Fig 1.
Schematic depiction of binary vector constructs used in this study.
Genes were cloned in frame with GBSSI transit peptide to allow amyloplast targeting and were driven by GBSSI promoter for tuber-specific expression. CBM20 and DSP represent the carbohydrate-binding module 20 domain and a dual specificity phosphatase of laforin protein. RB and LB stand for right and left borders, respectively. SBD, LK, Kan and 3’NOS stand for starch binding domain of cyclodextrin glycosyltransferase from Bacillus circulans, linker, kanamycin resistant gene and NOS terminator, respectively. The arrow represents cleavage site of the transit peptide. FLAG and HIS are two tags for protein quantification and HpaI, SpeI, XhoI, BglII and BamHI are restriction enzyme sites.
Fig 2.
Characterization of transgenic plants.
(a) Distribution of the individual transformants over the classes of the (engineered) laforin expression. The qRT-PCR analysis was performed in triplicate on all transformants, which are 27, 26 and 30 lines for D, CD and SD series, respectively. D, CD and SD represent DSP, CBM20-DSP and SBD-DSP transformants, respectively. N, L, M and H stand for none, low, medium and high expressors. (b) Accumulation level of SBD-DSP (SD) in transformants. Protein levels were determined using Western dot blot analysis with an anti-SBD antibody. The number above each dot stands for the different lines, while UT and ‘+’ represent negative and positive control, respectively. The intensity of dots shows the various protein levels. The corresponding gene expression level of each line obtained from qRT-PCR is indicated between brackets. For 90% of lines a good correlation between gene expression level and protein accumulation level was found.
Fig 3.
Light micrographs and SEM analyses showing starch granules morphology of control UT (A-C) and modified starches (D-I).
Starch granules were stained with a 20× diluted Lugol solution for light microscopy (A, D and G). Two different morphologies were observed in the modified starches from each series. Based on the colour of stained granules, the starches were classified into two categories: blue-stained group (D-F) and red-stained group (G-I).
Table 1.
Summary of different starch characteristics determined for the representative modified starches and the control UT.
Fig 4.
The suppression of GBSSI in transformants with red-stained granules.
(a) The correlation between amylose content and the relative expression of granule-bound starch synthase I (GBSSI). The qRT-PCR analysis was performed on control UT and 5 random-selected transgenic tubers from each series, containing transformants with 8 red-stained starches and 7 blue-stained starches. (b) Western blot analysis of GBSSI abundance in starches from different transformants. The corresponding amylose content of each line is shown between brackets.
Fig 5.
The phosphate content of starches from all transgenic series and control plants in both (a) Kardal and (b) amf backgrounds.
D, CD and SD stand for transgenic series containing DSP protein, the full-length laforin protein and SBD-DSP fusion protein, respectively. Amf represents the amylose-free potato background. All the analyses were performed in duplicate on transformants with (engineered) laforin expression, which are 24, 25, 28, 32, 25 and 29 lines for D, CD, SD, amfD, amfCD and amfSD series, respectively. Significant difference between each transgenic group and control was analysed using one-way ANOVA. Different letters (a-d) indicate statistically significant differences at p < 0.05.
Fig 6.
Boxplot presenting median granule size (d50) of starches from control UT and all transformants of each series.
All the analyses were performed in duplicate on all transformants, which are 24, 25 and 28 lines for D, CD and SD series, respectively. Significant difference between each transgenic group and control was analysed using one-way ANOVA. Different letters (a-c) indicate statistically significant differences at p < 0.05.
Fig 7.
Principal components biplot displaying the classification of starches from three transgenic series based on the starch characteristics.
Green vectors indicate the correlation between the different measured variables. P, phosphate content; To, onset temperature of gelatinization; Tp, peak temperature of gelatinization; Tc, conclusion temperature of gelatinization; MC, starch moisture content; ΔH, gelatinization enthalpy; AM, amylose content; d50, granule median size.
Fig 8.
Heat-map displaying the extent and direction of correlations (r) between starch compositional characters and starch physico-chemical properties in transgenic lines.
Correlations were statistically significant at r ≥ 0.22 and r ≤ -0.22. Blue colours show negative correlations and red colours show positive correlations. P, phosphate content; To, onset temperature of gelatinization; Tp, peak temperature of gelatinization; Tc, conclusion temperature of gelatinization; MC, starch moisture content; ΔH, gelatinization enthalpy; AM, amylose content; d50, granule median size.
Fig 9.
(a) The expression of the genes encoding key enzymes involved in starch metabolism and (b) correlation between starch phosphate content and gene expression and (c, d) between relative expressions of different genes.
qRT-PCR was performed on control UT and 5 random-selected transgenic tubers from each series, containing transformants with 8 red-stained starches and 7 blue-stained starches. The expression level of following genes were measured: glucan, water dikinase (GWD1), phosphoglucan, water dikinase (GWD3), starch phosphorylase (SP), β-amylase 9 (BAM9) and α-amylase 23 (AMY23). The values are expressed as the mean ± S.D. from three independent measurements. Statistical significances between each starch sample and the control determined by using t-test (*, p < 0.1; **, p < 0.01; ***, p < 0.001).