Characterization of the response of the Arabidopsis response regulator gene family to cytokinin

Abstract

We examined the expression of a family of Arabidopsis response regulators (ARR) and found that the steady-state levels of RNA for most are elevated very rapidly by cytokinin. Using nuclear run-on assays we demonstrated that this increase in ARR transcript levels in response to cytokinin is due, at least in part, to increased transcription. The start site of transcription for the ARR5 gene was identified using primer extension analysis. A DNA fragment comprised of 1.6 kb upstream of the ARR5 transcript start site conferred cytokinin-inducible gene expression when fused to a beta-glucuronidase reporter, confirming that the transcription rate of ARR5 is elevated by cytokinin. This reporter construct was also used to examine the spatial pattern of ARR5 expression. The highest levels of expression were observed in the root and shoot apical meristems, at the junction of the pedicle and the silique, and in the central portion of mature roots. The expression of ARR5 in the apical meristems was confirmed by whole mount in situ analysis of seedlings and is consistent with a role for cytokinin in regulating cell division in vivo.

Figure 1

The ARR gene family. A, Phylogenetic tree of ARR receiver domains. The tree was generated using the AllAll program at Molecular Biology Computational Resource at Baylor college of Medicine (http://cbrg.ing.ethz.ch/subsection3_1_1.html). The amino acid sequences of only the receiver domains of each gene was used (see B). This program uses a least-squared, heuristic method to generate trees. The accession numbers for all but ARR15, ARRR16, and ARR17 are presented in D'Agostino and Kieber (1999). The accession numbers for the novel ARR-predicted amino acid sequences are as follows: ARR15: AF305720; ARR16: AF305721; ARR17: AF305722. B, Alignment of type A ARR receiver domain amino acid sequences. Residues identical in >6 ARRs are blocked in black. The three residues that are invariant among all response regulators are marked with an asterisk, including the predicted Asp phosphorylation site, embedded in a conserved TDY sequence. C, A cartoon representation of the domain structures of the various type A ARR proteins. The ARRs corresponding to each of the four different structures is indicated to the left of each cartoon. The response regulator domains are represented by white rectangles and the C-terminal extensions by black rectangles. The different domains are depicted to scale and the various features of the C-terminal extensions are noted within the black rectangles.

Figure 2

Kinetics of ARR gene induction. A, Northern-blot analysis of 15 μg of total RNA from 3-d-old etiolated seedlings treated with 5 μm BA for various times (indicated in minutes above each lane) and hybridized with the indicated ARR probe (at left). The β-tubulin shown is a representative image and is not the loading control for all of the blots (see methods). B, Quantification of transcript levels from blot depicted in A. The signals from the northern blots were quantified with a PhosphorImager and normalized to the β-tubulin loading control. The highest level of expression for each probe was assigned a value of 100% and all other points were normalized to it (percentage of maximum).

Figure 3

Tissue-specific expression of ARR6 and ARR7. A and B, RNA was extracted from tissues treated with 5 μm BA (lanes 2, 4, 6, and 8) or buffer control (lanes 1, 3, 5, and 7) for 50 min. Fifteen micrograms of total RNA was separated by agarose gel electrophoresis, blotted to a nylon membrane, and hybridized with an ARR6 (A), ARR7 (B), or 18S rDNA (bottom of each inset) probe. Tissues analyzed were buds and young flowers (lanes 1 and 2), stems of young inflorescences (lanes 3 and 4), leaves from 2-week-old adult plants (lanes 5 and 6), and soil-grown roots (lanes 7 and 8). The signals were quantified using a PhosphorImager and the expression level in each tissue normalized to the 18S loading control. The highest signal was set to 100% and the other samples were normalized to it.

Figure 4

Nuclear run-on analysis of ARR genes in cytokinin-treated tissue. Leaves from 2-week-old adult plants were incubated with 5 μm BA for the indicated times. Intact nuclei were purified and used for in vitro transcription. Nuclear transcripts were hybridized to duplicate slot blots containing 200 ng of ARR5, ARR4, ARR6, ARR7, or β-tubulin cDNAs. The hybridizations were quantified using a PhosphorImager and the ARR signals normalized to the tubulin control. The resultant relative transcript levels were plotted as a function of time.

Figure 5

Identification of the transcription initiation site of the ARR5 gene. Primer extension analysis. A [³²P]-labeled ARR5 primer (see “Materials and Methods”) was hybridized with 50 μg of total RNA and extended using reverse transcriptase. The product was resolved on a sequencing gel (far left lane), and the mobility compared with an ARR5 genomic DNA sequencing reaction using the same ARR5 primer (lanes A, C, G, and T). The nucleotide sequence in the vicinity of the start site of transcription is shown to the right of the gel insert. The arrow marks the deduced transcription start site and the boxed residues mark the predicted TATA box. Below the gel inset is a schematic diagram of the ARR5 primer relative to the transcription and translation start sites and the primer extended product.

Figure 6

Transcriptional induction of ARR5 by cytokinin as demonstrated by promoter GUS fusions in transgenic plants. A, GUS staining of etiolated transgenic seedlings harboring an ARR5 promoter, but lacking the 5′-untranslated region fused to GUS (pIB-1TC). Seedlings were grown for 3 d in the dark on MS with no added hormone (left seedling) in the presence of 2.5 μm BA (right seedling). B, GUS staining of pIB-1TC transgenic seedlings were grown for 6 d in the light on MS containing no (left seedling), 0.5 μm BA (center seedling), or 5 μm BA (right seedling). C, Northern analysis of GUS expression in pIB-1TC transgenic seedlings following cytokinin treatment. Leaves from 2-week-old pIB-1TC plants were treated with 5 μm BA for various times and total RNA isolated. The RNA was analyzed by northern blotting and hybridized with a GUS (top inset, GUS), or β-tubulin (bottom inset, TUB). The signals were quantified using a PhosphorImager and the GUS signal was normalized to the β-tubulin loading control. The relative signal at time 0 was assigned a value of 1, and the subsequent time points plotted relative to it (fold-induction).

Figure 7

Pattern of ARR5 expression. A through I, Gus staining of stable Arabidopsis transformants harboring a pIB1.6-TC-GUS T-DNA. Apical portion of 7-d-old (A) and 15-d-old (B) seedlings grown in the light on MS. C, Close-up of the region corresponding to the apical meristem in a 10-d-old light-grown plant. D, Inflorescence of 4-week-old plants. E, Close-up of primary root tip from 7-d-old seedlings grown in the light on MS. F, Mature root/hypocotyl junction from 15-d-old seedlings grown in the light on MS. G, Central portion of root from same seedling as in F. H, Silique from 4-week-old plant, mature, but fully green. I, Silique that had just turned yellow from a 4-week-old plant. J through M, Whole mount in situ hybridization with ARR5. Five-day-old light-grown seedlings grown on MS were fixed and hybridized with a sense (J and L) or an antisense (K and M) digoxygenin-labeled ARR5 RNA probe (see “Materials and Methods”). The sense probe is a negative control and should not hybridize to the endogenous ARR5 transcript. Purple color indicates a positive reaction. The arrow in M indicates the shoot apical meristem staining evident with the antisense probe.