A novel role for non-muscle gamma-actin in skeletal muscle sarcomere assembly.

Existing models describing sarcomere assembly have arisen primarily from studies using cardiac muscle. In contrast to cardiac muscle, skeletal muscle differentiation is characterised by dramatic changes in protein expression, from non-muscle to muscle-specific isoforms before organisation of the sarcomeres. Consequently, little is understood of the potential influence of non-muscle cytoskeletal proteins on skeletal sarcomere assembly. To address this issue, transfectant (gamma33-B1) and control mouse C2 myoblasts were differentiated to form myotubes, and various stages of skeletal sarcomere assembly were studied. Organisation of non-muscle gamma-actin and co-localisation with sarcomeric alpha-actinin, an early marker of sarcomere assembly and a major component of Z lines, was noted. gamma-Actin was also identified in young myotubes with developing sarcomeric myofibrils in regenerating adult mouse muscle. Localisation of gamma-actin in a different area of the myotube to the muscle-specific sarcomeric alpha-actin also indicated a distinct role for gamma-actin. The effects of aberrant gamma-actin expression in other myoblast lines, further suggested a sequestering role for gamma-actin. These observations make the novel suggestion that non-muscle gamma-actin plays a role in skeletal sarcomere assembly both in vitro and in vivo. Consequently, a modified model is proposed which describes the role of gamma-actin in skeletal sarcomere assembly.

Gamma tropomyosin gene products are required for embryonic development.

The actin filament system is essential for many cellular functions, including shape, motility, cytokinesis, intracellular trafficking, and tissue organization. Tropomyosins (Tms) are rod-like components of most actin filaments that differentially affect their stability and flexibility. The Tm gene family consists of four genes, alphaTm, betaTm, gammaTm (Tm5 NM, where "NM" indicates "nonmuscle"), and deltaTm (Tm4). Multiple isoforms of the Tm family are generated by alternative splicing of three of these genes, and their expression is highly regulated. Extensive spatial and temporal sorting of Tm isoforms into different cellular compartments has been shown to occur in several cell types. We have addressed the function of the low-molecular-weight Tms encoded by the gammaTm gene by eliminating the corresponding amino-terminal coding sequences from this gene. Heterozygous mice were generated, and subsequent intercrossing of the F1 pups did not result in any viable homozygous knockouts. Genotype analysis of day 2.5 morulae also failed to detect any homozygous knockouts. We have failed in our attempts to delete the second allele and generate in vitro double-knockout cells, although 51 clones displayed homologous recombination back into the originally targeted locus. We therefore conclude that low-molecular-weight products from the gammaTm gene are essential for both embryonic development and cell survival.

Distinct localizations of tropomyosin isoforms in LLC-PK1 epithelial cells suggests specialized function at cell-cell adhesions.

At least eight nonmuscle, nonbrain tropomyosin isoforms have been described. We used antibodies, microinjection, and transfection to characterize their expression and localization in LLC-PK1 kidney epithelial cells and compared them with other cells. Similar to primary enterocytes, LLC-PK1 cells exhibited predominantly TM-1 and TM-3 of the high-molecular-weight (HMW) isoforms; TM-5 and TM-5b of the low-molecular-weight (LMW) isoforms. Neither TM-4 nor TM-5a was detectable in the LLC-PKI cells. Immunofluorescence studies revealed that HMW isoforms were localized only on stress fibers, not adhesion belts, whereas the adhesion belts were stained by LMW isoform antibodies. When exogenous proteins are introduced either by transfection or microinjection, the HMW isoforms do not incorporate into the adhesion belt, whereas the LMW isoforms can incorporate into the stress fibers, thus indicating there are different mechanisms at work for the selective localization. Temporal changes in the microfilament system of the LLC-PK1 cells were studied during differentiation in culture as defined by spectrin expression and F-actin architecture. Western blot analysis indicated that TM-5b is only expressed in the LLC-PK1 cells after a certain degree of maturation in culture, which suggests isoform switching after the cell-cell contacts are developed. Collectively these results demonstrate that epithelial cells express a complex pattern of TM isoforms, which exhibit differential localizations within the cells and different patterns of expression depending on their origin and stage of differentiation. The implication of differential localization of TM isoforms on their specific functions is discussed.

