Introduction

Records on Magnoliaceae plants can be traced back to the Mesozoic era. Until now, many Magnoliaceae plants have been favored by people for their ornamental characteristics. However, with the changes of the climate, forest community composition, and the succession of forests, deciduous Magnoliaceae species are facing endangerment in the wild1. Nowadays, the growth and distribution of endangered species of Magnoliaceae are mostly confined to narrow areas. In natural condition, Magnoliaceae plants was mainly adapted to coniferous and broad-leaved mixed communities, while declined to evergreen broad-leaved communities2,3,4. Magnolia sinostellata is an endangered species belonging to family Magnoliaceae. It is a deciduous shrub, listed in the Red List of Magnoliaceae since 2016, which can grow to 3 m in height and blossom in early spring (February and March) in subtropical regions5. M. sinostellata is endemic to a narrow area of China (mainly distributed in Jingning, Wenzhou county, Zhejiang Province), with extremely small population. M. sinostellata is a sun-loving plant, and it largely grows under canopy shade or nearby the brook of the north slope in its natural habitats as its seedling growth requires mild degree of shading5. Moreover, pre-investigations indicated that M. sinostellata mainly distributed in coniferous forests, sparsely distributed in broad-leaved forests and mixed forests4. As the dominant tree species in the upper layer of different forests are different, the light condition in the under story of different forest type can be varying. For instance, the light intensity of coniferous communities is higher than that of evergreen broad-leaved communities6. Previous research has simulated light intensities under different forest types using black shade net, finding that heavy shading (equivalent to the light intensity under a closed broad-leaved forest) suppressed M. sinostellata seedlings by impacting their chlorophyll metabolic pathway, photosynthesis and antioxidant systems7. These findings suggest that alternation in light environment caused by upper canopy shade might be the key factor that limiting the distribution and population renewal of M. sinostellata. Canopy shading of evergreen trees in the upper layer changes the light intensity as well as the light quality (ratio of red/far-red light, R/FR), which will affect the growth of deciduous plants in the understory6,8,9,10,11,12. As light intensity and light quality that reaches chloroplasts is essential for activation of photosynthesis, the primary effect of shading on plants is to affect their photosynthesis13. The R/FR ratio of natural light in undergrowth declined as photosynthetic pigments of vegetative canopy absorb red light14. Low R/FR ratio changes chlorophyll content and directly affect plant photosynthetic capacity15. Leaf occlusion of upper layer can also reduce light intensity in the undergrowth, which directly affects photosynthesis via inhibiting the expression of photosynthetic proteins and altering chloroplast ultrastructure16,17. Under weak light conditions, light captured by light harvesting complex (LHC) is limited18. In Zea may, photochemical quenching coefficient (qp) and effective quantum yield of PSII photochemistry (ΦPSII) were significantly reduced under low light19. Due to the defect in PSII, Electron transport rate through PS I (ETR I) is significantly blocked in Rice20. Thus, nicotinamide adenine dinucleotide phosphate (NADPH) generated for further CO2 assimilation is insufficient. Furthermore, the activity of ribulose 1,5-bisphosphate carboxylase/oxygenase (Rubisco), key enzyme that transforms CO2 into carbohydrates21, is repressed and its content is decreased under shade22, indicating that carbon assimilation was suppressed under shading in shade sensitive plants such as rice23, Cucumis sativus24 and maize25. While moderate shading increased both photosynthesis Rubisco activity in C. sativus24 and Sargassum henslowianum26, thereby enhancing carbon assimilation and stress tolerance. Shade tolerant plants Arachis hypogaea27 and Quercus robur28 can maintain rubisco activity and photosynthesis efficiency under shading environment.

In plant, the rate of photosynthetic CO2 fixation determines the rate of starch and sucrose synthesis and other carbon metabolism29. Sucrose generated via the Calvin cycle, which further secreted into reservoir organs to participate starch synthesis30. The limitation of activity of sucrose phosphate synthase and sucrose synthase in response to shading indicates that its starch and sucrose metabolism was also inhibited31. For shade sensitive plants, with the increase of shading degree, the carbon assimilation and metabolism capacities of seedlings decreases32, water usage and transpiration are also affected33, which affect plant growth and community regeneration. In contrast, the expression of genes regulating photosynthesis and carbon metabolism in shade tolerant plant Solidago canadensis increased to maintain carbon fixation under weak light condition34.

