Skip to main content
NCBI home page
Journal of Experimental Botany logoLink to Journal of Experimental Botany
. 2024 Sep 7;75(19):6167–6181. doi: 10.1093/jxb/erae371

Altered expression of a raspberry homologue of VRN1 is associated with disruption of dormancy induction and misregulation of subsets of dormancy-associated genes

Brezo Mateos 1,2,3,✉, Katharine Preedy 4, Linda Milne 5, Jenny Morris 6, Pete E Hedley 7, Craig Simpson 8, Robert D Hancock 9, Julie Graham 10
Editor: Michael Considine11
PMCID: PMC11480652  PMID: 39243357

Abstract

Winter dormancy is a key process in the phenology of temperate perennials. Climate change is severely impacting its course leading to economic losses in agriculture. A better understanding of the underlying mechanisms, as well as the genetic basis of the different responses, is necessary for the development of climate-resilient cultivars. This study aims to provide an insight into winter dormancy in red raspberry (Rubus idaeus L). We report the transcriptomic profiles during dormancy in two raspberry cultivars with contrasting responses. The cultivar ‘Glen Ample’ showed a typical perennial phenology, whereas ‘Glen Dee’ registered consistent dormancy dysregulation, exhibiting active growth and flowering out of season. RNA-seq combined with weighted gene co-expression network analysis identified gene clusters in both genotypes that exhibited time-dependent expression profiles. Functional analysis of ‘Glen Ample’ gene clusters highlighted the significance of the cell and structural development prior to dormancy entry as well the role of genetic and epigenetic processes such as RNAi and DNA methylation in regulating gene expression. Dormancy release in ‘Glen Ample’ was associated with up-regulation of transcripts associated with the resumption of metabolism, nucleic acid biogenesis, and processing signal response pathways. Many of the processes occurring in ‘Glen Ample’ were dysregulated in ‘Glen Dee’ and 28 transcripts exhibiting time-dependent expression in ‘Glen Ample’ that also had an Arabidopsis homologue were not found in ‘Glen Dee’. These included a gene with homology to Arabidopsis VRN1 (RiVRN1.1) that exhibited a sharp decline in expression following dormancy induction in ‘Glen Ample’. Characterization of the gene region in the ‘Glen Dee’ genome revealed two large insertions upstream of the ATG start codon. We propose that expression below detection level of a specific VRN1 homologue in ‘Glen Dee’ causes dormancy misregulation as a result of inappropriate expression of a subset of genes that are directly or indirectly regulated by RiVRN1.1.

Keywords: Dormancy, phenology, transcriptomics, vernalization, VRN1, WGCNA

The raspberry cultivar ‘Glen Dee’ exhibits aberrant winter dormancy status associated with undetectable expression of a VRN1 (RiVRN1.1 ) homologue, allowing the identification of dormancy-associated genetic modules that are regulated by RiVRN1.1 .

Introduction

Winter dormancy is an adaptive mechanism protecting temperate plants from abiotic stress. It is established at the end of summer as a response to shortening photoperiods and falling temperatures. Once the onset of dormancy is established, the plants require exposure to a genotype-dependent time at cold temperatures to allow resumption of growth. Dormancy induction is triggered by environmental cues, particularly daylength and temperature, which in many species exhibit an interaction that impacts not only the onset of growth cessation and bud formation but also the depth of bud dormancy (Olsen, 2010). Climate change is therefore decoupling the main environmental cues for dormancy induction, and the rise in average winter temperatures is compromising the fulfilment of cold requirements for dormancy release. This is leading to alterations in the phenology of dormancy, uneven budbreak, and frost damage. Subsequently, natural ecosystems as well as several temperate crops are being negatively impacted (Fitter and Fitter, 2002; Cleland et al., 2007; Amano et al., 2010; Ford et al., 2016; Mosedale et al., 2016; Tixier et al., 2019).

Red raspberry Rubus idaeus (L.) is a perennial berry crop belonging to the family Rosaceae. It is primarily cultivated across temperate areas, restricted mainly by the need for winter chill. Rubus idaeus is a diploid species with a relatively small genome (~300 Mb), although highly heterozygous (Price et al., 2023). Many of its commercial varieties have a biennial life cycle, requiring dormancy to fully transition to a reproductive stage. However, there are annual genotypes able to flower within the first year of growth. Cold exposure is still needed for optimum and consistent flowering (Foster et al., 2019). Both the type of cycle and the extent of the cold requirements have been key breeding targets in the species (Jennings, 1988). Frost damage and inconsistent budbreak have become increasingly frequent, causing significant economic losses (Heide and Sønsteby, 2011; Sønsteby and Heide, 2014; Graham and Simpson, 2018).

Raspberry dormancy research has been limited by the difficulty in phenotyping dormancy depth and the genetic resources available. Several quantitative trait locus (QTL) studies in biparental populations from annual × biennial cultivars have tackled the inheritance of the type of cycle. A single locus conferring annual fruiting was described in tetraploid blackberry (Rubus subgenus Rubus Watson.) (Castro et al., 2013), although its transferability to raspberry genetic maps has not been possible (Foster et al., 2019). Two QTLs (RiAF3 and RiAF4) associated with annual fruiting have been mapped in chromosomes 3 and 4, respectively (Jibran et al., 2019), of red raspberry. Candidate genes underlying QTLs were proposed based on their function and differential expression. A homologue of JUMONJI14 (JMJ14), encoding a H3K4 demethylase was identified in the region of RiAF3. PHYTOCHROME AND FLOWERING TIME 1 (PFT1), FLOWERING CONTROL LOCUS A (FCA), and AGAMOUS-LIKE 24 (AGL24) were the candidates proposed for RiAF4. More recently, several loci linked to the annual cycle were described on chromosomes 1, 2, 4, 5, and 6 (Graham et al., 2022).

Mazzitelli et al. (2007) explored the molecular mechanisms underlying dormancy release in raspberry in an early RNA microarray analysis. This study reported a significant abundance of genes involved in stress mechanisms throughout the entire process. An SVP-like MADS-box gene, RiMADS_01, was identified, showing a profile of constant down-regulation as dormancy was released. RiMADS_01 was subsequently mapped to a region on chromosome 5 that showed significant association with flowering time in a biparental population exhibiting differences in phenology (Graham et al., 2009). Interestingly, RiMADS_01 is homologous to dam6 of Prunus persica, proposed as a negative regulator of bud break (Jiménez et al., 2010).

A growing body of research has unveiled some of the main mechanisms involved in dormancy induction and release in other species. RNA-seq analysis, sometimes combined with epigenetic or metabolomic analysis, has contributed significantly to this area. However, some fundamental questions regarding signalling, underlying causes of the differences in cold requirements between genotypes, and the conservation of general mechanisms across taxa remain unsolved. Here, we report a comparative analysis of dormancy induction and release aimed to provide an insight into the processes occurring during dormancy in two raspberry genotypes with contrasting cold requirements, ‘Glen Ample’ and ‘Glen Dee’. We identify transcriptional misregulation in ‘Glen Dee’ associated with poor aberrant dormancy regulation. Genetic analysis revealed the presence of insertions in the promoter of a VRN1-like gene in ‘Glen Dee’ that we hypothesize is responsible for the lack of detectable expression, thereby identifying genetic modules associated with bud dormancy in raspberry that are dependent on VRN1-like expression.

Materials and methods

Plant material and dormancy assessments

The material for this study was collected from mature plants of ‘Glen Ample’, a high chilling requirement biennial cultivar (Mazzitelli et al., 2007), and ‘Glen Dee’, which we have previously identified as a low chilling requirement biennial cultivar. The plants were kept under commercial conditions in a polytunnel at The James Hutton Institute, in Invergowrie (56°27ʹ24.9″N 3°03ʹ56.1″W), Scotland. Plants were pruned, leaving two to three canes per root (stool), individually tagged. Tissue sampling was conducted fortnightly between 6 August 2021 and 14 March 2022. At each time point, four canes of each cultivar were randomly selected, covering the length of the tunnel. The axillary buds from the middle and top region of each cane were pooled and flash frozen in liquid nitrogen. All the material was collected between 10.00 h and 12.00 h to standardize the circadian variation.

The progression of dormancy was monitored through single-node tests (Velappan et al., 2022). Five canes of each cultivar were cut into nodes and transferred to a forcing environment. The nodes were suspended in water trays at 20 °C with light regimes of 16 h above 100 W m−2 for 14 d after which the number of dormant, dead, and active buds was recorded.

RNA sequencing and development of the reference transcriptome

Total RNA was isolated from the bud tissue using the Qiagen RNeasy kit according to the manufacturer’s instructions. Quality was checked using a Bioanalyzer 2100 (Agilent) prior to generating individual indexed libraries each from 1 µg RNA using the Stranded mRNA Prep kit (Illumina), as recommended. Libraries were quality controlled and pooled in equimolar quantities. The final pool was run at 750 pM with 2% PhiX control library on a NextSeq 2000 (Illumina) with a P3 300 cycle kit, according to the manufacturer’s guidelines, to generate 150 bp paired-end data. Data were demultiplexed, generating individual fastq files for each sample. A reference transcriptome was developed using the methodology described in Coulter et al. (2022). RNA-seq reads were then mapped onto a raspberry reference transcript dataset (RTD) using Salmon (Patro et al., 2017) to produce transcripts per million (TPM) quantifications.

