Genetic analysis of SUMOylation in Arabidopsis: conjugation of SUMO1 and SUMO2 to nuclear proteins is essential

Abstract

The posttranslational addition of small ubiquitin-like modifiers (SUMOs) to other intracellular proteins has been implicated in a variety of eukaryotic functions, including modifying cytoplasmic signal transduction, nuclear import and subnuclear compartmentalization, DNA repair, and transcription regulation. For plants, in particular, both genetic analyses and the rapid accumulation of SUMO conjugates in response to various adverse environmental conditions suggest that SUMOylation plays a key role in the stress response. Through genetic analyses of various SUMO conjugation mutants, we show here that the SUMO1 and SUMO2 isoforms, in particular, and SUMOylation, in general, are essential for viability in Arabidopsis (Arabidopsis thaliana). Null T-DNA insertion mutants affecting the single genes encoding the SUMO-activating enzyme subunit SAE2 and the SUMO-conjugating enzyme SCE1 are embryonic lethal, with arrest occurring early in embryo development. Whereas the single genes encoding the SUMO1 and SUMO2 isoforms are not essential by themselves, double mutants missing both are also embryonic lethal. Viability can be restored by reintroduction of SUMO1 expression in the homozygous sum1-1 sum2-1 background. Various stresses, like heat shock, dramatically increase the pool of SUMO conjugates in planta. This increase involves SUMO1 and SUMO2 and is mainly driven by the SUMO protein ligase SIZ1, with most of the conjugates accumulating in the nucleus. Taken together, it appears that SIZ1-mediated conjugation of SUMO1 and SUMO2 to other intracellular proteins is essential in Arabidopsis, possibly through stress-induced modification of a potentially diverse pool of nuclear proteins.

Figure 1.

Tissue-specific expression of SUMO and SUMO-conjugating enzymes. A and B, Transcript profiles of SUMOs and SUMO-conjugating enzymes by DNA microarray expression analysis. Data were extracted from Genevestigator (Zimmermann et al., 2004) for SUM1 to 3 and SUM5 (A), and SAE1a, SAE1b, SAE2, SCE1, and SIZ1 (B). C, Levels of SUMO, SUMO conjugates, and SUMO-conjugating enzymes in various Arabidopsis tissues. Tissues were extracted directly into 2 mL/g fresh weight of SDS-PAGE sample buffer, and equal volumes were subjected to SDS-PAGE and immunoblot analysis with anti-SUMO1, SAE1a, and SCE1 antibodies. An overexposure of the SUMO conjugate blot was included to better show the levels of free SUMO. Immunoblot analysis with antibodies against the 26S proteasome subunit PBA1 was included to show protein loading.

Figure 2.

Characterization of the sae1a-1 mutant. A, Gene diagram of the SAE1a and SAE1b genes. Exons and introns are represented by boxes (black, coding region; white, UTR) and lines, respectively. Number of amino acids in each protein is indicated on the right. Insertion positions of the sae1a-1 and sae1a-2 T-DNA mutants are shown. Arrows show the positions of the RT-PCR primers used in B. B, RT-PCR analysis of wild-type Col-0 and homozygous sae1a-1 seedlings. RNA was subjected to first-strand cDNA synthesis with primers specific to either the 3′-UTR of SAE1a (primer 5) or an internal site in SAE1a (primer 4) and then to PCR using the indicated gene-specific primers. RT-PCR of the PAE2 mRNA encoding the 26S proteasome α5 subunit and genomic PCR of the SAE1a loci were included as controls. C, RNA gel-blot analysis with a SAE1a probe of total RNA extracted from wild-type Col-0 and homozygous sae1a-1 plants. RNA gel-blot with a β-TUB4 probe was used to confirm equal RNA loading. D, Immunoblot detection of the SAE1 protein. Protein extracts from 7-d-old wild-type Col-0 and homozygous sae1a-1 plants were separated by SDS-PAGE and subjected to immunoblot analysis with affinity-purified anti-SAE1a antibodies. The anti-PBA1 antibody blot was included to confirm equal protein loading. E, Affinity-purified SAE1a antibodies preferentially recognize recombinant SAE1a protein. The indicated amount of recombinant SAE1a and SAE1b were separated by SDS-PAGE and either subjected to immunoblot analysis with anti-SAE1a antibodies (top) or silver-stained for protein (bottom).

Figure 3.

Genetic analysis of the single genes encoding the SAE2 subunit of the SUMO-activating enzyme and the SUMO-conjugating enzyme SCE1. A, Gene diagram of SAE2 and SCE1. Exons and introns are represented by boxes (black, coding region; white, UTR) and lines, respectively. Number of amino acids in each protein is indicated on the right. Insertion positions of the various T-DNA mutants are shown; insertions that result in null alleles are depicted in red, whereas insertions that produce viable homozygous plants are in green. Positions of the active-site Cys (C) are marked with arrowheads. B and C, Exon insertion mutants of SAE2 and SCE1 are embryonic lethal. B, Immature siliques from self-fertilized wild-type Col-0, SAE2/sae2-1, SCE1/sce1-5, and SCE1/sce1-6 plants. White and shriveled brown seeds have undergone developmental arrest. C, Microscopic examination of embryos from green seeds that presumably contain at least one wild-type allele of SAE2 or SCE1 and white/brown seeds that are presumably homozygous for the sae2-1 or sce1-5 mutations from the same heterozygous sae2-1 or sce1-5 silique. Predicted genotypes are indicated. D, RNA gel-blot analysis with a SCE1 probe of total RNA extracted from wild-type Col-0 and homozygous sce1-4 and sce1-7 plants. RNA gel-blot with a β-TUB4 probe was used to confirm equal RNA loading. E, Homozygous sce1-4 and sce1-7 seedlings have reduced levels of SCE1 proteins and accumulate less SUMO conjugates during heat stress. Seven-day-old liquid-grown wild-type Col-0, sce1-4, and sce1-7 plants were either kept at 24°C or subjected to 30-min heat shock at 37°C and returned to 24°C to recover for 30 min. Total protein extracts were separated by SDS-PAGE and subjected to immunoblot analysis with anti-SUMO1, SCE1a, and PBA1 antibodies. SUMO conjugates are indicated by the bracket.

