Figure 1.
SMO-1 and NHR-25 physically and genetically interact.
(A) NHR-25 fused to the Gal4 DNA binding domain (DB) interacted with wild type (WT) SMO-1 fused to the Gal4 activation domain (AD). No interaction was seen with empty vector (No insert). β-galactosidase (LacZ) and HIS3 (3AT; 3-aminotriazole) reporters were assayed, and yeast viability was confirmed by growth on a plate lacking leucine and tryptophan (-Leu-Trp). (B) DIC microscopy examining vulval morphology in animals of the indicated genotype. Characteristic protruding vulvae (Pvl) seen in nhr-25(RNAi) and smo-1(ok359) animals are indicated, as is the low penetrance multivulva phenotype (Muv) of smo-1(lf) animals. RNAi inactivation of nhr-25 in a smo-1(ok359) mutant resulted in vulvaless (Vul) animals. (C) Table providing scoring of the Pvl, Vul, Muv, and sterility phenotypes of the indicated genotypes. n = number of animals scored.
Figure 2.
The fully formed vulva of C. elegans is generated post-embryonically from cell divisions of three vulval precursor cells (VPCs). These three VPCs are denoted as P5.p, P6.p and P7.p and undergo a series of stereotyped divisions producing 22 cells. Cells arising from the P6.p precursor are designated as having the primary (1°) fate, while those arising from P5.p and P7.p precursors are designated as having the secondary (2°) fate. 2° cells generate vulA-D cells and 1° cells generate vulE and vulF. In early to mid L3 larvae, the proximity to a gonadal cell known as the anchor cell (AC) initiates vulval patterning by secretion of LIN-3/EGF [64]. The closest VPC to the AC (P6.p) receives the highest LIN-3/EGF dose, which activates LET-60/Ras signaling in P6.p [65]–[68], prompting it to adopt a 1° fate. This EGF-Ras signaling also induces P6.p to express the Notch ligand. The moderate level of LIN-3 received by neighboring P5.p and P7.p cells combined with lateral inhibition through the Notch pathway, induces P5.p and P7.p cells to adopt a 2° fate. In the L3 larval stage, the 1° and 2° cell lineages divide three times, and undergo a coordinated series of migrations and fusions during morphogenesis to complete vulval development [38]. Three VPCs (P3.p, P4.p and P8.p) normally adopt a 3° fate, which means that they divide once and fuse into an epidermal syncytial cell called hyp7, with the exception of P3.p of which about 50% of the lineage fuses without division (designated as S). Syncytial fate is designated S or SS in the figure. The pattern of cell division axes are depicted as L (longitudinal), T (transverse) and U (undivided).
Table 1.
Vulva cell lineage analyses in nhr-25(RNAi) and smo-1(lf) animals.
Figure 3.
Three lysines in NHR-25 are necessary for the interaction with SMO-1.
(A) NHR-25 fused to the Gal4 DNA binding domain (DB) interacted with wild type (WT) SMO-1 fused to the Gal4 activation domain (AD). No interaction was seen with empty vector (No insert), SMO-1 with the terminal di-glycine residues deleted (ΔGG), or SMO-1 with a β-sheet mutation (V31K). (B) The NHR-25 3KR (K165R K170R K236R) allele specifically blocked interaction with SMO-1, as both NHR-25 and NHR-25 3KR interacted with NHR-91. (C) Schematic of NHR-25 domain structure illustrating the DNA binding domain (DBD), hinge region, and ligand binding domain (LBD). The candidate SUMO acceptor lysines (K165, K170, K236) are indicated. (D) Mutating the indicated SUMO acceptor lysines to arginine in NHR-25 only abolished the interaction when all three were mutated (K165R K170R K236R). We note the non-reciprocality of our Y2H interactions: DB-NHR-25 interacted with AD-SMO-1 and AD-NHR-25 interacted with DB-NHR-91. Switching the Gal4 domains did not result in an interaction, as sometimes occurs in Y2H interactions [69]. β-galactosidase (LacZ) reporters were assayed in A, B, and D.
Figure 4.
In vitro sumoylation of NHR-25.
In vitro sumoylation reactions were resolved by SDS-PAGE and analyzed by either Coomassie staining (A,C) or immunoblotting with anti-NHR-25 antibody (B). (A and B) used recombinant human sumoylation enzymes (hE1, hE2, hSUMO1, hSENP1 SUMO protease), (C) used recombinant C. elegans CeUBC-9 and CeSMO-1 with hE1 and hSENP1. Substrates were recombinant 6×His-MBP-NHR-25 (amino acids 161–541; A,C), and the same construct in vitro transcribed and translated (B). In (B) an MBP control was in vitro transcribed and translated, as were the NHR-25 alleles 3KR (K165 K170R K236R) and 3EA(E167A E172A E238A). The positions of NHR-25, sumoylated NHR-25 and AOS1 (part of E1 heterodimer) are indicated. Size markers in kilodaltons (kDa) are provided.
Figure 5.
NHR-25(3KR) displays elevated activity in heterologous reporter assays.
(A) A luciferase reporter vector containing four Ftz-F1/NHR-25 binding sites was transfected into HEK293T cells along with a Renilla internal control and either Myc-NHR-25 (WT) or Myc-NHR-25(3KR) expression constructs. Relative luciferase activity was normalized to the internal Renilla control and empty expression vector (EV). Eight biological replicates from three independent experiments were analyzed and error bars indicate standard deviation. (**T-test p<0.01; *** p<0.0001) (B) Transfected cells were stained with anti-Myc antibody. NHR-25(3KR) does not affect NHR-25 levels or localization. Nuclei were visualized with DAPI staining and an overlay of Myc and DAPI staining is shown.
