Cas9-mediated mutagenesis of potato starch-branching enzymes generates a range of tuber starch phenotypes

Abstract

We investigated whether Cas9-mediated mutagenesis of starch-branching enzymes (SBEs) in tetraploid potatoes could generate tuber starches with a range of distinct properties. Constructs containing the Cas9 gene and sgRNAs targeting SBE1, SBE2 or both genes were introduced by Agrobacterium-mediated transformation or by PEG-mediated delivery into protoplasts. Outcomes included lines with mutations in all or only some of the homoeoalleles of SBE genes and lines in which homoeoalleles carried several different mutations. DNA delivery into protoplasts resulted in mutants with no detectable Cas9 gene, suggesting the absence of foreign DNA. Selected mutants with starch granule abnormalities had reductions in tuber SBE1 and/or SBE2 protein that were broadly in line with expectations from genotype analysis. Strong reduction in both SBE isoforms created an extreme starch phenotype, as reported previously for low-SBE potato tubers. HPLC-SEC and ¹ H NMR revealed a decrease in short amylopectin chains, an increase in long chains and a large reduction in branching frequency relative to wild-type starch. Mutants with strong reductions in SBE2 protein alone had near-normal amylopectin chain-length distributions and only small reductions in branching frequency. However, starch granule initiation was enormously increased: cells contained many granules of <4 μm and granules with multiple hila. Thus, large reductions in both SBEs reduce amylopectin branching during granule growth, whereas reduction in SBE2 alone primarily affects numbers of starch granule initiations. Our results demonstrate that Cas9-mediated mutagenesis of SBE genes has the potential to generate new, potentially valuable starch properties without integration of foreign DNA into the genome.

Keywords: Cas9-mediated mutagenesis; potato tuber; starch granule; starch structure; starch-branching enzyme.

Figure 1

Sequence alignments showing mutations in a, SBE1 and b, SBE2 in four plant lines. Exons are shaded in grey, and target sequences of sgRNAs 1, 2 and 9 are shown in bold letters. Long stretches of sequence are indicated by dots (‘…’); short nucleotide deletions by dashes, ‘‐’; longer deletions by continuous lines, ‘_’ and; single nucleotide insertions by dark shading. In‐frame deletions are boxed.

Figure 2

Occurrence of SBE1 and SBE2 proteins in extracts of tubers of wild‐type and mutant plants. The soluble fraction of extracts of mature tubers (samples from the mid‐cortical region half way along the tuber) were subjected to SDS‐PAGE on 7.5% acrylamide gels. Each lane contained 20 μg soluble protein. Gels were blotted onto PVDF membranes, probed with 1/1000 dilutions of SBE1 or SBE2 antiserum and developed using an anti‐rabbit antiserum and an enhanced chemiluminescence kit. Left lanes contain marker proteins; molecular masses are indicated in kDa. (a) Probed with SBE2 antiserum. (b) Probed with SBE1 antiserum. The occurrence of two bands of SBE1 protein has been observed previously (Jobling et al., 1999). Mutants 230‐51a and 230‐51b are derived from two independent shoots from the same callus and are genetically identical.

Figure 3

Starch granules from mutant and wild‐type plants viewed under normal (L) and polarized (R) light. (a) Wild‐type starch granules, (b) granules from mutant line 237/4‐44. (c) Granules from mutant line 230‐51, focussed to show huge numbers of tiny granules. (d) Granules from mutant line 212‐36. (e) Granules from mutant line 227‐25. Arrows indicate examples of cracking (red), nodular granules (yellow) and granules with multiple hila (green). Scale bars represent 50 μm.

Figure 4

Distributions of starch granule sizes between 4 and 100 μm. Results were obtained from analysis of purified tuber starch using a Multisizer 4e Coulter counter. Graphs show the % of total granules (y‐axis) of a particular diameter (x‐axis). (a) Wild‐type starch (blue) and three independently prepared samples of starch (yellow, grey, orange) from mutant line 230‐51. (b) Samples as for a, but showing two independently prepared starch samples from mutant line 227‐25. (c) Samples as for a, but showing three independently prepared starch samples from mutant line 212‐36. (d) Samples as for a, but showing three independently prepared starch samples from mutant line 237/4‐44.

Figure 5

Chain‐length distributions of debranched starch. Purified starch from mature tubers was debranched with isoamylase and subjected to HPLC‐SEC. The y‐axis shows weight distribution, based on the relationship between elution volume and hydrodynamic radius (logV _h) for pullulan standards. Note that the x‐axis is on a log scale. Each line represents an independently purified starch sample and is the mean of values from duplicates of that sample. (a) Starch from wild‐type (grey, dark blue and brown lines) and the sbe1 sbe2 mutant 237/4‐44 (pale blue, yellow and green lines). (b) Starch from a control line (black line), sbe2 mutant 230‐51 (green lines), sbe1 sbe2 mutant 212‐36 (orange lines) and sbe1 sbe2 mutant 227‐25 (blue lines).

Figure 6

Starch granules within isolated tuber cells. Tuber slices were incubated with 50 mm CDTA for 3 days; then, cells were separated by gentle pressure. (a) Wild type. (b) Mutant line 237/4‐44. (c, d) Mutant line 230‐51. (e) Mutant line 212‐36. (f) Mutant line 227‐25. Arrows illustrate examples of nodular granules (yellow) and clusters of tiny granules (blue). Scale bars represent 50 μm.