To determine the importance of PPD1 for the NB response, we compared wild-type and ppd1-null mutant lines lacking all functional copies of PPD1 in the photoperiod-sensitive hexaploid variety Paragon and in the tetraploid line Kronos-PS . We first measured heading date in these lines when grown in SD or LD conditions since germination. Under SD conditions, neither the wild type nor the ppd1-null mutants of either variety flowered within 150 d, when the experiment was terminated . Under LD conditions, the ppd1-null mutants headed 60 d and 34 d later than the wild type Kronos-PS and Paragon-PS lines respectively . In a separate experiment using slightly different conditions , we compared the effect of NBmax in photoperiod-sensitive and ppd1-null mutant lines. Kronos-PS and Paragon-PS plants headed on average at 73 d and 91 d under NB, respectively, but neither ppd1-null line flowered within 150 d, when the experiment was terminated . These results demonstrated that PPD1 plays a major role in the effect of NB and LD on heading time. We next assayed PPD-B1 and FT1 transcript levels in Kronos-PS and ppd1-null plants at four time points, including dusk, when these flowering time genes are normally expressed at high levels under LD . Plants were grown in SDs for 4 weeks, then either maintained in SD conditions or moved to NBmax conditions for 6 weeks. In Kronos-PS plants, PPD-B1 transcript levels were upregulated 1 h and 3 h after the start of the NB and to even higher levels at dusk . In the Kronos-PS plants kept under SD, PPD-B1 transcript levels were not upregulated during the night but showed an increase at dusk, although the levels were significantly lower than in plants that were exposed to multiple NBs . As expected, PPD-B1 transcripts in the Kronos ppd1-null mutants were not detected in either SD or NB conditions, confirming the specificity of the qRT-PCR primers used in this assay .
Consistent with previous results, FT1 transcripts were undetected in Kronos-PS plants under SD but were highly upregulated in NB conditions at all time points . However, in the ppd1-null mutant, FT1 transcripts were not detected in any sampled time points, including dusk,indoor garden under either SD or NB conditions.We next tested the effect of the timing of the NB on PPD-B1 induction by exposing SD-grown plants to a single NB at different times of the night. We hypothesized that maximal induction of PPD-B1 would coincide with the strongest acceleration of heading date . This hypothesis proved to be incorrect and, instead, we found that PPD-B1 was induced to progressively higher levels in accordance with the duration of the dark period preceding the NB . We first thought that the gradual accumulation of inactive Pr phytochromes in the nucleus resulting from dark reversion could explain the increased PPD-B1 induction with longer periods of darkness. However, plants treated with FR light immediately before a NB applied after 2 h of darkness did not exhibit increased PPD-B1 expression . This result suggested that the accumulation of Pr phytochromes in the nucleus was not responsible for the progressive induction of PPD-B1 with extended dark periods. We then thought that PPD-B1 induction could be associated with the de novo synthesis of phytochromes or other intermediate proteins during darkness. To test this hypothesis, we grew Kronos-PS plants in hydroponic solution and treated half of them with cycloheximide to block protein synthesis and left half of the plants untreated as a control. Consistent with previous results, control plants maintained in darkness showed noinduction of PPD-B1, whereas those exposed to a single NB exhibited strong up-regulation of PPD-B1 expression 2 h after the NB . The induction of PPD-B1 in response to NB was abolished in plants treated with cycloheximide , which demonstrates that the expression of PPD1 in response to light requires active protein synthesis during darkness. This experiment was performed twice with identical results.Many studies using NBs to characterize the effects of changing photoperiods on flowering time focused on SD plants, mainly because the inhibition of flowering by NB was found to be a simpler system of study than the acceleration of flowering by NB in LD plants .
Our characterization of the NB response in wheat highlights some of the similarities and differences between these two systems. In many SD plants, flowering is inhibited by NB and in rice; this effect is associated with the suppression of Hd3a transcription . When rice plants are moved from NB back to inductive SD photoperiods, this inhibition is lost and Hd3a expression returns to high levels. In wheat, NBs also affect the expression of FT1 and flowering time, although these responses are reversed. These results suggest SD and LD plants both respond to NB through regulatory mechanisms acting on FT expression. The opposite effect of NB on FT expression and flowering in rice and wheat is likely determined by the opposite roles of PPD1 in different grass species. In LD grasses, such as wheat and barley, PPD1 induces FT1 and accelerates flowering , whereas in SD grasses, such as sorghum and rice, PRR37 suppresses FT-like genes and delays flowering .In some SD plants, the suppression of flowering by a single R light NB is completely reversible by immediate exposure to FR light . In wheat, we found that a single FR exposure after NB had a limited effect on heading time . One-minute pulses of FR after 1-min pulses of white light were more effective , but did not completely abolish the acceleration of heading by NB . The partial effect of FR light on the NB acceleration of flowering is consistent with previous results in the LD grass barley . Finally, rice and wheat differ in the role of PHYC in the NB response. In rice, the NB response is completely abolished in plants carrying PHYB loss-of-function mutations but is unaffected by similar mutations in PHYC . By contrast, the NB response in wheat is abolished in both the phyB-null and phyC-null mutants . The different roles of PHYC on NB parallel the different roles played by this phytochrome in the photoperiodic response in wheat and rice. PHYC is a positive regulator of flowering time in some temperate grasses such as wheat, barley, and Brachypodium distachyon but has limited or no effect on flowering time in rice and Arabidopsis . These results suggest PHYC plays a more critical role in the photoperiod and NB response in the LD temperate grasses than in other plant species.Whereas a single NB is sufficient to repress flowering in rice and promote flowering in Lolium temulentum cv Ceres , multiple LDs are required to accelerate flowering in many temperate grasses .
