Shared and genetically distinct Zea mays transcriptome responses to ongoing and past low temperature exposure

Abstract

Background: Cold temperatures and their alleviation affect many plant traits including the abundance of protein coding gene transcripts. Transcript level changes that occur in response to cold temperatures and their alleviation are shared or vary across genotypes. In this study we identify individual transcripts and groups of functionally related transcripts that consistently respond to cold and its alleviation. Genes that respond differently to temperature changes across genotypes may have limited functional importance. We investigate if these genes share functions, and if their genotype-specific gene expression levels change in magnitude or rank across temperatures.

Results: We estimate transcript abundances from over 22,000 genes in two unrelated Zea mays inbred lines during and after cold temperature exposure. Genotype and temperature contribute to many genes' abundances. Past cold exposure affects many fewer genes. Genes up-regulated in cold encode many cytokinin glucoside biosynthesis enzymes, transcription factors, signalling molecules, and proteins involved in diverse environmental responses. After cold exposure, protease inhibitors and cuticular wax genes are newly up-regulated, and environmentally responsive genes continue to be up-regulated. Genes down-regulated in response to cold include many photosynthesis, translation, and DNA replication associated genes. After cold exposure, DNA replication and translation genes are still preferentially downregulated. Lignin and suberin biosynthesis are newly down-regulated. DNA replication, reactive oxygen species response, and anthocyanin biosynthesis genes have strong, genotype-specific temperature responses. The ranks of genotypes' transcript abundances often change across temperatures.

Conclusions: We report a large, core transcriptome response to cold and the alleviation of cold. In cold, many of the core suite of genes are up or downregulated to control plant growth and photosynthesis and limit cellular damage. In recovery, core responses are in part to prepare for future stress. Functionally related genes are consistently and greatly up-regulated in a single genotype in response to cold or its alleviation, suggesting positive selection has driven genotype-specific temperature responses in maize.

Keywords: Abiotic stress; Cold; Crossover interactions; Genotype environment interaction; Maize; RNA-Seq; Short read alignment.

Fig. 1

Experimental design. All plants were first germinated and grown at 24 °C /14 °C day/ night temperature. The cold treated plants were exposed to 14 °C / 2 °C day/night temperatures for three days starting on day 13 after planting and then returned to 24 °C /14 °C. Leaf samples for RNA extraction, depicted by the letter S, were taken 24 h into the cold exposure (D1) and 24 h after cold exposure (D4). Two control groups of plants were planted with one and three day delays, so the developmental stages of the cold and control treated plants were the same at sampling time

Fig. 2

The effects of alignment parameters on alignment attributes. As the frequency of concordant read alignments increases (x-axis), the frequency of discordant paired end read alignments increases (left y axis) and the number of variable nucleotides at single genomic positions increases (right y-axis). Alignment criteria set nine, the criteria set used in this study, results in concordant frequency of 87.3%, discordant frequency of 5%, and 2229 variable nucleotides. The data is from sample 1 chromosome 1 alignments

Fig. 3

Multidimensional scaling (MDS) representations of distances between gene transcript abundance estimates. a Projections of samples from plants grown at low temperatures (14 °C/2 °C) for 24 h, and their controls grown continuously at 24 °C/14 °C. b Projections of samples from plants grown for 24 h at 24 °C/14 °C following a 72 h exposure to 14 °C/2 °C stress. Plants’ controls were grown continuously at 24 °C/14 °C. Samples are labelled as CG60/CG102_stressed/control plants_R1/R2/R3, where R is replicate

Fig. 4

Biological process GO terms over-represented (Fisher Test, Benjamini-Hochberg adjusted p-value < 0.05) in upregulated (UP) and downregulated (DOWN) genes. D1 COLD-UP and COLD-DOWN refer to genes significantly up or down regulated in plants exposed to 24 h of cold, relative to controls; D4 COLD-UP and COLD-DOWN refer to genes significantly up or down regulated in plants exposed to 72 h of cold relative and 24 h of recovery, relative to controls. CG60-UP refers to transcripts high in inbred CG60 relative to CG102. The P values of significantly enriched GO terms with no significant child terms are shown

Fig. 5

Cellular component GO terms over-represented (Fisher Test, Benjamini-Hochberg adjusted p-value < 0.05) in upregulated (UP) and downregulated (DOWN) genes. D1 COLD-UP and COLD-DOWN refer to genes significantly up or down regulated in cold-grown plants relative to controls; D4 COLD-UP and COLD-DOWN refer to genes significantly up or down regulated in plants that had been exposed to cold relative to controls. CG60-UP refers to transcripts high in inbred CG60 relative to CG102. The P values of significantly enriched GO terms with no significant child terms are shown

Fig. 6

Examples of gene transcripts whose responses to cold differ between inbred lines. a and b plot the transcript abundances of two genes from control plants and plants grown 24 h in cold temperatures. c and d plot the transcript abundances of two genes from control plants and plants grown 24 h after the end of cold temperature exposure. A putative glyceraldehyde-3-phosphate dehydrogenase, GRMZM2G071630, (a) and a putative heat shock cognate 70 kDa protein 2, GRMZM2G428391, (b) increase in cold grown CG60 relative to control CG60 but do not increase in cold-grown CG102. These two genes are representative of the transcript changes amongst genes responsive to hydrogen peroxide (GO:00424542). (c) A histone H2A gene (GRMZM2G056231) is greatly downregulated in CG102 plants with past cold exposure relative to control plants, but it is not downregulated in CG60 plants with past cold exposure. This gene’s pattern is typical of genes involved in DNA replication and cell division. (d) The transcription factor R1 (GRMZM5G822829) is up-regulated in CG102 with past cold exposure but not up-regulated in CG60 with past cold exposure. A similar pattern was observed with other anthocyanin biosynthesis genes. Error bars represent the standard error