Related to ubiquitin 1 and 2 are redundant and essential and regulate vegetative growth, auxin signaling, and ethylene production in Arabidopsis

Abstract

Related to Ubiquitin (RUB)/Nedd8 is a ubiquitin-like protein that covalently attaches to cullins, a subunit of the SCF (for Skp, Cdc53p/Cul1, and F-box protein) complex, an E3 ubiquitin ligase, and has been shown to be required for robust function of the complex. The effects of reducing protein levels for two Rub proteins, RUB1 and RUB2, were characterized in Arabidopsis thaliana. T-DNA insertional null lines homozygous at a single RUB-encoding locus were analyzed and found to have a wild-type phenotype. A double mutant was never recovered. More than one-quarter of the progeny from the self-fertilization of plants with a single functional RUB-encoding gene died as embryos at the two-cell stage. Outcrosses demonstrated reduced inheritance of the null allele from both the male and female parent. Hemigglutinin-tagged forms of RUB1 and RUB2 conjugate to the same cullin protein, CUL1, and produce the same conjugation pattern. To further understand the function of the RUB proteins, a construct designed to produce a double-stranded RUB1 mRNA was introduced into plants, and three lines with reduced levels of RUB1- and RUB2-encoding mRNA and RUB1/2 protein content were analyzed in detail. Mature plants were severely dwarfed, seedlings were insensitive to auxin in root assays, and dark-grown seedlings had a partial triple-response phenotype that was suppressed when seedlings were grown on ethylene perception or synthesis inhibitors. The dsrub lines produced threefold to fivefold more ethylene than the wild type. This study illustrates that RUB1 and RUB2 are genetically and biochemically redundant and demonstrates that RUB1/2 proteins are essential for early embryonic cell divisions and that they regulate diverse processes.

Figure 1.

Authentic RUB1 and RUB2 mRNAs Are Eliminated in T-DNA Insertional Lines and Reduced in dsrub Lines. (A) Genomic representations of RUB1 and RUB2 and of the dsrub construct introduced into plants. Introns are represented by lines and exons by boxes; shaded boxes encode RUB1 or RUB2 protein, black boxes encode ubiquitin, and open boxes represent T-DNA (not to scale). rub1-1 (Col) has a T-DNA insert at 447 bp in intron 1, relative to the ATG, and rub1-2 (Ws) has a T-DNA insert at 128 bp in intron 1. rub2-1 (Col) and rub2-2 (Col) have T-DNA inserts at 387 and 810 bp, respectively. dsrub lines were created with a construct containing the RUB1 open reading frame in opposite directions, separated by an intron, under transcriptional control of the 35S promoter of Cauliflower mosaic virus (CaMV35S). (B) RT-PCR for RUB1, RUB2, and UBQ10 (polyubiquitin) with cDNA from Col (lanes 1 and 2), rub1-1 (lanes 3 and 4), rub2-1 (lanes 5 and 6), rub2-2 (lanes 7 and 8), Ws (lanes 9 and 10), and rub1-2 (lanes 11 and 12) seedlings. Asterisks indicate genomic PCR band for RUB2. The a and b are splice variants. Odd numbered lanes contain PCR reactions made from cDNA using reverse transcriptase, and even numbered lanes contain PCR reactions treated identically but with no reverse transcriptase added. (C) RT-PCR with total RNA from Col (lane 1), dsrub-1 (lane 2), dsrub-2 (lane 3), dsrub-3 (lane 4), and the transgenic control line dsrub-con (lane 5) using primers for RUB1 (top panel), RUB2 (middle panel), and UBQ10 (bottom panel). Lane 6 is identical to lane 1, except reverse transcriptase was not included. PCR using the same primers on Col genomic DNA is shown in lane 7. The numbers indicate size markers in base pairs.

Figure 2.

All Ovules Are Fertilized, but Some Arrest at the Two-Cell Stage. (A) Fully developed siliques from RUB1 rub2 (left), RUB1/rub1-1 rub2 (middle), and rub1 RUB2/rub2 (right). (B) Close up of two ovules from the same silique with the genotype RUB1/rub1-1 rub2 representing the large (left) and small (right) ovules observed. The e indicates location of the embryo, and n indicates location of the endosperm. Bars = 50 μm. (C) Close up of two ovules from the same silique with the genotype rub1-1 RUB2/ rub2 representing the large (left) and small (right) ovules observed. The e indicates location of the embryo, and n indicates location of the endosperm. Bars = 50 μm.

Figure 3.

3HA-RUB1 and 3HA-RUB2 Attach to the Same Proteins. (A) Dex-induced expression of 3HA-RUB1 and 3HA-RUB2 in seedlings has the same conjugation pattern and is unchanged by auxin treatment. Total plant extract (80 μg) from seedlings expressing 3HA-RUB1 (lanes 2 and 3) or 3HA-RUB2 (lanes 4 and 5) treated with 10 μM 2,4-D (lanes 2 and 4) or mock treated (lanes 3 and 5) for 30 min was reacted with anti-HA antibodies. Background bands (marked with asterisks) are determined by electrophoresis of extract from untreated Col seedlings (lane 1). The numbers indicate size markers in kilodaltons. (B) Expression of 3HA-RUB1 and 3HA-RUB2 at high levels leads to CUL1 existing in three forms. Immunoblot (IB) with anti-CUL1 antibodies of extracts from seedlings treated with dex to express 3HA-RUB1 (lane 1) or 3HA-RUB2 (lane 2) for 17 h illustrates the unmodified (fastest band), RUBx-modified (middle band), and 3HA-RUBx-modified (slowest band) forms of CUL1. Samples eluted from anti-HA antibody conjugated beads after incubation in extracts from seedlings dex treated to express 3HA-RUB1 (lane 3) or 3HA-RUB2 (lane 4) for 17 h immunoreacted with anti-CUL1 antibodies creating a band that comigrated with the slowest band from extracts (lanes 1 and 2). (C) 3HA-RUB1 and 3HA-RUB2 attach to CUL1 in a dex-dependent manner. Samples eluted from anti-HA antibody–conjugated beads after incubation in extracts from seedlings dex- or mock-treated for 2 h to express 3HA-RUB1 (lanes 3 and 5) or 3HA-RUB2 (lanes 4 and 6) still maintain a CUL1 band, in a dex-dependent manner. The conjugate pattern of CUL1 in the lines expressing the 3HA-RUBx dex-induced for 2 h is limited to only two bands (lanes 1 and 2).

