The actin-associating protein Tm5NM1 blocks mesenchymal motility without transition to amoeboid motility.

Cell migration is an integral component of metastatic disease. The ability of cells to transit between mesenchymal and amoeboid modes of migration has complicated the development of successful therapies designed to target cell migration as a means of inhibiting metastasis. Therefore, investigations of the mechanisms that regulate cell migration and render cells stationary are necessary. Tropomyosins are actin-associating proteins that regulate the activity of several effectors of actin filament dynamics. Previously, we have shown that the tropomyosin isoform Tm5NM1 stabilizes actin filaments and inhibits cell migration in a two-dimensional culture system. Here, we show that Tm5NM1 inhibits the mesenchymal migration of multiple cell lines in an isoform-specific manner. Tm5NM1 stimulates the downregulation of Src kinase activity and a rounded or elliptical morphology in three-dimensional collagen gels, and cells have dramatically reduced capacity to form pseudopodia. Importantly, we find that Tm5NM1 inhibits both the mesenchymal to amoeboid and amoeboid to mesenchymal transitions. Collectively, our data suggest that mimicking the action of Tm5NM1 overexpression represents an approach for effectively inhibiting the mesenchymal mode of migration.

Tropomyosins as interpreters of the signalling environment to regulate the local cytoskeleton.

A key regulator of cell morphology is the actin cytoskeleton and it has long been appreciated that the cytoskeleton is characteristically altered in cancer. Actin is organized into polymeric structures with distinct dynamics which in turn participate in a wide variety of cell processes including adhesion, migration, cell division and apoptosis. Despite displaying an altered actin cytoskeleton, transformed cells retain--and in many cases increase--their ability to adhere, move, divide and respond to apoptotic stimuli. Thus cancer cells maintain responsive actin cytoskeletons. Actin dynamics are regulated by numerous actin-binding proteins and chief among these are the tropomyosins which are core components of the microfilament. Recent advances in genomic and proteomic profiling confirm that Tm expression profiles are profoundly changed in transformed cells. It is therefore timely to review the role of Tms in the regulation of actin dynamics that pertain to crucial phenotypic changes in cancer. In this review we discuss how actin filaments containing different Tm isoforms respond to the activation of cell signalling pathways and consider the implications of this for cancer progression and therapy.

Functional analysis of the actin-binding protein, tropomyosin 1, in neuroblastoma.

Tropomyosin 1 (TM1) is downregulated in a number of transformed cell types, and exogenous expression of TM1 can restore actin organisation and reverse cellular transformation. We find that TM1 is also downregulated in human neuroblastoma cell lines, correlating with increasing malignancy. However, exogenous TM1 does not restore actin cytoskeleton organisation in neuroblastoma cells.

Alpha-skeletal actin induces a subset of muscle genes independently of muscle differentiation and withdrawal from the cell cycle.

Muscle differentiation is characterized by the induction of genes encoding contractile structural proteins and the repression of nonmuscle isoforms from these gene families. We have examined the importance of this regulated order of gene expression by expressing the two sarcomeric muscle actins characteristic of the differentiated state, i.e. alpha-skeletal and alpha-cardiac actin, in C2 mouse myoblasts. Precocious accumulation of transcripts and proteins for a group of differentiation-specific genes was elicited by alpha-skeletal actin only: four muscle tropomyosins, two muscle actins, desmin and MyoD. The nonmuscle isoforms of tropomyosin and actin characteristic of the undifferentiated state continued to be expressed, and no myosin heavy or light chain or troponin transcripts characteristic of muscle differentiation were induced. Stable transfectants displayed a substantial reduction in cell surface area and in the levels of nonmuscle tropomyosins and beta-actin, consistent with a relationship between the composition of the actin cytoskeleton and cell surface area. The transfectants displayed normal cell cycle progression. We propose that alpha-skeletal actin can activate a regulatory pathway linking a subset of muscle genes that operates independently of normal differentiation and withdrawal from the cell cycle.

