ScFT6: A Putative Candidate for Sugarcane Floral Inducer

One of the factors that can decrease sugarcane productivity is the owering, because it affects the quantity and quality of feedstock, due to sucrose consumption from the stem during inorescence emission. Photoperiodicity is the main environmental factor involved in sugarcane oral induction, which occurs by the integration of gene regulatory networks in response to environmental and endogenous stimuli. One of the genes involved in those regulatory networks is the FLOWERING LOCUS T (FT), which is considered a phloem-mobile signal that stimulates oral induction in the shoot apical meristem. This work aimed to identify and characterize homologs of the FT gene in sugarcane, as well as to determine the putative function of these genes during oral induction. From this perspective, we have conducted in silico analyses of putative FT orthologs in sugarcane, as well as the expression levels in different photoperiodic conditions in a 24-hours-day-cycle of ScFT6 in different plant tissues in contrasting cultivars in terms of owering time. Three new possible FT orthologs were found with high similarity to FT homologs in other species. Among three genes identied, we highlighted ScFT6, which has a conserved domain and amino acids at characteristic positions for the owering inducer phosphatidylethanolamine-binding protein gene family. Additionally, its expression occurs according to coincidental model, possibly being controlled by the circadian clock. Cultivars with distinct owering time behavior display variable expression for the ScFT6 gene, suggesting a possible genotypic relationship for its expression. Therefore, sugarcane has at least one putative orthologous gene in relation to FT that promotes oral induction. The oral transition is a critical step in the life cycle of sugarcane. The scarcity of information on oral genetic control makes it to use molecular tools to minimize yield damage. Here, we rst showed a homolog with high identity to the FT gene compatible with owering promote function due to association with its amino acid positions. Two other homologs, compatible with the TFL gene, were identied with putative owering repressor function in sugarcane, however, with lower identity values. regulators novel oral classied as either oral repressors or inducers, on the specic amino acids present in their sequences ScFT6 a residue at position 85, is characteristic of inductor FT, whereas ScTFL3 ScTFL4 domain is of TFL. These amino acid position differences imply ScFT6 as a oral inducer and ScTFL3 and ScTFL4 as oral repressors.


