The overexpression of the switchgrass (Panicum virgatum L.) genes PvTOC1-N or PvLHY-K affects circadian rhythm and hormone metabolism in transgenic Arabidopsis seedlings

Abstract

Switchgrass (Panicum virgatum L.) is a perennial C4 warm-season grass known for its high-biomass yield and wide environmental adaptability, making it an ideal bioenergy crop. Despite its potential, switchgrass seedlings grow slowly, often losing out to weeds in field conditions and producing limited biomass in the first year of planting. Furthermore, during the reproductive growth stage, the above-ground biomass rapidly increases in lignin content, creating a significant saccharification barrier. Previous studies have identified rhythm-related genes TOC1 and LHY as crucial to the slow seedling development in switchgrass, yet the precise regulatory functions of these genes remain largely unexplored. In this study, the genes TOC1 and LHY were characterized within the tetraploid genome of switchgrass. Gene expression analysis revealed that PvTOC1 and PvLHY exhibit circadian patterns under normal growth conditions, with opposing expression levels over time. PvTOC1 genes were predominantly expressed in florets, vascular bundles, and seeds, while PvLHY genes showed higher expression in stems, leaf sheaths, and nodes. Overexpression of PvTOC1 from the N chromosome group (PvTOC1-N) or PvLHY from the K chromosome group (PvLHY-K) in Arabidopsis thaliana led to alterations in circadian rhythm and hormone metabolism, resulting in shorter roots, delayed flowering, and decreased resistance to oxidative stress. These transgenic lines exhibited reduced sensitivity to hormones and hormone inhibitors, and displayed altered gene expression in the biosynthesis and signal transduction pathways of abscisic acid (ABA), gibberellin (GA), 3-indoleacetic acid (IAA), and strigolactone (SL). These findings highlight roles of PvTOC1-N and PvLHY-K in plant development and offer a theoretical foundation for genetic improvements in switchgrass and other crops.

Introduction

Switchgrass (Panicum virgatum L.) is a C4 perennial warm-season grass that thrives on marginal soils, making it an exemplary bioenergy crop due to its high-biomass production, high cellulose content, and broad environmental adaptability [1]. This plant also possesses high nitrogen use efficiency and can grow under rainfed conditions [2]. However, switchgrass experiences a slow-growing seedling stage, resulting in relatively low biomass yield during the planting year [3]. Without meticulous field management, the seedlings are often outcompeted by weeds, complicating large-scale cultivation [4]. In addition, rapid lignin accumulation post-flowering reduces the saccharification efficiency of the biomass [5]. Enhancing the quality and quantity of switchgrass biomass can be achieved by delaying flowering and extending the vegetative growth period, which is a highly desirable trait for a bioenergy crop [6]. Previous studies have indicated that the slow seedling development in switchgrass is linked to circadian rhythm genes such as TIMING OF CAB EXPRESSION 1 [TOC1] and LATE ELONGATED HYPOCOTYL [LHY]), which also regulate the transition from vegetative to reproductive growth [7, 8], but little is known about the precise regulatory functions of these genes. Therefore, it is crucial to understand the specific functions of these genes in switchgrass to identify potential targets for crop breeding.

The genes TOC1 (also known as PSEUDO-RESPONSE REGULATOR 1 [PRR1]) and LHY are central components of the plant circadian rhythm system. Together with CIRCADIAN CLOCK ASSOCIATED 1 (CCA1), they form a negative feedback regulatory loop that governs seedling morphogenesis [9]. TOC1 is a part of the PRR family, whose members are expressed sequentially throughout the day, approximately every 2–3 h: PRR9, PRR7, PRR5, PRR3, and TOC1 [10, 11]. Sequence similarity between these proteins is found in two specific regions: the N-terminal pseudoreceiver (PR) region and the C-terminal CCT domain [12]. LHY is part of the CCA1-like subfamily within the MYB-related family and includes a MYB domain, which typically has three conserved tryptophan residues, and a conserved amino acid motif (SHAQKFF) characteristic of the CCAl-like subfamily. LHY is generally co-expressed with CCA1, and these two proteins have some overlapping functions [13, 14].

Studies have demonstrated that LHY regulates carbon and nitrogen metabolism, directly influences seedling morphology, and modulates seedlings responses of seedlings to light and temperature [15, 16]. In addition, it contributes to yield hybrid vigor and is involved in mitochondrial retrograde signaling, impacting seedling development rates [17, 18]. LHY can self-regulate its expression, with structural overexpression inhibiting the transcription of both CCA1, resulting in significant disruptions to biological rhythms [19,20,21]. Furthermore, LHY and CCA1 bind to the promoter of the evening-expressed gene TOC1 to negatively regulate its expression. In this feedback loop, TOC1 functions as a positive regulator since the morning activation of CCA1/LHY depends on TOC1 [22]. However, overexpression of TOC1 leads to reduced expression of CCA1/LHY because it acts as a general transcriptional repressor, negatively regulating not only CCA1/LHY but also several other clock-related genes [19, 23, 24]. This complexity highlights TOC1’s multifaceted role in the core circadian feedback pathway. Both TOC1 and LHY, key circadian rhythm genes, exhibit diverse functions, and further investigation is needed to understand the effects of their functions and interactions on the plant development.

