Global Transcriptome Analysis Reveals the Role of the GI-CO-FT Pathway in Response to microgravity
Arabidopsis plants were cultivated in the General Biological Modular Culture System (GBMCS) aboard the CSS (Fig. 1a, b). The GBMCS is equipped with subsystems for atmospheric, temperature, and hydration regulation and a lighting system (Fig. 1c). Additionally, the GBMCS is fitted with four video cameras to facilitate image capture and to visually monitor the ongoing experiment. Seeds were germinated and grown in two culture chambers (CCs), CC1 and CC2, which were installed in GBMCS under microgravity onboard the CSS and under normal gravity on the ground, respectively (Fig. 1d, e). The images of Arabidopsis were taken by inflight cameras once every two hours to record the processes of growth and development (Fig. 2a-c; Supplementary Fig. 1). Ground control experiments were carried out in parallel with CCs in an engineering reproduction of the spaceflight GBMCS.
Compared with the ground controls, the time of bolting of WT plants grown in space appeared to be significantly delayed (Supplementary Fig. 1) as our previous reported9. To determine whether the microgravity response observed in WT was mediated by the key flowering time genes, the different flowering time plants that are representative of the photoperiodic floral pathways were exposed to microgravity in space. Mutant gi flowers late, while 35S:CO transgenic plants flower early relative to the WT. FG transgenic plants, which had been constructed in our previous study, flower early under a heat shock condition11. Downloaded images revealed that seedlings of WT, gi, 35S:CO and FG in CCs under microgravity in space and on ground all germinated well at day 3 after sowing. After germination, WT, gi, 35S:CO and FG plants developed rosette leaves under either in space or on ground conditions (Fig. 2c). Plants of all genotypes in space grew slowly with smaller cotyledon. Seedlings of gi mutant grown in space exhibited apparently elongated hypocotyls in comparison with their controls on ground, indicating a positive role of GI in inhibiting elongation of hypocotyls under microgravity condition. Mutant gi both grown in space and on ground didn’t flower during 25 days after sowing (Fig. 2d). Because there was not enough growth time on orbit, we couldn’t rule out the effect of microgravity on the flowering of gi mutant in space. 35S:CO transgenic plants were consistently early flowering both under microgravity in space and on the ground condition, while FG plants displayed a slight earlier in flowering under microgravity in space than those on ground (Supplementary Fig. 1 and Fig. 2). These results revealed a role of CO and FT to activate microgravity response in regulating Arabidopsis flowering in space.
To capture the global transcriptional impacts of microgravity under spaceflight on Arabidopsis growth, we undertook RNA sequencing (RNA seq) of leaves from plants grown in space and compared them to their ground controls. RNA was extracted from leaves of plants at day 25 after sowing in both CCs-space (sp) and CCs-ground (gr). The mean expression levels of genes in the space-grown samples were significantly different from those of the ground controls when we applied a false discovery rate (FDR) threshold of 0.01; these genes were selected as potential differentially expressed genes (DEGs) (Supplementary Table 1). Among these genes, we further refined our list by considering those with expression levels that varied more than five-fold (fold change [FC] ≥ 5) when comparing space samples to their ground controls. These analyses revealed that 4901 genes in WT, 4952 genes in gi mutant, 3871 genes in 35S:CO, and 3832 genes in FG transgenic plants were differentially expressed (Fig. 3a; Supplementary Table 2). To validate these RNA-seq data, we designed sequence-specific primers and performed real-time quantitative PCR (qRT-PCR). qRT-PCR for differential expression levels of selected genes, NUDT 21 (AT1G73540), DIC2 (AT4G24570) and glycine-rich protein (AT1G07135), match with the results analyzed by RNA-seq (Supplementary Fig. 3 and Table 2). In all genotypes, the proportion of up-regulated genes in response to spaceflight conditions consistently higher than that of down-regulated genes (Fig. 3a, b; Supplementary Table 2). Principal Component Analysis (PCA) highlighted marked differences between the transcriptomes of space-grown samples and their ground controls (Fig. 3c), suggesting the regulatory roles of GI, CO, and FT in microgravity responses of Arabidopsis plants.