To compete for sunlight, under vegetative shading, elongation response was triggered and flowering process was dramatically accelerated in shade-avoiding Arabidopsis35, cucumber36 and tomato37, etc. The change in light quality under canopy shade (low R/FR ratio) could be sensed by phytochrome photoreceptors14, which activate classical shade avoidance response subsequently. The low light intensity can also be detected in plants and cause elongation phenotype in shade sensitive soybean38 and Brassica39. Whereas, these responses can be detrimental, especially for agricultural crops and ornamental plants, because the reallocation of resources into vegetative growth reduced reproductive growth and crop yield40. In ornamental plants, shading inevitably altered their flower period, flower quality and numbers, which eventually destroyed their ornamental and economic values. For instance, weak light dramatically reduced flower quality, alkaloid yield and seed number of Papaver somniferum41. Although shading accelerated flowering in some plants, while in Paeonia lactiflora, shading led to delayed flowering date, as well as reduced flower fresh weight and faded flower color11. These studies showing that, even in shade sensitive species, the shade response can be species–dependent. In our previous study, we found that shading caused light deficiency boosted chlorophyll degradation, leaf chlorosis and senescence in M. sinostellata42, which also found in rice43 and Camellia sinensis44. While accelerated leaf abscission under weak light was only found in M. sinostellata and severely impaired its growth, which has not been reported in other species42. However, as the impact of light deficiency on reproductive growth of M. sinostellata and other vital metabolic mechanism still unknown, these findings were not able to explain how low light affects the population renewal of M. sinostellata, which eventually limiting its distribution in the wild. Furthermore, despite the light environment under canopy shade containing reduced light intensity and inconsistent light quality, in order to reduce the confounding factors, many studies mainly focused on impact of weak light on plant growth45,46,47. Accordingly, in this study, we continued investigating the effect of light deficiency on M. sinostellata.

Although numerous low-light responsive pathways, transcription factors, and R-genes has been identified in M. sinostellata, the hub genes in its light deficiency remains unknown and the weak light response mechanism still obscure. In this study, co-expressed modules and hub genes in transcriptomes of light deficiency treated and untreated M. sinostellata leaves were pinpointed via weighted gene co-expression network analysis (WGCNA). The functions of these genes were analyzed via KEGG, Gene Ontology (GO) and National Center for Biotechnology Information (NCBI), which shed more light on the mystery of light deficiency response mechanism of M. sinostellata. This study forms a theoretical basis for protection and conservation of M. sinostellata and other shade sensitive woody plants.

Results

Identification of the light deficiency-responsive core DEGs in M. sinostellata

Significant morphological and physiological changes were observed in M. sinostellta, relative to control group (CK), under light deficiency condition. The seedlings in CK group grew well and leave were fully expanded during the experiment. Five days after being subjected to light deficiency condition, seedlings began to wilt. By 15 d of low light treatment, the wilting of the seedling aggravated and severe leaf abscission was observed (Fig. S1). Moreover, the net photosynthetic rate (Pn) and photochemical efficiency (Fv/Fm) declined obviously during the treatment, indicating that photosynthesis ability of M. sinostellata impaired42.

In order to gain more insight into the weak light response mechanism, transcriptome sequencing was performed. A total of 181, 902 genes were detected via transcriptome sequencing, and the average comparison rate with the compared gene set was 80.49%. 246,481 non-redundant transcript sequences were obtained. The total length of these sequences is 270,112,156, the longest sequence is 20445 bp, the shortest sequence is 199 bp, and the average sequence length is 3420.95 bp. In this study, we analyzed the global gene expression profiles of M. sinostellata for light deficiency response using five datasets, including M-D0, CK-D5, LT-D5, CK-D15 and LT-D15, the details of which are shown in Table S1. Each dataset contains 3 replicates. Correlation analysis was performed among these 15 samples, which showed good reproductivity in the same group and a significance difference between control groups and shade treated groups (Fig. S2). This analysis suggests that these samples can be used for subsequent screening and analysis.