Circadian rhythm genes not only govern the circadian rhythms in plants but also orchestrate growth and development by regulating the expression of genes involved in plant hormone biosynthesis and signal transduction. Genome-wide studies using RNA extracted from seedlings have revealed that approximately 30% of all expressed genes are under the influence of the circadian rhythm [25,26,27]. Genes regulated by hormones such as abscisic acid (ABA), brassinosteroids (BRs), cytokinins (CKs), ethylene (ET), gibberellins (GAs), auxin, jasmonic acid (JA), and salicylic acid (SA) are particularly likely to be influenced by the circadian rhythm. In Arabidopsis thaliana, approximately 35–46% of circadian rhythm-regulated genes are associated with hormone signaling [28]. Chromatin immunoprecipitation studies have revealed that circadian-regulated proteins such as CCA1, TOC1, and PRRs bind to the promoters of hundreds of genes, including those influenced by plant hormones [29,30,31,32]. In this study, two TOC1 genes and two LHY genes were identified in the switchgrass genome, designated as PvTOC1-K (Pavir.1NG350900), PvTOC1-N (Pavir.1KG385300), PvLHY-K (Pavir.6KG070500), and PvLHY-N (Pavir.6NG060600). The expression patterns of these genes were analyzed across multiple tissues under various stress and hormone treatments. Further, the roles of PvTOC1-N and PvLHY-K in seedling development were investigated in Arabidopsis. Results indicated that PvTOC1-N and PvLHY-K share partial homology with AtTOC1 and AtLHY, respectively, and their overexpression can alter circadian rhythm and hormone metabolism in Arabidopsis. These findings offer new insights into the role of circadian genes in switchgrass, particularly in seedling development and vegetative growth, and help identifying strong candidates for future genetic enhancement of switchgrass and other crops.

Materials and methods

Plant materials and growth conditions

The Arabidopsis ecotype Columbia 0 (Col-0) was used for the experiments in this study. Arabidopsis seeds were incubated in sterile water at 4 °C for 3 days, then sown in soil and transferred to a growth chamber set at 25/22 °C day/night with a 16/8-h light/dark cycle (LD) for cultivation. The switchgrass variety used was the lowland tetraploid ‘Alamo’ from our laboratory. ‘Alamo’ was planted in the field at Northwest A&F University in Yangling, Shaanxi, China (east longitude 108°–108° 7′, north latitude 34° 12′–34° 20′) to study gene expression in seedling tissues.

PvTOC1 and PvLHY expression levels were detected in various switchgrass tissues (roots, stems, leaves, stem nodes, and leaf sheaths) at the E4 stage. Ears were collected when they reached 50 cm in length, and seeds were gathered during the late grain-filling period. All collected materials were flash-frozen in liquid nitrogen and stored at − 80 °C until further processing.

To study gene expression under different treatment conditions, ‘Alamo’ seedlings were grown hydroponically using Hoagland liquid medium. Seeds were first germinated for 7 days on filter paper in Petri dishes, then transferred to Hoagland liquid medium. Seedlings were grown in a growth chamber at 27/25 °C day/night with a 16/8-h light/dark cycle. Samples were collected from plants 36 days after germination treated with 100 μM gibberellic acid (GA), 50 μM abscisic acid (ABA), 200 mM sodium chloride (NaCl), or 20% (w/v) polyethylene glycol (PEG). Leaves from treated plants were collected at 6 h after treatment initiation. Untreated samples were collected simultaneously to serve as controls. The seedlings used in the photoperiod treatment were the same. Seedlings were acclimated to a 12-h light/12-h dark cycle at 22 °C for 10 days. After this period, they were switched to a continuous 12-h light/12-h light (LL) cycle. Samples were then collected every 4 h for 2 days, starting from the second full day under LL conditions [33].

Total RNA extraction and cDNA generation

Total RNA was isolated from plant samples using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. The concentration and purity of the extracted RNA were measured using a NanoDrop ND-1000 (NanoDrop, Wilmington, DE, USA). cDNA was synthesized from the RNA using a PrimeScript™ RT Kit (6210A and RR047A; TaKaRa, Dalian, China) following the manufacturer’s instructions.

Sequence analysis and isolation of PvTOC1 and PvLHY

The full-length coding sequences (CDSs) of PvTOC1 and PvLHY were amplified from cDNA synthesized from ‘Alamo’ seedlings. Primers specific to the two genes, incorporating XbaI and XhoI sites, were designed for one-step cloning based on the switchgrass reference genome available on the Phytozome database (https://phytozome-next.jgi.doe.gov/info/Pvirgatumvar_AP13HAP1_v6_1, Table A.1). Multiple sequence alignment was performed using Clustalx [34]. The isoelectric points of the proteins were calculated using the Expasy website (http://web.expasy.org/compute_pi/).

Quantitative real-time polymerase chain reaction (qRT-PCR)

RNA was extracted and cDNA was generated as previously described. Quantitative real-time PCR (qRT-PCR) was then performed on a QuantStudio™ 3 Flex Real-Time PCR System (Thermo Fisher Scientific, Waltham, MA, USA) using the SYBR Premix Ex Taq™ II Kit (RR820A; TaKaRa) with three technical replicates. EUKARYOTIC ELONGATION FACTOR 1α (PveEF-1α) and AtACTIN2 were used as positive controls to assess cDNA quality and as internal controls for gene expression normalization in switchgrass and Arabidopsis, respectively [35], using the ΔΔC_T method [36]. Primers were designed using Primer Premier 6 (http://www.premierbiosoft.com/primerdesign/index.html) and are listed in Additional file 1.

Arabidopsis transformation with PvTOC1- and PvLHY-overexpression vectors

The CDSs of PvTOC1 and PvLHY containing XbaI and XhoI sites were inserted into the plant expression vector mCherry-pGreenII OE, which harbored the mCherry and Bar genes for selection. The resulting recombinant plasmids were then introduced into Agrobacterium tumefaciens strain GV3101, followed by transformation of Arabidopsis plants using the floral-dip method [37]. Transformed seeds emitting red light were selected using the LUYOR-3415RG fluorescent protein excitation light source or a fluorescence microscope with a 580 nm excitation wavelength. Transgenic plants were confirmed by spraying with a 0.05% glufosinate-ammonium solution and through PCR using primers specific for BarF/BarR. Phenotypic observations were conducted for 10 randomly selected plants per homozygous line. Three independent homozygous transgenic lines were generated for each gene.

