Shade signals alter the expression of circadian clock genes in newly‐formed bioenergy sorghum internodes

Abstract Stem internodes of bioenergy sorghum inbred R.07020 are longer at high plant density (shade) than at low plant density (control). Initially, the youngest newly‐formed subapical stem internodes of shade‐treated and control plants are comparable in length. However, full‐length internodes of shade‐treated plants are three times longer than the internodes of the control plants. To identify the early molecular events associated with internode elongation in response to shade, we analyzed the transcriptome of the newly‐formed internodes of shade‐treated and control plants sampled between 4 and 6 hr after the start of the light period (14 hr light/10 hr dark). Sorghum genes homologous to the Arabidopsis shade marker genes ATHB2 and PIL1 were not differentially expressed. The results indicate that shade signals promote internode elongation indirectly because sorghum internodes are not illuminated and grow while enclosed with leaf sheaths. Sorghum genes homologous to the Arabidopsis morning‐phased circadian clock genes LHY, RVE, and LNK were downregulated and evening‐phased genes such as TOC1, PRR5, and GI were upregulated in young internodes in response to shade. We hypothesize that a change in the function or patterns of expression of the circadian clock genes is the earliest molecular event associated with internode elongation in response to shade in bioenergy sorghum. Increased expression of CycD1, which promotes cell division, and decreased expression of cell wall‐loosening and MBF1‐like genes, which promote cell expansion, suggest that shade signals promote internode elongation in bioenergy sorghum in part through increasing cell number by delaying transition from cell division to cell expansion.

. The large biomass and sugar yield of these species is in part due to their utilization of C4 photosynthesis (Mullet, 2017).
Because of its genetic diversity, adaptation to diverse climates, and small sequenced genome, sorghum is a model plant for studying stem growth, biomass accumulation, and sugar production in C4 crops (Mullet et al., 2014;Slewinski, 2012).
The stems of grasses are formed as a series of internodes generated by the activities of the lower section of the shoot apical meristem known as the rib zone (McKim, 2019;Serrano-Mislata & Sablowski, 2018). Internodes in bioenergy sorghum develop during the vegetative stage and further elongate when grown at high plant density where plants shade each other (Kebrom, McKinley, & Mullet, 2017). Shade signals also inhibit shoot branching and promote early flowering (Smith & Whitelam, 1997). These responses are collectively known as the shade avoidance syndrome (SAS) (Casal, 2013;Smith & Whitelam, 1997). The elongation of the stem of a plant in response to shade by its neighbors elevates the leaves out of shade to get sunlight for photosynthesis. Understanding how shade signals modulate stem elongation in bioenergy sorghum will be useful to increase the biomass and sugar yield of crops through agronomy, breeding, and genetic manipulation of the size of internodes.
Plants use photoreceptors to monitor their light environment and proximity to potential competitors for light. Plant photoreceptors that detect neighbor proximity and shade signals include the red (R) and far-red (FR) light absorbing phytochromes, blue light absorbing cryptochromes, and UV-B absorbing UVR8 (Ballare & Pierik, 2017).
Research on shade signaling has focused on phytochromes because of the ability of this small family of photoreceptors, encoded by up to five genes (PhyA to PhyE) in Arabidopsis and three genes in the grasses (PhyA, PhyB, and PhyC), to detect neighbor proximity and shade signals before a plant is completely shaded by its neighbors (Ballare & Pierik, 2017;Ballare, Scopel, & Sanchez, 1990;Mathews & Sharrock, 1996). Besides their role in shade signaling, phytochromes regulate many other aspects of plant growth and development, including seed germination and flowering time Franklin & Whitelam, 2004). Each phytochrome may play a major role at specific stages during plant growth and development. The shoot elongation response to shade signals is mediated mainly through the activities of phytochrome B (PhyB) (Franklin & Quail, 2010;Martinez-Garcia et al., 2010), which is localized in the cytoplasm in a R light absorbing inactive Pr form. Once it is activated by R light, PhyB is converted into a FR absorbing Pfr form, moves into the nucleus, and regulates the expression and activities of genes to modulate plant growth and development.
Plants absorb R light for photosynthesis and transmit or reflect FR light. At high plant density, the absorption of R and reflection of FR reduces the R-to-FR ratio. At low R:FR, PhyB is in the inactive R light absorbing Pr form, which signals plants to anticipate shading by their neighbors and thus initiate the shade avoidance developmental program including shoot elongation. When leaves are exposed to full sunlight, PhyB is in its active FR light absorbing Pfr form and shoot elongation is suppressed. Therefore, the proportion of Pr and Pfr form of PhyB in plants is proportional to the level of R and FR illumination in their microenvironment, and the extent of shoot elongation (Casal, 2013). In PhyB-deficient mutant plants, the shoot elongation developmental program is activated in all environments. The molecular mechanisms promoting shoot elongation in response to low R:FR lights have been investigated in more detail in Arabidopsis seedlings (reviewed in Ballare & Pierik, 2017;Fiorucci & Fankhauser, 2017;Wang & Wang, 2015).
Attenuation of the quality or intensity of light might change the timing of expression or activities of the core clock genes and promote or inhibit hypocotyl or shoot elongation (Wenden et al., 2011).
Although research on shoot elongation in response to shade has focused mainly on Arabidopsis seedlings, shade signals in crops growing at high plant density in the field arise in adult plants as neighbor proximity increases with higher planting density and enlargement of canopy size during development. Therefore, to identify the earliest molecular mechanisms regulating the response to shade signals in adult plants, we investigated the shoot elongation response of bioenergy sorghum inbred line R.07020 to high plant density treatments beginning at 60 days after planting. Our study identified differential regulation of core clock and clock-associated genes in newly-formed subapical internodes indicating that the circadian clock plays a role in shade-induced internode elongation in adult plants. The elongating stem tissues in sorghum are enclosed by several layers of leaves and sheaths and are not exposed to direct light. Therefore, we hypothesize that shade signals indirectly regulate the expression of clock genes and promote internode elongation in sorghum. The results highlight the need for more research on the role of the circadian clock in shade signaling, stem growth, and biomass accumulation in adult plants.