Pairwise comparison identified 17,943, 9102, 19,790 and 17,994 in M-D0 vs LT-D5, M-D0 vs LT-D15, CK-D5 vs LT-D5 and CK-D15 vs LT-D15 comparison groups, respectively (Log2FC > 1, Qvalue < 0.05). 4734 core DEGs were found in their cross-compared Venn diagrams, which were significantly induced or suppressed during the experiment (Fig. 1, Table S2). Then, we characterized these 4734 core DEGs to get more insight into their contributing molecular pathways. First, 4734 DEGs were subjected to GO analysis, which were classified into three groups and 33 subgroups (Fig. 2A). The biological process group can be divided into 18 subgroups, among which cellular process and metabolic process were the top two sub-groups involved the most genes. Four subgroups were related to cellular component, among which cellular anatomical entity and intracellular were the main subgroups involved most genes. Eleven subgroups constitute the molecular function group, and the catalytic activity and binding involved the most genes. GO enrichment analysis showed that most significant GO terms are related to photosynthesis. The top five enriched GO terms were (GO:0009522); photosynthesis, light harvesting (GO:0009765); photosystem (GO:0009521); photosynthesis (GO:0015979) and photosystem II (GO:0009523) (Fig. 2B). Then, we performed KEGG analysis to explore contributing pathways of 4734 DEGs (Fig. 3A). For KEGG classification, genes were annotated into five groups and 19 subgroups. Among metabolism group, global and overview maps, carbohydrate metabolism and energy metabolism subgroups were significantly enriched. Photosynthesis-antenna proteins was the most significantly enriched KEGG pathway. The top five frequently enriched pathways were as follows: photosynthesis-antenna proteins, galactose metabolism, phenylpropanoid biosynthesis, starch and sucrose metabolism and alanine, aspartate and glutamate metabolism (Fig. 3B).

Figure 1
figure 1

Venn diagram of differential expression genes in treated and control groups in M. sinostellata (R software, version 4.0.3, www.gnu.org/software/r/). Venn diagram showing 4734 core low light responsive DEGs among treated and control groups used in this study.

Full size image
Figure 2
figure 2

Gene ontology (GO) analysis of 4734 core DEGs (R software, version 4.0.3, www.gnu.org/software/r/). (A) GO classification of 4734 core DEGs. 4734 core DEGs were divided into three main categories and 36 subgroups. (B) GO enrichment of 4734 core DEGs. The top 20 GO Terms with the smallest Qvalue were selected to plot the chart.

Full size image
Figure 3
figure 3

KEGG analysis of core DEGs in M. sinostellata (R software, version 4.0.3, www.gnu.org/software/r/). (A) KEGG classification of core DEGs. The metabolism pathways contributed by core DEGs were divided into five groups and 19 subgroups. (B) KEGG enrichment of 4734 core DEGs. The top 20 KEGG pathways with the smallest Qvalue were selected to plot the chart, among which 'Photosynthesis-antenna proteins' was the most enriched pathway.

Full size image

WGCNA of core light deficiency responsive DEGs in M. sinostellata

To identify key shade responsive genes in M. sinostellata, WGCNA was performed to analysis 4734 core DEGs, which can identify modules of highly related genes. The selection of an optimal soft threshold power is an essential step to construct WGCNA. A network topology research of 1–20 was executed, and the scale independence and the average connectivity of the WGCNA relative equilibrium were determined. Threshold 18 was selected to construct a hierarchical clustering tree of DEGs (Fig. 4A). MEDiss Thres was set to 0.22 to merge similar modules and 4 modules was generated (Fig. 4B). Genes in grey module that cannot be assigned to any modules were not analyzed in the further study. Four different modules were generated by WGCNA including module 1, module 2, module 3 and module 4, which including 2008, 75, 2481, and 71 DEGs, respectively (Table 1). To identify co-expression similarity of modules, characteristic genes were calculated and clustered according to their correlation (Fig. 4C). We found that these 4 modules are divided into two categories: the first included module 1 and module 2 modules; the second included module 3 and module 4. Gene modules of the same category may have similar functions or contributing to the same regulatory mechanism. To investigate modules associated with light deficiency treatment, we plotted module-trait relationship heat map (Fig. 4D). This result shows that module 1 positively correlated with two weak light-treated groups (LT-D5, r = 0.7; LT-D15, r = 0.49) but negatively correlated with control groups (M-D0, r = − 0.40; CK-D5, r = − 0.38; CK-D15, r = − 0.41). Similarly, module 2 had a strong correlation with LT-D15 (r = 0.68), while had a negative correlation with M-D0 (r = − 0.26), CK-D5(r = − 0.25) and CK-D15 (r = − 0.25). In contrast, module 3 showed a positive correlation with control groups (M-D0, r = 0.38; CK-D5, r = 0.35; CK-D15, r = 0.49) but illustrated a significant negative correlation with treated groups (cor < − 0.5, p < 0.05). However, genes in module 4 showed no significant correlation with treated groups or control groups. These results suggest that DEGs in module 1 and module 2 mainly up-regulated and DEGs in module 3 down-regulated under light deficiency treatment. These results showed that module 1, module 2 and module 3 were significantly correlated with light deficiency treatment. To obtain further understand of the expression pattern of these three modules, heat maps of gene expression for these modules were generated along with eigengene expression values (Fig. 5). We observed that genes inmodule 1 were significantly responsive to light deficiency and mainly up regulated. In module 2, genes were slightly induced in LT-D5 while markedly up-regulated in LT-D15. In contrast, genes in module 3 mainly downregulated during the experiment.

