Contrasting growth, physiological and gene expression responses of Clematis crassifolia and Clematis cadmia to different irradiance conditions

Clematis crassifolia and Clematis cadmia Buch.-Ham. ex Hook.f. & Thomson are herbaceous vine plants native to China. C. crassifolia is distributed in shaded areas, while C. cadmia mostly grows in bright, sunny conditions in mountainous and hilly landscapes. To understand the potential mechanisms involved in the irradiance responses of C. crassifolia and C. cadmia, we conducted a pot experiment under three irradiance treatments with natural irradiation and two different levels of shading. Various growth, photosynthetic, oxidative and antioxidative parameters and the relative expression of irradiance-related genes were examined. In total, 15 unigenes were selected for the analysis of gene expression. The exposure of C. crassifolia to high irradiance resulted in growth inhibition coupled with increased levels of chlorophyll, increased catalase, peroxidase, and superoxide dismutase activity and increased expression of c144262_g2, c138393_g1 and c131300_g2. In contrast, under high irradiance conditions, C. cadmia showed an increase in growth and soluble protein content accompanied by a decrease in the expression of c144262_g2, c133872_g1, and c142530_g1, suggesting their role in the acclimation of C. cadmia to a high-irradiance environment. The 15 unigenes were differentially expressed in C. crassifolia and C. cadmia under different irradiance conditions. Thus, our study revealed that there are essential differences in the irradiance adaptations of C. crassifolia and C. cadmia due to the differential physiological and molecular mechanisms underlying their irradiance responses, which result from their long-term evolution in contrasting habitats.


Materials and Methods
plant materials and growth conditions. A pot experiment was established using shading nets of different thicknesses in a room at the Zhejiang Institute of Subtropical Crops, Zhejiang Province, China (N28°23′, E120°72′). Seeds of one-year-old healthy and uniform Clematis seedlings (C. crassifolia and C. cadmia, 8-to 10-leaf age) were individually transferred to plastic pots (16.5 cm inner diameter, 18 cm height, with holes in the bottom) filled with a substrate mixture of perlite:peat:black soil (2:5:3, v/v/v, 45 kg m −3 of organic manure). After six months, the potted plants were grown under a temperature range of 20-32 °C and three different irradiance levels, (i) natural irradiance (T1), (ii) moderate shade (T2), and (ii) heavy shade (T3) (Fig. 1). The different irradiances were established using different layers of commercial black shading net (resembling square tents, commonly used in agriculture), and the irradiance was measured using a digital lux meter (TES-1339R, Taiwan) for a week and averaged. Planted pots were rotated daily to minimize the effect of the environment and irrigated daily to keep the plants well-watered (the water level was kept at 75% of the field capacity of the soil). After three months of growth, leaves were selected, and a completely randomized design with five replications per treatment was established. Each irradiance treatment included the two species of Clematis, C. crassifolia (C1) and C. Cadmia (C2). The treatments were as follows: T1C1 (C. crassifolia grown under natural irradiance conditions); T2C1 (C. crassifolia grown under moderate shade irradiance conditions); T3C1 (C. crassifolia grown under heavy shade conditions); T1C2 (C. cadmia grown under natural irradiance conditions); T2C2 (C. cadmia grown under moderate shade irradiance conditions); and T3C2 (C. cadmia grown under heavy shade conditions). Thus, the experiment was set up in a split-plot design, in which irradiance levels constituted the main plots and Clematis species were the subplots, and contained five replicates with twenty-five pots each.
Leaf growth analysis. After three months of the differential irradiance treatments, one intact plant from each replication of each treatment was selected to observe and record the growth conditions of the whole plant. The leaf samples of Clematis from all treatments were collected for growth analysis. The leaf mass was weighed by electronic scale, and the leaf area was measured by an LA meter (LI-300, Li-Cor, Lincoln, NE, USA). The leaves were killed by keeping them at 105 °C for 15 min and then dried in an oven at 80 °C until they reached a constant dry weight. Specific leaf weight (SLM) was calculated by leaf dry mass divided by the leaf area for each plant 25 . chlorophyll content analysis. Chlorophyll content was measured as described by Lichtenthaler with a spectrophotometer (Shimadzu UV-2550, Kyoto, Japan) and expressed as mg g −1 fresh weight (FW) 26 .
