RNA as a signalling molecule - Intercellular Communication in Plants - page 69

10-15 cells on both sides of the veins (Himber et al. , 2003) (Plate 3.2C-E). These
results indicated that the extent of silencing movement outside of the vasculature
was directly dependent upon the activity of SDE1, which was presumably involved
in production of secondary signal molecules. They also suggested that the primary
signal molecules generated from the veins could move over 10-15 cells in the ab-
sence of relay-amplification by SDE1. Similar experiments carried out with the
sde3 mutant showed that movement of GFP silencing outside of the veins was more
extensive than in the sde1 mutant, but less pronounced than in a wt background,
indicating that SDE3 also contributes to the extensive cell-to-cell movement, but
to a lesser degree than SDE1 (Himber et al. , 2003) (see Plate 3.3A-D, following
page 146). This was in accordance with the previous demonstration that, unlike in
the sde1 mutant, initiation of sense-transgene silencing in plants deficient for the
SDE3 RNA helicase was not completely abolished, presumably because of residual
SDE1 activity (Dalmay et al. , 2000, 2001).
Cell-to cell movement of silencing in mature leaves. We saw in a previous section
that spontaneous triggering in Class-II Nia plants is first manifested by the emergence
of chlorotic spots on mature leaves, which are also observed upon bombardment of
Nia -homologous DNA in single cells (Palauqui et al. , 1996; Palauqui & Balzergue,
1999). Likewise, biolistic delivery of GFP DNA in cells of mature leaves of GFP
transgenic N. benthamiana leads to the rapid appearance of red fluorescent (i.e.
silenced) foci that are similar in size to the chlorotic spots of the Nia plants (Voinnet
et al. , 1998; Klahre et al. , 2002). Interestingly, in both systems, the diameter of
these clusters of silenced cells does not increase further throughout time, indicating
that silencing moves over a limited number of cells from its site of initiation in
mature leaves, a phenomenon referred to as localized movement of silencing .To
gain insight into this phenomenon, the agroinfiltration procedure was exploited in
the GFP plants (Himber et al. , 2003). In addition to the GFP sequences that trigger
silencing, the T-DNA used in those experiments carried a GUS transgene, whose
open reading frame (ORF) was interrupted by an intron. The presence of this intron
meant that GUS could be produced only within plant cells, upon T-DNA transfer
(see Plate 3.4A, following page 146).
The infiltrated patch rapidly became uniformly red fluorescent, indicating GFP
silencing had taken place. GUS histochemical blue staining was used to image
precisely the cells that had received the T-DNA, and therefore the cells in which
silencing had been triggered. GFP silencing clearly extended outside of the stained
patch, forming a border of red fluorescent tissue that followed exactly the shape of
the infiltrated patch itself (Plate 3.4B-D). Microscopic inspection and cell counting
revealed that the width of this red border was remarkably constant and affected 10-15
cells, which was strikingly reminiscent of the extent of silencing movement observed
near the veins of the sde1 Arabidopsis mutant. In further experiments, a cork borer
was used to remove most of the inner part of an infiltrated area immediately after
delivery of the Agrobacterium . This treatment did not modify the development and
extent of localized silencing at the edge of the infiltrated patch, indicating that the