Figure 4
figure 4

WGCNA module identification of 4734 core DEGs in M. sinostellata (R software, version 4.0.3, www.gnu.org/software/r/). (A) hierarchical clustering tree of 4734 core DEGs in M. sinostellata. (B) Dynamic tree cut and merged modules. (C) Dendrogram heatmap of four merged modules. (D) Module-trait relationship heat map. the correlation of the identified modules in control and shade treated groups. Red and green color illustrated positive and negative correlation with gene expression, respectively.

Full size image
Table 1 Details of four detected modules.
Full size table
Figure 5
figure 5

Eigengene expression pattern of key light deficiency associated modules (R software, version 4.0.3, www.gnu.org/software/r/). (A) Eigengene expression pattern of Module 1. (B) Eigengene expression pattern of Module 2 module. (C) Eigengene expression pattern of Module 3. Eigengene expression pattern is the optimal tool to summarize expression pattern of specific module.

Full size image

GO and KEGG pathways analysis of DEGs in key modules in M. sinostellata associated with weak light

The functions of genes under light deficiency treatment in specific modules and its contributing regulatory pathways were revealed by GO and KEGG analysis. GO analysis suggested that ‘N-acetyltransferase activity’ and ‘acetyltransferase activity’ as the most significantly enriched GO terms in module 1, which means genes in this module mainly response to weak light by regulating acetyltransferase activity (Fig. 6A). KEGG analysis in module 1 identified ‘Valine, leucine and isoleucine degradation’ and ‘Amino sugar and nucleotide sugar metabolism’ as the most enriched regulatory pathways (Fig. 6B). Concerning module 2, GO analysis displayed that ‘hydrolase activity’ and ‘hydrolyzing O-glycosyl compound’ were the top two enriched GO terms indicating that these genes regulating hydrolase-related metabolism (Fig. 6C). KEGG analysis demonstrated that ‘Starch and sucrose metabolism’ pathway was appreciably influenced by low light (Fig. 6D). These results matched the results of GO and KEGG analysis of 4734 core DEGs, indicating that these results are reliable. In module 3, GO analysis indicated that enriched GO terms of cellular component, biological process and molecular function were mainly related to photosynthesis, among which the top five GO terms were ‘photosystem I’, ‘photosynthesis, light harvesting’, ‘photosystem’, ‘photosynthesis’ and ‘thylakoid’ (Fig. 6E). KEGG analysis showed that ‘Photosynthesis-antenna proteins’ and ‘Photosynthesis’ were the most enriched pathways (Fig. 6F). These indicate that genes in module 3 mainly involved in photosynthesis.

Figure 6
figure 6

GO and KEGG enrichment analysis of Module 1, Module 2 and Module 3 (R software, version 4.0.3, www.gnu.org/software/r/). (A) GO enrichment analysis of Module 1. (B) KEGG pathway enrichment analysis of Module 1. (C) GO enrichment analysis of Module 2. (D) KEGG pathway enrichment analysis of Module 2. (E) GO enrichment analysis of Module 3. (F) KEGG pathway enrichment analysis of Module 3. The top 20 GO terms or KEGG pathways with the smallest Qvalue were selected to plot these charts.