Network construction

Clusters of genes with similar expression patterns were identified using weighted co-expression network analysis (WGCNA). Raw data from Salmon quantification were transformed into length-scaled TPM at gene level using the R package tximport (Soneson et al., 2015). Genes within the fifth percentile of variance were filtered, lowering the input from 37 297 to 34 941. Reads were normalized using the variance stabilizing transformation (vst) from the package DESeq2 (Love et al., 2014). Samples were clustered based on Euclidean distance to detect potential outliers. The soft threshold for the network construction was estimated based on the fit of the linear model regressing log[p(k)]∼log(k), where p(k) corresponds to the frequency distribution of the connectivity of the nodes. Six was used as consensus power for both genotypes. Co-expression networks were built for each genotype using the R package WGCNA (Langfelder and Horvath, 2008). Modules whose overall expression changed over time were identified by fitting a model using the function lmFit from the package limma (Ritchie et al., 2015), the eigengene of each cluster as response, and time point as predictor. Standard errors were smoothed using empirical Bayes (eBayes) from limma. Results were filtered using 0.001 as significance threshold.

The main processes represented in every cluster were identified through functional profiling using g:GOSt in g:Profiler (Kolberg et al., 2023). Arabidopsis was used as reference with the analysis run using default settings (s:SCS threshold algorithm for multiple testing correction, user threshold of 0.05).

Comparative analysis

Comparative analysis focused on the four clusters of genes up-regulated during dormancy induction in ‘Glen Ample’ (M1, M31, M55, and M9). Reads of both genotypes for each cluster were clustered applying the k-means method with two centres. Genes without reads in ‘Glen Dee’ were identified; from them, RiVRN1.1 was analysed further.

Sequencing RiVRN1.1

The sequence of RiVRN1.1 was BLAST (Zhang et al., 2000)-searched against the reference genome of ‘Glen Moy’ (Hackett et al., 2018) with default parameters, giving a unique hit in scaffold 3118. The genomic sequence of the gene plus a fragment of around a 1 kb at both ends was selected. Sequences of genomic DNA and mRNA were aligned using the MAFF algorithm in Benchling. Introns, exons, open reading frames (ORFs), and 3ʹ and 5ʹ untranslated regions (UTRs) were identified. The ORF closest to the transcription start was identified, and the homology of the predicted amino acid sequence to VRN1 was verified using BLAST. Primers within the 5ʹ and 3ʹ were designed using NCBI’s primer design tool. Their specificity was verified through BLAST with the default parameters against the ‘Glen Moy’ genome.

Sequencing primers were designed from the exons in the mRNA of ‘Glen Ample’ and used to sequence the coding region of both genotypes using Sanger sequencing. For sequencing the promoter and UTRs, an extra set of primers was designed using the reference genome of ‘Glen Moy’.

Sequencing the genomes of ‘Glen Ample’ and ‘Glen Dee’

High-quality genomic DNA (250 ng each) for cultivars ‘Glen Ample’, ‘Glen Mor’, and ‘Glen Dee’ was used to generate indexed sequencing libraries using the Illumina DNA Prep kit, as recommended. Following library QC on a Bioanalyzer 2100 (Agilent), libraries were pooled, loaded at 750 pM with 1% PhiX control, and sequenced as paired-end 150 bp reads on a NextSeq 2000 (Illumina) using a P1 sequencing kit, according to the manufacturer’s instructions. Fastq files were demultiplexed post-run on the NextSeq prior to analysis. Each sequence of genomic DNA data was mapped onto the Anitra genome using BWA-MEM (Li, 2013, Preprint). The SAM output was run through ‘sambamba view’ (Tarasov et al., 2015) using a filter to remove mappings with more than six mismatches to the reference. The output BAM files were sorted and had duplicate reads removed using samtools (Danecek et al., 2021) and sambamba, respectively.

Contigs containing RiVRN1.1 in the genomes of ‘Glen Moy’, ‘Glen Mor’, ‘Glen Ample’, and ‘Glen Dee’ were aligned using the MAFF algorithm in Benchling. The promoter region of ‘Glen Ample’ and ‘Glen Dee’ was confirmed through PCR and Sanger sequencing of the amplicon.

In addition to the data of ‘Glen Ample’ and ‘Glen Dee’, the expression of RiVRN1.1 was investigated in expression data from Jibran et al. (2019), publicly available and incorporated into the RTD. The data consist of RNA-seq analysis of axillary buds of ‘Wakefield’ and ‘Heritage’, biennial and annual cultivars, respectively.

Results

Winter phenology of ‘Glen Ample’ and ‘Glen Dee’

This study monitored the progression of dormancy in two genotypes on a bi-weekly basis between August 2021 and March 2022. The depth of dormancy was measured as the proportion of bud break after 14 d in a forcing environment. The genotypes analysed had contrasting chilling requirements, previously estimated as 1500 chilling hours for ‘Glen Ample’ (Mazzitelli et al., 2007) and 750 for ‘Glen Dee’ (Sutherland et al., 2019). ‘Glen Ample’ exhibited a rapid change in bud dormancy status, shifting from almost entirely active buds on 5 August to its maximum dormancy on 30 September (TP5), with ~65% of excised buds remaining dormant (Fig. 1A). Endodormancy was almost entirely released by 10 November (TP8). ‘Glen Dee’ exhibited a shallower entry into dormancy, with the majority of buds remaining active until 16 September. After this date, an increasing proportion of buds exhibited a dormant state, reaching a maximum dormancy on 15 October (TP6). However, even at maximum dormancy only about 35% of buds were dormant. As was observed for ‘Glen Ample’, dormancy was released in the first half of November (TP8), although in some canes a higher proportion of dormant buds were observed in late November and early December (TP9 and TP10), before almost all buds resumed activity for the remaining samples. A key observation was that isolated buds from ‘Glen Dee’ showed a much larger dormancy distribution than those isolated from ‘Glen Ample’, as can be seen from the distribution of replicates from the four canes. The dispersion of the measurements suggests that dormancy is a relatively synchronous process within individuals of ‘Glen Ample’. In contrast, canes of ‘Glen Dee’ varied up to between 12% and 81% of dormant buds at TP7. This seems particularly acute around the onset of dormancy. Indications that developmental processes were more widely perturbed were supported by the observation that flowering also appears to be dysregulated in ‘Glen Dee’ with flowers observed on field planted canes as late as 13 December (Fig. 1B). On the contrary, ‘Glen Ample’ canes appeared entirely dormant in the field with closed buds (Fig. 1C).

Fig. 1.

Fig. 1.

Open in a new tab

Dormancy in R. idaeus cv ‘Glen Ample’ and ‘Glen Dee’. (A) Profiles of dormancy depth in axillary buds of raspberry cultivars ‘Glen Dee’ and ‘Glen Ample’, sampled between August 2021 and February 2022. Dormancy depth was measured as the proportion of budbreak in node sections exposed to a forcing environment for 14 d. Each point corresponds to a cane (replicate). (B) Canes of ‘Glen Dee’ exhibiting active growth and flowering on 12 December 2022, suggesting a dysregulation of dormancy consistent over seasons. (C) Canes of ‘Glen Ample’ on 12 December 2022. This genotype showed a wild type dormancy response: the plants are senescent, the buds are covered with scaly leaves, and no flowering was registered.

Co-expression networks

To understand the mechanisms underpinning differences in dormancy behaviour between the two genotypes, RNA-seq was undertaken. RNA was isolated from bud tissue of a subset of six time points representing different dormancy status (Fig. 1A) and sequenced using Illumina. Transcript abundance was quantified using Salmon. The raspberry RTD contains 137 902 transcripts that mapped to 37 492 unique genes. The length-scaled TPM for every gene was estimated for the 48 samples. Genes whose variance fell within the fifth general percentile were filtered, leaving 34 941. Given the highly contrasting dormancy behaviours between the genotypes (Fig. 1A) the dataset was split for the analysis, and an independent network was built for each genotype. The network of ‘Glen Ample’ clustered the genes into 64 modules containing between 110 and 7962 genes that differed in their expression profiles. Each module’s profile was summarized in an eigengene, and 2472 genes were assigned to the grey module (not assigned to any specific expression profile). The network built from ‘Glen Dee’ samples comprised 56 clusters ranging in size from 117 to 6082 genes and 1966 genes were assigned to the grey module. Interestingly, both networks include two clusters with opposite profiles and a notably higher size: modules 1 and 2 of ‘Glen Ample’ (7962 and 5694 genes), and modules 1 and 2 of ‘Glen Dee’ (6082 and 6068 genes). To identify clusters that exhibited changes in transcript abundance over time, a limma model where ModuleEigengene∼time was fitted. Nine modules in the network of ‘Glen Ample’ and five of ‘Glen Dee’ were significant (P<0.001) (Figs 2, 3). Hub genes, highly connected nodes within each of the significant clusters, were identified. Supplementary Table S1 lists the top five hub genes for each module.

Fig. 2.

Fig. 2.

Open in a new tab

Gene expression in clusters differentially expressed in ‘Glen Ample’. WGCNA methodology produced a network of clusters of co-expressed genes. Clusters of interest fitted a limma model using time as predictor (P<0.001). Each cluster is identified by an eigengene (red) summarizing the profile of the genes within. Gene expression values are represented as TPM centred and scaled.

