Phosphoinositide signaling pathways in nuclei are associated with nuclear speckles containing pre-mRNA processing factors

Abstract

Phosphoinositide signal transduction pathways in nuclei use enzymes that are indistinguishable from their cytosolic analogues. We demonstrate that distinct phosphatidylinositol phosphate kinases (PIPKs), the type I and type II isoforms, are concentrated in nuclei of mammalian cells. The cytosolic and nuclear PIPKs display comparable activities toward the substrates phosphatidylinositol 4-phosphate and phosphatidylinositol 3-phosphate. Indirect immunofluorescence revealed that these kinases were associated with distinct subnuclear domains, identified as "nuclear speckles," which also contained pre-mRNA processing factors. A pool of nuclear phosphatidylinositol bisphosphate (PIP2), the product of these kinases, was also detected at these same sites by monoclonal antibody staining. The localization of PIPKs and PIP2 to speckles is dynamic in that both PIPKs and PIP2 reorganize along with other speckle components upon inhibition of mRNA transcription. Because PIPKs have roles in the production of most phosphatidylinositol second messengers, these findings demonstrate that phosphatidylinositol signaling pathways are localized at nuclear speckles. Surprisingly, the PIPKs and PIP2 are not associated with invaginations of the nuclear envelope or any nuclear membrane structure. The putative absence of membranes at these sites suggests novel mechanisms for the generation of phosphoinositides within these structures.

Figure 1

anti-PIPKIα and anti-PIPKIIα polyclonal antibodies (10 μg/ml) selectively detected 68- and 53-kDa proteins, respectively, by Western blotting total lysates prepared from NRK-49F, 2RA, and HeLa cells.

Figure 2

Indirect immunofluorescence of NRK-49F rat fibroblasts indicated nuclear localization of PIPKIα (A–C) and PIPKIIα (D–F). The speckled staining in nuclei was observed with methanol fixation (A and D), or with a brief preextraction using 0.2% Triton X-100 and then fixation with formaldehyde (B and E). Strong diffuse nuclear staining upon formaldehyde fixation was obtained when anti-PIPK antibodies were used at 10 μg/ml (C and F). However, this picture at lower concentrations progressed to speckles and then resolved into a pattern of smaller dots when 0.5 μg/ml anti-PIPKIα or 1 μg/ml anti-PIPKIIα antibodies were used (C and F, insets). Insets, thin optical sections of magnified view of the nuclei of human MG-63 cells. Fixation and staining protocols are detailed in MATERIALS AND METHODS. Bar, 10 μm.

Figure 3

Subcellular fractionation of HeLa cells showed that nuclei contained PIPKIα and PIPKIIα. HeLa cells were disrupted and separated into nuclear and crude cytosolic (cytosol) fractions by low-speed sedimentation. The nuclei were membrane stripped with 0.8% Triton X-100, washed, and pelleted (stripped nuclei). Stripped nuclei were then treated with 0.35 M KCl and separated into soluble (nuclear extract) and insoluble (postextracted nuclei) fractions by high-speed centrifugation. Equal amounts of protein from each fraction were used to do Western blots for both PIPKIα and PIPKIIα (A). The purity of the cellular fractions was assessed by reprobing the Western blots (B) with antibodies toward proteins from the ER, cytosol, and plasma membrane (PM), as described in MATERIALS AND METHODS.

Figure 4

Nuclear and cytosolic PIPKs exhibited similar kinase activities toward PI4P and PI3P as substrates. PIPKIα (A) and PIPKIIα (B) could be selectively immunoprecipitated from HeLa cytosol or nuclear extract. (A) PIPKIα was immunoprecipitated from either 200 μg of cytosol or 100 μg of nuclear extract with the rabbit anti-PIPKIα polyclonal antibody, followed by Western blotting with the PIPKIα C-terminal isoform-specific antibody. (B) The goat N19 peptide PIPKIIα isoform-specific antibody was used to immunoprecipitate PIPKIIα from 130 μg of cytosol or 65 μg of nuclear extract. PIPKIIα was then detected by blotting with the rabbit anti-PIP5KIIα polyclonal antibody. (C) PIPKIα and PIPKIIα, immunoprecipitated from 400 μg of cytosol or 200 μg of nuclear extract, were assayed for lipid kinase activity toward either PI4P or PI3P. The assays shown are representative of two or three immunoprecipitations from two different nuclear preparations. As controls, mock immunoprecipitations (mock IP) were performed in the absence of the antibody, and a sample of the antibody used for the immunoprecipitation was Western blotted (IgG).