Full size image

Hub genes identification in key Co-Expressed modules in M. sinostellata under light deficiency

To identify hub genes in the three modules, genes with a weight parameter over 0.4 were analyzed and visualized through Cytoscape 3.8.2 to construct interaction networks (Fig. 7). We identified 6, 7 and 8 hub genes in module 1, module 2 and module 3 respectively via integrated analysis results of MCODE, cytoHubba and Centiscape2.2 in Cytoscape3.8.2 (Table 2). Isoform_16555 (Anthesis Pomoting Factor 1, MsAPF1), isoform_15622 (DUF1644 domain-containing protein, MsSIZ1), isoform_210768 (Acyl-CoA N-acyltransferase protein, MsGNAT6), isoform_13861 (Detoxification 21, MsHMP21), isoform_16567 (Mitogen-activated protein kinase 10, MsCXIP4), and isoform_107196 (Unknown) were identified as hub genes in module 1, indicating that these genes have significant functions in Amino acids and nucleic acids metabolism under weak light. In module 2, isoform_10150 (Beta-glucosidase 18, MsBGL18), isoform_92874 (Basic 7S globulin, MsBg7S), isoform_192429 (Cytochrome P450 710A11, MsCYP710A11), isoform_238198 (Transcription factor TGA2.2, MsTGA2), isoform_55976 and isoform_152869 (Pathogenesis-related protein P2, MsPR4) might play central roles in carbon fixation related pathways in response to light deficiency.

Figure 7
figure 7

Co-expression network analysis of core light deficiency responsive modules (Cytoscape Version 3.8.2, https://cytoscape.org/). (A) Co-expression network of Module 1. Eight hub genes were identified in this module. (B) Co-expression network of Module 2. There are seven hub genes detected in this module. (C) Co-expression network of Module 3. Six hub genes were found in this module.

Full size image
Table 2 The hub genes detected in three modules.
Full size table

In module 3, isoform_10052 (FAS1 domain-containing protein, MsFLA15), isoform_12760 (UDP-rhamnose:rhamnosyltransferase1, MsUGT73C7), isoform_121918 (UDP-rhamnose:rhamnosyltransferase1, MsUGT71C1), isoform_108687 (Subtilisin-like protein protease SBT3.3, MsSBT3), isoform_11567 (FAS1 domain-containing protein, MsFLA17), isoform_28177 (L-type lectin-domain-containing protein, MsLECRK-V.1), isoform_119775 (NADPH-dependent FMN reductase, MsFMN) and isoform_47022 (Geraniol 8-hydroxylase-like protein, MsGHL) were identified as hub genes, suggesting that these genes exert vital functions in photosynthesis in M. sinostellata under low light conditions (Table S3).

To verify the expression of 21 hub genes under light deficiency treatment after 0d, 5d and 15d in M. sinostellata, quantitative reverse-transcription PCR (RT-qPCR) was performed. These results suggested that the expression level of all the hub genes were significantly altered during light deficiency treatment, which indicating that these genes were all light deficiency-responsive (Fig. 8). Interestingly, hub genes in module 1 and module 2 were all significantly up regulated during, while 8 hub genes in module 3 were all down regulated under light deficiency in M. sinostellata.

Figure 8
figure 8

Relative expression of 21 hub genes in M. sinostellata under weak light and control conditions (GraphPad Prism8.0, https://www.graphpad.com). Data are the means of three biological replicates and three technical replicates. The 2-ΔΔct method was employed to conduct the gene differential expression analysis.

Full size image

Discussion

Canopy shade is a major abiotic stress that affect plants growth and reproduction in the wild. In this study, we found that shading caused light deficiency impaired photosynthesis capacity, enhanced carbon and amino acid metabolism, and flowering related pathways of M. sinostellata through WGCNA of transcriptome data. Light deficiency has significant impact on morphology of M. sinostellta. Leaves and seedlings of M. sinostellta gradually wilt under light deficiency treatment, and severe leaf abscission was observed, which was distinct from low light responses of module plants such as Arabidopsis48 and rice46. Photosynthesis is essential for plant growth and development, and key photosynthetic indicators Pn and Fv/Fm were both decreased significantly in M. sinostellta42.