To understand the biological roles of these DEGs in response to microgravity, we undertook GO analysis for WT, gi, 35S:CO, and FG. We emphasized the top ten GO enrichment terms for both up- and down-regulated genes. Predominantly, up-regulated genes were associated with oxidant detoxification, while down-regulated genes were linked to the photosynthetic process, and these trends were consistent across gi, 35S: CO, FG, and WT (Fig. 4; Supplementary Table 4).For a deeper understanding of the GI-CO-FT module's impact on plant microgravity response, we compared the DEGs involved in flowering time regulation from gi, 35S:CO, and FG with those from WT. Expression of 1517 genes in gi, 798 genes in 35S:CO, and 873 genes in FG showed similarities with WT in response to microgravity in space (termed the "common-sp" group, Fig. 5a; Supplementary Table 5), while a larger number of DEGs in gi, 35S:CO, and FG exhibited different behaviors from what they did in WT. Specifically, 4844 genes in gi, 5427 genes in 35S:CO, and 5224 genes in FG demonstrated different expression responses to microgravity in space compared to those in WT (named 'df-sp' group, Fig. 5a; Supplementary Table 5). A significant overlap, constituting about 40% of the total, was observed among the "df-sp" DEGs from gi, 35S:CO, and FG (Fig. 5b).
The ten most significantly enriched GO terms for 'df-sp' genes in gi, 35S:CO, and FG primarily involved the photosynthetic process and responses to light stimuli for up-regulated genes, while down-regulated genes were associated with oxidant detoxification processes (Fig. 5c). Furthermore, we also identified genotype-specific GO term enrichments. For instance, gi-associated terms involved secondary metabolic processes and transmembrane transport, 35S:CO was linked to carbohydrate and auxin metabolic processes, and FG showed enrichment in UV response and chlorophyll metabolism. Importantly, subcellular localization of all DEGs across gi, 35S:CO, and FG revealed a pronounced association with chloroplast-related genes (Supplementary Table 6).
The response of flowering pathway genes to microgravity in space
Our previous studies demonstrated that the spaceflight environment notably decreased the expression of FT9,11. This suggests that the observed flowering delay in Arabidopsis plants in space might be attributed to modifications in the photoperiodic flowering network, where microgravity stress signals converge on FT expression. To gain a comprehensive view of the changed expression of flowering time genes among the whole genome transcriptional response to microgravity in space, DEGs that were involved in different flowering control pathways were identified. The number of spaceflight-responsive flowering genes identified in WT, gi, 35S:CO, and FG were 134, 132, 127, and 143 respectively (Supplementary Table 7). We categorized these transcripts based on their established or predicted functions in the flowering regulation pathway25. Most of these flowering genes regulated by microgravity in space were functionally categorized to be related to the photoperiodic response (44 genes, accounting for 28% of all DEGs in the flowering pathway). Subsequent classifications highlighted genes functioning in general (34 genes, 22%), hormone regulation (19 genes, 12%), the circadian clock (15 genes, 9%), vernalization (12 genes, 8%), flower development and meristem identity (13 genes, 8%), aging (12 genes, 8%), sugar metabolism (5 genes, 3%), and ambient temperature response (4 genes, 3%). This indicates that microgravity response signaling pathways concurrently integrate into various flowering regulation pathways (Fig. 6).
The circadian clock, an endogenous timekeeping mechanism, enables plants not only to respond to environmental changes but to predict and adapt to upcoming changes. Our previous space experiments suggested that the circadian oscillator might integrate spaceflight responses with photoperiodic signals in Arabidopsis grown in space11. In this study, clock transcriptional repressors (CIRCADIANCLOCKASSOCIATED1(CCA1) and LATE ELONGATED HYPOCOTYL (LHY)) were upregulated, while transcriptional activators (Arabidopsis pseudo-response regulator 3 (APRR3) and Arabidopsis pseudo-response regulator 7 (APRR7)) were downregulated in all four genotypes in space, compared to their ground controls. The expression of TIMING OF CAB EXPRESSION 1 (TOC1) was downregulated in WT but remained unchanged in gi, 35S:CO, and FG. In contrast, Arabidopsis pseudo-response regulator 9 (APRR9) expression was consistent in WT but upregulated in gi, 35S:CO, and FG. These findings suggest that the circadian oscillator participates in regulating the Arabidopsis microgravity response in both GI-CO-FT dependent and independent manners.
