Extended Vernalization Regulates Inflorescence Fate in Arabis alpina by Stably Silencing PERPETUAL FLOWERING1

Abstract

The alpine perennial Arabis alpina initiates flower buds during prolonged exposure to cold. In the accession Pajares, we demonstrate that the length of vernalization influences flowering time and inflorescence fate but does not affect the axillary branches that maintain vegetative growth. The expression of floral organ identity genes gradually increases in the main shoot apex during vernalization, correlating with an increase in floral commitment. In northern Arabidopsis (Arabidopsis thaliana) accessions, the length of vernalization modulates the stable silencing of the floral repressor FLOWERING LOCUS C (FLC). We demonstrate that expression of PERPETUAL FLOWERING1 (PEP1), the ortholog of FLC in A. alpina, is similarly influenced by the duration of the exposure to cold. Extended vernalization results in stable silencing of PEP1 in the inflorescence. In contrast, insufficient vernalization leads to PEP1 reactivation after cold treatment, which correlates with delayed flowering and the appearance of floral reversion phenotypes such as bracts and vegetative inflorescence branches. Floral reversion phenotypes are reduced in the pep1-1 mutant, suggesting that PEP1 regulates the fate of the inflorescence after vernalization. The effect of vernalization duration on stable silencing of PEP1 is specific to meristems that initiate flowering during cold treatment. Extended vernalization fails to silence PEP1 in young seedlings and axillary branches that arise from buds initiated during cold treatment, which remain vegetative. We conclude that the duration of vernalization in A. alpina differentially regulates PEP1 in the inflorescence and axillary branches. PEP1 has a dual role regulating meristem fate; it prevents meristems from flowering and antagonizes inflorescence development after vernalization.

Figure 1.

The length of vernalization determines inflorescence outgrowth and floral reversion in A. alpina. A, Schematic representation of a flowering A. alpina plant after vernalization. Axillary branches have different fates according to their position on the plant: (V1) axillary branches that flower and partially senesce, (V2) dormant axillary buds, and (V3) axillary vegetative branches. The inflorescence consists of the I1 zone with inflorescence branches and the I2 zone with solitary flowers. The I1 branches can be distinguished from the V3 branches by the bolting of the inflorescence stem. B, Flowering time of Pajares plants vernalized for 8, 12, 15, 18, 21, or 24 weeks. C, Length of the inflorescence in Pajares plants vernalized for 8, 12, 15, 18, 21, or 24 weeks measured 8 weeks after vernalization. D, Picture of A. alpina Pajares exposed to 8, 12, 15, 18, 21, or 24 weeks of vernalization (8w, 12w, 15w, 18w, 21w, or 24w) and subsequently grown for 3 weeks in an LD greenhouse. Bar, 10 cm. E, Percentages of flowering and vegetative inflorescence branches in the I1 zone after 8, 12, 15, 18, 21, and 24 weeks of vernalization followed by 8 weeks in a LD greenhouse. Close-up pictures of flowering (top) and vegetative (bottom) I1 inflorescence branches. Bar, 0.5 cm. F, Number of bracts in the I2 zone of plants exposed to 8, 12, 15, 18, 21, and 24 weeks of vernalization followed by 14 weeks in a LD greenhouse. Close-up picture of the bracts in the inflorescence of plants exposed to 12 weeks of vernalization followed by 14 weeks in a LD greenhouse. Bar, 3 cm. Arrows indicate two bracts as an example. The error bars represent the SD; n = 9–12. Letters indicate significant differences determined by omnibus Kruskal-Wallis test followed by pairwise multiple comparison using Mann-Whitney U test (α-value of 0.05). No letters indicate no significant differences.

Figure 2.

The duration of vernalization determines the commitment to floral identity of inflorescence and V3 branches in a basipetal pattern. A, Ratio of flowering branches after different periods of vernalization (8, 12, 15, 18, 21, and 24 weeks). Nodes are numbered from the bottom of the inflorescence. Positive values represent the I1 inflorescence zone. Negative values represent the V3 zone. The basis of the inflorescence bolted stem is indicated by zero (0). B, Mean length of inflorescence branches at consecutive nodes along the inflorescence in plants vernalized for 12 or 24 weeks followed by 8 weeks in a LD greenhouse. The error bars represent the SD; n = 11–12.

Figure 3.

