Overexpression of Arabidopsis ESR1 induces initiation of shoot regeneration

Abstract

Functional screening of an Arabidopsis cDNA library enabled the identification of a novel cDNA, ESR1 (for Enhancer of Shoot Regeneration), that can confer cytokinin-independent shoot formation when overexpressed in Arabidopsis root explants. Neither callus induction nor root formation was affected by ESR1 overexpression. ESR1 encodes a putative transcription factor with an AP2/EREBP domain. Surprisingly, ESR1 overexpression also greatly increased the efficiency of shoot regeneration from root explants in the presence of cytokinin, with a shift in the optimal cytokinin concentration required for this process. The effects of ESR1 overexpression on shoot regeneration are synergistic with those of cytokinin. Overexpression of ESR1 cannot induce callus formation or root formation, suggesting that its effects are specific to shoot formation. In wild-type Arabidopsis plants, ESR1 expression was induced by cytokinin. ESR1 transcript levels also increased transiently during shoot regeneration from root explants, most probably in response to cytokinin in the shoot-inducing medium. This transient increase occurred after the acquisition of competence for regeneration and before shoot formation, which is consistent with the physiological effects of ESR1 overexpression. Our results suggest that ESR1 may regulate the induction of shoot regeneration after the acquisition of competence for organogenesis.

Figure 1.

Effects of ESR1 Overexpression on Shoot Regeneration from Root Explants. (A) Representative plates 4 weeks after transformation. Arabidopsis root cultures transformed with the vector alone (pER10) or pER10-ESR1 were cultured on SIM with or without cytokinin (25 μM 2-ip) or an inducer of transcription (5 μM 17β-estradiol). (B) Numbers of transformed shoots from 0.1 g fresh weight of root culture. Data represent the average of four independent experiments, and error bars indicate standard deviations. Open columns show transformants with the vector alone (pER10), and cross-hatched columns show transformants with pER10-ESR1.

Figure 2.

Deduced Amino Acid Sequence of ESR1 and Comparison of the AP2/EREBP Domains of ESR1 with Those of Other AP2/EREBP Proteins. (A) Predicted amino acid sequence encoded by ESR1 cDNA. Amino acid numbers are indicated at right, and the AP2/EREBP domain is underlined. (B) Alignment of the AP2/EREBP domains of ESR1 and other representative AP2/EREBP proteins by the multiple alignment function of Lasergene (DNASTAR, Madison, WI). GenBank accession numbers are T02433 (EREBP3), BAA32421(AtERF4), Y09942 (AtEBP), AAD25937 (ABI4), CAB80440 (Aintegumenta), AAC13770 (APETALA2), and BAB11119 (Leafy petiole).

Figure 3.

Sensitivity of ESR1-Overexpressing Cells to Cytokinin. (A) Numbers of regenerated shoots in different concentrations of cytokinin. Ten root segments from transgenic seedlings carrying either pER8 or pER8-ESR1 were cultured on plates with MS medium supplemented with the indicated concentrations of 2-ip with or without an inducer (10 μM 17β-estradiol). Each value represents the average of three independent experiments, and bars indicate standard deviations. (B) RNA gel blot analysis of the ESR1 transcript. Four micrograms of total RNA was loaded on each lane and hybridized with an ESR1 cDNA probe or a 25S rRNA probe.

Figure 4.

Effects of ESR1 Overexpression on the Formation of Calli, Shoots, and Roots. (A) and (B) Root segments on RIM with (B) or without (A) inducer. (C) and (D) Root segments on SIM with (D) or without (C) inducer. (E) and (F) Root segments on CIM with (F) or without (E) inducer. 17β-Estradiol (10 μM) was used as an inducer in (B), (D), and (F). Root segments from a transgenic Arabidopsis line carrying a pER8-ESR1 construct were preincubated on CIM for 3 days and then transferred onto the media described above. Photographs were taken 4 weeks after transfer onto individual media. Bars = 0.5 cm.

Figure 5.

Inhibition of Shoot Regeneration Caused by Constitutive ESR1 Expression. (A) Phenotype of transformants obtained by root transformation. Root segments were transformed with pSK-ESR1, a 35S:ESR1 construct, after incubation on CIM for 3 days and then transferred onto SIM containing kanamycin (50 mg/L) and carbenicillin (300 mg/L) for 4 weeks. Six representative explants are shown. (B) Phenotype of seedling obtained by in planta transformation. Seedlings harboring a 35S:ESR1 construct were germinated on MS medium containing kanamycin and carbenicillin. The photograph was taken 6 weeks after germination. (C) Magnified root part of (B). (D) Magnified aerial part of (B). Bars = 1 cm.

Figure 6.

Time Course of ESR1 Expression during Shoot Regeneration. RNA gel blot analysis of the ESR1 transcript. Root segments were preincubated on CIM for 4 days and then transferred onto SIM. Total RNAs were prepared from root explants at the times indicated, and 10 μg of total RNA was loaded. RNA gel blots were hybridized to different cDNA probes. Numbers above the lanes indicate days (A) or hours (B) after transfer onto SIM. In (B), in the first lane from the right (CIM), after pretreatment on CIM, root segments were incubated on fresh CIM for 24 hr.

Figure 7.

Induction of ESR1 Expression by Cytokinin. (A) RNA gel blot analysis in seedlings. Two-week-old Arabidopsis plants grown on MS medium were transferred onto MS medium containing various phytohormones. After 24 hr, total RNAs were prepared and analyzed on a gel. Ten micrograms of total RNA was loaded and hybridized to different cDNA probes. Mock, fresh MS medium without any phytohormones; ACC, 1-aminocyclopropane-1-carboxylic acid (50 μM); IAA, indole-3-acetic acid (2.5 μM). 2-ip was used at a concentration of 5 μM. (B) Expression of ESR1 in response to cytokinin with or without 2,4-D pretreatment in root explants. Root segments were pretreated on MS medium with or without 2,4-D (2 μM) for 4 days (indicated as MS pretreatment or 2,4-D pretreatment, respectively) and then transferred onto MS medium containing the compound(s) indicated. After 12 hr of incubation, total RNAs were prepared. Ten micrograms of total RNA was loaded on each lane and hybridized to different cDNA probes. Concentrations were as follows: adenine, 50 μM; trans-zeatin, 5 μM; 2-ip, 5 μM; 6-benzyladenine (BA), 5 μM; kinetin, 5 μM.