Microgravity resulted in altered expression levels of photoperiod-responsive genes, with 22 upregulated and 12 downregulated in WT (Fig. 6). For example, the expression of basic helix-loop-helix (bHLH63) in spaceflight samples was downregulated in WT but upregulated only in 35S:CO, with no changes observed in gi mutants and FG plants in comparison with their controls on ground. Notably, FLAVIN-BINDING, KELCH REPEAT 1 (FKF1) remained unaltered in 35S:CO but was upregulated in WT, gi, and FG. Both bHLH63 and FKF1 have been reported to interact with CO to regulate flowering26–27. Thus, spaceflight-induced alterations in bHLH63 and FKF1 expression levels, especially in 35S:CO, suggest a potential co-regulation between bHLH63-FKF1 and CO in the regulation of photoperiodic flowering in response to microgravity in space.
In the Gibberellin (GA) pathway, spaceflight caused variations in the transcription levels of GA "activating enzymes", GIBBERELLIN 3-OXIDASE 1 (GA3OX1), GA20OX3, and DELLA proteins RGA-LIKE PROTEIN 3 (RGL3) in WT, while no effect was observed in the gi mutant. GA2OX3 and RGL3 were upregulated, whereas GA INSENSITIVE DWARF1A (GID1A) and GA20OX1 were downregulated specifically in 35S:CO plants. In contrast, three GA "activating enzymes" GA20OX1/2/3 notably changed transcription levels in FG in response to microgravity in space (Fig. 6 and Fig. 7).
Regarding the vernalization pathway, the expression of genes like WRKY34, FRIGIDA LIKE 2 (FRL2), SUPPRESSOR of FRI 4 (SUF4), and VERNALIZATION INSENSITIVE 3 (VIN3) displayed unconsistent variations in gi, 35S:CO, and FG in response to spaceflight compared to those in WT. WRKY34 contains two WRKY domains found in the promoters of numerous stress-related plant genes28. All of these genes are essential for regulating flowering in response to vernalization (Fig. 7).
In the aging pathway, spaceflight caused notable changes in the expression levels of 9 genes in WT. Among these genes, miR156 a/c, TARGET OF EARLY ACTIVATION TAGGED (EAT) 2 (TOE2) and TOE3, and SCHLAFMUTZE (SMZ) were upregulated, while SQUAMOSA PROMOTER BINDING PROTEIN-LIKE 3 (SPL3), SPL4, and SPL5 were downregulated (Fig. 6). MiR156 is known to mediate the repression of SPLs and inhibit Arabidopsis floral inductive signals29. The expression of SPL5 in gi, 35S:CO and FG plants grown in space remained unchanged in comparison with their controls on ground, but SCHNARCHZAPFEN (SNZ) was apparently downregulated specifically in gi compared to WT. Conversely, SPL5 was upregulated, and SPL9 was downregulated in 35S:CO in response to microgravity in space (Fig. 6 and Fig. 7).
In summary, spaceflight profoundly influences the expression of key genes in floral pathways. The GI-CO-FT module plays a significant role in the response of Arabidopsis to microgravity in space.
Patterns of cis-element enrichment unveil potential regulatory modules in the transcriptional control of GI-CO-FT pathway in response to microgravity in space
To determine if variations in the expression of GI, CO, and FT influence plant responses to microgravity in space, we employed plant regulomics analyses (bioinfo.sibs.ac.cn, Padjust<0.01) to find overrepresented motifs in the 2-kb upstream sequence (-2000 to + 100) of spaceflight DEGs in WT, gi, 35S:CO, and FG. A total of 214 potential cis-elements correspond to 22 transcription factor (TF) families were found. These corresponded closely with known elements linked to stress-responsive and developmental TFs, such as ethylene responsive factor (ERF), Basic (region) leucine zippers (bZIP), bHLH, MYB, Squamosa promoter Binding Protein (SBP), and WRKY (Table 1, Fig. 8, Supplementary Table 8). Of these cis-elements, 79% (169 members) were observed across all four genotypes (Fig. 8). Motifs highly enriched (Enrichment score > 150% and Padjust <0.01) among spaceflight-regulated genes prominently corresponded to ERF, bZIP, bHLH, and BES1 transcription factors (Table 2).