Gas exchange measurements. Healthy and fully developed leaves from each treatment were chosen for photosynthetic parameter measurements with a LI-6400 XT portable photosynthesis system (LI-COR Inc., Lincoln, NE, USA) equipped with a 6400-18 RGB LED light source. The measurements were carried out on sunny days from 9:30 to 11:00 am at an air concentration of 21% O 2 , 1200 µmol m −2 s −1 photosynthetically active radiation (PAR) white light, 400 µmol of CO 2 mol −1 of dry air, 55% relative humidity and a temperature range of 28-32 °C 27 . cellular membrane damage analysis. Cell membrane stability and integrity were expressed as the relative electrical conductivity (REC) and were measured as previously described 28 . The membrane peroxidation was expressed as the malondialdehyde (MDA) content and was measured by the method of Hodges et al. 29 .
peroxide (H 2 o 2 ) content and superoxide anion (o −2 ) production rate analysis. The H 2 O 2 content was determined with the method described by Patterson et al. 30 and expressed as mM mg −1 FW. The superoxide anion (O .−2 ) production rate was determined using the method of Wang and Luo 31 and expressed as µM min −1 mg −1 FW.
Antioxidant enzyme activity analysis. Leaf samples (1.0 g FW) were ground in a mortar containing 8.0 ml of grinding media consisting of 1% polyethylene pyrrole (PVP) and 50 mM phosphate buffer solution (pH 7.4) at 4 °C. Following centrifugation at 10,000 rpm for 15 min at 4 °C, the supernatants were collected to obtain crude enzymes. Total superoxide dismutase (SOD) activity was determined spectrophotometrically at 520 nm and expressed as U g −1 FW 32 . Catalase (CAT) activity was determined by monitoring the disappearance of H 2 O 2 , and specific CAT activity was expressed as U g −1 FW min −1 33 . The peroxidase (POD) activity was measured with the method described by Thomas et al., and specific POD activity was expressed as U g −1 FW min −1 34 .
Soluble protein content analysis. The soluble protein content was measured with Coomassie brilliant blue staining according to the method described by Bradford and was expressed as mg g −1 FW 35 .
Related gene expression analysis. We screened various genes related to the irradiance stress response through transcriptome sequencing to analyze gene expression (Table 1). Primer Premier 5 (www.Premierbiosoft. com/primerdesign) was used for designing primers. Table 2 shows the primer information for amplification of the analyzed genes. Total RNA was extracted from frozen and pulverized Clematis leaves using an RNeasy column (Qiagen USA, California, CA), and RNA samples for each replicate were pooled to obtain a single RNA sample for cDNA and cRNA probe preparation and expression profiling. qRT-PCR experiments were conducted as described by Gao  Data analysis. Statistical analysis was conducted using a two-factor analysis of variance with SPSS software version 16.0 (SPSS, Chicago, IL, U.S.). Tukey's multiple range test was applied to detect differences between the means. The data shown are the mean ± standard deviation (SD).