Fig. 3.

Fig. 3.

Open in a new tab

Gene expression in clusters differentially expressed in ‘Glen Dee’. WGCNA methodology produced a network of clusters of co-expressed genes. Clusters of interest fitted a limma model using time as predictor (P<0.001). Each cluster is identified by an eigengene (red) summarizing the profile of the genes within. Gene expression values are represented as TPM centred and scaled.

The network of ‘Glen Ample’ was composed of nine significant differentially expressed clusters. They could be grouped into two main branches, clusters that contained transcripts that showed a pattern of decline over time (M1, M31, M55, and M9) and those that contained transcripts exhibiting an increase in abundance over time (M40, M2, M3, M33, and M14). The first branch consisted of four clusters (M1, M31, M55, and M9) comprising transcripts abundant at the end of summer and progressively decreasing as dormancy is established (Fig. 2). This branch can be linked to mechanisms active within dormancy induction. Clusters M1 and M55 had the peak of expression in the first time point, corresponding to 5 August, and decreased in abundance falling to a low level around the onset of dormancy (TP5, 30 September 2021). M1 showed a constant reduction, reaching its minimum at the point of release (TP9, 23 November 2021). Contrastingly, the abundance of M55 rose slightly at that point. These two clusters were defined as the Induction I subphase of dormancy as their maximum expression was at the earliest time point and consistently declined as dormancy status increased. Clusters M9 and M31 showed high relative abundances during induction until TP5, when plants reached a high depth of dormancy. Both exhibited a sharp reduction toward the release (TP7 29 October 2021). Phenologically, these mechanisms correspond to a second subphase of dormancy induction (Induction II).

Clusters M33 and M14 registered low abundances throughout dormancy induction and peaked during dormancy release (TP7 29 October 2021). Both modules decreased at the end of dormancy release (TP9, 23 November 2021) to reach maximum abundance in active buds (TP13 18 January 2021). These two modules have been grouped as a subphase (Release I). M2 showed low levels of relative expression at induction, increasing consistently to a maximum at the end of dormancy release (TP9, 23 November 2021) and decreasing towards TP13 (18 January 2021). Genes within M3 exhibited low abundances during early induction, increasing gradually towards release. Interestingly, the pattern of expression of this cluster was highly inconsistent between replicates. Finally, M40 clustered genes with low abundancy throughout dormancy induction and release, peaking sharply at TP13 (18 January 2021) once buds were fully responsive.

The network of ‘Glen Dee’ could be similarly divided in two main branches of clusters, showing continuous up-regulation and down-regulation, respectively (Fig. 3). The first branch grouped M1 and M3 with high abundance during induction and decreasing towards release. Genes in M1 had maximum abundance at TP1 (5 August 2021), while M3 peaked at TP3 (2 September 2021). The transition between the two main phases occurred at TP7 (29 October 2021), a month later than ‘Glen Ample’. The clusters M19, M2, and M5 followed a similar profile, registering low abundance in the first time points and increasing expression over time. M5 and M19 reach maximum expression at TP13 (18 January 2021) after endodormancy was released. M2 experienced a sharp increase at TP9 (23 November 2021) that continued at TP13 (18 January 2021). The structure of the network contrasted with results from ‘Glen Ample’, which showed a greater level of resolution and consistency between replicates, implying a loss of some degree of control.

Functional analysis of gene clusters

Functional profiling using g:GOST allowed the identification of the main biological processes represented by every cluster. This, combined with the month-to-month resolution of the networks, provided a timeline for dormancy induction and release. As ‘Glen Ample’ established and released dormancy successfully and consistently, this network was used as reference.

Dormancy Induction I

This subphase includes genes in clusters M1 and M55, with maximum expression at the earliest time point and down-regulated towards dormancy onset. Cluster M1 grouped 7962 genes, 6177 of them having an associated TAIR homologue. g:GOST analysis revealed 14 driver terms regarding biological processes (Supplementary Table S2). Highest enrichment was detected for the GO term cell cycle processes (P=3.74 × 10−39), including functions such as cell division (mitosis), development of anatomical structures, and microtubule-based processes, suggesting active growth. In addition, organization or biogenesis of cell wall, morphogenesis of plant epidermis, microtubule-based movement (P=3.38 × 10−12), and formation of xylem and phloem (P=0.001), and protein transport were enriched. GO terms for metabolism were highlighted as drivers, including metabolism of lipids (P=3.63 × 10−17), aromatic amino acids, lactone (P=0.018), and phenylpropanoids (P=2.49 × 10−12). The main regulatory processes significantly enriched in this cluster were DNA methylation (P=0.020) and histone lysine methylation (P=0.005), as well as regulation of hormone levels (P=0.047). Genes involved in RNAi-mediated antiviral immune response were significantly over-represented (P=0.003).

Cluster M55 was considerably smaller, containing 166 genes of which 136 were annotated. The genes in M55 expressed similarly to M1, having maximum expression in August at early stages of dormancy induction and decreasing at dormancy onset. However, this cluster shows a small increase in expression towards the release (TP9), although inconsistent between replicates. All the enriched GO terms in this cluster were related to membrane transport processes and endoplasmic reticulum localization.

Dormancy Induction II

The second subphase corresponded to genes that reached maximum expression at later stages of induction. Module M31 grouped 267 genes, of which 223 had available annotation. g:GOST analysis showed significant enrichment in genes involved in carbohydrate catabolism (P=0.007), cell wall modification (P=0.016), and biosynthesis and metabolism of phenylpropanoids (P=0.046, P=0.007). This cluster grouped genes exhibiting high abundances during dormancy induction whose expression increased until maximum at the point of dormancy onset, falling in expression afterwards. Cluster M9 grouped 421 genes annotated from a total of 509. Functional analysis identified four significantly enriched terms, vesicle-mediated transport (P=1.88 × 10−10), protein catabolic process (P=5.898 × 10−6), protein K63-linked ubiquitination (P=0.045), and vacuolar proton-transporting V-type ATPase complex assembly (P=0.049). This suggests strong protein turnover during this stage, a likely consequence of intense gene expression during development of protective structures at earlier stages of dormancy.

Dormancy Release I

The earliest processes occurring during dormancy release corresponded to clusters M33 and M14, and M3, with 260, 434, and 947 genes, respectively, from which 204 and 268, and 604 were annotated. No significantly enriched GO terms were detected for cluster M14. Cluster M33 showed over-representation for cellular process (P=0.012), transport (P=0.026), and sulfur compounds biosynthetic processes (P=0.0332). Although not highlighted as a driver term, establishment of location within the cell was also significantly over-represented (P=0.037). Cluster M3 is up-regulated during all processes of release, but inconsistent between replicates. From significantly over-represented GO terms, three were highlighted as drivers: secondary metabolism (P=7.566 × 10−5), protein phosphorylation (P=8.65 × 10−7), and response to stress (P=3.16 × 10−17).

Dormancy Release II

A second subphase of dormancy release peaked at TP9, at the last stages of release. This contained cluster M2, a big cluster containing 5694 genes, 4675 of which had annotation available, and M3, smaller (947 genes, 604 annotated) and quite variable in profile among the samples. Genes in M2 registered low relative abundance during dormancy induction, increasing continuously to a maximum at TP9, and decreasing towards active tissue (TP13). g:GOST analysis provided 248 significantly enriched terms regarding biological processes detailed in Supplementary Table S2, 14 of which were highlighted as driver terms. The majority of driver terms were related to resumption of gene expression, such as mRNA metabolic process (P=1.81 × 10−21), nitrogen compound transport (P=1.79 × 10−10), response to organonitrogen compound (P=1.054 × 10−6), organophosphate biosynthesis (P=0.005), DNA-templated transcription initiation (P=0.008), cytoplasmic translation (P=0.009), translational initiation (P=0.010), and protein folding (P=3.154 × 10−5). Aligned to this, post-embryonic development (P=6.74 × 10−23) appeared as a driver term. Starch catabolism (P=0.0367) also appears as a driver term, suggesting a role as energy source.

Resumption of growth

Cluster M40 exhibited a peak in gene expression at the latest sampled time point once dormancy was fully released. This cluster was composed of 231 transcripts of which 171 genes were annotated. Functional profiling highlighted non-coding RNA (ncRNA) processing (P=4.045 × 10−5), macromolecule modification (P=0.008), and several GO terms referring to metabolic processes, as well as nitrogen compound and organic cyclic compounds metabolism specifically (P=0.005, P=0.019), all presumably associated with resumption of growth.

