11 Starch Synthesis in the Potato Tuber 263most successful transgenic approaches have resulted
from the overexpression of a bacterial AGPase (Stark
et al. 1991) and the Arabidopsisplastidial ATP/ADP
translocator (Tjaden et al. 1998), and the antisense
inhibition of a plastidial adenylate kinase (Regierer
et al. 2002) in potato tubers.
The majority of previous attempts to improve the
starch yield of potato tubers concentrated on the
expression of a more efficient pathway of sucrose
degradation, consisting of a yeast invertase, a bacte-
rial glucokinase, and a sucrose phosphorylase (Tre-
thewey et al. 1998, 2001). However, although the
transgenics exhibited decreased levels of sucrose
and elevated hexose phosphates and 3-PGA with
respect to wild type, these attempts failed. Tubers of
these plants even contained less starch than the wild
type, but showed higher respiration rates. Recent
studies have shown that as a consequence of the
high rates of oxygen consumption, oxygen tensions
fall to almost zero within growing tubers of these
transformants, possibly as a consequence of the high
energy demand of the introduced pathway, and this
results in a dramatic decrease in the cellular energy
state (Bologa et al. 2003, Geigenberger 2003b). This
decrease is probably the major reason for the unex-
pected observation that starch synthesis decreases in
these lines. In general, oxygen can fall to very low
concentrations in developing sink organs like potato
tubers and seeds, even under normal environmental
conditions (Geigenberger 2003b, Vigeolas et al.
2003, van Dongen et al. 2004). The consequences of
these low internal oxygen concentrations for meta-
bolic events during storage product formation have
been ignored in metabolic engineering strategies.
Molecular approaches to increase internal oxygen
concentrations could provide a novel and exiting
route for crop improvement.
Another failed attempt at increasing tuber starch
accumulation was the overexpression of a heterolo-
gous sucrose transporter from spinach under the
control of the CaMV 35S promoter (Leggewie et al.
2003). The rationale behind this attempt was that it
would increase carbon partitioning toward the tuber;
however, in the absence of improved photosynthetic
efficiency this was not the case.
With respect to the plastidial pathway for starch
synthesis, much attention has been focused on AG-
Pase. Analysis of potato lines exhibiting different
levels of reduction of AGPase due to antisense inhi-
bition have been used to estimate flux control coeffi-
cients for starch synthesis of between 0.3 and 0.55
for this enzyme (Geigenberger et al. 1999a, Sweet-
love et al. 1999), showing that AGPase is colimiting
for starch accumulation in potato tubers. In addition
to having significantly reduced starch content in the
tubers, these lines also exhibit very high tuber su-
crose content, produce more but smaller tubers per
plant, and produce smaller starch grains than the
wild type (Müller-Röber et al. 1992; see also later
discussion and Table 11.2). To date one of the most
successful approaches for elevating starch accumu-
lation in tubers was that of Stark et al. (1991), who
overexpressed an unregulated bacterial AGPase.
This manipulation resulted in up to a 30% increase
in tuber starch content. However, it should be noted
that the expression of exactly the same enzyme
within a second potato cultivar did not significantly
affect starch levels (Sweetlove et al. 1996), indicat-
ing that these results might be highly context de-
pendent. Another more promising route to increase
starch yield would be to manipulate the regulatory
network leading to posttranslational redox activation
of AGPase (see Fig. 11.4). Based on future progress
in this field, direct strategies can be taken to modify
the regulatory components leading to redox regula-
tion of starch synthesis in potato tubers.
More recent studies have shown that the adeny-
late supply to the plastid is of fundamental impor-
tance to starch biosynthesis in potato tubers (Loef
et al. 2001, Tjaden et al. 1998). Overexpression of
the plastidial ATP/ADP translocator resulted in in-
creased tuber starch content, whereas antisense inhi-
bition of the same protein resulted in reduced starch
yield, modified tuber morphology, and altered starch
structure (Tjaden et al. 1998). Furthermore, incuba-
tion of tuber discs in adenine resulted in a consider-
able increase in cellular adenylate pool sizes and a
consequent increase in the rate of starch synthesis
(Loef et al. 2001). The enzyme adenylate kinase (EC
2.7.4.3) interconverts ATP and AMP into 2 ADP.
Because adenylate kinase is involved in maintaining
the levels of the various adenylates at equilibrium, it
represents an interesting target for modulating the
adenylate pools in plants. For this reason, a molecu-
lar approach was taken to downregulate the plastidi-
al isoform of this enzyme by the antisense technique
(Regierer et al. 2002). This manipulation led to a
substantial increase in the levels of all adenylate
pools (including ATP) and, most importantly, to a
record increase in tuber starch content up to 60%