A mutation in alpha-tropomyosin(slow) affects muscle strength, maturation and hypertrophy in a mouse model for nemaline myopathy.

Nemaline myopathy is a hereditary disease of skeletal muscle defined by a distinct pathology of electron-dense accumulations within the sarcomeric units called rods, muscle weakness and, in most cases, a slow oxidative (type 1) fiber predominance. We generated a transgenic mouse model to study this disorder by expressing an autosomal dominant mutant of alpha-tropomyosin(slow) previously identified in a human cohort. Rods were found in all muscles, but to varying extents which did not correlate with the amount of mutant protein present. In addition, a pathological feature not commonly associated with this disorder, cytoplasmic bodies, was found in the mouse and subsequently identified in human samples. Muscle weakness is a major feature of this disease and was examined with respect to fiber composition, degree of rod-containing fibers, fiber mechanics and fiber diameter. Hypertrophy of fast, glycolytic (type 2B) fibers was apparent at 2 months of age. Muscle weakness was apparent in mice at 5-6 months of age, mimicking the late onset observed in humans with this mutation. The late onset did not correlate with observed changes in fiber type and rod pathology. Rather, the onset of muscle weakness correlates with an age-related decrease in fiber diameter and suggests that early onset is prevented by hypertrophy of fast, glycolytic fibers. We suggest that the clinical phenotype is precipitated by a failure of the hypertrophy to persist and therefore compensate for muscle weakness.

Different electrophoretic techniques produce conflicting data in the analysis of myocardial samples from dilated cardiomyopathy patients: protein levels do not necessarily reflect mRNA levels.

A variety of electrophoretic techniques were used to search for potential causes of human dilated cardiomyopathy (DCM). Northern blots were used to quantify alpha-cardiac and alpha-skeletal muscle actins, and beta-myosin heavy chain mRNAs which are the predominant expressed isoform species. We found a wide range of mRNA levels expressed in both DCM and nondiseased (ND) samples of left ventricles. However, sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) gels of the same heart samples revealed a stable and constant ratio of actin and myosin. Dystrophin deficiency might account for the DCM symptoms and so dystrophin levels of DCM and ND samples were evaluated using Western blots probed with monoclonal antibodies for the N-, C- and mid-rod portions of this protein. We found that dystrophin levels were constant in all 29 DCM and 5 ND samples suggesting that dystrophin deficiency is probably not a contributing cause. We explored the possibility that terminal failure may be due to an apoptotic-like event in the cardiomyocytes. Zymograms of DCM and ND samples revealed a significant increase in DNase I activity in the DCM group compared to the ND samples. These data raise the possibility that end-stage failure may be associated with apoptosis.

Characterization of human myosin light chains 1sa and 3nm: implications for isoform evolution and function.

We have isolated a cDNA clone for the human slow-twitch muscle isoform myosin light-chain 1slow-a (MLC1sa) from a skeletal muscle library and for the human nonmuscle isoform myosin light-chain 3nonmuscle (MLC3nm) from a fibroblast library. The nucleotide sequence of both isoforms was determined, and isoform-specific probes were constructed. In addition, MLC1sa was subsequently isolated from the fibroblast library. MLC1sa and MLC3nm were found to be very closely related to each other and distant from all other myosin light-chain isoforms so far described. We concluded that MLC1sa arose by duplication of MLC3nm rather than from any other isoform. A comparison was made between all human myosin light chains described to date and a model proposed for the evolution of this multigene family. A comparison between human and chicken myosin light-chain isoforms showed that human isoforms are more similar to their chicken counterparts than to human MLC1sa. The expression of MLC1sa and MLC3nm was studied in humans, rabbits, mice, and rats. MLC1sa was detected at the onset of both human and murine myogenesis in vitro. With development, MLC1sa may be replaced by the other slow-twitch muscle isoform, 1sb, in slow-twitch skeletal muscle, but the proportion of MLC1sa to 1sb expression varies between different species. MLC1sa was detected in nonmuscle cells in humans, mice, and rats. MLC3nm was the major nonmuscle alkaline myosin light chain in all species tested, but its pattern of expression in nonmuscle tissues was not identical to that of beta- or gamma-actin. We have shown that in the human, as in the chicken, one exon is spliced out of the MLC3nm transcript in smooth muscle to give an alternative product. We concluded that all alkali myosin light-chain isoforms may be functionally different.