Comparative analysis of ‘Glen Dee’ and ‘Glen Ample’

The overall mechanisms dysregulated during dormancy induction in ‘Glen Dee’ were investigated using the gene network of ‘Glen Ample’ as reference. As ‘Glen Dee’ is unable to fully establish dormancy, analysis focused on the four clusters of genes up-regulated during dormancy induction in ‘Glen Ample’. For each cluster, reads of both genotypes were merged and reclustered using k-means with two centres. This split the reads into a centre containing reads from ‘Glen Ample’ and genes of ‘Glen Dee’ with similar profiles, and a second centre containing genes with altered profiles in ‘Glen Dee’. gOST analysis from gProfiler identified ontology terms over-represented as potential dysregulated mechanisms (Supplementary Table S3). During early induction cell cycle, microtubule-based process and movement, lipid, lactone and phenylpropanoids, and histone lysine methylation showed altered expression. Contrastingly with ‘Glen Dee’, these mechanisms did not shift to down-regulation towards the maximum of dormancy but stayed up-regulated for much of the experiment. In addition to these mechanisms, other highlighted processes such as photosynthesis (P=0.03683) and stomatal movement (P=0.0405) were unique to this genotype. Dysregulated genes in module 55 were significantly enriched in genes associated with the endosome (GO:0005768, P=0.034379), and those in module 31 were significantly enriched in genes associated with apoplast, chloroplasts, and plastids (GO:0048046, GO:0009536, GO:0009507). Terms for synthesis of phenylpropanoids and flavonoids were also over-represented (WP:WP1538, GO:0009698; P=0.011596, P=0.011054), as well as copper ion binding, NADP+ binding, and oxidoreductase activity (GO:0005507, GO:0070401, GO:0016491; P=0.017675, P=0.024927, P=0.025497). Re-clustering module 9 did not produce a separation, where the reads of ‘Glen Ample’ fall together in one of the clusters, but rather split them evenly between the two. Equally, the reads of ‘Glen Dee’ distributed evenly.

Genes not expressed in ‘Glen Dee’ were identified as potential candidates regulating dormancy in ‘Glen Ample’. Module 55 of ‘Glen Ample’ contained 166 genes of which three were not detected in ‘Glen Dee’. These genes were not annotated in the reference transcriptome. Module 9 of ‘Glen Ample’ contained 506 genes, 10 of which were not detected in ‘Glen Dee’. Four of them have alignments in BLAST, but their Arabidopsis homologues are proteins of unknown function (AT5G45530.1 and AT2G36430.1), and the other two have a hAT dimerization domain (AT5G33406.1) and an FBD-like domain (AT1G51055.1) but no known function. Module 31 of ‘Glen Ample’ contained 267 genes, four of which were not detected in the transcriptome of ‘Glen Dee’. None produced significant alignments. Module 1 of ‘Glen Ample’ contains 7962 genes, of which 148 were not detected in ‘Glen Dee’ samples. Twenty-four of these transcripts had homologues in Arabidopsis as determined by significant BLAST hits (Table 1).

Table 1.

Genes in cluster M1 of ‘Glen Ample’ not detected in ‘Glen Dee’

Gene Homologue Description
rirtd3_HiC_scaffold_1100G000080 AT2G01050.1 Zinc ion binding, nucleic acid binding
rirtd3_HiC_scaffold_1118G000160 AT3G17080.1 Plant self-incompatibility protein S1 family
rirtd3_HiC_scaffold_1G002370 AT3G42170.1 BED zinc finger, hAT family dimerization domain
rirtd3_HiC_scaffold_1G020110 AT5G45520.1 Leucine-rich repeat (LRR) family protein
rirtd3_HiC_scaffold_1G021000 AT4G08685.1 SAH7 Pollen Ole e 1 allergen and extensin family protein
rirtd3_HiC_scaffold_1G026130 AT3G18990.1 VRN1 REM39 AP2/B3-like transcriptional factor family protein
rirtd3_HiC_scaffold_270G000220 AT4G03460.1 Ankyrin repeat family protein
rirtd3_HiC_scaffold_2G010200 AT2G21340.1 MATE efflux family protein
rirtd3_HiC_scaffold_2G014650 AT5G53130.1 CNGC1 ATCNGC1 cyclic nucleotide gated channel 1
rirtd3_HiC_scaffold_2G047170 AT1G64185.1 Lactoylglutathione lyase/glyoxalase I family protein
rirtd3_HiC_scaffold_330G000110 AT1G65780.1 P-loop containing nucleoside triphosphate hydrolases superfamily protein
rirtd3_HiC_scaffold_3G038200 AT1G08390.1 Unknown protein RecQ-mediated genome instability protein 2
rirtd3_HiC_scaffold_3G039040 AT4G14060.1 Polyketide cyclase/dehydrase and lipid transport superfamily protein
rirtd3_HiC_scaffold_4G010850 AT5G57590.1 BIO1 adenosylmethionine-8-amino-7-oxononanoate transaminases
rirtd3_HiC_scaffold_4G010900 AT1G17230.1 Leucine-rich receptor-like protein kinase family protein
rirtd3_HiC_scaffold_5G007100 AT2G34730.1 Myosin heavy chain-related
rirtd3_HiC_scaffold_5G016800 AT4G12500.1 Bifunctional inhibitor/lipid-transfer protein/seed storage 2S albumin superfamily protein
rirtd3_HiC_scaffold_5G040920 AT5G36930.2 Disease resistance protein (TIR-NBS-LRR class) family
rirtd3_HiC_scaffold_5G045300 AT1G27170.1 Transmembrane receptors ATP binding
rirtd3_HiC_scaffold_6G038650 AT4G32208.1 Heat shock protein 70 (Hsp 70) family protein
rirtd3_HiC_scaffold_775G000010 AT1G54690.1 HTA3 H2AXB G-H2AX GAMMA-H2AX gamma histone variant H2AX
rirtd3_HiC_scaffold_7G001240 AT5G24280.1 GMI1 gamma-irradiation and mitomycin c induced 1
rirtd3_HiC_scaffold_7G002170 AT5G24280.1 GMI1 gamma-irradiation and mitomycin c induced 1
rirtd3_HiC_scaffold_7G009310 AT3G53960.1 Major facilitator superfamily protein peptide/nitrate transporter plant
Open in a new tab

Genes were annotated using Arabidopsis as reference. M1 groups genes with maximum levels of expression at early dormancy induction, decreasing expression throughout the remainder of the time course.

Characterizing RiVRN1.1

Twenty-eight genes were up-regulated during dormancy in ‘Glen Ample’ and not expressed in ‘Glen Dee’. rirtd3_HiC_scaffold_1G026130 was analysed further due to its molecular function as transcription factor and its homology with VERNALIZATION1 (VRN1) in Arabidopsis. VRN1 is involved in vernalization and transition between vegetative and reproductive stages, and is widely characterized in Arabidopsis (reviewed in Banerjee et al., 2017), as well as barley (Deng et al., 2015) and wheat (reviewed in Milec et al., 2023). This gene homologue was highly expressed in ‘Glen Ample’ during dormancy induction up to TP3 (Fig. 4), but transcripts sharply declined at TP5 and remained close to the limits of detection for the remainder of the time course. There were four splice variants, with rirtd3_HiC_scaffold_1G026130.RTD.1 being the longest and most abundant transcript. Alignment of the mRNA sequences of ‘Glen Ample’ with the reference genome of ‘Glen Moy’ (Hackett et al., 2018) showed no mismatches.

Fig. 4.

Fig. 4.

Open in a new tab

Expression of RiVRN1.1 (rirtd3_HiC_scaffold_1G026130) in ‘Glen Ample’ and ‘Glen Dee’ samples. Four isoforms were detected in ‘Glen Ample’, reaching the maximum accumulated expression at TP3 (2 September 2021). No reads of any RiVRN1.1 transcript were detected in samples from ‘Glen Ample’. TPM, transcripts per million.

The RiVRN1.1 isoforms of ‘Glen Ample’ were aligned to identify the splicing sites. The coding region of ‘Glen Ample’ and ‘Glen Dee’ was re-sequenced using Sanger sequencing. Although substitutions were detected, most mutations were synonymous with ‘Glen Moy’, and no new stop codons or shifts in reading frame were observed in the ‘Glen Ample’ and ‘Glen Dee’ genomes relative to ‘Glen Moy’. The genomes of ‘Glen Ample’ and ‘Glen Dee’ were sequenced using Illumina to access the promoter regions and gene environment. The sequencing produced a total of 13 410 248 bp for ‘Glen Ample’, 11 870 486 for ‘Glen Mor’, and 11 605 377 for ‘Glen Dee’, reaching 0.05× coverage for an estimated genome size of 291 000 000 bp (Davik et al., 2022). Alignment of contigs containing rirtd3_HiC_scaffold_1G026130 revealed two insertions in the promoter region of the gene in ‘Glen Dee’. The most distal insertion was 747 bp in length and was located 535 bp upstream of the mRNA (Supplementary Dataset S1). Sequence and presence of the insertion were confirmed through amplicon sequencing. The second insertion was 301 bp and located 54 bp upstream of the beginning of the mRNA in ‘Glen Dee’ (Fig. 5). Both insertions were confirmed to be homozygous through PCR and gel visualization. In addition to this, the genome of ‘Glen Dee’ exhibited a 3 bp deletion at the predicted translational start, absent in the genomes of ‘Glen Ample’ and ‘Glen Moy’.

Fig. 5.

Fig. 5.

