pICln associates with several cytosolic proteins in vivo, including Sm proteins. (Left panel) MDCK cell cytosolic proteins purified by binding to immobilized GST (lane 2) or GST-pICln (lane 3). Lane 1, molecular size standards. To better visualize nonspecifically bound proteins, lane 2 was loaded with five times more sample than lane 3. (Middle panel) Proteins coimmunoprecipitated with endogenous pICln from the cytosolic fraction of [³⁵S]methionine-labeled MDCK cells. (Right panel) MDCK cell cytosolic extract was immunoprecipitated with a monoclonal anti-Sm antibody and immunoblotted with a polyclonal anti-pICln antibody.

pICln is not a component of snRNPs. (a) SnRNPs were immunoprecipitated from MDCK cell cytosolic (C) or nuclear (N) extracts with TMG-specific antibody or pICln-specific antibody. The immunoprecipitate, the supernatant (Sup), and the untreated extract were analyzed by immunoblotting with a monoclonal pICln-specific antibody. (b) MDCK cell cytosolic (C) and nuclear (N) extracts were immunoprecipitated with antibodies specific for pICln, TMG, Sm, and an unrelated antigen. Ctrl Ab, control antibody. RNA from the immunoprecipitates and from untreated extract was deproteinized, electrophoresed on a denaturing polyacrylamide gel, and transferred to a nylon membrane. U1 RNA was detected by hybridization to a U1-specific probe. U1 content in the pICln immunoprecipitate was negligible and indistinguishable from the control immunoprecipitate, in contrast to prominent U1 signals in TMG and Sm immunoprecipitates.

pICln inhibits Sm binding to U RNA. (a) Oocytes were injected with buffer, GST, or GST-xICln, followed by a mixture of ³²P-labeled U1, U2, U4, U5, and U6 RNA. The oocytes were then homogenized, and Sm-associated U RNA was immunoprecipitated with the anti-Sm antibody Y12. The immunoprecipitate and supernatant were deproteinized and analyzed on a denaturing polyacrylamide gel. Each sample represents a pool of four oocytes. Ctrl indicates the mixture of U RNAs prior to injection. The different amounts of each U RNA reflect different amounts of U RNA produced from each template in in vitro transcription reactions. (b) Quantitation of the results of panel a. The amount of U RNA present in the immunoprecipitate and supernatant was quantitated by measuring radioactivity in gels on a phosphorimager. The percentage of U RNA found in the precipitate of pICln-injected oocytes was plotted relative to the percentage of U RNA in the precipitate in the buffer-injected control oocytes. •, significant inhibition due to GST-xICln injection compared to buffer or GST controls (P < 0.03). Each column represents the average (error bar, standard error of the mean) for five (GST) or six (buffer, GST-xICln) samples. (c) Bandshift analysis of Sm binding to U RNA. Oocytes were injected with GST (G) or GST-xICln (I) and subsequently injected with either ³²P-labeled U1 (1), U5 (5), or U1Δ (Δ), a U1 mutant that lacks the Sm binding site. The oocytes were homogenized, and either buffer, anti-Sm antibody, or anti-pICln antibody was added. An aliquot was analyzed by nondenaturing polyacrylamide gel electrophoresis (top panel). RNA was recovered from another aliquot of the same bandshift reaction mixture and analyzed by denaturing gel electrophoresis (bottom panel). The square bracket indicates U RNA with reduced mobility due to protein binding. The arrow indicates complexes with very low mobility due to antibody binding. The positions of free U1, U5, and U1Δ probes are shown.

pICln interferes with SMN binding to Sm proteins. (a) Fusion proteins used, separated on an SDS–10% polyacrylamide gel and stained with Coomassie blue. (b) In vitro-transcribed and -translated, ³⁵S-labeled SmB′, SmD1, and SmD3 were preincubated with 12 μg of either Trx, Trx-ICln, or Trx-SMN. The mixture was then incubated with 1 μg of either GST, GST-SMN, or GST-ICln bound to glutatione beads, and the beads were vigorously washed. Bound Sm proteins were analyzed by SDS-polyacrylamide gel electrophoresis. (c) Quantitation of the results in panel b. A phosphorimager was used to measure the amount of each Sm protein bound in the presence of Trx or inhibitor protein (Trx-SMN or Trx-ICln). Relative binding was calculated as the percent of Sm protein bound in the presence of inhibitor compared to the amount bound in the presence of Trx (i.e., binding in the presence of Trx was taken as 100%). Each column represents the average (error bar, standard deviation) of five to six independent measurements.