Figure 6.
The NHR-25 DBD is robustly sumoylated.
(A and B) In vitro sumoylation reactions were resolved by SDS-PAGE and Coomassie stained. A 6×His-MBP-NHR-25 (amino acids 1–173) substrate was used and incubated with hE1, hE2, and either CeSMO-1 (A) or methyl-hSUMO1 (B) for the indicated time in minutes. Methyl-hSUMO1 is a modified protein that blocks SUMO chain formation. Recombinant hSENP1 SUMO protease was included in (B) to demonstrate that bands reflected sumoylated species. A size standard in kilodaltons (kDa) is provided. (C) Schematic of sumoylation sites within NR5 family proteins. DNA-binding domains (DBD), hinge, and ligand-binding domains (LBD) are indicated. Sumoylation sites based on SUMOplot prediction and conservation in multi-species alignments are shaded red. SUMO acceptor lysines confirmed by in vitro biochemistry or cell-based sumoylation assays are shaded blue. The DBD-hinge fragment used in (A and B) is underlined.
Figure 7.
Sumoylation inhibits NHR-25-dependent transcription in vivo.
(A) NHR-25 binds to canonical SF-1 target sequences. Sequence of the wild-type (WT) and mutant (MUT) MIS and CYP11A1 binding sites used are shown on top. Bases altered in the MUT sequences are underlined. The annealed 2×NR5RE oligonucleotides were incubated with combinations of the following: sumoylation enzymes (hUbc9+CeSMO-1) with or without hE1 enzyme, and NHR-25 DBD substrate. Recombinant hSENP1 SUMO protease was included to demonstrate that bands reflected sumoylated species. The corresponding proteins in the EMSA were detected by anti-MBP immunoblotting (input). The positions of unsumoylated and sumoylated NHR-25 DBD are indicated. (B) Animals carrying an 8×NR5RE (WT)::NLS::3×Venus transgene as an extrachromosomal array were generated. No Venus expression was detected in transgenic animals on vector RNAi (i and ii). The nematode body is outlined in (i), and the corresponding differential interference contrast (DIC) image of the same animal is provided (ii). Representative Venus expression in transgenic animals treated with smo-1 RNAi (iii–vii). Expression was observed in seam cells at L4 (iii), in seam cells and hyp7 at L3 (iv), in hyp7 at L4 (v), in the AC and vulF at early L4 (vi), and in developing vulval cells at L3 (vii). Fluorescent and DIC images were merged in vi and vii. (C and D) Transgenic animals expressing the Venus reporter in at least one of the following tissues: seam cells, hyp7, or vulval cells; were scored. Reduction of nhr-25 function either by RNAi (C) or by ku217 mutation (D) reduced the 8×NR5RE (WT) reporter activity following smo-1 RNAi. (n) number of animals scored. (E) Mutations (MUT) in NR5RE completely eliminated Venus expression following smo-1 RNAi. DIC (i and iii) images corresponding to Venus fluorescence images (ii and iv, respectively) are provided. Positions of hypodermal nuclei (ii) and the developing vulva (iv) are outlined. (F) NHR-25::GFP is expressed in nuclei of seam cells and hyp7 (i) and the developing vulva (ii), similar to 8×NR5RE (WT)::NLS::3×Venus reporter expression. All animals are positioned with the anterior to the left.
Figure 8.
Overexpression of unsumoylated NHR-25 causes multivulva induction.
(A) Table providing scoring of overall multivulva (Muv) induction in the indicated strains/genotypes, as well as induction in individual VPCs. Number of animals (n) scored for each strain genotype is provided. Use of brackets denotes transgenic genotypes. (B) Graphical representation of the overall percentage of animals for each strain that display Muv induction of any VPC.
Figure 9.
NHR-25::GFP (OP33) expression during vulval development.
Expression in 1-cell stage Pn.p cells (A), in 2-cell stage Pn.px cells (B) and 4-cell stage Pn.pxx cells (C) in wild type and in smo-1(RNAi) animals (E). Higher levels and ectopic expression of NHR-25 were seen in P4.px and P8.px(x) in a smo-1(RNAi) background (E). Expression at the bell stage in wild type and smo-1(RNAi) animals (D,F). Ectopic expression in the AC observed in smo-1(RNAi) animals. Arrowheads indicate the position of the AC, red asterisk indicates the position of the invaginated vulva. Colored bars indicate 1° (red), 2° (yellow), and 3° (blue) lineages, as described in Figure 2.
Figure 10.
Ratio of sumoylated to unsumoylated NHR-25 and 3° cell fate.
After the first round of cell division, VPCs adopt 1°, 2°, and 3° fates and NHR-25 accumulates in a gradient. The highest NHR-25 levels are in 1° fated cells, lower NHR-25 levels are in 2° fated cells, and the lowest levels are in 3° fated cells. Sumoylation output is a reflection of the combined activities of the sumoylation machinery and the SUMO proteases. In this model, sumoylation output is limiting, and the NHR-25 gradient results in a gradient of unsumoylated NHR-25. 1° cells have the highest ratio of unsumoylated to sumoylated NHR-25, and the ratio decreases as NHR-25 levels drop in 2° and 3° VPCs. The dashed line indicates the constant amount of sumoylated NHR-25 produced by limiting, steady-state sumoylation. At a particular threshold, enough sumoylated NHR-25 relative to unsumoylated NHR-25 allows 3° cells to either adopt and/or maintain the correct fate.