Non-erythroid protein 4.1R (4.1R) consists of a complex family of isoforms. We have shown that 4.1R isoforms localize at the mitotic spindle/spindle poles and associate in a complex with the mitotic-spindle organization proteins Nuclear Mitotic Apparatus protein (NuMA), dynein, and dynactin. We addressed the mitotic function of 4.1R by investigating its association with microtubules, the main component of the mitotic spindles, and its role in mitotic aster assembly in vitro. 4.1R appears to partially co-localize with microtubules throughout the mitotic stages of the cell cycle. In vitro sedimentation assays showed that 4.1R isoforms directly interact with microtubules. Glutathione S-transferase (GST) pull-down assays using GST-4.1R fusions and mitotic cell extracts further showed that the association of 4.1R with tubulin results from both the membrane-binding domain and C-terminal domain of 4.1R. Moreover, 4.1R, but not actin, is a mitotic microtubuleassociated protein; 4.1R associates with microtubules in the microtubule pellet of the mitotic asters assembled in mammalian cell-free mitotic extract. The organization of microtubules into asters depends on 4.1R in that immunodepletion of 4.1R from the extract resulted in randomly dispersed microtubules. Furthermore, adding a 135-kDa recombinant 4.1R reconstituted the mitotic asters. Finally, we demonstrated that a mitotic 4.1R isoform appears to form a complex in vivo with tubulin and NuMA in highly synchronized mitotic HeLa extracts. Our results suggest that a 135-kDa non-erythroid 4.1R is important to cell division, because it participates in the formation of mitotic spindles and spindle poles through its interaction with mitotic microtubules.

The nonerythrocyte isoform of the cytoskeletal protein 4.1R (4.1R) is associated with morphologically dynamic structures during cell division and has been implicated in mitotic spindle function. In this study, we define important 4.1R isoforms expressed in interphase and mitotic cells by RT-PCR and mini-cDNA library construction. Moreover, we show that 4.1R is phosphorylated by p34cdc2 kinase on residues Thr60 and Ser679 in a mitosis-specific manner. Phosphorylated 4.1R135 isoform(s) associate with tubulin and Nuclear Mitotic Apparatus protein (NuMA) in intact HeLa cells in vivo as well as with the microtubule-associated proteins in mitotic asters assembled in vitro. Recombinant 4.1R135 is readily phosphorylated in mitotic extracts and reconstitutes mitotic aster assemblies in 4.1R-immunodepleted extracts in vitro. Furthermore, phosphorylation of these residues appears to be essential for the targeting of 4.1R to the spindle poles and for mitotic microtubule aster assembly in vitro. Phosphorylation of 4.1R also enhances its association with NuMA and tubulin. Finally, we used siRNA inhibition to deplete 4.1R from HeLa cells and provide the first direct genetic evidence that 4.1R is required to efficiently focus mitotic spindle poles. Thus, we suggest that 4.1R is a member of the suite of direct cdc2 substrates that are required for the establishment of a bipolar spindle.

The tightly regulated production of distinct erythrocyte protein 4.1R isoforms involves differential splicing of 3 mutually exclusive first exons (1A, 1B, 1C) to the alternative 3 splice sites (ss) of exon 2/2. Here, we demonstrate that exon 1 and 2/2 splicing diversity is regulated by a transcription-coupled splicing mechanism. We also implicate distinctive regulatory elements that promote the splicing of exon 1A to the distal 3 ss and exon 1B to the proximal 3 ss in murine erythroleukemia cells. A hybrid minigene driven by cytomegalovirus promoter mimicked 1B-promoter-driven splicing patterns but differed from 1A-promoter-driven splicing patterns, suggesting that promoter identity affects exon 2/2 splicing. Furthermore, splicing factor SF2/ASF ultraviolet (UV) cross-linked to the exon 2/2 junction CAGAGAA, a sequence that overlaps the distal U2AF 35 -binding 3 ss. Consequently, depletion of SF2/ASF allowed exon 1B to splice to the distal 3 ss but had no effect on exon 1A splicing. These findings identify for the first time that an SF2/ASF binding site also can serve as a 3 ss in a transcript-dependent manner. Taken

Alternative splicing events altering the 5′ end of the red blood cell protein 4.1 (4.1R) mRNA are critical for modulating the expression of both the 135 kD and 80 kD isoforms. Three mutually exclusive 5′ exons, 1A, 1B, and 1C, are transcribed from their respective promoter and exhibit differential splicing to exon 2′/2 splice acceptor sites. Exon 1A splices to the distal 3′ splice site (3′ ss) excluding exon 2′ and encoding for the 80 kD isoform. Exons 1B and 1C splice to the proximal 3′ ss, including exon 2′ in mature mRNA and encoding for the 135 kD form. We investigated the regulatory mechanism involved in the selection of the alternative 3′ ss, using 1A, 1B, and 1C minigene constructs containing intronic sequences downstream of each exon joined with intronic sequences upstream of exon 2′. We positioned either CMV or the respective native promoter upstream of the minigene and analyzed the spliced products for exon 2′ expression in mouse (MELC or C2C12) or human (HeLa or RD) cells. When under the control of the CMV promoter, all 1A, 1B, and 1C minigenes resulted in the inclusion of exon 2′. However, when under the control of its respective promoter, the minigene mimicked its endogenous splicing pattern, suggesting that promoter identity influences the alternative exon 2′ splicing decision. To confirm this hypothesis, we switched 1A and 1B promoters in their respective minigene constructs. Replacement of the 1A promoter by the 1B promoter resulted in the inclusion of exon 2′ in the 1A minigene. Conversely, replacement of the 1B promoter by the 1A promoter caused increased exclusion of exon 2′ in the 1B minigene, confirming that alternative splicing of exon 2′ is sensitive to the type of promoter. The current model on the modulation of alternative splicing by promoters suggests that the promoter might control alternative splicing via the regulation of polymerase II (pol II) elongation or processivity. To test whether the same mechanism applies to exon 2′/2 splicing, we treated 1A and 1B minigene-transfected cells with transcription elongation inhibitor 5,6-dichloro-1-b-D-ribofuranosylbenzimidazole (DRB). Inhibition of transcription elongation did not affect splicing of the 1B minigene, but it enhanced exon 2′ inclusion in the 1A minigene. The distal 3′ ss is a stronger splice site than the proximal 3′ ss. A highly processive elongating pol II favors the simultaneous presentation of both sites to the splicing machinery, a situation in which the distal 3′ ss out-competes the proximal 3′ ss, resulting in exon 2′ exclusion. Conversely, a slow pol II processivity on the 1B promoter favors the selection of the proximal 3′ ss and inclusion of exon 2′. Taken together, our results show that the alternative splicing of exon 2′/2 is tightly coupled to promoter architecture. Inclusion or exclusion of exon 2′ is achieved through a coordinated action of transcription and splicing.