Principal component (PC) analysis using preprocessed and normalized gene counts from genome-guided transcriptome assembly data of control (CT) and 12 (Trt12), 24 (Trt24), and 48 (Trt48) h of cold acclimation of African bermudagrass genotype OKC1163.
Volcano plots show the distribution and the number of the differentially expressed genes under 12 h of cold acclimation (A), 24 h of cold acclimation (B), and 48 h of cold acclimation (C). Upregulated and downregulated genes are shown in blue and red, respectively.
UpSet plot showing the comparison of unique number of upregulated and downregulated genes after 12, 24, and 48 h of cold acclimation of African bermudagrass genotype OKC1163.
Top significantly enriched Gene Ontology terms in the set of differentially expressed genes of 12-h cold acclimation vs. control. (A) Set of upregulated genes. (B) Set of downregulated genes. BP = biological process, CC = cellular component, MF = molecular function.
Top significantly enriched Gene Ontology terms in the set of differentially expressed genes of 24-h cold acclimation vs. control. (A) Set of upregulated genes. (B) Set of downregulated genes. BP = biological process, CC = cellular component, MF = molecular function.
Top significantly enriched Gene Ontology terms in the set of differentially expressed genes of 48-h cold acclimation vs. control. (A) Set of upregulated genes. (B) Set of downregulated genes. BP = biological process, CC = cellular component, MF = molecular function.
Venn diagram of common and unique set of differentially expressed genes across three cold acclimation treatments (12, 24, and 48 h) located in previously identified consistent quantitative trait loci regions reported by Yu et al. (2022). (A) Upregulated genes. (B) Downregulated genes. CT = Control.
Click on author name to view affiliation information
African bermudagrass (Cynodon transvaalensis Burtt Davy) is a fine-texture warm-season grass that often crosses with common bermudagrass [C. dactylon (L.) Pers] to create turf-type hybrid bermudagrass. African bermudagrass can adapt to colder climates than its origin in the southwestern Transvaal in South Africa, which has been used to improve winter hardiness of interspecific hybrid bermudagrass cultivars. However, the underlying genomic mechanisms that respond to cold temperatures are not fully elucidated in the species. Accordingly, the objective of this study was to identify differentially expressed genes (DEGs) in African bermudagrass genotype ‘OKC1163’ during different stages of cold acclimation. Plants were exposed to 8/2 °C day and night (12/12 hour) light cycles as cold acclimation treatment, while the control group was 24/20 °C with 12/12 hour day and night light cycles. Leaf tissues from three replications were collected at 12, 24, and 48 hours of treatment and from the control to identify DEGs. Compared with the untreated control, a total of 1707 genes were differentially expressed at 12 hours, including 967 upregulated and 740 downregulated genes, respectively. At 24 hours, a total of 2277 genes were differentially expressed, including 1140 upregulated and 1137 downregulated genes. At 48 hours, a total of 2691 genes were differentially expressed, including 1429 upregulated and 1262 downregulated genes. The most significant Gene Ontology enrichments were phosphorus metabolic process and response to abscisic acid, crucial biological functions that play significant roles in the adaptive response of plants to cold temperature stress. Integrating this information with previous winter hardiness quantitative trait loci, several candidate genes were consistently identified to be associated with cold acclimation, such as multiple cytochrome P450 and several ethylene-responsive transcription factors. This study offers insights into the transcriptional changes occurring during cold acclimation in African bermudagrass that can be used in designing molecular markers for selecting winter-hardy bermudagrass genotypes.
African bermudagrass (Cynodon transvaalensis Burtt Davy) originated in the Transvaal Valley in South Africa and was first documented by botanist Joseph Burtt Davy (Harlan 1971). African bermudagrass has unique characteristics, such as fine-textured leaves, slender stolons, short internodes, yellowish-green color, and small stature but high-density sod compared with other Cynodon spp. (Harlan et al. 1970; Juska and Hanson 1964). Although African bermudagrass can be used on golf course putting greens and tennis courts, its high water and nutrient demands, reduced turf quality in summer, and purple color under cool temperatures limit its use as turfgrass. The primary use of African bermudagrass, however, is to cross with common bermudagrass [C. dactylon (L.) Pers] to produce turf-type hybrid bermudagrasses, from which many modern high-quality cultivars have been developed (Schwartz et al. 2018; Wu et al. 2020).
Turf-type bermudagrasses are widely esteemed in the turf sector due to their high turfgrass quality, rapid establishment, and extraordinary ability to tolerate abiotic stresses such as traffic, heat, salt, and drought (Gerken 1994). Due to their advantageous characteristics, hybrid bermudagrasses are the preferred choice for sports fields, golf courses, and home lawns in the southern and transitional climatic regions of the United States (Beard 1973; Harlan et al. 1970). However, when grown in the transitional climatic zone, particularly at latitudes higher than 36°N, turf-type bermudagrasses are susceptible to low-temperature winterkill, resulting in a loss of use and revenue and substantial cost to re-establish turf. Therefore, using enhanced winter-hardy bermudagrasses would greatly benefit turfgrass management in the transitional climatic region (Gatschet et al. 1994; Rutledge et al. 2009; Taliaferro 1995). As a result, breeding programs have focused on developing bermudagrass cultivars that possess enhanced tolerance to freezing temperatures and outstanding turfgrass quality characteristics (Anderson and Taliaferro 2002; Dunne et al. 2019). Bermudagrass cultivars, such as ‘OKC1119’ (marketed as Latitude 36®), ‘OKC1134’ (NorthBridge®), ‘OKC1131’ (Tahoma 31®), and ‘OSU3920’ were released by Oklahoma State University with improved winter hardiness and expanded the adaption region of bermudagrass further north to USDA plant cold hardiness zones 6a and 6b (Wu et al. 2012a, 2012b, 2020, 2023) and 5a and 5b in recent years. Developing winter-hardy high-quality interspecific hybrid bermudagrass relies on elite parents. Generally, it is presumed the common bermudagrass parent contributes to winter hardiness, while the African bermudagrass parent contributes to turfgrass quality characteristics. However, this generalization may not be warranted due to inherently good winter hardiness in African bermudagrass, as well as genetic variations associated with winter hardiness within the species (Dunne et al. 2019; Harlan et al. 1970; Yu et al. 2022).
