Integrated Analysis of the Transcriptome and Metabolomics Reveals an Essential Role for Auxin in Hypocotyl Elongation During End-of-Day Far-Red Treatment of Cucurbita Moschata

induced by EOD-FR. The IAA content in leaves increased signi�cantly after EOD-FR treatment, and the content of IAA in leaves of the NPA-T group was the highest. When NPA was applied alone, the IAA content of leaves and hypocotyls decreased signi�cantly. The above results indicate that EOD-FR treatment can increase free IAA levels in the plant, especially in the hypocotyl, but that this effect is counteracted by NPA. This shows that the level of free IAA in the hypocotyl is an important factor in promoting cell expansion and hypocotyl elongation


Background
The control of plant morphology is critical to commercial greenhouse transplant production [1].As an important transplant production technique, grafting plays a key role in controlling soil-borne diseases and improving plant resistance to abiotic stress during crop cultivation [2].In vegetable grafting propagation, the use of rootstocks with long hypocotyls and uniform growth can improve grafting speed and prevent vulnerable scions from coming into contact with the soil during and after transplant [1].However, because seasonal variation in greenhouse environmental conditions affects elongation, the continuous production of rootstocks with long and uniform hypocotyls remains a signi cant challenge [3].
Light is essential for plant growth and development, serving as an energy source for photosynthesis and an environmental cue for photomorphogenesis [4].Numerous studies have con rmed that the use of different light quality treatments signi cantly affects early plant morphology, speci cally the promotion of seedling hypocotyl elongation [5,6] [4,5].In recent years, arti cial light sources have become a new type of 'intelligent' equipment in agricultural production facilities, particularly for vegetable production [7].Based on the production goals and crop light demand, optimum light conditions for crop growth can be provided.New opportunities for the production of long hypocotyl rootstocks in growth facilities are now possible.
The use of far-red light at the end of the day to affect plant hypocotyl length has been studied by biologists for many years [1,3,8,9].The technique is based on plants' use of phytochrome to sense the red/far-red ratio (R/FR) of the external light environment [10].When FR light is absorbed, the structure of phytochrome changes reversibly from active Pfr to Pr [11].Decreased Pfr content in the nucleus leads to an increase in the activity of phytochrome interacting factors (PIFs) [12].In this way, phytochrome perceives information on the surrounding light and correspondingly mediates the expression level of genes regulated by PIF [13].EOD-FR can mimic growth in low R:FR conditions by decreasing Pfr levels before the onset of darkness, thereby maintaining a low Pfr throughout the dark period [14].
Previous studies have shown that EOD-FR mediates the hypocotyl elongation of tomato, pumpkin and other crops without affecting plant dry weight, stem diameter, or seedling index.There is exibility in the selection of FR light intensity and duration to provide a su cient FR light dose (light intensity × duration).
When the dose of FR light falls within a certain range, the response of hypocotyl length can be described accurately by the Michaelis-Menten saturation curve [3].Therefore, EOD-FR has signi cant potential for the morphological regulation of seedlings in growth facilities [1,3].
Although research on EOD-FR is continuously being performed, there have been few studies on the mechanisms by which EOD-FR promotes hypocotyl elongation.A variety of phytohormones are known to regulate elongation growth in response to the R:FR environment [13,15].Among them, auxin is thought to be one of the most important factor [16,17], but its speci c role in EOD-FR-induced elongation has not been studied in detail.Moreover, although researchers have previously used RNA-seq to analyze differences in the transcriptome of Platycodon grandi orum after EOD-FR treatment [18], the auxin signaling pathway is extremely complex, and there are many auxin response genes in a single family.The complexities of auxin response involve the modulation of both transcriptional and metabolic networks, and the use of RNA-seq alone has limitations in predicting the central metabolic pathways and pivotal reactions that underlie this phenomenon [19].
In recent years, the use of metabolomics techniques for the quantitative analysis of small molecular weight plant metabolites has become increasingly widespread.These techniques enable researchers to explore the relationships between metabolite levels and phenotypic changes, providing new insights into multiple phenomena [20].
Pumpkin, the vegetable of Cucurbitaceae, has stronger root system, disease resistance and stress resistance, so it is widely used as rootstock in the grafting of melons [21].Here, we performed metabolite and transcriptional pro ling of pumpkin (Cucurbita moschata) hypocotyls exposed to EOD-FR and combined these results with related physiological measurements to investigate the role of auxin in EOD-FR-mediated hypocotyl elongation and identify key genes and pathways.The expression of a subset of genes was veri ed by quantitative real-time polymerase chain reaction (qRT-PCR).The resulting data expand our understanding of auxin synthesis and signal transduction during EOD-FR-mediated elongation at the metabolic and molecular levels.They also provide valuable mechanistic information for the application of EOD-FR in greenhouse production settings.