Arabidopsis phenotypic analysis

Wild-type (WT) and transgenic seedlings at the 30-day-old stage were used in the phenotypic analysis. The expression of PvTOC1-N or PvLHY-K was detected and the number of rosette leaves were counted. Relative chlorophyll content in the leaves was measured using a SPAD-502 chlorophyll meter. Root tip cells of 6-day-old seedlings were observed via confocal microscopy. For this observation, Arabidopsis seeds were surface-sterilized, incubated at 4 °C in the dark for 3 days, then grown on ½ × MS plates. After 4 days of growth, uniformly sized seedlings were selected and transferred to new ½ × MS plates. After 2 days, root tips were stained with 10 μg/mL propidium iodide (PI) for 2–3 min. The morphology of root apex cells was examined using a laser confocal microscope with an excitation wavelength: 543 nm. The length of the root apical meristem was measured, and the cell morphologies of the root apex, quiescent center, meristem, and mature zone were observed.

Wild-type (WT) and transgenic seedling root lengths were measured following stress treatments in ½ × MS medium. The medium was supplemented with NaCl (150 mM, 200 mM, or 250 mM), mannitol (250 mM or 300 mM), ABA (30 μM, 60 μM, or 90 μM), or varying nitrogen levels (½ × MS medium without nitrogen or with three times the nitrogen content). Surface-sterilized seeds were incubated on ½ × MS medium for 4 days. Seedlings of uniform size were then selected and transferred to the treated medium for 7 days. Root length and survival rates were assessed in 20 seedlings per line, with untreated roots serving as controls for each treatment. Each treatment was performed in three biological replicates.

Seedling root lengths were also measured after growth in ½ × MS medium supplemented with various hormones or hormone inhibitors. The seedlings were grown as described previously for the stress treatments. The hormone treatments included 1-naphthylacetic acid (NAA) at concentrations of 0.1 nM, 0.25 nM, or 1 nM; gibberellic acid (GA) at 30 μM, 60 μM, or 90 μM; N-(phenylmethyl)-9H-purin-6-amine (6-BA) at 0.05 μM, 0.5 μM, or 2 μM; and rac-GR24 (GR24) at 1 μM, 10 μM, or 100 μM. The hormone inhibitor treatments were 2,3,5-triiodobenzoic acid (TIBA) at 0.1 μM, 0.5 μM, or 2.5 μM; paclobutrazol (PAC) at 0.15 μM, 0.3 μM, or 0.6 μM; and lovastatin at 0.1 μM, 0.5 μM, or 2.5 μM.

Chlorophyll fluorescence measurements

Well-grown leaves from WT and transgenic seedlings were wrapped in wet gauze for 15 min to become dark-adapted, and chlorophyll fluorescence parameters were then measured using a FluorCam multispectral fluorescence imaging system (Eco-tech, Beijing, China). To assess responses to oxidative stress, leaves were soaked in a 200 mM H₂O₂ solution in the dark for 2 h before additional chlorophyll fluorescence measurements. Post-oxidative stress treatment, the leaves were divided into two groups; one group was exposed to light for 0.5 h and the other for 1 h (both at 24,000 lx and 23 °C). Following the light treatment, the leaves were dark-adapted for 15 min before measuring chlorophyll fluorescence parameters. Measurements were taken from 20 leaves per line, with 3 biological replicates for each experiment.

Statistical analysis was performed on various photosynthetic parameters to assess plant photosynthetic capacity and oxidative stress-induced damage. The parameters analyzed included maximum quantum efficiency of photosystem II (QYmax), steady-state quantum efficiency of photosystem II (QY_Lss), steady-state light-adapted photochemical quenching (qL_Lss) which indicates fluorescence quenching caused by photosynthesis and steady-state non-photochemical quenching (i.e., fluorescence quenching caused by heat dissipation) (NPQ_Lss).

Statistical analysis

The data were analyzed using SPSS Statistics version 22.0 (IBM, Armonk, NY, USA) to determine significant differences between genotypes and treatment groups. One-way analysis of variance (ANOVA) and Duncan’s multiple range tests were employed to calculate significance, with thresholds set at p < 0.05 and p < 0.01. Mean values from biological triplicate experiments were displayed and plotted using GraphPad Prism 8 (GraphPad Software, San Diego, CA, USA) and Excel 2019.

Results

Identification and sequence analysis of PvTOC1 and PvLHY

In a previous study analyzing transcriptomic data from switchgrass seedlings with different growth rates, we observed that TOC1 was upregulated while LHY was downregulated in slow-growing seedlings [7]. The homologs of PvTOC1 and PvLHY were found on the switchgrass K and N genomes. The amino acid sequence of PvTOC1s showed 93% similarity with OsTOC1 and 66% similarity with AtTOC1 (Additional file 2). The nucleic acid sequences of Pavir.1NG350900 (PvTOC1-N) and Pavir.1KG385300 (PvTOC1-K) had a 97% similarity, and their amino acid sequences had over 98% similarity. Due to the high similarity in both DNA and amino acid sequences, we selected PvTOC1-N, for further functional studies. The full-length coding sequence (CDS) of PvTOC1-N is 1566 bp long, encoding a protein of 522 amino acids with a molecular mass of 127.6 kDa and an isoelectric point of 4.97.

Similar to TOC1, switchgrass contained two homologous LHY genes on the K and N genomes: Pavir.6KG070500 (PvLHY-K) and Pavir.6NG060600 (PvLHY-N). Sequence analysis revealed a 42-bp deletion beginning at position 732 in the PvLHY-N sequence compared to PvLHY-K (Additional file 3). Both proteins contain a complete MYB domain, and the conserved SHAQKFF domain of the CCA1-like subfamily. The deletion in PvLHY-N does not cause a frameshift mutation, suggesting that both homologs may be functional. The amino acid similarity of PvLHYs proteins is approximately 90% with OsLHY and 74% with AtLHY (Additional file 3). PvLHY-K was selected for further study due to its higher sequence similarity with orthologs in rice. The full-length CDS of PvLHY-K is 2163 bp long, encoding a protein of 720 amino acids with a molecular mass of 79.0 kDa and an isoelectric point of 6.12.

