Figure 1.

Genomic architecture and allelic organization of the D. melanogaster Smn gene, and its role in snRNP assembly. (A) D. melanogaster Smn is a single-exon gene. Smn^73Ao (herein referred to as Smn^A) and Smn^B are missense mutations in the conserved Y-G box described previously (Chan et al., 2003). Transposon insertions are marked by open triangles. Smn^C and Smn^D are piggyBac insertions at +407 and +58 bp from the translation start, respectively. Smn^E is a P element insertion at −94 bp, which is upstream of the putative transcription start site. Smn^E2 and Smn^E33 are imprecise excision alleles derived from the mobilization of Smn^E. (B) Western blot of embryonic and adult lysates. A dilution series of the embryonic lysate indicated that adults have ∼30-fold less dSMN than embryos. Anti-SNF antibody was used as the loading control. (C) S2 cells were left untreated (Mock) or incubated with dsRNAs targeting either LacZ or Smn. Cytoplasmic extracts were collected 3 d after treatment, and Western blotting confirmed efficient knockdown of dSMN compared with the loading control. (D) Radiolabeled U1 snRNA transcripts were incubated in cytoplasmic extracts and immunoprecipitated with α-Sm monoclonal antibody Y12 to assay for Sm-core assembly. Incubations of wild-type U1 snRNA (+) at the nonpermissive temperature (4°C) or a U1 construct containing a deletion in the Sm protein binding site (Δ) at the permissive temperature (22°C) served as the negative controls. RNAi knockdown of dSMN significantly disrupted Sm-core assembly compared with mock and LacZ dsRNA transfections. (E) After dsRNA treatment for 6 d (two doses of dsRNA), S2 cells were transfected with either GFP alone or GFP-SmB. Immunoprecipitation using anti-GFP antibodies, followed by Northern analysis of U1 and U2 snRNAs, indicated that GFP alone did not bring down detectable amounts of snRNA, the amounts immunoprecipitated by GFP-SmB after RNAi knockdown of dSMN were at least twofold less than in the control (LacZ) knockdown. (F) Expression profile of dSMN in the lethal alleles described in A. All of the lethal alleles are essentially protein nulls, although residual levels of dSMN in lysates derived from the Smn^A and Smn^B alleles varied from preparation to preparation (not depicted). Anti-SNF antibody was used as the loading control. We note that wild-type embryonic lysates were competent for the Sm-core assembly assay shown in D, but larval, pupal, and adult lysates were incompetent, which is consistent with previous results from other species (Gabanella et al., 2005; Wan et al., 2005). (G) Northern blot of total larval RNAs from wild-type (WT), Smn^C homozygous (Smn^C), or heterozygous (Het^C) mutants. 2 μg of RNA were loaded in each lane; comparable levels of U1, U2, and U4 snRNAs were detected (U3 was used as a control). Molecular weight markers are in kilodaltons.