- Email: [email protected]
trans Isomerase Regulates Flowering Time in Arabidopsis
Molecular Cell
Article Pin1At Encoding a Peptidyl-Prolyl cis/trans Isomerase Regulates Flowering Time in Arabidopsis Yu Wang,1,2 Chang Liu,1,2 Daiwen Yang,1 Hao Yu,1,2,* and Yih-Cherng Liou1,* 1Department
of Biological Sciences, Faculty of Science Life Sciences Laboratory, 1 Research Link National University of Singapore, Singapore *Correspondence: [email protected] (H.Y.), [email protected] (Y.-C.L.) DOI 10.1016/j.molcel.2009.12.020 2Temasek
SUMMARY
Floral transition in plants is regulated by an integrated network of flowering genetic pathways. We show that an Arabidopsis PIN1-type parvulin 1, Pin1At, controls floral transition by accelerating cis/trans isomerization of the phosphorylated Ser/ Thr-Pro motifs in two MADS-domain transcription factors, SUPPRESSOR OF OVEREXPRESSION OF CO 1 (SOC1) and AGAMOUS-LIKE 24 (AGL24). Pin1At regulates flowering, which is genetically mediated by AGL24 and SOC1. Pin1At interacts with the phosphorylated AGL24 and SOC1 in vitro and with AGL24 and SOC1 in vivo and accelerates the cis/trans conformational change of phosphorylated Ser/Thr-Pro motifs of AGL24 and SOC1. We further demonstrate that these Ser/Thr-Pro motifs are important for Pin1At function in promoting flowering through AGL24 and SOC1 and that the interaction between Pin1At and AGL24 mediates the AGL24 stability in the nucleus. Taken together, we propose that phosphorylation-dependent prolyl cis/trans isomerization of key transcription factors is an important flowering regulatory mechanism. INTRODUCTION Floral transition in Arabidopsis is coordinately controlled by multiple genetic pathways in response to various developmental and environmental cues (Koornneef et al., 1998; Levy and Dean, 1998; Mouradov et al., 2002). The autonomous and gibberellic acid (GA) pathways are responsive to endogenous developmental and physiological state, while the photoperiod and vernalization pathways monitor the alteration of environmental signals such as day length and temperature. These genetic pathways ultimately converge at the floral pathway integrators, which in turn regulate the activity of flower meristem identity genes to initiate the transition from vegetative to reproductive growth. Regulatory genes acting at the convergence point of the multiple floral induction pathways include two MADS-box transcription factors, SOC1 and AGL24 (Lee et al., 2000; Liu et al., 2008; Michaels et al., 2003; Samach et al., 2000; Yu et al., 2002).
Peptidyl-prolyl cis/trans isomerases (PPIases) are enzymes that accelerate energetically unfavorable cis/trans isomerization of the peptide bond preceding a proline (Hunter, 1998; Kiefhaber et al., 1990). PPIases include four structurally distinct subfamilies: cyclophilins, FK506-binding proteins, parvulins, and PP2A phosphatase activator (Lu et al., 2007). Members from the parvulin subfamily, such as PIN1 from human and ESS/PTF1 from Saccharomyces cerevisiae, have been shown to be essential for cell cycle and growth (Hanes et al., 1989; Hani et al., 1995; Lu et al., 1996). In the parvulin subfamily, PIN1-like PPIases are the only type of PPIases that specifically recognize phosphorylated Ser/Thr residues preceding proline (pSer/Thr-Pro) and catalyze the conformational change of the phosphorylated substrates, thereby controlling their function (Hsu et al., 2001; Huang et al., 2001; Liou et al., 2002, 2003; Lu et al., 1996, 1999; Pastorino et al., 2006; Ranganathan et al., 1997; Stukenberg and Kirschner, 2001; Yaffe et al., 1997). Pin1At was early-identified as the only PIN1-type PPIase from Arabidopsis (He et al., 2004; Landrieu et al., 2000). Sequence comparison showed that PIN1 homologs from human, yeast, and Drosophila consist of two domains, an N-terminal WW regulatory domain and a C-terminal PPIase catalytic domain, whereas plant PIN1s, including Pin1At, contain only a PPIase domain with four additional amino acids (Figure S1) (Yao et al., 2001). Although several Pin1 homologs have been identified in plants (Landrieu et al., 2000; Metzner et al., 2001; Yao et al., 2001), their substrates and biological function in plants are so far unknown. Here, we show that Pin1At affects floral transition in Arabidopsis and that phosphorylation-dependent prolyl cis/ trans isomerization of key transcription factors such as SOC1 and AGL24 by Pin1At emerges as an important flowering regulatory mechanism.
RESULTS Pin1At Promotes Flowering To investigate the biological function of Pin1At, we first examined the effect of modulating Pin1At expression in Arabidopsis. Since insertion mutants were not available in public resources, transgenic plants with antisense suppression of Pin1At were generated. Among 56 transgenic lines containing Pin1At N-terminal antisense fragment driven by a double cauliflower mosaic virus 35S promoter (Pin1At-AS), 30 plants showed late flowering under long days (LDs) (Figures 1A and 1B). Under
112 Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc.
Molecular Cell Protein Isomerization Mediates Flowering
Figure 1. Pin1At Regulates Flowering Time in Arabidopsis (A and B) Wild-type plants (A) show earlier flowering than representative Pin1At antisense plants (line 18-3) (B) at 35 days after germination under long days. (C and D) Wild-type plants (C) show later flowering than representative 35S:Pin1At plants (line 22-5) (D) at 28 days after germination under long days. (E and F) Flowering time of Pin1At transgenic plants under long days (E) and short days (F). The number of rosette leaves on the main shoot represents flowering time. Values presenting the mean ± standard deviation were scored from at least 15 plants of each genotype. Significant differences in comparison with wild-type plants are indicated with asterisks: *p < 0.05; **p < 0.01, by Student’s t test. (G) Effect of antisense suppression of Pin1At (line 18-3) on overexpression of various flowering promoters under long days. Error bars indicate standard deviation. Significant differences between with and without antisense suppression of Pin1At are indicated with asterisks: *p < 0.05, by Student’s t test. (H) Effect of overexpression of Pin1At (line 22-5) on various flowering mutants under long days. Error bars indicate standard deviation. Significant differences between with and without overexpression of Pin1At are indicated with asterisks: *p < 0.05, by Student’s t test.
both long and short days (SDs), representative Pin1At-AS plants with reduced expression of Pin1At flowered much later than wild-type plants (Figures 1E, 1F, and S2A). On the contrary, among 39 transgenic plants containing Pin1At cDNA under the control of 35S promoter (35S:Pin1At), 21 showed early flowering under LDs (Figures 1C and 1D). Representative 35S:Pin1At lines with overexpression of Pin1At showed early flowering under both LDs and SDs (Figures 1E, 1F, and S2B). In addition, these plants also exhibited serrated rosette and cauline leaves (Figure 1D). These observations suggest that Pin1At at least affects flowering and leaf development. The Photoperiod and Vernalization Pathways Affect Pin1At Expression To examine the flowering pathways that regulate Pin1At, we examined its expression in various environmental conditions and flowering mutants. In wild-type Col plants under LDs, Pin1At expression gradually increased during floral transition occurring from 9 to 13 days after germination (Figure 2A) and decreased afterwards. On the contrary, its expression was not significantly changed under SDs within 21 days after germination. These results suggest that the photoperiod pathway affects Pin1At
expression. Further examination of Pin1At expression revealed that its expression was affected by CONSTANS (CO), but not FLOWERING LOCUS T (FT) (Figure S3). Pin1At expression was also affected by vernalization (Figure 2B), but not by autonomous pathway mutants and GA treatment (Figures S4 and S5). Notably, dramatic downregulation of a flowering repressor, FLOWERING LOCUS C (FLC), in FRI FLC, wherein the dominant allele of FRIGIDA (FRI) causes high expression of FLC by vernalization, did not enhance the upregulation of Pin1At (Figure 2B), indicating that vernalization regulates Pin1At in an FLC-independent manner. Regulation of Pin1At by the vernalization pathway partly explains that manipulation of Pin1At expression affects flowering time in SDs, under which the LD photoperiod pathway is not activated. Pin1At Regulates Flowering Time through AGL24 and SOC1 Since the photoperiod and vernalization pathways converge on the regulation of Pin1At expression, we speculated that Pin1At function may be relevant to flowering regulators located downstream of several genetic pathways, such as SOC1, FT, and AGL24 (Kardailsky et al., 1999; Kobayashi et al., 1999; Lee et al., 2000; Michaels et al., 2003; Samach et al., 2000; Yu et al., 2002). Thus, we performed genetic crossing to identify potential interacting regulators of Pin1At. While downregulation of Pin1At in Pin1At-AS caused late flowering, it did not
Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc. 113
Molecular Cell Protein Isomerization Mediates Flowering
Figure 2. Pin1At Expression Pattern and Protein Localization (A) Temporal expression of Pin1At in wild-type plants grown under long days (LD) and short days (SD). Error bars indicate standard deviation. (B) Effect of vernalization on Pin1At expression. For vernalization treatment, seeds were sown on Murashige and Skoog (MS) agar plates and incubated at 4 C under low light levels for 8 weeks. The relative expression of FLC, SOC1, and Pin1At in 9-day-old seedlings grown in LDs was compared. Quantitative real-time PCR data (A and B) were normalized against the TUB2 expression. Error bars indicate standard deviation. (C–F) In situ localization of Pin1At expression. Sections of shoot apices of 11-day-old (C) and 35-day-old (E) plants were hybridized with Pin1At antisense probe, while the corresponding control sections (D and F) were hybridized with the sense probe. Scale bars, 25 mm. (G–J) Subcellular localization of Pin1At-GFP fusion protein in Arabidopsis protoplasts. GFP localization is observed in protoplasts transfected with 35S:Pin1At-GFP (G) and a control 35S:GFP construct (I). Their respective chlorophyll red autofluorescence is also shown (H and J).
affect early flowering conferred by constitutive expression of CO and FT (Figure 1G). On the contrary, early flowering of 35S:SOC1 and 35S:AGL24 was delayed in Pin1At-AS background (Figure 1G). These observations imply that promotion of flowering by SOC1 and AGL24 is partially dependent on Pin1At. We further crossed 35S:Pin1At with several late-flowering loss-of-function mutants (Figure 1H). 35S:Pin1At promoted flowering in co-1 mutants, indicating that Pin1At could act downstream of or in parallel with CO. Comparatively, 35S:Pin1At
only slightly promoted flowering in soc1-2 and agl24-1 mutants. In particular, early flowering of 35S:Pin1At was completely suppressed by soc1-2 agl24-1 (Figure 1H), suggesting that 35S:Pin1At function is mediated by AGL24 and SOC1. We further found that the mRNA expression of AGL24 and SOC1 was not affected by suppression or overexpression of Pin1At (Figure S6A). Pin1At mRNA expression was also not significantly affected by AGL24 and SOC1 (Figure S6B). These observations demonstrate that Pin1At and flowering time regulators SOC1 and AGL24 are mutually dependent for their function in flowering, which is, however, not mediated by mutual transcriptional regulation. Pin1At mRNA was expressed in all the tissues examined, with the highest expression in cauline and rosette leaves (Figure S7). Pin1At transcripts were localized in the shoot apical meristem during the floral transition (Figures 2C and 2D) and later in the inflorescence shoot apical meristem and emerging floral meristems (Figures 2E and 2F). Expression of AGL24 and SOC1 was also shown to be upregulated in the shoot apical meristem during the floral transition (Michaels et al., 2003; Yu et al., 2002; Lee et al., 2000; Samach et al., 2000), which is concomitant with the Pin1At expression, indicating that they could interact to control flowering time. We then investigated the subcellular localization of Pin1At by Arabidopsis protoplast transfection assay using a fusion construct of Pin1At-GFP. Like 35S:GFP, Pin1At-GFP localized in both the cytoplasm and nucleus (Figures 2G–2J), implying that Pin1At may exert its function in multiple cell organelles. AGL24 and SOC1 Are Phosphorylated during Plant Development To investigate whether AGL24 and SOC1 interact with Pin1At, we first tested their phosphorylation status, because PIN1-type isomerases bind only to phosphorylated substrates (Hani et al., 1999; Yaffe et al., 1997). Immunoblot analysis of protein extracts from 35S:AGL24-6HA, where AGL24-6HA retains the same
114 Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc.