In plants, the process of physiological changes in response to low, but nonfreezing temperatures, refers to cold acclimation. During cold acclimation, plants undergo a series of regulation of specific genes and metabolic changes that lead to different biochemical, physiological, and morphological adjustments (Guo et al. 2018; Satyakam et al. 2022). Cold acclimation is an essential step for the development of freezing tolerance in bermudagrass (Janská et al. 2010). Physiological changes, such as accumulation of osmoprotectants including proline, soluble sugars, and nonstructural carbohydrates like starch and fructans, play an important role in protecting plant cells by stabilizing membranes and proteins during freezing stress. These molecules serve as energy reserves during cold stress and recovery. Also, during cold acclimation, plants can alter the lipid composition of the cell membrane, making the cell membrane maintain integrity under cold stress (Fontanier et al. 2020; Samala et al. 1998). Prolonged exposure to cold before freezing temperatures allows for more comprehensive adjustments, significantly increasing cold tolerance (Janská et al. 2010). Limited information related to cold acclimation timing has been studied in bermudagrass. Therefore, determining the optimal length of cold acclimation period for bermudagrass is critical for maximizing cold tolerance.
A complex array of molecular mechanisms triggered by low nonfreezing temperatures is critical to mitigate damage and ensure plant survival at freezing temperatures (Brown et al. 2023; Fontanier et al. 2020; Satyakam et al. 2022). Previous studies have found that one of the primary responses under cold stress in Arabidopsis is the upregulation of cold-responsive genes mediated by specific transcription factors, including C-repeat binding factors (CBFs) and dehydration-responsive element–binding transcription factors (Thomashow et al. 1997; Zhang and Xia 2023). Also, it has been found that signal transduction pathways involving calcium ions (Ca2+), mitogen-activated protein kinases, antifreeze proteins, heat shock proteins, cold shock proteins, and reactive oxygen species (ROS) are important in initiating signaling cascades that regulate gene expression and metabolic changes that collectively allow plants to preserve cellular integrity and metabolic equilibrium when faced with low temperature stress (Eskandari et al. 2020; Sadura and Janeczko 2024; Sasaki and Imai 2012; Satyakam et al. 2022).
Several studies have investigated transcriptome profiling in common bermudagrass under cold acclimation, identifying key genes, including the NAC (NAM, ATAF1/2 and CUC2), CBF, and WRKY transcription factor families, which are significantly upregulated and play crucial roles in cold stress responses (Hu et al. 2020; Huang et al. 2017, 2022; Zhu et al. 2015). NAC transcription factors, one of the largest groups of plant-specific regulators, are significantly upregulated in common bermudagrass during cold acclimation and downregulated in nonacclimated plants (Nuruzzaman et al. 2015; Zhu et al. 2015). Similarly, CBF genes play a crucial role in cold tolerance. Overexpression of the CBF gene in transgenic Arabidopsis thaliana has been demonstrated to significantly enhance cold tolerance, whereas suppression of CBF1 reduces low-temperature tolerance (Sun et al. 2008). Zhu et al. (2015) also reported that CBF transcription factors are overexpressed during cold acclimation. In addition, in common bermudagrass, the cold-induced WRKY transcription factor CdWRKY2 has been identified as a key regulator of the cold stress response. This gene functions by coordinately activating pathways involved in sucrose synthesis and cold-responsive signaling via CdCBF1 during cold stress (Huang et al. 2022). These findings suggest that transcription factors are key players in the molecular mechanisms that facilitate common bermudagrass to acclimate to cold conditions. However, little information is available for African bermudagrass.
Transcriptomic approach is effective in understanding the molecular pathways and identifying differentially expressed genes (DEGs) under cold acclimation. However, the large amount of data provided limits the utilization of critical information in plant improvement. Integration of transcriptome and quantitative trait loci (QTL) mapping has been a method to identify potential candidate genes because traditional map-based cloning using near isogenic line population is impossible in open pollinate species (Jaganathan et al. 2020). Recently, Brown et al. (2023) conducted a transcriptomic study of the response to low temperature in zoysiagrass and verified results from previous QTL mapping and proteomics studies. In this study, many DEGs were identified under cold acclimation. Specifically, genes involved in pathways, such as photosynthesis and plant hormone signal transduction, are linked to cold response identified from the winter-hardiness QTL region (Brown et al. 2023). Our previous studies on a winter-hardy African bermudagrass ‘OKC1163’–derived population using a forward genetic approach successfully identified genomic regions associated with winter survivability traits (Yu et al. 2022). However, due to the relatively small mapping population size and winter survivability being highly influenced by environments, the relatively large QTL intervals hinder the identification of candidate genes and their application in developing molecular markers in marker-assisted selection. Additionally, a significant gap in transcriptomic resources remains, particularly concerning the mechanisms underlying cold acclimation in African bermudagrass. Therefore, the objectives of this study were 1) to identify DEGs in response to cold acclimation in African bermudagrass using RNA-sequencing profiling and differential gene expression analysis and 2) to identify potential candidate genes by integrating transcriptome data with previous QTL mapping results.