The transcriptome analysis results confirmed these results subsequently. Genes in module 3 were all down-regulated and mostly involved in photosynthesis related pathways (Fig. 6E,F), indicating that its photosynthesis was inhibited. The suppressed photosynthesis under weak light also have been reported in various plants, including Arabidopsis49, cucumber24, maize50, etc. The most direct effect of impaired photosynthesis is the decreased production of glucose, followed by blocked carbon mechanism, and finally impact plant growth51. More importantly, glucose and its secondary metabolites sucrose and fructan are the main soluable sugar to maintain cell membrane and turgor under adverse stress52. Various studies have found that low light decreased content and expression of regulated genes of soluble sugars, superoxide dismutase (SOD) and peroxidase (POD) in cucumber53, Olea europaea54, M. sinostellata7, showing that their stress tolerance ability weakened. However, these genes mainly located in the downstream of stress tolerance molecular mechanism in M. sinostellata, and the hub genes in its upstream network remains obscure. In module 3, eight hub genes were identified and all regulating stress tolerance related pathways. MsUGT73C7 (isoform_12760) and MsUGT91C1 (isoform_121918) belongs to UDP-glucose: glycosyltransferase (UGT) family, most of which can be induced by abiotic stresses55. MsUGT73C7 and MsUGT91C1 reduced the expression of superoxide dismutase synthesis genes (isoform_15962), the function of which is to protect plants under stress by scavenging reactive oxygen and nitrogen species. Lectin-domain-containing protein and NADPH-dependent FMN reductase were reported to have essential roles in stress response56,57. MsLECRK-V.1 (Lectin-domain-containing protein, isoform_28177) was predicted to down regulate the expression of MsFMN (NAD(P)H dehydrogenase, isoform_119775), which both able to regulate plant basal resistance58,59. FAS1 domain-containing proteins have important function in plant development and abiotic stress response60, among which MsFLA15 (isoform_10052) and MsFLA17 (isoform_11567) can both repress the expression of MsFMN. One of the vital functions of MsFMN is to impact utilization efficiency of photosynthetic products and plant cell wall synthesis via inhibiting the expression of cellulose synthase synthesis genes (isoform_48903), which is also an important factor to cope with various stresses61. Subtilisin-like protein protease and Geraniol 8-hydroxylase-like protein is associated with plant pathogen resistance and MAPK signaling pathways, respectively62,63. MsSBT3 (Subtilisin-like protein protease, isoform_108687) can attenuate the expression of the MsGHL (Geraniol 8-hydroxylase-like protein, isoform_47022) under low light condition. These results verified that the stress tolerance ability in M. sinostellata under light deficiency depressed, which are essential for plant adaptation and survival.

In contrast, the expression levels of genes in module 1 and module 2 were all induced, which mainly regulating amino acid metabolism and carbon metabolism, respectively. As amino acid metabolism is strongly correlated with carbon metabolism64, the expression patterns of module 1 and module 2 were quite similar. Generally, the decrease in glucose production due to the inhibited photosynthesis under light deficiency will block the starch and sucrose metabolism and further carbon metabolism, such as peanut8 and wheat65. While shade tolerant plants can maintain high photosynthesis efficiency and carbon metabolism to support their life activities under light deficiency due to lengthy adaption and evolution66. In shade sensitive soybean, enhancing photosynthesis as well as carbon metabolism via silicon treatment can improve its growth under weak light condition8. Surprisingly, in M. sinostellata, although photosynthesis related genes inhibited under light deficiency, the genes in module 2 related to starch and sucrose metabolism and other carbon metabolism were significantly induced. The expressions of coding genes of principal enzymes in starch and sucrose metabolism including sucrose synthase and sorbitol dehydrogenase were all activated (Table S5). One theory is that low light intensity impact plant growth via a negative imbalance in carbon metabolism67. Owing to the enhanced starch and sucrose metabolism, the carbon imbalance state in M. sinostellata under weak light was much worse. The six hub genes identified in this module make this carbon metabolic molecular interacting network clearer. Basic 7S globulin is multifunctional and have essential roles in starch and sucrose pathways68. Beta-glucosidase 18 (MsBGL18) up-regulated the expression of MsBg7S (Basic 7S globulin, isoform_92874) and MsCYP710A11 (Cytochrome P450 710A11, isoform_192429) in the sucrose and starch metabolism pathway in M. sinostellata. MsBGL18 can also controls starch and sucrose metabolism through beta-glucosidase in M. sinostellata, which homology was reported involved in various pathways, including activation of chemical defense compounds, phytohormones, and metabolites69. Interestingly, MsPR4 (Pathogenesis-related protein, isoform_152869) regulated key enzymes such as beta-glucosidase, beta-amylase and trehalose 6-phosphate synthase to promote the metabolism of sucrose and starch under shade. Unknown protein (isoform_55976) is unique in M. sinostellata was also predicted to regulate the metabolism of sucrose and starch via activating the activity of beta-glucosidase and threonine aldolase.

