Characterisation of the cold acclimation process in spring and winter cereals

Wado, May Saleh (2020) Characterisation of the cold acclimation process in spring and winter cereals. PhD thesis, University of Glasgow.

Abstract

Vast tracts of viable agricultural land are located in the northern latitudes of Eurasia and North America. They experience very high seasonal productivity due to long warm days and plentiful rainfall remain, but are largely uncultivated due to late spring or early autumn frosts. Wild grasses survive these frost events and this has prompted intensive investigation into how to improve the cold tolerance of domesticated small grained cereals. This thesis presents work to investigate the relative importance of two environmental factors (i.e. non-freezing low night temperatures and night length (or photoperiod)) on cold acclimation in two diploid species from economically important crops; i.e. rye (Secale cereal, L.) and barley (Hordeum vulgare, L.).
A winter (California) and spring (Belgravia) cultivar of barley, and two cold-tolerant cultivars of rye (Forage and Kapitan), were grown to the three week stage over short warm nights (8hr at 16C), and then exposed to long nights alone (15hr at 16C), cool nights alone (8hr at 4C), or both. The controls were raised under conditions of 8hr 16C nights. The degree of cold tolerance conferred by these treatments was subsequently assessed 1, 2, 3, and 4 weeks post acclimation. Cold hardiness was assessed by measuring electrolyte leakage (cell lysis) from the leaf tissues, after exposure to a range of freezing temperatures, using modification of the Lethal Temperature 50 (LT50) method (temperature that produces 50% cell lysis). These experiments reported a major difference in the cold acclimation processes of rye and barley. Consistent with reports in the literature; in both rye cultivars, cool night temperatures alone (34-38% improvement) and long night periods alone (<11% improvement) triggered a significant cold acclimation response. However, when applied together the response was strongly synergistic, i.e. more-than-additive (up to 16.2C improvement, 100%). In contrast, with barley long nights alone (12-18% improvement) and cool night temperatures alone (80-88% improvement) there was conferred frost resistance, but when applied together (up to 11.2C improvement, 100%) there was limited evidence of synergism, as the effect of temperature and night length appeared to be additive.
The transcriptional events that underpin the cold acclimation process in Arabidopsis are well characterised, and detailed temperature-sensing and photoperiod-sensing pathways have been proposed along with the crosstalk between two networks. In addition, it is well established that parts of these Arabidopsis pathways involve the action of the phytohormone Abscisic acid (ABA), whilst others do not. In particular, the process of Vernalization, as triggered by AtVRN1, and the ICE-CBF-COR regulon has been implicated in conferring at least 30% of full cold tolerance on Arabidopsis. Homologues of several VRN and ICE-CBF-COR have been identified in the Triticeae, and so semi-quantitative (sq) RT-PCR was used with acclimated and non-acclimated leaf tissue to establish whether similar transcriptional events accompany the cold acclimation processes in Arabidopsis and barley. This approach provided inconclusive results, partly due to the scale of the experiments (over 800 samples). The experiments on HvVRN1, however, demonstrated that transcription occurred at a similar level in non-acclimated and cold acclimated leaves; thereby casting doubt on the universality of the Arabidopsis model in higher plants.
A transcriptome-wide approach, RNA-Seq, was selected to establish the pattern of transcriptional changes that accompanies cold acclimation in barley across three separate experiments. Winter barley cv. California plants were grown in non-acclimating conditions for three weeks, and then cold acclimated for up to four weeks. Leaf tissue was harvested two hours post-dawn and mRNA extracted, cDNA synthesised and between 30 and 50 million 75bp reads made for each replicate sample, using paired end Illumina NextSeq 500 technology. A novel analysis ‘pipeline’ was then written and the University of Glasgow Galaxy server used to process the reads and align them with the barley cv. Morex cDNA genome (63.7 Mb; IBSC, v2 2017) using the Kallisto package (Bray et al. 2016). Alignment with a cDNA genome dramatically reduced computational overheads; these are much greater when alignment is made to a large genome such as barley (5.3 Gb haploid genome). After indexing each of the reads to a specific barley sequence, the Kallisto output was analysed using an R-Script routine, incorporating the statistical analysis package DESeq 2 (Anders and Huber, 2010). Filtering was then applied to the DESeq 2 output in a spreadsheet to identify those sequences whose expression differed from that of the controls using the following criteria: 1, significantly different using a Benjamini-Hochberg corrected t-test; 2, greater than a two-fold change; abundance (BaseMean) score of over 1.0. The resulting data files were then analysed using set theory and Euler (Venn) diagrams to identify the common sequences amongst the treatments. Where sequences appeared in multiple data sets, a novel statistical method, a Hypergeometric Mass Function (HMF) was calculated, to estimate the probability that the multiple observations were real and not the result of random transcriptional events.
