Genetic control of meristem arrest and life span in Arabidopsis by a FRUITFULL-APETALA2 pathway

Abstract

Monocarpic plants have a single reproductive cycle in their lives, where life span is determined by the coordinated arrest of all meristems, or global proliferative arrest (GPA). The molecular bases for GPA and the signaling mechanisms involved are poorly understood, other than systemic cues from developing seeds of unknown nature. Here we uncover a genetic pathway regulating GPA in Arabidopsis that responds to age-dependent factors and acts in parallel to seed-derived signals. We show that FRUITFULL (FUL), a MADS-box gene involved in flowering and fruit development, has a key role in promoting meristem arrest, as GPA is delayed and fruit production is increased in ful mutants. FUL directly and negatively regulates APETALA2 expression in the shoot apical meristem and maintains the temporal expression of WUSCHEL which is an essential factor for meristem maintenance.

Fig. 1

FUL regulates GPA timing. a ful mutants produce more fruits than WT. b The increase in the number of fruits produced by the main inflorescence in ful mutants does not depend on the genetic background or on the reduced seed production of ful fruits, since other mutants with similar reduced fertility (like hec3 or fer/+) undergo GPA-like WT. Asterisks (***) indicate a significant difference (P < 0.001) from WT according to Student’s t test. c ful mutants produce more fruits because GPA is delayed and the period of fruit production is extended. d Induced sterility by continuous manual pruning of developing flowers causes increased flower production in WT and ful mutants. Asterisks (***) indicate a significant difference (P < 0.001) from the corresponding untreated plants according to Student’s t test. e While sterility in WT leads to the formation of a terminal flower, in ful mutants sterility only leads to a further delayed GPA. n = 10 for all genotypes in b, c, d

Fig. 2

WUS expression is temporally extended in the SAM of ful and ap2-170 mutants. Whole mount in situ hybridization of WUS mRNA in SAMs from different genetic backgrounds shortly after floral transition (0 weeks after bolting- 0 wab), 2 weeks after bolting (2 wab) and 4 weeks after bolting (4 wab), a time point when wild-type plants have undergone GPA and stop producing new floral meristems, while mutants are still proliferative. Arrows indicate the SAM. Bars = 100 µm

Fig. 3

FUL controls GPA timing by regulating AP2 activity in the SAM. a L170.2, a mutant from a second-site mutagenesis on ful-1, shows a dramatically delayed GPA, with plants producing more than 100 fruits and usually dying of other causes before GPA can take place; L170.2 carries a mutation in the miR172 binding site of AP2 (renamed as ap2-170). When segregated from the ful-1 background, the ap2-170 mutation also delays GPA. n = 10 for all genotypes. b L170.2 bears highly indeterminate fruits, but not ful-1 or ap2-170 single mutants; ap2-170 mutants or heterozygous ap2-170/ap2-2 plants do not show the typical homeotic defects of ap2-2 mutants. c Induction of AP2¹⁷⁰ expression in arrested plants reactivates SAM and flower and fruit production. d Overexpression of the AP2¹⁷⁰ allele causes highly indeterminate flowers both in WT and ful backgrounds, while overexpression of wild-type AP2 does not cause evident phenotypic defects in flowers or fruits. Similar indeterminate flowers to those produced by AP2¹⁷⁰ overexpression are produced by combining the ap2-170 allele with a FUL:VP16 allele. FUL:VP16 also causes mild indetermination in fruits, especially in a ful mutant background. e Transcript levels of AP2 are significantly elevated in the inflorescence meristems of ful mutants. Asterisks (*) indicate a significant difference (P < 0.001) from WT according to Student’s t test

Fig. 4

AP2 and AP2-like genes control GPA timing. a ChIP-qPCR experiment showing the enrichment of the different fragments relative to REF1 in ChIP samples from IM tissue of ap1 cal pFUL::FUL:GFP plants. Fragment enrichment was calculated as a percentage of the corresponding input sample. See Supplementary Data 1 for information about the fragments. Enrichment is detected for AP2, SNZ, TOE1 and TOE3, but not for WUS and the MIR172 genes. b In the gene trap line GT.100845, the activity of a miR-independent GUS reporter inserted in the 5’-UTR of the AP2 gene is elevated both in ful and FUL:VP16 backgrounds. All panels correspond to heterozygotes AP2 (WT)/AP2-GT.100845. c Luciferase reporter assay in N. benthamiana leaves co-transformed with 35S::FUL or 35S::FUL:VP16 and different reporter lines where the AP2 promoter fragments enriched in FUL ChIP-seq results (AP2-I and AP2-II) and other non-enriched fragment (AP2-III) drive the expression of the Luc gene. Ren expression is used as internal control and Luc/Ren ratios are represented. d Binding of FUL to different DNA probes in an EMSA assay (as indicated on top of the lanes). The absence or presence of FUL (e, empty or F, FUL) is indicated at the bottom. The AP2-I fragment was loaded on both gels. The arrows indicate the shifts that can be attributed to binding of a FUL homodimer or a FUL homotetramer. More information about the probes can be found in Supplementary Data 1. e Transcript levels of AP2-like genes bound by FUL in inflorescence apices of wild-type and ful mutants: AP2, SNZ and TOE1 are significantly overexpressed in ful, while transcript levels of MIR172 genes are not affected. f ap2-12 mutants produce a similar number of flowers than WT in spite of severely reduced fertility (noted by seed number per fruit), but much less than the unrelated sterile nga^quad mutant; 35S::miR172 plants, where expression of AP2 and related genes is reduced, show further reduced flower production. The delayed GPA observed in ful mutants is not fully suppressed by the ap2-12 mutation, but almost completely in the 35S::miR172 background. Shaded bars and shaded boxes in seed numbers indicate genotypes where fertility is above the threshold affecting GPA timing. Superscript letters indicate a significant difference (P < 0.001) from Col (a), ap2-12 (b) and ful-2 (c) controls, respectively, according to t test. g Induced sterility by continuous pruning of developing flowers causes increased flower production in WT and ap2-2 mutants. Error bars represent the SE of four biological replicas in the case of expression analyses, three replicas for the ChIP-qPCR and six replicas for the Luc assays. Significant differences from the control (t test, P < 0.05) are indicated with an asterisk in a, c, e, g). n = 10 for all genotypes in f, g

Fig. 5

Model for temporal regulation of inflorescence meristem maintenance. WUS levels in the inflorescence SAM decrease with plant age. WUS expression is under positive regulation of AP2 and AP2-like factors in the SAM and negative regulation of an unknown signal from developing flowers/fruits. It has been previously shown that during the plant’s life cycle, levels of miR156 decrease with age, concomitantly increasing the expression of members of the SPL family and of miR172. Hence, AP2 levels in the SAM of the inflorescence would decrease with time by the combined repressing action of miR172 and FUL, which is strongly upregulated by SPL factors after floral transition. The decrease in AP2 levels and the increasing repressing signal derived from fruits cause WUS to eventually turn off in the SAM. In ful mutants or in the ap2-170 mutants, AP2 and WUS are expressed in the SAM much longer, and therefore, GPA is delayed, while in loss-of-function ap2/ap2-like mutants, GPA occurs earlier. In sterile mutants, while the FUL-AP2 module should not change with age, the repressing signal from fruits is not active and WUS remains to be expressed longer