| Response of bioenergy sorghum inbred R.07020 plants to high plant density (shade)
Stem internodes of the bioenergy sorghum inbred line R.07020 are formed during the vegetative phase and elongate in response to high plant density or shade signals (Kebrom et al., 2017). As shown in Figure 1, internodes elongated when the plants were grown at high density in the field. When plant density was reduced by trimming some of the shoots to ground level leaving a solitary shoot, internode elongation was suppressed. Regrowth of new shoots from the stubble increased plant density and promoted elongation of newly-formed internodes in the solitary shoot. A series of short internodes between elongated internodes of the stem of the solitary F I G U R E 1 Internode elongation in response to high plant density (shade) in bioenergy sorghum genotype R.07020. (a) R.07020 plants were grown in the field at high density. During early vegetative stage, the surrounding shoots of some of the plants were trimmed to ground level to reduce plant density (low density). Subsequently the height of plants at low density was reduced due to inhibition of internode elongation. (b) The plants at low density eventually elongated as the plant density increased due to regrowth of shoots from the stubble and their plant height was similar to the plants at high density. A series of short internodes between elongated internodes (inset, b) in the stem of plants from low density indicates inhibition of internode growth in R.07020 at low density shoot indicates that the final length of each internode is determined by prevailing growing conditions during its developmental window.
The response of R.07020 plants to high plant density (shade) was investigated in detail by growing plants in pots in growth chambers in a 3.5 m 2 growth area. The plants were grown in two growth chambers at low density with ample space between plants to avoid mutual shading. Tillers were also removed as they appeared in order to reduce shading. At 60 days after planting ( ternode below the shoot apex in shade-treated plants was slightly longer but not significantly different from the corresponding internode in the control. This could be due to the advanced age of the internode prior to shade treatment, and thus it was less responsive to growth-promoting factors. There was no consistent response in the growth of leaf blades in response to shading (Figure 3c). The sheaths of the five fully expanded upper leaves in the shade-treated plants were significantly longer than the sheaths in the control (Figure 3d).
Since the length of the youngest newly-formed subapical internode in the shade-treated plants was comparable to the corresponding internode in the control plants (Figure 3b), the identification of differentially expressed genes (DEGs) and associated developmental changes that distinguish these internodes could potentially identify key early regulators promoting internode elongation in response to shade signals. Therefore, the microscopic and RNA-seq transcriptome studies focused on the youngest newly-formed subapical internodes of shade-treated and control plants. Mitotic cells were present in both internode tissues. Therefore, although the internode in the shade-treated plant at maturity will be at least three times longer than the internode in the control, shade signal had little effect on the developmental status of the internode tissues at their early stages of development.