The environmental stimuli and alternation of metabolic states can impact plant flowering70. Interestingly, the effect of shading on flowering period is variant. Low R/FR ratio under vegetative shade accelerated flowering in Arabidopsis71 but delayed flowering in Medicago sativa72. In addition, shading caused low light intensity induced flowering related genes in peanut8, while postponed flowering process of Odontonema strictum73 and Paeonia lactiflora11. The expression of seven hub genes in module 1 were all increased, and mostly controlled flowering related genes, showing that flowering was promoted in M. sinostellata under light deficiency. DUF1644 domain-containing proteins are abiotic stress responsive in rice74 and Anthesis pomoting factor 1 (AtARF1) have essential roles in plant flowering in Arabidopsis75. In M. sinostellata, abiotic stress responsive MsSIZ1 (DUF1644 domain-containing protein, isoform_15622)74 up regulated the expression of Anthesis pomoting factor MsAPF1 (isoform_16555), which has vital function in flowering. MsHMP21 (Detoxification 21 isoform X1, isoform_13861) enhanced the expression of important flowering-time gene Constans (isoform_25713 and isoform_21721) together with MsAPF1. Acyl-CoA N-acyltransferase protein was reported regulating the reproductive growth and flower bud differentiation in Hordeum vulgare76. Hub gene isoform_107196 (Unknown) was predicted to induce MsGNAT6 (Acyl-CoA N-acyltransferase protein, isoform_210768) in M. sinostellata.

In general, light deficiency impaired photosynthesis, stress tolerance, and accelerated carbon metabolism and flowering in M. sinostellata. When the photosynthesis capacity is damaged, the enhancement in carbon metabolism exacerbated the consuming of plant carbon reservoir. Also, for perennial plants, early flowering was harmful for accumulation of reserves for resume growth in the next season72. Therefore, M. sinostellata is hyper-sensitive to light deficiency, which might be the main cause that limiting its distribution and population renewal in forests. Hub genes isoform_107196 in module 1 and isoform_55976 in module 2 were unique to M. sinostellata, and their functions will be investigated in our future work. This study provides new insight into light deficiency response mechanism in M. sinostellata and laid a firm foundation for further protection and conservation of Magnolia and other shade sensitive woody plants.

Methods

Plant material and experimental design

The M. sinostellata seedlings used in this study were collected by Yamei Shen’s research group, Zhejiang Agriculture & Forestry University, Hangzhou, Zhejiang Province, China (latitude 30°26′ N, longitude119°73′ E) and cultivated in Qingshan Lake Garden Center, Hangzhou, Zhejiang Province, China (latitude 30°25′ N, longitude119°81′ E), which both have been approved by Forestry Bureau of Zhejiang Province, P. R. China. In addition, experimental research on M. sinostellata was strictly comply with the IUCN Policy Statement on Research Involving Species at Risk of Extinction and the Convention on the Trade in Endangered Species of Wild Fauna and Flora. The seedlings were placed in artificial climate room (luminance 1400 ± 30 μmol m−2 s 1, photoperiod 14 h light, temperature 25 ± 2 °C, humidity 40–60%) in Zhejiang Agriculture & Forestry University throughout the experiment. In order to simulate the closed canopy of forest, the shade experimental set-up was built with three layers of black shade net (25% full light, luminance 350 ± 30 μmol m−2 s−1) and bamboo poles. The control group seedlings were unsheltered (100% full light, luminance 1400 ± 30 μmol m−2 s−1), and the other conditions remained unchanged. There were no leaves or stems overlapping between plants, each treatment had 3 replicates. Plants were subjected to light deficiency condition for 5 (LT-D5), and 15 days (LT-D15) or kept in the control condition at indicated time points (CK-D5; CK-D15). To analysis dynamic changes of gene expression and reduce error, samples of 0 day (M-D0) used in this study were evenly collected from leave samples of control and treated groups of 0 day. These samples were frozen in liquid nitrogen and stored at − 80 °C for further experiments and RNA-seq analyses. At sampling, collections were performed from 3 plants, and each sample collection was repeated 3 times for biological replicates.