Introduction
Sugarcane (Saccharum spp.) is one of the most relevant world crops due to its high production potential in both food and biofuels industrial sectors. Many countries have stood out as world sugarcane producers, with ethanol produced from it being the main source of renewable energy. The drive for this type of energy will grow stronger over time, as it is an e cient substitute for fossil fuels (1; 2; 3), However, a drawback of this commodity is the owering process, which affects the quantity and quality of raw material produced, due to the consumption of sucrose present in the stem during in orescence formation (4; 5).
Sugarcane owering results in the pith formation into the stalks (the dehydration of the sugarcane stalk internal tissues), which leads to a decrease of internodes density and sucrose contents. Consequently, a high ber relative percentage and reducing sugars become imminent. This process results in a low-quality broth in the power plant due to the high reducing sugar content and suspension bers, which reduces the value of raw material (6). Then, there is a relationship between genotypes with higher in orescence formation and higher pith process (4), generating considerable economic losses to the production chain.
Photoperiod is the main environmental factor affecting sugarcane oral induction. The ideal photoperiod varies from 12 to 12.5 hours when the day is shortening (Shortening-Day). This period must coincide with the time when the plants have already completed the vegetative period or juvenile phase (7). Natural photoperiodic characteristics have been simulated in the development of research for arti cial oral induction (8; 9). These researches have contributed to genetic improvement programs. However, the genetic network for owering control of this species is still incipient, considering both different genotypic responses and its complex genome, principally by the perception of photoperiodic stimulus genes.
Usually, plants perceive photoperiodic stimulus by the leaves, which send signals that regulate the owering transition process in the shoot apex. In Arabidopsis thaliana, a long-day plant, the perception between long and short days is made by the interaction between light signaling, circadian rhythms and expression of CONSTANS (CO) (10). The CO protein is a transcriptional regulator capable of binding to DNA sequences promoting owering under long-day conditions in A. thaliana. Under these conditions, CO stimulates the expression of another transcription factor, the FLOWERING LOCUS T (FT), a phloem-mobile signal that stimulates oral induction in shoot apical meristem (SAM) (11; 12).
FT is a member of the phosphatidylethanolamine binding protein (PEBP) gene family. Furthermore, TERMINAL FLOWER1 (TFL1), another gene in this family, is also involved in owering and 60% of its amino acid sequence is identical to FT. Whereas FT acts in oral induction, TFL1 inhibits the action of LEAFY (LFY) and APETALA1 (AP1), which are genes that participate in the oral induction, preserving SAM in its vegetative stage (13; 14; 15). As well as FT, TFL1 function is also conserved among several species (15).
Long-day plants, as well as short-day plants, have their owering process regulated by genes from the PEBP family. Rice (Oryza sativa L.), for instance, has a pair of FT orthologs known as Heading-date3 (Hd3a) and RICE FLOWERING LOCUS T1 (RFT1), which are transcribed and translated in leaves. Their proteins are translocated through phloem to SAM, where they regulate its differentiation into oral meristem (16; 17). Hd3a is responsible for promoting owering on short days, while RFT1 is expressed under long-day conditions and, although not preventing owering, causes its delay (18; 19; 20).
In sugarcane, two possible FT and TFL homologous genes, which belong to the PEPB protein family, have already been identi ed (21). ScTFL1, which belongs to the TFL1-like subfamily, causes owering delay and increases vegetative phase length when overexpressed in A. thaliana. ScFT1 has a possible involvement in meristematic activity controlling owering time and formation of reproductive organs (21).
The comparison of partial sequences of several genes similar to ScFT -from the EST sugarcane database (SUCEST) (22) -indicated that other candidates may be involved in oral induction (21). The identi cation and characterization of those genes by making use of heterologous expression analyses have provided important contributions to understanding owering regulation. Nevertheless, elucidating the mechanisms that control owering in sugarcane is still a missing information, despite its economic importance.
Although the discovery of genes linked to oral repression has provided a breakthrough in the face of one of the biggest productivity problems of sugarcane, until the present work, there was no evidence of genes promoting owering, bringing new perspectives. Here we identify and characterize one novel sugarcane PEBP member and show that it is under photoperiod and circadian clock control, determining it like a putative oral inducer gene in sugarcane. In the search to elucidate the sugarcane oral induction mechanism through a photoperiodic pathway, we show the possible functional relation among putative sequences related to FT, from bioinformatics, gene expression, and morphological characterization from two genotypes.

Material And Methods
In silico analysis Sugarcane putative orthologs FT were identi ed by computational analyses in available RNA-seq libraries with nearly 48 million paired-end reads deposited at the NCBI SRA (National Center for Biotechnology Information -Sequence Read Archive) under accession numbers SRR1979656 through SRR1979669 and SRR1974519 (23). Adapter removal and quality control were performed using a Trimmomatic software version 0.36 (24), resulting in approximately 43 million paired-end reads (91%) with high quality. Transcript assembly was performed using the Trinity assembler version 2.3.2 (25). By the end of the process, 80647 possible transcripts with an N50 value of 1392 and an average length of 1065 bases were produced.
Identi cation of potential coding transcripts as well as their respective polypeptide chains was performed using the TransDecoder version 3.0.1 program (26). After protein prediction, FT sequences characterized in Arabidopsis thaliana and Setaria viridis were aligned against all predicted protein sequences using the BLAST tool version 2.4.0 (27). A new prediction step was performed from BLAST results transcripts using the ORFinder tool (http://www.ncbi.nlm.nih.gov) and GeneScan (28).
Multiple alignment was performed between the possible predicted sugarcane FT (29) and homologs of other related species using ClustalW (30). After the alignment, a phylogenetic tree was inferred using the MEGA 5.2.2 program (31) following the neighbor-joining comparison model (32) and using the Jones-Taylor-Thornton (JTT) distance matrix. The tree node reliability test was accessed via 2000 randomizations.
The protein sequences obtained were analyzed by the Conserved Domains Database (CDD) tool (33) to identify, characterize and compare them with protein domains present in FT sequences from species already characterized.
From the sequences obtained with in silico analyses, the ampli cation primers for the ScFT6 were designed using the online tool OligoPerfect (Available at http://tools.thermo sher.com/content.cfm? pageid=9716). The designed primers quality was veri ed using Oligo Analyzer3.1 software (34) at the IDT (Integrated DNA Technologies) company website. Two primer pairs were designed: (Table 1) one for PCR ampli cation (Polymerase Chain Reaction) and the other for real-time quantitative PCR (RT-qPCR) gene expression analyses.