Splicing of two internal and four carboxyl-terminal alternative exons in nonmuscle tropomyosin 5 pre-mRNA is independently regulated during development.

Four nonmuscle tropomyosin isoforms have been reported to be produced from the rat Tm5 gene by alternative splicing (Beisel, K. W., and Kennedy, J. E. (1994) Gene (Amst.) 145, 251-256). In order to detect additional isoforms that might be expressed from that gene, we used reverse transcriptase-polymerase chain reaction assays and evaluated the presence of all product combinations of two alternative internal exons (6a and 6b) and four carboxyl-terminal exons (9a, 9b, 9c, and 9d) in developing and adult rat brain. We identified five different combinations for exon 9 (9a + 9b, 9a + 9c, 9a + 9d, 9c, and 9d), and the exon combinations 9a + 9c and 9a + 9d were previously unreported. Each of these combinations existed with both exon 6a and exon 6b. Thus, the rat brain generates at least 10 different isoforms from the Tm5 gene. Northern blot hybridization with alternative exon-specific probes revealed that these isoforms were also expressed in a number of different adult rat tissues, although some exons are preferentially expressed in particular tissues. Studies of regulation of the 10 different Tm5 isoform mRNAs during rat brain development indicated that no two isoforms are coordinately accumulated. Furthermore, there is a developmental switch in the use of exon 6a to exon 6b from embryonic to adult isoforms. TM5 protein isoforms show a differential localization in the adult cerebellum.

Differential regulation of actin depolymerizing factor and cofilin in response to alterations in the actin monomer pool.

Myoblasts, transfected with a human gene encoding a beta-actin point mutation, down-regulate expression of actin depolymerizing factor (ADF) and its mRNA. Regulation is posttranscriptional. Expression of cofilin, a structurally similar protein, and profilin, CapG, and tropomodulin is not altered with increasing mutant beta-actin expression. Myoblasts expressing either human gamma-actin or the mutant beta-actin down-regulate the endogenous mouse actin genes to keep a constant level of actin mRNA, whereas the gamma-actin transfectants do not down-regulate ADF. Thus, ADF expression is regulated differently from actin expression. The mutant beta-actin binds to ADF with about the same affinity as normal actin; however, it does not assemble into normal actin filaments. The decrease in ADF expression correlates with an increase in the unassembled actin pool. When the actin monomer pool in untransfected myoblasts is increased 70% by treatment with latrunculin A, synthesis of ADF and actin are down-regulated compared with cofilin and 19 other proteins selected at random. Increasing the actin monomer pool also results in nearly complete phosphorylation of both ADF and cofilin. Thus, ADF and cofilin are coordinately regulated by posttranslational modification, but their expression is differentially regulated. Furthermore, expression of ADF is responsive to the utilization of actin by the cell.

Tropomyosin localization reveals distinct populations of microfilaments in neurites and growth cones.

The functional and structural differences between neurites and growth cones suggests the possibility that distinct microfilament populations may exist in each domain. Tropomyosins are integral components of the actin-based microfilament system. Using antibodies which detect three different sets of tropomyosin isoforms, we found that the vast majority of tropomyosin was found in a microfilament-enriched fraction of cultured cortical neurons, therefore enabling us to use the antisera to evaluate compositional differences in neuritic and growth cone microfilaments. An antibody which reacts with all known nonmuscle isoforms of the alpha Tms gene (Tm5NM1-4) stains both neurites and growth cones, whereas a second antibody against the isoform subset, Tm5NM1-2, reacts only with the neurite. A third antibody which reacts with the Tm5a/5b isoforms encoded by a separate gene from alpha Tms was strongly reactive with both neurites and growth cones in 16-h cultures but only with the neurite shaft in 40-h cultures. Treatment of neurons with cytochalasin B allowed neuritic Tm5NM1-2 to spread into growth cones. Removal of the drug resulted in the disappearance of Tm5NM1-2 from the growth cone, indicating that isoform segregation is an active process dependent on intact microfilaments. Treatment of 40-h cultures with nocodazole resulted in the removal of Tm5NM1-2 from the neurite whereas Tm5a/5b now spread back into the growth cone. We conclude that the organization of Tm5NM1-2 and Tm5a/5b in the neurite is at least partially dependent on microtubule integrity. These results indicate that tropomyosin isoforms Tm5NM1-2, Tm5NM3-4, and Tm5a/5b mark three distinct populations of actin filaments in neurites and growth cones. Further, the composition of microfilaments differs between neurites and growth cones and is subject to temporal regulation.