Figure 4.

RUB1/2 Protein Levels Are Decreased in dsrub Lines. (A) RUB1/2 affinity-purified antibodies specifically react with GST-RUB1. Immunoblot (IB) of purified GST (lane 1) and GST fusion proteins: GST-RUB1 (lane 2), GST-RUB3 (lane 3), and GST-UBQ (lane 4) probed with affinity-purified anti-RUB1/2 antibodies (bottom panel). The antibodies were also tested on 100× (lane 5) and 200× (lane 6) GST-UBQ. An anti-GST IB (top panel) verifies protein levels (lane 6 has 400× GST-UBQ). (B) Affinity-purified anti-RUB1/2 antibodies show specificity against whole plant extract. Immunoblot analysis of purified ubiquitin (UBQ) (lane 1), purified RUB1 (lane 2), Col protein extract (lane 3), and Col protein extract enriched with purified ubiquitin (lane 4). The anti-RUB1/2 antibodies (top panel) detect endogenous RUB1/2 protein, and the anti-ubiquitin antibodies (bottom panel) verify the presence of purified ubiquitin, as well as visualizing endogenous ubiquitin in Col extract (lane 3). (C) Immunoblot with anti-RUB1/2 antibodies on 200 μg of total protein extracted from Col (lanes 2 and 7), dsrub-con (lane 3), dsrub-3 (lane 4), axr1-13 (lane 5), and dsrub-1 (lane 8). Purified RUB1 is used as a positive control (lanes 1, 6, and 9). Lane 7 contains half the protein as lane 8. (D) The dsrub lines have a decreased ratio of unmodified CUL1 to modified CUL1. Immunoblot with anti-CUL1 antibodies of extracts from seedlings of control lines, Col (lane 3) and dsrub-con (lane 2), dsrub-1 (lane 1), dsrub-2 (lane 4), dsrub-3 (lane 5), and axr-13 (lane 6). Coomassie blue stain of identically loaded samples serves as a loading control (bottom panel).

Figure 5.

Growth of dsrub Plants Is Significantly Slower and Overall Size Is Severely Reduced. (A) to (C) Seedlings at 3 weeks: Col (A), dsrub-4 (B), and dsrub-5 (C). Bar = 0.5 cm. (D) to (F) Plants at 5 weeks: Col (D), dsrub-6 (E), and dsrub-5 (F). Bar = 1.0 cm. (G) to (I) Plants at 8 weeks: dsrub-3 (G), dsrub-1 (H), and dsrub-2 (I). Bar = 1.0 cm. (J) Col (left), dsrub-3 (middle), and axr1-13 (right) at 8 weeks. Bar = 2.0 cm.

Figure 6.

dsrub Seedlings Exhibit Insensitivity to Exogenous Auxin. (A) Inhibition of primary root growth by 0.1 μM 2,4-D. The primary root of Col, dsrub-1, dsrub-2, dsrub-3, and axr1-13 seedlings grown on 2,4-D for 10 d is presented as a percentage of the growth of seedlings on germination media (GM) for the same length of time. (B) Increase in the number of lateral roots by 0.1 μM 2,4-D. The number of lateral roots on the primary root of Col, dsrub-1, dsrub-2, dsrub-3, and axr1-13 seedlings grown on GM for 10 d is subtracted from the number on the same primary root grown on 2,4-D or KOH alone for an additional 6 d. Each bar represents the average ± se of two experiments with a minimum of 10 seedlings each. Single asterisks represent lines that are statistically different from Col with a P < 0.0001, and double asterisks represent lines that are statistically different from Col with a P < 0.02.

Figure 7.

Four-Day-Old, Dark-Grown dsrub Seedlings Exhibit a Partial Triple Response That Is Reversed by Inhibitors of the Ethylene Pathway. Seedlings after germination on GM plates: Col (A), dsrub-1 (B), dsrub-2 (C), dsrub-3 (D), and axr1-13 (H); Col on GM with 50 μM ACC (E); dsrub-1 on GM with 100 μM AgNO₃ (F); dsrub-1 on GM with 5 μM AVG (G). The apical hook of dsrub-1 (I) and Col (J) is magnified. Bars = 1 mm.

Figure 8.

Dark-Grown dsrub Seedlings Overproduce Ethylene. Col, axr1-13, dsrub-1, dsrub-2, and dsrub-3 seedlings were germinated and grown in GC vials for 4 d. The amount of ethylene (nL) produced per fresh weight of seedlings (mg) is indicated as the mean ± se of triplicate injections from at least three experiments. All lines are statistically different from Col (Student's t test; P < 0.001).