Open in a new tab

Alignment of genomic sequences of ‘Glen Moy’ (reference genome and PCR product), ‘Glen Ample’, and ‘Glen Dee’, illustrating the second insertion located in the promoter area of RiVRN1.1. The PCR product from RiVRN1 of ‘Glen Ample’, ‘Glen Moy’, and ‘Glen Dee’ was Sanger sequenced and aligned with the reference genome of ‘Glen Moy’. ‘Glen Moy’ and ‘Glen Ample’ showed a high level of conservation. ‘Glen Dee’ has a 301 bp insertion 54 bp upstream of the start of the UTR of ‘Glen Ample’ (*) and 162 bp upstream of the start codon (**). In addition, ‘Glen Dee’ has a substitution 93 bp upstream the start codon, and a deletion of three bases where ‘Glen Ample’ and ‘Glen Moy’ have two consecutive ORFs.

The expression of RiVRN1.1 was investigated in gene expression data from the cultivars ‘Heritage’ and ‘Wakefield’ produced by Jibran et al. (2019) incorporated into the RTD (Supplementary Fig. S1). ‘Heritage’ is an annual cultivar, while ‘Wakefield’ is biennial. The four isoforms of the gene were present in both cultivars, with rirtd3_HiC_scaffold_1G026130.RTD.1 being predominant. Its expression values were higher in ‘Wakefield’, with a mean of 19.35 TPM, than ‘Heritage’, with 9.60. The samples were collected in late spring/early summer, ahead of the peak of expression we report in early August.

Twelve more genes showed homology with VRN1 of Arabidopsis (Supplementary Dataset S2) and their expression levels varied during dormancy, suggesting their cues and functions may differ. Alignment of their mRNA sequences showed relatively low identities, ranging from 61.95% to 86.5%, to the predominant isoform of rirtd3_HiC_scaffold_1G026130 (rirtd3_HiC_scaffold_1G026130.RTD.1).

Discussion

We investigated the genetic mechanisms underlying dormancy in two raspberry cultivars with contrasting phenology. Reportedly, ‘Glen Ample’ and ‘Glen Dee’ differ in their chilling requirements, which would translate in different timings of dormancy release. Bi-weekly monitoring of the dormancy status revealed differences not only in the timing or the cold requirements, but in the profile of dormancy and consistency among different individuals. In addition, blooming out of season was recorded in ‘Glen Dee’ as late as 12 December 2021 (Fig. 1B, C).

The onset and release of dormancy occurred earlier than expected based on the literature in both genotypes. This may be due to most of the published data being assessed in whole canes and therefore incorporating effects of paradormancy. In our experiment, endodormancy was released before the end of winter and subsequent frost risk.

A model for dormancy in Rubus idaeus.

The time-resolved co-expression network of ‘Glen Ample’ was used to build a model for dormancy in raspberry. The process is divided into two main phases, induction and release, controlled by independent mechanisms. Induction initiates at the end of summer, in early August, and completes late September.

Induction can be divided into two subphases. Its earliest stage, Induction I, involves metabolic, cellular organization, and gene silencing processes. Key metabolic processes include the organization of the cell wall or biogenesis, as well as potentially linked processes such as metabolism of phenylpropanoids, or aromatic amino acid metabolic processes, and their precursors. Primary metabolic processes are represented by lipid metabolism. During this stage histone lysine methylation, DNA methylation, and RNAi-mediated immune response genes are over-represented. These processes likely initiate silencing of gene expression that would lead to growth arrest. Involvement of histone and DNA methylation in the establishment of winter dormancy has been widely described in other temperate species (Conde et al., 2013; Ríos et al., 2014; Rothkegel et al., 2020; Chen et al., 2022). The role of RNAi mechanisms in the process is less studied, although variations in the miRNA linked to QTLs controlling chilling requirements have been reported in peach (Barakat et al., 2012). Interestingly, homologues of some of the genes detected in our analysis, such as ARGONAUT 1 (AGO1) or HUA ENHANCER 1 (HEN1) play a central role in the silencing of gene expression during seed dormancy in Arabidopsis (Jones-Rhoades and Bartel, 2004; Allen et al., 2005; Tognacca and Botto, 2021).

A second group of mechanisms peak a month later, around the time the plants reach the midpoint of dormancy (Induction II). Cell wall modification and linked processes, such as phenylpropanoid metabolism, are active. General metabolism is represented by catabolism of carbohydrates. Other processes include vesicle-based transport, membrane docking, organelle localization by membrane tethering, and processes associated with endoplasmic reticulum (ER) organization (GO:0051643, GO:2261817, GO:0090158). The potential significance of the latter may relate to the ER being an integral component of the plasmodesmata (Nicolas et al., 2017), which have been strongly implicated in dormancy induction and release (Rinne et al., 2011).

At this stage, processes linked to protein turnover are strongly represented in the cells. This could be a consequence of growth and development of protective structures occurring in earlier stages and could potentially have an additional regulatory role, as has been proposed during the release of seed dormancy (Oracz and Stawska, 2016). This group of genes have a second peak in expression about the time of dormancy release.

Onset of dormancy coincides with a shift in mechanisms in the transcription data. Release of dormancy was a gradual process, less synchronous between individuals than induction. The earliest stage, named here Release I, contains genes up-regulated at the maximum levels of dormancy. The main processes include secondary metabolism, phosphorylation of proteins, and abiotic stress signalling, particularly hypoxia. Mazzitelli et al. (2007) reported several genes involved in stress response linked to dormancy in raspberry. Resemblance of gene expression in dormant buds and response to anoxic stress has been previously drawn (Considine and Foyer, 2014) and proposed to be involved in release through increase in ROS (Beauvieux et al., 2018). Here, this group of genes shows low relative abundances through dormancy onset and induction, peaking sharply during release. Timing of up-regulation and maximum abundance seems to be inconsistent between replicates (Fig. 2) but does not reflect the consistency in the release of dormancy at corresponding time points (Fig 1A). This GO term includes CCG-BINDING PROTEIN 1 (CBP1), EXORDIUM-LIKE 2 (EXL2), ATRMA3, ETHYLENE RESPONSE DNA BINDING FACTOR 1 (EDF1), and CYTOCHROME P450 (CYP707A3), reported in gene regulatory networks linked to dormancy (Tarancón et al., 2017).

After the initial stages of anoxic stress, endodormancy is released between October and late November. Gene expression is resumed as is inferred by over-representation of genes involved in transcription, translation, and protein folding. Starch catabolism is one of the main processes occurring at this stage, potentially an energy source for cell machinery resumption after the period of low carbon input from short photoperiods. However, a growing body of evidence suggests a pivotal role of non-structural carbohydrates in dormancy release signalling (Gibon et al., 2004; Bolouri Moghaddam and Van den Ende, 2013; Palacio et al., 2014; Tixier et al., 2018, 2019). This has been hypothesized to rely on the effect of temperature on the enzymatic equilibrium between starch and sugars (Tixier et al., 2019). Our data show a constant increase in expression of this group of genes from the onset of dormancy to the resumption of growth, following the accumulation of chilling. Further analysis is needed to clarify the role of starch degradation as cause or consequence of release of dormancy. Upon resumption of growth, a group of genes involved in regulation of non-coding DNA and different aspects of metabolism become up-regulated, overlapping with the processes previously described.

RiVRN1.1 is a key to the establishment of dormancy

The insertions in the promoter region of RiVRN1.1 in ‘Glen Dee’ are associated with reduced dormancy and early flowering, confirmed over two following seasons. RiVRN1.1 registers high relative abundances during dormancy induction in ‘Glen Ample’. After reaching maximum mid-induction, expression falls abruptly, remaining at low abundance for the rest of the time course (Fig. 4). The cultivar ‘Glen Dee’ exhibits two insertions of 774 bp and 300 bp, at the distal and proximal regions of the promoter (Fig. 5), and no expression was detected in our data. RiVRN1.1 is an orthologue of VERNALIZATION 1 (VRN1), a MADS-box transcription factor holding a central role in the vernalization pathway in Arabidopsis by repressing FLOWERING LOCUS C (FLC) (Levy et al., 2002; Milec et al., 2023). Links between regulation of vernalization and dormancy through epigenetic mechanisms have been previously reviewed (Horvath, 2009; Considine and Foyer, 2014; Ríos et al., 2014; Maurya and Bhalerao, 2017).

In addition to the flowering and dormancy phenotypes, the gene expression data of ‘Glen Dee’ show mechanisms involved in photosynthesis, active growth, and stomatal movement, active throughout the winter (Figs 1B, 3). Therefore, dormancy induction seems to be lacking a key point of regulation. Comparative analysis, taking the network of ‘Glen Ample’ as a model, identified processes altered during dormancy induction in ‘Glen Dee’. Most processes identified in the ontology analysis are clustered in the same module as RiVRN1.1. These include regulatory elements, such as hormones, histone lysine methylation, DNA methylation, and RNAi silencing.