F I G U R E 2
Shoot and stem growth of of bioenergy sorghum genotype R.07020 grown at low plant density (control) and high plant density (shade). Shade treatment (crowding potted plants) was started at 60 days after planting and the plants were photographed after two weeks. Dry lower leaves were removed from the plants before photographing

| Transcriptome changes in newly-formed sorghum internodes in response to shade
Internodes in sorghum and most other grass species develop while enclosed by several layers of leaves and sheath, and as a consequence, internodes are not directly illuminated. Therefore, it is possible that internodes in these species respond to shade signals perceived by leaves that are exposed to direct light that is modified when plants are shaded.  (Table S1). About 90.1% and 91.3% of the reads from the shade-treated and control plants, respectively, F I G U R E 3 Response of bioenergy sorghum genotype R.07020 to shade. R.07020 plants were grown in growth chambers in pots at low density until 60 days after planting (60 DAP). A set of plants were crowded for the next ten days by bringing the pots closer to simulate a high plant density growing conditions (shade). At sampling (70 DAP

F I G U R E 4
Longitudinal and cross-section microscopic images of the youngest newly-formed sub-apical internodes of bioenergy sorghum genotype R.07020 grown at low plant density (control) or high plant density (shade). The internode samples were stained with Safranin and alcian blue were aligned to the sorghum V3 genome (DOE-JGI, http://phyto zome. jgi.doe.gov/).

Differentially expressed genes in the newly-formed subapical in-
ternode of shade-treated plants were identified using the following criteria: at least two-fold higher or lower than the expression level in the corresponding control internode, false discovery rate (FDR) <0.01, and average RPKM ≥2 either in the control or the shadetreated or both internodes. Prior to analyzing the DEGs between shade-treated and control internodes, to validate our RNA-seq data, we compared the transcriptome of the newly-formed subapical in-  (Table S3). A total of 129 genes were upregulated and 200 genes were downregulated in response to shade (Table S4). A higher number of these DEGs function in cell wall metabolism, stress response, regulation of transcription, protein synthesis, plant development, and transport (Table S4). First, we evaluated the expression of sorghum homologs of the Arabidopsis PHYTOCHROME RAPIDLY REGULATED (PAR) genes such as ATHB2 and PIL1 that are rapidly induced by shade (Martinez-Garcia et al., 2010;Roig-Villanova et al., 2006). None of the sorghum genes similar to the Arabidopsis PAR genes were differentially regulated in the subapical internodes in response to shade.
Shoot elongation in response to shade is also associated with an increase in the expression of cell wall loosening genes which promote cell elongation (Sasidharan, Keuskamp, Kooke, Voesenek, & Pierik, 2014). Of the 15 differentially expressed cell wall-related transcripts, 14 were downregulated in the first subapical internodes of shade-treated plants (Tables S3 and S4). Most of the downregulated genes encode pectin lyase-like superfamily proteins and expansins. These genes function in cell wall loosening (Cosgrove, 2016).
In Arabidopsis, genes encoding cell wall loosening enzymes such as pectinesterases and pectin-lyases were upregulated at later stages in response to shade (Devlin, Yanovsky, & Kay, 2003).
To further identify molecular pathways associated with internode elongation in response to shade, we performed gene ontology (GO) enrichment analysis of the differentially expressed sorghum transcripts using the corresponding Arabidopsis gene IDs (Table S3).
As shown in Table 1, GO terms for circadian rhythm (GO:0007623), regulation of circadian rhythm (GO:0042752), and GO terms for various stress responses such as response to hydrogen peroxide (GO:0042542), response to salt stress (GO:0009651), protein folding (GO:0006457), and response to water deprivation (GO:0009414) were overrepresented. It is possible that the growth response of the sorghum internodes in response to shade is linked to differential expression of the clock genes. Therefore, we looked at the patterns of expression and function of the differentially expressed clock-related genes in more detail.
TA B L E 1 Gene ontology (GO) enrichment analysis of genes differentially expressed in response to shade in the youngest newly-formed sub-apical internode of bioenergy sorghum R.07020. Go terms with FDR values < 0.01 were selected for further analysis