Molecular Cell Protein Isomerization Mediates Flowering
biological function as AGL24 in promoting flowering (Liu et al., 2008), revealed two specific bands recognized by HA antibody (Figure 3A). The lower faint band with a molecular mass of approximately 31 kDa was consistent with the predicted size of the fusion protein AGL24-6HA. As phosphorylated proteins often have slower mobility on an SDS-PAGE gel, the dominant upper band was probably the phosphorylated form of AGL246HA. To test this, 35S:AGL24-6HA protein extracts were treated with calf intestinal alkaline phosphatase (CIAP) for different time points. Immunoblot analysis showed that the amount of the major (upper) band was gradually reduced (Figure 3C), resulting in the increased intensity of the lower band. This suggests that the slow- and fast-migrating bands represent the phosphorylated and nonphosphorylated forms of AGL24-6HA, respectively. Similarly, immunoblot analysis of protein extracts from 35S:SOC1-9Myc, where SOC1-9Myc promotes flowering as SOC1 (Liu et al., 2008), also showed two bands with different mobility (Figure 3B). After CIAP treatment, the intensity of the slow- and fast-migrating bands was reduced and increased (Figure 3D), respectively, indicating that SOC1-9Myc proteins also exist as the phosphorylated and nonphosphorylated forms. We further performed immunoblot analysis to examine the temporal phosphorylation status of AGL24 and SOC1 in developing plants (Figure 3E). The phosphorylated form of AGL246HA was prominently expressed in 35S:AGL24-6HA over the developmental stages from young vegetative plants (day 4) to bolting plants (day 28). The phosphorylated form of SOC19Myc was highly expressed in 35S:SOC1-9Myc at all stages examined, while the amount of its unphosphorylated form gradually increased. Thus, phosphorylation of AGL24 and SOC1 proteins occurs in developing plants, indicating that their function is likely regulated by the phosphorylation/dephosphorylation mechanism. Pin1At Interacts with Phosphorylated SOC1 and AGL24 To examine if Pin1At directly interacts with SOC1 and AGL24, we performed GST pull-down assays. The phosphorylated AGL246HA and SOC1-9Myc from 35S:AGL24-6HA and 35S:SOC19Myc plants, respectively, were pulled down by GST-Pin1At, but not by GST control beads (Figures 3F and 3G). Because the amount of AGL24-6HA and SOC1-9Myc unphosphorylated forms was low in the plant extracts, CIAP-treated proteins were further tested if Pin1At also interacts with the unphosphorylated proteins (Figures 3H and 3I). Although the unphosphorylated proteins increased significantly after CIAP treatment, they were not pulled down by GST-Pin1At. These results indicate that Pin1At specifically interacts with the phosphorylated SOC1 and AGL24 in vitro. Furthermore, bimolecular fluorescence complementation (BiFC) analysis (Ohad et al., 2007), which examines protein-protein interactions through monitoring the fluorescence emitted by reconstitution of an enhanced yellow fluorescent protein from two fragments (N- and C-terminal halves) fused to two interacting proteins, revealed the direct interaction of Pin1At-AGL24 (Figure 3J) and Pin1AtSOC1 (Figure 3K) in the nuclei of living plant cells. Coimmunoprecipitation analysis on the protein extracts from 35S:AGL24-6HA 35S:Pin1At further confirmed the in vivo interaction of AGL24 and Pin1At (Figure 3L).
As PIN1-type isomerases specifically recognize phosphorylated Ser/Thr-Pro motifs, we examined whether phosphorylation of AGL24 and SOC1 occurs at Ser/Thr-Pro motifs using the mitotic phosphoprotein monoclonal-2 (MPM-2) antibody that specifically recognizes phosphorylated Ser/Thr-Pro motifs in many phosphoproteins (Davis et al., 1983). MPM-2 antibody detected the phosphorylated forms of AGL24-6HA and SOC19Myc, but not the unphosphorylated forms resulting from CIAP treatment (Figures 3M and 3N), suggesting that the Thr-Pro motif in AGL24 and Ser-Pro motifs in SOC1 are phosphorylated. Pin1At Catalyzes the Conformational Change of Ser/Thr-Pro Motifs in AGL24 and SOC1 We further used nuclear magnetic resonance (NMR) spectroscopy to examine if Pin1At catalyzes the cis/trans conformational exchange of the phosphorylated Ser/Thr-Pro motifs in SOC1 and AGL24 at per residue resolution. To this end, we synthesized the phosphorylated/nonphosphorylated peptides—SLIIF(p)SPKGKLYE, which were derived from one of the two Ser-Pro motifs in SOC1, and SSYDSG(p)TPLEDDSD, which were derived from the only Thr-Pro motif in AGL24—and determined their conformational exchanges in response to Pin1At. Because of slow exchange between proline cis and trans conformations, many residues in both phosphorylated and nonphosphorylated peptides displayed two distinct sets of 1H signals in the ROESY (rotating frame Overhauser effect spectroscopy) and TOCSY (total correlation spectroscopy) spectra (Figures 4A, 4B, S8, and S9). Only partial assignment of the peptide signals was obtained on the basis of the TOCSY and ROESY experiments due to severe signal overlap. The cis and trans populations of the peptides, regardless of their phosphorylation status, were 10% and 90%, respectively, as estimated from the intensities of well-resolved peaks in the one-dimensional 1H spectra. In the absence of Pin1At, no exchange cross peaks were observed in the ROESY spectra of the phosphorylated AGL24 and SOC1 peptides (Figures 4A and S8A), indicating that the exchange between the cis and trans conformations is too slow on the NMR timescale (<0.1 s1). In contrast, in the presence of Pin1At, we detected the exchange peaks resulting from proline isomerization in AGL24- and SOC1-phosphorylated peptides (Figures 4B and S8B). Taking the intensities of cross and diagonal peaks of Leu204 and Thr202 amide protons from the phosphorylated AGL24 peptides, we further calculated the isomerization rates from cis to trans (kctcat) and from trans to cis (ktccat) conformations (Figures 4C and 4D). The values of kctcat and ktccat obtained from Itt and Itc intensities for Thr202 were 8.7 and 0.8 s1, respectively, while the respective values for Leu204 were 9.8 and 1.3 s1 (Figures 4E and 4F). The results measured from two different residues were consistent and revealed that the exchange rate from the cis to trans conformation (kctcat) was about 8–10 times larger than that from the trans to cis conformations (ktccat), which was expected from the trans and cis population ratio (9:1). The isomerization rates catalyzed by Pin1At (kctcat and ktccat) were significantly higher than those without Pin1At (kct and ktc) (Figure 4F), demonstrating that isomerization of phosphorylated AGL24 peptides was dramatically enhanced by Pin1At (Figure 4G). On the contrary, the nonphosphorylated peptides (both AGL24 and SOC1) displayed no exchange peaks
Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc. 115
Molecular Cell Protein Isomerization Mediates Flowering
Figure 3. Phosphorylation of AGL24 and SOC1 and Their Interaction with Pin1At (A) AGL24-6HA proteins exist as two forms (arrowhead and arrow) in 35S:AGL24-6HA. Total proteins extracted from 2-week-old plants were analyzed by immunoblot with anti-HA antibody. (B) SOC1-9Myc proteins exist as two forms (arrowhead and arrow) in 35S:SOC1-9Myc. Total proteins extracted from 2-week-old plants were analyzed by immunoblot with anti-Myc antibody. (C) CIAP treatment of proteins immunoprecipitated from 35S:AGL24-6HA. After CIAP treatment, phosphorylated AGL24-6HA protein (arrowhead) was converted into the unphosphorylated form (arrow). (D) CIAP treatment of proteins immunoprecipitated from 35S:SOC1-9Myc. After CIAP treatment, phosphorylated SOC1-9Myc protein (arrowhead) was converted into the unphosphorylated form (arrow). (E) Temporal status of AGL24 and SOC1 phosphorylation in 4- to 28-day-old 35S:AGL24-6HA and 35S:SOC1-9Myc plants, respectively. (F and G) GST pull-down assay of interaction of Pin1At with phosphorylated AGL24-6HA (F) and SOC1-9Myc (G). GST or GST-Pin1At beads were incubated with the protein extracts from 35S:AGL24-6HA or 35S:SOC1-9Myc plants. The beads were subjected to immunoblot analysis with anti-HA (F) or anti-Myc (G) antibody in the upper panel. Arrowhead and arrow indicate the phosphorylated and unphosphorylated form, respectively. (H and I) GST pull-down assay of interaction of Pin1At with unphosphorylated AGL24-6HA (H) and SOC1-9Myc (I). The proteins immunoprecipitated from 35S:AGL24-6HA or 35S:SOC1-9Myc using anti-HA or anti-Myc, respectively, were pretreated with CIAP and incubated with GST or GST-Pin1At beads. The beads were subsequently analyzed as in (F and G).
116 Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc.
Molecular Cell Protein Isomerization Mediates Flowering
Figure 4. Pin1At Catalyzes Cis/Trans Isomerization of pThr-Pro Motifs in AGL24 (A and B) Selected region of two-dimensional ROESY spectra of the phosphorylated AGL24 peptide in the absence (A) or presence (B) of Pin1At at a mixing time of 110 ms. Negative and positive peaks are displayed in red and blue, respectively. Diagonal peaks from cis and trans conformers are indicated by cc and tt, respectively, while exchange peaks resulting from Pin1At-catalyzed isomerization are labeled by ct and tc. Note that ROE and exchange cross peaks are positive and negative, respectively. (C and D) Dependence of ratios of cross/diagonal peak intensities for cis and trans Leu204-HN (C) and Thr202HN (D) on ROE mixing times. Experimental data are fit to a two-site exchange model, and the fitting curves are indicated by solid lines. (E) Enzyme kinetic scheme of Pin1At catalytic isomerization of the phosphorylated Thr202-Pro bond of AGL24. (F) Pin1At-catalyzed isomerization rates from cis to trans (kctcat) and from trans to cis (ktccat) generated from independent fitting of the intensity ratios of ROESY peaks from (C) and (D). (G) Structural model represents the catalytic isomerization of Pin1At on phosphorylated Thr202-Pro bond of AGL24. In the absence of Pin1At (in blue), the conformational changes between cis (trans) and trans (cis) are very slow, while in the presence of Pin1At (in red), the conformational changes of prolyl-peptidyl bond of AGL24 are dramatically accelerated.
in both the presence and absence of Pin1At (Figure S9). These findings, together with the protein interaction results shown by GST pull-down and BiFC assays (Figures 3F to 3K), suggest
that Pin1At binds to the phosphorylated Ser/ Thr-Pro motifs in AGL24 and SOC1 and catalyzes their cis/trans isomerization. Compared to their homologs in other species, plant PIN1 homologs contain an extra four amino acid insertion in their PPIase domains (Figure S1). Deletion of these four amino acids has been shown to abolish the ability of Pin1At to rescue the lethal phenotype of yeast Ess1 mutation under nonoverexpression conditions, indicating its critical role in mediating the substrate interaction of Pin1At (Yao et al., 2001). We thus constructed the Pin1At mutant protein (mPin1At) with deletion of these four amino acids to test their function in mediating Pin1At enzyme activity. While GST-mPin1At was still able to bind to the Myc-tagged AGL24 and SOC1 (Figure S10), NMR analysis revealed that there was no isomerization of AGL24 peptide by mPin1At (Figure S11), suggesting that these plant-specific amino acids are essential for the enzyme activity of Pin1At.
(J and K) BiFC analysis of the interaction between Pin1At and AGL24 (J) or SOC1 (K). DAPI, fluorescence of 40 ,6-diamino-2-phenylindol; EYFP, fluorescence of enhanced yellow fluorescent protein; Merge, merge of DAPI and EYFP. (L) In vivo interaction between Pin1At and AGL24-6HA. Nuclear extracts from 35S:Pin1At 35S:AGL24-6HA double transgenic plants were immunoprecipitated by either anti-Myc or anti-HA agarose beads. The coimmunoprecipitated proteins were analyzed using anti-Pin1At antibody. (M and N) Phosphorylation of AGL24 and SOC1 on the Ser/Thr-Pro motifs. The proteins immunoprecipitated from 35S:AGL24-6HA (M) or 35S:SOC1-9Myc (N) were treated with or without CIAP. The resulting products were subjected to immunoblot analysis with anti-HA (M) or anti-Myc (N) antibody in the upper panel and with the MPM-2 antibody in the lower panels (M and N).
Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc. 117
Molecular Cell Protein Isomerization Mediates Flowering
Figure 5. Mutation of pSer/Thr-Pro Motifs in AGL24 and SOC1 Affects Flowering Time (A) Schematic diagram of the mutations in AGL24 and SOC1 proteins. There are four characteristic domains of MADS-box transcription factors, including MADS-box (M), intervening (I), keratin-like (K), and C-terminal (C) domains. Arrows indicate the sites containing pSer/Thr-Pro motifs. The mutated amino acids are indicated by asterisks. (B) Distribution of flowering time in T1 transgenic plants carrying the wildtype AGL24 gene (gAGL24) and its mutated form (gAGL24-AP) in the agl24-1 mutant background. (C) Distribution of flowering time in T1 transgenic plants carrying the wildtype SOC1 gene (gSOC1) and its mutated form (gSOC1-2AP) in the soc12 mutant background. (D) Flowering time of 35S:Pin1At is dependent on the pSer/Thr-Pro motifs in AGL24 and SOC1. Values presenting the mean ± standard deviation were scored from at least 15 plants of each genotype. Significant differences in comparison with wild-type plants are indicated with asterisks: *p < 0.05, by Student’s t test.
To test whether this four amino acid insertion is relevant to Pin1At function in the control of flowering time, we have overexpressed the mutated Pin1At in Arabidopsis and obtained 25 transgenic plants. All of these transgenic plants showed similar flowering time as wild-type plants under LDs (data not shown), which is different from the early flowering phenotypes exhibited by overexpression of wild-type Pin1At (Figure 1E), further indicating that Pin1At function in flowering time control depends on its isomerization function.
