Comparison of the transcriptomic responses of two Chrysanthemum morifolium cultivars to low light

Low light is a primary regulator of chrysanthemum growth. Our aim was to analyse the different transcriptomic responses of two Chrysanthemum morifolium cultivars to low light. We conducted a transcriptomic analysis of leaf samples from the ‘Nannonggongfen’ and ‘Nannongxuefeng’ chrysanthemum cultivars following a 5-day exposure to optimal light (70%, control [CK]) or low-light (20%, LL) conditions. Gene Ontology (GO) classification of upregulated genes revealed these genes to be associated with 11 cellular components, 9 molecular functions, and 15 biological processes, with the majority being localized to the chloroplast, highlighting the role of chloroplast proteins as regulators of shading tolerance. Downregulated genes were associated with 11 cellular components, 8 molecular functions, and 16 biological processes. Heat map analyses suggested that basic helix–loop–helix domain genes and elongation factors were markedly downregulated in ‘Nannongxuefeng’ leaves, consistent with the maintenance of normal stem length, whereas no comparable changes were observed in ‘Nanonggongfen’ leaves. Subsequent qPCR analyses revealed that phytochrome-interacting factors and dormancy-associated genes were significantly upregulated under LL conditions relative to CK conditions, while succinate dehydrogenase 1, elongated hypocotyls 5, and auxin-responsive gene of were significantly downregulated under LL conditions. These findings suggest that LL plants were significantly lower than those of the CK plants. Low-light tolerant chrysanthemum cultivars may maintain reduced indole-3-acetic acid (IAA) and elongation factor expression as a means of preventing the onset of shade-avoidance symptoms.


Introduction
Chrysanthemum morifolium is among the most popular ornamental plants globally, and it is frequently grown in shaded greenhouses in China. The morphological characteristics of plants are profoundly impacted by light intensity such that lower levels of light exposure can affect key growth, anatomy [1,2], photosynthesis [3][4][5], antioxidant activity [6,7], and sucrose metabolism [8,9]. The author's previous research found that weak light seriously affected the growth of chrysanthemums [10,11].The molecular mechanisms governing the low-light (LL) responses of the 'Nannongxuefeng' and 'Nannonggongfen' chrysanthemum cultivars are currently lacking. Transcriptomic analyses offer a powerful approach that can be used to explore gene expression in a wide range of plant species. As such, this study conducted such transcriptomic analyses in order to understand the mechanisms whereby chrysanthemum plants adapt to low light.
C. morifolium is an allohexaploid plant (2n = 6 × = 54), and owing to such complexity there is no extant wholegenome sequence available for this plant [12]. RNA sequencing (RNA-seq), however, has been used to explore the biological properties of these plants, with the first transcriptomic study of C. morifolium responses to dehydration stress having been published in 2013 [13]. Such analyses have also enabled researchers to identify differentially expressed genes (DEGs) in C. morifolium plants that are 1 3 responsive to low-light conditions [14][15][16]. One deep transcriptomic analysis determined that phytochrome interacting factors (PIFs) likely serve as direct drivers of auxin biosynthesis, conjugation, transport, perception, and signalingrelated genes [17]. Early termination of leaf proliferation can occur as a consequence of extended exposure to a low red to far-red light ratio (R/FR) through mechanisms dependent upon the ATHB2 and ATHB4 HD-Zip II transcription factors [15]. Applied transcriptomic analysis methods have also been employed to explore C. morifolium responses upon exposure to UV-B light [18], the mechanisms regulating carotenoid accumulation [19], responses to dodder invasion [20], and other biologically relevant pathways. No studies, however, have explored C. morifolium transcriptomic responses to low-light exposure.
As the ability of these plants to grown under low-light conditions can impact the ornamental value of C. morifolium, there is clear commercial relevance to the elucidation of the molecular mechanisms governing such growth responses. Prior studies of C. morifolium have revealed shade-avoidance behaviors to emerge on day 7 of low-light treatment [11]. To better understand the transcriptomic changes associated with such behaviors, we herein used an RNA-seq approach to identify DEGs associated with exposure to supraoptimal (70%) and very low (20%) natural light conditions in the 'Nannongxuefeng' and 'Nannonggongfen' chrysanthemum cultivars. Leaves were collected from these plants at 11:00 AM on the fifth day of shade exposure, before the manifestation of overt shade-avoidance behaviors. In addition to identifying upregulated and downregulated DEGs in these plants, we performed Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses to understand the biological functions of these key low-light exposure-related genes. Together, these results may offer novel insights into the genetic determinants of low-light sensitivity in chrysanthemum cultivars, thereby guiding future research or selective breeding efforts.

