Abstract. The contractile ring in dividing animal cells is formed primarily through the reorganization of existing actin filaments (Cao, L.-G., and Y.-L. Wang. 1990. J. Cell Biol. 110:1089-1096), but it is not clear whether the process involves a random recruitment of diffusible actin filaments from the cytoplasm, or a directional movement of cortically associated filaments toward the equator. We have studied this question by observing the distribution of actin filaments that have been labeled with fluorescent phaUoidin and microinjected into dividing normal rat kidney (NRK) cells. The labeled filaments are present primarily in the cytoplasm during prometaphase and early metaphase, but become associated extensively with the cell cortex 10-15 min before the onset of anaphase. This process is manifested both as an increase in cortical fluorescence intensity and as movements of discrete aggregates of actin filaments toward the cortex. The concentration of actin fluorescence in the equatorial region, accompanied by a decrease of fluorescence in polar regions, is detected 2-3 min after the onset of anaphase. By directly tracing the distribution of aggregates of labeled actin filaments, we are able to detect, during anaphase and telophase, movements of cortical actin filaments toward the equator at an average rate of 1.0 #m/min. Our results, combined with previous observations, suggest that the organization of actin filaments during cytokinesis probably involves an association of cytoplasmic filaments with the cortex, a movement of cortical filaments toward the cleavage furrow, and a dissociation of filaments from the equatorial cortex.

Abstract. The dynamic intra-nuclear localization of MRP RNA, the RNA component of the ribonucleoprotein enzyme RNase MRP, was examined in living cells by the method of fluorescent RNA cytochemistry (Wang, J., L.-G. Cao, Y.-L. Wang, and T. Pederson. 1991. Proc. Natl. Acad. Sci. USA. 88:7391-7395). MRP RNA very rapidly accumulated in nucleoli after nuclear microinjection of normal rat kidney (NRK) epithelial cells. Localization was specifically in the dense fibrillar component of the nucleolus, as revealed by immunocytochemistry with a monoclonal antibody against fibrillarin, a known dense fibrillar component protein, as well as by digital optical sectioning microscopy and 3-D stereo reconstruction. When MRP RNA was injected into the cytoplasm it was not imported into the nucleus. Nuclear microinjection of mutant MRP RNAs revealed that nucleolar localization requires a sequence element (nucleotides 23-62) previously implicated as a binding site for a nucleolar protein, the To antigen. These resuits demonstrate the dynamic localization of MRP RNA in the nucleus and provide important insights into the nucleolar targeting of MRP RNA. cessing) was initially identified as an endoribonu-,Ik ~ clease that produces, in vitro, RNA primers for the initiation of mitochondrial DNA replication (Chang and Clayton, 1987a). When RNase MRP was subsequently discovered to be a ribonucleoprotein enzyme and its RNA (MRP RNA; see Fig. 1) component was sequenced (Chang and Clayton, 1987b), it turned out to be identical to a previously defined nucleolar RNA, termed 7-2 RNA (Reddy et al., 1981;Gold et al., 1989;Yuan et al., 1989). MRP RNA (7-2 RNA) is essential for RNase MRP enzymatic activity in vitro and is encoded by a nuclear gene (Chang and Clayton, 1987b;Yuan et al., 1989; Tooper and Clayton, 1990) that is transcribed by RNA polymerase III (Chang and Clayton, 1989). MRP RNA has been detected by in situ hybridization in both mitochondria and nucleoli of mouse cardiac myocytes (Li et al., 1994). The great majority of MRP RNA (7-2 RNA) in the cell fractionates with nucleoli (Reddy et al., 1981 Chu et al., 1994;Lygerou et al., 1994). In HeLa cells, nuclear MRP RNA is associated with large nucleolar structures sedimenting at N80 S (Kiss et al., 1992). MRP RNA-containing RNP particles can be immunoprecipitated by human To (Reddy et al., 1983), Th (Hashimoto and Steitz, 1983), and Wa (Reimer et al., 1988) sera (To, Th, and Wa representing patient codes for autoimmune sera that contain antibodies directed against the MRP RNA-containing RNP particle). The protein recognized by anti-To and anti-Wa antibodies is a ~40,000-mol wt nucleolar protein (Reimer et al., 1988). The 40,000-mol wt To autoantigen-binding domain was subsequently localized to nucleotides 21-64 of human MRP RNA (Yuan et al., 1991).In the present investigation, we have examined the subcellular localization of fluorescently tagged MRP RNA after microinjection into the nucleus of living cells. This method, which we term fluorescent RNA cytochemistry, is integrated with high re...

