PIF7 is a master regulator of thermomorphogenesis in shade | Panda Anku

Synergistic effect of low R/FR and elevated temperature on hypocotyl growth

Previous studies suggest that both low R/FR and warm temperatures promote growth through the phyB-PIF-auxin signaling module5,11,12,25,26,30,35. We asked what the response to both signals would be when presented simultaneously. Seedlings were grown in constant white light (WL) at 21 °C for 3 days. Afterward, they were kept in WL at 21 °C (21WL) or moved either to WL at 30 °C (30WL), to low R/FR at 21 °C (21FR), or to low R/FR at 30 °C (30FR, warm shade) for additional 3–5 days (Fig. 1a). We chose 30 °C for our experimental set-up since we found that maximal hypocotyl elongation in response to elevated temperature occurred between 28 and 30 °C (Supplementary Fig. 1a).

Fig. 1: Synergistic hypocotyl elongation under combined shade and warm temperature.
figure 1

a Scheme of the experimental design. Seedlings were grown in 21 °C with constant simulated white light (WL) for 3 days and then moved to 30 °C (30WL), 21 °C with shade (21FR, R/FR = 0.6), 30 °C with shade (30FR, R/FR = 0.6) or kept at 21 °C white light (21WL) for three more days. b Representative images of 6-day-old Arabidopsis wild-type seedlings grown as described in (a). Scale bar = 2 mm. c Hypocotyl length of wild-type seedlings during the first 4 days after transfer to the different conditions. Insert is an enlargement of the first hours. The average values +/− SE are shown, n = 8 or more seedlings per time point (for the exact number in each sample, see Source data). d Representative SEM images of hypocotyl epidermal cell lengths at the apex, base, and middle of 6-day-old seedlings grown as described in (a). We repeated the experiment five times with similar results. Scale bar = 100 µm e. Hypocotyl epidermal cell length was determined from confocal images of 8–9 seedlings per condition and 2–3 cells per segment. The number of cells (n) is shown under each box. f Hypocotyl length of 6-day-old wild-type seedlings grown for 3 days at 21WL and then moved to the indicated temperature and light. The response to 21FR is shown for two different chambers used for the experiment. The number of seedlings (n) is shown on top of each box. g Hypocotyl length of 6-day-old wild-type seedlings grown for 3 days at 21WL, then moved to the indicated conditions. In gray, the plates were transferred from 21FR to 30WL or from 30WL to 21FR, after the indicated time and stayed there for the rest of the experiment. The number of seedlings (n) is shown under each box. In bg, PAR = 70 µmol m−2 s−1. In eg, Different letters denote statistical differences (p < 0.05) among samples as assessed by one-way ANOVA and Tukey HSD. Boxes indicate the first and third quartiles and the whiskers indicate the minimum and maximum values, the black lines within the boxes indicate the median values and gray dots mark the individual measurements.

In agreement with previous reports, 21FR and 30WL promoted similar elongation of the hypocotyl (Fig. 1b, c). However, the treatment with both signals simultaneously (30FR) resulted in a much longer hypocotyl than either stimulus alone. We detected this effect as early as 4 h after the transfer to 30FR (Fig. 1c). This synergistic elongation was reflected in epidermal cell elongation throughout the hypocotyl (Fig. 1d, e), and was detectable at 24 °C, with maximum impact at 28–30 °C (Fig. 1f). Since the interaction between temperature and low R/FR most likely occurs in many crops grown in high density, we wanted to confirm that the synergistic growth response is conserved in crop plants. We found that the synergistic response also occurs in tomato (Solanum lycopersicon) and Nicotiana benthamiana (Supplementary Fig. 1b–f), emphasizing the importance of identifying the factors and the mechanisms that drive this interaction.

We next asked if the synergistic growth response observed at 30FR requires simultaneous light and temperature signals or if it can be achieved by consecutive treatments. We transferred seedlings between 21FR and 30WL, and between 30WL and 21FR at different time points and measured hypocotyl lengths. Only the seedlings that were exposed to simultaneous low R/FR and high temperature showed the synergistic elongation response (Fig. 1g, Supplementary Fig. 2a, b). Sequential presentation of these stimuli failed to enhance growth. We, therefore, conclude that simultaneous sensing of low R/FR and warm temperature is required for this enhanced growth response. Additionally, we noticed that in low light conditions (30 µmol m−2 s−1), the synergistic interaction between low R/FR and the warm temperature was suppressed (Supplementary Fig. 3a), indicating that this response is light intensity dependent.