Plant materials, growth, and treatment
The 'Nannongxuefeng' (XF, low-light resistant) and 'Nannonggongfen' (GF, low-light sensitive) C. morifolium cultivars were obtained from an experimental field at Shangqiu Normal University. Uniform cuttings were collected for each cultivar and were grown in plugs containing a 1:1 peat and perlite mixture for 15 days, after which rooted seedlings were transferred to 4 L pots filled with a 2:1:1 combination of garden soil, perlite, and vermiculite that was treated with a nutrient solution. Seedlings were randomly selected to be subjected to either 70% or 20% natural light conditions (control and low-light groups; CK and LL, respectively). Commercial black shading nets were used to reduce light exposure in a net-house (2.5 m high × 4 m long × 4 m wide) for 21 days beginning on July 20, 2019, with 60 pots per treatment condition. No significant differences in temperature were observed between treatment conditions, with average canopy temperatures remaining within < 1.0 °C of one another. other, and no consistent differences in temperature were detected between the treatments. The respective average maximum/minimum temperature, photoperiod (day/ night), and humidity were 30 °C/21 °C, 14/10 h, and 70%. In addition to natural rainfall, supplemental water was applied to all plants as appropriate. On the fifth day, apex leaves from 10 plants per treatment condition were collected from 10:30 to 12:00 AM, snap-frozen, and shipped to the Huada Genetics Company for transcriptome sequencing.

Internode and petiole length measurements
Five plants per treatment condition from different pots were selected, and measurements of the internodes of the fourth and fifth fully expanded leaves and the petioles of the fourth fully expanded leaves were taken.

RNA sequencing
The BGIS EQ-500 platform was used for sequencing, generating 122.59 Gb of raw data. Following assembly and the removal of redundant sequences, 104,769 unigenes were obtained, and 55,189 coding sequences (CDS) were detected using TransDecoder. Transcript expression was assessed based on the number of fragments per kilobase per million mapped reads (FPKM), with FPKM≧1 being indicative of reliability. Differences in expression levels between samples were assessed with DEseq2, with P < 0.05 serving as the threshold for the identification of differentially expressed genes (DEGs). DEGs were further classified into functional categories through GO and KEGG enrichment analyses conducted with the Phyper function in R. False discovery rate (FDR) approach was then used to correct P values, with FDR≦0.01 being indicative of significant enrichment. Samples were analyzed in triplicate. qPCR A total of 5 genes in three replicate plants were selected for qPCR-based validation of RNA-seq data using primers designed using the Primer 5.0 software (Applied Biosystems) and synthesized by Genewiz (Tianjin, China). A Pri-meScript RT reagent kit (Takara, Dalian, China) was used to prepare cDNA from 1 μg of total RNA, and qPCR was subsequently conducted with a Thermo PR096 Real-Time PCR instrument (Applied Biosystems) and 2 × SYBR green PCR master mix (Applied Biosystems). Samples were analyzed in triplicate and results were reported as means ± SE (n = 3). Primers used in this study are shown in Additional File 1. The 2 −ΔΔCT method was used to assess relative gene expression [21].

Statistical analysis
Samples were selected randomly for quantitative evaluation. Data were compared via one-way ANOVAs using SPSS v21.0 (SPSS, IL, USA), with means being compared by Duncan's test at a 5% significance level.

Chrysanthemum seedling responses to low-light conditions
On day 21 of low-light exposure, chrysanthemum plants of the GF cultivar exhibited thinner leaves and weaker growth relative to corresponding control (CK) plants, whereas no such morphological changes were evident for plants of the XF cultivar (Fig. 1). The low-light-sensitive GF plants exhibited increased internode and petiole lengths, in contrast to low-light-resistant XF (Fig. 2).
The response curves for different chrysanthemum varieties grown under low light treatment conditions are shown in Fig. 3. When photosynthetically active radiation was low (PAR ≤ 200 μmol m −2 s −1 ), the net photosynthetic rate of leaves exhibited an almost linear increase with changing light intensity, and the net photosynthetic rate of XF CK plants was significantly higher than that of plants under low light treatment (P < 0.05). Similarly, GF CK plants exhibited a net photosynthetic rate that was significantly higher than that of plants under low light conditions (P < 0.01). When the photosynthetically active radiation was 1200-1500 μmol m −2 s −1 , the net photosynthetic rate of XF LL plants tended to plateau and to reach the light saturation point. This trend manifested earlier in GF LL plants, indicating that GF plants were more significantly affected by low light.