Mutagenesis of Ser/Thr-Pro Motifs in SOC1 and AGL24 Affects Flowering Time If Pin1At effect on flowering is mediated through its interaction with phosphorylated Ser/Thr-Pro motifs in SOC1 and AGL24, mutagenesis of these motifs would affect flowering time. Thus, we substituted Thr202 with Ala in the Thr-Pro motif of AGL24 and Ser49 and Ser195 with Ala in the Ser-Pro motifs of SOC1 in their respective genomic constructs (Figure 5A). Previous studies have demonstrated that the native AGL24 and SOC1 genomic fragments used in this study are able to largely rescue late flowering of their respective loss-offunction mutants (Liu et al., 2008; Samach et al., 2000). The average flowering time of agl24-1 or soc1-2 mutants transformed with their respective native genomic fragments was around 14 rosette leaves, which was comparable with that of wild-type plants (12 rosette leaves) (Figures 5B and 5C). However, agl24-1 mutants transformed with the mutated Thr-Pro motif construct, gAGL24-AP, and soc1-2 mutants transformed with the mutated Ser-Pro motif construct, gSOC1-2AP, flowered much later with around 18 and 22 rosette leaves, respectively (Figures 5B and 5C), demonstrating that mutation of Ser/Thr-Pro motifs in AGL24 and SOC1 delays flowering. To study if the mutations in Ser/Thr-Pro motifs affect the ability of AGL24 and SOC1 as substrates of Pin1At-mediated isomerization, we further tested whether Pin1At could catalyze the isomerization of the mutated Thr-Pro motif in AGL24. To this end, we synthesized a peptide, SSYDSGAPLEDDSD, in which TP was mutated into AP, and examined its conformational change by NMR spectroscopy. No conformational exchange was observed in the absence or presence of Pin1At (Figure S12), suggesting that the Thr-Pro motif is essential for AGL24 to serve as a Pin1At substrate. To confirm that Pin1At affects flowering through these Ser/ Thr-Pro motifs, we further crossed the representative transgenic lines (agl24 + gAGL24, agl24 + gAGL24-AP, soc1 + gSOC1, and soc1 + gSOC1-2AP) that showed an average flowering time of their respective populations with 35S:Pin1At agl24-1 and 35S:Pin1At soc1-2. While overexpression of Pin1At promotes
118 Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc.
Molecular Cell Protein Isomerization Mediates Flowering
Figure 6. Pin1At Regulates Flowering Time by Mediating AGL24 Protein Stability (A) AGL24 protein level in the nucleus is affected by Pin1At. Western blot analysis using anti-AGL24 antibody detects AGL24 protein expression (arrowhead) only in the nuclear fraction extracted from 2-weekold wild-type, Pin1At-AS, and 35S:Pin1At plants. * indicates a nonspecific signal. The purity of the nuclear and cytosolic fraction is examined by western blot analysis using anti-Histone 3 antibody, as previously published (Lee et al., 2008). (B) Nuclear AGL24 protein exists mainly as the phosphorylated form. After CIAP treatment of proteins in the nuclear fraction of wild-type plants, phosphorylated AGL24 protein (arrowhead) is converted into the unphosphorylated form (arrow). (C) Pin1At increases endogenous AGL24 protein stability. Wild-type, Pin1At-AS, and 35S:Pin1At plants were germinated and grown on MS solid medium for 2 weeks before being transferred to MS liquid medium supplemented with 100 mM cycloheximide. After treatment for the indicated times, the seedlings were harvested for western blot analysis using anti-AGL24 antibody. * indicates a nonspecific signal, which also serves as a loading control. The relative AGL24 protein levels were quantified as shown in the lower panel. (D) A working model of regulation of floral transition by Pin1At. In response to environmental signals, the vernalization and photoperiod pathways regulate the transcriptional levels of Pin1At. Pin1At interacts with the phosphorylated AGL24 and SOC1 and accelerates the cis/trans conformational change of phosphorylated Ser/Thr-Pro motifs of AGL24 and SOC1. Pin1At regulates flowering time through mediating the function of AGL24 and SOC1, which includes the control of the nuclear AGL24 stability.
flowering, this effect was compromised by the mutations of Ser/ Thr-Pro motifs in SOC1 and AGL24 (Figure 5D), thus corroborating that the Ser/Thr-Pro motifs in AGL24 and SOC1 are indispensable for Pin1At function in promoting flowering. Pin1At Mediates AGL24 Protein Stability Similar to the expression levels of AGL24 and SOC1 (Figure S6A), the localization of AGL24 and SOC1 in the shoot apical meristem during floral transition was also not affected by suppression or overexpression of Pin1At (Figure S13), further suggesting that Pin1At does not regulate AGL24 and SOC1 at the transcriptional level. To elucidate how Pin1At affects the function of AGL24 in the control of flowering time, we examined AGL24 protein expression in the nuclear and cytosolic fractions extracted from 2-week-old wild-type, Pin1At-AS, and 35S: Pin1At plants by western blot analysis using a specific antiAGL24 antibody (Liu et al., 2009). AGL24 protein expression was detected only in the nuclear fraction, with a higher level in 35S:Pin1At and a lower level in Pin1At-AS (Figure 6A), suggesting that Pin1At activity regulates phosphorylated AGL24 protein accumulation in the nucleus. As CIAP treatment of the nuclear protein from wild-type plants resulted in a shift of the upper band to a lower band with an expected size of AGL24 (Figure 6B), the upper band detected by AGL24 antibody before CIAP treatment should be the phosphorylated form of AGL24. This demonstrates that the endogenous AGL24 protein exists dominantly as the phosphorylated form, which is consistent with the observation on AGL24-6HA in 35S:AGL24-6HA (Figures 3A and 3E).
As human PIN1 regulates the turnover of many interacting proteins (Lu and Zhou, 2007), it is possible that Pin1At regulates the AGL24 level through mediating its protein stability. To test this possibility, we treated 2-week-old seedlings with a protein synthesis inhibitor, cycloheximide, and found that AGL24 had a half-life of about 2–3 hr in the absence of new protein synthesis (Figure 6C). Downregulation of Pin1At in Pin1At-AS plants decreased AGL24 protein stability with a half-life of less than 2 hr, whereas upregulation of Pin1At in 35S:Pin1At increased AGL24 half-life to 4–6 hr, suggesting that Pin1At indeed interacts with the phosphorylated AGL24 to regulate its stability in the nucleus. DISCUSSION The transition from vegetative to reproductive growth in plants is regulated by multiple flowering pathways in response to various developmental and environmental signals. This dynamic process involves spatial and temporal regulation of intermolecular interactions and enzymatic reactions. Although a number of flowering time regulators have been reported, whether and how protein conformational change affects floral transition remains unknown. In this study, we have found that the Arabidopsis PPIase, Pin1At, controls flowering through modulating cis/trans isomerization of the phosphorylated Ser/Thr-Pro motifs in two MADS-domain flowering regulators, SOC1 and AGL24 (Figure 6D). Because SOC1 and AGL24 synergistically integrate environmental and endogenous signals from several flowering pathways (Lee
Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc. 119
Molecular Cell Protein Isomerization Mediates Flowering
et al., 2000, 2008; Liu et al., 2008; Michaels et al., 2003; Samach et al., 2000; Yu et al., 2002), conformational regulation of these two proteins by Pin1At after phosphorylation delineates a hitherto unknown signaling mechanism that mediates the transition from the vegetative to reproductive growth in Arabidopsis. Extensive studies on human PIN1 have suggested that PIN1catalyzed isomerization of phosphorylated substrates has a profound impact on the function of proteins involved in many cellular processes (Liou et al., 2002; Lu, 2004; Lu and Zhou, 2007). For example, human PIN1 accelerates conformational change of the amyloid precursor protein (APP), thus regulating APP processing and amyloid-b production (Pastorino et al., 2006). The results presented here provide the evidence that PIN1-catalyzed isomerization of key transcription factors plays an indispensable role in regulating developmental transitions in plants. The conformational dynamics of AGL24 and SOC1 are tightly regulated by Pin1At. When Pin1At is downregulated, isomerization of these proteins from cis to trans and trans to cis forms is decelerated. Under this circumstance, if the cis or trans isomers are depleted in the cell, restoration of the equilibrium between cis and trans isomers would be delayed. However, if Pin1At is overexpressed, the isomerization rate is markedly increased, resulting in a rapid restoration of the equilibrium between cis or trans isomers of these proteins (Figure 4). The fact that Pin1At promotes flowering indicates that the accelerated isomerization of phosphorylated (Ser/Thr)-Pro motifs in AGL24 and SOC1 facilitates flowering, whereas the decelerated isomerization delays this transition. This is in line with the observation that Pin1At increases the stability of nuclear AGL24 protein that is a dosage-dependent flowering promoter (Michaels et al., 2003; Yu et al., 2002), suggesting that isomerization of phosphorylated AGL24 by Pin1At at least regulates AGL24 protein stability in the nucleus. Although we have not yet been able to examine SOC1 protein stability due to unsuccessful production of its specific antibodies, the similar molecular features of AGL24 and SOC1 and the analogous protein interaction between SOC1 and Pin1At imply that Pin1At may regulate the function of SOC1 like AGL24. Both SOC1 and AGL24 are MADS-domain proteins, which belong to a group of conserved transcription factors that play fundamental roles in developmental control and signal transduction in eukaryotes (Ng and Yanofsky, 2001; Riechmann and Meyerowitz, 1997; Theissen, 2000). Our findings on conformational regulation of SOC1 and AGL24 by Pin1At not only reveal an essential regulatory mechanism in plant development, but also shed light on the identification of PIN1 substrates and their relevant biological processes in animals. While this study suggests that prolyl cis/trans isomerases are evolutionarily conserved enzymes in plants and animals, how the postphosphorylation regulation mediated by these isomerases evolves to meet specific developmental needs in different organisms remains an intriguing question. EXPERIMENTAL PROCEDURES Plant Materials Seeds of Arabidopsis thaliana ecotype Columbia or Landsberg erecta were grown in LDs (16 hr light/8 hr dark) or SDs (8 hr light/16 hr dark) at 23 C ± 2 C.
The flowering time was measured by scoring the total number of rosette leaves after bolting. Construction of Binary Vectors To generate 35S:Pin1At construct, the full-length Pin1At cDNA fragment containing 44 bp 30 untranslated region was amplified using primers Pin1At-F1 and Pin1At-R1 (Table S1). The amplified fragment was digested and cloned into the pGreen 0229-35S binary vector (Yu et al., 2004). Based on this construct, we generated 35S:mPin1At, in which the four plant-specific amino acids (as shown in Figure S1) were deleted using the QuikChange II XL-Site-Directed Mutagenesis Kit (Stratagene; La Jolla, CA). To generate the Pin1At antisense construct, a 216 bp N-terminal fragment was amplified by Pin1At-F2 and Pin1At-R2 (Table S1). This fragment was also cloned into the pGreen 022935S binary vector. For the complementation test, the plasmids containing gAGL24 and gSOC1 were produced as previously described (Liu et al., 2008). Two genomic constructs, gAGL24-AP and gSOC1-2AP, which contain the mutations from Thr residue to Ala in AGL24 and Ser residue to Ala in SOC1, respectively, were further generated using the QuikChange II XL-Site-Directed Mutagenesis Kit (Stratagene). For the subcellular localization study, the Pin1At full-length cDNA was fused at the N terminus of GFP in the modified pGreen 0229-35S vector. Expression Analysis Quantitative real-time PCR was performed in triplicates on 7900HT Fast RealTime PCR system (Applied Biosystems; Foster City, CA) with SYBR Green PCR Master Mix (Applied Biosystems) (Liu et al., 2007). Semiquantitative PCR and nonradioactive in situ hybridization were performed as previously described (Yu et al., 2004). Primer sequences used for gene expression analysis were listed in Table S1. Protoplast Transfection Arabidopsis protoplast transfection by pGreen 0229-GFP and pGreen 0229Pin1At-GFP was performed as previously published (Sheen, 2001). The GFP localization and chlorophyll red autofluorescence were observed under the fluorescence light microscopy (Zeiss LSM510; Oberkochen, Germany). Alkaline Phosphatase Treatment of AGL24 and SOC1 Proteins Proteins were extracted from 2-week-old wild-type, 35S:AGL24-6HA, and 35S:SOC1-9Myc plants and immunoprecipitated with anti-AGL24, anti-HA, and anti-Myc antibodies (Santa Cruz Biotechnology; Santa Cruz, CA), respectively. After immunoprecipitation, the protein A/G beads (Santa Cruz Biotechnology) were washed twice with alkaline buffer (50 mM Tris-HCl, 1 mM MgCl2 [pH 8.8]) and then once with CIAP buffer with protease inhibitors. The beads were resuspended in CIAP buffer and incubated with CIAP at 37 C. Mocktreated control samples were incubated in the same volume of the reaction buffer without CIAP. The samples were subsequently spun for 1 min at 13,000 rpm to pellet the bead-bound complex. After removing the supernatant, 23 SDS protein sample buffer was added for further immunoblot analysis. BiFC Analysis Full-length cDNAs of SOC1, AGL24, and Pin1At were cloned into primary pSAT1 vectors (Tzfira et al., 2005). The resulting cassettes, including a fusion protein and a constitutive promoter, were cloned into pGreen binary vector HY105 and transformed into Agrobacterium. For BiFC experiments, 3-weekold tobacco (Nicotiana benthamiana) leaves were coinfiltrated with Agrobacterium as previously described (Sparkes et al., 2006). GST Pull-Down Assay GST-Pin1At fusion construct was prepared by cloning the coding region of Pin1At into the SpeI-XhoI sites of pET42a expression vector. Based on this construct, we generated the GST-mPin1At fusion construct using the same method for generating 35S:mPin1At. E. coli BL21 cells transformed with the plasmids were induced by IPTG, harvested, and lysed. After centrifugation, the supernatant was incubated with glutathione-agarose beads (Amersham Biosciences; Piscataway, NJ). The beads with the bound GST-Pin1At were washed for subsequent pull-down assays.
120 Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc.