PvTOC1 and PvLHY expression in switchgrass

The expression profiles of the two PvTOC1 genes from the Phytozome database were consistent with each other, showing high expression levels in both the floret and the vascular bundle (Additional file 4A). During the E4 stage in vegetative organs and in reproductive organs, expression analysis revealed that both PvTOC1 genes were highly expressed in seeds, with PvTOC1-K showing higher expression (Fig. 1A). The genes most strongly co-expressed with each PvTOC1 gene varied; Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis indicated that those co-expressed with PvTOC1-K were enriched in the “ribosome biogenesis in eukaryotes” and “circadian rhythm” pathways, while those with PvTOC1-N were enriched in the “RNA transport” and “ribosome biogenesis in eukaryotes” pathways (Additional file 5A). Both PvLHYs genes showed high expression in stems, leaf sheaths, and nodes, with PvLHY-K being more highly expressed than PvLHY-N (Fig. 1D, Additional file 4B). Similar to PvTOC1, the genes most strongly co-expressed with each PvLHY gene differed. For PvLHY-K, these genes were enriched in the “circadian rhythm” and “glyoxylate and dicarboxylate metabolism” pathways, while for PvLHY-N, they were enriched in the “circadian rhythm” and “carbon fixation in photosynthetic organisms” pathways (Additional file 5B).

Fig. 1

Expression pattern analysis of TOC1s and LHYs in switchgrass. A Expression patterns of PvTOC1s in multiple organs. B Expression patterns of PvTOC1s under various treatments for 6 h. C Expression patterns of PvTOC1s in switchgrass over time. D Expression patterns of PvLHYs in multiple organs. E Expression patterns of PvLHYs under various treatments for 6 h. F Expression patterns of PvTOC1s in switchgrass over time. R: root; S: stem; L: leaf; SN: stem node; SH: leaf sheath; P: panicle; SD: seed. The lowercase letters indicate statistically significant groups at p < 0.05

Under various treatments, the PvTOC1 genes exhibited similar expression profiles, and were both inhibited by ABA (p < 0.05, Fig. 1B). Conversely, both PvLHY genes were significantly induced by PEG treatment, with PvLHY-K expression being inhibited by GA (p < 0.05, Fig. 1E). As key components of rhythmic oscillations, these genes displayed consistent cycles of increased and decreased expression levels over time. Notably, PvTOC1 and PvLHY genes showed opposing expression patterns over time, although the two copies of each gene had similar expression patterns (Fig. 1C, F). Overall, PvLHY-K was expressed at higher levels than PvLHY-N.

Effects of PvTOC1-N or PvLHY-K overexpression on Arabidopsis growth and development

Based on the analysis of endogenous PvTOC1 and PvLHY genes, the more highly expressed gene from each pair (PvTOC1-N or PvLHY-K) was transformed into Arabidopsis to study their effects on plant growth and development. Transgenic lines expressing PvTOC1-N or PvLHY-K were confirmed via PCR, showing high expression levels of these genes (Fig. 2A, F).

Fig. 2

Phenotypic analysis of plants overexpressing PvTOC1-N and PvLHY-K. A PvTOC1-N expression in wild-type (WT) and PvTOC1-N-ox plants. B Rosette leaf number in WT and PvTOC1-N-ox plants. C Root lengths of WT and PvTOC1-N-ox plants grown on a ½ × MS culture medium. D Root meristem length in WT and PvTOC1-N-ox plants. E Flowering ratios of WT and PvTOC1-N-ox plants. F PvLHY-K expression in WT and PvLHY-K-ox plants. G Phenotypes of WT and PvTOC1-N-ox plants 45 days after sowing. Scale bar = 7 cm. H Root lengths of WT and PvLHY-K-ox plants on ½ × MS culture medium. I Phenotypes of WT and PvLHY-K-ox plants 45 days after sowing. Scale bar = 7 cm. J Root tip phenotypes of WT and PvLHY-K-ox plants. Scale bar = 50 μm. The lowercase letters indicate statistically significant groups at p < 0.05

To investigate the effects of PvTOC1-N or PvLHY-K overexpression on Arabidopsis seedling growth and development, various phenotypic indicators were measured, including rosette leaf number, flowering time, root length, and root tip morphology (Fig. 2, Additional file 6). Plants overexpressing PvTOC1-N (PvTOC1-N-ox) had a significantly higher number of rosette leaves (p < 0.05, Fig. 2B) and flowered late, with one of the three transgenic lines flowering approximately 3 days later than the WT (Fig. 2E, F). In addition, PvTOC1-N-ox plants had significantly shorter roots after 7 days of growth on ½ × MS medium compared to the WT (p < 0.05, Fig. 2C). PI staining of the roots revealed that PvTOC1-N-ox plants had longer root apical meristems (p < 0.05, Fig. 2D, Additional file 6).

Compared to WT plants, those overexpressing PvLHY-K (PvLHY-K-ox) showed no significant differences in leaf shape, leaf number, or chlorophyll content, but they did bloom slightly later (Fig. 2I). PvLHY-K-ox plants had shorter roots than WT plants after 7 days of growth on ½ × MS medium (p < 0.05, Fig. 2H). PI revealed that the root apical meristem was longer in two of the transgenic lines (L1 and L3) compared to WT (p < 0.05, Additional file 6B), and the cell arrangement in the quiescent center of the root tip differed from that of WT in one transgenic line (L2) (Fig. 2J).