Results
plant growth and development. The C. crassifolia and C. cadmia grown under different irradiances for three months are shown in Fig. 2. Irradiance had different effects on the leaf growth of C. crassifolia and C. cadmia. Additionally, irradiance levels significantly affected the leaf growth of Clematis (p < 0.05) ( Table 3). After three months of irradiance treatment, the total leaf fresh weight per seedling of C. crassifolia plants increased by 8.5% (p < 0.05) and 21.78% (p < 0.05), respectively, under T2 and T3 light treatments compared with that of plants subjected to T1 irradiance treatments. Conversely, the leaf fresh weight of C. cadmia decreased by 25.82% (p < 0.05) and 52.23% (p < 0.05) under T2 and T3 irradiances, respectively, compared with that under

photosynthetic pigments. Irradiance had a notable influence on the photosynthetic pigment content in
Clematis leaves (Fig. 3). There were higher Chla, Chlb, and carotenoid concentrations in both C. cadmia and C. crassifolia plants under T2 and T3 light treatments than in those grown under natural sunlight (T1). In C. crassifolia, the Chla and Chlb content increased by 112.6% (p < 0.05) and 181.6% (p < 0.05) and by 138.2% (p < 0.05) and 204.4% (p < 0.05), respectively, under T2 and T3 light treatments compared with the Chla and Chlb content in plants under natural light intensity (T1). While a similar trend appeared in the leaves of C. cadmia plants, all three pigments were at higher levels in C. cadmia than in C. crassifolia across all irradiance conditions. The carotenoid contents in C. cadmia leaves increased by 31.8% (p < 0.05) and 82.8% (p < 0.05), respectively, under T2 and T3 irradiance compared with the carotenoid contents in C. cadmia leaves under T1. Specifically, the Chla/b ratio significantly decreased (p < 0.05) with decreasing irradiance in C. cadmia leaves, while no significant differences in Chla/b ratio were observed in C. crassifolia plants across the different irradiance treatments. photosynthetic parameters. The leaf photosynthetic parameters in C. cadmia and C. crassifolia were strikingly affected by the irradiance levels ( Table 4). The C. crassifolia plants exposed to natural irradiance (T1) showed less leaf net photosynthesis (Pn) (p < 0.05), stomatal conductance (Gs) (p < 0.05), and intercellular CO 2 concentration (Ci) (p < 0.05) and a lower transpiration rate (Tr) (p < 0.05) than those under T2 and T3 irradiance. Pn dramatically increased (p < 0.05) with decreasing irradiance in C. crassifolia plants, whereas a decrease (p < 0.05) in Pn in C. cadmia plants was observed when irradiance decreased from T1 to T3. Even under the same natural   production rate were notably different between C. cadmia and C. crassifolia leaves (Fig. 5). The H 2 O 2 content in C.   www.nature.com/scientificreports www.nature.com/scientificreports/ crassifolia leaves decreased by 65.6% (p < 0.05) and 87.03% (p < 0.05) under T2 and T3 irradiance, respectively, compared with that under natural light intensity (T1), whereas H 2 O 2 content increased by 210.3% (p < 0.05) and 679.3% (p < 0.05) in C. cadmia leaves under T2 and T3 irradiance, respectively. (Fig. 5). Similar responses were observed in the O 2 − production rate. A marked decrease was observed in the CAT (p < 0.05) and SOD activity (p < 0.05) in C. crassifolia growing under T2 and T3 light treatments compared with those in the control plants growing under natural irradiance (Fig. 6). Conversely, there were 92.8% (p < 0.05) and 140.4% (p < 0.05) increases in CAT activity and 29.8% (p < 0.05) and 13.5% (p > 0.05) increases in POD activity under T2 and T3 irradiance, respectively, in C. cadmia compared with the activity under natural light intensity (T1). Interestingly, the SOD activity in C. crassifolia leaves grown under natural irradiance (T1) was higher (p < 0.05) than that in the other light treatments, and no obvious differences (p > 0.05) were found between the other irradiance treatments.