Molecular Cell Protein Isomerization Mediates Flowering
NMR Analysis All NMR experiments were performed on a Bruker 800 MHz NMR spectrometer at 25 C. All spectra were recorded on peptide samples containing 20 mM phosphate buffer (90% H2O and 10% D2O [pH 6.5]) with or without Pin1At or mPin1At. The peptide concentration was 2.4 mM, and the Pin1At concentration was 0.03 or 0 mM for the samples of phosphorylated or nonphosphorylated AGL24 peptides (SSYDSG[p]TPLEDDSD) (GL Biochem; Shanghai, China), respectively. The peptide concentration was 0.15 mM, and the Pin1At concentration was 0.004 or 0 mM for the samples of phosphorylated and nonphosphorylated SOC1 peptides (SLIIF[p]SPKGKLYE), respectively, because these two peptides have relatively low solubility in aqueous solution. For all experiments, 256 3 512 complex points were acquired with spectral widths of 7200 3 9600 Hz in t1 3 t2 dimensions and an interscan delay of 1 s. For phosphorylated AGL24 peptide, ROESY spectra (Bax and Davis, 1985a) were acquired at a series of mixing times: 30, 45, 60, 75, 90, and 110 ms, and a spin-lock field strength of 4 kHz and 16 scans. For nonphosphorylated AGL24 peptide, one ROESY spectrum was recorded at a mixing time of 110 ms, a spin-lock field strength of 4 kHz, and 16 scans, while TOCSY experiments (Bax and Davis, 1985b) were carried out at a mixing time of 75 ms and 8 scans. For phosphorylated and nonphosphorylated SOC1 peptides, ROESY spectra were recorded at a mixing time of 110 ms, a spin-lock field strength of 4 kHz, and 96 scans, while TOCSY spectra were acquired at a mixing time of 75 ms and 64 scans. Ratios of cross to diagonal peak intensities for cis and trans conformations in the ROESY spectra depend on the forward (kctcat) and backward (ktccat) rate constants of a two-state exchange process (cis 4 trans) and the mixing times (tm). Changes of the ratios with the mixing times were fit using Equations 1 and 2 to determine the forward and backward rates (Pastorino et al., 2006): cat cat (1) Ict =Icc = kcat ct ½expðkex tm Þ 1= ktc expðkex tm Þ + kct cat cat Itc =Itt = kcat tc ½expðkex tm Þ 1= kct expðkex tm Þ + ktc
(2)
where Icc and Itt are the intensities of diagonal peaks for cis and trans conformers, Ict and Itc are the intensities of cross peaks from cis to trans and cat cat + ktc . from trans to cis conformers, and kex = kct SUPPLEMENTAL INFORMATION Supplemental Information includes one table and 13 figures and can be found with this article online at doi:10.1016/j.molcel.2009.12.020.
Davis, F.M., Tsao, T.Y., Fowler, S.K., and Rao, P.N. (1983). Monoclonal antibodies to mitotic cells. Proc. Natl. Acad. Sci. USA 80, 2926–2930. Hanes, S.D., Shank, P.R., and Bostian, K.A. (1989). Sequence and mutational analysis of ESS1, a gene essential for growth in Saccharomyces cerevisiae. Yeast 5, 55–72. Hani, J., Stumpf, G., and Domdey, H. (1995). PTF1 encodes an essential protein in Saccharomyces cerevisiae, which shows strong homology with a new putative family of PPIases. FEBS Lett. 365, 198–202. Hani, J., Schelbert, B., Bernhardt, A., Domdey, H., Fischer, G., Wiebauer, K., and Rahfeld, J.U. (1999). Mutations in a peptidylprolyl-cis/trans-isomerase gene lead to a defect in 30 -end formation of a pre-mRNA in Saccharomyces cerevisiae. J. Biol. Chem. 274, 108–116. He, Z., Li, L., and Luan, S. (2004). Immunophilins and parvulins. Superfamily of peptidyl prolyl isomerases in Arabidopsis. Plant Physiol. 134, 1248–1267. Hsu, T., McRackan, D., Vincent, T.S., and Gert de Couet, H. (2001). Drosophila Pin1 prolyl isomerase Dodo is a MAP kinase signal responder during oogenesis. Nat. Cell Biol. 3, 538–543. Huang, H.K., Forsburg, S.L., John, U.P., O’Connell, M.J., and Hunter, T. (2001). Isolation and characterization of the Pin1/Ess1p homologue in Schizosaccharomyces pombe. J. Cell Sci. 114, 3779–3788. Hunter, T. (1998). Prolyl isomerases and nuclear function. Cell 92, 141–143. Kardailsky, I., Shukla, V.K., Ahn, J.H., Dagenais, N., Christensen, S.K., Nguyen, J.T., Chory, J., Harrison, M.J., and Weigel, D. (1999). Activation tagging of the floral inducer FT. Science 286, 1962–1965. Kiefhaber, T., Quaas, R., Hahn, U., and Schmid, F.X. (1990). Folding of ribonuclease T1. 2. Kinetic models for the folding and unfolding reactions. Biochemistry 29, 3061–3070. Kobayashi, Y., Kaya, H., Goto, K., Iwabuchi, M., and Araki, T. (1999). A pair of related genes with antagonistic roles in mediating flowering signals. Science 286, 1960–1962. Koornneef, M., Alonso-Blanco, C., Peeters, A.J., and Soppe, W. (1998). Genetic Control of Flowering Time in Arabidopsis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 49, 345–370. Landrieu, I., De Veylder, L., Fruchart, J.S., Odaert, B., Casteels, P., Portetelle, D., Van Montagu, M., Inze´, D., and Lippens, G. (2000). The Arabidopsis thaliana PIN1At gene encodes a single-domain phosphorylation-dependent peptidyl prolyl cis/trans isomerase. J. Biol. Chem. 275, 10577–10581. Lee, H., Suh, S.S., Park, E., Cho, E., Ahn, J.H., Kim, S.G., Lee, J.S., Kwon, Y.M., and Lee, I. (2000). The AGAMOUS-LIKE 20 MADS domain protein integrates floral inductive pathways in Arabidopsis. Genes Dev. 14, 2366–2376.
ACKNOWLEDGMENTS We thank I. Lee, R. Amasino, M. Yanofsky, A. Samach, G. Coupland, and D. Weigel for providing seeds of various flowering mutants; S. Gelvin for providing pSAT1 vectors; S.Q. Yao and H.Y. Sun for providing some peptides; Y.C. Lai for protein expression; and B.C. Karthik for NMR analysis. We also thank T. Ito, J. Lim, and J.R. Dinneny for critical reading of the manuscript. This work was supported by a grant (06/1/21/19/473) from the Biomedical Research Council, the Agency for Science, Research and Technology, Singapore to Y.C.L. and a grant (T208B3113) from the Ministry of Education, Singapore and the intramural research funds from Temasek Life Sciences Laboratory to H.Y. L.C. was supported by the Singapore Millennium Foundation. Received: April 29, 2009 Revised: August 10, 2009 Accepted: October 14, 2009 Published: January 14, 2010
Lee, J., Oh, M., Park, H., and Lee, I. (2008). SOC1 translocated to the nucleus by interaction with AGL24 directly regulates leafy. Plant J. 55, 832–843. Levy, Y.Y., and Dean, C. (1998). The transition to flowering. Plant Cell 10, 1973–1990. Liou, Y.C., Ryo, A., Huang, H.K., Lu, P.J., Bronson, R., Fujimori, F., Uchida, T., Hunter, T., and Lu, K.P. (2002). Loss of Pin1 function in the mouse causes phenotypes resembling cyclin D1-null phenotypes. Proc. Natl. Acad. Sci. USA 99, 1335–1340. Liou, Y.C., Sun, A., Ryo, A., Zhou, X.Z., Yu, Z.X., Huang, H.K., Uchida, T., Bronson, R., Bing, G., Li, X., et al. (2003). Role of the prolyl isomerase Pin1 in protecting against age-dependent neurodegeneration. Nature 424, 556–561. Liu, C., Zhou, J., Bracha-Drori, K., Yalovsky, S., Ito, T., and Yu, H. (2007). Specification of Arabidopsis floral meristem identity by repression of flowering time genes. Development 134, 1901–1910.
REFERENCES
Liu, C., Chen, H., Er, H.L., Soo, H.M., Kumar, P.P., Han, J.H., Liou, Y.C., and Yu, H. (2008). Direct interaction of AGL24 and SOC1 integrates flowering signals in Arabidopsis. Development 135, 1481–1491.
Bax, A., and Davis, D.G. (1985a). 2D ROESY with cw spinlock for mixing phase sensitive using States-TPPI method. J. Magn. Reson. 63, 207–213.
Liu, C., Xi, W., Shen, L., Tan, C., and Yu, H. (2009). Regulation of floral patterning by flowering time genes. Dev. Cell 16, 711–722.
Bax, A., and Davis, D.G. (1985b). MLEV-17-based two-dimensional homonuclear magnetization transfer spectroscopy. J. Magn. Reson. 65, 355–360.
Lu, K.P. (2004). Pinning down cell signaling, cancer and Alzheimer’s disease. Trends Biochem. Sci. 29, 200–209.
Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc. 121
Molecular Cell Protein Isomerization Mediates Flowering
Lu, K.P., and Zhou, X.Z. (2007). The prolyl isomerase PIN1: a pivotal new twist in phosphorylation signalling and disease. Nat. Rev. Mol. Cell Biol. 8, 904–916. Lu, K.P., Hanes, S.D., and Hunter, T. (1996). A human peptidyl-prolyl isomerase essential for regulation of mitosis. Nature 380, 544–547. Lu, P.J., Wulf, G., Zhou, X.Z., Davies, P., and Lu, K.P. (1999). The prolyl isomerase Pin1 restores the function of Alzheimer-associated phosphorylated tau protein. Nature 399, 784–788. Lu, K.P., Finn, G., Lee, T.H., and Nicholson, L.K. (2007). Prolyl cis-trans isomerization as a molecular timer. Nat. Chem. Biol. 3, 619–629. Metzner, M., Stoller, G., Ru¨cknagel, K.P., Lu, K.P., Fischer, G., Luckner, M., and Ku¨llertz, G. (2001). Functional replacement of the essential ESS1 in yeast by the plant parvulin DlPar13. J. Biol. Chem. 276, 13524–13529. Michaels, S.D., Ditta, G., Gustafson-Brown, C., Pelaz, S., Yanofsky, M., and Amasino, R.M. (2003). AGL24 acts as a promoter of flowering in Arabidopsis and is positively regulated by vernalization. Plant J. 33, 867–874. Mouradov, A., Cremer, F., and Coupland, G. (2002). Control of flowering time: interacting pathways as a basis for diversity. Plant Cell 14 (Suppl ), S111–S130. Ng, M., and Yanofsky, M.F. (2001). Function and evolution of the plant MADSbox gene family. Nat. Rev. Genet. 2, 186–195. Ohad, N., Shichrur, K., and Yalovsky, S. (2007). The analysis of protein-protein interactions in plants by bimolecular fluorescence complementation. Plant Physiol. 145, 1090–1099. Pastorino, L., Sun, A., Lu, P.J., Zhou, X.Z., Balastik, M., Finn, G., Wulf, G., Lim, J., Li, S.H., Li, X., et al. (2006). The prolyl isomerase Pin1 regulates amyloid precursor protein processing and amyloid-beta production. Nature 440, 528–534. Ranganathan, R., Lu, K.P., Hunter, T., and Noel, J.P. (1997). Structural and functional analysis of the mitotic rotamase Pin1 suggests substrate recognition is phosphorylation dependent. Cell 89, 875–886. Riechmann, J.L., and Meyerowitz, E.M. (1997). MADS domain proteins in plant development. Biol. Chem. 378, 1079–1101.
Samach, A., Onouchi, H., Gold, S.E., Ditta, G.S., Schwarz-Sommer, Z., Yanofsky, M.F., and Coupland, G. (2000). Distinct roles of CONSTANS target genes in reproductive development of Arabidopsis. Science 288, 1613–1616. Sheen, J. (2001). Signal transduction in maize and Arabidopsis mesophyll protoplasts. Plant Physiol. 127, 1466–1475. Sparkes, I.A., Runions, J., Kearns, A., and Hawes, C. (2006). Rapid, transient expression of fluorescent fusion proteins in tobacco plants and generation of stably transformed plants. Nat. Protoc. 1, 2019–2025. Stukenberg, P.T., and Kirschner, M.W. (2001). Pin1 acts catalytically to promote a conformational change in Cdc25. Mol. Cell 7, 1071–1083. Theissen, G. (2000). Plant biology. Shattering developments. Nature 404, 711–713. Tzfira, T., Tian, G.W., Lacroix, B., Vyas, S., Li, J., Leitner-Dagan, Y., Krichevsky, A., Taylor, T., Vainstein, A., and Citovsky, V. (2005). pSAT vectors: a modular series of plasmids for autofluorescent protein tagging and expression of multiple genes in plants. Plant Mol. Biol. 57, 503–516. Yaffe, M.B., Schutkowski, M., Shen, M., Zhou, X.Z., Stukenberg, P.T., Rahfeld, J.U., Xu, J., Kuang, J., Kirschner, M.W., Fischer, G., et al. (1997). Sequencespecific and phosphorylation-dependent proline isomerization: a potential mitotic regulatory mechanism. Science 278, 1957–1960. Yao, J.L., Kops, O., Lu, P.J., and Lu, K.P. (2001). Functional conservation of phosphorylation-specific prolyl isomerases in plants. J. Biol. Chem. 276, 13517–13523. Yu, H., Xu, Y., Tan, E.L., and Kumar, P.P. (2002). AGAMOUS-LIKE 24, a dosage-dependent mediator of the flowering signals. Proc. Natl. Acad. Sci. USA 99, 16336–16341. Yu, H., Ito, T., Wellmer, F., and Meyerowitz, E.M. (2004). Repression of AGAMOUS-LIKE 24 is a crucial step in promoting flower development. Nat. Genet. 36, 157–161.