Plant material and growth conditions
The sugarcane seedlings from the AgMusa System™ were kindly provided by BASF. We used three cultivars: RB85-5156 and RB 85-5453 both with frequent owering; and CTC 9003 with rare owering. The seedlings were kept at greenhouse with shade netting of 50% spacing until transplanting. These seedlings were used for four experiments, including gene expression analysis in different tissues (1), in a 24-hour-cycle (2), in different photoperiods (3), and contrasting cultivars (4). Therefore, the experiments were conducted using pots in growth chambers for the arti cial photoperiodic treatment impositions.
Moreover, the greenhouse (with shade netting of 50% spacing) was used for the acclimatization, and the experimental area (outdoor) was used for the contrasting cultivars evaluation.
The seedlings were transplanted to pots (40 L for the experiment under contrasting photoperiods and 25 L for the experiment over 24-hours) with equal amounts of clay soil and sand (3:2), following fertilizations being performed in a 30-day-interval throughout the experimental periods, watered every two days to keep at the eld capacity. Two shoots were kept in each pot and any other tillers were eliminated with pruning shears once a week during the experimental period. Fully expanded leaf +3 (35), immature leaf, stalk (from the internode surrounded by +3 leaf) and apical meristem were used for the gene expression analysis.

Qualitative gene expression analysis in different sugarcane tissues
Sugarcane seedlings (RB85-5156 cultivar) were kept in full sun (outdoor) and evaluated at six months of age. Samples from leaf (+3), immature leaves, stalk and apical meristem of sugarcane shoots were collected for the gene expression analysis by the semi-quantitative PCR.
Quantitative gene expression analysis over time (24-hour) in sugarcane leaves Sugarcane seedlings (RB85-5156 cultivar), after transplanted to 25 L pots were grown by 78 days at greenhouse (with shade netting of 50% spacing). For the treatment imposition, they were transferred to the growth chambers running under short-day conditions (11 hours light, 13 hours dark) with a light ux of 400 µmol m -2 s -1 using metal vapor lamps (HQI 400W E40TUB, OSRAM), during 30 days. Plants were also kept watered every two days keeping the eld capacity. After 30 days of treatment, fully expanded leaves (+3) were collected at 06, 12, 18 and 24 hours for gene expression analysis by RT-qPCR. The experiment was a completely randomized design (CRD) with four treatments (photoperiodic sample times), using three biological repetitions (three plants by the chamber) with two technical replicates.

Quantitative gene expression analysis in Shortening and Lengthening-Day conditions
The seedlings (RB 85-5156), after transplantation to 40 L pots, were grown by three months at greenhouse (with shade netting of 50% spacing), then they were transferred to growth chambers and kept in 6.2 m 2 controlled light (400 µmol m -2 s -1 ) with three metal vapor lamps (HQI 400W E40TUB ® , OSRAM) in neutral day conditions (12 hours light and 12 hours dark) for acclimatization. After twenty days of acclimatization, two treatments were imposed to the plants: one group of plants was subjected to a Shortening-Day condition (with one-minute light reduction a day) and the other to a Lengthening-Day