Transcriptome sequencing and data analysis

To identify shade responsive genes, RNA sequencing was performed to analyze transcriptome gene expression profiles in 15 samples. These 15 samples were divided into five groups, including three control groups (M-D0, CK-D5 and CK-D15) and two light deficiency-treated groups (LT-D5 and LT-D15). Total RNA of 15 samples were proceed by mRNA enrichment method or rRNA removal method. The purified RNA was fragmented with the interrupted buffer and reversed with random N6 primer, and then synthesized into cDNA two-strand to form double-stranded DNA. The ends of synthetic double-stranded DNA are filled in and the 5'end is phosphorylated. The 3'end forms a sticky end with an ‘A’ protruding, and then a bubbly linker with a protruding ‘T’ on the 3'end is connected. The ligation product is amplified by PCR with specific primers. The PCR product is heat-denatured into single-stranded, and then the single-stranded DNA is circularized with a bridge primer to obtain a single-stranded circular DNA library. DNBSEQ platform was employed to sequence the libraries. The R package (edgeR v3.16) was employed to identify the differentially expressed genes (DEGs) between weak light-treated and control samples. Genes fulfilled stringent criteria were identified as DEGs (fold-change > 2 and q value < 0.05, with false discovery rate (FDR) less than 0.05). The function and involved pathways of the DEGs were classified according to the GO and KEGG77 annotation results and official classification. Phyper function in R software was used for enrichment analysis.

Weighted gene co-expression network construction and hub genes detection

The weighted gene co-expression network was constructed using WGCNA package in R software to further analysis gene functions and contributing pathways in response to light deficiency in M. sinotellata. The 4734 core DEGs of 15 samples were used to construct this network. The similarity matrix was calculated by identifying the Pearson correlation coefficient between all gene pairs. The correlation matrix was transformed by soft-thresholding process to mimic the scale-free topology. The adjacency matrix was converted into a topological overlap matrix (TOM), and all coding sequences were hierarchically clustered by TOM similarity algorithm. The co-expression gene modules of the gene dendrogram were detected by the dynamic tree cut method, which using a height-cut less than 0.22. The module membership (MM) and the significance gene (GS) were calculated and used to confirm the distinguished modules. Module network visualization was performed by Cytoscape 3.8.2 with a cut-off of weight parameter set at 0.4. MCODE, cytoHubba and Centiscape2.2 in Cytoscape3.8.2 were used to identify hub genes.

RT-qPCR analysis of hub genes expression

Total RNA was extracted from the leaves of M. sinostellata in treated and control groups of 0 d, 5 d and 15 d using ultra-clean polysaccharide and phenol plant RNA purification kit. 1 μg total RNA was converted to first-strand cDNA using the Prime Script RT master Mix. RT-qPCR were performed with Light Cycler 480 II (Roche) using BCG qPCR Master Mix. Results were analyzed with the software Light Cycler 480 SW 1.5.1. Experiments was performed under the following conditions: 95 °C for 30 s; 95 °C for 5 s and 58 °C for 30 s. A total of 40 cycles were performed. After the end of the program, melting curves were generated (65–95 °C, 0.2 °C increment). Relative gene expression was calculated using 2−∆∆Ct method and M. sinostellata EF1-α was employed as the reference gene5. DNA primers used are listed in Table S4. Each sample testing was repeated at 3 times.

Statistical analysis

Statistical analyses and graphing of venn diagrams, GO and KEGG classification77 and enrichment plots, diagrams of WGCNA modules, eigengene expression pattern of modules were performed using R software (version 4.0.3, www.gnu.org/software/r/). Hub genes in these three modules were visualized through Cytoscape (Version 3.8.2, https://cytoscape.org/). GraphPad Prism (GraphPad Prism8.0, https://www.graphpad.com) software were used for statistical analyses and bar charts plotting of RT-qPCR analysis of hub genes.

Ethical approval

Magnolia sinostellata seedlings collected from Qingshan Lake Garden Center in Hangzhou, Zhejiang Province, China, have both been approved by Forestry Bureau of Zhejiang Province, P. R. China. In addition, experimental research on M. sinostellata was strictly comply with the IUCN Policy Statement on Research Involving Species at Risk of Extinction and the Convention on the Trade in Endangered Species of Wild Fauna and Flora.