Identification and developmental expression of Xenopus laevis SUMO proteases

Abstract

SUMO proteins are small ubiquitin-related modifiers. All SUMOs are synthesized as propeptides that are post-translationally cleaved prior to conjugation. After processing, SUMOs become covalently conjugated to cellular targets through a pathway that is similar to ubiquitination. Ubiquitin like protein proteases/Sentrin specific proteases (Ulp/SENPs) mediate both processing and deconjugation of SUMOs. The action of Ulp/SENPs makes SUMOylation a highly dynamic post-translational modification. To investigate how Ulp/SENPs are regulated in a developmental context, we isolated and characterized all Ulp/SENPs in Xenopus laevis. Xenopus possess homologues of mammalian SENP3, 5, 6 and 7. All of these enzymes reacted with HA-tagged vinyl sulfone derivatives of SUMO-2 (HA-SU2-VS) but not SUMO-1 (HA-SU1-VS), suggesting that they act primarily on SUMO-2 and -3. In contrast, Xenopus possess a single member of the SENP1/SENP2 subfamily of Ulp/SENPs, most closely related to mammalian SENP1. Xenopus SENP1 reacted with HA-SU1-VS and HA-SU2-VS, suggesting that it acts on all SUMO paralogues. We analyzed the mRNA and protein levels for each of the Ulp/SENPs through development; we found that they show distinct patterns of expression that may involve both transcriptional and post-transcriptional regulation. Finally, we have characterized the developmental function of the most abundant Ulp/SENP found within Xenopus eggs, SENP3. Depletion of SENP3 using morpholino antisense oligonucleotides (morpholinos) caused accumulation of high molecular weight SUMO-2/3 conjugated species, defects in developing embryos and changes in the expression of some genes regulated by the transforming growth factor beta (TGF-beta) pathway. These findings collectively indicate that SUMO proteases are both highly regulated and essential for normal development.

Figure 1. Abundance and paralogue preference of Ulp/SENP in XEEs.

(A) XEEs were incubated with buffer (left), HA-SU1-VS (middle) or HA-SU2-VS (right), and subjected to SDS-PAGE and immunoblotting with anti-HA antibodies. Free SUMO vinyl sulfones and adducts with individual SENPs are indicated. (B) XEEs were incubated with buffer (left lanes), HA-SU1-VS (middle lanes) or HA-SU2-VS (right lanes). HA-tagged proteins were purified by affinity chromatography, resolved on SDS- PAGE and visualized by Silver staining (left panel) or by immunoblotting with anti-HA antibodies (right panel). Bands recognized by the anti-HA antibodies were identified by mass spectrometry as indicated to the right of each panel. “*” indicates that SENP7 was not identified by mass spectrometry, but could be readily detected by immunoblotting with an antibody specific to SENP7. (C) In vitro translated Flag-SENP5 and Flag-SENP5-C/S were incubated with HA-SU1-VS or HA-SU2-VS. The reactions were monitored by immunoblotting with anti-HA (Upper panel) or anti-Flag (Lower panel) antibodies. “▪” indicates the position of SENP1-VS in the upper panels. Arrow indicates the position of unreacted Flag-SENP5 or Flag-SENP5-C/S.

Figure 2. RT-PCR assay showed that mRNA levels for different SENPs family proteins vary during developmental stages.

Total RNA from whole animals at the indicated stages was analyzed by RT-PCR using specific primers for different SENPs and Histone H4 (control). The interval corresponding to the midblastula transition (MBT) is indicated by the arrow. UF indicates samples from unfertilized eggs, while individual stages are indicated numerically, proceeded by the abbreviation St.

Figure 3. Developmental regulation of Xenopus SENP levels.

For all panels, bands corresponding to unreacted SENPs are indicated with the full name of each protein. SENPs within adducts are indicated with the following abbreviations: S1 = SENP1. S3 = SENP3. S6 = SENP6. S7 = SENP7. SUMO moieties within adducts are indicated with the following abbreviations: SU1 = SUMO-1. SU2 = SUMO-2. UF indicates samples from unfertilized eggs, while individual stages are indicated numerically, proceeded by the abbreviation St. (A) Equal amount of protein from embryos of the indicated stages were incubated with excess HA-SU2-VME. Reactions were subjected to SDS-PAGE and Western blotting with anti-HA or -Actin antibodies. “▪” indicates unidentified bands. (B) Equal amount of protein from embryos of the indicated stages were incubated with excess HA-SU2-VME (right panels) or without HA-SU2-VME (left panels). The reactions were subjected to SDS-PAGE and Western blotting with anti-SENP3, SENP5, SENP6 and SENP7 antibodies. While we did not observe detectable bands on the blot with the anti-SENP5 antibodies, in vitro translated Flag-SENP5 was clearly recognized by this antibody (additional lane in right panel). (C) Equal amount of protein from embryos of the indicated stages were incubated with excess HA-SU1-VS. Reactions were subjected to SDS-PAGE and Western blotting with anti-HA or -Actin antibodies. “▪” Indicates unidentified bands. (D) Equal amount of protein from embryos of the indicated stages were incubated with excess HA-SU2-VME (right panel), HA-SU1-VS (lower left panel) or without addition (upper left panel). The reactions were subjected to SDS-PAGE and Western blotting with anti-SENP1 antibodies.

Figure 4. SENP3 is essential for development of Xenopus laevis embryos.

(A) One cell embryos were injected with morpholinos, and subsequently harvested at Stage 39. Equal amounts of protein from Control-MO (left) or SENP3-MO (right) embryos were subjected to SDS-PAGE and Western blotting with anti-SENP3 or -actin antibodies. (B) Samples as in (A) were subjected to SDS-PAGE and Western blotting with anti-SUMO-2 and –SUMO-1 antibodies, as indicated. Note the substantial accumulation of high molecular weight SUMO-2-conjugated species after SENP3 knockdown, while there is essentially no change in SUMO-1 conjugates. (C) Morphological defects of SENP3-MO injected embryos were categorized according to defects in body patterning and eye development, as indicated (lower three panels). The morphology of embryos injected with Control-MO is shown in the upper panel. (D) Quantitation of morphological body deformities of SENP3-MO injected embryos. Bars indicate standard error of the mean values. p values are indicated.

Figure 5. Inhibition of TGF-Beta signalling by SENP3 knockdown.

(A) One cell embryos were injected with morpholinos, and harvested at Stage 39. The cDNA was generated from total RNA of tadpoles. The levels of expression for the indicated mRNAs were evaluated for Control-MO or SENP3-MO populations by qRT-PCR, and normalized to that of the control gene, elongation factor-1α, and the mean of the Control-MO was set to 1. One non-regulated gene, ornithine decarboxylase (ODC), was analyzed by qRT-PCR to confirm equivalent inputs of total mRNA. For individual genes, three independent samples were analyzed, and all comparisons had p values of <0.05. (B) Schematic representation of BMP signalling. SENP3 depletion disrupts expression of BMP target genes, MSX1 and Vent1. Asterisks indicate genes that are downregulated in the absence of SENP3.