Molecular analysis
After sampling, the tissues were ground with liquid nitrogen and stored in Deep Freezer. RNA extraction was performed following the Qiazol protocol (QIAzol ® Lysis Reagent, Qiagen) adapted for sugarcane tissues. After RNA extraction the samples were treated with DNA-free kit (Ambion ® , Thermo Fisher Scienti c) to eliminate residual DNA. The integrity of the samples was accessed by 1.2% agarose gel electrophoresis, stained with uorescent nucleic acid dye (GelRed ® Nucleic Acid Gel, Thermo Fisher Scienti c), and visualized with a photo-imager (UVITEC Cambridge). Afterwards, the samples were quanti ed in a spectrophotometer (Biochrom NanoVue ® Plus) at A260, A230 and A280 nm to determine the quantity and quality of the extracted material. Then, the samples were treated with DNA-free Ambion kit to eliminate any residual DNA from the samples. Samples with high integrity and purity were used for cDNA synthesis with High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems™).
An assay was conducted to obtain the standard curve for the e ciency and the primer dilution. Expression of ScFT6 was characterized in different tissues by semi-quantitative PCR making use of the primers described in Table 1 and the cDNA obtained from each sampled tissue, being visualized through electrophoresis and by photo-imager.
The RT-qPCR was conducted in a Rotor-Gene Q Real-Time PCR (Qiagen) with the SYBR Green detection system. The cDNA volume used for each reaction was determined after quantifying primer e ciency, by using 10 μL of reaction volume per sample with MasterMix SYBR Green ® (Qiagen).
ScFT6 gene expression data were obtained from Rotor-Gene Series Pure Detection Software (Version 2.0.2), and relative expression was determined according to Pfa (36). Relative expression was performed by the comparative ΔΔCT (Cycle Threshold) method obtained from the reference genes GAPDH and eEF-1α (37).

Statistical analysis of data
The experiments were conducted under split-plot system on time with two cultivars contrasting to the owering. Data obtained from the biometric measurements were submitted to analysis of variance by the F test (p≤0.05) and, in case of the signi cance of the analysis of variance, the means were compared by the Tukey test (p≤0.05) or submitted to regression analysis (p≤0.05) when the interaction between cultivars and evaluation times occurred. Gene expression data was analyzed using the Rotor-Gene Series Pure Detection Software, and relative expression was calculated according to comparative ΔΔCT (36). The raw data of gene expression were modeled using an LMM (Linear Mixed Model) to estimate the level of signi cance of the interaction between treatments (38).