The first ‘Condition’ experiment was designed to identify components of temperature-sensing and photoperiod-sensing pathways; non-acclimated three-week-old cv. California plants were exposed to cold acclimating conditions for a further two weeks (low night temperature alone, long night periods alone, both together, and controls – neither; n=3 biological replicates, 12 samples in total). Set analysis was permitted for the groups of genes to be identified known to be responsive to cool night temperatures (T) alone, long nights alone (P) alone, and to a combination of both only (I). For the upregulated genes, 18 responded to T alone and could be placed in a temperature-sensing pathway, 8 appeared to be triggered through the P pathway, and 6 were tentatively placed at the intersections (I) between both pathways. For down regulated sequences, 16 appeared to respond to long nights alone (P), but none could be unambiguously attributed to cool night temperatures alone or interactions between both conditions. The HMFs could be estimated for the up regulated I and T sequences, and were found to be p<3.7 10-11. It is almost certain that these 24 I and T sequences respond to cold acclimating conditions. Full lists of the genes were identified as ‘significant’ by the B-H and HMF methods provided.
The second RNA-Seq experiment compared the transcriptome of the cv. California leaves exposed to non-acclimating and fully acclimating conditions (long, cool nights) for up to four weeks (‘Time-series’ experiment). Forty-eight of the up regulated TPI sequences that were identified in the first (‘Condition’) experiment were also identified as up regulated TPI sequences in the ‘Time-series’ experiment, and a corresponding 22 were identified in both experiments as down regulated. Unsurprisingly, the greatest number of changes was recorded at the end of the first week of acclimation, and this declined over time, so that the fewest changes were observed by week 4. In total, 3,697 up regulated and 5,059 down regulated sequences were identified as significant (p<0.1 in the Benjamini-Hochberg corrected t-test method incorporated into DESeq 2, whereas 48 up- and 22 down regulated sequences were calculated to have an HMF of p< 9.8 10-12. Lists of the genes identified as ‘significant’ by the B-H and the more rigorous HMF methods were also provided.
The third RNA-Seq experiment compared the transcriptome of cv. California leaves that were non-acclimated or cold acclimated by the application of exogenous ABA (10-4M) for up to 4 weeks (‘ABA Time-series’ experiment). The plants were not exposed to long nights or cool night temperatures. Measurements of the cold hardiness of leaves (LT50 measurements) showed ABA treated leaves developed a cold tolerance comparable to plants exposed to long, cool nights for the same period of time suggesting that, unlike Arabidopsis, ABA is involved in all the cold signalling pathways in barley. Transcript profiling of the control and ABA treated leaves showed 1,356 significant up regulated and 3,542 down regulated sequences (Benjamini-Hochberg statistic of p<0.1). Of these, 306 of the up regulated sequences were also found in the ‘Time-series’ experiment, and an HMF of p< 8.18 10-11 was calculated. Similarly, the down regulated 1,343 were also found in the ‘Time-series’ down regulated data set, and the corresponding HMF was p< 5.74 10-11. The lists of genes identified as ‘significant’ in the B-H and the more rigorous HMF methods are provided.
During the ‘corrections’ stage of the preparation of this thesis Gene Ontology (GO) analysis became available on-line for analysing barley sequences. This offers an independently cross-checked annotation for gene function and KEGG (Kyoto Encyclopaedia for Genes and Genomes) pathway analysis. It is beyond the scope of this thesis to provide a detailed GO analysis for the thousands of sequences that have been identified by RNA-Seq and reported here, but several of those confirmed by the HMF as highly significant changes are discussed in detail in the Discussion. These include a DNA helicase (HORVU1Hr1G019290), which is a homologue of a helicase involved in the animal Wnt signalling pathway. Another up regulated cold acclimation sequence encodes a long non-coding RNA (lncRNA; HORVU6Hr1G005050); the role of lncRNAs in the vernalisation pathway of Arabidopsis and small grained cereals is currently emerging, but the role of this lncRNA remains to be established.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Keywords: Cold acclimation, Spring and winter cereals, RNAseq, ABA, LT50, vernalisation.
• Subjects: Q Science > QP Physiology; S Agriculture > S Agriculture (General)
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Molecular Biosciences
• Supervisor's Name: Dominy, Dr. Peter
• Date of Award: 2020
• Depositing User: Mrs May Wado
• Unique ID: glathesis:2020-81623
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 03 Sep 2020 08:37
• Last Modified: 03 Sep 2020 08:47
• URI: https://theses.gla.ac.uk/id/eprint/81623