Results
Effects of EOD-FR light on growth and hypocotyl cell morphology of pumpkin seedlings Hypocotyls were signi cantly longer in EOD-FR-treated plants than in control plants (Fig. 1a).However, there were no signi cant differences in stem diameter or above-and belowground fresh and dry weights between the CK and T. After treatment with inhibitors, not only was hypocotyl elongation inhibited, but stem diameter and dry and fresh weights also decreased signi cantly (Fig. 1b).Using axial sections of hypocotyl cells (Fig. 2a), we observed the length and width of parenchyma cells in the four treatments (Fig. 2b).Result showed hypocotyl parenchyma cells expanded signi cantly after EOD-FR treatment.

Changes in auxin level after EOD-FR treatment
To further explore the role of auxin in EOD-FR-mediated hypocotyl elongation and cell expansion, we measured IAA in leaves and hypocotyls from the different treatments.With LC-MS, we found that the level of free IAA in hypocotyls of the EOD-FR seedlings was signi cantly higher than that of other treatments (Fig. 3).Treatment with the auxin transport inhibitor NPA eliminated the increase in free IAA induced by EOD-FR.The IAA content in leaves increased signi cantly after EOD-FR treatment, and the content of IAA in leaves of the NPA-T group was the highest.When NPA was applied alone, the IAA content of leaves and hypocotyls decreased signi cantly.The above results indicate that EOD-FR treatment can increase free IAA levels in the plant, especially in the hypocotyl, but that this effect is counteracted by NPA.This shows that the level of free IAA in the hypocotyl is an important factor in promoting cell expansion and hypocotyl elongation.

Metabolomics analyses and identi cation of differentially abundant metabolites
To investigate the source of free IAA synthesis in hypocotyls under EOD-FR, we analyzed their metabolic pro les using LC-MS/MS.A principal component analysis (PCA) model was constructed (Additional le 7) to study the relationships between metabolite levels and EOD-FR by R (Used package including vegan, stats and ggplot2).Based on PCA analysis of metabolites identi ed in the positive and negative ion modes, we found that metabolite accumulation in the hypocotyls was quite different between CK and T.
We speci cally identi ed metabolites (Additional le 1) related to auxin synthesis.Compared with the CK, EOD-FR plants accumulated more tryptophan and indole: their abundances increased 2.14-fold and 2.05fold, respectively.Tryptophan is an important precursor for auxin-IAA, and indole is a precursor for tryptophan [22].In addition, the content of 3-indole acrylic acid (IAcrA), another product of the tryptophan metabolic pathway, also increased signi cantly after EOD-FR treatment.
We compared the correlation coe cients of tryptophan, indole, and IAA metabolites measured above, and found that the correlations between them were greater than 0.9 (Additional le 8), it strongly suggesting that newly produced free IAA was synthesized by a tryptophan-dependent IAA synthesis pathway.

RNA-seq analysis
Transcriptome pro le comparisons were performed with the same experimental materials used for metabolic pro ling.The sizes of the sequenced libraries ranged from 40,704,360-54,037,758 bp (Additional le 2), and the Q30 percentage (percentage of sequences with an error rate <0.1%) was over 93% for each library.A total of 44.19 G of raw sequencing data was generated, and 32,205 expressed unigenes were identi ed.Overall, the RNA-seq data were of high quality and could be used for further analysis.
Identi cation of DEGs in hypocotyls after EOD-FR treatment DEGs analysis indicated that 1968 genes were upregulated and 833 genes were downregulated in hypocotyls after EOD-FR treatment (Additional le 3, Additional le 9 and Additional le 10).GO enrichment analysis of upregulated genes (Additional le 4 and Additional le 11) demonstrated that GO terms related to cell division and cell wall metabolism were signi cantly enriched.Terms such as cell wall biogenesis, plant-type cell wall, cell wall, cell population proliferation, xyloglucan metabolic process, xyloglucan:xyloglucosyl transferase activity, and plant-type secondary cell wall biogenesis were among the top 20 enriched terms.
The metabolic pathways associated with the DEGs were explored further by KEGG enrichment analysis.The KEGG enrichment scatterplot showed that starch and sucrose metabolism (K00500) and amino sugar and nucleotide sugar metabolism (K00520) were the two pathways with the highest degree of enrichment.Sixty-four and 44 genes were involved in these two signi cantly enriched pathways, respectively (Additional le 5 and Additional le 11).