African bermudagrass genotype ‘OKC1163’ was selected because it showed better winter hardiness compared with long-time industry standard ‘Tifway’ and ‘Latitude 36®’ bermudagrass (National Turfgrass Evaluation Program 2017). Sprigs were transplanted into 12 plastic pots (12.7 × 12.7 × 12.7 cm) using soilless media (Berger BM2, Berger, QC, Canada) in Ridge Road greenhouse facilities at Oklahoma State University (OSU) in Stillwater, OK. The greenhouse was set with a day/night temperature of 32/28 °C, with a 12-h photoperiod (0600 to 1800 HR) providing 900 mmol·m−2·s−1 photosynthetically active radiation (PAR). The plants were watered daily to promote establishment and trimmed at 2.54-cm height weekly. Water soluble fertilizer (20N–20P–20K; Jack’s Professional Water-Soluble Fertilizer; J.R. Peters, Inc., Allentown, PA) was applied every 2 weeks using a cylinder with at a rate of 85 ml of the fertilizer solution containing 2 g·L−1 of the product. After 4 months of establishment, the plants were transferred to a growth chamber (PGC Flex growth chamber; Conviron, Winnipeg, MB, Canada) in the Controlled Environment Research Laboratory at OSU. The temperature in the growth chamber was adjusted to 24/20 °C day/night, with a 12-h photoperiod (0600 to 1800 HR) at a PAR of 900 mmol·m−2·s−1 as the control. After 2 weeks of preacclimation, the treatment plants were transferred to a chamber exposed to 8/2 °C day and night (12/12 h) light cycles with 900·mmol·m−2·s−1 PAR for cold acclimation treatment. Three plants were first transferred to the acclimation chamber, and another three plants were transferred after 24 and 36 h, respectively. Twelve hours later, about 50 mg of whole leaf tissue evenly collected from three replications were sampled from plants subjected to 12 h (Trt12), 24 h (Trt24), and 48 h (Trt48) of cold acclimation treatment in the acclimation chamber and control (CT) from plants in the preacclimation chamber. Leaf tissues were immediately frozen in liquid nitrogen and placed into 2-mL centrifuge tubes and stored at −80 °C until RNA extraction.
Total RNA was extracted from frozen leaves using the RNeasy plant mini kit (Qiagen, Hilden, Germany) following the manufacturer’s instructions. After quality (agarose gel electrophoresis), integrity (Agilent 2100; Agilent, Santa Clara, CA, USA), and quantity (NanoDrop ND-1000; NanoDrop, Wilmington, DE, USA) checks, mRNA was purified from total RNA using poly-T oligo-attached magnetic beads and fragmented in a fragmentation buffer. After fragmentation, the first cDNA was synthesized using random hexamer primers and reverse transcriptase, followed by the second strand cDNA synthesis with dUTPs. The library was completed following purification, end repair, Poly A tailing, adapter ligation, size selection, enzyme digestion, amplification, and purification. Before sequencing, library quality was checked with Qubit fluorometer and real-time polymerase chain reaction for quantification and bioanalyzer for size distribution detection using an Agilent 5400. Quantified libraries were pooled and sequenced on an Illumina HiSeq System (Illumina Inc., San Diego, CA, USA) to generate 150-bp pair-ended sequences per each treatment and replicate. The library preparation and sequencing were performed by Novogene (Novogene Corporation Inc., Sacramento, CA, USA).
To remove the adapters and low-quality reads of each RNA raw read sample in fastq format, the program fastp was used (Chen 2023). All the downstream analyses were based on clean data with high quality. High-quality and trimmed RNA-sequencing reads were mapped to the African bermudagrass reference genome using HISAT2 v2.0.5 (Cui et al. 2021; Kim et al. 2015). Alignment outputs of each sample generate a sequence alignment map in .sam format that was converted to a binary alignment map in a .bam format for space efficiency. All mapped reads of each sample were assembled in a reference-based approach using StringTie, version v1.3.3b (Pertea et al. 2015). To estimate and quantify the transcript expression levels, reads that were mapped to each gene were counted using the program featureCounts v1.5.0-p (Liao et al. 2014). Fragments per kilobase of transcript per million fragments mapped of each transcript was calculated based on the length of the transcript and reads count mapped to this gene. Differential gene expression analysis between two conditions was performed using R with the DESeq2 package, version 1.20.0 (Varet et al. 2016). Genes were considered differentially expressed based on a 5% false discovery rate, a P value of ≤0.05, and a log fold change of 2.
Gene Ontology (GO) enrichment analysis of highly DEGs was implemented by using the clusterProfiler R package, in which gene length bias was corrected. GO terms with corrected P values less than 0.05 were considered significantly enriched by DEGs (Yu et al. 2012). GO categories, including biological process, cellular component, and molecular function, were selected for the enrichment analysis. ShinyGO 0.77 graphical tool was also used to test statistical enrichment of DEGs (Ge et al. 2020). GO enrichment bubble plots were generated using the graphical interface SRplot (Tang et al. 2023).