PIF7 is necessary and sufficient for the synergistic growth response

The importance and dominance of PIF4, and to some extent PIF5, in thermomorphogenesis has been shown in many studies15,16,35,36. Recently, PIF7, which has a dominant role in response to low R/FR, was also found to have a role in the thermomorphogenesis response19,20,23,25. We, therefore, asked which PIFs are essential for the synergistic response of elevated temperatures under low R/FR conditions. We used a genetic approach and examined this response in multiple pif mutants. In agreement with published data, we found that PIF7 and PIF4 played dominant roles in response to low R/FR and to warm ambient temperature, respectively (Fig. 2a). The pif7 mutant showed reduced hypocotyl elongation in warm temperatures, but this phenotype was not as strong as observed in pif4 or pif45 mutants. Surprisingly, we found that pif4 or pif45 mutants resembled wild-type seedlings in 30FR, while pif7 eliminated the synergistic elongation. This phenotype was slightly enhanced in the pif457 mutant (Fig. 2a). These results suggest that PIF7 has the dominant function when the elevated temperature is sensed together with low R/FR (30FR), while PIF4 and PIF5 are minor contributors. This pattern was also apparent under long-day conditions at 28 °C (Supplementary Fig. 3b). Interestingly, while pif457 suppressed the long hypocotyl phenotype of phyB in 21WL or 21FR, it only partially suppressed the phyB long hypocotyl phenotype at 30WL (Supplementary Fig. 3c, d), indicating that part of the response to 30WL in phyB plants is PIF457-independent, as shown before37.

Fig. 2: PIF7 plays a dominant role in the synergistic response to low R/FR and warm ambient temperature.
figure 2

a Hypocotyl length of 6-day-old Arabidopsis pif mutant seedlings. The number of seedlings (n) is shown under each box. b Hypocotyl length of 6-day-old Arabidopsis PIF7:PIF7-4xMYC (pif457) seedlings. The number of seedlings (n) is shown under each box. The growth conditions are the same as described in Fig. 1a, PAR = 70 µmol m−2 s−1. Different letters denote statistical differences (p < 0.05) among samples as assessed by one-way ANOVA and Tukey HSD test. Boxes indicate the first and third quartiles and the whiskers indicate the minimum and maximum values, the black lines within the boxes indicate the median values and gray dots mark the individual measurements.

The activity of PIF7 in thermomorphogenesis is dependent on PIF4 and PIF519. We asked if this relationship persisted in response to 21FR or 30FR. We tested a transgenic line expressing PIF7 driven by its native promoter (PIF7:PIF7-4xMYC) in the pif457 background23. This line failed to rescue elongation in 30WL but fully rescued elongation in 21FR or 30FR, perfectly phenocopying the pif45 mutant phenotype (Fig. 2a, b). The direct comparison between pif457 and pif45 suggests that PIF7 activity depends on PIF4 in 30WL, but does not require PIF4 or PIF5 in 21FR or for the exaggerated growth response to 30FR. Also evident by the complementation of pif457 by PIF7:PIF7-4xMYC in 21FR and 30FR conditions but not in 30WL. In addition, we found that PIF4 and PIF5 transcripts were upregulated in 30WL and 30FR while PIF7 expression did not respond to changes in growth conditions, suggesting different post-transcriptional mechanisms governing PIF4, PIF5, and PIF7 activity in 30WL and 30FR conditions (Supplementary Fig. 4a).