Putative FT ortholog genes in sugarcane
The alignment of the already characterized FT sequences of A. thaliana and Setaria viridis against the predicted proteins assembled from the public RNA-seq libraries resulted in 51 possible homologous sequences. The sugarcane sequences with identity above 50% and query cover above 70% were selected. Previous studies have already identi ed a TFL1 ortholog (ScTFL1) and ve FT-like genes (ScFT1, ScFT2, ScFT3, ScFT4 and ScFT5) in sugarcane (21). Here, three sequences compatible with these characteristics were identi ed, being named ScFT6, ScTFL3 and ScTFL4 with 174, 173 and 197 amino acids, respectively. Both mRNA and amino acid sequences were indexed with the following access codes: MN458470.1; MN458471.1; MN458472.1; respectively, at NCBI database. These three sequences were aligned with homologous sequences of FT gene in A. thaliana and different species related (Table 2).
Three possible homologs of FT and TFL genes identi ed in sugarcane showed high identity values in comparison with FT genes from other species, highlighting the ScFT6 gene that showed the highest values. The protein sequence analysis identi ed PEBP conserved domain presence (Figure 1). The presence of the PEBP domain indicates the involvement of these genes as regulators of signaling complexes in growth and development (39).
Multiple alignment analysis allowed to explore the relationships between the possible predicted sugarcane FT, putative FT previously identi ed (29) and homologs of other related species by ClustalW. We veri ed the structure and the putative functions of the obtained sequences ( Figure 2). Regarding the possible FT genes identi ed in sugarcane, ScFT6 has a tyrosine residue (Y) at position 85 that characterizes the owering-inducing FT genes. The other two (ScTFL3 and ScTFL4) have a histidine domain (H) that is characteristic of TFL-like genes considered to be oral repressors (12; 13). For sugarcane FTs, conserved motifs were observed between ScFT6 relative to the previously identi ed ScFT3 and ScFT4 genes. In contrast, amino acid differences for ScFT1 and ScFT2 genes corroborate with possible antagonistic functions.
Phylogenetic analyses allowed the division of the putative FTs in sugarcane into four clades (Figure 3). The rst clade represented by AtFT and SvFT, including ScFT5 (21) and ScFT6. The second clade, composed by Hd3a homologs, grouped genes characterized in other monocotyledons with ScFT3 and ScFT4 previously investigated (21). The third clade is represented by ZCN8, including ScFT1 and ScFT2 (21). Finally, the fourth clade is divided into two subclades, one represented by AtMFT, with no sugarcane member, and the second subclade, where ScTFL3 and ScTFL4 were clustered, was represented by AtTFL and contained owering repressor members. ScFT6 shares a greater identity with FT from A. thaliana, an LD plant, and farther from Hd3A from O. sativa, an SD plant. Both ScFT6 and AtFT share a monophyletic group, in contrast to the TFL sequences identi ed in sugarcane, as well as to the TFL orthologous of Sorghum bicolor, Hordeum vulgare, include A. thaliana.
PCR analysis of RNA isolated from different sugarcane tissue samples revealed that the ScFT6 is transcribed in mature leaves and SAM (Figure 4), as well as its homologs in A. thaliana, where it can induce owering by its expression in leaf phloem cells or SAM (40; 41). The same pattern is observed in rice, where Hd3a (FT homolog) is highly expressed in inductive photoperiodism in leaf tissues and SAM (42; 16). A more detailed analysis of the temporal pattern of expression showed that the gene is expressed according to the day-night cycle and photoperiod.

ScFT6 gene expression over time (24-hour) in sugarcane leaves
The expression of ScFT6 in sugarcane, a putative FT homolog, follows a similar pattern of relative expression along the day ( Figure 5). The results expressed in estimates of the fold-change showed that ScFT6 is highly expressed throughout the day and has its expression signi cantly reduced during the night. The results showed that the expression increased from 12:00 p.m. to 6:00 p.m., decreasing until midnight. Differences between midnight and 6:00 a.m. were not detected (Supplementary Figure 2). With this expression pattern, we can infer that the ScFT6 gene has its expression controlled by circadian rhythms, as well as its owering-inducing counterparts in other species (15).

ScFT6 gene expression in Shortening-Day and Lengthening-Day conditions
Relative expression of ScFT6 varied as a function of day-length (Figure 6). The results of the estimates of fold-change showed a signi cant increase in expression of ScFT6 under the shortening-day condition (Supplementary Figure 3). The data evidenced that ScFT6 was about 31 times more expressed 30 days after the imposition of the photoperiod reduction. These results support the hypothesis of its function as a owering inducer corresponding to inductive photoperiod sugarcane oral. Whereas there is strong functional evidence as to the level of identity between the ScFT6 and their counterparts, similar responses are attributed to gene expression in photoperiodic inductive conditions (16; 43; 44).
The owering process in sugarcane occurs by induction (9; 29) and subsequent reallocation of sugar reserves present in the stalk to the oral organs (7). The enhanced vegetative development is characterized by the larger average stalk diameter and average stalk height that predominated in the cultivar RB 85-5453 compared with CTC 9003 (Figure 7). These factors are possible contributions that can lead to different frequencies owering in these cultivars. Moreover, the cultivar RB 85-5453 displayed lower senescence and increased leaf area at the beginning of the cultivation period ( Figure 8). This result can indicate discrepancies in favorable characteristics to support the transition of reproductive development. Taken together, the results suggest differences in sucrose enhanced production and storage, ensuring energy source for oral organ development.
The relative expression levels of ScFT6 evidenced differences between RB 85-5453 and CTC 9003 cultivars (Figure 9). These results showed the effect of shorter natural photoperiods on different cultivars. In this case, the recorded photoperiod varied from 11:50 to 10:35 h of light, below the supposed critical photoperiod for sugarcane. Although the cultivar RB 85-5453 (the most responsive to the inductive photoperiod of owering) showed higher levels of ScFT6 expression, the fold-change estimates, based in LMM, did not indicate signi cant differences between cultivars along this time (Supplementary Figure 4). The expression levels of ScFT6 accumulated throughout the day, having baseline levels during the early morning. More pronounced differences between the two contrasting cultivars may be more evident at peak times of expression, such as in the afternoon hours. Thus, the collection period in the early morning may have contributed to these results.