pICln interacts with Sm proteins in vitro. (a) (Left panel) ³⁵S-labeled Sm protein in vitro translation products input into binding assays. (Right panel) GST-ICln fusion proteins used in binding assays, separated on an SDS–10% polyacrylamide gel and stained with Coomassie blue. GST-hICln deletion mutants are labeled with the numbers of the pICln amino acids present in the fusion protein. (b) GST or GST-ICln immobilized on glutathione-Sepharose beads was incubated with in vitro-translated ³⁵S-labeled Sm. After extensive washing, bound proteins were analyzed by polyacrylamide gel electrophoresis and visualized by using a phosphorimager. In vitro translation of SmD3 consistently resulted in two products, both of which bound to GST-ICln. (b) Deletion analysis of pICln-Sm interaction. In vitro-translated SmB′, SmD1, or SmD3 was incubated with deletion mutants of pICln fused to GST and immobilized on glutathione-Sepharose beads. Input levels of in vitro-translated proteins are shown in panel a, and binding to GST or GST-ICln from the same experiment is shown in panel b. ³⁵S-labeled proteins were visualized by using a phosphorimager.

The majority of cytosolic SmB/B′ is bound to pICln. (a) pICln was immunoprecipitated from MDCK cell cytosolic extract with a purified polyclonal anti-pICln antibody. More than 95% of pICln was immunoprecipitated, as demonstrated by immunostaining the supernatant with an anti-pICln antibody (top panel). Immunostaining the pICln immunoprecipitate with anti-Sm protein antibody shows that SmB/B′ coimmunoprecipitated with pICln (bottom panel). An unrelated antibody (Ctrl) did not immunoprecipitate pICln or SmB/B′. (b) Immunodepletion of pICln results in depletion of SmB/B′. Cytosolic extract from MDCK cells was immunodepleted of pICln as shown in panel a. SmB/B′ content in the extract prior to (No antibody) and after immunoprecipitation with pICln or unrelated antibody was assessed by Western blotting with Y12 antibody (insert) and quantitated by densitometry in four independent experiments. The amount of SmB/B′ in pICln-depleted samples was normalized to the amount in control samples in each experiment. Treatment of sample with an unrelated antibody and protein A beads did not change SmB/B′ content in the extract. Bars indicate standard errors of the means.

pICln inhibits nuclear import of U RNA. (a) Oocytes were injected with buffer, GST, or GST-xICln and then with ³²P-labeled, in vitro-transcribed U1, U2, U4, and U5 RNAs. RNA was recovered from nuclear (N) and cytoplasmic (C) fractions and analyzed by denaturing polyacrylamide gel electrophoresis. Each sample represents a pool of five oocytes. The lane labeled control contains the mixture of RNAs prior to injection. U RNAs imported into the nucleus were slightly smaller than the injected U RNAs due to 3′ end trimming (27). The standard is a labeled RNA that was added in equal amounts to the nuclear and cytoplasmic fractions following dissection to ensure equal recovery during processing of the samples. (b) Quantitation of the effect of GST-xICln on individual U RNAs. The amount of each U RNA in the nuclear and cytoplasmic fractions of GST or GST-xICln injected oocytes was determined by analyzing gels, such as shown in panel a, on a phosphorimager. Results are displayed as the percentage of nuclear import of each U RNA in GST-xICln injected oocytes relative to GST injected oocytes. Each column represents the average (error bar, standard deviation) of two independent experiments.

Model of pICln and SMN regulation of Sm protein assembly on U RNA. pICln is depicted as inhibiting Sm core domain assembly and U RNA nuclear import by preventing Sm interaction with SMN. Hypermethylation and nuclear import of U RNA occur after Sm core domain assembly, although the timing of these events relative to SMN dissociation is not known.