Next, we asked if the dominant function of PIF7 and the minor role of PIF4 observed under combined temperature and light changes are unique to low R/FR conditions. To test this, we chose Low Blue Light (LBL) since growth under LBL was shown to be regulated mainly by PIF438,39. We combined LBL with warm ambient temperature, finding that while LBL and warm temperature interacted synergistically, PIF4 played the major role, while PIF7’s contribution was smaller (Fig. 3a). Consistent with these results, and similar to 30WL, the PIF7:PIF7-4xMYC construct failed to fully rescue the elongation of pif457 mutant in LBL at 21 °C or 29 °C (Fig. 3a). We hypothesized that the functions of PIF7 in low R/FR and PIF4 in LBL and thermomorphogenesis might result from their interactions with phyB and cry1. PIF7 being potentially regulated by phyB alone, while PIF4 is controlled by phyB and cry138,39,40,41. To test this hypothesis, we compared hypocotyl elongation of cry1 and pif mutants in response to 21FR, 30WL, and 30FR. We found that pif7 suppressed the long hypocotyl phenotype of cry1 in all tested conditions (Fig. 3b). This indicates that PIF7 works downstream of both phyB and cry1. We confirmed our finding in high monochromatic blue light (Supplementary Fig. 4b)42.

Fig. 3: PIF7 acts downstream of both PHYB and CRY1.
figure 3

a Hypocotyl length of 6-day-old wild-type Arabidopsis seedlings grown for 3 days at 21 °C, then moved to the indicated conditions. Long-day white light (WL, fluorescent bulbs, PAR = 60 µmol m−2 s−1), or Low Blue Light (LBL, PAR = 56 µmol m−2 s−1, Supplementary Data 1). Temperature: constant 21 °C or 29 °C. The number of seedlings (n) is shown under each box. b Hypocotyl length of 6-day-old Arabidopsis cry1 and pif mutant seedlings. The number of seedlings (n) is shown under each box. c Hypocotyl length of 6-day-old phyB seedlings expressing 35S:PHYB-GFP, 35S:PHYBSer86Asp-YFP (S86A), or 35S:PHYBSer86Ala-YFP (S86D). The number of seedlings (n) is shown under each box. In b, c, growth condition is the same as described in Fig. 1a, PAR = 70 µmol m−2 s−1. Different letters denote statistical differences (p < 0.05) among samples as assessed by one-way ANOVA and Tukey HSD test. Boxes indicate the first and third quartiles and the whiskers indicate the minimum and maximum values, the black lines within the boxes indicate the median values and gray dots mark the individual measurements.

All together these results demonstrated that PIF7 has a dominant role in response to elevated temperature when combined with low R/FR. Surprisingly, PIF4 has a minor role in the combined response to high temperature and low R/FR, while maintaining a major role in the combined response to warm temperature and LBL. This indicated separate mechanisms for processing warm temperatures depending on light quality.

The response to 30FR condition is independent of phyB thermal reversion

The phyB-PIF-auxin signaling pathway is central to growth responses in low R/FR and warm temperatures. PIF7 is regulated mainly by phyB23,41, and we found that PIF7 is the major PIF required for the synergistic response of low R/FR and warm ambient temperature. Therefore, we examined the contribution of phyB to this response and asked whether changes in PIF7 activity can explain the synergistic response to 30FR. We found that the relative response to 30WL in phyB mutants was similar to wild-type seedlings (Supplementary Fig. 3c, d). In addition, phyB abundance was similar in all growth conditions (Supplementary Fig. 3e, f). However, relative elongation growth in response to low R/FR at either 21 °C or 30 °C was compromised in phyB mutants (Fig. 3c, Supplementary Fig. 3c, d). As the phyB thermal reversion rate was shown to increase with increasing ambient temperature11,12, we asked if this phenomenon might contribute to the synergistic response to 30FR. We tested transgenic lines that overexpress PHYB with a modified S86 phosphorylation site leading to faster reversion (phyBS86D-YFP) or slower reversion (phyBS86A-YFP)43. Phosphorylation of this serine residue varies based on low R/FR and temperature conditions44. Plants overexpressing phyB-GFP or phyBS86A-YFP had shorter hypocotyls than controls in all tested conditions (Fig. 3c). However, overexpression of phyBS86D-YFP, led to a wild-type-like hypocotyl phenotype in response to 21FR and 30FR but failed to elongate in 30WL (Fig. 3c). These results suggest that in our light conditions (continuous white light), the effect of increased temperature on phyB activity is minor relative to the impact of low R/FR light. In addition, the synergistic effect seen in response to 30FR cannot be explained by decreased phyB activity under simultaneous low R/FR and warm temperature.