| Differential expression of circadian clock genes in sorghum internodes in response to shade
Of the 329 differentially expressed transcripts in the newly-formed subapical internode of bioenergy sorghum in response to shade annotated by MapMan, at least 24 are core circadian clock, clockassociated, or clock-regulated transcripts (Figures 5-7). and functionally similar to RVE4 and RVE8, and RVE2 is similar to RVE1 and RVE7 (Rawat et al., 2011). The peak expression of RVE8 in Arabidopsis precedes dawn and the rve8 mutant develops a longer hypocotyl when grown in low or medium fluence rate (Rawat et al., 2011). The shoots of triple rve4 6 8 Arabidopsis seedlings and adult plants are larger than the wild-type indicating the RVE8-like genes suppress growth (Gray, Shalit-Kaneh, Chu, Hsu, & Harmer, 2017). Therefore, the downregulation of the SbRVE6 gene in the sorghum internodes in response to shade might contribute to the enhanced elongation growth of the sorghum stem internodes.
The peak circadian expression of RVE2, and its homologs RVE1 and RVE7, is before dawn (Rawat et al., 2009). RVE2 expression is induced by light and repressed by overexpression of CCA1, a homolog of LHY and RVE8 (Zhang et al., 2007). RVE2 mutants flower earlier while their hypocotyl growth is not different from wild-type (Zhang et al., 2007). However, overexpression of RVE2 promotes hypocotyl elongation (Rawat et al., 2009 (Table S3) In contrast to the downregulation of morning genes, several circadian clock and clock-associated evening-phased genes, in addition to SbTOC1, were upregulated in the sorghum internode in response to shade ( Figure 6, Table S3). These include upregulation of two sorghum genes, Sobic.002g275100.1 (SbPRR5a) and Sobic.005G044400.1 (SbPRR5b), homologous to the Arabidopsis

| Differential expression of circadian clock-associated or clock-controlled genes in sorghum internodes in response to shade
Several clock-associated or clock-controlled genes were differentially expressed in sorghum internodes in response to shading ( Figure 7, Table S3). The expression of genes associated with morning core clock genes was downregulated and the expression of genes associated with evening core clock genes was upregulated. A sorghum gene, Sobic.009G113400.1, homologous to the EID1 gene of Arabidopsis was downregulated by shade ( Figure 7a). The EID1 gene functions in phytochrome signaling and light input pathway to the core clock (Dieterle, Zhou, Schafer, Funk, & Kretsch, 2001;Muller, Zhang, Koornneef, & Jimenez-Gomez, 2018). A mutant EID1 allele in cultivated tomato selected during domestication from wild ancestors reduced the speed of the clock and enabled the cultivation of tomato in higher latitudes . A sorghum gene,

Sobic.002G116000.1, homologous to the Arabidopsis Granule Bound
Starch Synthase1 (GBSS1) gene that functions in starch biosynthesis was 6.6-fold lower in the shade-treated internode (Figure 7b). The expression of GBSS1 is regulated by CCA1 and LHY, with peak expression in the morning (Moraes et al., 2019;Ortiz-Marchena et al., 2014;Tenorio, Orea, Romero, & Merida, 2003). A sorghum gene (SbHY5, Sobic.004G085600.1) homologous to the Arabidopsis ELONGATED HYPOCOTYL 5 (HY5) was downregulated in the sorghum internode in response to shade (Figure 7c). The HY5 gene in Arabidopsis encodes a basic leucine zipper (bZIP) transcription factor that regulates the expression of many genes that function in diverse physiological and developmental processes, including photomorphogenesis, circadian clock, light and hormone signaling, and cell elongation (Gangappa & Botto, 2016). The expression of HY5 is regulated by light perceived by the plant photoreceptors. HY5 functions as a signaling molecule that transduces the R:FR light status to roots van Gelderen et al., 2018). In addition, HY5 transduces red and blue light signals to the core clock (Hajdu et al., 2018).
About 30 differentially expressed transcripts were assigned to the stress bin (Bin 30) of the MapMan (Table S4). Six of these genes were upregulated and 24 downregulated in the first subapical stem internode of bioenergy sorghum in response to shade (Table S4).
In summary, RNA-seq transcriptome analysis revealed differential expression of circadian clock genes in the newly-formed subapical sorghum internodes in response to shade. Interestingly, shading downregulated the expression of morning genes and upregulated the expression of evening genes around 4 hr after the start of the 14-hr-long light period.