Axoplasmic transport of transfer RNA in the chick optic system.

It has previously been shown that 4S RNA is transported in the optic nerve of the chick, but that no movement of rRNA can be detected. The 4S component behaved as though it were composed mainly of transfer RNA (tRNA), but the possibility remained that it could contain significant amounts of material resulting from RNA degradation. The transport of this 4S component has been examined in more detail to determine its nature. In addition, the transported material was examined to establish whether the transport of tRNA is a general phenomenon or that there are only a limited number of species involved. This was done using the same principles applied in the previous study; i.e., the specific activities of separated 4S RNA species appearing in the optic tectum 4 days after intraocular injection of [3H]uridine were compared with that of 5S RNA, a nontransported species. The separation was accomplished using 2.8-5-10-17% slab polyacrylamide gels, and 18 separate regions of 4S species could be identified. The results show that at least most, if not all 4S RNA species are transported. In a separate series of experiments the 4S RNA was aminoacylated and again separated on slab gels. In this instance, the RNA was labelled with [3H]uridine and the aminoacyl component with [14C]amino acids. Gel profiles of these dual-labelled components showed excellent correspondence between the two labels, demonstrating that 4S RNA species could be aminoacylated and were therefore tRNA species.(ABSTRACT TRUNCATED AT 250 WORDS)

The nerve growth factor induced differentiation of a pheochromocytoma PC12 cell line.

Cessation of cell division and DNA synthesis are not required for beta NGF-induced neurite outgrowth from PC12 cells. However, beta NGF-mediated increases in RNA and protein levels closely correlate with the size and number of neurites per cell suggesting that these changes influence some quantitative aspects of neurite outgrowth. RNA levels are increased due to an increase in transcriptional rates. It is likely that the increase in transcription is triggered by the beta NGF-mediated increase in the amounts of the nucleoside triphosphates which serve as substrates for transcription.

Nerve growth factor-induced stimulation of alpha-aminoisobutyric acid uptake in PC12 cells: relationship to plasma membrane receptor occupancy.

Nerve growth factor (NGF) stimulates the uptake rate of the nonmetabolized amino acid alpha-aminoisobutyric acid (AIB) in the clonal PC12 pheochromocytoma cell line by 40-70%. This effect reaches a maximum after a 1-hour incubation with the hormone and then drops over 50%, reaching a minimum after 4 hours of NGF administration. Longer exposure to the hormone leads to a gradual rise in stimulation, and by 24 hours, the cells regain about 80% of the original 1-hour rate. Results of NGF-binding studies indicate that stimulation of AIB uptake follows closely behind the amount of NGF bound to the low-affinity NGF receptors. Dose-response experiments indicate that full stimulation occurs biphasically. Within the NGF concentration range of 0.1 ng/ml to 1 ng/ml, the AIB uptake rate is 30% of the maximum 1-hour response. At 2.5 ng/ml NGF, the stimulation jumps to about 75% maximal response while 100% response is reached between 5 ng/ml and 50 ng/ml NGF. We suggest that the two states of the NGF plasma membrane receptor on PC12 cells may both be involved in mediating NGF stimulation of AIB uptake.

Nerve growth factor-induced stimulation of alpha-aminoisobutyric acid uptake in PC 12 cells: evaluation of its biological significance.