Next, we asked whether PIF7 activity increases in 30FR. Since PIF7 is regulated by phosphorylation, we tested whether there is a difference in PIF7 phosphorylation between 21FR and 30FR. We found that PIF7 was dephosphorylated as early as 10 min in the low R/FR and stayed dephosphorylated as long as it remained in low R/FR in both temperatures (Fig. 4a, Supplementary Fig. 5). PIF7 was rapidly phosphorylated when transferred from 21FR to 30WL (Fig. 4b). This result suggests that PIF7 is dephosphorylated similarly by low R/FR in 21 °C and 30 °C. It was recently shown that PIF7 protein level increases during the thermomorphogenesis response19,20. We did not observe difference in PIF7 levels when seedlings were transferred from 21WL to 21FR versus 30FR (Fig. 4a, Supplementary Fig. 5). We, therefore, asked whether PIF7 is more active in promoting the transcription of its targets in 30FR compared to 21FR or 30WL. We performed a global transcriptomic analysis using RNA-sequencing (RNA-seq) and chromatin immunoprecipitation-sequencing (ChIP-seq) and compared the expression levels of direct PIF7 targets under 21WL, 21FR, 30WL, or 30FR. The majority of PIF7 direct targets were upregulated in both 21FR and 30FR (Fig. 4c–f). In agreement with this result, PIF7 binding to the DNA was similar in 21FR and 30FR (Fig. 4g, h, Supplementary Data 4). These results suggest that PIF7 has the same activity under 21FR and 30FR conditions. Interestingly, in the 30WL condition, PIF7 binds to very similar sets of genes as in 21FR and 30FR conditions, while almost no binding was observed in 21WL (Fig. 4g, h, Supplementary Data 4). In addition, PIF7 binding in the 30WL condition was insufficient to rescue the pif457 mutant response or promote the expression of target genes such as IAA29 and ATHB2 (Figs. 2b, 4d, e, h). These results indicate that while warm temperature promotes the binding of PIF7 to its targets, its activity depends on the interaction with additional co-factors.

Fig. 4: PIF7 activity is comparable in low R/FR at low or high temperature (21FR vs 30FR conditions).
figure 4

a Immuno-detection of PIF7-MYC Protein levels using anti-MYC antibody. Total protein extract was collected from 3-day-old PIF7:PIF7-4xMYC (pif457) seedlings grown in 21WL (70 µmol m−2 s−1) plus the indicated time in the stated conditions. Anti-ACTIN blots are shown below as loading controls. This experiment was done one time. b Same as (a) except protein extract was collected after 24 h in the indicated starting condition plus the stated time in the treatment condition. This experiment was done one time. ce Expression of PIL1 (c), IAA29 (d), and ATHB2 (e) in 3-day-old seedlings grown at 21 °C in 21WL (70 µmol m−2 s−1) plus the indicated time in the stated conditions. The expression from the RNA-seq data is shown as normalized counts per million reads mapped (cpm), and as the average of two biological replicates per condition. Color dots mark the individual measurements. f Relative expression of the top 50 PIF7-bound genes in 30FR that are also differentially expressed in at least one condition of the RNA-seq (FDR < 0.05, n = 31). Log2FC: Log2 Fold Change relative to 21WL at each time point. Different letters denote statistical differences (p < 0.05) within time points assessed by one-way ANOVA and Tukey HSD. Boxes indicate the first and third quartiles and the whiskers indicate the minimum and maximum values, the black lines within the boxes indicate the median values and gray dots mark the individual measurements. g Venn diagram comparing the top 200 PIF7-bound genes in 21FR, 30WL, and 30FR. Bound genes were annotated if a peak was found 2 kb upstream to the TSS or in the gene body in both replicates of each growth condition. All top 200 genes in 30FR were also bound in 21FR (all genes) and all top 200 genes in 21FR were also bound in 30FR (all genes) except two (see Supplementary Data 4 for the top 200 genes in each condition). h Visualization of PIF7-MYC binding to the IAA29 and ATHB2 promoters. Note that differences in binding peak intensities between replicates might be due to differences in library preparation and data analysis between replicates 1 and 2 (see “Methods” section). Light and temperature conditions in ag were the same as in Fig. 1.

