Phloem unloading after anthesis is symplastic
In open flowers (late stage 13), we demonstrated that
symplastic unloading is switched on again. Strongest
fluorescence was visible in an area which was defined as
phloem unloading area in an earlier publication (Stadler et
al. 2005a). A weaker signal was visible in all cells of the
integuments all around the embryo sac (Fig. 2c, d). The
switch-on of symplastic unloading was independent of
fertilization. We can conclude that increasing sink strength
due to energy demand of developing embryos is not the
reason for the observed massive influx of phloem-mobile
fluorescent tracers into developing seeds. What is the
anatomical basis for the switch-on of symplastic phloem
unloading? Are there new plasmodesmata formed into
existing sieve elements or are new sieve elements formed?
232 D. Werner et al.
The formation of new sieve elements in line with the
induction of symplastic unloading had also been described
in the roots of Arabidopsis after nematode infection (Hoth
et al. 2005). However, between stage 12 and stage 13, the
ovules are only 10–15 h, and a comparison of pSUC2-
tmGFP9 ovules of both stages indicated that no new SE/CC
complexes had formed (Fig. 2a, h). In root tips, the sieve
elements which mediate symplastic unloading are protophloem
sieve elements (Stadler et al. 2005b). Protophloem
sieve elements in dicotyledons are not accompanied by
companion cells. The cells which were highly labeled by
MP17-GFP in fertilized ovules rather resembled metaphloem
sieve elements in that they did not contain nuclei
and were located right next to small cells which contain big
nuclei as is typical for SE/CC complexes.
Our data suggest that the onset of symplastic unloading
is mediated by a new formation of plasmodesmata or also a
structural change of existing plasmodesmata in sieve
elements rather than a new formation of sieve elements.
However, it is not clear how sieve elements which do not
contain nuclei can form plasmodesmata or even change the
structure of existing plasmodesmata. The weak fluorescence
of the companion cell in the stage 13 ovule shown in
Fig. 2h (arrow) could be interpreted as a sign for ongoing
maturation. Maturation of the adjacent sieve element might
be accompanied by the widening of the plasmodesmata of
the last sieve element and development of plasmodesmata
to surrounding cells. It would be interesting to study the
presence or absence of callose in stage 12 and stage 13
ovules by staining with aniline blue. In addition, the
presence and structure of plasmodesmata at the postphloem
pathway should be studied by TEM analyses
during various developmental stages in ovules.
Our data altogether showed that the phloem unloading
mode in ovules of A. thaliana is a highly regulated
process. It is likely tightly connected with changes in the
architecture of plasmodesmata. Components which are
involved in the formation of new plasmodesmata or
factors which regulate the opening of existing plasmodesmata,
for example, by regulation of callose-degrading
enzymes, are not known so far. It is a challenge for the
future to identify genes which are involved in these
processes