cisternae providing the enzymes entered the
perforations being formed during local wall digestion and
were finally “trapped” in them by the surrounding wall, the
formed connections would be sPD with appressed ER, if
not, the simple PD without ER would be formed; if the re-
leased enzymes were very strong or remained active for a
long time, the perforations formed will be large, leading to
the formation of CC (Wang et al., 1998). Our present results
indicated that pectinase might be co-released with cellu-
lase from the same ER-vesicle (ERV) into the wall, where, in
a closely cooperative manner with cellulase activity, it se-
lectively degraded pectin in the local middle lamellae at the
forming sites of sPDs and CCs. Its localization within the
“Y”- and “H”-shaped sPDs (Figs.13, 16) clearly showed
that in PMCs such complex sPDs could be formed by rather
directly simple way of enzymatic perforations that did not
require the process of wall rebuilding. Moreover, removal
of pectin in middle lamellae by pectinase resulted in lateral
fusion of two neighboring PDs, producing “H”-shaped PD
as shown in Figs.12 and 13.
3.4 Cytoplasm channel formation is associated with
cytomixis
CCs are distinct from PDs (Bisalpufra and Stein, 1966;
Robards, 1975) in their sizes, structures and functions. They
are non-branched pores lined by plasmalemma, and gener-
ally 200-600 nm in diameter, and can reach up to 3 500 nm
at its maximum, which are much larger than those of PDs (30
-50 nm) (Weiling, 1965; Bisalpufra and Stein, 1966; Heslop-
Harrison, 1966; Baquar and Husain, 1969, Wang et al., 2002).
In both PMCs (Figs.21, 22) and tapetal cells (Figs.19, 20),
cytomixis could be observed once the cytoplasmic chan-
nels had been formed. Cytomixis between tapetal cells has
also been observed in lily under light microscope (Cheng
et al., 1964). In the present investigation, we found that
cytomixis appeared to closely follow the fully formation of
CC but not precede it. These results suggested that, as
huge intercellular connections much larger than PDs, CCs
were closely related to the phenomenon of cell-to-cell mi-
gration of cytoplasmic organelle (such as mitochondria,
plastids and vesicles, etc.) and nuclear material (Bisalpufra
and Stein, 1966; Heslop-Harrison, 1966; Baquar and Husain,
1969; Cheng et al., 1975; Wang et al., 2002).
Acknowledgements: We thank Prof. WANG Xin-Yu for
his inspiration, Mr. PAN You-Fu and Miss NIU Yu-Hong
for their assistance in photography. We are grateful to
Mr. LU Wei for his assistance in electron microscopic
observation, and Miss WANG Dai-Si for her help in ultra-
thin sectioning.
References:
Allen R D, Nessler C L. 1984. Cytochemical localization of pec-
tinase activity in Laticifers of Nerium oleander L.
Protoplasma, 119: 74-78.
Bal A K. 1974. Cellulase. Hayat M A. Electron Microscopy of
Enzymes. Vol. 3. New York: Van Nostrand Reihold. 68-76.
Baquar S R, Husain S A. 1969. Cytoplasmic channels and chro-
matin migration in the meiocytes of Arnebia hispidissima (Sieb)
DC. Ann Bot, 33: 821-831.
Bisalpufra T, Stein J R. 1966. The development of cytoplasmic
bridges in Volvox aureus. Can J Bot, 44: 1697-1702.
Cheng K-C, Nieh H-W , Yang C-L , Wang Y-X . 1964. Chromatin
extrusion in pollen mother cells in relation to mechanical injury
and fixing fluids during microsporogenesis. Acta Bot Sin, 12:
289-308. (in Chinese with English abstract)
Cheng K-C, Nieh H-W , Yang C-L , Wang Y-X, Zhou Y-S, Chen J-
S . 1975. Light and electron microscopical observation on
cytomixis and the study of its relation to variation and evolution.