PvTOC1-N-ox and PvLHY-K-ox plants responded differently to stress treatments compared to WT

We next evaluated the stress tolerance of PvTOC1-N-ox and PvLHY-K-ox plants by adding ABA, NaCl, mannitol, or nitrogen to the ½ × MS growth plates. Compared with the WT, PvTOC1-N-ox plants displayed longer roots under exogenous ABA treatment (p < 0.05 at 30 μM and 60 μM, Fig. 3A, C). However, we found no significant differences between WT and PvTOC1-N-ox plants under any of the remaining stress treatments tested.

Fig. 3

Stress tolerance in plants overexpressing PvTOC1-N and PvLHY-K. A Relative root length of wild-type (WT) and PvTOC1-N-ox plants after abscisic acid (ABA) treatment. B Root growth phenotypes of WT and PvTOC1-N transgenic lines on a ½ × MS culture medium. The diameter of the grid above the plate corresponds to 1.5 cm. C Phenotypes of WT and PvTOC1-N-ox plants after 30 μM or 60 μM ABA treatment. D Relative root lengths of WT and PvLHY-K-ox plants after ABA treatment. E Root growth phenotypes of WT and PvLHY-K transgenic lines on a ½ × MS culture medium. F WT and PvLHY-K-ox plant phenotypes after treatment with 60 μM or 90 μM ABA. The lowercase letters indicate statistical significance groups at p < 0.05. The relative root length corresponds to the ratio of the root length of the line on the stress treatment culture medium to the root length on the ½ × MS culture medium

PvLHY-K-ox plants were less sensitive to ABA than WT plants (p < 0.05 at 60 μM and 90 μM, Fig. 3D, F). On nitrogen-free ½ × MS medium (0 N), PvLHY-K-ox roots were longer than those of WT plants, indicating that PvLHY-K-ox plants were less affected by nitrogen-starvation stress (Additional file 7). When grown on medium with triple the standard nitrogen concentration, growth was severely inhibited in all genotypes. After 7 days of 250 mM NaCl treatment, both WT and PvLHY-K-ox plants died; however, under 150 mM NaCl conditions, PvLHY-K-ox plants had a higher survival rate than WT plants (p < 0.05, Additional file 7), indicating higher salt tolerance. There were no significant differences between PvLHY-K-ox and WT plants under mannitol treatment.

PvTOC1-N-ox and PvLHY-K-ox plants had low resistance to oxidative stress

Photosynthesis plays a crucial role in seedling growth. Therefore, we measured chlorophyll fluorescence in PvTOC1-N-ox, PvLHY-K-ox, and WT plants. Under normal growth conditions, QYmax was > 0.8 in WT and PvTOC1-N-ox plants, indicating healthy growth. Compared to WT plants, qL_Lss was slightly higher in PvTOC1-N-ox, suggesting slightly stronger photosynthetic activity in the transgenic plants; NPQ_Lss values, indicating light protection in the seedlings, were low in both WT and PvTOC1-N-ox, suggesting high photosynthetic activity and healthy growth. Treatment with H₂O₂ for 2 h did not significantly alter the chlorophyll fluorescence parameters of WT and PvTOC1-N-ox plants. However, QYmax, QY_Lss, and qL_Lss were significantly reduced in H₂O₂-treated WT and PvTOC1-N-ox plants after 0.5 h or 1 h of light exposure. At this point, QYmax, QY_Lss, and qL_Lss were lower in PvTOC1-N-ox, indicating higher sensitivity to oxidative stress and decreased photosynthetic activity. The increased NPQ after light exposure suggested activation of light protection in the seedlings following oxidative stress. NPQ was also lower in PvTOC1-N-ox compared to WT plants, indicating slightly lower photoprotective ability in PvTOC1-N-ox (Fig. 4A).

Fig. 4

Chlorophyll fluorescence parameters in wild-type, PvTOC1-N-ox and PvLHY-K-ox plants. A, B Maximum quantum efficiency of photosystem II (QYmax), steady-state quantum efficiency of photosystem II (QY_Lss), steady-state light-adapted photochemical quenching (qL_Lss), and steady-state non-photochemical quenching (NPQ_Lss) in A WT and PvTOC1-N-ox; and B WT and PvLHY-K-ox plants subjected to oxidative stress conditions. The lowercase letters indicate statistically significant groups at p < 0.05

Under normal growth conditions, the chlorophyll fluorescence parameters of PvLHY-K-ox plants were similar to those of PvTOC1-N-ox plants. Photosynthetic parameters in PvLHY-K-ox plants did not significantly change after 2 h of treatment with H₂O₂ in the dark. However, after the H₂O₂-treated WT and PvLHY-K-ox plants were exposed to light for 0.5 h or 1 h, QY_Lss decreased significantly (p < 0.05) and NPQ_Lss increased significantly (p < 0.05) in both lines. This indicated that the plants were under oxidative stress, and photoprotection was activated in response to the light treatment. After 1 h of light exposure, the chlorophyll fluorescence parameters of PvLHY-K-ox significantly differed from those of WT plants (p < 0.05, Fig. 4B), indicating severe oxidative stress in PvLHY-K-ox. Thus, PvLHY-K overexpression made seedlings more sensitive to oxidative stress and affected their photosynthetic efficiency and activity.

PvTOC1-N-ox and PvLHY-K-ox plants displayed distinct responses to hormones and their respective inhibitors