To functionally annotate the genes and novel transcripts generated after transcriptome reconstruction, transcripts in .fasta format were extracted from the reference annotation file and reference genome file of African bermudagrass using TrasnDecoder program. Ten QTL associated with winter survivability and spring green-up traits were identified in the previous study (Yu et al. 2022). Two genomic regions overlapped with consistently identified spring green-up and winterkill QTL, QCTSG3 and QCTWK2 on chromosome 3 (54.0 to 55.4 Mbp) and QCTSG1 and QCTWK1 on chromosome 9 (15.2 to 17.6 Mbp) were selected to annotate identified DEGs. In addition, another consistent spring green-up QTL, QCTSG2 on chromosome 1 (2.2 to 4.7 Mbp) and QCTSG5 and QCTWK3 overlapped region on chromosome 4 (6.0 to 6.7 Mbp) were searched for DEGs. All searching intervals were determined based on the physical location of flanking peak logarithm of the odds value single nucleotide polymorphisms. Differentially expressed genes within these two genomic regions were extracted and compared with the National Center for Biotechnology Information (NCBI) nucleotide nr database using Blast-2.10.1+ with default settings on the OSU Pete High-Performance Computer. Predicted proteins/genes were searched against the NCBI, Swissport, and UniPort databases for functional annotation.
Illumina sequencing yielded an average of 61,709,316 raw reads across replicates for the control treatment. For the plants exposed to cold acclimation, averages of 60,897,046, 58,845,356, and 40,747,918 raw reads across replicates were yielded for the samples collected at 12, 24, and 48 h, respectively. After filtering and removing low-quality bases and adapter sequences, the data set was refined, resulting in a set of 58,360,646 (94.6%) read sequences averaged across three replicates for the control treatment. Filtered read sequences corresponding to the plants exposed to cold acclimation across replicates were 58,350,670 (95.8%), 58,108,190 (98.7%), and 40,462,368 (99.3%) reads for the samples collected at 12, 24, and 48 h, respectively.
To generate the genome-guided transcriptome assembly and the predicted transcript abundances, the program StringTie was used. A total of 31,030 transcripts were assembled in a single master transcriptome from all the libraries, including CT, Trt12, Trt24, and Trt48. Based on the number of mapped reads, the correspondence assembled genes were 28,444, and a total of 2586 new transcripts were discovered.
A principal component analysis plot (PCA) was generated using normalized gene counts from the genome-guided transcriptome assembly data to visualize patterns and correlations among treatments (Fig. 1). There is a clear separation in the gene count data between control treatment (CT) and the gene count data of samples under cold acclimation collected at 12, 24, and 48 h. A total of 56.43% of the variation between samples was explained by principal component 1, indicating that CT showed significant differences compared with Trt12, Trt24, and Trt48 (Fig. 1). principal component 2, the second most significant source of variation, explains 26.03% of the total variance in the gene count data between treatments. The normalized gene count data corresponding to the cold acclimation treatments showed less variation compared with the normalized gene count data compared with the control treatment (Fig. 1).
Citation: HortScience 60, 5; 10.21273/HORTSCI18443-25
Differentially expressed genes were identified by comparing the CT vs. each cold acclimation treatment. The numbers of identified nonredundant annotated DEGs were 1707, 2277, and 2691 for Trt12 vs. CT, Trt24 vs. CT, and Trt48 vs. CT, respectively (Supplemental Table 1). A total of 740 and 967 genes, 1137 and 1140 genes, and 1262 and 1429 genes were downregulated and upregulated, respectively, for the plants exposed for 12, 24, and 48 h of cold acclimation, as compared with the control (Fig. 2). The volcano plots in Fig. 2 illustrate the differential expression of genes in plants exposed to cold acclimatization for Trt12 (Fig. 2A), Trt24 (Fig. 2B), and Trt48 (Fig. 2C). Each point represents a gene, with the x-axis indicating the log2 fold change in expression and the y-axis showing the −log10 of the adjusted P value, highlighting the statistical significance of the expression changes.
Citation: HortScience 60, 5; 10.21273/HORTSCI18443-25
Unique and shared DEGs among three cold acclimation treatments are shown in Fig. 3. The largest intersection (513 genes) occurs in the combination of the upregulated genes of Trt12, Trt24, and Trt48. A total of 477 unique downregulated genes of Trt48 represent the second largest group followed by 460 unique upregulated genes expressed in Trt48 and unique downregulated genes expressed in Trt24 (Fig. 3). Variations and relationships among upregulated and downregulated genes between treatments highlight both the extent of shared genetic responses and the specificity of gene expression changes, providing insights into the molecular mechanisms of cold acclimation.
Citation: HortScience 60, 5; 10.21273/HORTSCI18443-25
Differentially expressed genes were categorized into enriched GO terms to describe functional annotations for their biological process (BP), cellular component (CC), and molecular function (MF). For the upregulated genes expressed in Trt12 vs. CT, a total of 199, 52, and 71 significant GO terms were designated for BP, CC, and MF, respectively. The main categories for BP were phosphorus metabolic process (GO:0006793), response to organic substance (GO:0010033), defense response (GO:0006952), and response to abscisic acid (GO:0009737) (Fig. 4A). The most significantly enriched GO terms for CC were chloroplast (GO:0009507), organelle subcompartment (GO:0031984), and plasmodesma (GO:0009506). The most significantly enriched GO terms for MF were catalytic activity (GO:0140096), phosphotransferase activity (GO:0016773), and kinase activity (GO:0016301) (Fig. 4A). For the downregulated genes expressed in Trt12 vs. CT, a total of 137, 46, and 56 significant GO terms were designated for the BP, CC, and MF, respectively. The main categories for BP were lipid metabolic process (GO:0006629), response to abiotic stimulus (GO:0009628), and proteolysis (GO:0006508). The most significantly enriched GO terms for CC were chloroplast envelope (GO:0009941), plastid envelope (GO:0009526), and chloroplast thylakoid (GO:0009534). The most significantly enriched GO terms for MF were catalytic (GO:0140096), transporter (GO:0005215), and peptidase activity (GO:0008233) (Fig. 4B).