In the clonal PC12 pheochromocytoma cell line, the observed effects of nerve growth factor (NGF) on the uptake rates of alpha-aminoisobutyric acid (AIB) depend upon the assay conditions employed. In orthodox uptake assays calling for serum removal prior to the addition of AIB, 50 ng/ml NGF causes a stimulation of uptake of 62% after 40 minutes and 46% after 24-hours exposure. However, serum stimulates AIB uptake to a similar extent and the effects of serum and NGF are not additive. An uptake assay which directly measures the AIB uptake experienced by PC12 cells undergoing NGF-induced morphological differentiation was therefore employed. When compared with control cells growing in serum-containing medium, NGF-induced differentiating PC12 calls experience (1) only a transient and modest increase in AIB uptake, and (2) a significant long-term decrease in AIB uptake under conditions optimal for differentiation. It is concluded that NGF-directed neurite outgrowth is not mediated by NGF effects on amino acid uptake via the A-system.

The action of nerve growth factor and dibutyryl adenosine cyclic 3':5'-monophosphate on rat pheochromocytoma reveals distinct stages in the mechanisms underlying neurite outgrowth.

The clonal rat pheochromocytoma, PC12, responds to nerve growth factor (NGF) and dibutyryl adenosine cyclic 3':5'monophosphate (dbc AMP) by the elevation of cellular protein and RNA levels in all three parameters that are additive or greater than the sum of those caused by either agent alone, indicating that the mechanisms by which the two agents act to produce these changes are distinct. The concentration of dbcAMP required for half-maximal stimulation of these changes is also different for each, while NGF is active in all instances at 10(-11) M. PC12 cells initially generate neurites slowly in response to NGF and, at the same time, develop the capacity to regenerate neurites rapidly, a process termed priming. The cells, however, are not primed by dbcAMP nor does it influence the ability of NGF to prime them. Time lapse cinematography demonstrates that both NGF and dbcAMP each have unique effects on cellular morphology. The latter produces' rapid, unstable neurite initiation but does not promote sustained neurite extension. In contrast, NGF has immediate effects at the cell surface with no neurite initiation but later produces sustained neurite outgrowth. Utilizing dbcAMP, it is possible to dissect four morphological response of PC12 cells to NGF. Three of these may be mechanistically closely tied to occupancy of the NGF plasma membrane receptor.

Isolation and characterization of human actin genes.

We have utilized cloned actin genes from Drosophila melanogaster and from chicken to isolate 12 actin gene fragments from a human DNA library. Each of these 12 clones was shown to contain actin coding regions by its ability to selectively hybridize to human actin mRNA as assayed by in vitro translation. The translation product was judged to be actin on the basis of its comigration with authentic actins when electrophoresed on one- and two-dimensional NaDodSO4/polyacrylamide gels and on the basis of its partial proteolysis products. Determination of the sizes and order of the fragments generated by restriction endonuclease digestion of each of these recombinant phages allows us to conclude that they are nonallelic and are from nonoverlapping regions of the genome. We have used these cloned human actin genes and the Drosophila and chicken actin gene clones to show that the human genome contains 25-30 EcoRI fragments homologous to actin genes and that, among three nonconsanguineous individuals tested, none of these fragments exhibit length or restriction-site polymorphism.

Differential and coordinate regulation of the eukaryotic small molecular weight RNAs.

We have detected within the adult rat cerebral cortex 9 identifiable small molecular weight RNAs, K, K', L, Q, A, C, D, G', and H. These molecules are electrophoretically similar to the small RNAs found in other mammalian cells. Metabolic turnover and subcellular fractionation studies demonstrate that each of these small RNAs behave in a manner distinguishable from the other 8. In addition, 2 of the small RNAs, K and A, are under considerable developmental regulation. The results obtained suggest that no 2 of these molecules are present exclusively within the same subcellular structure. The cellular concentrations of the small RNAs were analyzed in 7 different adult rat tissues which varied 10-fold in their cellular RNA concentrations. The cellular levels of the cytoplasmic RNAs, 28 S, 18 S, and 5 S rRNA, 4 S tRNA, K, and L, were found to be a function of the total cellular RNA concentration. In contrast, the cellular levels of the nuclear RNAs, A, C, D, and H, varied slightly between tissues and were independent of the total cellular RNA concentration. However, these variations in the levels of the nuclear RNAs were significant in that they revealed these molecules are coordinately regulated.