The activity of VRN1 in Arabidopsis has been linked to changes in the methylation patterns of histone H3 (Bastow et al., 2004). Several genes altered in ‘Glen Dee’ from the histone lysine methylation pathway are involved in silencing mechanisms, such as CHROMOMETHYLASE 3 (CMT3), ARGONAUTE 4 (AGO4), PDP1, ATRFC, KRYPTONITE (KYP), and SDG34. AGO4 was proposed to act as repressor of dormancy in wheat seeds (Singh et al., 2013; Katsuya-Gaviria et al., 2020). In transcriptomic studies in poplar, leafy spurge, and tea tree, AGO4 shows down-regulation during bud endodormancy (Horvath et al., 2008; Matzke and Mosher, 2014; Howe et al., 2015; Hao et al., 2017). KYP encodes a methyltransferase known to hold a role in seed dormancy by regulating negatively DOG1 and ABI3 (Zheng et al., 2012). Genes regulating flowering time were present in the cluster of altered expression profiles. These include EARLY FLOWERING MYB PROTEIN (EFM), RBBP5 LIKE (RB), ETHYLENE INSENSITIVE 6 (EIN6) and TRAUCO (TRO), as well as regulators of FLC, EARLY FLOWERING IN SHORT DAYS (EFS) (Kim et al., 2005), PLANT HOMOLOGOUS TO PARAFIBROMIN (Park et al., 2010), PWWP DOMAIN PROTEIN 1 (PDP1) (Zhou et al., 2018), and VERNALIZATION INDEPENDENCE 3 (VIP3) (Zhang et al., 2003). RBBP5 is a component of the COMPASS complex, and its silencing supresses the activity of FLC, leading to early flowering (Jiang et al., 2011). EIN6 (REF6) inhibits dormancy through the catabolism of ABA in seeds of Arabidopsis (Chen et al., 2020). EFM is a key repressor of FT integrating light and temperature stimuli (Yan et al., 2014). Interestingly, EFM expression is promoted by SHORT VEGETATIVE PHASE (SVP), homologous to the DORMANCY-ASSOCIATED MADS-BOX (DAM) family.

Other mechanisms and genes could be contributing to the dysregulation of dormancy observed in ‘Glen Dee’. Chromatin accessibility through histone methylation is one of the central processes controlling dormancy induction (Chen et al., 2022). An orthologue of H2A.X was found among the genes involved in dormancy induction in ‘Glen Ample’ showing no detectable expression levels in ‘Glen Dee’ (Table 1). This gene encodes a gamma-induced variant of histone H2A in Arabidopsis. h2a.x mutants have been linked to tissue-specific hypomethylation in the endosperm of Arabidopsis (Frost et al., 2023). However, its mechanism of action remains unsolved and, to our knowledge, has not been studied in vegetative meristems.

Conclusions

This study provides an overview of the mechanisms underlying raspberry dormancy. The WGCNA methodology identifies changes in gene expression on a month-by-month basis. Comparative analysis of transcriptomic and phenological data from ‘Glen Ample’ and ‘Glen Dee’ identified a general dysregulation of dormancy induction in the latter, which leads to growth and flowering out of season. Analysis of transcriptomic and genomic data identified a VRN1-like gene (RiVRN1.1) as the likely candidate for these responses. The genome of ‘Glen Dee’ exhibits two insertions in the proximal and distal promoter regions of RiVRN1.1 that might cause disruption and be linked to its not-detectable expression. Our findings provide a framework for molecular analysis of dormancy in raspberry and suggest the role of RiVRN1.1 as a key point in the regulation of dormancy induction. Further analysis would be necessary to understand its role, and shed light onto the mechanisms linking vernalization and dormancy.

Supplementary data

The following supplementary data are available at JXB online.

Fig. S1. Expression of RiVRN1.1 in axillary buds from the cultivars ‘Heritage’ (annual) and ‘Wakefield’ (biennial). Data from Jibran et al. (2019).

Table S1. Top five genes with most connections (‘hubgenes’) in the differentially expressed gene clusters of ‘Glen Ample’ and ‘Glen Dee’.

Table S2. Summary of the functional analysis of the gene clusters differentially expressed in ‘Glen Ample’.

Table S3. Summary of the functional analysis of the of dysregulated genes in ‘Glen Dee’ corresponding to clusters 1, 31, and 55 of ‘Glen Ample’.

Dataset S1. Alignment of the genomic scaffolds of ‘Glen Moy’, ‘Glen Dee’, ‘Glen Ample’, and ‘Glen Mor’ containing RiVRN1.1.

Dataset S2. Alignment of the mRNA of the VRN1-like genes identified in the analysis.

erae371_suppl_Supplementary_Tables_S1-S14_Datasets_S1-S3
erae371_suppl_supplementary_tables_s1-s14_datasets_s1-s3.zip (896.4KB, zip)

Acknowledgements

The authors would like to thank Kay Smith, Linzi Ross, and John Fuller for their advice on processing the samples, Charlotte Hurst for her advice on obtaining long amplicons for Sanger sequencing, and Dominic Williams for his feedback on the network construction.

Contributor Information

Brezo Mateos, Cell and Molecular Sciences, The James Hutton Institute, Invergowrie, Dundee DD2 5DA, UK; School of Biology, Biomedical Sciences Research Complex, University of St Andrews, North Haugh, St Andrews KY16 9ST, UK; Biomathematics and Statistics Scotland, The James Hutton Institute, Invergowrie, Dundee DD2 5DA, UK.

Katharine Preedy, Biomathematics and Statistics Scotland, The James Hutton Institute, Invergowrie, Dundee DD2 5DA, UK.

Linda Milne, Informational and Computational Sciences, The James Hutton Institute, Invergowrie, Dundee DD2 5DA, UK.

Jenny Morris, Cell and Molecular Sciences, The James Hutton Institute, Invergowrie, Dundee DD2 5DA, UK.

Pete E Hedley, Cell and Molecular Sciences, The James Hutton Institute, Invergowrie, Dundee DD2 5DA, UK.

Craig Simpson, Cell and Molecular Sciences, The James Hutton Institute, Invergowrie, Dundee DD2 5DA, UK.

Robert D Hancock, Cell and Molecular Sciences, The James Hutton Institute, Invergowrie, Dundee DD2 5DA, UK.

Julie Graham, Cell and Molecular Sciences, The James Hutton Institute, Invergowrie, Dundee DD2 5DA, UK.

Michael Considine, University of Western Australia, Australia.

Author contributions

BM, JG, RDH, KP, and CS: conceptualization; BM: formal analysis; JG, RDH, and KP: supervision; LM: software, data curation; PEH and JM: methodology, BM: writing—original draft; JG, RDH, KP, CS, LM, and PEH: writing—review and editing.

Conflict of interest

The authors declare they have no conflicts of interest.

Funding

BM acknowledges the Mylnefield trust for receipt of a PhD scholarship. Work was funded by the Mylnefield Trust and the Rural and Environmental Science and Analytical Services Division of the Scottish Government under the 2022-2027 Strategic Research Programme.

Data availability

The sequencing data from the RNA-seq experiment, as well as the genomic DNA from ‘Glen Ample’, ‘Glen Dee’, and ‘Glen Mor’ are deposited in the European Nucleotide Archive under the project accession PRJEB77730. The phenotypic data can be made available upon request to the authors, as well as the plant material provided the appropriate licenses from James Hutton Ltd.