Discussion
The ScFT6 gene is a putative sugarcane oral inducer The oral transition is a critical step in the life cycle of sugarcane. The scarcity of information on oral genetic control makes it di cult to use molecular tools to minimize yield damage. Here, we rst showed a homolog with high identity to the FT gene compatible with owering promote function due to association with its amino acid positions. Two other homologs, compatible with the TFL gene, were identi ed with putative owering repressor function in sugarcane, however, with lower identity values.
ScFT6 had the conserved PEBP domain, which is present in genes that act, in overall terms, as regulators of signaling complexes in growth and development (39), mainly as integrators in genetic regulatory networks, controlling the transition from vegetative to reproductive phase (11; 45; 46). These genes may represent novel regulators present in sugarcane, highlighting the ScFT6 as a rst oral promote found in such a crop. PEBP domain genes, involved in plant transition from vegetative to reproductive phase, can be classi ed as either oral repressors or inducers, depending on the speci c amino acids present in their sequences (47). ScFT6 had a tyrosine residue (Y) at position 85, which is characteristic of inductor FT, whereas ScTFL3 and ScTFL4 had a histidine domain (H) that is characteristic of TFL. These amino acid position differences imply ScFT6 as a oral inducer and ScTFL3 and ScTFL4 as oral repressors.
In several plant species, histidine (H) is the most important amino acid to the PEBP family members that have a TFL1-like function, as well as tyrosine (Y) is key for FT-like function members (13). Despite the importance of residues in position 85 for the promotion of a full phenotypic expression, the protein has also to be complete. Loss of residues in other regions has shown alterations in both TFL and FT functions as well (12; 48). In some species, FT-like genes can repress or induce owering. In addition to position 85 in exon 2, positions 134 and 139 in exon 4 are also important to determine either the induced or repressed phenotype expressed by those genes (49; 50).
In species that have owering repressor FT homologs, most FTs inducers have a tyrosine (Y) at position 134 (based on alignment with A. thaliana FT protein) and another non-tyrosine residue, is found in repressive FTs. In sugarcane, ScFT1 is proven to be a owering repressor (21), and has a phenylalanine residue (F) at position 134. In addition to ScFT1, the ScFT2, ScFT7 and ScFT8 genes have different residues of tyrosine (Y) at position 134. Already ScFT3, ScFT4, ScFT5 ScFT6 have a tyrosine residue (Y) at position 134, characteristic of owering-inducing FT homologues.
Regarding position 139 in the amino acid sequences of FT homologs, the presence of a conserved tryptophan residue (W) in this speci c position most likely, can allow owering induction. On another standpoint, FTs with a different amino acid in that position can inhibit owering (49; 50). The presence of tryptophan (W) in sugarcane FT homologs in position 139 was observed only in ScFT3, ScFT4, ScFT5 and ScFT6, thus those protein-coding genes are probably FTs that induce owering.
The phylogenetic results suggest that ScFT6 possibly acts as a owering inducer in sugarcane and can be classi ed as FT-like due to the presence of tyrosine residue (Y) at positions 85 and 134 and tryptophan (W) at position 138 and by its clustering pattern in the phylogenetic tree. ScFT6 appears to have diverged from the other FTs in Saccharum spp., especially in relation to ScFT1, ScFT2, ScFT3 and ScFT4, with antagonistic possible functions. In the monocots group, FT genes may be a more diverse group than TFL genes, suggest that they could have acquired functional roles in delaying owering (51; 52), for example, ScFT1 (21). While the ScTFL3 and ScTFL4 genes were suggested as possible owering repressors and classi ed as TFL-like due to the presence of a histidine residue (H) at position 85 and their respective clustering pattern in the phylogenetic tree.
The circadian clock controls ScFT6 gene expression The circadian clock acts indirectly on FT expression (53). According to the coincidence model, there is a straight correlation between light and CONSTANS stability in a way that continuous sunlight exposition leads to its accumulation. CO is a transcription factor that, when presented in high number, promotes owering by inducing FT expression in long-day plants subjected to this speci c condition (54) leading ultimately to owering induction (55; 56). This same regulatory network also inhibits owering of shortday plants by repressing FT homologs when a high number of CO are present (42; 57). By the coincidence model, it is possible to determine that the FT gene had its expression controlled by the circadian clock and, thus, its expression peak overlaps with the sunlight dynamics, both in long and shortday plants (55; 58). Determination of owering time in short and long-day plants is dependent on light signaling, where the signal must coincide with sensitive phases established by endogenous circadian clock components (59; 60). FT and its homologs, including the possible owering-inducing ScFT6 gene, are controlled by such signals, and their expression is indirectly regulated by light-controlled transcription factors as well as the duration of light exposure (17; 59).
Sugarcane genes involved in the circadian clock can be compared with genes already characterized in other species and by transgenic lines. Some of these genes showed greater similarities about the patterns seen in short-day plants, grown under non-inductive photoperiods (61). Considering a 24-hour cycle, these genes have expression patterns that control the change from vegetative to reproductive growth and can be headed by ScFT6 signaling.