To understand the effects of PvTOC1-N or PvLHY-K overexpression on Arabidopsis seedling growth and development, WT and transgenic seedlings were treated with various hormones, including NAA, 6-BA, GA, or GR24; and hormone inhibitors, such as TIBA, lovastatin, or PAC. Following these treatments, we measured root growth in each respective seedling (Figs. 5 and 6). We found that NAA (0.1–1 nM) slightly promoted WT and PvTOC1-N root growth compared to untreated plants, but there were no differences between both (Additional file 8). In addition, our results showed that TIBA, which inhibits the polar transport of auxin, prevented Arabidopsis root growth [38, 39], while PvTOC1-N-ox plants were less sensitive to 0.5 μM TIBA than WT plants (p < 0.05, Fig. 5A). Treatment with exogenous 6-BA or lovastatin (an inhibitor of cytokinin synthesis) also inhibited Arabidopsis root growth in a concentration-dependent manner [40, 41]. Importantly, we found that PvTOC1-N-ox plants were less sensitive to 6-BA and lovastatin treatments than WT plants, and that this difference was significant at 0.05 μM 6-BA (Figs. 5B, G; Additional file 8B). Previous studies showed that the concentration of GA required to regulate root growth is lower than that necessary to regulate bud development, and 30–90 μM GA did not significantly promote root elongation [42]. We found that this concentration slightly promotes root elongation in PvTOC1-N-ox but not in WT plants (Fig. 5C). The GA biosynthesis inhibitor PAC also inhibited Arabidopsis root growth [43, 44], but our observations showed that PvTOC1-N-ox plants were less sensitive to PAC compared to WT plants (p < 0.05 at 0.15 μM, Fig. 5D, G). GR24, the most widely used synthetic strigolactone (SL), inhibited WT root growth at concentrations of 1–100 μM, but promoted PvTOC1-N-ox root growth, with an inverse correlation between GR24 concentration and root growth (p < 0.05 at 100 μM, Fig. 5E, G).

Fig. 5

Effects of hormone and hormone inhibitor treatments on wild-type (WT) and PvTOC1-N overexpression (PvTOC1-N-ox) plant roots. A–E Comparison of relative root lengths between WT and PvTOC1-N-ox plants following treatment with A 2,3,5-triiodobenzoic acid (TIBA), B N-(phenylmethyl)-9H-purin-6-amine (6-BA), C gibberellin (GA), D paclobutrazol (PAC), or E rac-GR24 (GR24). F, G Root length phenotypes observed in WT and PvTOC1-N-ox plants treated with F 0.5 μM TIBA and 0.05 μM 6-BA or G 90 μM GA and 100 μM GR24. The lowercase letters indicate statistically significant groups at p < 0.05. Relative root length is defined as the ratio of root length measured on hormone or hormone inhibitor treatment culture medium to that observed on ½ × MS culture medium

Fig. 6

Effects of hormone and hormone inhibitor treatments on wild-type (WT) and PvLHY-K overexpression (PvLHY-K-ox) plants. A–E Relative root lengths of WT and PvLHY-K-ox plants following treatment with A 1-naphthylacetic acid (NAA), B gibberellin (GA), C N-(Phenylmethyl)-9H-purin-6-amine (6-BA), D lovastatin, or E rac-GR24 (GR24). F Root length phenotypes observed in WT and PvLHY-K-ox plants treated with 0.05 μM 6-BA and 0.5 μM lovastatin. Lowercase letters denote statistically significant groups at p < 0.05. Relative root length is defined as the ratio of the root length measured on the hormone or hormone inhibitor treatment culture medium to that observed on ½ × MS culture medium

PvLHY-K-ox and PvTOC1-N-ox plants exhibited similar responses to hormone and inhibitor treatments. PvLHY-K-ox plants displayed a slight decrease in root lengths with increasing concentrations of NAA (Fig. 6A). GA treatment promoted root elongation in PvLHY-K-ox plants, although this promotion effect decreased with higher GA concentrations (p < 0.05 at 60 μM, Fig. 6B). PvLHY-K-ox plants exhibited reduced sensitivity to 6-BA/lovastatin treatment compared to WT plants, particularly at lower concentrations (p < 0.05, Fig. 6C, D, F). Application of 1 μM GR24 significantly stimulated root growth in PvLHY-K-ox plants but not in WT plants (p < 0.05, Fig. 6E). No significant differences were observed between WT and PvLHY-K-ox plants following TIBA or PAC treatment (Additional file 8C, D).

Effects of PvTOC1-N or PvLHY-K overexpression on hormone-related genes

The results above suggest that the transgenic overexpression lines responded differently to hormone or hormone inhibitor treatments compared to WT plants. Consequently, we investigated the effects of PvTOC1-N or PvLHY-K overexpression on genes involved in hormone biosynthesis and signal transduction pathways. In PvTOC1-N-ox plants, the rate-limiting ABA biosynthesis gene 9-CIS-EPOXY CAROTENOID DIOXYGENASE (NCED) was upregulated (p < 0.05), along with negative ABA-response regulators ABA-INSENSITIVE 1 (ABI1) and ABA-INSENSITIVE 5 (ABI5) (Fig. 7A). In addition, PvTOC1-N-ox plants exhibited upregulation of a member of the DELLA family (GA INSENSITIVE [GAI]), a key enzyme in GA synthesis (GA20 OXIDASE, GA20ox), the auxin biosynthetic gene YUCCA8, and a gene controlling root meristem size, TIME FOR COFFEE (TIC), compared to WT plants (p < 0.05, Fig. 7B, C).

Fig. 7

Expression of hormone- and circadian rhythm-related genes in wild-type (WT), PvTOC1-N overexpression (PvTOC1-N-ox) and PvLHY-K overexpression (PvLHY-K-ox) plants. A–C Expression of genes related to the A abscisic acid (ABA), B gibberellin (GA), and C auxin pathways in WT and PvTOC1-N-ox plants. D–G Expression of genes related to the D ABA, E auxin, F strigolactone, and G GA pathways in WT and PvLHY-K-ox plants. H, I Expression levels of AtTOC1 and AtLHY in H WT and PvTOC1-N-ox and I WT and PvLHY-K-ox plants. The lowercase letters indicate statistically significant groups at p < 0.05

In PvLHY-K-ox plants, the expression of NCED, ABI1, ABI5, and GA20ox, were all significantly downregulated compared to the WT (p < 0.05, Fig. 7D, G). In addition, the canonical strigolactone (SL) biosynthetic enzymes MORE AXILLARY GROWTH 1 (MAX1) (also known as CYP711A) and DWARF 27 (D27) were downregulated in PvLHY-K-ox plants. MAX1 showed significantly lower expression levels in all three transgenic lines compared to the WT (Fig. 7F). The differential expression of these hormone pathway-related genes may explain the varied responses to hormones observed in the transgenic lines.