Soluble protein content. The soluble protein in the C. cadmia and C. crassifolia leaves differed significantly (p < 0.05) under different light intensities (Fig. 7). A significant increase (p < 0.05) was observed in the soluble protein content from the T1 to the T3 irradiance treatments in C. crassifolia. In contrast, there was a dramatic reduction (by 35.3% (p < 0.05) and 48.87% (p < 0.05)) in the soluble protein content under the T2 and T3 irradiance treatments, respectively, in C. cadmia compared with that under natural light intensity (T1). Fig. 8, the genes in Clematis showed differential expression in response to various irradiance levels. A total of 15 genes were analyzed for differences in gene expression in C. cadmia and C. crassifolia leaves under three irradiance levels. Three of those genes, including c136757_g1, c144230_g2 and c145729_g1, are involved in the biosynthesis of phytochrome and chlorophyll in chloroplasts. The expression of these genes was significantly different (p < 0.05) in C. cadmia and C. crassifolia under the three  www.nature.com/scientificreports www.nature.com/scientificreports/ irradiance levels. The genes c144230_g2 and c144262_g2, which encode the receptor proteins in the Photosystem II core, were upregulated in C. crassifolia from T1 to T3, while the same genes were upregulated more than 5-fold (p < 0.05) in C. cadmia under T2 irradiance compared their levels under T1. The genes c131300_g2 and c133872_g1, which belong to the abscisic acid receptor family, were differentially upregulated in C. crassifolia and C. cadmia in response to the shaded conditions. A similar trend was observed in c133872_g1, c144154_g3 and c145729_g1 expression. The gene c139448_g1, which is associated with auxin response, first showed upregulation and then downregulation in C. crassifolia but showed no significant difference in C. cadmia. Similarly, the gene c142929_g5, which is also associated with auxin response, first showed upregulation and then downregulation in C. crassifolia and C. cadmia as the irradiance changed from T1 to T3.

Discussion
Excessive or deficient irradiance is the most common environmental stress factor that affects a series of physiological, developmental and biochemical changes in plants, including molecular and cellular responses 37 . However, plants are also known to cope with these stress conditions by adjusting their metabolism and physiology and making specific changes in gene expression 38 . Differences in plant growth, carbon allocation, physiology and gene www.nature.com/scientificreports www.nature.com/scientificreports/ expression have been investigated for species suffering from different levels of irradiance stress 39 . In this study, we have illustrated the growth, physiological mechanisms and changes in some gene expression patterns of C. cadmia and C. crassifolia in response to different irradiance levels.  www.nature.com/scientificreports www.nature.com/scientificreports/ Irradiance that is too intense can be stressful, resulting in reduced growth, leaf fading and foliage yellowing, as shown in Fig. 2, where the C. crassifolia plants under natural irradiance (T1) experienced growth inhibition accompanied by lower leaf fresh weight. This is probably because higher irradiance can lead to photoinhibition and other adverse reactions in C. crassifolia. However, C. cadmia showed increased growth vigor and leaf fresh weight under higher irradiance conditions, which demonstrated that C. crassifolia and C. cadmia prefer entirely different light environments. This difference was largely caused by long-term adaptation to their contrasting ecological niches. Moreover, the specific leaf weight in the two Clematis species decreased with the decrease in irradiance. This is a common response in species suffering from low irradiance, indicating that plants increase light energy capture by increasing dry weight per unit leaf area; this strategy has been observed in Torreya grandis 25 , Camptotheca acuminata 40 , and wheat 41 .
Plant leaf systems possess numerous fundamental functions for adapting to different irradiances, including chlorophyll content changes and photosynthesis. The changes in pigment content in Clematis are shown in Fig. 3. The decreasing irradiance elevated the content of carotenoids, chlorophyll a and chlorophyll b in both species of Clematis but reduced the Chla/b ratio in C. cadmia. These findings were consistent with observations in C. acuminata 40 , suggesting that a long exposure to high light caused pigments to degrade in C. crassifolia, while the content of chlorophyll b, as an antenna chlorophyll, increased in C. cadmia to absorb more light energy for photosynthesis; thus, the Chla/b ratio underwent a sharp decline in C. cadmia 40 . Photosynthesis is correlated with chlorophyll content in plants. In our study, decreasing irradiance significantly increased the Pn, Ci, Tr, and Gs values in C. crassifolia but lowered those values in C. cadmia. These results indicated that C. cadmia could take full advantage of the energy under high light (T1) and suitable growth conditions, but higher light intensity (T1) would lead to an immediate interruption of growth and excessive light energy accumulation in C. crassifolia 42 .