We concluded that deactivation of phyB by low R/FR, dephosphorylation of PIF7, and activation of PIF7-mediated transcription are necessary for the enhanced elongation at 30FR. However, these steps cannot explain the synergistic effect observed in 30FR since we observed these low R/FR responses independent of the ambient temperature.

Auxin-dependent hypocotyl elongation differs in response to 21FR, 30WL, and 30FR

Both low R/FR and warm temperature promote growth through the canonical phyB-PIF signaling pathway, which controls the levels of auxin19,25,35,45,46,47. Therefore, control of the auxin levels and auxin sensitivity are appealing candidates to explain the synergistic growth response at 30FR. To test this, we determined the levels of auxin (indole-3-acetic acid, IAA) and IAA metabolites under our growth conditions. We observed a slight increase in IAA levels in response to 21FR and 30FR after 3 h. In addition, we observed a substantial increase in the levels of several IAA conjugates. IAA-Glutamate (IAA-Glu) accumulated in response to 30FR and 21FR, while IAA-Aspartate (IAA-Asp) increased only in response to 30FR. The changes in IAA-Glu and IAA-Asp levels correlated well with the expression of the GH3 genes, which regulate their accumulation (Supplementary Fig. 6a, b)48,49. In response to 30WL, we only observed a change in the IAA-glucose (IAA-glc) level. However, we did not detect changes in the expression levels of the enzymes that generate this conjugate: UGT84B1 (below detection) and UGT74D150,51,52 (Supplementary Fig. 6a, Supplementary Data 2). The accumulation of IAA conjugates suggests an increase in IAA after exposure to 21FR, 30WL, or 30FR, most likely prior to the time point of sampling. Next to test the contribution of auxin and its transport to hypocotyl elongation under these conditions, we applied NPA (N−1-naphthylphthalamic acid, an inhibitor of auxin transport) or peo-IAA (an inhibitor of auxin perception). We found that growth in all four conditions is dependent on auxin transport and auxin perception, as previously reported for low R/FR and thermomorphogenesis (Supplementary Fig. 6c)27,53. These results also indicate that the accumulation of IAA and its conjugates in response to low R/FR and to high temperature is highly dynamic. We also examined the role of the YUCCA (YUC) auxin biosynthetic enzymes. We used a high order yucca mutant, yucca2589, which is known to suppress the response to low R/FR30. The yucca2589 mutant strongly suppressed the response to 21FR but only slightly suppressed the response to 30WL. The response to 30FR was similar to 30WL (Supplementary Fig. 6d). While YUC2,5,8 and 9 were shown to be important for thermomorphogenesis under long-day conditions19, our results suggest that in constant light, they are primarily involved in the low R/FR response, with only a minor role in the temperature response.

Given the minor differences observed in auxin biosynthesis among our conditions, we asked if the increased elongation under 30FR may be due to increased sensitivity to auxin. HSP90 proteins were shown to increase the sensitivity to auxin during the response to warm temperature by stabilizing the TIR1 auxin receptor54. In addition, in our transcriptomic analysis, we found that many HEAT SHOCK PROTEINS (HSPs) and HEAT SHOCK FACTORS (HSFs) were upregulated in 30WL and 30FR (Supplementary Fig. 7a, Supplementary Data 3). We hypothesized that HSP90s and HSFs might explain the synergistic response at 30FR. To test this hypothesis, we blocked HSP90 activity using Geldanamycin (GDA) and, in agreement with published data, we observed dose-dependent inhibition of hypocotyl elongation in 30WL and 30FR (Supplementary Fig. 8a)54. Next, we decided to test the SUPPRESSOR OF G2 ALLELE SKP1b (SGT1b/eta3), which is crucial for HSP90 activity55. In contrast to our results with GDA treatment, the eta3 mutant showed only a minor decrease in hypocotyl length at 30WL and 30FR and still responded in a synergistic manner to 30FR (Supplementary Fig. 8b). To address this discrepancy, we used the shade responsive PIL1 promoter (PIL1p, Supplementary Fig. 8c) to conditionally overexpress HSFA2, HSP90.1, HSP90.2, and a dominant-negative mutant form HSP90.2(D80N), which binds but does not release its client proteins. We hypothesized that if HSFA2 and HSP90 promote growth in response to warm temperatures, shade-dependent overexpression will promote further hypocotyl growth in 21FR, while HSP90.2(D80N) will suppress growth in 30FR. However, while the expression of these proteins was induced by low R/FR, we could not detect enhancement of growth by HSFA2, HSP90.1 or HSP90.2, or inhibition of growth by HSP90.2(D80N). All lines behaved like the wild-type control under all conditions (Supplementary Fig. 8d, e). Therefore, we cannot confirm or reject the hypothesis that HSP90s are responsible for the synergistic response to low R/FR and warm temperature.

