Starch synthase 4 is essential for coordination of starch granule formation with chloroplast division during Arabidopsis leaf expansion

Arabidopsis thaliana mutants lacking the SS4 isoform of starch synthase have strongly reduced numbers of starch granules per chloroplast, suggesting that SS4 is necessary for the normal generation of starch granules. To establish whether it plays a direct role in this process, we investigated the circumstances in which granules are formed in ss4 mutants. Starch granule numbers and distribution and the accumulation of starch synthase substrates and products were investigated during ss4 leaf development, and in ss4 mutants carrying mutations or transgenes that affect starch turnover or chloroplast volume. We found that immature ss4 leaves have no starch granules, but accumulate high concentrations of the starch synthase substrate ADPglucose. Granule numbers are partially restored by elevating the capacity for glucan synthesis (via expression of bacterial glycogen synthase) or by increasing the volumes of individual chloroplasts (via introduction of arc mutations). However, these granules are abnormal in distribution, size and shape. SS4 is an essential component of a mechanism that coordinates granule formation with chloroplast division during leaf expansion and determines the abundance and the flattened, discoid shape of leaf starch granules.

Oligonucleotide primers used in this study.
Primers used in the selection of arc mutants were as described in Crumpton-Taylor et al. (2012). Gene-specific sequences for Gateway® primers are in bold.
Starch synthase activities and chlorophyll contents of ss4 mutants.
Measurements were made on mature, non-flowering rosettes. Starch synthase activity was measured on freshly prepared leaf extracts by a modification of the resin method of Jenner et al. (1994).
Values are means ± SE of measurements on five rosettes for each genotype.  Table S3.
ADPglucose contents of mature and immature leaves of ss4 mutants.
ADPglucose contents of two different ss4 T-DNA insertion mutants, and of mature and young leaves from the same, mature, non-flowering rosettes. Values are means ± SD of measurements on five plants for each genotype for the batch of plants in the upper part of the

Fig. S1
Characterization of ss4 mutants. (a) Immunoblot of an SDS-polyacrylamide gel of ss4-3 mutant leaves and mature and young leaves of wild-type plants. M is g y g yp p molecular mass markers, masses indicated in kDa. Each of the remaining lanes contains 40 µl extract from a separate plant All extracts contained the same mg contains 40 µl extract from a separate plant. All extracts contained the same mg tissue per ml extraction medium. Equal loadings were confirmed by SDS-PAGE f ll d b C i I t tBl TM t i i f th t t ( t h ) Th followed by Coomassie InstantBlue TM staining of the same extracts (not shown). The blot was probed with purified SS4 antibodies, raised against a unique 14-amino-acid peptide. (b) Changes in starch, sucrose glucose and fructose over 24 h in mature, non-flowering rosettes grown under 12 h light, 12 h dark (darkness from 12 to 24 h g g g , ( after dawn). Values are means of measurements on six rosettes, bars are SE. Open circles wild-type (Col) Filled triangles ss4 1 Filled squares ss4 3 (c) Fresh circles, wild-type (Col). Filled triangles, ss4-1.  were grown under sterile conditions in square polystyrene Petri dishes (100 x 15 (f) (e) (d) mm) on 0.7% (w/v) agar containing the nutrients described in Haughn and g Somerville (1986) with 1% (w/v) sucrose added Seeds were surface-sterilised and  , and the merged g p g ( ) p y ( ) g images (right). (c) GS and endogenous starch synthase activities detected by non-denaturing PAGE Soluble extracts of leaves were loaded onto native 7 5% polyacrylamide gels containing PAGE. Soluble extracts of leaves were loaded onto native 7.5% polyacrylamide gels containing 0.3% (w/v) glycogen. For wild-type (Col) and ss3ss4, lanes contain material from 100 µg fresh weight The three lanes for the GS expressing line contain material from 25 50 and 100 µg weight. The three lanes for the GS-expressing line contain material from 25, 50 and 100 µg fresh weight. After electrophoresis and incubation in a medium containing 1 mM ADPG, activities were detected by iodine staining. GS and endogenous SS1 and SS3 activities are marked. (d) Chain-length distribution of starch isolated from wild-type plants (black symbols) and ss3ss4 lines expressing Agrobacterium GS (grey symbols) Left GS-2-2 Right and ss3ss4 lines expressing Agrobacterium GS (grey symbols). Left, GS 2 2. Right, GS-5-3. Starch was debranched with isoamylase and pullulanase and analysed by HPAEC PAD Peak areas ere s mmed and the areas of indi id al peaks e pressed as HPAEC-PAD. Peak areas were summed and the areas of individual peaks expressed as a percentage of the total. Values are means ± SE of measurements on four (wild-type, GS-5-3) or three (GS-2-2) independent samples.    Values are means of measurements on six to eight plants. Bars are SE. Values with the same letter are not statistically significantly different (p >0.05, Student's t-test). (i) Daily starch y g y (p ) ( ) y turnover in genotypes shown in (h). Turnover is end-of-day minus end-of-night starch contents calculated from (h) (j) Starch synthase activities detected by non-denaturing PAGE contents, calculated from (h). (j) Starch synthase activities detected by non denaturing PAGE. Soluble extracts of leaves (equivalent fresh weight in each lane) were loaded onto native 7 5% polyacrylamide gels containing 0 3% (w/v) glycogen For all genotypes except ss4 the 7.5% polyacrylamide gels containing 0.3% (w/v) glycogen. For all genotypes except ss4, the two lanes contain extracts from separate plants. After electrophoresis and incubation in a medium containing 1 mM ADPG, activities were detected by iodine staining. Note that band pattern and intensity is essentially the same in all genotypes. p y y g y p   Figure 7(a) were derived. The first harvest (day 0) was immediately prior to dexamethasone (dex) application, 10 h into a 12-h light period. Dex was applied daily at this time point for the next ten days. Each harvest was immediately prior to dex   Note that the pattern of leaf t h t t ith t t l f i i il t th t f ild t l t th starch content with respect to leaf age is similar to that of wild-type plants rather than ss4 plants (compare with Fig. 2b, c). (b) Immunoblot of an SDSpolyacrylamide gel of extracts of rosettes of a wild-type plant and five lines of ss4 plants expressing GFP-tagged SS4. The positions of molecular mass markers are p p g gg p indicated at the right, in kDa. Each lane contains 60 µl extract from a separate plant All lanes are from the same gel and blots were developed together and for plant. All lanes are from the same gel, and blots were developed together and for the same length of time. All extracts contained the same mg tissue per ml extraction di Th bl t b d ith ifi d SS4 tib di i d i t i medium. The blot was probed with purified SS4 antibodies, raised against a unique 14-amino-acid peptide. SS4 appears as two bands in wild-type extracts (see Fig.  S1a and Roldán et al., 2007). In the transgenic lines SS4 has reduced mobility because of its GFP tag (~30 kDa). No band is present at this position in g ( ) p p immunoblots of extracts of ss4 mutants (see Figs 7, S1a). (c) Starch contents of wild-type leaves ss4 mutant leaves and leaves of two lines of ss4 plants wild-type leaves, ss4 mutant leaves, and leaves of two lines of ss4 plants expressing GFP-tagged SS4 at the end of the day (black) and the end of the night ( hit ) L f i th t d l f 16 th ld t l f V l f (white). Leaf one is the youngest and leaf 16 the oldest leaf. Values are means of measurements on four plants. Error bars are SE.