122 Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc.
Article Pin1At Encoding a Peptidyl-Prolyl cis/trans Isomerase Regulates Flowering Time in Arabidopsis Yu Wang,1,2 Chang Liu,1,2 Daiwen Yang,1 Hao Yu,1,2,* and Yih-Cherng Liou1,* 1Department
of Biological Sciences, Faculty of Science Life Sciences Laboratory, 1 Research Link National University of Singapore, Singapore *Correspondence: [email protected] (H.Y.), [email protected] (Y.-C.L.) DOI 10.1016/j.molcel.2009.12.020 2Temasek
SUMMARY
Floral transition in plants is regulated by an integrated network of flowering genetic pathways. We show that an Arabidopsis PIN1-type parvulin 1, Pin1At, controls floral transition by accelerating cis/trans isomerization of the phosphorylated Ser/ Thr-Pro motifs in two MADS-domain transcription factors, SUPPRESSOR OF OVEREXPRESSION OF CO 1 (SOC1) and AGAMOUS-LIKE 24 (AGL24). Pin1At regulates flowering, which is genetically mediated by AGL24 and SOC1. Pin1At interacts with the phosphorylated AGL24 and SOC1 in vitro and with AGL24 and SOC1 in vivo and accelerates the cis/trans conformational change of phosphorylated Ser/Thr-Pro motifs of AGL24 and SOC1. We further demonstrate that these Ser/Thr-Pro motifs are important for Pin1At function in promoting flowering through AGL24 and SOC1 and that the interaction between Pin1At and AGL24 mediates the AGL24 stability in the nucleus. Taken together, we propose that phosphorylation-dependent prolyl cis/trans isomerization of key transcription factors is an important flowering regulatory mechanism. INTRODUCTION Floral transition in Arabidopsis is coordinately controlled by multiple genetic pathways in response to various developmental and environmental cues (Koornneef et al., 1998; Levy and Dean, 1998; Mouradov et al., 2002). The autonomous and gibberellic acid (GA) pathways are responsive to endogenous developmental and physiological state, while the photoperiod and vernalization pathways monitor the alteration of environmental signals such as day length and temperature. These genetic pathways ultimately converge at the floral pathway integrators, which in turn regulate the activity of flower meristem identity genes to initiate the transition from vegetative to reproductive growth. Regulatory genes acting at the convergence point of the multiple floral induction pathways include two MADS-box transcription factors, SOC1 and AGL24 (Lee et al., 2000; Liu et al., 2008; Michaels et al., 2003; Samach et al., 2000; Yu et al., 2002).
Peptidyl-prolyl cis/trans isomerases (PPIases) are enzymes that accelerate energetically unfavorable cis/trans isomerization of the peptide bond preceding a proline (Hunter, 1998; Kiefhaber et al., 1990). PPIases include four structurally distinct subfamilies: cyclophilins, FK506-binding proteins, parvulins, and PP2A phosphatase activator (Lu et al., 2007). Members from the parvulin subfamily, such as PIN1 from human and ESS/PTF1 from Saccharomyces cerevisiae, have been shown to be essential for cell cycle and growth (Hanes et al., 1989; Hani et al., 1995; Lu et al., 1996). In the parvulin subfamily, PIN1-like PPIases are the only type of PPIases that specifically recognize phosphorylated Ser/Thr residues preceding proline (pSer/Thr-Pro) and catalyze the conformational change of the phosphorylated substrates, thereby controlling their function (Hsu et al., 2001; Huang et al., 2001; Liou et al., 2002, 2003; Lu et al., 1996, 1999; Pastorino et al., 2006; Ranganathan et al., 1997; Stukenberg and Kirschner, 2001; Yaffe et al., 1997). Pin1At was early-identified as the only PIN1-type PPIase from Arabidopsis (He et al., 2004; Landrieu et al., 2000). Sequence comparison showed that PIN1 homologs from human, yeast, and Drosophila consist of two domains, an N-terminal WW regulatory domain and a C-terminal PPIase catalytic domain, whereas plant PIN1s, including Pin1At, contain only a PPIase domain with four additional amino acids (Figure S1) (Yao et al., 2001). Although several Pin1 homologs have been identified in plants (Landrieu et al., 2000; Metzner et al., 2001; Yao et al., 2001), their substrates and biological function in plants are so far unknown. Here, we show that Pin1At affects floral transition in Arabidopsis and that phosphorylation-dependent prolyl cis/ trans isomerization of key transcription factors such as SOC1 and AGL24 by Pin1At emerges as an important flowering regulatory mechanism.
RESULTS Pin1At Promotes Flowering To investigate the biological function of Pin1At, we first examined the effect of modulating Pin1At expression in Arabidopsis. Since insertion mutants were not available in public resources, transgenic plants with antisense suppression of Pin1At were generated. Among 56 transgenic lines containing Pin1At N-terminal antisense fragment driven by a double cauliflower mosaic virus 35S promoter (Pin1At-AS), 30 plants showed late flowering under long days (LDs) (Figures 1A and 1B). Under
112 Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc.
Molecular Cell Protein Isomerization Mediates Flowering
Figure 1. Pin1At Regulates Flowering Time in Arabidopsis (A and B) Wild-type plants (A) show earlier flowering than representative Pin1At antisense plants (line 18-3) (B) at 35 days after germination under long days. (C and D) Wild-type plants (C) show later flowering than representative 35S:Pin1At plants (line 22-5) (D) at 28 days after germination under long days. (E and F) Flowering time of Pin1At transgenic plants under long days (E) and short days (F). The number of rosette leaves on the main shoot represents flowering time. Values presenting the mean ± standard deviation were scored from at least 15 plants of each genotype. Significant differences in comparison with wild-type plants are indicated with asterisks: *p < 0.05; **p < 0.01, by Student’s t test. (G) Effect of antisense suppression of Pin1At (line 18-3) on overexpression of various flowering promoters under long days. Error bars indicate standard deviation. Significant differences between with and without antisense suppression of Pin1At are indicated with asterisks: *p < 0.05, by Student’s t test. (H) Effect of overexpression of Pin1At (line 22-5) on various flowering mutants under long days. Error bars indicate standard deviation. Significant differences between with and without overexpression of Pin1At are indicated with asterisks: *p < 0.05, by Student’s t test.
both long and short days (SDs), representative Pin1At-AS plants with reduced expression of Pin1At flowered much later than wild-type plants (Figures 1E, 1F, and S2A). On the contrary, among 39 transgenic plants containing Pin1At cDNA under the control of 35S promoter (35S:Pin1At), 21 showed early flowering under LDs (Figures 1C and 1D). Representative 35S:Pin1At lines with overexpression of Pin1At showed early flowering under both LDs and SDs (Figures 1E, 1F, and S2B). In addition, these plants also exhibited serrated rosette and cauline leaves (Figure 1D). These observations suggest that Pin1At at least affects flowering and leaf development. The Photoperiod and Vernalization Pathways Affect Pin1At Expression To examine the flowering pathways that regulate Pin1At, we examined its expression in various environmental conditions and flowering mutants. In wild-type Col plants under LDs, Pin1At expression gradually increased during floral transition occurring from 9 to 13 days after germination (Figure 2A) and decreased afterwards. On the contrary, its expression was not significantly changed under SDs within 21 days after germination. These results suggest that the photoperiod pathway affects Pin1At
expression. Further examination of Pin1At expression revealed that its expression was affected by CONSTANS (CO), but not FLOWERING LOCUS T (FT) (Figure S3). Pin1At expression was also affected by vernalization (Figure 2B), but not by autonomous pathway mutants and GA treatment (Figures S4 and S5). Notably, dramatic downregulation of a flowering repressor, FLOWERING LOCUS C (FLC), in FRI FLC, wherein the dominant allele of FRIGIDA (FRI) causes high expression of FLC by vernalization, did not enhance the upregulation of Pin1At (Figure 2B), indicating that vernalization regulates Pin1At in an FLC-independent manner. Regulation of Pin1At by the vernalization pathway partly explains that manipulation of Pin1At expression affects flowering time in SDs, under which the LD photoperiod pathway is not activated. Pin1At Regulates Flowering Time through AGL24 and SOC1 Since the photoperiod and vernalization pathways converge on the regulation of Pin1At expression, we speculated that Pin1At function may be relevant to flowering regulators located downstream of several genetic pathways, such as SOC1, FT, and AGL24 (Kardailsky et al., 1999; Kobayashi et al., 1999; Lee et al., 2000; Michaels et al., 2003; Samach et al., 2000; Yu et al., 2002). Thus, we performed genetic crossing to identify potential interacting regulators of Pin1At. While downregulation of Pin1At in Pin1At-AS caused late flowering, it did not
Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc. 113
Molecular Cell Protein Isomerization Mediates Flowering
Figure 2. Pin1At Expression Pattern and Protein Localization (A) Temporal expression of Pin1At in wild-type plants grown under long days (LD) and short days (SD). Error bars indicate standard deviation. (B) Effect of vernalization on Pin1At expression. For vernalization treatment, seeds were sown on Murashige and Skoog (MS) agar plates and incubated at 4 C under low light levels for 8 weeks. The relative expression of FLC, SOC1, and Pin1At in 9-day-old seedlings grown in LDs was compared. Quantitative real-time PCR data (A and B) were normalized against the TUB2 expression. Error bars indicate standard deviation. (C–F) In situ localization of Pin1At expression. Sections of shoot apices of 11-day-old (C) and 35-day-old (E) plants were hybridized with Pin1At antisense probe, while the corresponding control sections (D and F) were hybridized with the sense probe. Scale bars, 25 mm. (G–J) Subcellular localization of Pin1At-GFP fusion protein in Arabidopsis protoplasts. GFP localization is observed in protoplasts transfected with 35S:Pin1At-GFP (G) and a control 35S:GFP construct (I). Their respective chlorophyll red autofluorescence is also shown (H and J).
affect early flowering conferred by constitutive expression of CO and FT (Figure 1G). On the contrary, early flowering of 35S:SOC1 and 35S:AGL24 was delayed in Pin1At-AS background (Figure 1G). These observations imply that promotion of flowering by SOC1 and AGL24 is partially dependent on Pin1At. We further crossed 35S:Pin1At with several late-flowering loss-of-function mutants (Figure 1H). 35S:Pin1At promoted flowering in co-1 mutants, indicating that Pin1At could act downstream of or in parallel with CO. Comparatively, 35S:Pin1At
only slightly promoted flowering in soc1-2 and agl24-1 mutants. In particular, early flowering of 35S:Pin1At was completely suppressed by soc1-2 agl24-1 (Figure 1H), suggesting that 35S:Pin1At function is mediated by AGL24 and SOC1. We further found that the mRNA expression of AGL24 and SOC1 was not affected by suppression or overexpression of Pin1At (Figure S6A). Pin1At mRNA expression was also not significantly affected by AGL24 and SOC1 (Figure S6B). These observations demonstrate that Pin1At and flowering time regulators SOC1 and AGL24 are mutually dependent for their function in flowering, which is, however, not mediated by mutual transcriptional regulation. Pin1At mRNA was expressed in all the tissues examined, with the highest expression in cauline and rosette leaves (Figure S7). Pin1At transcripts were localized in the shoot apical meristem during the floral transition (Figures 2C and 2D) and later in the inflorescence shoot apical meristem and emerging floral meristems (Figures 2E and 2F). Expression of AGL24 and SOC1 was also shown to be upregulated in the shoot apical meristem during the floral transition (Michaels et al., 2003; Yu et al., 2002; Lee et al., 2000; Samach et al., 2000), which is concomitant with the Pin1At expression, indicating that they could interact to control flowering time. We then investigated the subcellular localization of Pin1At by Arabidopsis protoplast transfection assay using a fusion construct of Pin1At-GFP. Like 35S:GFP, Pin1At-GFP localized in both the cytoplasm and nucleus (Figures 2G–2J), implying that Pin1At may exert its function in multiple cell organelles. AGL24 and SOC1 Are Phosphorylated during Plant Development To investigate whether AGL24 and SOC1 interact with Pin1At, we first tested their phosphorylation status, because PIN1-type isomerases bind only to phosphorylated substrates (Hani et al., 1999; Yaffe et al., 1997). Immunoblot analysis of protein extracts from 35S:AGL24-6HA, where AGL24-6HA retains the same
114 Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc.