References

  1. Allen E, Xie Z, Gustafson AM, Carrington JC.. 2005. MicroRNA-directed phasing during trans-acting SiRNA biogenesis in plants. Cell 121, 207–221. doi: 10.1016/J.CELL.2005.04.004 [DOI] [PubMed] [Google Scholar]
  2. Amano T, Smithers RJ, Sparks TH, Sutherland WJ.. 2010. A 250-year index of first flowering dates and its response to temperature changes. Proceedings of the Royal Society Section B: Biological Sciences 277, 2451–2457. doi: 10.1098/rspb.2010.0291 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Banerjee A, Wani SH, Roychoudhury A.. 2017. Epigenetic control of plant cold responses. Frontiers in Plant Science 8, 1643. doi: 10.3389/fpls.2017.01643 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Barakat A, Sriram A, Park J, Zhebentyayeva T, Main D, Abbott A.. 2012. Genome wide identification of chilling responsive microRNAs in Prunus persica. BMC Genomics 13, 481. doi: 10.1186/1471-2164-13-481 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bastow R, Mylne JS, Lister C, Lippman Z, Martienssen RA, Dean C.. 2004. Vernalization requires epigenetic silencing of FLC by histone methylation. Nature 427, 164–167. doi: 10.1038/NATURE02269 [DOI] [PubMed] [Google Scholar]
  6. Beauvieux R, Wenden B, Dirlewanger E.. 2018. Bud dormancy in perennial fruit tree species: a pivotal role for oxidative cues. Frontiers in Plant Science 9, 657. doi: 10.3389/fpls.2018.00657 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bolouri Moghaddam MR, Van den Ende W.. 2013. Sugars, the clock and transition to flowering. Frontiers in Plant Science 4, 22. doi: 10.3389/fpls.2013.00022 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Castro P, Stafne ET, Clark JR, Lewers KS.. 2013. Genetic map of the primocane-fruiting and thornless traits of tetraploid blackberry. Theoretical and Applied Genetics 126, 2521–2532. doi: 10.1007/s00122-013-2152-3 [DOI] [PubMed] [Google Scholar]
  9. Chen H, Tong J, Fu W, et al. 2020. The H3K27me3 demethylase RELATIVE OF EARLY FLOWERING6 suppresses seed dormancy by inducing abscisic acid catabolism. Plant Physiology 184, 1969–1978. doi: 10.1104/pp.20.01255 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chen W, Tamada Y, Yamane H, Matsushita M, Osako Y, Gao-Takai M, Luo Z, Tao R.. 2022. H3K4me3 plays a key role in establishing permissive chromatin states during bud dormancy and bud break in apple. The Plant Journal 111, 1015–1031. doi: 10.1111/tpj.15868 [DOI] [PubMed] [Google Scholar]
  11. Cleland E, Chuine I, Menzel A, Mooney H, Schwartz M.. 2007. Shifting plant phenology in response to global change. Trends in Ecology & Evolution 22, 357–365. doi: 10.1016/j.tree.2007.04.003 [DOI] [PubMed] [Google Scholar]
  12. Conde D, González-Melendi P, Allona I.. 2013. Poplar stems show opposite epigenetic patterns during winter dormancy and vegetative growth. Trees 27, 311–320. doi: 10.1007/s00468-012-0800-x [DOI] [Google Scholar]
  13. Considine MJ, Foyer CH.. 2014. Redox regulation of plant development. Antioxidants & Redox Signaling 21, 1305–1326. doi: 10.1089/ARS.2013.5665 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Coulter M, Entizne JC, Guo W, et al. 2022. BaRTv2: A highly resolved barley reference transcriptome for accurate transcript-specific RNA-seq quantification. The Plant Journal 111, 1183–1202. doi: 10.1111/tpj.15871 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Danecek P, Bonfield JK, Liddle J, et al. 2021. Twelve years of SAMtools and BCFtools. GigaScience 10, giab008. doi: 10.1093/gigascience/giab008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Davik J, Roen D, Lysoe E, Buti M, Rossman S, Alsheikh M, Aiden EL, Dudchenko O, Sargent DJ.. 2022. A chromosome-level genome sequence assembly of the red raspberry (Rubus idaeus L.). PLoS One 17, e0265096. doi: 10.1371/journal.pone.0265096 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Deng W, Casao MC, Wang P, Sato K, Hayes PM, Finnegan EJ, Trevaskis B.. 2015. Direct links between the vernalization response and other key traits of cereal crops. Nature Communications 6, 5882. doi: 10.1038/ncomms6882 [DOI] [PubMed] [Google Scholar]
  18. Fitter AH, Fitter RSR.. 2002. Rapid changes in flowering time in British plants. Science 296, 1689–1691. doi: 10.1126/science.1071617 [DOI] [PubMed] [Google Scholar]
  19. Ford KR, Harrington CA, Bansal S, Gould PJ, Clair JBS.. 2016. Will changes in phenology track climate change? A study of growth initiation timing in coast Douglas-fir. Global Change Biology 22, 3712–3723. doi: 10.1111/gcb.13328 [DOI] [PubMed] [Google Scholar]
  20. Foster TM, Bassil NV, Dossett M, Leigh Worthington M, Graham J.. 2019. Genetic and genomic resources for Rubus breeding: a roadmap for the future. Horticulture Research 6, 116. doi: 10.1038/s41438-019-0199-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Frost JM, Lee J, Hsieh PH, et al. 2023. H2A.X promotes endosperm-specific DNA methylation in Arabidopsis thaliana. BMC Plant Biology 23, 585. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Gibon Y, Bläsing OE, Palacios-Rojas N, Pankovic D, Hendriks JHM, Fisahn J, Höhne M, Günther M, Stitt M.. 2004. Adjustment of diurnal starch turnover to short days: depletion of sugar during the night leads to a temporary inhibition of carbohydrate utilization, accumulation of sugars and post-translational activation of ADP-glucose pyrophosphorylase in the following light period. The Plant Journal 39, 847–862 doi: 10.1111/j.1365-313X.2004.02173.x [DOI] [PubMed] [Google Scholar]
  23. Graham J, Hackett CA, Smith K, Woodhead M, Hein I, McCallum S.. 2009. Mapping QTLs for developmental traits in raspberry from bud break to ripe fruit. Theoretical and Applied Genetics 118, 1143–1155. doi: 10.1007/s00122-009-0969-6 [DOI] [PubMed] [Google Scholar]
  24. Graham J, Simpson C.. 2018. Developmental transitions to fruiting in red raspberry. In: Hytönen T, Graham J, Harrison R, eds. The Genomes of rosaceous berries and their wild relatives. Compendium of plant genomes. Cham: Springer, 199–212. doi: 10.1007/978-3-319-76020-9_14 [DOI] [Google Scholar]
  25. Graham J, Smith K, Mackenzie K, Milne L, Jennings N, Mateos B, Hackett C.. 2022. Developmental QTL in a red raspberry primocane × biennial raspberry population that exhibit primocane fruiting. Journal of Horticulture 9, 308. doi: 10.35248/2376-0354.22.9.1000308 [DOI] [Google Scholar]
  26. Hackett CA, Milne L, Smith K, Hedley P, Morris J, Simpson CG, Preedy K, Graham J.. 2018. Enhancement of Glen Moy × Latham raspberry linkage map using GbS to further understand control of developmental processes leading to fruit ripening. BMC Genetics 19, 59. doi: 10.1186/s12863-018-0666-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Hao X, Yang Y, Yue C, Wang L, Horvath DP, Wang X.. 2017. Comprehensive transcriptome analyses reveal differential gene expression profiles of Camellia sinensis axillary buds at para-, endo-, ecodormancy, and bud flush stages. Frontiers in Plant Science 8, 553. doi: 10.3389/fpls.2017.00553 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Heide OM, Sønsteby A.. 2011. Physiology of flowering and dormancy regulation in annual- and biennial-fruiting red raspberry (Rubus idaeus L.) – a review. Journal of Horticultural Science and Biotechnology 86, 433–442. doi: 10.1080/14620316.2011.11512785 [DOI] [Google Scholar]
  29. Horvath D. 2009. Common mechanisms regulate flowering and dormancy. Plant Science 177, 523–531. doi: 10.1016/j.plantsci.2009.09.002 [DOI] [Google Scholar]
  30. Horvath DP, Chao WS, Suttle JC, Thimmapuram J, Anderson JV.. 2008. Transcriptome analysis identifies novel responses and potential regulatory genes involved in seasonal dormancy transitions of leafy spurge (Euphorbia esula L.). BMC Genomics 9, 536. doi: 10.1186/1471-2164-9-536 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Howe GT, Horvath DP, Dharmawardhana P, Priest HD, Mockler TC, Strauss SH.. 2015. Extensive transcriptome changes during natural onset and release of vegetative bud dormancy in Populus. Frontiers in Plant Science 6, 989. doi: 10.3389/fpls.2015.00989 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Jennings DL. 1988. Raspberries and blackberries: their breeding, diseases and growth. London: Academic Press. [Google Scholar]
  33. Jiang D, Kong NC, Gu X, Li Z, He Y.. 2011. Arabidopsis COMPASS-Like complexes mediate histone H3 lysine-4 trimethylation to control floral transition and plant development. PLoS Genetics 7, e1001330. doi: 10.1371/journal.pgen.1001330 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Jibran R, Spencer J, Fernandez G, Monfort A, Mnejja M, Dzierzon H, Tahir J, Davies K, Chagné D, Foster TM.. 2019. Two loci, RiAF3 and RiAF4, contribute to the annual-fruiting trait in Rubus. Frontiers in Plant Science 10, 1341. doi: 10.3389/fpls.2019.01341 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Jiménez S, Li Z, Reighard GL, Bielenberg DG.. 2010. Identification of genes associated with growth cessation and bud dormancy entrance using a dormancy-incapable tree mutant. BMC Plant Biology 10, 25. doi: 10.1186/1471-2229-10-25 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Jones-Rhoades MW, Bartel DP.. 2004. Computational identification of plant microRNAs and their targets, including a stress-induced miRNA. Molecular Cell 14, 787–799. doi: 10.1016/J.MOLCEL.2004.05.027 [DOI] [PubMed] [Google Scholar]