Citation: HortScience 60, 5; 10.21273/HORTSCI18443-25
For the upregulated genes expressed in Trt24 vs. CT, a total of 292, 44, and 96 significant GO terms were designated for the BP, CC, and MF, respectively. The main categories for BP were protein phosphorylation (GO:0006468), response to organic substance (GO:0010033), response to abiotic stimulus (GO:0009628), and response to endogenous stimulus (GO:0009719) (Fig. 5A). The most significantly enriched GO terms for CC were endomembrane system (GO:0012505), cell–cell junction (GO:0005911), and plasmodesma (GO:0009506). The most significantly enriched GO terms for MF were catalytic (GO:0140096), protein kinase (GO:0004672), and phosphotransferase activity (GO:0016773) (Fig. 5A).
Citation: HortScience 60, 5; 10.21273/HORTSCI18443-25
For the downregulated genes expressed in Trt24 vs. CT, a total of 140, 66, and 43 significant GO terms were designated for BP, CC, and MF, respectively. The main categories for BP were protein photosynthesis (GO:0019684), response to light stimulus (GO:0009416), organic substance transport (GO:0071702), and response to abiotic stimulus (GO:0009628) (Fig. 5B). The most significantly enriched GO terms for CC were chloroplast (GO:0009507), chloroplast thylakoid (GO:0009534), and plastid thylakoid (GO:0031976). The most significantly enriched GO terms for MF were catalytic activity (GO:0140096), transporter activity (GO:0005215), and cytoskeletal protein binding (GO:0008092) (Fig. 5B).
For the upregulated genes expressed in Trt48 vs. CT, a total of 394, 68, and 114 significant GO terms were designated for the BP, CC, and MF, respectively. The main categories for BP were phosphorus metabolic process (GO:0006793), response to oxygen (GO:1901700), response to abiotic stimulus (GO:0009628), and response to endogenous stimulus (GO:0009719) (Fig. 6A). The most significantly enriched GO terms for CC were endomembrane system (GO:0012505), Golgi apparatus (GO:0005794), and bounding membrane (GO:0098588).
Citation: HortScience 60, 5; 10.21273/HORTSCI18443-25
The most significantly enriched GO terms for MF were catalytic activity (GO:0140096), adenyl nucleotide binding (GO:0030554), and ATP binding (GO:0005524) (Fig. 6A). For the downregulated genes expressed in Trt48 vs. CT, a total of 166, 61, and 81 significant GO terms were designated for BP, CC, and MF, respectively. The main categories for BP were nucleic acid metabolic process (GO:0090304), photosynthesis (GO:0015979), transmembrane transport (GO:0055085), and lipid metabolic process (GO:0006629) (Fig. 6B). The most significantly enriched GO terms for CC were chloroplast (GO:0009507), plastid thylakoid (GO:0031976), and organelle subcompartment (GO:0031984). The most significantly enriched GO terms for MF were catalytic activity (GO:0140096), transporter activity (GO:0005215), and isomerase activity (GO:0016853) (Fig. 6B). More details of GO enrichment analysis are described in Supplemental Table 2.
A total of 59, 94, and 110 DEGs were found under 12-, 24-, and 48-h cold acclimation periods, respectively, from four previously identified consistent QTL regions associated with winter hardiness (Table 1). There is a progressive increase in both upregulated and downregulated genes as the duration of cold acclimation increases from 12 to 48 h in QTL regions on chromosomes 3, 4, and 9. This suggests that the gene expression response becomes more pronounced with prolonged cold exposure, but on chromosomes 3 and 9, there were higher numbers of differentially expressed genes (both upregulated and downregulated) across all time points compared with chromosome 4. However, in the QTL region on chromosome 1, there was an increased number of DEGs at 24 h, but the gene expression response decreased at 48 h.
A total of 30 and 7 common genes located in the previously mentioned QTL regions were consistently upregulated and downregulated, respectively, across all three treatments (Table 2). Certain genes are uniquely upregulated or downregulated at specific time points, indicating the dynamic nature of gene regulation in response to cold acclimation over time (Fig. 7). Common upregulated genes located within QTL regions across three acclimation periods were associated with diverse biological functions, as determined by functional annotation (Table 2). CYQ32 and C94C1 encode cytochrome P450 81Q32 and cytochrome P450 94C1, respectively, which are involved in the oxidative metabolism of various substrates. The NPR4 gene encodes an ankyrin repeat-containing protein, NPR4, implicated in protein–protein interactions and signal transduction pathways. AKRC9 encodes a chloroplast-localized NADPH-dependent aldo-keto reductase, which plays a crucial role in reducing aldehydes and ketones to maintain cellular redox balance. URT1 encodes a putative UDP-rhamnose synthase, a key enzyme in glycosylation processes. The IPCS gene encodes phosphatidylinositol inositolphosphotransferase, essential for the synthesis of complex sphingolipids. CSE encodes caffeoylshikimate esterase, an enzyme central to the lignin biosynthesis pathway, which has been proposed to play important roles in signal transduction, membrane stability, host–pathogen interactions, and stress responses (Sperling and Heinz 2003). ERG1 encodes an elicitor-responsive protein involved in plant defense mechanisms. FER1 encodes ferredoxin-1, which is critical for electron transfer during photosynthesis, while LAC6 encodes laccase-6, a multicopper oxidase likely involved in lignin degradation and other oxidative reactions (Table 2).