Identi cation of key cell wall protein genes associated with cell expansion
Plant growth is inseparable from cell expansion and division, a fact that is con rmed by the morphological changes in hypocotyl cells and the transcriptome results documented here.Likewise, the expansion of plant cells is inseparable from the regulation of various cell wall modi cation proteins [23].Expansins (EXPs) and endotransglucosylase/hydrolases (XTHs) play an crucial role in the regulation of cell wall relaxation and remodeling [24,25].Therefore, we identi ed these two critical cell wall modi cation proteins from the transcriptome data and examined their expression.Twelve EXP homologs and 13 XTH homologs were identi ed based on BLAST results from the NCBI NR database (Fig. 4), and their expression was signi cantly upregulated after EOD-FR treatment.Therefore, EOD-FR appears to stimulate the activities of XTHs and EXPs, which are important for cell expansion and hypocotyl elongation.

Identi cation of auxin-related DEGs
Auxin has been shown to play key roles in the promotion of hypocotyl elongation in low R:FR environments.Therefore, we identi ed DEGs related to auxin synthesis and auxin signaling pathways using the GO, KEGG, and NR databases.
We also identi ed 14 DEGs encoded auxin-responsive proteins, including four AUX/IAAs, six SAURs, one ABP1, one GH3, and one LIR1, ten auxin transporter-like genes, including genes encoding six in ux carriers (AUX-LIKE), two e ux carriers (PIN and PIN-LIKE), two ATP-binding cassette transporters (ABCB) and one WALLS ARE THIN 1 (WAT1) transporter.Finally, we identi ed two genes that indirectly affect auxin levels; these two CBL-interacting serine/threonine protein kinase related genes were also upregulated under EOD-FR treatment.The expression levels of these genes are presented in the form of a heatmap ( Fig. 5).

Integrated analysis of the transcriptome and metabolomics
We calculated Pearson correlation coe cients of indole, tryptophan, and IAA with auxin direct-related genes and drew a corresponding heatmap (Fig. 6).Seventeen genes had correlation coe cients greater than 0.8 with tryptophan, and 10 of these genes had correlation coe cients greater than 0.9.Six genes had correlation coe cients greater than 0.8 with indole, and three had correlation coe cients greater than 0.9.Fifteen genes had correlation coe cients greater than 0.8 with IAA, and four had correlation coe cients greater than 0.9.The correlations of CmoCh19G001500 (YUCC8, a key enzymes in auxin synthesis), CmoCh06G001560 (endoplasmic reticulum transporter PIN-LIKES 5), CmoCh19G007280 (auxin-induced protein 15A), and CmoCh07G008650 (IAA-amino acid hydrolase ILR1-like 3) with the three metabolites were more than 0.8.Finally, based on the reported auxin regulatory network in Arabidopsis and other plants, we proposed a hypothetical pathway by which auxin regulates pumpkin hypocotyl elongation under EOD-FR (Fig. 7).

qRT-PCR
qRT-PCR was used to validate the RNA-seq transcriptomic data.We selected six DEGs for qRT-PCR veri cation (Additional le 12), and their relative expression measured with qRT-PCR was consistent with their FPKM values in the RNA-seq dataset (R 2 = 0.8607, P < 0.01) (Additional le 13), indicating that the RNA-seq data were reliable for further analysis.To verify the effect of auxin on gene expression, we selected eight auxin-related genes and measured their expression in the CK, T, NPA-T, and NPA-CK treatments with qRT-PCR, as shown in Fig. 8.After NPA treatment, the expression of the eight genes decreased signi cantly.All the primers designed are listed in Table S3.