Effects of PvTOC1-N or PvLHY-K overexpression on Arabidopsis circadian rhythm

Previous studies have established feedback regulation between TOC1 and LHY/CCA1; LHY and CCA1 negatively regulate TOC1 expression, while TOC1 overexpression inhibits LHY and CCA1 expression [19, 24, 45]. To assess whether PvTOC1-N and PvLHY-K overexpression affected the circadian clock in Arabidopsis, WT and transgenic plants acclimated to a 12-h LD cycles were sampled every 4 h under LL conditions to measure AtLHY and AtTOC1 expression. In PvTOC1-N-ox lines compared to WT, AtTOC1 was highly expressed, whereas AtLHY was expressed at lower levels (Fig. 7H). A 180-degree phase shift was observed between AtTOC1 and AtLHY expression. In PvLHY-K-ox lines, both AtTOC1 and AtLHY were expressed at lower levels compared to WT (Fig. 7I). Thus, these two genes had different effects on the circadian rhythm when overexpressed.

Discussion

Most organisms possess inherent time-keeping abilities; many circadian-related phenomena persist even when external time cues are absent, indicating that these phenomena arise from endogenous circadian rhythms [8]. Feedback regulation between TOC1 and LHY/CCA1, identified as core components of the circadian rhythm in many plant species, plays a crucial role [20,21,22]. In bioenergy crops, such as switchgrass, the speed of seedling development and control over flowering time are essential for optimizing biomass production. A comprehensive understanding of the mechanisms governing plant circadian rhythms could be utilized to regulate reproductive growth, nutritional status, and ultimately achieve high-biomass crop varieties. It could also aid in improving seedling establishment in switchgrass. In this study, TOC1 and LHY were identified as key circadian rhythm genes in switchgrass, and their functions in plant growth and development were analyzed in Arabidopsis.

The two PvTOC1 and two PvLHY genes in switchgrass showed circadian-regulated expression patterns

Expression pattern analysis revealed that the two PvTOC1 homologs exhibited similar expression patterns to each other, as did the two PvLHY homologs. However, comparison of the two PvTOC1 genes to the two PvLHY genes showed opposing expression patterns. PvTOC1 genes were highly expressed primarily in reproductive organs, such as florets and seeds, whereas PvLHY genes were highly expressed in vegetative organs (Fig. 1A, D, Additional file 4). Both PvTOC1 and PvLHY genes showed circadian rhythmicity in switchgrass, with their expression levels upregulated and downregulated at opposite times of the day, consistent with previous findings that LHY negatively regulates TOC1 expression (Fig. 1C, F) [24, 46]. Interestingly, the expression levels of PvTOC1-K and PvLHY-K were higher than their corresponding genes on the N chromosome. PvTOC1 genes were repressed by ABA treatment, while PvLHY genes were induced by PEG treatment. The two PvLHYs responded differently to GA and NaCl treatments: PvLHY-K responded to GA treatment, whereas PvLHY-N responded to NaCl treatment (Fig. 1E). It suggests that PvLHY-K and PvLHY-N may share some overlapping circadian rhythm functions but also possess distinct roles in regulating switchgrass responses to different environmental factors. Each PvLHY and each PvTOC1 gene exhibited a unique set of co-expressed genes, indicating functional differences between the homologs and suggesting that each gene may play different roles in developmental processes (Additional file 5).

PvTOC1-N or PvLHY-K overexpression affected Arabidopsis circadian rhythm

CCA1 and LHY negatively regulate TOC1 expression and can also regulate their own expression. Constitutive overexpression of either LHY or CCA1 represses the transcription of both genes, leading to generalized circadian dysrhythmia [19,20,21]. Previously, it has been demonstrated that the overexpression of the maize (Zea mays) genes ZmCCA1b or ZmCCA1a disrupts the circadian rhythm of Arabidopsis by inhibiting the expression of circadian rhythm-related genes [47, 48]. In this study, we analyzed the expression levels of AtTOC1 and AtLHY in WT, PvTOC1-N-ox, and PvLHY-K-ox plants. We found that PvLHY-K overexpression significantly suppressed AtTOC1 and AtLHY (Fig. 7I) indicating that PvLHY-K can regulate circadian rhythm in Arabidopsis. However, the effects of TOC1 on CCA1 and LHY are complex: TOC1 is necessary to activate CCA1 and LHY expression in the early morning, but TOC1 overexpression in Arabidopsis inhibits CCA1 and LHY expression [19, 22,23,24]. In this study, PvTOC1-N overexpression repressed AtLHY. Endogenous AtTOC1 was expressed at higher levels in PvTOC1-N-ox than in WT plants (Fig. 7H, I). This might be attributed to reduced TOC1 inhibition due to low AtLHY expression. Previous studies have shown that increased rhythmic TOC1 expression delays the circadian rhythm, whereas constitutive TOC1 overexpression completely disrupts rhythmicity [19]. In our study, PvTOC1-N overexpression in Arabidopsis did not completely abolish rhythmicity, likely due to the high rhythmic expression of endogenous AtTOC1.