Citation: HortScience 60, 5; 10.21273/HORTSCI18443-25
Exposure to low but nonfreezing temperatures, known as cold acclimation, is crucial for developing and enhancing freezing tolerance in plants (Guy 1990; Thomashow 1998). Bermudagrass undergoes intricate molecular regulatory networks throughout the process of adapting to chilling temperature. Specialized receptors on the cell membrane detect low temperature signals, which then activate a complex cascade of reactions that involve thousands of genes and changes in their expression levels. This sequence of events triggers further physiological reactions and the activation of genes that respond to stress, resulting in the development of defensive adaptations throughout the plant (Lei et al. 2014; Varshney et al. 2011; Zhu et al. 2015). Although previous studies evaluated the winter hardiness and reported QTL associated with winter hardiness, the genomic and molecular responses of African bermudagrass to low temperatures at the molecular level remain unclear (Dunne et al. 2019; Yu et al. 2022). In this study, we used the winter-hardy African bermudagrass genotype ‘OKC1163’, which demonstrated 50% survival after exposure to −26.1 °C air temperature (Yu et al. 2022). This study investigated gene profiling under cold acclimation using leaf tissues, providing valuable insights into understanding molecular mechanisms of winter hardiness in African bermudagrass.
To identify biological pathways that govern the responses of differentially expressed genes to cold stress, GO category enrichment analysis was performed. GO analysis indicated that the most significant GO terms “phosphorus metabolic process,” “response to abiotic stimulus,” “response to abscisic acid,” and “response to endogenous stimulus,” were shared among upregulated genes expressed under three cold acclimation stages (12, 24, and 48 h). These shared biological processes highlight the complex and interconnected nature of cold acclimation in African bermudagrass, in which multiple pathways and signals work together to help the grass survive winter. Phosphorus is an important nutrient involved in energy transfer, signaling, and macromolecule biosynthesis such as nucleic acids and phospholipids (Cockefair 1931). During cold stress, phosphorus metabolism can be adjusted to regulate energy production and usage, allowing the plant to better cope with the reduced metabolic activity caused by low temperatures. Several studies have shown that phosphate improves the cold tolerance in plants (Gao et al. 2022; Nie et al. 2015; Schlüter et al. 2013; Wang et al. 2023). The GO term referred to the response to endogenous stimulus (GO:0009719) indicates any process that induces a modification in the condition or functioning of an organism or cell due to a stimulus that originates from within the organism (Zhang et al. 2018). These endogenous stimuli might include activation of hormones such as abscisic acid (ABA) and ethylene, changes in cellular redox status, or shifts in the levels of secondary metabolites that signal the need to adjust to cold conditions (Gusta et al. 2005; Lissarre et al. 2010; Shi and Yang 2014; Swamy and Smith 1999).
The most significant GO terms enriched among the downregulated genes expressed under all three cold acclimation stages (12, 24, and 48) included “lipid metabolic process” and “photosynthesis.” The functionality and survival of a plant cell under cold stress relies on membrane integrity. The thylakoid membrane structure consists of galactolipids including monogalactosyldiacylglycerol and digalactosyldiacylglycerol (Douce and Joyard 1990; Yadav 2010). The plasma membrane comprises phospholipids, such as phosphatidylcholines, phosphatidylethanolamines, phosphatidylglycerols, phosphatidylinositols, phosphatidylserines, phosphatidic acids, and glycerolipids (Bhattacharya 2022). Fontanier et al. (2020) reported an increase of galactolipids after a prolonged exposure to chilling stress in bermudagrass genotypes. Even though lipids have an important structural role in mitigating the impact of cold stress, the downregulation of genes involved in lipid metabolism might indicate a strategic reduction in the synthesis of certain lipids that are not essential under cold stress to conserve energy for more critical processes like producing stress proteins and other protective compounds (Bhattacharya 2022; Janská et al. 2010; Wei et al. 2006; Zhang et al. 2020). Plants under cold stress can inhibit the enzymes involved in photosynthesis, reducing the efficiency of light capture and causing oxidative stress due to the imbalance between energy absorption and utilization. The upregulation of carbon metabolism under low-temperature conditions has been linked to reduced inhibition of photosynthesis (Ensminger et al. 2006). Studies found that during the initial stages of cold stress in plants, there is a repression of the genes involved in photosynthesis (Huner et al. 1993; Peng et al. 2015).
The upregulation of genes associated with the GO term “cell wall macromolecule metabolic process” after 48 h of cold exposure in African bermudagrass highlights the critical role that the cell wall plays in plant response to cold stress and cold acclimation. The cell wall serves as both a physical barrier and a central center for cellular signaling. Adjustments in the cellular membrane can initiate signaling cascades that stimulate the expression of stress response genes, including those implicated in the process of adapting to cold temperatures (Rajashekar and Lafta 1996, Sasidharan et al. 2011; Takahashi et al. 2024).