Based on enrichment scores across the four genotypes, these potential spaceflight-responsive TFs can be divided into two groups. Group I encompass 26 identified TFs all from the ERF family, exhibiting high enrichment scores across genotypes. This suggests a universal microgravity-response regulatory module among the four genotypes. Group II contains specific subsets of genes (e.g., BRI1-EMS-SUPPRESSOR1 (BES1), bHLH, bZIP, ERF, and SBP) that showed high enrichment scores in one or more of the genotypes (Table 2). Notably, five ERF members (ERF027, COOPERATIVELY REGULATED BY ETHYLENE AND JASMONATE 1 (CEJ1), RELATED TO AP2 1 (RAP2.1), DEHYDRATION RESPONSE ELEMENT-BINDING PROTEIN 26 (DREB26) and ERF015) and four bZIP members (ABA-RESPONSIVE ELEMENT BINDING PROTEIN 3 (AREB3), ABSCISIC ACID RESPONSIVE ELEMENT-BINDING FACTOR 1 (ABF1), bZIP28 and Abscisic acid-insensitive (ABI5)) showed enrichment in gi, 35S:CO, and FG, but not in WT. This implies a GI-CO-FT associated regulatory module in response to microgravity in space. Remarkably, 35S:CO showed a greater number of specific enriched TFs in microgravity response compared to gi and FG (Table 2).
The interaction network between microgravity regulated cis-element TFs and regulators of flowering induction
The blueprints for development and response to environment in plants are encoded by cis-regulatory elements (CREs)30, which are often provides an indirection for co-regulation among particular sets of genes. To explore the potential co-regulation of the microgravity response and the flowering induction in space, TFs corresponding to a set of CREs highly enriched among microgravity responsive genes in gi, 35S:CO, and FG, respectively were identified (Group II in Table 2). Within the clock pathway, 15 clock pathway genes were potentially regulated by 12 interconnected TFs in Group II (Fig. 9a). The phytochrome-interacting factor PHYTOCHROME-INTERACTING FACTOR7 (PIF7) emerged as a central interactor of microgravity-responsive clock pathway proteins. Additionally, interactions encompassing the Abscisic Acid (ABA) signaling pathway (namely ABI5 and ABF1 in Group II) were identified. These findings suggest that light regulation and ABA signaling might influence the clock pathway in response to microgravity in space.
For the photoperiod pathway, 13 cis-element TFs in Group II emerged as interactors (Fig. 9b). Here, PIF1, PIF7, and BZR2 showed interactions with phytochromes (PHYTOCHROME A (PhyA), PhyB, PhyC, and PhyD) and cryptochromes (CRYPTOCHROME 1 (CRY1) and CRY2). Meanwhile, ABI5 and ABF1 engaged with nuclear factor Y (NF-YA1, NF-YC3, and NF-YC4) TFs. This pattern indicates a microgravity-induced interplay between light, ABA, and BR signals in modulating the photoperiod pathway.
In the GA pathway, PIF1, BZR2, ABI5, and ABF2 in Group II (Table 2) showed interactions with GA hormone signaling pathway DELLA proteins (GIBBERELLIC ACID INSENSITIVE (GAI), RGL1, RGL2, RGL3, RGA) and with GA enzymes (GA20OXs, GA2OXs, and GA3OXs) (Fig. 9c). These findings hint a possible crosstalk among GA, ABA, BR and light signaling pathways in response to microgravity in Arabidopsis during flowering in space.
In aging pathway, SPL11, which was highly enriched specifically in 35:CO in response to microgravity (Table 2), showed interactions with TARGET OF EAT 2 (TOE2), TOE3, SMZ, and SNZ (Fig. 9d). TOE2/TOE3 were reported to be controlled by the miR156-SPL regulation module of developmental timing31–32, while SMZ and SNZ are important regulators of flowering time in response to nitrate33. This result indicates that SPL11 could play as an important role in regulating expression of TOE2/TOE3 and SMZ/SNZ to control flowering of Arabidopsis plants in space.