Cassava shrunken-2 homolog MeAPL3 determines storage root starch and dry matter content and modulates storage root postharvest physiological deterioration

Abstract

Among the five cassava isoforms (MeAPL1-MeAPL5), MeAPL3 is responsible for determining storage root starch content. Degree of storage root postharvest physiological deterioration (PPD) is directly correlated with starch content. AGPase is heterotetramer composed of two small and two large subunits each coded by small gene families in higher plants. Studies in cassava (Manihot esculenta) identified and characterized five isoforms of Manihot esculenta ADP-glucose pyrophosphorylase large subunit (MeAPL1-MeAPL5) and employed virus induced gene silencing (VIGS) to show that MeAPL3 is the key isoform responsible for starch and dry matter accumulation in cassava storage roots. Silencing of MeAPL3 in cassava through stable transgenic lines resulted in plants displaying significant reduction in storage root starch and dry matter content (DMC) and induced a distinct phenotype associated with increased petiole/stem angle, resulting in a droopy leaf phenotype. Plants with reduced starch and DMC also displayed significantly reduced or no postharvest physiological deterioration (PPD) compared to controls and lines with high DMC and starch content. This provides strong evidence for direct relationships between starch/dry matter content and its role in PPD and canopy architecture traits in cassava.

Keywords: ADP-glucose pyrophophorylase; Cassava; Dry matter; MeAPL3; Postharvest physiological deterioration; Starch.

Fig. 1

Amino acid sequences of cassava and selected plant ADP-glucose pyrophophorylase large subunit (APL) conserved regions and expression patterns in major organs. a Amino acid sequences of catalytic site, b amino acid sequences of Glucose-1-Phospahte (Glu-1-P) binding region. The conserved amino acid residues [R/K/ or Q] at position 102 and [K/T] at position 112 (amino acid position based on AtAPL1 numbering) in catalytic sites in a and conserved [K] residue at position 271 in Glu-1-P binding site in b are shown in rectangular boxes. *indicates conserved amino acid residues across the 20 APL genes. Accession numbers cassava MeAPL1 (Manes.01G236700), MeAPL2 (Manes.03G182100), MeAPL3 (Manes.11G085500), MeAPL4 (Manes.15G025400), MeAPL5 (GenBank MN734216); Arabidopsis AtAPL1 (AT5G19220), AtAPL2 (AT1G27680), AtAPL3 (AT4G39210), AtAPL4 (AT2G21590); Sweet potato IbAPL1 (AFL55396); IbAPL2 (AFL55397), IbAPL3 (AFL55398), IbAPL4 (AFL55399); Potato StAPL1 (XP_006365120), StAPL2 (XP_015163910), StAPL3 (NP_001275395) and rice OsAPL1 (Q6AVT2), OsAPL2 (Q7G065), OsAPL3 (Q688T8) and OsAPL4 (Q0D7I3) (c) expression pattern of the five cassava APL genes in selected organs extracted from cassava expression atlas (Wilson et al. 2017) showing abudance of MeAPL3 in storage roots. Bars show SD (n = 2–3)

Fig. 2

Functionality of East African cassava mosaic virus (EACMV-K201) based virus induced gene silencing (VIGS) in cassava storage roots. a Dry matter, b total carotenoid content and c transverse slices of transgenic storage roots harvested from EC20 plants and wild-type TME 7S after challenge with GFP-VIGS and Patatin-VIGS construct. EC20-08 and EC20-11 transgenic events co-express crtB and DXS transgenes, each driven by the patatin promoter (Beyene et al. 2018). Silencing of crtB and DXS expression resulting from targeting the patatine promoter with Patatin-VIGS abolishes carotenoid accumulation in storage roots (b, c) and restores DMC to wild-type levels (a), GFP-VIGS has no target in the cassava genome and does not affect carotenoid accumulation or DMC. Bars show SD (n = 4–6)