| D ISCUSS I ON
Bioenergy sorghum plants produce a large amount of biomass, which is mainly accumulated in the stem internodes (Olson et al., 2012).
Understanding the physiological and molecular mechanisms controlling stem growth in bioenergy sorghum will help to modify stem growth and biomass accumulation of crops as desired. Therefore, we investigated the growth of stem internodes in the bioenergy sorghum inbred R.07020. Our study revealed that internode elongation in R.07020 during the vegetative phase is a response to high plant density, a typical growth response of plants anticipating shading by their neighbors known as the shade avoidance response (Casal, 2013;Martinez-Garcia et al., 2010;Smith & Whitelam, 1997). It appears that this phenomenon remained unnoticed in the bioenergy sorghum R.07020 because like any other agronomic crop it is grown in the field at high plant density. However, by changing planting density in the field and growth chambers, we discovered that internode elongation in R.07020 plants is a response to mutual shading at high planting density, and stem elongation is suppressed when the plants are grown at low planting density in the absence of shade signals from neighbor plants.
Internodes in sorghum are initiated sequentially from the rib zone of the shoot apical meristem and transition through different developmental stages until they become fully elongated and stop growing. As in maize, the first four subapical internodes in sorghum plants are at different developmental stages, from the youngest newly-formed first subapical internode to the more mature and longer fourth subapical internode (Kebrom et al., 2017;Morrison, Kessler, & Burton, 1994). As shown in Figure 3, the first subapical internodes in shade-treated and control plants were comparable in length. In addition, microscopic analysis of the stem tissues indicates that the internodes were at similar developmental stages. However, at maturity, the length of the internode in the shade-treated plants will be at least three times longer than the internode in the control.
We hypothesized that key genes that initiate internode elongation in response to shade might be activated in the newly-formed subapical internode before any visible change in growth. Therefore, to identify the molecular mechanisms that regulate the early events of internode elongation in response to shade, we analyzed the transcriptome of the first newly-formed subapical internodes in shadetreated and control sorghum plants.
Molecular mechanisms that enhance shoot elongation in response to shade signals have been investigated in more detail in younger plants of Arabidopsis and other species. Several PAR genes that are markers of shade signaling such as ATHB2 and PIL1 are rapidly upregulated in response to shade in these species (Ciolfi et al., 2013;Procko et al., 2014). Unlike the detailed study in young plants, the response of adult plants to shade has not been studied in detail (Nozue et al., 2015). Furthermore, in sorghum and other monocots, the molecular mechanisms promoting stem elongation in response to shade could be different from eudicots because the stem internodes in sorghum grow while enclosed by several layers of young leaves and sheaths, reducing exposure to direct light.
Consistent with these PAR genes that are markers of perception of shade signals were not among the 353 DEGs in the sorghum internodes in response to shade. The results indicate that shade signals in bioenergy sorghum might be perceived by leaves and indirectly promote internode elongation.
Hypocotyl elongation in response to shade signals in Arabidopsis seedlings involves inactivation of PhyB, which allows PIFs to transcribe auxin biosynthesis genes (Hornitschek et al., 2012;Lorrain et al., 2008;Muller-Moule et al., 2016). An increase in auxin production promotes cell elongation, and thus hypocotyl elongation, through increasing the expression of cell wall loosening genes such as xyloglucan endotransglucosylase hydrolase (XTH) (Sasidharan et al., 2010). In the newly-formed subapical sorghum internodes, there was little change in the expression of hormone biosynthesis or signaling genes in response to shade. In addition, shading downregulated several cell wall loosening genes in sorghum internodes.
It is likely that differential expression of hormone biosynthesis and signaling genes occurs once the internode starts to elongate.
However, the downregulation of cell wall loosening genes, the upregulation of ERF109-like gene that promotes vascular cell division (Etchells et al., 2012), and the downregulation of MBF1-like gene that promotes cell expansion in Arabidopsis leaves (Tojo et al., 2009) suggest internode elongation in response to shade in the bioenergy sorghum R.07020 is associated with an increase in cell division and/ or a delay in cell expansion. Consistent with this, the expression of a sorghum gene homologous to the Arabidopsis CycD1 gene was upregulated in response to shade. Interestingly, a mutation in the Arabidopsis CycD1 gene delays the onset of cell division (Masubelele et al., 2005). And, the CycD1 in Antirrhinum majus accelerates entry into the mitotic cell cycle and rate of growth (Koroleva et al., 2004).
Organ growth in plants is due to both cell division and cell elongation (Beemster & Baskin, 1998). Therefore, internode elongation in response to shade in bioenergy sorghum could be in part due to an increase in the rate and phase of cell division leading to a higher cell number in the elongated internodes.
Differential expression of circadian clock and clock-associated genes indicates the early molecular events promoting internode elongation in response to shade in the bioenergy sorghum R.07020 involves alterations in the activities of the circadian clock.
Interestingly, the expression of sorghum genes homologous to the Arabidopsis morning genes LHY, RVE6, RVE2, and LNK was downregulated, and evening genes TOC1, PRR5, and GI was upregulated in the sorghum internode in response to shade. Also, the expression patterns of genes that function in the input and output pathways of the clock were similar to the expression patterns of the core clock genes; morning expressed genes were downregulated and evening expressed genes were upregulated (Figure 7). The Arabidopsis core clock genes CCA1/LHY and TOC1 reciprocally inhibit each other's expression (McClung, 2019). It is possible that, in the bioenergy sorghum internodes, shade signals indirectly inhibited the expression of key morning expressed genes such as LHY during the early hours of the morning, and thus the evening genes were released from repression. Also, it is possible that shade signals shifted the phase of the clock in the sorghum internodes. This is consistent with far-red induced phase shift of the clock in Arabidopsis seedlings (Wenden et al., 2011;Yanovsky, Mazzella, Whitelam, & Casal, 2001).
There are a few studies that indicate a role for the plant circadian clock in gating enhanced shoot growth in response to shade signals.
For example, Sellaro, Pacin, and Casal (2012) reported that shade signal applied in the morning did not promote shoot elongation in Arabidopsis; whereas shade signal applied in the afternoon, when the expression of LHY and CCA1 was low, promoted shoot elongation. In the lhy-cca1 mutant, shoot elongation was promoted by FR light treatment applied in the morning as well as in the afternoon.
However, unlike in Arabidopsis, the expression of morning genes such as LHY in the sorghum internodes was downregulated by shade.