ScFT6 expression is modi ed by photoperiod
We showed that the ability of sugarcane mature leaf to respond to shortening-day is essential for ScFT6 expression. The effect of reducing the photoperiod on ScFT6 transcription is suggested by the lightdependent regulation promoted, together with the regulation of the circadian clock, on the stability of a probable CO homologous protein (53; 62; 63). However, this regulation establishes the response of ScFT6 during the shortening of the day duration, while CO is under control over long days.
The inductive photoperiod for sugarcane may have a higher connection with a regulation upstream of a possible Hd1 homolog, which is responsive to short-day conditions, as detected in O. sativa. In these photoperiodic conditions, the expression of the Hd3a origen is regulated positively by Hd1, suggesting that Hd3a is under the control of Hd1 (18). Thus, the mechanism linked to the actuation of the putative ScFT6 origen, under shortening-day, can be more similar to the genetic control dependent on light and the circadian clock of short-day plants.
The phenotypic result of owering in sugarcane, for the shortening of the photoperiod, can be veri ed when using arti cial photoperiodic growth chambers under controlled conditions (8). Under conditions of natural photoperiod (disregarding temperature and water and nutritional regime variations), owering induction is determined shortly after the start of the day's shortening period, with complete owering (anthesis) varying overtime throughout the year (64). This process can be certi ed by the results obtained with the interruption of the response, using the control management (inhibitors) of owering in the eld, which is recommended especially during this period. The synchronicity between the oral phenotype displayed in arti cial and natural conditions supports the classi cation of sugarcane as responsive in shortening-day condition. During the shortening of the photoperiod, the increase in ScFT6 expression indicates the possible role of this gene as origen under shortening-day control.
Several studies have shown signi cant progress on the molecular mechanisms involved in the control of plant photoperiod and the circadian clock. In sugarcane, the knowledge of these mechanisms is still incipient. In parallel research, the identi cation and characterization of genes upstream of ScFT should be the next targets of this study. In that case, we expect to establish greater precision in the molecular connections of the sugarcane owering process.
Sugarcane varieties display alternate behaviors regarding ScFT6 expression Different sugarcane cultivars exhibited distinct behavior regarding owering induction and are affected in different manners by photoperiod. There are differences regarding the number of days required for owering stimulus within the inductive photoperiod, and the required stimulus amount for the induction varies among cultivars (65). There are cultivars in which owering induction happens, but there is no complete development of the oral organs due to suboptimal photoperiodic regimes (66) and other stressful environmental conditions. In those cases, the feedstock production is compromised leading to severe economic losses.
Although owering is not desirable at a sugarcane eld, because owering potentially limits productivity, it is necessary for reproduction and the breeding of cultivars. In contrast, some desired crosses remain challenging due to unresponsive plant varieties. Therefore, it is likely that each cultivar's sensitivity to the photoperiod is a genetic characteristic involving ScFT6 expression, allowing this gene to be a research tool, as a marker, for selection and also to use cultivars with better commercial characteristics.
The ScFT6 expression levels, when evaluated throughout 24-hour-cycle, demonstrated baseline levels during the early morning. The level of transcripts detected in this period is low when comparing both cultivars in this period. The cultivar with frequent owering had higher transcriptional level, although not considered signi cant by LMM for fold-change re ned comparisons. The plants were evaluated after the supposed critical photoperiod, the transcript levels were not su cient for signi cant expression differences. In our results, the plants were analyzed during the extreme shortening of the day, after the critical photoperiod until mid-June. Under conditions of natural photoperiod, the plants would probably respond to the critical photoperiod from 12:00 pm to 12:30 p.m. We believe that in contrasting cultivars, the owering induction associated to the photoperiod may be determined under a speci c time and the beginning of the shortening period of the day.
When considering the control of the ScFT6 expression pattern, studies with genetic manipulation in sugarcane may help to obtain non-responsive cultivars to oral induction by the photoperiod. It will be necessary to check for polymorphisms of ScFT6, in a way that is silenced its function in owering without harm in other functions that are not linked directly to oral induction. In this sense, it will be possible to reduce or even eliminate the detrimental effect of owering on sugarcane productivity.
The expression patterns of ScFT6 should be assessed under different developmental stages, including in different tissues. These factors are fundamental to a better comprehension of its physiological role.
Therefore, ScFT6 gene recognition, within the owering pathway, consists of a perspective to the resolution of one of the biggest problems that impact the sugarcane culture yield during its maturation process. Discrepancies between cultivars may be linked to complete induction for the oral transition upon reaching an ScFT6 critical concentration possible limited by physiological characteristics.