Discussion
Emerging evidence indicates that auxin plays an important role in the regulation of plant elongation in a low R:FR environment.However, the speci c mechanisms by which auxin mediates hypocotyl elongation in response to EOD-FR are uncertain.Previous studies found that (1) EOD-FR treatment signi cantly increased seedling hypocotyl elongation but had no effect on stem diameter, biomass, or seedling index [1,3], and (2) that the immediate cause of hypocotyl elongation was the expansion of hypocotyl cells [18].The results of the present experiment con rm these conclusions.
We found that the polar auxin transport (PAT) inhibitor NPA completely abolished EOD-FR-induced elongation and caused signi cant decreases in stem diameter, dry weight, fresh weight, and hypocotyl IAA content.It is worth mentioning that the level of IAA in NPA-T leaves was signi cantly higher than that of other treatments.We speculate that EOD-FR treatment increased leaf IAA levels, but this IAA could not be transported to the hypocotyl by PAT and therefore accumulated in the leaves.The increase of free IAA content in hypocotyls after EOD-FR treatment therefore appears to be closely related to PAT activity.
Next, we performed a detailed investigation of the mechanisms by which auxin affects hypocotyl elongation under EOD-FR, using a combination of transcriptome pro ling, metabolomics analysis, and physiological measurements.

Auxin is essential for EOD-FR-mediated hypocotyl elongation
As early as 2008, Csukasi et al. had discovered TAA1, a key enzyme in the rst step of auxin biosynthesis under low R:FR conditions [16].Since then, increasing numbers of researchers have performed in-depth studies to demonstrate that auxin plays a vital role in low-R:FR-mediated elongation.However, EOD-FR is a special form of far-red light treatment, and few studies have addressed its relationship to auxin.Here, we have shown that auxin is essential for EOD-FR-mediated hypocotyl elongation.Several lines of evidence support this conclusion.
First, after EOD-FR treatment, the level of free IAA in hypocotyls increased signi cantly during EOD-FR treatment.However, hypocotyl elongation mediated by EOD-FR was abolished in the presence of the PAT inhibitor NPA.Second, we identi ed a large number of DEGs related to auxin biosynthesis, transport, and response in EOD-FR-treated plants (Fig. 5).Third, the content of tryptophan, a key auxin synthesis precursor, increased signi cantly after EOR-FR treatment (Additional le 1).Taken together, these results strongly suggest that auxin plays an essential role in the regulation of hypocotyl elongation in response to EOD-FR.

The pathway of auxin biosynthesis after EOD-FR treatment
There are multiple auxin biosynthetic pathways in plants, and the TAA-YUC pathway (IPyA pathway) appears to be the primary pathway that responds to low R:FR [16].TAA1 catalyzes the rst step in this pathway [26], and a family of enzymes encoded by YUCCA genes catalyze the second step [27].YUCCA appears to mediate the rate-limiting step in IAA synthesis [28,29].
RNA-seq data showed that the expression of one TAA1-related TRYPTOPHAN AMINOTRANSFERASE RELATED 2 (TAR2) gene and two YUCCA genes were upregulated under EOD-FR.Here, we found a signi cant correlation between YUCC8 expression levels and the abundance of three auxin-related metabolites and previous studies have shown that YUCC8 is closely related to the response of plants to dark environments and that its expression increases signi cantly in low R:FR environments [30].Therefore, YUCC8 may be a key gene that speci cally responds to EOD-FR.All of these results indicate that the TAA-YUC pathway is the main route for free IAA synthesis in response to EOD-FR treatment.
Increased L-tryptophan content in hypocotyls under EOD-FR also con rms the above conjecture [16].
Tryptophan is an important precursor of auxin synthesis, and its content is closely related to the level of plant auxin.It has previously been shown that IAA levels in rice calli and potato buds with excessive tryptophan synthesis increased 57-and 39-fold, respectively [31,32].
Moreover, the expression of a gene encoding N-(5'-phosphoribosyl) anthranilate isomerase 1 (trpF) in the tryptophan synthesis pathway and the abundance of the tryptophan precursor indole [22] also increased signi cantly after EOD-FR treatment.These may be important mechanisms underlying the increased tryptophan observed in plants under EOD-FR.The trpF appears to be a key gene of tryptophan metabolism induced by EOD-FR.