Differential expression of circadian clock genes in response to shade
in sorghum internodes provide additional evidence for the role of the circadian clock in mediating shoot elongation in response to shade.
Noticeable differential expression of circadian clock genes has not been reported in transcriptome studies of SAS in Arabidopsis seedlings and in the stem of tomato and other species (Cagnola, Ploschuk, Benech-Arnold, Finlayson, & Casal, 2012;Devlin et al., 2003). However, morning genes were downregulated and evening genes were upregulated in elongating maize seedlings grown with simulated shade (Wang et al., 2016). Unlike in the current study in sorghum internodes, shade signal marker genes such as ZmHB53, which is an ortholog of the Arabidopsis ATHB2 gene, were also differentially expressed in the maize seedlings (Wang et al., 2016). It is likely that plant tissues sampled for the transcriptome analysis of maize seedlings were composed of leaves exposed to light and young tissues enclosed with leaves and sheath. In addition, Salter, Franklin, and Whitelam (2003) showed that the expression of PIL1, a shade marker gene, was upregulated in the morning in response to shade, while maximal elongation response was documented in the evening when the evening clock gene TOC1 was expressed. The authors concluded that both PIL1 and TOC1 are required for elongation in response to shade. Therefore, the regulation of the circadian clock and response to shade in plant tissues that are exposed to light could be different from tissues that are not exposed to light.
In tissues that are illuminated with direct light, the simultaneous effect of photoreceptors on shade signaling and clock resetting in the morning might suppress the role of the clock in promoting shoot elongation in response to shade. Consistent with this, it appears that maximal shoot elongation in response to shade in Arabidopsis is delayed until in the afternoon when the level of morning genes is reduced (Sellaro et al., 2012). In sorghum, shade signals promoted internode and sheath elongation but did not increase the growth of leaf blades that are illuminated ( Figure 3). Interestingly, the sheath in sorghum also grows partially enclosed by an older sheath. It is possible that the enclosed section of the sheath, and not the illuminated section, elongated in response shade.
A link between enhanced shoot growth and reduced expression of CCA1-like gene, which is a homolog of LHY, has been reported in maize (Ko et al., 2016). Interestingly, the CCA1 gene in maize is homologous to the LHY gene of sorghum identified among DEGs in the current study. The ears in maize develop while they are enclosed with leaves and may not perceive light. The morning genes such as CCA1 and LHY were downregulated at all times in maize ears (Hayes et al., 2010). Interestingly, while the expression of the LHY gene in maize ears was low over a 24 hr period, the cycling of morning genes with lower amplitude has been documented. In the current study, we