Several of these genes have been implicated in the response to cold acclimation. For instance, common upregulated genes associated with cold stress in African bermudagrass among three treatments include evm.TU.LG03 homologous to RLK7, evm.TU.LG09.1604 homologous to RGLG1, and evm.TU.LG09.3085 homologous to HY5. RLK7 codes for a receptor-like protein kinase 7, which plays a crucial role in plant responses under cold stress (Chen et al. 2021; Ye et al. 2017). Genomic and transcriptome analysis in several plant species, including A. thaliana, wild soybean (Glycine soja), paper mulberry (Broussonetia papyrifera), and zoysiagrass (Zoysia spp. Willd.), revealed that RLK proteins play essential roles in plant growth, development, signal transduction, and stress adaptation, specifically cold stress (Chen et al. 2021; Su et al. 2022; Xuan et al. 2013; Yang et al. 2014). The protein E3 ubiquitin-protein ligase is encoded by the gene RGLG1, which functions as a positive regulator of ABA signaling (Shi and Yang 2014; Wu et al. 2016). ABA plays a crucial role in regulating various physiological processes, especially those related to stress responses, growth, and development. Under cold stress, the levels of ABA increase, and this accumulation activates downstream cold tolerance mechanisms, including gene regulation of cold-responsive genes, accumulation of osmoprotectants, and protein synthesis (Gusta et al. 2005). The gene long hypocotyl 5 (HY5) in A. thaliana regulates the expression of about one-third of genes in A. thaliana, and most of these genes might be involved in molecular responses to cold conditions (Perea-Resa et al. 2017). Zhang et al. (2020) found that silencing the HY5 gene in tomato (Solanum lycopersicum “Ailsa Craig”) myb15 mutants resulted in an increased sensitivity to the cold stress.
After 48 h of cold exposure, African bermudagrass upregulated unique genes associated to cold acclimation including evm.TU.LG03.3774, evm.TU.LG03.3873, and evm.TU.LG09.929. These genes identified in African bermudagrass are homologous to WRKY51, RBOHF, and PER11 in A. thaliana, respectively. WRKY51 codes for a transcription factor protein that plays a crucial role in response to cold stress in plants (Zhang et al. 2016). RNA sequencing showed that overexpressing the CdWRKY2 gene improved cold tolerance in transgenic Arabidopsis and bermudagrass hairy roots, whereas silencing CdWRKY2 expression increases cold susceptibility. These results indicate that CdWRKY2 functions as a positive regulator of cold stress tolerance by simultaneously activating sucrose synthesis pathways involving CdSPS1 and signaling pathways dependent on CdCBF1 (Huang et al. 2022). RBOHF is a gene that codes for respiratory burst oxidase homolog protein F in A. thaliana. This protein is involved in ABA-induced stomatal closing and ABA-ROS-dependent signaling (Desikan et al. 2006). PER11 is a gene that encodes peroxidase 11 in A. thaliana; these types of enzymes are important in response to environmental stress such as cold stress by reducing the accumulation of hydrogen peroxide (Nourredine et al. 2015). These outcomes align with the discoveries of the previous and current literature in other grasses and model plants. However, further analyses are needed to investigate the precise molecular pathways of cold-acclimated African bermudagrass at the gene and protein levels.
A typical approach for screening potential genes is to integrate QTL information with gene expression variations (Lin et al. 2019). In this study, previously characterized QTL regions associated with winter hardiness in African bermudagrass were used to identify potential candidate DEGs involved in cold-acclimation pathways. Through the examination of genome sequences in these specific QTL regions and the combination of the DEGs revealed in this study, a total of 170 genes distributed throughout four QTL researching regions were identified (Table S3). When examining the genes located in previously identified QTL areas that exhibited high abundance during cold acclimation at 12, 24, and 48 h, two consistently identified genes, evm.TU.LG03.4542 and evm.TU.LG03.4561, shared homology with CYQ32 and C94C1, respectively. These genes encode for cytochrome P450. Proteins from cytochrome P450 complex are related to the regulation of important cell processes that affect plant growth, development, and defense responses (Jun et al. 2015). Extreme temperature fluctuations can lead to disruption in respiration and photosynthesis, resulting in oxidative damage due to the generation of ROS. One of the main roles of cytochrome P450 enzymes is to regulate nonenzymatic antioxidants like carotenoids, flavonoids, and hormones, as well as in activating antioxidant enzymes (Pandian et al. 2020). Studies have shown that the genes CYP73A (transcinnamate 4-monooxygenase), CYP75A (flavonoid 3′,5′-hydroxylase), and CYP75B (flavonoid 3′-monooxygenase) were significantly upregulated in response to cold stress in perennial ryegrass (Lolium perenne L.) and tall fescue [Festuca arundinacea (Schreb.) Darbysh.] (Tao et al. 2017). A different transcriptome study revealed that cold-tolerant sorghum genotypes upregulated CYP99A1 and CYP709C1 genes under cold stress (Chopra et al. 2015).
Two DEGs, evm.TU.LG03.4609, which shared homology with the gene ERF110, and evm.TU.LG01.621, which is homologous to ERF60, were greatly upregulated after 12, 24, and 48 h of cold acclimation. These two genes are ethylene-responsive transcription factors that have been found to be related to abiotic responses including cold tolerance in Arabidopsis, grapevine (Vitis), rice (O. sativa), zoysiagrass, and Medicago falcata (Brown et al. 2023; Cheng et al. 2013; Sun et al. 2016; Tian et al. 2011; Zhuo et al. 2018). In common bermudagrass, studies have shown that silencing of the ethylene responsive factor CdERF1 reduced cold resistance (Hu et al. 2020). In addition, effects of ethylene on cold tolerance on bermudagrass and zoysiagrass have been documented (Hu et al. 2016; Zhang et al. 2022). These findings confirm that the ethylene-responsive transcription factors play roles in bermudagrass response to low temperature stress. In addition, several DEGs showed potential function in response to cold temperature stress. For instance, the ATP-binding cassette (ABC) transporter releases energy through binding and hydrolysis of ATP to achieve transmembrane transport of substrates, including peptides, sugars, amino acids, alkaloids, vitamins, inorganic molecules, organic molecules, glutathione, and cellular metabolites (Guo et al. 2022). The ABC transporter has been found to have regulated cold tolerance in maize (Zea mays L.) (Yang et al. 2024). There are two DEGs identified as cytochrome b561 and dopamine β-monooxygenase N-terminal (DOMON) domain–containing proteins (CYBDOMs). CYBDOMs are ferrireductases that regulate apoplastic iron deposition, which adjusts root growth and abiotic stress response (Clúa et al. 2024a, 2024b). CYBDOMs may play roles in energy dissipation systems that enable plants to withstand abiotic stress, such as salt stress in rice and drought stress in wild watermelon (Citrullus lanatus sp.) (Deng et al. 2023; Nanasato et al. 2005). It is likely that CYBDOMs identified in this study also modulate cold stress in African bermudagrass.