Auxin transport is enhanced by EOD-FR treatment
Based on changes in plant morphology and auxin levels after NPA treatment, we believe that PAT is necessary for hypocotyl elongation mediated by EOD-FR.PAT is accomplished through the concerted action of PIN-FORMED (PIN) e ux carriers, P-GLYCOPROTEIN/ATP-BINDING CASSETTE B4 (PGP/ABCB) family members, and auxin in ux AUXIN1/LIKEAUXIN1 (AUX/LAX) transporters, all are important for increasing IAA transport to speci c cell types [33].
Polarly distributed PINs have been shown to play an important role in plant elongation and growth under low R:FR [34].Here, we found that the expression of PIN4-related genes was signi cantly upregulated after EOD-FR treatment, consistent with the ndings of Takemura et al in a study of Platycodon grandi orum [18].PIN4 may therefore be a crucial e ux carrier protein for EOD-FR-mediated elongation.
In addition, a gene encoding PIN-like (PILS) protein, which is structurally similar to PIN, was also identi ed in our transcriptome data.PILS localizes to the endoplasmic reticulum and participates in the dynamic balance of auxin in that organelle [35].We found that the expression of PILS5 was signi cantly downregulated 7.7-fold in response to EOD-FR.Decreased PILS5 expression seems to suggest a cumulative decrease in IAA in the ER, causing more IAA to ow into the nucleus and in uence various biological processes [36], but further studies are needed to identify speci c underlying mechanisms.
Unlike PINs, ABCB proteins are involved in the ATP-dependent in ux and e ux of auxin and are located uniformly throughout the plasma membrane [37].In the sav4 mutant with a basic auxin transport de ciency, auxin e ux mediated by ABCB is blocked, and the plant's ability to respond to low R:FR is signi cantly impaired [38], indicating that ABCB is necessary for the low R:FR response.Here, two genes encoding proteins similar to ABC transporter B family member 19 (ABCB19) were signi cantly upregulated under EOD-FR.It has been reported that ABCB19 can affect the elongation of hypocotyls [39] and stabilize PIN proteins on the plasma membrane to coordinate the regulation of auxin e ux [40].In addition to auxin e ux carriers, AUX1-LIKE (LAX) auxin in ux carriers were also signi cantly upregulated after EOD-FR treatment.Studies on AUX/LAX in other plants have shown that hypocotyl length is signi cantly reduced in the PaLAX1 mutant compared to that of wild type [41], and there is a close relationship between PIN and LAX [42].At present, there are few studies of AUX/LAX genes under low R:FR conditions, but it is likely that they play an indispensable role in EOD-FR-mediated hypocotyl elongation.Here, we identi ed six LAX-like genes that may have a role in the EOD-FR response (Fig. 5).We also identi ed a gene related to vacuolar auxin transport, WALLS ARE THIN 1 (WAT1), whose encoded protein is involved in maintaining the dynamic balance of intracellular auxin [43].
In addition to carriers that directly participate in auxin transport, two CBL-interacting serine/threonine protein kinase (CIPK) related genes were signi cantly up-regulated after EOD-FR treatment.CIPKs can participate in auxin transport by regulating gene expression; for example, the tobacco high-expression CIPK mutant CaCIPK6T1878-90 shows increased basic auxin transport [44].
Here, we showed that many genes related to auxin transport are signi cantly upregulated after EOD-FR treatment; this may enhance the transport of free auxin to the hypocotyl, thereby increasing its IAA level.Synergy among various auxin transport carriers is probably required to maintain the auxin concentrations necessary for regulation of plant growth and hypocotyl elongation in response to EOD-FR.At present, the speci c functions of these DEGs remain to be studied in pumpkin.