The molecular composition of neuronal microfilaments is spatially and temporally regulated.

The actin-based microfilament system is thought to play a critical role in neuronal development. We have determined specific changes in the composition of microfilaments accompanying neuronal morphogenesis. By using specific antibodies against the isoforms for tropomyosin (Tm) (Tm-5 and TmBr-1/-3) and actin (beta- and gamma-actin), we found that during early morphogenesis in vivo immature growing axons contain beta- and gamma-actin and Tm-5. In particular, Tm-5 is exclusively located in the immature axonal processes relative to the neuronal cell body. In contrast, beta-actin and Tm-5 are absent in mature, quiescent axons. This developmental loss from axons is associated with an approximately twofold downregulation of beta-actin and Tm-5 levels in the brain; gamma-actin levels do not change, and this molecule is widely distributed throughout neurons during development. The loss of beta-actin and Tm-5 from axons is accompanied by a progressive appearance of TmBr-1/-3. This apparent replacement of Tm-5 with TmBr-1/-3 occurs over a 2 d time period during rat embryonic hindbrain development and is conserved in evolution between birds and mammals. The loss of Tm-5 from axons involves a redistribution of this molecule to the cell soma and dendrites. These findings suggest that specialized microfilament domains are associated with the development and maintenance of neuronal polarity. We conclude that these Tm isoforms and beta-actin are subject to specific patterns of segregation associated with axonal development and neuronal differentiation. This provides a potential molecular basis for the temporal and spatial specificity of microfilament function during neuronal differentiation.

Intracellular localization of tropomyosin mRNA and protein is associated with development of neuronal polarity.

Neuronal differentiation involves extensive rearrangement of the cytoskeleton, including the actin-based microfilament system, and establishment of molecular compartments within the neuron. The intracellular distribution of tropomyosin (Tm) mRNA in vivo and in vitro has been examined and correlated with protein targetting. The mRNAs encoding two Tm isoforms were found to be differentially localized in developing neurons. Tm-5 mRNA is localized to the axonal pole of differentiating embryonic rat neurons, in contrast to TmBr-2 mRNA distribution throughout the cell body. Tm-5 mRNA is transported into the axon of differentiating primary cultured neurons. This mRNA localization is developmentally regulated and correlates with the targeting of Tm-5 protein to growing axons. Tm-5 colocalizes with a subset of neuronal microfilaments associated with the initiation and maintenance of outgrowth. The segregation of Tm-5 is the earliest known marker of neuronal polarity and may play a role in the establishment of polarity.

Impact of altered actin gene expression on vinculin, talin, cell spreading, and motility.

Previous studies have demonstrated a strong correlation between the expression of vinculin and the shape and motility of a cell (Rodriguez Fernandez et al., 1992a, b, 1993). This hypothesis was tested by comparing the expression of vinculin and talin with the motility of morphologically altered myoblasts. These mouse C2 myoblasts were previously generated by directly perturbing the cell cytoskeleton via the stable transfection of a mutant-form of the beta-actin gene (beta sm) and three different forms of the gamma-actin gene; gamma, gamma minus 3'UTR (gamma delta'UTR), and gamma minus intron III (gamma delta IVSIII) (Schevzov et al., 1992; Lloyd and Gunning, 1993). In the case of the beta sm and gamma-actin transfectants, a two-fold decrease in the cell surface area was coupled, as predicted, with a decrease in vinculin and talin expression. In contrast, the gamma delta IVSIII transfectants with a seven-fold decrease in the cell surface area showed an unpredicted slight increase in vinculin and talin expression and the gamma delta 3'-UTR transfectants with a slight increase in the cell surface area showed no changes in talin expression and a decrease in vinculin expression. We conclude that changes in actin gene expression alone can impact on the expression of vinculin and talin. Furthermore, we observed that these actin transfectants failed to show a consistent relationship between cell shape, motility, and the expression of vinculin. However, a relationship between talin and cell motility was found to exist, suggesting a role for talin in the establishment of focal contacts necessary for motility.

