3. DiscussionPrevious studies on th

3. Discussion
Previous studies on the SlRab11a antisense lines clearly
demonstrated a delayed fruit softening phenotype [25]. This
phenotypewas repeated in the lines used for this paper. However, a
mechanism for this phenotype has not so far been elucidated. Early
studies, using northern blot analysis, suggested that the target Rab
gene showed increased expression during fruit ripening [25] and as
a result, previous attempts to identify a mechanism for the delayed
softening focused on the ripening phase of fruit development. In
this paper specific primers to SlRab11a were used and the quantitative
PCR data presented here showed levels of SlRab11a mRNA
were highest at around 15 dpa and then dropped during fruit
development, reaching a minimum at the breaker stage. The difference
between these results and the previous northern blot results
is probably due to those early results being influenced by
other members of the large family of highly homologous RabA
genes in plants [22]. There was no increased expression of SlRab11a
associated with fruit ripening. The observation that SlRab11a
expression was highest during early fruit development, suggests
that the changes in the antisense fruit that result in the firm
phenotype might occur mainly during development of the fruit,
rather than during ripening. This therefore raised the question of
whether the G4 antisense affected the deposition of the cell wall
during the expansion and subsequent development periods prior to
ripening of the fruit in such a way that it responded differently to
hydrolytic enzymes during ripening or was intrinsically stronger. If
so, the enzymatic differences compared with the wild type, shown
by Lu et al. [25] might be only part of the explanation for the
firmness of the fruit. Itwas therefore interesting to study the nature
of the cell wall during the developmental phase and to compare
wild type with antisense lines.
Of particular interest, was the cell wall composition in wild type
fruit. In this study, pectin was found to be the dominant polymer
in tomato fruit at breaker, which is consistent with established
literature [26,27]. However, these studies focused on cell wall
composition during the post breaker stage. Data in this study show
that at 15 dpa pectin levels represent the smallest proportion of the
cell wall polymers. From 15 dpa onwards pectin proportionally
increased to become the dominant cell wall polymer. In respect
to other polymers, cellulose remained relatively consistent levels
throughout, while hemicellulose was proportionally reduced leading
up to breaker. There are several possible ways of interpreting
these data. One possibility is that the levels of cellulose and hemicelluloses
are decreasing or remaining constant,however this seems
unlikely. A second possibility is that all three components, cellulose,
hemicelluloses and pectin, are being deposited in the walls during
development but that the rate of pectin deposition is increasing.
There is also the third possibility that turnover of thewall material is
also being differentially regulated during fruit development. It is
interesting to note in this context that pectic polysaccharides,
xyloglucans and proteins may be remobilised out of the cell wall
after deposition [28e34]. Finally, it is not impossible to conceive that
these data occur through a combination of the second and third
mechanisms. It is interesting to note that the period between 15 and
35 dpa is associated with expansion of the fruit. It is thus possible to
postulate that the change in cell wall composition is in some way
directly related to the “switch” from cell division, which is predominant,
during the first 2e3 weeks after pollination, to that of
expansion of the fruit over the following 3e5 weeks [32]. Although
showing a similar trend in cell wall compositional changes during
development fruit from the G4 antisense line demonstrated differences
in pectin levels from the ACþþ control.
It has long been postulated that pectin is made and then
esterified while in transit between the Golgi and cell wall [33] and