Molecular Cell Protein Isomerization Mediates Flowering
biological function as AGL24 in promoting flowering (Liu et al., 2008), revealed two specific bands recognized by HA antibody (Figure 3A). The lower faint band with a molecular mass of approximately 31 kDa was consistent with the predicted size of the fusion protein AGL24-6HA. As phosphorylated proteins often have slower mobility on an SDS-PAGE gel, the dominant upper band was probably the phosphorylated form of AGL246HA. To test this, 35S:AGL24-6HA protein extracts were treated with calf intestinal alkaline phosphatase (CIAP) for different time points. Immunoblot analysis showed that the amount of the major (upper) band was gradually reduced (Figure 3C), resulting in the increased intensity of the lower band. This suggests that the slow- and fast-migrating bands represent the phosphorylated and nonphosphorylated forms of AGL24-6HA, respectively. Similarly, immunoblot analysis of protein extracts from 35S:SOC1-9Myc, where SOC1-9Myc promotes flowering as SOC1 (Liu et al., 2008), also showed two bands with different mobility (Figure 3B). After CIAP treatment, the intensity of the slow- and fast-migrating bands was reduced and increased (Figure 3D), respectively, indicating that SOC1-9Myc proteins also exist as the phosphorylated and nonphosphorylated forms. We further performed immunoblot analysis to examine the temporal phosphorylation status of AGL24 and SOC1 in developing plants (Figure 3E). The phosphorylated form of AGL246HA was prominently expressed in 35S:AGL24-6HA over the developmental stages from young vegetative plants (day 4) to bolting plants (day 28). The phosphorylated form of SOC19Myc was highly expressed in 35S:SOC1-9Myc at all stages examined, while the amount of its unphosphorylated form gradually increased. Thus, phosphorylation of AGL24 and SOC1 proteins occurs in developing plants, indicating that their function is likely regulated by the phosphorylation/dephosphorylation mechanism. Pin1At Interacts with Phosphorylated SOC1 and AGL24 To examine if Pin1At directly interacts with SOC1 and AGL24, we performed GST pull-down assays. The phosphorylated AGL246HA and SOC1-9Myc from 35S:AGL24-6HA and 35S:SOC19Myc plants, respectively, were pulled down by GST-Pin1At, but not by GST control beads (Figures 3F and 3G). Because the amount of AGL24-6HA and SOC1-9Myc unphosphorylated forms was low in the plant extracts, CIAP-treated proteins were further tested if Pin1At also interacts with the unphosphorylated proteins (Figures 3H and 3I). Although the unphosphorylated proteins increased significantly after CIAP treatment, they were not pulled down by GST-Pin1At. These results indicate that Pin1At specifically interacts with the phosphorylated SOC1 and AGL24 in vitro. Furthermore, bimolecular fluorescence complementation (BiFC) analysis (Ohad et al., 2007), which examines protein-protein interactions through monitoring the fluorescence emitted by reconstitution of an enhanced yellow fluorescent protein from two fragments (N- and C-terminal halves) fused to two interacting proteins, revealed the direct interaction of Pin1At-AGL24 (Figure 3J) and Pin1AtSOC1 (Figure 3K) in the nuclei of living plant cells. Coimmunoprecipitation analysis on the protein extracts from 35S:AGL24-6HA 35S:Pin1At further confirmed the in vivo interaction of AGL24 and Pin1At (Figure 3L).
As PIN1-type isomerases specifically recognize phosphorylated Ser/Thr-Pro motifs, we examined whether phosphorylation of AGL24 and SOC1 occurs at Ser/Thr-Pro motifs using the mitotic phosphoprotein monoclonal-2 (MPM-2) antibody that specifically recognizes phosphorylated Ser/Thr-Pro motifs in many phosphoproteins (Davis et al., 1983). MPM-2 antibody detected the phosphorylated forms of AGL24-6HA and SOC19Myc, but not the unphosphorylated forms resulting from CIAP treatment (Figures 3M and 3N), suggesting that the Thr-Pro motif in AGL24 and Ser-Pro motifs in SOC1 are phosphorylated. Pin1At Catalyzes the Conformational Change of Ser/Thr-Pro Motifs in AGL24 and SOC1 We further used nuclear magnetic resonance (NMR) spectroscopy to examine if Pin1At catalyzes the cis/trans conformational exchange of the phosphorylated Ser/Thr-Pro motifs in SOC1 and AGL24 at per residue resolution. To this end, we synthesized the phosphorylated/nonphosphorylated peptides—SLIIF(p)SPKGKLYE, which were derived from one of the two Ser-Pro motifs in SOC1, and SSYDSG(p)TPLEDDSD, which were derived from the only Thr-Pro motif in AGL24—and determined their conformational exchanges in response to Pin1At. Because of slow exchange between proline cis and trans conformations, many residues in both phosphorylated and nonphosphorylated peptides displayed two distinct sets of 1H signals in the ROESY (rotating frame Overhauser effect spectroscopy) and TOCSY (total correlation spectroscopy) spectra (Figures 4A, 4B, S8, and S9). Only partial assignment of the peptide signals was obtained on the basis of the TOCSY and ROESY experiments due to severe signal overlap. The cis and trans populations of the peptides, regardless of their phosphorylation status, were 10% and 90%, respectively, as estimated from the intensities of well-resolved peaks in the one-dimensional 1H spectra. In the absence of Pin1At, no exchange cross peaks were observed in the ROESY spectra of the phosphorylated AGL24 and SOC1 peptides (Figures 4A and S8A), indicating that the exchange between the cis and trans conformations is too slow on the NMR timescale (<0.1 s1). In contrast, in the presence of Pin1At, we detected the exchange peaks resulting from proline isomerization in AGL24- and SOC1-phosphorylated peptides (Figures 4B and S8B). Taking the intensities of cross and diagonal peaks of Leu204 and Thr202 amide protons from the phosphorylated AGL24 peptides, we further calculated the isomerization rates from cis to trans (kctcat) and from trans to cis (ktccat) conformations (Figures 4C and 4D). The values of kctcat and ktccat obtained from Itt and Itc intensities for Thr202 were 8.7 and 0.8 s1, respectively, while the respective values for Leu204 were 9.8 and 1.3 s1 (Figures 4E and 4F). The results measured from two different residues were consistent and revealed that the exchange rate from the cis to trans conformation (kctcat) was about 8–10 times larger than that from the trans to cis conformations (ktccat), which was expected from the trans and cis population ratio (9:1). The isomerization rates catalyzed by Pin1At (kctcat and ktccat) were significantly higher than those without Pin1At (kct and ktc) (Figure 4F), demonstrating that isomerization of phosphorylated AGL24 peptides was dramatically enhanced by Pin1At (Figure 4G). On the contrary, the nonphosphorylated peptides (both AGL24 and SOC1) displayed no exchange peaks
Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc. 115
Molecular Cell Protein Isomerization Mediates Flowering
Figure 3. Phosphorylation of AGL24 and SOC1 and Their Interaction with Pin1At (A) AGL24-6HA proteins exist as two forms (arrowhead and arrow) in 35S:AGL24-6HA. Total proteins extracted from 2-week-old plants were analyzed by immunoblot with anti-HA antibody. (B) SOC1-9Myc proteins exist as two forms (arrowhead and arrow) in 35S:SOC1-9Myc. Total proteins extracted from 2-week-old plants were analyzed by immunoblot with anti-Myc antibody. (C) CIAP treatment of proteins immunoprecipitated from 35S:AGL24-6HA. After CIAP treatment, phosphorylated AGL24-6HA protein (arrowhead) was converted into the unphosphorylated form (arrow). (D) CIAP treatment of proteins immunoprecipitated from 35S:SOC1-9Myc. After CIAP treatment, phosphorylated SOC1-9Myc protein (arrowhead) was converted into the unphosphorylated form (arrow). (E) Temporal status of AGL24 and SOC1 phosphorylation in 4- to 28-day-old 35S:AGL24-6HA and 35S:SOC1-9Myc plants, respectively. (F and G) GST pull-down assay of interaction of Pin1At with phosphorylated AGL24-6HA (F) and SOC1-9Myc (G). GST or GST-Pin1At beads were incubated with the protein extracts from 35S:AGL24-6HA or 35S:SOC1-9Myc plants. The beads were subjected to immunoblot analysis with anti-HA (F) or anti-Myc (G) antibody in the upper panel. Arrowhead and arrow indicate the phosphorylated and unphosphorylated form, respectively. (H and I) GST pull-down assay of interaction of Pin1At with unphosphorylated AGL24-6HA (H) and SOC1-9Myc (I). The proteins immunoprecipitated from 35S:AGL24-6HA or 35S:SOC1-9Myc using anti-HA or anti-Myc, respectively, were pretreated with CIAP and incubated with GST or GST-Pin1At beads. The beads were subsequently analyzed as in (F and G).
116 Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc.
Molecular Cell Protein Isomerization Mediates Flowering
Figure 4. Pin1At Catalyzes Cis/Trans Isomerization of pThr-Pro Motifs in AGL24 (A and B) Selected region of two-dimensional ROESY spectra of the phosphorylated AGL24 peptide in the absence (A) or presence (B) of Pin1At at a mixing time of 110 ms. Negative and positive peaks are displayed in red and blue, respectively. Diagonal peaks from cis and trans conformers are indicated by cc and tt, respectively, while exchange peaks resulting from Pin1At-catalyzed isomerization are labeled by ct and tc. Note that ROE and exchange cross peaks are positive and negative, respectively. (C and D) Dependence of ratios of cross/diagonal peak intensities for cis and trans Leu204-HN (C) and Thr202HN (D) on ROE mixing times. Experimental data are fit to a two-site exchange model, and the fitting curves are indicated by solid lines. (E) Enzyme kinetic scheme of Pin1At catalytic isomerization of the phosphorylated Thr202-Pro bond of AGL24. (F) Pin1At-catalyzed isomerization rates from cis to trans (kctcat) and from trans to cis (ktccat) generated from independent fitting of the intensity ratios of ROESY peaks from (C) and (D). (G) Structural model represents the catalytic isomerization of Pin1At on phosphorylated Thr202-Pro bond of AGL24. In the absence of Pin1At (in blue), the conformational changes between cis (trans) and trans (cis) are very slow, while in the presence of Pin1At (in red), the conformational changes of prolyl-peptidyl bond of AGL24 are dramatically accelerated.
in both the presence and absence of Pin1At (Figure S9). These findings, together with the protein interaction results shown by GST pull-down and BiFC assays (Figures 3F to 3K), suggest
that Pin1At binds to the phosphorylated Ser/ Thr-Pro motifs in AGL24 and SOC1 and catalyzes their cis/trans isomerization. Compared to their homologs in other species, plant PIN1 homologs contain an extra four amino acid insertion in their PPIase domains (Figure S1). Deletion of these four amino acids has been shown to abolish the ability of Pin1At to rescue the lethal phenotype of yeast Ess1 mutation under nonoverexpression conditions, indicating its critical role in mediating the substrate interaction of Pin1At (Yao et al., 2001). We thus constructed the Pin1At mutant protein (mPin1At) with deletion of these four amino acids to test their function in mediating Pin1At enzyme activity. While GST-mPin1At was still able to bind to the Myc-tagged AGL24 and SOC1 (Figure S10), NMR analysis revealed that there was no isomerization of AGL24 peptide by mPin1At (Figure S11), suggesting that these plant-specific amino acids are essential for the enzyme activity of Pin1At.
(J and K) BiFC analysis of the interaction between Pin1At and AGL24 (J) or SOC1 (K). DAPI, fluorescence of 40 ,6-diamino-2-phenylindol; EYFP, fluorescence of enhanced yellow fluorescent protein; Merge, merge of DAPI and EYFP. (L) In vivo interaction between Pin1At and AGL24-6HA. Nuclear extracts from 35S:Pin1At 35S:AGL24-6HA double transgenic plants were immunoprecipitated by either anti-Myc or anti-HA agarose beads. The coimmunoprecipitated proteins were analyzed using anti-Pin1At antibody. (M and N) Phosphorylation of AGL24 and SOC1 on the Ser/Thr-Pro motifs. The proteins immunoprecipitated from 35S:AGL24-6HA (M) or 35S:SOC1-9Myc (N) were treated with or without CIAP. The resulting products were subjected to immunoblot analysis with anti-HA (M) or anti-Myc (N) antibody in the upper panel and with the MPM-2 antibody in the lower panels (M and N).
Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc. 117
Molecular Cell Protein Isomerization Mediates Flowering
Figure 5. Mutation of pSer/Thr-Pro Motifs in AGL24 and SOC1 Affects Flowering Time (A) Schematic diagram of the mutations in AGL24 and SOC1 proteins. There are four characteristic domains of MADS-box transcription factors, including MADS-box (M), intervening (I), keratin-like (K), and C-terminal (C) domains. Arrows indicate the sites containing pSer/Thr-Pro motifs. The mutated amino acids are indicated by asterisks. (B) Distribution of flowering time in T1 transgenic plants carrying the wildtype AGL24 gene (gAGL24) and its mutated form (gAGL24-AP) in the agl24-1 mutant background. (C) Distribution of flowering time in T1 transgenic plants carrying the wildtype SOC1 gene (gSOC1) and its mutated form (gSOC1-2AP) in the soc12 mutant background. (D) Flowering time of 35S:Pin1At is dependent on the pSer/Thr-Pro motifs in AGL24 and SOC1. Values presenting the mean ± standard deviation were scored from at least 15 plants of each genotype. Significant differences in comparison with wild-type plants are indicated with asterisks: *p < 0.05, by Student’s t test.
To test whether this four amino acid insertion is relevant to Pin1At function in the control of flowering time, we have overexpressed the mutated Pin1At in Arabidopsis and obtained 25 transgenic plants. All of these transgenic plants showed similar flowering time as wild-type plants under LDs (data not shown), which is different from the early flowering phenotypes exhibited by overexpression of wild-type Pin1At (Figure 1E), further indicating that Pin1At function in flowering time control depends on its isomerization function.