  37. Katsuya-Gaviria K, Caro E, Carrillo-barral N, Iglesias-fernández R.. 2020. Reactive oxygen species (ROS) and nucleic acid modifications during seed dormancy. Plants 9, 679. doi: 10.3390/plants9060679 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Kim SY, He Y, Jacob Y, Noh YS, Michaels S, Amasino R.. 2005. Establishment of the vernalization-responsive, winter-annual habit in Arabidopsis requires a putative histone H3 Methyl transferase. The Plant Cell 17, 3301–3310. doi: 10.1105/TPC.105.034645 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Kolberg L, Raudvere U, Kuzmin I, Adler P, Vilo J, Peterson H.. 2023. G:Profiler—interoperable web service for functional enrichment analysis and gene identifier mapping (2023 update). Nucleic Acids Research 51, W207–W212. doi: 10.1093/NAR/GKAD347 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Langfelder P, Horvath S.. 2008. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics 9, 559. doi: 10.1186/1471-2105-9-559 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Levy YY, Mesnage S, Mylne JS, Gendall AR, Dean C.. 2002. Multiple roles of Arabidopsis VRN1 in vernalization and flowering time control. Science 297, 243–246. doi: 10.1126/science.1072147 [DOI] [PubMed] [Google Scholar]
  42. Li H. 2013. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. ArXiv 1303.3997 [Preprint]. [Google Scholar]
  43. Love MI, Huber W, Anders S.. 2014. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology 15, 550. doi: 10.1186/s13059-014-0550-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Matzke MA, Mosher RA.. 2014. RNA-directed DNA methylation: an epigenetic pathway of increasing complexity. Nature Reviews Genetics 15, 394–408. doi: 10.1038/nrg3683 [DOI] [PubMed] [Google Scholar]
  45. Maurya JP, Bhalerao RP.. 2017. Photoperiod- and temperature-mediated control of growth cessation and dormancy in trees: a molecular perspective. Annals of Botany 120, 351–360. doi: 10.1093/aob/mcx061 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Mazzitelli L, Hancock RD, Haupt S, et al. 2007. Co-ordinated gene expression during phases of dormancy release in raspberry (Rubus idaeus L.) buds. Journal of Experimental Botany 58, 1035–1045. doi: 10.1093/jxb/erl266 [DOI] [PubMed] [Google Scholar]
  47. Milec Z, Strejčková B, Šafář J.. 2023. Contemplation on wheat vernalization. Frontiers in Plant Science 13, 1093792. doi: 10.3389/fpls.2022.1093792 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Mosedale JR, Abernethy KE, Smart RE, Wilson RJ, Maclean IMD.. 2016. Climate change impacts and adaptive strategies: lessons from the grapevine. Global Change Biology 22, 3814–3828. doi: 10.1111/gcb.13406 [DOI] [PubMed] [Google Scholar]
  49. Nicolas WJ, Grison MS, Bayer EM.. 2017. Shaping intercellular channels of plasmodesmata: the structure-to-function missing link. Journal of Experimental Botany 69, 91–103. doi: 10.1093/jxb/erx225 [DOI] [PubMed] [Google Scholar]
  50. Olsen JE. 2010. Light and temperature sensing and signaling in induction of bud dormancy in woody plants. Plant Molecular Biology 73, 37–47. doi: 10.1007/s11103-010-9620-9 [DOI] [PubMed] [Google Scholar]
  51. Oracz K, Stawska M.. 2016. Cellular recycling of proteins in seed dormancy alleviation and germination. Frontiers in Plant Science 7, 209910. doi: 10.3389/FPLS.2016.01128/BIBTEX [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Palacio S, Hoch G, Sala A, Körner C, Millard P.. 2014. Does carbon storage limit tree growth? New Phytologist 201, 1096–1100. doi: 10.1111/nph.12602 [DOI] [PubMed] [Google Scholar]
  53. Park S, Oh S, Ek-Ramos J, Nocker S van.. 2010. PLANT HOMOLOGOUS TO PARAFIBROMIN is a component of the paf1 complex and assists in regulating expression of genes within H3K27ME3-enriched chromatin. Plant Physiology 153, 821–831. doi: 10.1104/PP.110.155838 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Patro R, Duggal G, Love MI, Irizarry RA, Kingsford C.. 2017. Salmon provides fast and bias-aware quantification of transcript expression. Nature Methods 14, 417–419. doi: 10.1038/nmeth.4197 [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Price RJ, Davik J, Fernandéz FF, et al. 2023. Chromosome-scale genome sequence assemblies of the ‘Autumn Bliss’ and ‘Malling Jewel’ cultivars of the highly heterozygous red raspberry (Rubus idaeus L.) derived from longread Oxford nanopore sequence data. PLoS One 18, e0285756. doi: 10.1371/journal.pone.0285756 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Rinne PLH, Welling A, Vahala J, Ripel L, Ruonala R, Kangasjärvi J, van der Schoot C.. 2011. Chilling of dormant buds hyperinduces FLOWERING LOCUS T and recruits GA-inducible 1,3-β-glucanases to reopen signal conduits and release dormancy in Populus. The Plant Cell 23, 130–146. doi: 10.1105/tpc.110.081307 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Ríos G, Leida C, Conejero A, Badenes ML.. 2014. Epigenetic regulation of bud dormancy events in perennial plants. Frontiers in Plant Science 5, 247. doi: 10.3389/fpls.2014.00247 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, Smyth GK.. 2015. Limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Research 43, e47. doi: 10.1093/NAR/GKV007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Rothkegel K, Sandoval P, Soto E, Ulloa L, Riveros A, Lillo-Carmona V, Cáceres-Molina J, Almeida AM, Meneses C.. 2020. Dormant but active: chilling accumulation modulates the epigenome and transcriptome of Prunus avium during bud dormancy. Frontiers in Plant Science 11, 1115. doi: 10.3389/fpls.2020.01115 [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Singh M, Singh S, Randhawa H, Singh J.. 2013. Polymorphic homoeolog of key gene of RdDM pathway, ARGONAUTE4_9 class is associated with pre-harvest sprouting in wheat (Triticum aestivum L.). PLoS One 8, e77009. doi: 10.1371/journal.pone.0077009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Soneson C, Love MI, Robinson MD.. 2015. Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences. F1000Research 4, 1521. doi: 10.12688/f1000research.7563.1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Sønsteby A, Heide OM.. 2014. Cold tolerance and chilling requirements for breaking of bud dormancy in plants and severed shoots of raspberry (Rubus idaeus L.). Journal of Horticultural Science and Biotechnology 89, 631–638. doi: 10.1080/14620316.2014.11513131 [DOI] [Google Scholar]
  63. Sutherland L, Pearson S, Jennings N.. 2019. Modelling the dormancy and chilling requirements in raspberry. Coventry: Agriculture and Horticulture Development Board. https://horticulture.ahdb.org.uk/knowledge-library/modelling-the-dormancy-and-chilling-requirements-in-raspberry [Google Scholar]
  64. Tarancón C, González-Grandío E, Oliveros JC, Nicolas M, Cubas P.. 2017. A conserved carbon starvation response underlies bud dormancy in woody and herbaceous species. Frontiers in Plant Science 8, 254142. doi: 10.3389/FPLS.2017.00788/BIBTEX [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Tarasov A, Vilella AJ, Cuppen E, Nijman IJ, Prins P.. 2015. Sambamba: fast processing of NGS alignment formats. Bioinformatics 31, 2032–2034. doi: 10.1093/bioinformatics/btv098 [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Tixier A, Gambetta GA, Godfrey J, Orozco J, Zwieniecki MA.. 2019. Non-structural carbohydrates in dormant woody perennials; the tale of winter survival and spring arrival. Frontiers in Forests and Global Change 2, 18. doi: 10.3389/ffgc.2019.00018 [DOI] [Google Scholar]
  67. Tixier A, Orozco J, Roxas AA, Earles JM, Zwieniecki MA.. 2018. Diurnal variation in nonstructural carbohydrate storage in trees: remobilization and vertical mixing. Plant Physiology 178, 1602–1613. doi: 10.1104/pp.18.00923 [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Tognacca RS, Botto JF.. 2021. Post-transcriptional regulation of seed dormancy and germination: current understanding and future directions. Plant Communications 2, 100169. doi: 10.1016/J.XPLC.2021.100169 [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Velappan Y, Chabikwa TG, Considine JA, Agudelo-Romero P, Foyer CH, Signorelli S, Considine MJ.. 2022. The bud dormancy disconnect: latent buds of grapevine are dormant during summer despite a high metabolic rate. Journal of Experimental Botany 73, 2061–2076. doi: 10.1093/jxb/erac001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Yan Y, Shen L, Chen Y, Bao S, Thong Z, Yu H.. 2014. A MYB-domain protein EFM mediates flowering responses to environmental cues in Arabidopsis. Developmental Cell 30, 437–448. doi: 10.1016/j.devcel.2014.07.004 [DOI] [PubMed] [Google Scholar]
  71. Zhang H, Ransom C, Ludwig P, Nocker SV.. 2003. Genetic analysis of early flowering mutants in Arabidopsis defines a class of pleiotropic developmental regulator required for expression of the flowering-time switch FLOWERING LOCUS C. Genetics 164, 347–358. doi: 10.1093/genetics/164.1.347 [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Zhang Z, Schwartz S, Wagner L, Miller W.. 2000. A greedy algorithm for aligning DNA sequences. Journal of Computational Biology 7, 203–214. [DOI] [PubMed] [Google Scholar]
  73. Zheng J, Chen F, Wang Z, Cao H, Li X, Deng X, Soppe WJJ, Li Y, Liu Y.. 2012. A novel role for histone methyltransferase KYP/SUVH4 in the control of Arabidopsis primary seed dormancy. New Phytologist 193, 605–616. doi: 10.1111/j.1469-8137.2011.03969.x [DOI] [PubMed] [Google Scholar]
  74. Zhou JX, Liu ZW, Li YQ, Li L, Wang B, Chen S, He XJ.. 2018. Arabidopsis PWWP domain proteins mediate H3K27 trimethylation on FLC and regulate flowering time. Journal of Integrative Plant Biology 60, 362–368. doi: 10.1111/jipb.12630 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

erae371_suppl_Supplementary_Tables_S1-S14_Datasets_S1-S3
erae371_suppl_supplementary_tables_s1-s14_datasets_s1-s3.zip (896.4KB, zip)

Data Availability Statement

The sequencing data from the RNA-seq experiment, as well as the genomic DNA from ‘Glen Ample’, ‘Glen Dee’, and ‘Glen Mor’ are deposited in the European Nucleotide Archive under the project accession PRJEB77730. The phenotypic data can be made available upon request to the authors, as well as the plant material provided the appropriate licenses from James Hutton Ltd.

Articles from Journal of Experimental Botany are provided here courtesy of Oxford University Press

RESOURCES