The transcripts of floral meristem identity genes accumulate in the apices of 6-week-old plants that initiate flowering during cold treatment. A, C, E, G, and I, Expression level of AaSOC1, AaFUL, AaTFL1, AaLFY, and AaAP1 in the shoot apical meristem of 3- and 6-week-old Pajares plants at the end of different periods of vernalization. Three- or six-week-old plants were vernalized for 8, 18, or 24 weeks. Apices were harvested at the end of the vernalization treatments. B, D, F, H, and J, Expression level of AaSOC1, AaFUL, AaTFL1, AaLFY, and AaAP1 in V3 axillary branches. Six-week-old plants were vernalized for 8, 18, or 24 weeks and transferred to a LD greenhouse for 5 weeks. The data are the means of two or three biological replicates, and error bars represent the SD. Letters above the columns indicate significant differences determined by multiple pairwise comparisons using Benjamini-Hochberg-corrected P values (α-value of 0.05). Graphs with no letters indicate no significant differences.

Figure 4.

Extended vernalization silences PEP1 in the inflorescence. A, PEP1 expression in the main shoot apex of 6-week-old plants before (0), during (1, 3, 5, 8, 12, and 18 weeks), and after 18 weeks of vernalization. After vernalization, main shoot apices were harvested 1, 2, 3, and 4 weeks after the plants were transferred back to a LD greenhouse. PEP1 expression was also measured in the apices of axillary V3 branches after vernalization. B, PEP1 expression at the end of vernalization in the apices of plants vernalized for 8, 18, and 24 weeks (0 indicates expression at the end of vernalization) and after vernalization (1, 3, and 5 weeks in a LD greenhouse). C, PEP1 expression in the apices of axillary V3 branches vernalized for 8, 18, or 24 weeks and transferred back to a LD greenhouse for 5 weeks. D, PEP1 expression in the main shoot apex of 3-week-old seedlings at the end of 8, 18, or 24 weeks of vernalization (0 indicates expression at the end of vernalization) and after 5 weeks in a LD greenhouse. Letters above the columns indicate significant differences determined by multiple pairwise comparisons using Benjamini-Hochberg-corrected P values (α-value of 0.05) and Hochberg-GT2. Graphs with no letters indicate no significant differences. All data are the means of two or three biological replicates, and error bars represent the SD.

Figure 5.

AaFT1 transcript gradually accumulates in the apices of 6-week-old plants after different cold treatment periods. A, AaFT1 expression level in the shoot apical meristem of 3- and 6-week-old Pajares plants. Plants were vernalized for 8, 18, or 24 weeks (8w, 18w, and 24w), and apices were harvested at the end of each vernalization treatment (0 indicates expression at the end of vernalization) and 5 weeks after vernalization. B, AaFT1 expression level in the apices of V3 axillary branches vernalized for 8, 18, or 24 weeks and transferred back to a LD greenhouse for 5 weeks. Letters above the columns indicate significant differences determined by multiple pairwise comparisons using Benjamini-Hochberg-corrected P values (α-value of 0.05). Graphs with no letters indicate no significant differences. The data are the means of two or three biological replicates, and error bars represent the SD.

Figure 6.

Floral reversion phenotypes are reduced in the pep1 mutant. Phenotypes were measured in wild-type Pajares (WT) and pep1-1 plants exposed to 8, 12, 15, 18, and 21 weeks of vernalization. A, Time to flower emergence; B, number of bracts; C, percentage of flowering I1 inflorescence branches (FB); D, number of siliques in the I1 branches; E, number of siliques in the I2 inflorescence zone. All measurements in B to E were performed at the end of flowering. The error bars represent the SD; n = 9–12. Asterisks indicate significant differences between wild-type and the pep1-1 mutant at each time point determined by multiple pairwise Bonferroni tests (α-value of 0.05).

Figure 7.

Schematic representation of a model for the genetic regulation of flowering and vegetative branches in a fully vernalized A. alpina plant. Flowering is initiated during vernalization in the apical meristem of the main shoot and in the V1 axillary branches (orange circle). V3 vegetative branches arise from buds below the inflorescence initiated during cold treatment (green circle). The expression of PEP1 and AaTFL1 is downregulated in the main shoot apical meristem, whereas the expression of AaSOC1, AaFUL, AaLFY, and AaAP1 is upregulated (triangles). After the return to LD greenhouse conditions, the I1 and I2 zones of the inflorescence develop and the V1 branches flower. PEP1 remains stably silenced in the shoot apex, and AaFT1 expression is upregulated (triangle). The length of vernalization influences the expression levels of meristem identity genes during vernalization and of PEP1 and AaFT1 after vernalization. In contrast, the V3 branches show high PEP1 and AaTFL1 expression levels irrespective of the duration of vernalization.