| CON CLUS ION
The current study identified that shade signals promote internode elongation in the bioenergy sorghum genotype R.07020. Since internodes in bioenergy sorghum grow while enclosed by leaves, we hypothesize that shade signals indirectly alter the function or patterns of expression of the circadian clock genes to accelerate internode elongation. In tissues that are not illuminated, it is possible that the expression of clock genes is modulated by other factors.
For example, sucrose can modulate the function of the circadian clock (Dalchau et al., 2011;Philippou, Ronald, Sanchez-Villarreal, Davis, & Davis, 2019), and downregulation of CCA1 in Arabidopsis leaves enhanced chlorophyll and starch content and growth vigor (Ni et al., 2009). Therefore, changes in carbohydrate metabolism and partitioning in response to shade signals may alter the patterns and timing of expression of clock genes in stem internodes of bioenergy sorghum.

| Plant materials and growing conditions
Seed of bioenergy sorghum genotype R.07020 were planted in 3-gallon pots filled with the commercial growth mix MVP and field soil (3:1) in two growth chambers, 12 pots in each chamber. The growth chambers were supplied with incandescent and florescent lamps producing about 350 µmol m −2 s −1 photosynthetically active radiation (PAR). The temperature in the growth chambers was

| Microscopic study of internode samples
Longitudinal and cross-sections from the median region of the first subapical internodes of shade-treated and control plants were prepared for microscopic study of the development of the stem tissues.
The internode sections were fixed in FAA overnight and stored in 70% ethanol. Subsequent tissue dehydration, embedding in paraffin, sectioning to 10 µm, and staining with alcian blue and safranin were performed at the histopathology laboratory at Texas A&M University School of Veterinary Medicine. The slides were scanned at 20X using Nanozoomer HT digital slide scanner, and the images were viewed using NDP.view2 software (Hamamatsu Photonics).

| RNA-seq library preparation, sequencing, and analyses
To identify the early molecular events that mediate internode elongation in response to shade signals in the bioenergy sorghum inbred line R.07020, RNA-seq libraries were prepared from the newly-formed Workbench User Manual (CLC bio). Initially, transcripts with CPM values <2 for any of the nine biological replicates were removed from the analysis. Next, TMM normalization of the retained transcripts was performed to calculate the effective library sizes required for further normalization steps and CPM was again calculated using the TMM effective library sizes. These CPM values were then log-transformed to obtain normality of the expression distribution. Next, working with logCPM-normalized data, z-score normalization was performed across all transcripts for each biological replicate. These z-scores were used as input into PCA.
To identify DEGs, sequence reads were aligned to the sorghum V3 genome (DOE-JGI, http://phyto zome.jgi.doe.gov/) and statistically analyzed using EdgeR in the CLC bio workbench (CLC bio). A MapMan mapping file was created using Mercator by Blast search of the sorghum bicolor V3.1 transcripts against the Arabidopsis proteome, and DEGs were annotated using the MapMan software (Lohse et al., 2014;Thimm et al., 2004). GO overrepresentation test was conducted using the corresponding Arabidopsis IDs of the sorghum DEGs for GO biological process in GO ontology database.

This work was funded by the DOE Great Lakes Bioenergy Research
Center (DOE BER Office of Science DE-FC02-07ER64494). Also,

THK was supported by the Texas A&M System Chancellor's Research
Initiative for the Center for Computational Systems Biology and by the USDA-NIFA Evans-Allen funds at the Prairie View A&M University.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.

AUTH O R CO NTR I B UTI O N
THK and JEM conceived the research project; THK performed the research; and THK, BAM, and JEM analyzed the data. THK, BAM, and JEM wrote the manuscript. All authors read and approved the final manuscript.