Sorghum is a promising alternative to maize for bioenergy production in Europe; however, its use is currently limited by poor adaptation to low temperatures during and after germination. We collected multi-trait phenotype data under optimal and suboptimal temperatures in a genetically diverse recombinant inbred line (RIL) mapping population showing contrasting segregation patterns for pre-and post-emergence chilling tolerance. Germination, emergence, seedling development, root architecture and seedling survival were assessed in two different seedlots. Emergence and root establishment were found to be the key determinants of development and survival under chilling stress. Highly interactive epistatic quantitative trait loci (QTL) hotspots, including a previously unknown QTL on Sb06 with a significant effect on prolonged chilling survival, were found to regulate different physiological mechanisms contributing to maintenance of growth and development despite the chilling temperatures. The major QTL regions harbour promising candidate genes with known roles in abiotic stress tolerance. Identification of loci in the QTL hotspot regions conferring maintenance of cell division and growth under early chilling stress represents a promising step towards breeding for successful establishment of sorghum in temperate climates.

Promising genome regions for improving cold tolerance of sorghum were identified on chromosomes SBI-01, SBI-03, SBI-07, and SBI-10. Chlorophyll fluorescence had no major effect on growth rates at low temperatures. Developing fast growing sorghum seedlings is an important breeding goal for temperate climates since low springtime temperatures are resulting in a prolonged juvenile development. The adaptation of sorghum to tropical and subtropical highlands gives hint for certain genetic variation. The goals of the present study were to detect marker-trait associations for leaf and dry matter growth rate and for chlorophyll fluorescence and content (SPAD) in relation to temperature. A diversity set comprising 194 genotypes was tested in eight controlled environments with temperatures ranging from 9.4 to 20.8 °C. Significant marker-trait associations (p < 0.05) were identified for each individual temperature regime and on the parameters of regression analyses describing the responses of growth or chlorophyll related traits to temperatures. The diversity set was fingerprinted with 171 diversity array technology (DArT) and 31 simple-sequence repeat (SSR) markers. SSRs were used to analyze the population structure while association studies were performed on DArT markers. Promising marker-trait associations for growth rates in relation to temperature were detected on chromosomes SBI-01, SBI-03, SBI-07, and SBI-10. Many promising loci were also significantly associated to the results obtained in individual low-temperature environments. Marker-trait associations for chlorophyll content and fluorescence did occasionally co-locate to those for growth during juvenile development but there was no evidence supporting our hypothesis that seedling growth at low temperatures is largely influenced by SPAD or fluorescence.

Among the major limitations for cultivating biomass sorghum in temperate regions is low temperature in spring that results in low and non-uniform emergence. The adaptation of sorghum to tropical and subtropical highlands gives hint of genetic variation in cold tolerance during emergence. The objective of the present study was to detect marker-trait associations for parameters describing the emergence process under different temperature regimes. A diversity set comprising 194 genotypes was tested in nine controlled environments with temperatures ranging from 9.4 to 19.9 °C. The genotypes were fingerprinted with 171 DArT markers. A piecewise linear regression model carried out on cumulative emergence was used to estimate genotype mean performance across environments and to carry out stability analysis on the parameters of the regression model. Base temperature (T (b)) and thermal time required for emergence (E (TS)) were determined based on median time to emergence data. Identified QTL positions were compared to marker-trait associations for final emergence percentages under low (FEP(cold)) and normal (FEP(normal)) temperatures. QTL for mean final emergence percentage (FEP), FEP(cold) and FEP(normal,) T (b) and E (TS) were detected on SBI-01. Other QTL-rich regions were located on SBI-03, SBI-04, SBI-06, SBI-08, and SBI-09. Marker-trait associations for T (b) and E (TS) co-localized to QTL for the across environment stability of FEP and the median time to emergence or emergence rate, respectively. We conclude that genome regions on six chromosomes highly influencing cold tolerance during emergence are promising for regional association studies and for the development of stable markers for marker-assisted selection.

Improved cold tolerance during the juvenile phase is a major breeding goal to develop new sorghum cultivars suitable as an alternative energy crop in temperate regions. The objectives of this study were to identify marker‐trait associations for cold tolerance in a sorghum diversity panel fingerprinted with 2620 single nucleotide polymorphism (SNP) markers and to detect quantitative trait loci (QTL) in two F2:3 populations. Traits of interest were dry matter growth rate (DMGR), leaf appearance rate (LAR), chlorophyll content (SPAD) and chlorophyll fluorescence (Fv′/Fm′ and ФPSII) in relation to temperature. The association panel comprised 194 genotypes, while the F2:3 populations consisted of 80 and 92 genotypes. All populations were tested under controlled conditions in a minimum of four temperature regimes ranging from 9.4°C to 20.8°C. QTL were identified for means across environments and regression parameters describing temperature effects. Several marker‐trait associations for mean (m) DMGR, base temperature (Tb) of SPAD and ФPSII and temperature effect on LAR were validated by QTL detected in population 1 or 2. Promising QTL regions were found on chromosomes SBI‐01, SBI‐02, SBI‐03, SBI‐04, SBI‐06 and SBI‐09, among them genomic regions where candidate genes involved in the C‐repeat binding pathway or encoding cold‐shock proteins are located.