Induction of anchorage independent growth and serum resistance in immortalized human bronchial epithelial cells by alteration of the cytoskeleton.

Malignant transformation is frequently accompanied by obvious changes in cytoarchitecture, but the importance of these changes has been difficult to assess in view of the large number of other cellular changes that also occur. In this study, we transfected the SV40-immortalized human bronchial epithelial cell line, BEAS-2B, with human wild-type beta or gamma actin gene expression plasmids to induce cytoskeletal changes and to determine whether this was associated with altered cellular growth properties. Cells expressing the exogenous full-length actin genes underwent a fibroblastoid change in morphology which was reflected in changes in their pattern of actin cable organization, and acquired both the ability to grow under anchorage-independent conditions and resistance to the normal growth inhibitory effects of fetal bovine serum. These phenotypic changes correlated with changes in actin mRNA levels, but not with changes in actin protein levels. The phenotypically altered cells were not tumorigenic when injected subcutaneously in athymic nude mice, and they retained the ability to suppress the tumorigenic potential of a lung carcinoma cell line, HuT-292. Therefore, alteration of the cytoskeleton of immortalized human bronchial epithelial cells resulted in the acquisition of some properties commonly found in malignant cells, but did not result in tumorigenicity.

Differential regulation of tropomyosin isoform organization and gene expression in response to altered actin gene expression.

Phenotypically altered C2 myoblast cells, generated by the stable transfection of human nonmuscle actin genes (Schevzov, G., C. Lloyd, and P. Gunning. 1992. J. Cell Biol. 117:775-786), exhibit a differential pattern of tropomyosin cellular organization and isoform gene expression. The beta-actin transfectants displaying a threefold increase in the cell surface area, showed no significant changes in the pattern of organization of the high M(r) tropomyosin isoform, Tm 2, or the low M(r) tropomyosin isoform, Tm 5. In contrast, the gamma- and beta sm-actin gene transfectants, exhibiting a twofold decrease in the cell surface area, had an altered organization of Tm 2 but not Tm 5. In these actin transfectants, Tm 2 did not preferentially segregate into stress fiber-like structures and the intensity of staining was greatly diminished. Conversely, a well-defined stress fiber-like organization of Tm 5 was observed. The pattern of organization of these tropomyosin isoforms correlated with their expression such that a profound decrease in Tm 2 expression was observed both at the transcript and protein levels, whereas Tm 5 remained relatively unchanged. These results suggest that relative changes in nonmuscle actin gene expression can affect the organization and expression of tropomyosin in an isoform specific manner. Furthermore, this apparent direct link observed between actin and tropomyosin expression suggests that nonpharmacological signals originating in the cytoskeleton can regulate cytoarchitectural gene expression.

High level expression of transfected beta- and gamma-actin genes differentially impacts on myoblast cytoarchitecture.