Figure 5

PIPKs colocalized with components of the mRNA-processing machinery in nuclear speckles. Methanol-fixed rat NRK fibroblasts were double-labeled with anti-PIPKIα or anti-PIPKIIα polyclonal antibodies and human Sm antiserum (recognizes components of small nuclear RNA-binding proteins). Thin optical sections obtained by confocal scanning laser immunofluorescence microscopy are shown. Colocalization is represented by yellow in the overlays. Bar, 10 μm.

Figure 6

Epitope-tagged PIPKIIα and PIPKIIβ localized to nuclei and nuclear speckles when expressed in cultured fibroblasts. Indirect immunofluorescence was performed on formaldehyde-fixed human 2RA fibroblasts transiently overexpressing FLAG epitope-tagged PIPKIIα or PIPKIIβ. Double labeling with anti-PIPKIIα or PIPKIIβ antibodies and anti-FLAG M2 antibodies showed primarily diffuse nuclear localization of the expressed kinases (top row). Localization of overexpressed kinases to the speckles was revealed by a very brief preextraction of the cells with 0.2% Triton X-100 (bottom row) and staining with anti-FLAG antibody. Speckles were costained with the human Sm serum. Bar, 10 μm.

Figure 7

PIPKs colocalized with PIP₂ in nuclear speckles. Prepermeabilized, formaldehyde-fixed human 2RA fibroblasts were double labeled with anti-PIPKIα or anti-PIPKIIα polyclonal antibodies and anti-PIP₂ mAb AM212 (Miyazawa et al., 1988). Thin optical sections were obtained by confocal laser scanning microscopy. Colocalization is represented by yellow in the overlays. Bar, 10 μm.

Figure 8

The specificity of the nuclear PIP₂ signal was demonstrated by complete inhibition of staining after preincubation of the AM212 mAb with excess of PI4,5P₂ liposomes but only partial inhibition by PI4P or PI3,4P₂. Bar, 10 μm.

Figure 9

PIPKs and PIP₂ were not associated with invaginations of the nuclear envelope. Invaginations of the nuclear envelope can transverse nuclei (Fricker et al., 1997) and produce a series of dots (arrows) seen here in thin optical sections of methanol-fixed human 2RA fibroblasts labeled with biotin-conjugated Con A (a lectin that binds mannose residues of nuclear envelope glycoproteins). Thin optical sections of the triple-labeling with anti-PIPKIα (A) or anti-PIPKIIα (E) polyclonal antibodies, Con A (C and G) and anti-PIP₂ mAb AM212 (B and F) are shown. The overlay of PIPK and Con A staining patterns is shown in yellow (D and H). Bar, 10 μm.

Figure 10

Association of PIPKs and PIP₂ with nuclear speckles is dynamic. Treatment of 2RA cells with a 10 μg/ml concentration of the transcriptional inhibitor α-amanitin for 4 h caused reorganization of splicing-related nuclear speckles as detected by Sm antiserum into a few large dots (A, compare with Figure 5). Likewise, treatment of NRK cells with a 100 μM concentration of the transcriptional inhibitor DRB for 4 h caused reorganization of Sm speckles into large dots or a scattered array of small dots (B). Cells were triple labeled to show the changes in PIPKIα (A), PIPKIIα (B), and PIP₂ distribution upon treatment with the inhibitors. The cells were fixed with 4% formaldehyde after 0.2% Triton X-100 preextraction. The PIPK and Sm staining patterns were overlaid (yellow) to demonstrate colocalization. Bar, 10 μm.