EOD-FR signi cantly affects the component of auxin signal transduction in hypocotyl cells
The cellular response associated with free auxin was transduced through a speci c signal transduction pathway [45].In the nucleus, auxin regulates the expression of downstream genes by speci cally activating auxin response factors (ARFs) through the SCFTIR1/AFB-AUX/IAA pathway [46], in turn regulating the expression of downstream genes.It has been reported that the AUX/IAA, GH3, and SAUR gene families are the main downstream response genes induced by auxin in the nucleus [47], and low-R:FR-mediated elongation is closely related to their expression in Arabidopsis [48,49].
Relevant studies have shown that the stability of AUX/IAA proteins is enhanced in the PhyA mutant, weakening the low-R:FR-mediated elongation reaction [50].Furthermore, the expression of AUX22 is signi cantly lower in auxin-insensitive mutants than in wild-type plants under low R:FR conditions [51].Here, we identi ed four AUX/IAA genes (Fig. 5) , including two AUX22 genes, suggesting that AUX22 may affect hypocotyl elongation in response to EOD-FR treatment.Two AUX/IAA genes (IAA11 and IAA14) were also signi cantly up-regulated and may be important candidate genes in the EOD-FR response.
The GH3 gene family is related to the formation of auxin conjugates and regulates plant auxin homeostasis.Hypocotyls of Arabidopsis with GH3.17 de ciency accumulated more IAA and exhibited increased sensitivity to a low R:FR environment, resulting in longer hypocotyls [17].In our study, the expression of GH3.6 decreased signi cantly 9.35-fold after EOD-FR treatment.Recent research shows that overexpression of GH3.6 in the Arabidopsis mutant dwarf in light 1-D (d 1-D) produces strong developmental phenotypes such as short hypocotyls [52].Here, up-regulation of ILR1 and downregulation of GH3.6 suggest that the mechanisms underlying increased free IAA in hypocotyls under EOD-FR treatment are complex, although the regulation of IAA homeostasis clearly plays a certain role.
In recent years, SAURs have been show to play an important role in the auxin-related regulation of apoplast acidi cation [53].Speci cally, SAUR proteins inhibit the activity of phosphatases, thereby preventing plasma membrane H + ATPase dephosphorylation and activating H + ATPases, which in turn leads to apoplast acidi cation and the promotion of cell wall protein activity [54].We found that several SAUR genes were signi cantly up-regulated after EOD-FR treatment (Fig. 5).SAUR71 has been shown to regulate hypocotyl stele growth in Arabidopsis [55], and SAUR50 has been shown to promote hypocotyl elongation in the dark [56].The increased expression of SAUR genes likely contributed to apoplast acidi cation in hypocotyl cells after EOD-FR treatment, in turn affecting plant growth.At present, the functions of many SAUR proteins remain unknown, but as the largest family of auxin early response genes, SAURs are undoubtedly very important for EOD-FR-mediated hypocotyl elongation.
In addition to the typical nuclear auxin signal transduction pathway, some auxin-regulated processes are controlled by the independent TIR1/AFB-AUX/IAA pathway [57].The related molecular basis has not yet been clari ed, but auxin binding protein 1 (ABP1) is involved in some of these processes [57,58].In response to low R:FR, the expression levels of several genes regulated by auxin and shading in abp1-5 mutants were 3-5 times lower than those in the Col wild type, indicating that ABP1 is also an important factor in FR response [59].We found that the expression of auxin receptor abp19a-related genes increased signi cantly after EOD-FR treatment.Although many studies have shown that ABP1 has an important role in plant growth, more in-depth studies are needed to analyze its role in the EOD-FR response.In summary, after EOD-FR treatment, there were signi cant differences in the expression of different auxin response genes in cells, which presumably led to physiological phenomena such as cell expansion.

EOD-FR treatment enhances the activity of cell wall proteins
Hypocotyl elongation induced by EOD-FR is closely related to cell wall relaxation and activity.In this process, cell wall proteins play an important role in cell expansion [23].Research shows that XTHs act on xylan chains, relax the cell wall, accelerate cell wall remodeling, and regulate cell expansion [60].EXPs cause wallloosening by disrupting non-covalent interactions between cellulose micro brils and matrix polysaccharides [61].
Previous studies have shown that the expression levels of XTH22 in the doc1/BigArabidopsis mutant are signi cantly lower than those in the wild type under low R:FR [48], indicating that increased auxin levels in EOD-FR-treated plants may be an important reason for upregulated XTH expression.Moreover, it has been shown that XTH9 expression increases signi cantly under low R:FR in Arabidopsis [62].
At present, XTHs are thought to be the main cell wall modi ers during plant response to low R:FR, but studies on Arabidopsis show that EXPs can enhance or supplement the function of XTHs [62].Recent research shows that the expression of several EXP-related genes is upregulated after far-red light treatment in Brassica napus, further con rming the interaction between FR and EXPs [63].Here, we identi ed 13 XTH genes and 12 EXP genes (Fig. 3), including XTH22 and XTH9.All were signi cantly upregulated after EOD-FR treatment, perhaps in response to apoplast acidi cation [64] induced by SAUR.