We have studied the nuclear localization of rhodamine-labeled pre-mRNA after microinjection into nuclei of cultured rat kidney epithelial cells. Intranuclear localization of the injected RNA was followed in the living cells by fluorescence microscopy and digital image processing. Injected human P-globin pre-mRNA became localized in 30-60 discrete nuclear sites that were coincident with li defined by monoclonal antibodies against small nuclear ribonudeoproteins (Sm) or another spliceosome component RNA and premRNA (5,6). snRNPs have also been shown to be colocalized with pre-mRNA transcripts on insect polytene chromosomes (7,8). However, the nuclear sites at which splicing actually takes place have not been defined. In the present investigation, our objective was to determine whether exogenous pre-mRNA has an affinity for a particular element of nuclear structure. We have developed a system in which rhodamine-labeled pre-mRNA is introduced into the nucleus of normal rat kidney (NRK) epithelial cells, and its intranuclear distribution is followed by fluorescence microscopy of the living cells.MATERIALS AND METHODS Rhodamine Labeling of RNA. A human f-globin premRNA containing exon 1, intron 1, and 209 nucleotides of exon 2 (9) was transcribed from BamHI-cut plasmid SP64-H38A6 with SP6 RNA polymerase in the presence of 500 ,uM ATP, CTP, GTP, and UTP and 50 AM 5-(3-aminoallyl)-UTP (Sigma) (10). The RNA was recovered by ethanol precipitation after phenol and chloroform extraction and was coupled to tetramethylrhodamine-6-isothiocyanate (Molecular Probes) as described (11). The coupling reactions contained 25 mM sodium bicarbonate buffer (pH 9.0) and 16% (vol/vol) dimethylformamide. The rhodamine-RNA coupling reaction was carried out for 8-12 hr in the dark at 20'C in a tube mounted on a rocking platform. After the reaction, the RNA was purified by gel filtration through either Sephadex G-25 or Bio-Gel P-60 (with comparable results). RNA from the peak column fractions was concentrated by ethanol precipitation and then further purified by electrophoresis in an 8% polyacrylamide gel containing 7 M urea.To evaluate the efficiency of rhodamine conjugation, a preparation of H,8A RNA was transcribed in a sufficient amount for an A260 reading. After rhodamine conjugation, the rhodamine-RNA was purified by Bio-Gel P-60 gel filtration, ethanol precipitation, and electrophoresis. After elution from the gel and ethanol precipitation, the RNA concentration was measured as A2w. and the amount of coupled dye was measured by spectrofluorimetry at an excitation wavelength of 536 nm. The average value for three independent RNA preparations was 7.9 molecules ofrhodamine per molecule of RNA. The rhodamine coupling efficiency was calculated by considering that the 497-nucleotide HfA RNA contains 128 uridines (12) and by assuming that 1/11th of these uridines, or 11.6, carry aminoallyl groups (because transcription was done with 500 ,uM UTP and 50 1LM aminoallyl-UTP). Thus, the rhodamine coupling efficiency can be estimated t...

Abstract. Cytokinesis of animal cells involves the formation of the circumferential actin filament bundle (contractile ring) along the equatorial plane. To analyze the assembly mechanism of the contractile ring, we microinjected a small amount of rhodamine-labeled phalloidin (rh-pha) or rhodamine-labeled actin (rhactin) into dividing normal rat kidney cells, rh-pha was microinjected during prometaphase or metaphase to label actin filaments that were present at that stage. As mitosis proceeded into anaphase, the labeled iliamerits became associated with the cortex of the cell. During cytokinesis, rh-pha was depleted from polar regions and became highly concentrated into the equatorial region. The distribution of total actin illaments, as revealed by staining the whole cell with fluorescein phaUoidin, showed a much less pronounced difference between the polar and the equatorial regions. The sites of de novo assembly of actin filaments during the formation of the contractile ring were determined by microinjecting rh-actin shortly before cytokinesis, and then extracting and fixing the cell during mid-cytokinesis. Injected rhodamine actin was only slightly concentrated in the contractile ring, as compared to the distribution of total actin filaments. Our results indicate that preexisting actin filaments, probably through movement and reorganization, are used preferentially for the formation of the contractile ring. De novo assembly of filaments, on the other hand, appears to take place preferentially outside the cleavage furrow.C YTOKINESIS of animal cells has attracted the attention of cell biologists for more than a century (for recent reviews see Rappaport, 1986;Mabuchi, 1986). However, its mechanism still remains unclear. Studies over the past twenty years have indicated that actin and myosin II are likely to be involved in this process. For example, actin filaments show a parallel, bundle-like arrangement in the cleavage furrow, forming a "contractile ring" (Perry et al., 1971;Forer and Behnke, 1972;Schroeder, 1973). Myosin II molecules exist as minifilaments in this region, and appear more concentrated than in other regions of the cell (Fujiwara and Pollard, 1976;Yumura and Fukui, 1985;Maupin and Pollard, 1986). Various observations indicate that the integrity of actin and myosin structures is crucial for cytokinesis. Cytochalasin B, which binds to the ends of actin filaments and may also induce filament breakage, inhibits cytokinesis with a concomitant disappearance of the contractile ring (Schroeder, 1970(Schroeder, , 1972. The involvement of myosin II has been convincingly demonstrated with the microinjection of antibodies against myosin II (Mabuchi and Okuno, 1977;Kiehart et al., 1982), and with the disruption of myosin II expression after genetic manipulations (De Lozanne and Spudich, 1987;Knecht and Loomis, 1987). These treatments effectively inhibited cytokinesis while allowing karyokinesis to proceed. However, many critical questions concerning the roles of actin and myosin in cytokinesis need to be cla...

Abstract. Myosin light chain kinase (MLCK) is thought to regulate the contractile activity in smooth and nonmuscle cells, and may play an important role in controlling the reorganization of the actin-myosin cytoskeleton during cell division. To test this hypothesis we have microinjected the 61-kD catalytic fragment of MLCK into mitotic cells, and examined the effects of unregulated MLCK activity on cell division. The microinjection of active 61 kD causes both a significant delay in the transit time from nuclear envelope breakdown to anaphase onset, and an increase in motile surface activity during and after metaphase. Control experiments with intact MLCK or with inactive catalytic fragment suggest that these effects are specifically in-ACTIN and myosin undergo precisely regulated reorganlzations in cultured animal cells during mitosis,