Conclusions
Our results show that sugarcane has at least one possible FT-homologous gene that can induce owering in sugarcane. This homolog is characterized as ScFT6, the putative origen gene in sugarcane.     Gene expression pro le of ScFT6 in sugarcane leaves by RT-qPCR. Data was generated over 24-hours.
The expression values were obtained from three biological repetitions, and expression difference over the period of the experiment was calculated in relation to a normalizing sample (12:00 hours). Interval bars indicate the standard error. Expression values were normalized by the reference genes GAPDH and eEF1α.
* indicates a signi cant difference from the Linear Mixed Model analyses; signi cance level was determined based on the con dence intervals of the estimated fold-change values (p ≤ 0,05).

Figure 6
Gene expression pro le of ScFT6 in sugarcane leaves by RT-qPCR. The bars represent the difference in expression between two photoperiodic conditions relative to a normalizing sample (day length). The expression values in each treatment were obtained from four biological repetitions and the error bars represent their standard error. Expression values were normalized by the reference genes GAPDH and eEF1α. * indicates a signi cant difference from the Linear Mixed Model analyses; signi cance level was determined based on the con dence intervals of the estimated fold-change values (p ≤ 0,05).  The symbol (*) indicates statistical difference between means by F test at p ≤ 0.05.

Figure 9
Relative expression pro le of the ScFT6 gene in +3 leaves of two sugarcane varieties obtained by RT-qPCR. Color bars represent the difference in expression between varieties compared to a normalizing sample along the experimental period (CTC 9003 -180 days). Each genotype expression was obtained from three biological repetitions and the bars show their standard errors. GAPDH and eEF1α genes were used as reference genes. The line to second Y-axis indicates day length over the experimental period.

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