Most temperate grasses show some acceleration of flowering after being exposed to 4 to 8 LD photoperiods, but full saturation of this response requires 12 to 16 d of exposure to LD . These results are consistent with our observations for wheat, where 6 to 10 LDs induced a mild acceleration in flowering, but the greatest acceleration in flowering was seen in plants exposed to 12 or more LDs . The acceleration in heading time in response to increasing numbers of NBs was similar to that observed in response to increasing numbers of LDs,but the effects were smaller and at least 15 NBs were required to initiate the acceleration of flowering . These results are consistent with the existence of a PPD1- independent photoperiod pathway, which may be more responsive to LDs than to NBs. In Arabidopsis, the induction of the transition from the vegetative to the reproductive apex also requires cycles of FT induction repeated over several days. However, while 4 to 5 LDs are sufficient to saturate the acceleration of flowering in Arabidopsis more than 20 LDs are required in wheat . Possible explanations for the requirement of multiple NBs or LDs to induce FT1 in wheat include a gradual accumulation of a flowering promoter, a gradual reduction of a flowering repressor, or a gradual change in epigenetic marks in some of the involved genes. No correlation was detected between the number of NBs and transcript levels of ZCCT2 , suggesting that this gene is not critical for the observed changes in FT1 in this genetic background . Similarly,hydroponic farming PPD1 transcript levels did not increase in response to multiple NBs, indicating that the putative accumulating factor is unlikely to be a regulator of PPD1 transcription. However, it is still possible that the number of NBs affect the levels of active PPD1 protein. To test this hypothesis, we have initiated the generation of transgenic wheat plants expressing an HA-tagged PPD1 protein. It is also possible that proteins other than PPD1 also play a role in the regulation of FT1 in response to multiple NBs.Interestingly, we found that the magnitude of PPD-B1 induction by NBs was proportional to the length of darkness preceding the NB. This phenomenon appears to be unrelated to the accumulation of Pr phytochrome protein arising from dark reversion, since exposure to FR light prior to NBs had no effect on the subsequent induction of PPD1 by light . Instead, we found that treating plants with cycloheximide during the night abolished the NB up-regulation of PPD1 , which suggests that the induction of PPD1 by light is dependent on active protein synthesis during darkness.
One possibility is that the de novo synthesis of Pr isoforms of PHYB and/or PHYC during darkness is correlated with the strength of PPD1 induction. During longer periods of darkness, newly synthesized PHYB and PHYC proteins would accumulate to higher levels, so that subsequent light signals would result in stronger induction of PPD1. An alternative possibility is the de novo synthesis and dark accumulation of a PHYB/PHYC-induced transcription factor required for the activation of PPD1. This may include one or more PIFs, which have been shown to act as coactivators of light-induced genes in some cases . Additional experiments will be required to test these hypotheses and to identify the darksynthesized protein responsible for the increased activation of PPD1 with longer periods of darkness.Despite the stronger NB induction of PPD1 following longer periods of darkness, PPD1 transcript levels were not directly correlated with heading date. The greatest effect of NB on heading date was observed when the NB was timed to coincide with the middle of the night, even though PPD1 transcript levels were lower at this point than after NBs applied later in the night. This dependence on the time of the night suggests that PPD1 activity may be gated by circadian clock-regulated genes. The existence of a gating mechanism is also supported by the fact that although PPD1 transcription is induced during the light phases of both SD and LD, FT1 transcription is only observed under LD photoperiods . Furthermore, rhythmic sensitivities for NB-induced flowering have been observed in other LD grasses . L. temulentum cv Ceres plants, which are induced to flower by a single LD cycle, showed two phases of high sensitivity to NB when SD-grown plants were moved to constant darkness. The first phase occurred between 4 and 8 h from the start of the darkness period, and the second one was approximately 20 to 24 h later, suggesting the involvement of a circadian rhythm in the control of flowering in L. temulentum . Similar experiments would be challenging to perform in wheat because of the requirement for multiple NBs to induce flowering. It is tempting to speculate that the regulation of FT1 expression by PPD1 may function in a manner analogous to the regulation of FT by CO in Arabidopsis. In Arabidopsis, FT is induced only in LD conditions when the transcriptional peak of CO coincides with light, which is required to stabilize the CO protein . In wheat, FT1 induction and flowering may be determined by the coincidence of an external signal with an internal rhythm mediated by the circadian clock. In addition to this putative role in gating the effect of PPD1, the circadian clock is known to be involved in the regulation of PPD1 expression . Plants carrying loss-of function mutations in EARLY FLOWERING3 exhibited elevated expression of PPD1 and earlier flowering under both LD and SD .