Accordingly, excessive light energy could impair the photosynthetic electron transport chain and interrupt PSII electron transport. Excess ROS can be produced by the direct transfer of excitation energy from chlorophyll, causing membrane permeability dysfunction and lipid peroxidation 43 . In this study, as C. crassifolia and C. cadmia were grown under different irradiance levels for three months, we observed that high light caused higher H 2 O 2 content and a higher O 2 − production rate in C. crassifolia, thereby resulting in a sharp increase in MDA content and coincident increases in the relative electrical conductivity (REC); however, C. cadmia showed the opposite response. This phenomenon clearly suggests that an inability to adapt to light intensity could lead to either a photooxidative or an inadequate stress response in plants, closely accompanied by ROS overproduction and ultimately leading to damage to plant membranes 44 . This result was in agreement with the effects of photoinactivation and lipid peroxidation described by Takahama and Nishimura 45 . Substantial evidence has indicated that plants can alleviate and repair the damage caused by ROS through antioxidant enzymes, including SOD, POD, and CAT 40,46,47 . Our research confirmed this point in Clematis and showed that the POD, SOD and CAT enzyme activity increased when plants suffered from irradiance stress.
Proteins encoded by various plant genes, including most of the known enzymes (such as POD, APX and GST) and some compatible modulators of metabolic processes, perform many biological functions 48 . Our study found that light intensity significantly impacted the soluble protein content in Clematis; the soluble protein content increased in C. crassifolia grown under low irradiance but decreased in C. cadmia under the same conditions. Similar results were reported in microalgae 49 and wheat 50 , which suggests that soluble protein content could represent the physiological status of plants to some extent 51 .
In general, plants respond to stresses with changes in the expression of related genes that encode proteins that provide protective effects in plants 52 . Our study focused on the responses of C. cadmia and C. crassifolia to different irradiances and screened 15 light-related genes to analyze their expression (Fig. 8). Overall, the 15 genes were differentially expressed under irradiance stress in C. cadmia and C. crassifolia. After three months of irradiance treatment, low light induced upregulated gene expression of numerous genes in C. crassifolia, which are involved in pigment biosynthesis, Photosystem II core protein synthesis, glycometabolism and auxin response protein synthesis. Our results in C. crassifolia are consistent with previous observations of some shade-tolerant plants, such as Anoectochilus formosanus 38 . This suggested that these genes contribute to adaptation to low light in C. crassifolia. PYL receptors perceive ABA intracellularly and, as a result, form ternary complexes initiating ABA signaling 53,54 . In this study, the PYL receptors were upregulated in C. cadmia grown under low light, which indicated that the C. cadmia plants were under stress when growing in low light (T2, T3) and that low light could activate ABA-responsive gene expression in the protoplasts, contributing to the adjustment of C. cadmia to the adverse environment. Interestingly, the superoxide dismutase [Cu-Zn] pathway gene was also upregulated in C. cadmia under low light, which indicated that C. cadmia increases the biosynthesis of SOD under low light. However, C. crassifolia showed the opposite response.
In summary, the present results confirm that the promotion of C. crassifolia growth under low irradiance may be related to the changes in gene expression in response to irradiance, the increased chlorophyll content and photosynthetic rate and the decreased ROS content. In addition, C. crassifolia and C. cadmia exhibit different physiological and metabolic characteristics, as well as antioxidant defense and gene expression patterns, under different irradiance conditions. This difference was largely due to the contrasting irradiance conditions in the habitats of the different species. Long-term evolution led to the differences between the species. Therefore, we proposed the hypothesis that irradiance might regulate the expression patterns of genes in these plants and might be an important environmental factor that affects the direction of plant evolution. The hypothesis will be tested in future work.