Chrysanthemum gene expression profiles in response to low light
Relative to the low-light-resistant XF cultivar, the low-lightsensitive GF cultivar exhibited substantially more DEGs (both upregulated and downregulated) upon exposure to LL conditions (Fig. 4). When comparing shared DEGs between XF and GF plants (LL vs. CK conditions), 343 and 182 commonly up-and downregulated genes were identified, respectively (Figs. 4b, c).

GO classification of common DEGs
We next evaluated the transcriptomic responses of XF and GF plants to better understand how they respond to Fig. 1 Effects of different light treatments on chrysanthemum growth LL conditions. Overall, similar LL-induced DEG profiles were observed in leaf samples from both cultivars, with a slightly higher fraction of genes involved in extracellular responses, transcription regulator activity, and signal transduction activity being observed in GF plants relative to XF plants. A number of resistance-associated genes were among DEGs identified in both XF and GF plants (Fig. S1, Supplementary material). Resistance-related genes that were preferentially expressed in XF plants included those encoding light-harvesting proteins and protein phosphatases. Enrichment analyses revealed upregulated genes to be primarily associated with photosynthesis, including 9 proteins related to photosystem I (GO: 0009522), 7 with pectatelyase activity (

Analysis of low-light-related DEGs
To identify regulatory pathways related to low-light responses in these two chrysanthemum cultivars, heat map analyses were next conducted, revealing significant downregulation of the basic helix-loop-helix domain genes (unigene 4998) and elongation factor (CL1856) in XF plants (Fig. S2, Supplementary material), suggesting a role for these genes in the maintenance of normal stem length.
Candidate DEG validation. To validate our RNA-seq data, we next selectively assessed gene expression patterns for five DEGs via qPCR, including the plant hormone signal transduction-related genes XP_021983913.1 (phytochrome interacting factor 7, PIF7), XP_022023437.1 (elongated hypocotyls 5, HY5), and XP_022017432.1 (auxin-responsive protein IAA1-like, IAA) (KEGG Orthology), the mitochondrial electron transport-related gene XP_022017432.1 (succinate dehydrogenase assembly factor 1, SDH1) (Gene Orthology), and the dormancy-related gene DOR (Swissprot) (Fig. S2, Supplementary material). We observed PIF7 upregulation in response to LL conditions in both cultivars, although it was expressed at levels roughly four-fold higher in GF plants relative to XF plants (Fig. S3a, Supplementary   material). HY5 was significantly downregulated in LL plant samples relative to CK plant samples (Fig. S3b, Supplementary material), with low-light similarly significantly downregulating IAA (Fig. S3c, Supplementary material) and SDH1 (Fig. S3d, Supplementary material) in both tested cultivars. DOR-associated gene expression was increased in response to LL conditions in both cultivars, with the increase in GF plants being roughly 32-fold higher than that observed in XF plants (Fig. S3e, Supplementary material).