Conclusion
In summary, we conclude that auxin plays an essential role in hypocotyl elongation mediated by EOD-FR treatment and propose the hypothetical mechanism presented in Fig. 9. EOD-FR treatment can signi cantly increase the expression of genes related to auxin synthesis, transport, and signal transduction.Our RNA-seq and metabolite pro les combined with previous studies suggest that the TAA-YUC pathway is both a low-R:FR-mediated auxin synthesis pathway and also the primary pathway by which EOD-FR promotes auxin synthesis.Our ndings provide a new perspective on the role of auxin in hypocotyl elongation induced by EOD-FR and provide abundant resources for further investigation of the molecular mechanisms by which auxin controls plant growth in response to FR light.

Plant materials and sample collection
The Cucurbita moschata variety 'Shang 2' (Certi cation number: No. 2010001 of Shaanxi Vegetable Registration, provided by Shaanxi Academy of Agricultural Sciences, YangLing, China) was used as the experimental material.Seeds were soaked in hot water at 55 ℃ for 10 min with continuous stirring.When the water temperature dropped below 30 ℃, the seeds were soaked for 6 hours.After the seeds had germinated in a constant temperature incubator at 30 ℃, seedlings with the same sprouting length were sown into a 50-hole seedling plate and grown with seedling substrate.The plants were grown under arti cial light (14 hours light at 28 ℃, 10 hours dark at 18℃).After their hypocotyl hooks had emerged from the soil, the hypocotyls had straightened, and the cotyledons were fully unfolded, seedlings with uniform growth were selected for use in the experiment.Twenty seedlings were used for each treatment, and the treatment durations and intensities are shown in Table 1.Four different treatments were set, far red light treatment(T), far red light plus inhibitor treatment(NPA-T), without far-red light treatment (CK) and inhibitor treatment(NPA-CK).The auxin inhibitor NPA (2 μM) was applied 30 min before treatment.
Hypocotyl lengths of all plants were measured daily with a steel ruler.The leaves and hypocotyls of eight plants were rapidly cut and frozen in liquid nitrogen for subsequent RNA sequencing and metabolite detection.After 6 days of treatment, six plants were randomly selected for measurement of plant height, stem diameter, and the dry and fresh weights of above-and belowground parts; three plants were used for the preparation of hypocotyl para n sections.

Auxin quanti cation
Auxin quanti cation was performed as described by Du et al [65] with minor modi cations.Plant tissues stored at −80 °C were frozen in liquid nitrogen and crushed into a powder, then extracted with 1 mL methanol/water/formic acid (15:4:1, v/v/v).The combined extracts were evaporated to dryness under a nitrogen gas stream, reconstituted in 100 μL 80% methanol (v/v), and ltered through a 0.22 μm lter for further analysis.
The ESI source ran at 550 °C and a 5500 V ion spray voltage.The curtain gas was set at 35.0 pounds per square inch (psi), and the collision gas was set to medium.Declustering potential and collision energy for individual multiple reaction monitoring (MRM) transitions was performed with further optimization.A speci c set of MRM transitions were monitored for each period based on the plant hormones eluted within this period.Three technical replicates and three biological replicates were injected into the HPLC for auxin quanti cation.To verify the accurate identi cation and quanti cation of IAA, arti cial checks of HPLC and ion results were conducted to identify the consistency of each sample.A standard curve with a regression coe cient >0.99 was used to calculate IAA levels.