Mutagenesis of Ser/Thr-Pro Motifs in SOC1 and AGL24 Affects Flowering Time If Pin1At effect on flowering is mediated through its interaction with phosphorylated Ser/Thr-Pro motifs in SOC1 and AGL24, mutagenesis of these motifs would affect flowering time. Thus, we substituted Thr202 with Ala in the Thr-Pro motif of AGL24 and Ser49 and Ser195 with Ala in the Ser-Pro motifs of SOC1 in their respective genomic constructs (Figure 5A). Previous studies have demonstrated that the native AGL24 and SOC1 genomic fragments used in this study are able to largely rescue late flowering of their respective loss-offunction mutants (Liu et al., 2008; Samach et al., 2000). The average flowering time of agl24-1 or soc1-2 mutants transformed with their respective native genomic fragments was around 14 rosette leaves, which was comparable with that of wild-type plants (12 rosette leaves) (Figures 5B and 5C). However, agl24-1 mutants transformed with the mutated Thr-Pro motif construct, gAGL24-AP, and soc1-2 mutants transformed with the mutated Ser-Pro motif construct, gSOC1-2AP, flowered much later with around 18 and 22 rosette leaves, respectively (Figures 5B and 5C), demonstrating that mutation of Ser/Thr-Pro motifs in AGL24 and SOC1 delays flowering. To study if the mutations in Ser/Thr-Pro motifs affect the ability of AGL24 and SOC1 as substrates of Pin1At-mediated isomerization, we further tested whether Pin1At could catalyze the isomerization of the mutated Thr-Pro motif in AGL24. To this end, we synthesized a peptide, SSYDSGAPLEDDSD, in which TP was mutated into AP, and examined its conformational change by NMR spectroscopy. No conformational exchange was observed in the absence or presence of Pin1At (Figure S12), suggesting that the Thr-Pro motif is essential for AGL24 to serve as a Pin1At substrate. To confirm that Pin1At affects flowering through these Ser/ Thr-Pro motifs, we further crossed the representative transgenic lines (agl24 + gAGL24, agl24 + gAGL24-AP, soc1 + gSOC1, and soc1 + gSOC1-2AP) that showed an average flowering time of their respective populations with 35S:Pin1At agl24-1 and 35S:Pin1At soc1-2. While overexpression of Pin1At promotes
118 Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc.
Molecular Cell Protein Isomerization Mediates Flowering
Figure 6. Pin1At Regulates Flowering Time by Mediating AGL24 Protein Stability (A) AGL24 protein level in the nucleus is affected by Pin1At. Western blot analysis using anti-AGL24 antibody detects AGL24 protein expression (arrowhead) only in the nuclear fraction extracted from 2-weekold wild-type, Pin1At-AS, and 35S:Pin1At plants. * indicates a nonspecific signal. The purity of the nuclear and cytosolic fraction is examined by western blot analysis using anti-Histone 3 antibody, as previously published (Lee et al., 2008). (B) Nuclear AGL24 protein exists mainly as the phosphorylated form. After CIAP treatment of proteins in the nuclear fraction of wild-type plants, phosphorylated AGL24 protein (arrowhead) is converted into the unphosphorylated form (arrow). (C) Pin1At increases endogenous AGL24 protein stability. Wild-type, Pin1At-AS, and 35S:Pin1At plants were germinated and grown on MS solid medium for 2 weeks before being transferred to MS liquid medium supplemented with 100 mM cycloheximide. After treatment for the indicated times, the seedlings were harvested for western blot analysis using anti-AGL24 antibody. * indicates a nonspecific signal, which also serves as a loading control. The relative AGL24 protein levels were quantified as shown in the lower panel. (D) A working model of regulation of floral transition by Pin1At. In response to environmental signals, the vernalization and photoperiod pathways regulate the transcriptional levels of Pin1At. Pin1At interacts with the phosphorylated AGL24 and SOC1 and accelerates the cis/trans conformational change of phosphorylated Ser/Thr-Pro motifs of AGL24 and SOC1. Pin1At regulates flowering time through mediating the function of AGL24 and SOC1, which includes the control of the nuclear AGL24 stability.
flowering, this effect was compromised by the mutations of Ser/ Thr-Pro motifs in SOC1 and AGL24 (Figure 5D), thus corroborating that the Ser/Thr-Pro motifs in AGL24 and SOC1 are indispensable for Pin1At function in promoting flowering. Pin1At Mediates AGL24 Protein Stability Similar to the expression levels of AGL24 and SOC1 (Figure S6A), the localization of AGL24 and SOC1 in the shoot apical meristem during floral transition was also not affected by suppression or overexpression of Pin1At (Figure S13), further suggesting that Pin1At does not regulate AGL24 and SOC1 at the transcriptional level. To elucidate how Pin1At affects the function of AGL24 in the control of flowering time, we examined AGL24 protein expression in the nuclear and cytosolic fractions extracted from 2-week-old wild-type, Pin1At-AS, and 35S: Pin1At plants by western blot analysis using a specific antiAGL24 antibody (Liu et al., 2009). AGL24 protein expression was detected only in the nuclear fraction, with a higher level in 35S:Pin1At and a lower level in Pin1At-AS (Figure 6A), suggesting that Pin1At activity regulates phosphorylated AGL24 protein accumulation in the nucleus. As CIAP treatment of the nuclear protein from wild-type plants resulted in a shift of the upper band to a lower band with an expected size of AGL24 (Figure 6B), the upper band detected by AGL24 antibody before CIAP treatment should be the phosphorylated form of AGL24. This demonstrates that the endogenous AGL24 protein exists dominantly as the phosphorylated form, which is consistent with the observation on AGL24-6HA in 35S:AGL24-6HA (Figures 3A and 3E).
As human PIN1 regulates the turnover of many interacting proteins (Lu and Zhou, 2007), it is possible that Pin1At regulates the AGL24 level through mediating its protein stability. To test this possibility, we treated 2-week-old seedlings with a protein synthesis inhibitor, cycloheximide, and found that AGL24 had a half-life of about 2–3 hr in the absence of new protein synthesis (Figure 6C). Downregulation of Pin1At in Pin1At-AS plants decreased AGL24 protein stability with a half-life of less than 2 hr, whereas upregulation of Pin1At in 35S:Pin1At increased AGL24 half-life to 4–6 hr, suggesting that Pin1At indeed interacts with the phosphorylated AGL24 to regulate its stability in the nucleus. DISCUSSION The transition from vegetative to reproductive growth in plants is regulated by multiple flowering pathways in response to various developmental and environmental signals. This dynamic process involves spatial and temporal regulation of intermolecular interactions and enzymatic reactions. Although a number of flowering time regulators have been reported, whether and how protein conformational change affects floral transition remains unknown. In this study, we have found that the Arabidopsis PPIase, Pin1At, controls flowering through modulating cis/trans isomerization of the phosphorylated Ser/Thr-Pro motifs in two MADS-domain flowering regulators, SOC1 and AGL24 (Figure 6D). Because SOC1 and AGL24 synergistically integrate environmental and endogenous signals from several flowering pathways (Lee
Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc. 119
Molecular Cell Protein Isomerization Mediates Flowering
et al., 2000, 2008; Liu et al., 2008; Michaels et al., 2003; Samach et al., 2000; Yu et al., 2002), conformational regulation of these two proteins by Pin1At after phosphorylation delineates a hitherto unknown signaling mechanism that mediates the transition from the vegetative to reproductive growth in Arabidopsis. Extensive studies on human PIN1 have suggested that PIN1catalyzed isomerization of phosphorylated substrates has a profound impact on the function of proteins involved in many cellular processes (Liou et al., 2002; Lu, 2004; Lu and Zhou, 2007). For example, human PIN1 accelerates conformational change of the amyloid precursor protein (APP), thus regulating APP processing and amyloid-b production (Pastorino et al., 2006). The results presented here provide the evidence that PIN1-catalyzed isomerization of key transcription factors plays an indispensable role in regulating developmental transitions in plants. The conformational dynamics of AGL24 and SOC1 are tightly regulated by Pin1At. When Pin1At is downregulated, isomerization of these proteins from cis to trans and trans to cis forms is decelerated. Under this circumstance, if the cis or trans isomers are depleted in the cell, restoration of the equilibrium between cis and trans isomers would be delayed. However, if Pin1At is overexpressed, the isomerization rate is markedly increased, resulting in a rapid restoration of the equilibrium between cis or trans isomers of these proteins (Figure 4). The fact that Pin1At promotes flowering indicates that the accelerated isomerization of phosphorylated (Ser/Thr)-Pro motifs in AGL24 and SOC1 facilitates flowering, whereas the decelerated isomerization delays this transition. This is in line with the observation that Pin1At increases the stability of nuclear AGL24 protein that is a dosage-dependent flowering promoter (Michaels et al., 2003; Yu et al., 2002), suggesting that isomerization of phosphorylated AGL24 by Pin1At at least regulates AGL24 protein stability in the nucleus. Although we have not yet been able to examine SOC1 protein stability due to unsuccessful production of its specific antibodies, the similar molecular features of AGL24 and SOC1 and the analogous protein interaction between SOC1 and Pin1At imply that Pin1At may regulate the function of SOC1 like AGL24. Both SOC1 and AGL24 are MADS-domain proteins, which belong to a group of conserved transcription factors that play fundamental roles in developmental control and signal transduction in eukaryotes (Ng and Yanofsky, 2001; Riechmann and Meyerowitz, 1997; Theissen, 2000). Our findings on conformational regulation of SOC1 and AGL24 by Pin1At not only reveal an essential regulatory mechanism in plant development, but also shed light on the identification of PIN1 substrates and their relevant biological processes in animals. While this study suggests that prolyl cis/trans isomerases are evolutionarily conserved enzymes in plants and animals, how the postphosphorylation regulation mediated by these isomerases evolves to meet specific developmental needs in different organisms remains an intriguing question. EXPERIMENTAL PROCEDURES Plant Materials Seeds of Arabidopsis thaliana ecotype Columbia or Landsberg erecta were grown in LDs (16 hr light/8 hr dark) or SDs (8 hr light/16 hr dark) at 23 C ± 2 C.
The flowering time was measured by scoring the total number of rosette leaves after bolting. Construction of Binary Vectors To generate 35S:Pin1At construct, the full-length Pin1At cDNA fragment containing 44 bp 30 untranslated region was amplified using primers Pin1At-F1 and Pin1At-R1 (Table S1). The amplified fragment was digested and cloned into the pGreen 0229-35S binary vector (Yu et al., 2004). Based on this construct, we generated 35S:mPin1At, in which the four plant-specific amino acids (as shown in Figure S1) were deleted using the QuikChange II XL-Site-Directed Mutagenesis Kit (Stratagene; La Jolla, CA). To generate the Pin1At antisense construct, a 216 bp N-terminal fragment was amplified by Pin1At-F2 and Pin1At-R2 (Table S1). This fragment was also cloned into the pGreen 022935S binary vector. For the complementation test, the plasmids containing gAGL24 and gSOC1 were produced as previously described (Liu et al., 2008). Two genomic constructs, gAGL24-AP and gSOC1-2AP, which contain the mutations from Thr residue to Ala in AGL24 and Ser residue to Ala in SOC1, respectively, were further generated using the QuikChange II XL-Site-Directed Mutagenesis Kit (Stratagene). For the subcellular localization study, the Pin1At full-length cDNA was fused at the N terminus of GFP in the modified pGreen 0229-35S vector. Expression Analysis Quantitative real-time PCR was performed in triplicates on 7900HT Fast RealTime PCR system (Applied Biosystems; Foster City, CA) with SYBR Green PCR Master Mix (Applied Biosystems) (Liu et al., 2007). Semiquantitative PCR and nonradioactive in situ hybridization were performed as previously described (Yu et al., 2004). Primer sequences used for gene expression analysis were listed in Table S1. Protoplast Transfection Arabidopsis protoplast transfection by pGreen 0229-GFP and pGreen 0229Pin1At-GFP was performed as previously published (Sheen, 2001). The GFP localization and chlorophyll red autofluorescence were observed under the fluorescence light microscopy (Zeiss LSM510; Oberkochen, Germany). Alkaline Phosphatase Treatment of AGL24 and SOC1 Proteins Proteins were extracted from 2-week-old wild-type, 35S:AGL24-6HA, and 35S:SOC1-9Myc plants and immunoprecipitated with anti-AGL24, anti-HA, and anti-Myc antibodies (Santa Cruz Biotechnology; Santa Cruz, CA), respectively. After immunoprecipitation, the protein A/G beads (Santa Cruz Biotechnology) were washed twice with alkaline buffer (50 mM Tris-HCl, 1 mM MgCl2 [pH 8.8]) and then once with CIAP buffer with protease inhibitors. The beads were resuspended in CIAP buffer and incubated with CIAP at 37 C. Mocktreated control samples were incubated in the same volume of the reaction buffer without CIAP. The samples were subsequently spun for 1 min at 13,000 rpm to pellet the bead-bound complex. After removing the supernatant, 23 SDS protein sample buffer was added for further immunoblot analysis. BiFC Analysis Full-length cDNAs of SOC1, AGL24, and Pin1At were cloned into primary pSAT1 vectors (Tzfira et al., 2005). The resulting cassettes, including a fusion protein and a constitutive promoter, were cloned into pGreen binary vector HY105 and transformed into Agrobacterium. For BiFC experiments, 3-weekold tobacco (Nicotiana benthamiana) leaves were coinfiltrated with Agrobacterium as previously described (Sparkes et al., 2006). GST Pull-Down Assay GST-Pin1At fusion construct was prepared by cloning the coding region of Pin1At into the SpeI-XhoI sites of pET42a expression vector. Based on this construct, we generated the GST-mPin1At fusion construct using the same method for generating 35S:mPin1At. E. coli BL21 cells transformed with the plasmids were induced by IPTG, harvested, and lysed. After centrifugation, the supernatant was incubated with glutathione-agarose beads (Amersham Biosciences; Piscataway, NJ). The beads with the bound GST-Pin1At were washed for subsequent pull-down assays.
120 Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc.
