11 Starch Synthesis in the Potato Tuber 265breeding and transgenic approaches have been uti-
lized for modification of starch properties (specifi-
cally amylose, amylopectin, and phosphate content;
Sene et al. 2000, Kossmann and Lloyd 2000). How-
ever, in contrast to the situation described above for
yield, the use of natural mutants has been far more
prevalent in this instance.
Quantitative trait loci (QTL) analyses have recent-
ly been adopted for identifying the genetic factors
underlying starch structure. Recombinant inbred
lines were produced from parental maize lines of dif-
fering starch structure, and the amylose, amylopec-
tin, and water-soluble fractions were analyzed. The
loci linked to these traits were located on a genetic
map consisting of RFLP (restriction fragment length
polymorphism) markers (Sene et al. 2000). Using
this strategy, candidate genes were identified that
influenced starch structure. It is clear that application
of these techniques may facilitate future breeding
approaches to generate modified starch properties.
In addition to plant breeding approaches the use
of transgenesis has been of fundamental importance
for understanding and influencing starch structure
(see Table 11.2). Potato tubers with decreased ex-
pression of AGPase (Müller-Röber et al. 1992) had
decreased amylose contents, down to about 60% of
that found in wild type, and smaller starch granules
(Lloyd et al. 1999b). The reduction in amylose con-
tent is most likely due to the decreased levels of
ADP-glucose found in these plants, since this leads
to selective restriction of granule-bound versus solu-
ble starch synthases, the latter having a higher affin-
ity for ADPGlc (Frydman and Cardini 1967). Sim-
ilarly, when the plastidial adenylate supply was
altered by changes in the expression of the amylo-
plastidial ATP/ADP translocator, not only the starch
content, but also its structural properties were al-
tered (Tjaden et al. 1998, Geigenberger et al. 2001).
Tubers of ATP/ADP-translocator overexpressing
lines had higher levels of ADP-glucose and starch
with higher amylose content, whereas the opposite
was true for antisense lines. Furthermore, micro-
scopic examination of starch grains revealed that
their size in antisense tubers was considerably de-
creased (by 50%) in comparison with the wild type.
Antisensing granule-bound starch synthase I
(GBSSI) in potato tubers resulted in a starch that
was almost free of amylose (Kuipers et al. 1994).
This type of starch has improved paste clarity and
stability with potential applications in the food and
paper industries (Jobling 2004). In contrast to this,
the amylose content of potato starch could not be
increased above a value of 25.5% on overexpression
of GBSS, hinting that this enzyme is not limiting in
amylose production (Flipse et al. 1994, 1996). It has
additionally been demonstrated that amylose can
be completely replaced by a branched material that
exhibits properties in between those of amylose and
amylopectin following plastid-targeted expression
of bacterial glycogen synthases (Kortstee et al.
1998).
The roles of the other isoforms of starch synthase
(SS) are, if anything, less clear. A dramatic reduc-
tion in the expression of SSI in potato had absolute-
ly no consequences on starch structure (Kossmann
et al. 1999). The effects of modifying SSII were
observed in pea seeds mutated at the rug5 locus,
which encodes this protein. These mutants display a
wrinkled phenotype that is often caused by de-
creased embryo starch content. Starch granules in
these plants exhibit a striking alteration in morphol-
ogy in that they have a far more irregular shape and
exhibit a reduction in medium chain length glucans
coupled to an increase in short- and long-chain glu-
cans. Despite the dramatic changes observed in the
pea mutant, the down-regulation of SSII in the pota-
to tuber had only minor effects, including a 50%
reduction in the phosphate content of the starch and
a slight increase in short chain length glucans (Koss-
mann et al. 1999, Lloyd et al. 1999a). However, it is
worth noting that SSII only contributes 15% of the
total soluble SS activity within the tuber. Finally, a
third class of SS has also been identified and has
been reported to be the major form in potato tubers
(Marshall et al. 1996). When this isoform (SSIII)
was downregulated by the creation of transgenic
potato plants, deep fissures were observed in the
starch granule under the electron microscope, most
probably due to GBSS making longer glucan chains
in this background (Fulton et al. 2002). However,
the starch was not altered in its amylose content, nor
did it display major differences in chain length dis-
tribution of short chains within the amylopectin
(Marshall et al. 1996, Lloyd et al. 1999a). However,
dramatic changes were observed when the structure
of the side chains was studied, with an accumulation
of shorter chains observed in the transgenic lines
(Lloyd et al. 1999a). In addition to this, the phos-
phate content of the starch was doubled in this trans-
formant (Abel et al. 1996).