Alex C Ogbonna, Luciano Rogerio Braatz de Andrade, Ismail Y Rabbi, Lukas A Mueller, Eder Jorge de Oliveira, Guillaume J Bauchet

Plant Journal; 2021 Feb;105(3):754-770.  doi: 10.1111/tpj.15071. 

Abstract

Manihot esculenta (cassava) is a root crop originating from South America that is a major staple in the tropics, including in marginal environments. This study focused on South American and African germplasm and investigated the genetic architecture of hydrogen cyanide (HCN), a major component of root quality. HCN, representing total cyanogenic glucosides, is a plant defense component against herbivory but is also toxic for human consumption. We genotyped 3354 landraces and modern breeding lines originating from 26 Brazilian states and 1389 individuals were phenotypically characterized across multi-year trials for HCN. All plant material was subjected to high-density genotyping using genotyping by sequencing. We performed genome-wide association mapping to characterize the genetic architecture and gene mapping of HCN. Field experiments revealed strong broad- and narrow-sense trait heritability (0.82 and 0.41, respectively). Two major loci were identified, encoding for an ATPase and a MATE protein, and contributing up to 7 and 30% of the HCN concentration in roots, respectively. We developed diagnostic markers for breeding applications, validated trait architecture consistency in African germplasm and investigated further evidence for the domestication of sweet and bitter cassava. Fine genomic characterization revealed: (i) the major role played by vacuolar transporters in regulating HCN content; (ii) the co-domestication of sweet and bitter cassava major alleles are dependent upon geographical zone; and (iii) the major loci allele for high HCN in M. esculenta Crantz seems to originate from its ancestor, M. esculenta subsp. flabellifolia. Taken together, these findings expand our insights into cyanogenic glucosides in cassava roots and its glycosylated derivatives in plants.

See: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7898387/

Figure 2. Genome‐wide association study (GWAS) of HCN for Latin American (LA) germplasm. (a) Manhattan plot from a mixed linear model (MLM‐LOCO) with the chromosome on which the candidate SNP is located excluded from calculating the genetic relationship matrix (GRM). The Bonferroni significance threshold is shown in red. A quantile–quantile plot is inserted to demonstrate the observed and expected −log₁₀ P for HCN. The red circle indicates the candidate SNP. (b) locuszoom plot showing the HCN chromosome 16‐associated region (−log₁₀ P) around the candidate gene. The two rows above the plot show genomic coverage at the locus, with each vertical tick representing direct genotyping from GBS and HapMap single‐nucleotide polymorphisms (SNPs). Each circle represents an SNP, with the color of the circle indicating the correlation between that SNP and the candidate SNP at the locus (purple). Light‐blue lines indicate the estimated recombination rate (hot spots) in GBS. The middle panel shows 36 single point mutations (red are deleterious) between the region spanning ccMATE1 and ccMATE2. The bottom panel shows the annotated genes at each locus in cassava genome version 6.1. The red and black rectangles indicate Manes.16G007900 and Manes.16G008000, respectively, with a Pearson correlation coefficient of 0.96 (r ²) between both genes. The scheme presents the gene model, with the position of the associated SNP within the 4th exon indicated. (c and d) Box plots showing candidate SNP effects for HCN between each genotype class for the top markers, S14_6050078 and S16_773999, respectively.