Agriculture Reference
In-Depth Information
cells located in the outmost periphery of the delivered area are sufficient to produce
the local signal (Himber
et al.
, 2003). Taking also into account the bombardment
data, these results suggest that short-distance movement of silencing over 10-15
adjacent cells can be elicited from one single cell. Based on the fact that those
experiments involved tissues in which SDE1 was presumably active, it is surprising
that silencing did not expand further by means of relay-amplification. This situation
is similarly encountered with the limited increase in size of primary viral lesions on
inoculated leaves, before systemic movement of the pathogen. With both silencing
and virus movement, the size restriction of primary foci is most likely explained by
the source status of the delivered organs (Santa Cruz
et al.
, 1998).
Molecular requirements for cell-to-cell movement of RNA silencing.
The molecu-
lar requirements for cell-to-cell movement of RNA silencing were studied in both
N. benthamiana
and
Arabidopsis
(Himber
et al.
, 2003). In the former system, co-
delivery of
Agrobacterium
cultures producing viral-encoded silencing suppressors
was used to alter the silencing phenotype normally elicited in the infiltrated patch by
GFP constructs. Molecular analysis of RNA extracted from such silenced patches
shows that the ectopic and transgenic GFP mRNAs were below detection limit and
that there was high accumulation of both 21 nt and 24 nt species of GFP siRNAs. In
addition, the border of red fluorescent tissue, characteristic of the cell-to-cell move-
ment of silencing from the infiltrated region to the surrounding cells, was readily
detected. However, when those experiments were repeated in the presence of the P19
suppressor protein of tomato bushy stunt virus (TBSV), silencing in the patch was
completely abolished: GFP siRNAs of both size classes were undetectable and the
levels of GFP and GFP mRNA were high. As expected from the absence of silenc-
ing within those tissues, the development of the red border was also abolished. Co-
treatments with the P1 suppressor of rice yellow mottle virus (RYMV), in contrast,
did not prevent the onset of GFP silencing within the infiltrated tissue, as indicated
by the complete lack of GFP mRNA. However, although the levels of 21 nt long GFP
siRNAs were high in those tissues, the 24 nt long siRNAs was below detection limit.
Interestingly, and despite the absence of the 24 nt siRNAs, cell-to-cell movement
of GFP silencing was as extensive as it was in leaves that had been infiltrated with
GFP alone (Himber
et al.
, 2003). These experiments indicated that the occurrence
of the 21 nt siRNA at the source of silencing initiation was sufficient for cell-to-cell
movement of silencing to take place at the edge of the patch. Note, however, that
these experiments did not rule out the possibility that, although not necessary, the
24 nt siRNAs may also be sufficient for short-distance spread to take place.
As discussed in Section 3.4.1.2, high levels of
Nia
transcripts were mandatory
for the development of co-suppression upon graft transmission of the
Nia
silencing
signal into the scions. To test if, similarly, the cell-to-cell movement of GFP silencing
from an infiltrated patch to the surrounding cells required the presence of the GFP
mRNA in recipient cells, a second set of agroinfiltration experiments was carried
out (Himber
et al.
, 2003). These experiments involved sequential infiltrations of
the GFP
Agrobacterium
strain in leaves of wt
N. benthamiana
such that the second