Acta Botanica Sinica 植物学报 Vol.46 No.12 20041452
Acta Bot Sin, 17: 60-69. (in Chinese with English abstract)
Cheng K-C , Nieh H-W , Chen S-W , Jian L-C , Sun L-H , Sun D-
L. 1987. Studies on the secondary formation of plasmodesmata
between the pollen mother cells of lily before cytomixis. Acta
Biol Exp Sin, 20: 1-11. (in Chinese with English abstract)
Ding B, Haudenshield J S, Hull R J, Wolf S, Beachy K N, Lucas W
J. 1992. Secondary plasmodesmata are specific sites of local-
ization of the tobacco mosaic virus movement protein in
transgenic tobacco plants. Plant Cell, 4: 915-928.
Ding B, Lucas W J. 1996. Secondary plasmodesmata: biogenesis,
special functions and evolution. Smallwood M, Knox J P,
Bowles D J. Membranes: Specialized Functions in Plants.
Oxford: BIOS Scientific Publisher. 489-506.
Ehlers K, Kollmann R. 2001. Primary and secondary
plasmodesmata: structure, and origin, and functioning.
Protoplasma, 216: 1-30.
Gates R R. 1908. A study of reduction in Oenothera rubrinervis.
Bot Gaz, 46: 1-34.
Heslop-Harrison J. 1966. Cytoplasm connexions between an-
giosperm meiocytes. Ann Bot, 30: 221-230.
Jones M G K. 1976. The origin and development of
plasmodesmata. Gunning B E S, Robards A W. Intercellular
Communication in Plants: studies on Plasmodesmata. Berlin:
Springer-Verlag. 81-105.
Karnovsky M J. 1965. A formaldehyde-glutaraldehyde fixation
of high osmolality for use in electron microscopy. J Cell Biol,
27: 137.
Kollmann R, Glockmann C. 1985. Studies in graft unions Ⅰ:
plasmodesmata between cells of plants belonging to different
unrelated taxa. Protoplasma, 124: 224-235.
Kollmann R, Glockmann C. 1991. Studies on graft unions Ⅲ: on
the mechanism of secondary formation of plasmodesmata at
the graft interface. Protoplasma, 165: 71-85.
Kollmann R, Yang S, Glockmann C. 1985. Studies on graft unions
Ⅱ: continuous and half plasmodesmata in different regions of
the graft interface. Protoplasma, 126: 19-29.
Kragler F, Lucas W J, Monzer J. 1998. Plasmodesmata: dynamics, (Managing editor: WANG Wei)
domains and patterning. Ann Bot, 81: 1-10.
Li A-M, Wang Y-R, Wu H . 2004. Cytochemical localization of
pectinase: the cytochemical evidence for resin ducts formed by
schizogeny in Pinus massoniana. Acta Bot Sin, 46: 443-450.
Lucas W J, Ding B, van der Schoot C. 1993. Plasmodesmata and
the supracellular nature of plants. New Phytol, 125: 435-476.
Nie X-W, Wang Y-X , Cheng K-C . 1984. Transmission and scan-
ning electron microscopic observation on the plasmodesmal
channels between the pollen mother cells of lily. Acta Bot Sin,
26: 34-37. (in Chinese with English abstract)
Owen H A, Makaroff C A. 1995. Ultrastructure of microsporo-
genesis and microgametogenesis in Arabidopsis thaliana (L.)
Heynh ecotype Wassilewskija (Brassicaceae). Protoplasma,
185: 7-21.
Robards A W. 1975. Plasmodesmata. Ann Rev Plant Physiol, 26:
13-29.
Robards A W, Lucas W J. 1990. Plasmodesmata. Annu Rev Plant
Physiol Plant Mol Biol, 41: 369-419.
Wang X Y, Guo G Q, Nie X W, Zheng G C. 1998. Cytochemical
localization of cellulase activity in pollen mother cells of David
lily during meiotic prophase Ⅰ and its relation to secondary
formation of plasmodesmata. Protoplasma, 204: 128-138.
Wang X Y, Nie X W, Guo G Q, Pan Y F, Zheng G C (Cheng K C).
2002. Ultrastructural characterization of the cytoplasmic chan-
nel formation between pollen mother cells of David lily.
Caryologia, 55: 161-169.
Weiling F. 1965. Light and electron microscopical observation on
cytomixis and its possible relation to photocytosis. Planta,
67: 182-212.