Analysis of Arabidopsis glucose insensitive mutants, gin5 and gin6, reveals a central role of the plant hormone ABA in the regulation of plant vegetative development by sugar

Abstract

Sugars have signaling roles in a wide variety of developmental processes in plants. To elucidate the regulatory components that constitute the glucose signaling network governing plant growth and development, we have isolated and characterized two Arabidopsis glucose insensitive mutants, gin5 and gin6, based on a glucose-induced developmental arrest during early seedling morphogenesis. The T-DNA-tagged gin6 mutant abrogates the glucose-induced expression of a putative transcription factor, ABI4, previously shown to be involved in seed-specific abscisic acid (ABA) responses. Thus, ABI4 might be a regulator involved in both glucose- and seed-specific ABA signaling. The characterization of the gin5 mutant, on the other hand, reveals that glucose-specific accumulation of ABA is essential for hexokinase-mediated glucose responses. Consistent with this result, we show that three ABA-deficient mutants (aba1-1, aba2-1, and aba3-2) are also glucose insensitive. Exogenous ABA can restore normal glucose responses in gin5 and aba mutants but not in gin6 plants. Surprisingly, only abi4 and abi5-1 but not other ABA-insensitive signaling mutants (abi1-1, abi2-1, and abi3-1) exhibit glucose insensitivity, indicating the involvement of a distinct ABA signaling pathway in glucose responses. These results provide the first direct evidence to support a novel and central role of ABA in plant glucose responses mediated through glucose regulation of both ABA levels by GIN5 and ABA signaling by GIN6/ABI4.

Figure 1

Phenotype of gin5 and gin6 mutants. Seedlings of WS wild-type (A,D), gin5 (B,E), and gin6 (C,F) grown for 10 days under 16:8 hr photoperiod in GM medium supplemented with 7% glucose (A–C) or 7% mannitol (D–E). Phenotype of 4-week-old WS wild-type (G), gin5 (H), and gin6 (I) plants grown in Metro Mix 200 in 15:9 photoperiod.

Figure 2

Molecular characterization of the GIN6 gene and phenotypic analysis of abi mutants. (A) The GIN6 locus. A region (shaded box) of the BAC T07M07 shows a 100% sequence identity with the genomic flanking sequence of the T-DNA insertion site (vertical arrow) of the gin6 mutant. The horizontal arrows indicate the different primers used to determine the integrity of the sequences flanking the insertion site. The ABI4 gene begins at position 42,623 and gene 11 ends at position 48,323. (B) Glucose induction of the ABI4 transcript is diminished in gin6. Ethidium bromide-stained RT-PCR product (974 bp) of ABI4 from WS wild-type grown on 2% (lane 1) or 7% glucose (lane 2) and from the gin6 mutant grown on 2% (lane 3) or 7% glucose (lane 4). The RT-PCR product of the APT1 gene (478 bp) was used as a control. (C) Glucose insensitivity is detected in abi4-1 and abi5-1 but not inabi1-1, abi2-1, or abi3-1. Seedlings were grown on GM medium containing 7% (abi5) or 7.5% (others) glucose for one week under 16:8 hr light:dark photoperiod.

Figure 3

ABA-mediated responses in gin5 and gin6 mutants. gin5 and wild-type plants were grown under well-watered conditions for 3 weeks and then subjected to drought stress by withholding water for 5 days. The plants were then sprayed with 10 ml of either water (A,B) or 3μm of ABA water solution (C) on days 3 and 5 of the stress treatment. (D) Dormancy analysis. Germination of seeds from WS wild-type (broken line) and the gin5 mutant (solid line) were examined without cold treatment (crosses; ▪) or with a 24-hr cold treatment (○, ▴). Germination was scored when the radicle tip had fully emerged from the seed coat. These data represent the average of three independent experiments evaluating 80 seeds. The standard error was smaller than the symbol size. (E) ABA dose response in germination. Wild-type (broken line), gin5 (solid line, circles) and gin6 (solid line, triangles) seeds were analyzed. The data correspond to the average of two different experiments each containing around 50 seeds per data point. The number of germinated seeds was expressed as the percentage of fully germinated seeds of the total seeds plated. (F) Phenotype of aba mutants in the presence of high glucose. Ler and Col wild-type seedlings, aba1-1, aba3-2, and aba2-1 were grown in the presence of 7% glucose on the GM medium for 7 days. The two wild-type ecotypes were included as control for the corresponding aba mutants.

Figure 4

The effect of ABA and high glucose on the phenotype of gin5 and aba1-1 mutants. WS (A,C), gin5 (E,G,I), Ler (B,D), and aba1-1 (F,H,J) plants were grown on 4, 5, 6, 6.5, and 7% glucose (A–H) or 4, 5, 6, 6.5 and 7% mannitol (I,J), in the absence (A,B,E,F) or the presence of 100 nm ABA (C,D,G–J) for 10 days. gin6 and abi5-1 mutants were grown in the presence of 7% glucose without (K,M) or with (L,N) a 100 nm ABA. Representative plants are shown in each condition. The number below each plant indicates the percentage of plants displaying a green phenotype for each of the conditions used.

Figure 5

Expression of glucose- and ABA-modulated genes in gin5 and gin6 mutants. (A) RNA blot analysis. Total RNA (10 μg) of WS wild-type (lanes 1,2), gin5 (lanes 3,4), and gin6 (lanes 5,6) from 19-day-old seedlings grown in the presence of 2% (lanes 1,3,5) or 7% (lanes 2,4,6) glucose was used for RNA blots. Glucose repression of CAB1 and PC and activation of PAL were examined. Ethidium bromide-stained ribosomal RNA was used as a loading control. (B) RT–PCR analysis. Total RNA (5 μg) from WS wild-type (lanes 1,2) or gin6 (lanes 3,4) seedlings grown in 2% (lanes 1,3) or 7% (lanes 2,4) glucose was used to synthesize cDNA. A portion (1/80) of the synthesized cDNA was used to amplify and quantitate COR15a (accession no. U01377), EM6 (accession no. Z11923 S61761), ADH (accession no. M12196), and ZEP1 (accession no. T45502) gene transcripts. The sizes of PCR products are 210 bp, 302 bp, 398 bp, and 261 bp, respectively.

Figure 6

gin5 acts downstream of the AtHXK gene in the glucose signaling pathway. (A) Analysis of the glucose response. Seedlings of gin5 (panel 1), 35S-AtHXK (panel 2), and 35S-AtHXK gin5 (panel 3) were grown in the presence of 7% glucose for 8 days. (B) Immunoblot analysis. The HXK protein levels of the wild-type (lane 1), 35S-AtHXK transgenic plant (lane 2), the gin5 mutant (lane 3) and independent F₂ individuals displaying glucose insensitivity from the 35S-AtHKX gin5 cross (lanes 4–9) are shown. Protein was extracted from 15-day-old seedlings grown on GM medium with 7% glucose.

Figure 7

A model for the role of ABA in the HXK-dependent glucose signaling network.