Fig. 3

mRNA expression of cassava ADP-glucose pyrophosphorylase large subunit genes in VIGS challenged plants. Expression of a MeAPL1, b MeAPL2, c MeAPL3, d MeAPL4 and e MeAPL5 genes was determined by qRT-PCR in leaves and storage roots of TME 7S plants challenged with the respective VIGS-vectors for silencing and GFP-VIGS as control. Samples of storage roots and leaves were collected at harvest 12 weeks post-innoculation. Bars show SD of three biological replicates. Note graphs are not to the same scale

Fig. 4

Determination of the relative importance of cassava ADP-glucose pyrophosphorylase large subunit genes in DMC and carbohydrate accumulation in cassava storage roots using EACMV-K201 VIGS. a Dry matter, b starch, c glucose and d sucrose content of cassava storage roots after challenge with VIGS vectors targeting MeAPL1, MeAPL2, MeAPL3, MeAPL4 genes with GFP as control. CMD susceptible plants of cassava variety TME 7S were challenged with VIGS vectors and plants harvested 10 weeks later. Bars are SD of five biological replicates per line; ** and *** stand for significant difference, respectively, at p ≤ 0.01 and p ≤ 0.001. Student's t‐test compared to the GFP-VIGS control

Fig. 5

Effect of silencing of cassava MeAPL3 on storage root production. a Cassava storage root number, b storage root yield, c images of storage roots from different transgenic lines, empty-vector and wild-type TME 7S controls. Transgenic lines were generated using p8384 an empty vector control (p8384) and p8388 that expresses MeAPL3 (sense). Plant were harvested 17 weeks after planting in the greenhouse, assessed for storage root number and yield. Bars are SD of five biological replicates per line; * and ** stand for significant difference, respectively, at p ≤ 0.05 and p ≤ 0.01. Student's t‐test compared to the wild-type TME 7S control

Fig. 6

Effect of silencing MeAPL3 on DMC and plant phenotype. a DMC b leaf angle in selected transgenic lines and wild-type control c plant shows partial petiole drooping of lower leaves in transgenic line 8388-01 and severe drooping in 8388-04 compared to petiole angles of wild-type TME 7S. Transgenic lines were generated using an empty vector control (p8384) and p8388 that express MeAPL3 (sense). Plants were harvested from the greenhouse and assessed for DMC 17 weeks after planting. Average petiole angle of the middle canopy per plant was determined 10 weeks after planting by measuring leaf angle from position 8th, 9th and 10th counted downwards from the top-most leaf. Four plants per line was evaluated. Bars in (a) show SD of five biological replicates and Bars in (b) show SD of four biological replicates per lines. *, ** and *** stand for significant difference, respectively, at p ≤ 0.05, p ≤ 0.01 and p ≤ 0.001. Student's t‐test compared to the wild-type TME 7S control

Fig. 7

Effect of silencing MeAPL3 on starch and soluble sugar content. a starch b sucrose, c glucose, d raffinose content and e starch staining of cross-sections of cassava storage roots from selected 8388 transgenic lines, empty vector control (8384) and wild-type TME 7S. Plants were harvested from the greenhouse for starch determination 17 weeks after planting as described in Fig. 5. Bars show SD of three biological replicates

Fig. 8

Effect of silencing MeAPL3 on cassava postharvest physiological deterioration (PPD). a transverse slices of storage roots of transgenic lines showing low or no PPD (left panel) and transgenic and wild-type with high PPD (right panel) b PPD score of storage roots. Transgenic lines were generated using an empty vector control (p8384) and p8388 that expresses MeAPL3 (sense). Storage roots were harvested from the greenhouse 17 weeks after planting, washed and dried with paper towels, and stored at 25 ^oC in paper bags for three days. Roots cut transversely at proximal, middle and distal ends for PPD evaluation. PPD at day 0 is not shown as no PPD was detected at that time. Bars are SD of three biological replicates