Although cold acclimation is a critical determinant of winter hardiness, the process involves multiple organs in African bermudagrass. Leaves play a key role in sensing cold temperatures and initiating physiological cascades that enable plants to develop hardiness (Knight and Knight 2012). However, winter survival and spring regrowth of bermudagrasses primarily depend on organs such as crowns, stolons, and rhizomes. However, most transcriptomic studies on bermudagrass under low temperatures have relied on leaf tissues for RNA extraction (Hu et al. 2020; Huang et al. 2017, 2022; Zhu et al. 2015). Future research should explore the responses of crowns, stolons, and rhizomes to cold acclimation and deacclimation to achieve a holistic understanding of the plant’s reaction to low temperatures.
This study identified numerous genes and pathways involved in the cold stress and acclimation response, providing valuable information about the molecular and genetic mechanisms behind cold acclimation in African bermudagrass. Additionally, these findings validated certain results of the previous QTL mapping studies, as the overlapping genes, such as cytochrome P450 and ethylene-responsive transcription factors, present significant potential for selection in breeding programs focused on improving cold tolerance in African bermudagrass by designing molecular markers used in the selection.
Principal component (PC) analysis using preprocessed and normalized gene counts from genome-guided transcriptome assembly data of control (CT) and 12 (Trt12), 24 (Trt24), and 48 (Trt48) h of cold acclimation of African bermudagrass genotype OKC1163.
Volcano plots show the distribution and the number of the differentially expressed genes under 12 h of cold acclimation (A), 24 h of cold acclimation (B), and 48 h of cold acclimation (C). Upregulated and downregulated genes are shown in blue and red, respectively.
UpSet plot showing the comparison of unique number of upregulated and downregulated genes after 12, 24, and 48 h of cold acclimation of African bermudagrass genotype OKC1163.
Top significantly enriched Gene Ontology terms in the set of differentially expressed genes of 12-h cold acclimation vs. control. (A) Set of upregulated genes. (B) Set of downregulated genes. BP = biological process, CC = cellular component, MF = molecular function.
Top significantly enriched Gene Ontology terms in the set of differentially expressed genes of 24-h cold acclimation vs. control. (A) Set of upregulated genes. (B) Set of downregulated genes. BP = biological process, CC = cellular component, MF = molecular function.
Top significantly enriched Gene Ontology terms in the set of differentially expressed genes of 48-h cold acclimation vs. control. (A) Set of upregulated genes. (B) Set of downregulated genes. BP = biological process, CC = cellular component, MF = molecular function.
Venn diagram of common and unique set of differentially expressed genes across three cold acclimation treatments (12, 24, and 48 h) located in previously identified consistent quantitative trait loci regions reported by Yu et al. (2022). (A) Upregulated genes. (B) Downregulated genes. CT = Control.
Contributor Notes
This work was funded, in part, by the United States Department of Agriculture, National Institute of Food and Agriculture Hatch Project (OKL03255), Oklahoma State University, Horticultural and Landscape Architecture Department funding, and Oklahoma Agricultural Experiment Station.
Principal component (PC) analysis using preprocessed and normalized gene counts from genome-guided transcriptome assembly data of control (CT) and 12 (Trt12), 24 (Trt24), and 48 (Trt48) h of cold acclimation of African bermudagrass genotype OKC1163.
Volcano plots show the distribution and the number of the differentially expressed genes under 12 h of cold acclimation (A), 24 h of cold acclimation (B), and 48 h of cold acclimation (C). Upregulated and downregulated genes are shown in blue and red, respectively.
UpSet plot showing the comparison of unique number of upregulated and downregulated genes after 12, 24, and 48 h of cold acclimation of African bermudagrass genotype OKC1163.
Top significantly enriched Gene Ontology terms in the set of differentially expressed genes of 12-h cold acclimation vs. control. (A) Set of upregulated genes. (B) Set of downregulated genes. BP = biological process, CC = cellular component, MF = molecular function.
Top significantly enriched Gene Ontology terms in the set of differentially expressed genes of 24-h cold acclimation vs. control. (A) Set of upregulated genes. (B) Set of downregulated genes. BP = biological process, CC = cellular component, MF = molecular function.
Top significantly enriched Gene Ontology terms in the set of differentially expressed genes of 48-h cold acclimation vs. control. (A) Set of upregulated genes. (B) Set of downregulated genes. BP = biological process, CC = cellular component, MF = molecular function.
Venn diagram of common and unique set of differentially expressed genes across three cold acclimation treatments (12, 24, and 48 h) located in previously identified consistent quantitative trait loci regions reported by Yu et al. (2022). (A) Upregulated genes. (B) Downregulated genes. CT = Control.