Characterization of the association of two small molecular weight RNAs with eukaryotic polysomes.

Previous studies of the adult rat cerebral cortex have shown that two of the small molecular weight RNAs, K and L, are primarily cytoplasmic in location and are present in significant quantities in the microsomal fraction. We have examined their appearance in microsomal preparations. The results demonstrate that both K and L are, in fact, associated with polyribosomes. In particular, it is shown that (i) K and L are present in both membrane-bound and free polysomes; (ii) removal of microsomal membranes with Triton X-100 does not alter their association with microsomal polysomes; (iii) all of the K and almost one-half of the L associated with rat liver polysomes are resistant to high salt washing; and (iv) K and L are present in rabbit reticulocyte-free polysome preparations. The protein synthesis inhibitors cycloheximide and aurintricarboxylic acid significantly alter the association of K and L with rabbit reticulocyte-free polysomes. It is concluded that K and L are transiently associated with polysomes during the process of protein synthesis.

Differential and synergistic actions of nerve growth factor and cyclic AMP in PC12 cells.

When a clonal line of rat pheochromocytoma (PC12) was exposed to beta-nerve growth factor (beta NGF), N6, O2-dibutyryl adenosine 3':5' cyclic monophosphate (Bt2cAMP), or a combination of the two, 10, 26, or 70% of the cell clumps, respectively, displayed neurites after 1.d. Increases in the cellular RNA concentration were also found to be additive or greater when both agents were present. Neurites induced by Bt2cAMP alone were not maintained after replacement with beta NGF. The degree of potentiated neurite outgrowth was a function of the time of simultaneous exposure to both agents. The initiation of neurite outgrowth in the presence of Bt2cAMP was independent of RNA synthesis, in contrast to that induced by beta NGF alone. We conclude that beta NGF-induced initiation of morphological differentiation of these cells is not mediated by a cAMP-dependent mechanism. Consideration of Bt2cAMP effects upon other cell lines suggest that Bt2cAMP causes a rapid, but unstable, reorganization of the PC12 cytoskeleton, resulting in the initiation of neurite outgrowth from these cells. In contrast, beta NGF alone achieves a more stable cytoskeleton reorganization by an RNA synthesis-dependent mechanism.

Nerve growth factor-induced differentiation of PC12 cells: evaluation of changes in RNA and DNA metabolism.

The relationship between both DNA and RNA metabolism and the nerve growth factor (betaNGF)-induced differentiation of PC12 cells was investigated. Cells remained in log phase growth for 4 days after exposure to betaNGF. DNA synthesis was similarly unaffected by betaNGF during this time. Thereafter, a gradual decrease in DNA synthesis was observed. In contrast, as many as 40% of the cells display neurites after 4 days exposure to betaNGF. The implication, that cells which have begun to differentiate morphologically may still synthesize DNA, was confirmed using autoradiography. The cellular RNA concentration was elevated significantly within 1 day of exposure to betaNGF and continued to increase for up to 6 days. Dose-response analysis revealed that this increase in cellular RNA, both rRNA and tRNA, closely correlates with the size and number of neurites per cell rather than with the cell's capacity to grow a neurite per se. A parallel rise in the cellular protein concentration was observed also. It was concluded, therefore, that (1) cessation of cell division is not a prerequisite for the initiation of the morphological differentiation of these cells and (2) the elevation of the cellular RNA content influences some quantitative aspects of neurite outgrowth.