Figure 4.

SUMO1 and SUMO2 are essential in Arabidopsis. A, Gene diagram of the SUM1 and SUM2 genes and location of the T-DNA insertions (arrowheads) in sum1-1 and sum2-1. Exons and introns are represented by boxes (black, coding region; white, UTR) and lines, respectively. Number of amino acids in each protein is indicated on the right. Arrows show the positions of the RT-PCR primers used in B. B, RT-PCR analysis of wild-type Col-0 and homozygous sum1-1 and sum2-1 seedlings. RNA was subjected to first-strand cDNA synthesis with primers specific to either the 3′-UTR of SUM1 or SUM2 and then to PCR using the indicated gene-specific primers. RT-PCR of the PAE2 mRNA and genomic PCR of the SUM1 and SUM2 loci were included as controls. C, Double-homozygous sum1-1 sum2-1 plants are embryonic lethal. Shown are immature siliques from self-fertilized homozygous sum1-1 and sum2-1 plants and a self-fertilized plant that was heterozygous for sum1-1 and homozygous for sum2-1. White and shriveled brown seeds have undergone developmental arrest. D, Microscopic examination of embryos from green seeds that presumably contain at least one wild-type allele of SUM1 and white/brown seeds that are presumably homozygous for the sum1-1 from the same silique from a SUM1 sum1-1 sum2-1 sum2-1 plant. Predicted genotypes are indicated.

Figure 5.

Heat shock-induced accumulation of SUMO conjugates in various mutant combinations affecting SUM1 and SUM2. Seedlings were grown in liquid medium, subjected to 37°C heat shock for 30 min, and then returned to 24°C to recover. Seedlings collected at the indicated times were homogenized directly in SDS-PAGE sample buffer and the extracts were separated via SDS-PAGE and subjected to immunoblot analysis with anti-SUMO1 antibodies. An overexposure of the SUMO conjugate blot was included to better show the levels of free SUMO. The anti-PBA1 antibody blot was included to confirm equal protein loading. SUMO conjugates are indicated by the bracket.

Figure 6.

siz1 mutants show dramatic growth defects at maturity. Plants with the indicated genotypes were grown under SD for 10 weeks (top) or under LD for 8 weeks (bottom).

Figure 7.

The SIZ1 SUMO protein ligase is required for heat shock-induced conjugation of SUMO1/2. Plants of the indicated genotypes were grown for 7 d on solid medium, then transferred to liquid culture for 4 d and subjected to 37°C heat shock for 30 min, and then returned to 24°C to recover. Seedlings collected at the indicated times were homogenized directly in SDS-PAGE sample buffer and the extracts were subjected to immunoblot analysis with anti-SUMO1, SAE1a, and SCE1 antibodies. Anti-PBA1 blot was used to confirm equal protein loading. SUMO conjugates are indicated by the bracket. A, Kinetics of SUMO conjugate accumulation following 37°C heat shock. B, Direct comparison of SUMO conjugate levels. Seedlings were either kept at 24°C or subjected to 30-min heat shock at 37°C and returned to 24°C to recover for 30 min.

Figure 8.

Complementation of the sum1/sum1 sum2/sum2 mutant with SUM1. A, Diagrams of SUM1 and SUM2 loci and the SUM1 transgene (SUM1-T) used for complementation. Dashed black lines represent UTRs, gray boxes represent promoter regions, white boxes indicate genomic regions, solid lines represent introns, and black boxes represent exons. Hatched boxes represent vector DNA. T-DNA insertions are marked by triangles. Numbered arrows show the positions of primers used for PCR genotyping in B. B, Genomic PCR of wild-type Col-0, and heterozygous and homozygous sum1-1 sum2-1 plants with and without SUM1-T. C, Heat shock-induced accumulation of SUMO conjugates in the double-homozygous sum1-1 sum2-1 plants harboring SUM1-T as compared to wild-type Col-0 and sum1-1 SUM1 sum2-1 sum2-1 plants. Seedlings were either kept at 24°C or subjected to 30-min heat shock at 37°C and returned to 24°C to recover for 30 min.

Figure 9.

Heat-induced SUMO conjugates are nuclear localized. Seven-day-old liquid-grown Col-0 seedlings were either kept at 24°C or heat shocked at 37°C for 30 min. Nuclei (N) were purified away from cytosol (S) by Percoll gradient centrifugation of total extracts (T) and the resulting samples were subjected to SDS-PAGE and immunoblot analysis with anti-SUMO1, histone-3 (H3), PUX1, or Rubisco large subunit antibodies. The asterisks identify potential dimeric and trimeric chains of free SUMO of 36 and 50 kD in the heat-shocked samples. Thirty-three micrograms of protein were loaded into each of the T and S lanes and 12 μg were loaded into each of the N lanes.