Discussion
Light exposure is the primary climatic determinant of plant growth [1,11], with changes in irradiance levels having the potential to impact plant growth, morphology, and anatomy [22]. While many studies of how low-light exposure alters plant physiology have been published to date, the molecular basis for such physiological changes is not well understood. Herein, we found that proteins encoded by most upregulated LL exposure-related DEGs were associated with the chloroplast and involved in photosynthesis and oxidative stress-related pathways. In particular, these proteins are associated with photosystem I/II functionality and associated with the chloroplast thylakoid membrane, suggesting a direct role for these proteins as mediators of chrysanthemum In contrast, proteins downregulated under LL conditions were involved in processes associated with chorismate and aromatic amino acid biosynthesis and were also related to chloroplasts, indicating that these proteins may be involved in LL resistance (Table S3, Supplementary material). However, more research is necessary to understand the importance of the differential regulation of these genes in the context of chrysanthemum LL exposure and resistance. We detected apparent relationships between chrysanthemum LL tolerance and altered plant hormone signal transduction and MAPK signaling pathway activities, in line with prior reports [23,24]. We also observed increases in succinate dehydrogenase, glycerol-3-phosphate transporter and chaperone protein biosynthesis, and E3 ubiquitin-protein ligase pathway activity-related proteins under LL conditions in the low-light tolerant XF cultivar, whereas the opposite was observed for the low-light sensitive GF cultivar (Fig.  S2, Supplementary material). Light-activated photoreceptors are involved in downstream complex activation, repressing its E3 ubiquitin ligase activity and photoreceptor stabilization [25]. We observed several E3 ubiquitin ligase activityrelated genes that were upregulated in XF plant leaves under LL conditions, whereas in GF plants these same genes were downregulated upon LL exposure (Fig. S2, Supplementary  material). Increased E3 ubiquitin protein levels can promote photomorphogenesis under low light, benefiting short stem maintenance responses. In LL-sensitive soybean cultivars, lignin content and related synthase activity levels were markedly reduced in response to low-light stress [26], consistent with our data and suggesting that this may be a fundamental pathway involved in crop low-light tolerance.
Previous work has indicated important roles for hormones including ABA, auxin, cytokinins, ethylene, gibberellins, and brassinosteroid in shade-avoidance response (SAR) induction [27], with ethylene, gibberellic acid, auxin, and brassinosteroids interacting through complex mechanisms to control cellular elongation under varying light conditions [28]. Overexpression of brassinosteroid biosynthetic genes in tomatoes has been shown to result in brassinosteroid transcript upregulation, gibberellin and endogenous BRdownregulation, and enhanced germination, lateral root development, and CO 2 assimilation [29]. Herein, we identified certain auxin response-related genes that were associated with cell enlargement and tryptophan metabolism. We also observed lower levels of the key brassinosteroid signal transduction-related transcription factors EIN and BZR1 in the low-light tolerant XF cultivar relative to the low-light sensitive GF cultivar, suggesting that the genes may be involved in the regulation of chrysanthemum's resistance to low light at the transcriptional level.
PIF transcription factors are required for the phytochrome-mediated SAR [30]. Elongation growth is inhibited by HY5, which competitively binds to PIF4 chromatin targets rather than regulating PIF4 transcription [31]. PIF7 promotes auxin biosynthesis and auxin redistribution [32], with auxins being essential mediators of the SAR such that auxin activation can modulate gene expression to promote shoot cell elongation [33,34]. AXR3/IAA17 mutant Arabidopsis thaliana plants exhibit reduced shade-induced hypocotyl growth [35]. Herein, we found LL treatment to enhance PIF7 expression, with a roughly four-fold increase in the expression of this gene in GF plants relative to XF plants (Fig. S3a, Supplementary material). SDH functions at the interface between the oxidative phosphorylation and TCA cycle pathways [36], with lower Arabidopsis SDH levels serving to suppress primary root elongation with early lateral root emergence [37]. SAR behaviors may thus be suppressed in low-light tolerant chrysanthemum cultivars through the maintenance of low IAA and elongation factor expression levels (CL1856). In contrast, exposure of the low-light-sensitive plants to low-light conditions severely disrupted and photosynthetic carbon assimilation (Fig. 2), resulting in profound growth inhibition and other SAR symptoms (Fig. 1).

Conclusions
Comparative transcriptome analysis revealed 1620 upregulated and 1207 downregulated genes between the GF LL and GF CK groups and 889 commonly upregulated and 556 downregulated genes between the XF LL and XF CK groups. GO classification and KEGG pathway analyses of these DEGs showed that most proteins encoded by the upregulated genes were localized in the chloroplast and involved in multiple biological processes associated with photosynthesis. This result indicated that these chloroplast proteins play an important role in increasing shading tolerance in chrysanthemum. The proteins encoded by the downregulated genes were assigned to 11 cellular components, classified into 8 molecular functions, and involved in 16 biological processes. Further analysis revealed that the proteins were associated with plant hormone signal transduction, photosynthesis-antenna, and shading regulation pathway. Therefore, our results provide new insights into the transcriptional alterations in low-light resistant and low-light sensitive chrysanthemum cultivars under low-light stress. These insights help unravel the underlying molecular mechanism of the response of C. morifolium to low-light stress.