PvTOC1-N or PvLHY-K overexpression altered hormone metabolism in Arabidopsis

The circadian rhythm directly influences plant hormone responses, and many hormone-related genes are regulated by circadian rhythm genes [49]. Therefore, we analyzed the expression levels of genes involved in hormone biosynthesis and signal transduction pathways in the transgenic overexpression lines. Genes related to the ABA, GA, and IAA pathways showed differential expression in PvTOC1-N-ox compared to WT plants, while genes related to the ABA, GA, and SL pathways exhibited differential expression in PvLHY-K-ox plants (Figs. 7, 8). Interestingly, both PvTOC1-N-ox and PvLHY-K-ox plants showed decreased sensitivity to exogenous ABA treatment compared to WT plants, consistent with previous studies [50,51,52,53]. Although plants overexpressing either gene exhibited differential expression of key genes in the ABA metabolic pathway, their expression profiles differed from each other and from WT plants (Fig. 8A). LHY can directly inhibit NCED expression [53], and overexpressing PvLHY-K significantly repressed NCED. NCED was upregulated in PvTOC1-N-ox, which may have been related to the low endogenous AtLHY expression. TOC1 and LHY directly regulate separate sets of key genes in the ABA signaling pathway [50, 53], but they appeared to have different effects on ABI1 and ABI5; ABI1 and ABI5 were downregulated in PvLHY-K-ox but showed a tendency to be upregulated in PvTOC1-N-ox plants. These findings suggest that TOC1 and LHY play distinct roles in the ABA signaling pathway. TOC1 may not only affect plant responses to ABA by regulating other ABA-related genes but also could indirectly affect ABA biosynthesis through interactions with LHY.

Fig. 8

Gene interactions in wild-type (WT), PvTOC1-N overexpression (PvTOC1-N-ox), and PvLHY-K overexpression (PvLHY-K-ox) plants. A Gene expression levels in WT, PvTOC1-N-ox, and PvLHY-K-ox plants. B Functional patterns of PvTOC1-N and PvLHY-K expression in switchgrass. The arrows indicate the results of this study combined with previous studies

ABA and GA play crucial and often antagonistic roles in regulating plant growth and development [54, 55]. ABI5 participates in an interaction with DELLA to regulate plant ABA homeostasis [56], and can also regulate GA20ox expression as part of the GA biosynthetic pathway [57]. In this study, GAI was found to be upregulated in PvTOC1-N-ox plants, whereas GA20ox was downregulated in PvLHY-K-ox plants (Fig. 8A). This suggests that overexpressing PvTOC1-N or PvLHY-K altered the circadian rhythm, influencing ABA and GA homeostasis in Arabidopsis.

In PvTOC1-N-ox plants, the auxin metabolism-related genes YUCCA8 and TIC were upregulated, and the root meristem size (which can be altered by auxin accumulation) was increased. The enhanced root meristem size of PvTOC1-N-ox plants may result from auxin overproduction due to high expression of YUCCA8 and TIC, which interact with PIN genes [58,59,60]. Moreover, TIC is known to interact with MYC2 and negatively regulate JA signaling [61], suggesting that TIC upregulation may lead to changes in JA signaling in PvTOC1-N-ox plants.

PvTOC1-N or PvLHY-K overexpression reduced oxidative stress tolerance in Arabidopsis

TOC1 acts as a molecular link between environmental information and circadian clock output; plants that overexpress TOC1 have significantly enhanced light responsiveness [19]. Chlorophyll fluorescence assays showed that either PvTOC1-N or PvLHY-K overexpression enhanced photochemical quenching (qL_Lss) in Arabidopsis and improved photosynthetic characteristics, but the ability to cope with oxidative stress was significantly reduced. Following oxidative stress treatment, the photosynthetic characteristics and photoprotective capacity were notably decreased in transgenic lines. This finding aligns with previous studies indicating that altering the circadian rhythm can reduce plant adaptability to the external environment and decrease stress resistance [62]. Indeed, the importance of circadian rhythm has been demonstrated in phytoplankton and higher plants: organisms with circadian rhythms that match the external environment have competitive advantages [62,63,64]. Overexpression of PvTOC1-N or PvLHY-K altered the Arabidopsis circadian rhythm and external coordination, impacting the progression of seedling development. In switchgrass, PvTOC1-N and PvLHY-K play a role in circadian rhythm and receive environmental signals transmitted by receptors, influencing plant hormone homeostasis and stress resistance, thereby regulating seedling development and the flowering process (Fig. 8B).

In this study, it is evident that the functions of PvTOC1-N and PvLHY-K are relatively conserved in Arabidopsis, but there are also differences. Overexpression of DhLHY (LHY in Doritaenopsis) or PbLHY (LHY in Pear) in Arabidopsis significantly delays flowering; however, the inhibition of flowering caused by overexpression of PvLHY-K is not as pronounced (Fig. 2I) [65, 66]. This difference may arise from distinct functions between PvLHY-K and PvLHY-N, or differences in flowering regulation between monocots and dicots. The enrichment analysis of co-expressed genes of the K chromosome group and N chromosome group reveals differences, and these genes respond to different stress treatment, indicating functional differences of TOC1s and LHYs on the two chromosome groups (Additional file 5, Fig. 1B, E). Therefore, further studies are needed to understand their functions in switchgrass.

Conclusions

PvTOC1 and PvLHY genes are core regulators of circadian rhythm in switchgrass, exhibiting opposing expression patterns. PvTOC1 genes are highly expressed primarily in reproductive organs, while PvLHY genes are highly expressed in vegetative organs. Overexpressing PvTOC1-N or PvLHY-K in Arabidopsis resulted in delayed flowering, shorter roots, decreased resistance to oxidative stress, and lower sensitivity to hormone and hormone inhibitor treatment. Furthermore, PvTOC1-N or PvLHY-K overexpression disturbed the circadian rhythm and altered the expression of genes associated with hormone metabolism in Arabidopsis: in PvTOC1-N-ox, genes related to the ABA, GA, and IAA pathways were differentially expressed, whereas genes related to the ABA, GA, and SL pathways were differentially expressed in PvLHY-K-ox plants. PvTOC1-N-ox and PvLHY-K-ox plants exhibited the same response to exogenous ABA treatment, but overexpression of PvTOC1-N or PvLHY-K had different effects on genes involved in ABA biosynthesis and signal transduction pathways. These findings lay a theoretical foundation for the genetic improvement of switchgrass or other crops through the modulation of TOC1 and LHY expression.