Metabolomics Pro ling Analysis
Freeze-dried sample powder (0.1 g) of hypocotyl from CK and T were extracted with 50% methanol , then centrifuged and ltered at 4000 g and 4 °C.All chromatographic separations were performed using an ultra-performance liquid chromatography (UPLC) system (SCIEX, UK).An ACQUITY UPLC T3 column (100 mm × 2.1 mm, 1.8 µm, Waters, UK) was used for the reversed phase separation.A high-resolution tandem mass spectrometer TripleTOF 5600+ (SCIEX, UK) was used to detect metabolites eluted from the column.The Q-TOF was operated in both positive and negative ion modes.The mode of data collection was IDA (information dependent acquisition).To evaluate the stability of the LC-MS during the whole acquisition, a quality control sample (pooled from all samples) was run after every 10 experimental samples.The acquired MS data were pre-processed with XCMS software.LC−MS raw data les were converted into mzXML format and then processed using XCMS, CAMERA, and metaX toolbox implemented in R software.Each ion was identi ed by combining retention time (RT) and m/z removed.The MS fragment data from the mass spectrometer were matched with a secondary library of in-house metabolite standards, and metabolites with similarity >80% were extracted.An online database was used to annotate the metabolites by matching the exact molecular mass data (m/z) of the samples with those from the database.Statistical analysis was performed by t-test with multiple testing correction to obtain Q-values for individual metabolites, and VIP values were obtained using partial least squares discriminant analysis (PLS-DA).Signi cantly different metabolites were identi ed using thresholds of >2-fold difference, Q <0.05, and VIP >1.
Transcriptome sequencing CK and T were selected for RNA-SEQ analysis.Total RNA was extracted using Trizol reagent (Invitrogen, CA, USA) following the manufacturer's procedure.Then the RNA were reverse-transcribed to create 6 cDNA libraries in accordance with the protocol for the TruSeq Stranded mRNA Library Prep Kit (Illumina, San Diego, USA).And then we performed the paired-end sequencing on an Illumina Hiseq 4000 (LC-Bio Technology CO., Ltd., Hangzhou, China).Differential expression analysis was performed following previously described methods [66].In brief, the fragments per kilobase per million reads (FPKM) method was used to calculate the expression levels of genes.The differentially expressed genes were selected with log 2 (fold change) >1 or log 2 (fold change) <-1 and p value < 0.05 by the R package edgeR.

Annotation and identi cation of DEGs
We annotated the DEGs by BLASTing against three public databases [NR(Non-Redundant Protein Sequence Database ), GO(Gene Ontology), and KEGG( Kyoto Encyclopedia of Genes and Genomes)].
DEGs were then subjected to GO functional enrichment analysis and KEGG pathway analysis by R. GO terms or KEGG pathways with a Bonferroni-corrected P-value <0.05 were considered to be signi cantly enriched.

Integrated analysis of DEGs and differentially abundant metabolites
To further clarify the relationships among DEGs, differentially abundant metabolites, and hypocotyl elongation, we calculated the Pearson correlations between key metabolites and key genes and created a corresponding heatmap.
Quantitative real-time PCR (qRT-PCR) Gene-speci c primers for 14 genes were designed based on the sequences obtained from RNA-seq data (Additional le 6).At the end of the reaction, the dissociation curve was analyzed, and the speci city of the primers was evaluated.Actin primers used were the following: CmaActin_Forward:CTGCTGAGATGAACAAAAGGTC, CmaActin_Reverse: TGTGCAGTAGTACTTGGTGGTCTC. Relative gene expression levels were calculated using the 2                 PrB, which leads to a decrease in PfrB in the nucleus, thereby relieving the inhibition of PIF.PIF promotes the expression of downstream genes related to auxin synthesis.Based on the experimental data, we conclude that the tryptophan-dependent TAA-YUC pathway is likely to be the main pathway of IAA synthesis induced by EOD-FR.After synthesis, auxin is transported to the hypocotyls by way of the polar auxin transport system and promotes the transcription of downstream auxin response genes.Among them, the products of SAUR genes can promote the activity of H+ATPase, leading to apoplast acidi cation and increased activity of wall proteins such as XTHs and EXPs.This promotes cell expansion and eventually leads to signi cant hypocotyl elongation.PrB, which leads to a decrease in PfrB in the nucleus, thereby relieving the inhibition of PIF.PIF promotes the expression of downstream genes related to auxin synthesis.Based on the experimental data, we conclude that the tryptophan-dependent TAA-YUC pathway is likely to be the main pathway of IAA synthesis induced by EOD-FR.After synthesis, auxin is transported to the hypocotyls by way of the polar auxin transport system and promotes the transcription of downstream auxin response genes.Among them, the products of SAUR genes can promote the activity of H+ATPase, leading to apoplast acidi cation and increased activity of wall proteins such as XTHs and EXPs.This promotes cell expansion and eventually leads to signi cant hypocotyl elongation.

Supplementary Files
This is a list of supplementary les associated with this preprint.Click to download.

Figures
Figures

Figure 1 Change
Figure 1

Figure 1 Change
Figure 1

Figure 4 Expression
Figure 4

Figure 4 Expression
Figure 4

Figure 5 Expression
Figure 5

Figure 5 Expression
Figure 5

Figure 7 Overview
Figure 7

Figure 7 Overview
Figure 7

Figure 9 A
Figure 9