The impact of the human beta- and gamma-actin genes on myoblast cytoarchitecture was examined by their stable transfection into mouse C2 myoblasts. Transfectant C2 clones expressing high levels of human beta-actin displayed increases in cell surface area. In contrast, C2 clones with high levels of human gamma-actin expression showed decreases in cell surface area. The changes in cell morphology were accompanied by changes in actin stress-fiber organization. The beta-actin transfectants displayed well-defined filamentous organization of actin; whereas the gamma-actin transfectants displayed a more diffuse organization of the actin cables. The role of the beta-actin protein in generating the enlarged cell phenotype was examined by transfecting a mutant form of the human beta-actin gene. Transfectant cells were shown to incorporate the aberrant actin protein into stress-fiber-like structures. High level expression of the mutant beta-actin produced decreases in cell surface area and disruption of the actin microfilament network similar to that seen with transfection of the gamma-actin gene. In contrast, transfection of another mutant form of the beta-actin gene which encodes an unstable protein had no impact on cell morphology or cytoarchitecture. These results strongly suggest that it is the nature of the encoded protein that determines the morphological response of the cell. We conclude that the relative gene expression of beta- and gamma-actin is of relevance to the control of myoblast cytoarchitecture. In particular, we conclude that the beta- and gamma-actin genes encode functionally distinct cytoarchitectural information.

Transfection of nonmuscle beta- and gamma-actin genes into myoblasts elicits different feedback regulatory responses from endogenous actin genes.

We have examined the role of feedback-regulation in the expression of the nonmuscle actin genes. C2 mouse myoblasts were transfected with the human beta- and gamma-actin genes. In gamma-actin transfectants we found that the total actin mRNA and protein pools remained unchanged. Increasing levels of human gamma-actin expression resulted in a progressive down-regulation of mouse beta- and gamma-actin mRNAs. Transfection of the beta-actin gene resulted in an increase in the total actin mRNA and protein pools and induced an increase in the levels of mouse beta-actin mRNA. In contrast, transfection of a beta-actin gene carrying a single-point mutation (beta sm) produced a feedback-regulatory response similar to that of the gamma-actin gene. Expression of a beta-actin gene encoding an unstable actin protein had no impact on the endogenous mouse actin genes. This suggests that the nature of the encoded actin protein determines the feedback-regulatory response of the mouse genes. The role of the actin cytoskeleton in mediating this feedback-regulation was evaluated by disruption of the actin network with Cytochalasin D. We found that treatment with Cytochalasin D abolished the down-regulation of mouse gamma-actin in both the gamma- and beta sm-actin transfectants. In contrast, a similar level of increase was observed for the mouse beta-actin mRNA in both control and transfected cells. These experiments suggest that the down-regulation of mouse gamma-actin mRNA is dependent on the organization of the actin cytoskeleton. In addition, the mechanism responsible for the down-regulation of beta-actin may be distinct from that governing gamma-actin. We conclude that actin feedback-regulation provides a biochemical assay for differences between the two nonmuscle actin genes.

Cloning and expression of a cDNA coding for a human monocyte-derived plasminogen activator inhibitor.

Human monocyte-derived plasminogen activator inhibitor (mPAI-2) was purified to homogeneity from the U937 cell line and partially sequenced. Oligonucleotide probes derived from this sequence were used to screen a cDNA library prepared from U937 cells. One positive clone was sequenced and contained most of the coding sequence as well as a long incomplete 3' untranslated region (1112 base pairs). This cDNA sequence was shown to encode mPAI-2 by hybrid-select translation. A cDNA clone encoding the remainder of the mPAI-2 mRNA was obtained by primer extension of U937 poly(A)+ RNA using a probe complementary to the mPAI-2 coding region. The coding sequence for mPAI-2 was placed under the control of the lambda PL promoter, and the protein expressed in Escherichia coli formed a complex with urokinase that could be detected immunologically. By nucleotide sequence analysis, mPAI-2 cDNA encodes a protein containing 415 amino acids with a predicted unglycosylated Mr of 46,543. The predicted amino acid sequence of mPAI-2 is very similar to placental PAI-2 (3 amino acid differences) and shows extensive homology with members of the serine protease inhibitor (serpin) superfamily. mPAI-2 was found to be more homologous to ovalbumin (37%) than the endothelial plasminogen activator inhibitor, PAI-1 (26%). Like ovalbumin, mPAI-2 appears to have no typical amino-terminal signal sequence. The 3' untranslated region of the mPAI-2 cDNA contains a putative regulatory sequence that has been associated with the inflammatory mediators.