If simultaneous low R/FR and warm temperature specifically modify auxin sensitivity, an increase in the expression of auxin response genes could be expected. To test whether the auxin response differs between 21FR, 30WL, and 30FR, we compared the expression of IAA29, YUC8, YUC9, SMALL AUXIN UP RNA 19 (SAUR19), and SAUR22, all of which are upregulated in response to auxin, low R/FR, or warm temperature25,26,27,30,32,35,39,56,57. In 21FR and 30FR, RT-qPCR showed that the expression levels of IAA29, SAUR19, SAUR22, YUC8, and YUC9 were massively increased (6–70-fold over 21WL), while in response to 30WL, their expression was only slightly increased (1.5–3-fold over 21WL) (Figs. 4d, 5a). We observed similar patterns in the whole transcriptome data (see cluster 3 in Supplementary Fig. 7a and Supplementary Data 3), as well as a narrower comparison of SAUR genes (Fig. 5b). In addition, examination of SAUR19p:GUS and DR5:GUS reporters lines showed that these changes occur in cotyledons and in hypocotyls (Fig. 5c and Supplementary Fig. 6e). In agreement with the finding that PIF7 has a dominant role in response to 21FR and 30FR, we found that the expression of YUC8 in these conditions was well correlated with PIF7 activity (Fig. 5d). SAUR22 expression, on the other hand, was dependent on PIF4, 5, and 7, but its expression levels did not tightly correlate with hypocotyl elongation: We observed no difference in expression between the pif7 mutant and wild-type controls in 21FR or 30FR conditions, although pif7 hypocotyls in these conditions are significantly shorter than wild-type hypocotyls (Figs. 2a, 5d). Overall, these results suggest that the magnitude of the auxin response differs between 21WL and 30WL but cannot explain the increased growth in 30FR compared to 21FR.

Fig. 5: Auxin response is higher in shaded conditions than in warm temperature and is predominantly regulated by PIF7 in shade at 21 or 30 °C.
figure 5

a Expression of SAUR22, SAUR19, YUC9, and YUC8 in 3-d-old seedlings grown in constant white light (70 µmol m−2 s−1) at 21 °C plus the indicated time in the stated conditions. Relative expression assayed using RT-qPCR relative to the reference gene IPP2. The average values of three biological replicates per condition +/− SE shown. b Relative expression of all SAUR genes that show differential expression in at least one condition assayed by RNA-seq (Log2FC < −1 or >1, FDR < 0.05). Expression in 3-day-old seedlings (grown in constant simulated white light- 70 µmol m−2 s−1 at 21 °C) plus the indicated time in each of the conditions, shown as Log2FC relative to 21WL. Different letters denote statistical differences (p < 0.05) between conditions, within each time point as determined by one-way ANOVA and Tukey HSD test. Boxes indicate the first and third quartiles and the whiskers indicate the minimum and maximum values, the black lines within the boxes indicate the median values and gray dots mark the individual measurements n = 18 genes. c Images of GUS-stained whole Arabidopsis seedlings carrying a SAUR19p:GUS reporter. Plants were grown in constant simulated white light (70 µmol m−2 s−1) at 21 °C plus 6 h or 8 h in the conditions indicated on the left. d Expression of YUC8 and SAUR22 in 3-d-old wild-type, pif45, pif7, and pif457 seedlings grown in constant simulated white light (70 µmol m−2 s−1) at 21 °C plus 4 h in the stated condition. Gene expression was assayed using RT-qPCR relative to the reference gene IPP2 and normalized to the expression in wild type at 21WL. The average values of three biological replicates per condition +/− SE are shown. Different letters denote statistical differences (p < 0.05) among samples as assessed by one-way ANOVA and Tukey HSD test.