Molecular Cell Protein Isomerization Mediates Flowering
NMR Analysis All NMR experiments were performed on a Bruker 800 MHz NMR spectrometer at 25 C. All spectra were recorded on peptide samples containing 20 mM phosphate buffer (90% H2O and 10% D2O [pH 6.5]) with or without Pin1At or mPin1At. The peptide concentration was 2.4 mM, and the Pin1At concentration was 0.03 or 0 mM for the samples of phosphorylated or nonphosphorylated AGL24 peptides (SSYDSG[p]TPLEDDSD) (GL Biochem; Shanghai, China), respectively. The peptide concentration was 0.15 mM, and the Pin1At concentration was 0.004 or 0 mM for the samples of phosphorylated and nonphosphorylated SOC1 peptides (SLIIF[p]SPKGKLYE), respectively, because these two peptides have relatively low solubility in aqueous solution. For all experiments, 256 3 512 complex points were acquired with spectral widths of 7200 3 9600 Hz in t1 3 t2 dimensions and an interscan delay of 1 s. For phosphorylated AGL24 peptide, ROESY spectra (Bax and Davis, 1985a) were acquired at a series of mixing times: 30, 45, 60, 75, 90, and 110 ms, and a spin-lock field strength of 4 kHz and 16 scans. For nonphosphorylated AGL24 peptide, one ROESY spectrum was recorded at a mixing time of 110 ms, a spin-lock field strength of 4 kHz, and 16 scans, while TOCSY experiments (Bax and Davis, 1985b) were carried out at a mixing time of 75 ms and 8 scans. For phosphorylated and nonphosphorylated SOC1 peptides, ROESY spectra were recorded at a mixing time of 110 ms, a spin-lock field strength of 4 kHz, and 96 scans, while TOCSY spectra were acquired at a mixing time of 75 ms and 64 scans. Ratios of cross to diagonal peak intensities for cis and trans conformations in the ROESY spectra depend on the forward (kctcat) and backward (ktccat) rate constants of a two-state exchange process (cis 4 trans) and the mixing times (tm). Changes of the ratios with the mixing times were fit using Equations 1 and 2 to determine the forward and backward rates (Pastorino et al., 2006): cat cat (1) Ict =Icc = kcat ct ½expðkex tm Þ 1= ktc expðkex tm Þ + kct cat cat Itc =Itt = kcat tc ½expðkex tm Þ 1= kct expðkex tm Þ + ktc
(2)
where Icc and Itt are the intensities of diagonal peaks for cis and trans conformers, Ict and Itc are the intensities of cross peaks from cis to trans and cat cat + ktc . from trans to cis conformers, and kex = kct SUPPLEMENTAL INFORMATION Supplemental Information includes one table and 13 figures and can be found with this article online at doi:10.1016/j.molcel.2009.12.020.
Davis, F.M., Tsao, T.Y., Fowler, S.K., and Rao, P.N. (1983). Monoclonal antibodies to mitotic cells. Proc. Natl. Acad. Sci. USA 80, 2926–2930. Hanes, S.D., Shank, P.R., and Bostian, K.A. (1989). Sequence and mutational analysis of ESS1, a gene essential for growth in Saccharomyces cerevisiae. Yeast 5, 55–72. Hani, J., Stumpf, G., and Domdey, H. (1995). PTF1 encodes an essential protein in Saccharomyces cerevisiae, which shows strong homology with a new putative family of PPIases. FEBS Lett. 365, 198–202. Hani, J., Schelbert, B., Bernhardt, A., Domdey, H., Fischer, G., Wiebauer, K., and Rahfeld, J.U. (1999). Mutations in a peptidylprolyl-cis/trans-isomerase gene lead to a defect in 30 -end formation of a pre-mRNA in Saccharomyces cerevisiae. J. Biol. Chem. 274, 108–116. He, Z., Li, L., and Luan, S. (2004). Immunophilins and parvulins. Superfamily of peptidyl prolyl isomerases in Arabidopsis. Plant Physiol. 134, 1248–1267. Hsu, T., McRackan, D., Vincent, T.S., and Gert de Couet, H. (2001). Drosophila Pin1 prolyl isomerase Dodo is a MAP kinase signal responder during oogenesis. Nat. Cell Biol. 3, 538–543. Huang, H.K., Forsburg, S.L., John, U.P., O’Connell, M.J., and Hunter, T. (2001). Isolation and characterization of the Pin1/Ess1p homologue in Schizosaccharomyces pombe. J. Cell Sci. 114, 3779–3788. Hunter, T. (1998). Prolyl isomerases and nuclear function. Cell 92, 141–143. Kardailsky, I., Shukla, V.K., Ahn, J.H., Dagenais, N., Christensen, S.K., Nguyen, J.T., Chory, J., Harrison, M.J., and Weigel, D. (1999). Activation tagging of the floral inducer FT. Science 286, 1962–1965. Kiefhaber, T., Quaas, R., Hahn, U., and Schmid, F.X. (1990). Folding of ribonuclease T1. 2. Kinetic models for the folding and unfolding reactions. Biochemistry 29, 3061–3070. Kobayashi, Y., Kaya, H., Goto, K., Iwabuchi, M., and Araki, T. (1999). A pair of related genes with antagonistic roles in mediating flowering signals. Science 286, 1960–1962. Koornneef, M., Alonso-Blanco, C., Peeters, A.J., and Soppe, W. (1998). Genetic Control of Flowering Time in Arabidopsis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 49, 345–370. Landrieu, I., De Veylder, L., Fruchart, J.S., Odaert, B., Casteels, P., Portetelle, D., Van Montagu, M., Inze´, D., and Lippens, G. (2000). The Arabidopsis thaliana PIN1At gene encodes a single-domain phosphorylation-dependent peptidyl prolyl cis/trans isomerase. J. Biol. Chem. 275, 10577–10581. Lee, H., Suh, S.S., Park, E., Cho, E., Ahn, J.H., Kim, S.G., Lee, J.S., Kwon, Y.M., and Lee, I. (2000). The AGAMOUS-LIKE 20 MADS domain protein integrates floral inductive pathways in Arabidopsis. Genes Dev. 14, 2366–2376.
ACKNOWLEDGMENTS We thank I. Lee, R. Amasino, M. Yanofsky, A. Samach, G. Coupland, and D. Weigel for providing seeds of various flowering mutants; S. Gelvin for providing pSAT1 vectors; S.Q. Yao and H.Y. Sun for providing some peptides; Y.C. Lai for protein expression; and B.C. Karthik for NMR analysis. We also thank T. Ito, J. Lim, and J.R. Dinneny for critical reading of the manuscript. This work was supported by a grant (06/1/21/19/473) from the Biomedical Research Council, the Agency for Science, Research and Technology, Singapore to Y.C.L. and a grant (T208B3113) from the Ministry of Education, Singapore and the intramural research funds from Temasek Life Sciences Laboratory to H.Y. L.C. was supported by the Singapore Millennium Foundation. Received: April 29, 2009 Revised: August 10, 2009 Accepted: October 14, 2009 Published: January 14, 2010
Lee, J., Oh, M., Park, H., and Lee, I. (2008). SOC1 translocated to the nucleus by interaction with AGL24 directly regulates leafy. Plant J. 55, 832–843. Levy, Y.Y., and Dean, C. (1998). The transition to flowering. Plant Cell 10, 1973–1990. Liou, Y.C., Ryo, A., Huang, H.K., Lu, P.J., Bronson, R., Fujimori, F., Uchida, T., Hunter, T., and Lu, K.P. (2002). Loss of Pin1 function in the mouse causes phenotypes resembling cyclin D1-null phenotypes. Proc. Natl. Acad. Sci. USA 99, 1335–1340. Liou, Y.C., Sun, A., Ryo, A., Zhou, X.Z., Yu, Z.X., Huang, H.K., Uchida, T., Bronson, R., Bing, G., Li, X., et al. (2003). Role of the prolyl isomerase Pin1 in protecting against age-dependent neurodegeneration. Nature 424, 556–561. Liu, C., Zhou, J., Bracha-Drori, K., Yalovsky, S., Ito, T., and Yu, H. (2007). Specification of Arabidopsis floral meristem identity by repression of flowering time genes. Development 134, 1901–1910.
REFERENCES
Liu, C., Chen, H., Er, H.L., Soo, H.M., Kumar, P.P., Han, J.H., Liou, Y.C., and Yu, H. (2008). Direct interaction of AGL24 and SOC1 integrates flowering signals in Arabidopsis. Development 135, 1481–1491.
Bax, A., and Davis, D.G. (1985a). 2D ROESY with cw spinlock for mixing phase sensitive using States-TPPI method. J. Magn. Reson. 63, 207–213.
Liu, C., Xi, W., Shen, L., Tan, C., and Yu, H. (2009). Regulation of floral patterning by flowering time genes. Dev. Cell 16, 711–722.
Bax, A., and Davis, D.G. (1985b). MLEV-17-based two-dimensional homonuclear magnetization transfer spectroscopy. J. Magn. Reson. 65, 355–360.
Lu, K.P. (2004). Pinning down cell signaling, cancer and Alzheimer’s disease. Trends Biochem. Sci. 29, 200–209.
Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc. 121
Molecular Cell Protein Isomerization Mediates Flowering
Lu, K.P., and Zhou, X.Z. (2007). The prolyl isomerase PIN1: a pivotal new twist in phosphorylation signalling and disease. Nat. Rev. Mol. Cell Biol. 8, 904–916. Lu, K.P., Hanes, S.D., and Hunter, T. (1996). A human peptidyl-prolyl isomerase essential for regulation of mitosis. Nature 380, 544–547. Lu, P.J., Wulf, G., Zhou, X.Z., Davies, P., and Lu, K.P. (1999). The prolyl isomerase Pin1 restores the function of Alzheimer-associated phosphorylated tau protein. Nature 399, 784–788. Lu, K.P., Finn, G., Lee, T.H., and Nicholson, L.K. (2007). Prolyl cis-trans isomerization as a molecular timer. Nat. Chem. Biol. 3, 619–629. Metzner, M., Stoller, G., Ru¨cknagel, K.P., Lu, K.P., Fischer, G., Luckner, M., and Ku¨llertz, G. (2001). Functional replacement of the essential ESS1 in yeast by the plant parvulin DlPar13. J. Biol. Chem. 276, 13524–13529. Michaels, S.D., Ditta, G., Gustafson-Brown, C., Pelaz, S., Yanofsky, M., and Amasino, R.M. (2003). AGL24 acts as a promoter of flowering in Arabidopsis and is positively regulated by vernalization. Plant J. 33, 867–874. Mouradov, A., Cremer, F., and Coupland, G. (2002). Control of flowering time: interacting pathways as a basis for diversity. Plant Cell 14 (Suppl ), S111–S130. Ng, M., and Yanofsky, M.F. (2001). Function and evolution of the plant MADSbox gene family. Nat. Rev. Genet. 2, 186–195. Ohad, N., Shichrur, K., and Yalovsky, S. (2007). The analysis of protein-protein interactions in plants by bimolecular fluorescence complementation. Plant Physiol. 145, 1090–1099. Pastorino, L., Sun, A., Lu, P.J., Zhou, X.Z., Balastik, M., Finn, G., Wulf, G., Lim, J., Li, S.H., Li, X., et al. (2006). The prolyl isomerase Pin1 regulates amyloid precursor protein processing and amyloid-beta production. Nature 440, 528–534. Ranganathan, R., Lu, K.P., Hunter, T., and Noel, J.P. (1997). Structural and functional analysis of the mitotic rotamase Pin1 suggests substrate recognition is phosphorylation dependent. Cell 89, 875–886. Riechmann, J.L., and Meyerowitz, E.M. (1997). MADS domain proteins in plant development. Biol. Chem. 378, 1079–1101.
Samach, A., Onouchi, H., Gold, S.E., Ditta, G.S., Schwarz-Sommer, Z., Yanofsky, M.F., and Coupland, G. (2000). Distinct roles of CONSTANS target genes in reproductive development of Arabidopsis. Science 288, 1613–1616. Sheen, J. (2001). Signal transduction in maize and Arabidopsis mesophyll protoplasts. Plant Physiol. 127, 1466–1475. Sparkes, I.A., Runions, J., Kearns, A., and Hawes, C. (2006). Rapid, transient expression of fluorescent fusion proteins in tobacco plants and generation of stably transformed plants. Nat. Protoc. 1, 2019–2025. Stukenberg, P.T., and Kirschner, M.W. (2001). Pin1 acts catalytically to promote a conformational change in Cdc25. Mol. Cell 7, 1071–1083. Theissen, G. (2000). Plant biology. Shattering developments. Nature 404, 711–713. Tzfira, T., Tian, G.W., Lacroix, B., Vyas, S., Li, J., Leitner-Dagan, Y., Krichevsky, A., Taylor, T., Vainstein, A., and Citovsky, V. (2005). pSAT vectors: a modular series of plasmids for autofluorescent protein tagging and expression of multiple genes in plants. Plant Mol. Biol. 57, 503–516. Yaffe, M.B., Schutkowski, M., Shen, M., Zhou, X.Z., Stukenberg, P.T., Rahfeld, J.U., Xu, J., Kuang, J., Kirschner, M.W., Fischer, G., et al. (1997). Sequencespecific and phosphorylation-dependent proline isomerization: a potential mitotic regulatory mechanism. Science 278, 1957–1960. Yao, J.L., Kops, O., Lu, P.J., and Lu, K.P. (2001). Functional conservation of phosphorylation-specific prolyl isomerases in plants. J. Biol. Chem. 276, 13517–13523. Yu, H., Xu, Y., Tan, E.L., and Kumar, P.P. (2002). AGAMOUS-LIKE 24, a dosage-dependent mediator of the flowering signals. Proc. Natl. Acad. Sci. USA 99, 16336–16341. Yu, H., Ito, T., Wellmer, F., and Meyerowitz, E.M. (2004). Repression of AGAMOUS-LIKE 24 is a crucial step in promoting flower development. Nat. Genet. 36, 157–161.
122 Molecular Cell 37, 112–122, January 15, 2010 ª2010 Elsevier Inc.