Our data set revealed few genes that were differentially expressed only in the 30FR condition during the first 6 h. Upon closer analysis, we found that not a single gene was differently expressed across all time points (Supplementary Fig. 7b, c, e and Supplementary Data 3). This suggests that there are no unique gene sets differentially expressed in response to 30FR. In agreement with this, most of these genes were not bound by PIF7 (Supplementary Fig. 7d). However, the dynamics of expression of a relatively small number of genes that are differently regulated in 30FR, such as EXP8 and IAA19, did vary between conditions (Supplementary Fig. 7f).

Warm temperature enhances auxin-mediated hypocotyl growth

The minor changes observed in auxin response gene expression, auxin synthesis, and auxin signaling between 30FR and 21FR led us to hypothesize that increased temperature might increase the efficiency of auxin in promoting growth. To test this, we treated seedlings with the auxin analog picloram and measured hypocotyls after 3 days on the drug (treatment starting day 2: Fig. 6a, b, or day 3: Supplementary Fig. 9a). Without picloram, hypocotyls at 30WL were nearly identical to those at 21FR. However, we observed that elongation growth in response to increasing picloram concentration was stronger at 30WL than at 21WL or 21FR. Even at the lowest concentrations, picloram caused an increase in hypocotyl length in 30WL and mimicked those of untreated seedlings at 30FR. We did not observe a response of this magnitude at 21WL or 21FR. These treatments suggest that high temperatures enhances auxin-mediated hypocotyl growth (Fig. 6a, b).

Fig. 6: Warm temperature enhances the response to auxin.
figure 6

a Dose–response curves for hypocotyl length of 5-day-old wild-type Arabidopsis seedlings treated with the synthetic auxin picloram. Seedlings were grown in constant simulated white light (70 µmol m−2 s−1) at 21 °C (21WL) for 2 days, then moved to plates with the indicated picloram concentration or DMSO control (0) and the indicated growth condition for an additional 3 days. Data represent mean +/− SE; n = 16 seedlings or more per sample (for the exact number in each sample, see Source data). b Hypocotyl length in response to picloram shown in (a), normalized to the hypocotyl length of mock treatments in each growth condition. Data represent mean +/− SE; n = 16 seedlings or more per sample (for the exact number in each sample, see Source data). c A model of hypocotyl elongation in response to low R/FR, warm temperature, and both signals simultaneously. During warm temperatures and high R/FR (30WL), the abundance of PIF4 and to some extent PIF7 increases, while both PIFs are still repressed by cry1 and phyB. This leads to a slight increase in auxin levels which initiate the growth response but depends on additional unknown factors (X). During the response to low R/FR (21FR), the repression of phyB over PIF7 is removed and auxin is induced, which is the main driving force that promotes hypocotyl elongation. In this condition, PIF4 plays a minor role (represented by faint font). When plants sense low R/FR and warm temperature simultaneously (30FR), auxin production increases as in the low R/FR, while the warm temperature activates factor X, boosting the effect of auxin in the hypocotyl resulting in extensive growth. The dashed line represents the increased dark reversion rate of phyB in cycling conditions.

Together, our results indicate that the role of auxin in thermomorphogenesis is different from its role in low R/FR. While 21FR caused a dramatic increase in auxin response, 30WL led only to a moderate increase. However, both signals led to similar hypocotyl elongation phenotypes. Therefore, our data indicate that while auxin is necessary to initiate thermomorphogenesis, growth at 30WL depends on additional unknown factors that increase the efficiency of auxin in promoting elongation. Hence, in seedlings that sensed low R/FR and elevated temperature simultaneously, low R/FR might induce auxin production while warm temperature might induce the expression or the activity of other factors that likely work downstream to auxin and increasing the effects of auxin on hypocotyl growth (Fig. 6c). The interaction between the high level of auxin response and increase in auxin efficiency in regulating growth, results in the observed hyper-elongated phenotype in 30FR.

Leave a Comment