News and Publications

Actin Dynamics in Cell Morphogenesis and Function

V.M. Fowler, R.S. Fischer, K.L. Fritz-Six, N.J. Greenfield,* M. Mardahl-Dumesnil, J. Moyer, K. Sampson

* UMDNJ, Piscataway, NJ

Regulation of actin dynamics at the ends of filaments determines the organization and turnover of actin cytoskeletal structures and is critical for cell motility and cell architecture. For example, when cells change shape or crawl during cellular and tissue morphogenetic movements, new actin filaments are rapidly assembled at the barbed ends of the filaments and disassembled at the pointed ends during extension of lamellipodia or filopodia. In contrast, in differentiated cells such as striated muscle and red blood cells, actin filaments are organized into regular architectural arrays that persist for the lifetime of the cell and are important for maintenance of cell shape, mechanical properties, and physiologic function. Our goal is to elucidate and compare the distinct regulatory mechanisms that control the polymerization and dynamics of actin filaments in the rapidly turning over filaments of motile cells with the regulatory mechanisms involved in the stable, long-lived actin filaments of nonmotile cells. Specifically, we focus on the regulation of the pointed ends of filaments by the actin-capping protein tropomodulin.

Tropomodulins are a conserved family of proteins of approximately 40 kD that cap the pointed ends of actin filaments. These proteins are present in vertebrates, flies, and worms. The function of tropomodulin in vivo is best understood in striated muscle cells, where the protein is associated with the free, pointed ends of thin filaments in sarcomeres. Previously, we showed that tropomodulin directly controls the length of thin filaments in living cardiac muscle cells by transiently binding to the pointed ends of the filaments and competing for the addition of actin monomer to the ends. Increasing the level of tropomodulin prevents net monomer association at pointed ends, resulting in shorter thin filaments. In contrast, inhibition of the association of actin monomers at the barbed ends does not affect filament length.

These results led to the unexpected conclusion that the precise lengths of thin filaments in muscle are maintained via ongoing regulation of actin dynamics at the pointed ends of the filaments. A related conclusion is that the lengths of filaments in cardiac muscle cells are not predetermined to a unique length and thus most likely are not determined by a molecular ruler mechanism.

These conclusions were made possible by use of a novel fluorescence-based method we developed, distributed deconvolution, to measure (1) the incorporation of rhodamine-actin at the ends of filaments and (2) the lengths of filaments on the basis of phallacidin staining of F-actin. This computational averaging method enabled us to determine the distribution and intensity of fluorescent probes along thin filaments in living cardiac muscle cells with an accuracy and precision similar to the accuracy and precision of electron microscopy. In related experiments in which we used this method, we detected significant variations in the lengths of thin filaments among individual myofibrils isolated from the posterior latissimus dorsi muscle of an adult chicken. The results were again inconsistent with the idea that a ruler molecule (e.g., nebulin) strictly determines the lengths of thin filaments in this muscle. However, because tropomodulin can bind to the end of nebulin located at the pointed ends of thin filaments, the interaction between tropomodulin and nebulin may contribute to the regulation of filament length in some muscles.

In more recent work, we found that capping of pointed ends by tropomodulin plays an important role during myofibril assembly: it specifies the final filament lengths in mature myofibrils. We used sanpodo, the Drosophila homolog of tropomodulin, to investigate the function of tropomodulin during muscle development in Drosophila indirect flight muscle. We found that transient increases in sanpodo at any time during the development of the muscle led to an irreversible block in elongation of preexisting thin filaments. Flies with a preponderance of abnormally short thin filaments could not fly.

These results indicate that thin filaments elongate from their pointed ends during the assembly of myofibrils and that pointed ends are dynamically capped at endogenous levels of sanpodo so as to allow elongation, similar to the situation in cardiac muscle cells. However, transient increases in sanpodo levels during myofibril assembly in the development of the indirect flight muscle converted sanpodo from a dynamic to a permanent cap.

Clues to a molecular mechanism that could convert tropomodulin from a dynamic to a permanent pointed cap at the end of filaments come from studies on the unique actin-capping properties of tropomodulin in vitro. Namely, alone among all capping proteins, tropomodulins bind tropomyosin, and the affinity of tropomodulin for pointed ends is upregulated more than 1000-fold in the presence of tropomyosin. In the presence of tropomyosin, tropomodulin capping of actin pointed ends is no longer dynamic, suggesting that tropomyosin could be a structural regulator of the capping activity of tropomodulin at pointed ends. We are investigating this possibility by examining the structure of tropomodulin alone and in complex with tropomyosin, in collaboration with N. Greenfield, UMDNJ, Piscataway, NJ. We used circular dichroism spectroscopy to study the interaction of a tropomodulin fragment containing residues 1-130 with a designed N-terminal tropomyosin peptide. Our results indicated that tropomodulin residues 1-130 are almost completely unstructured in solution and that interaction with the tropomyosin peptide causes a conformational change in the tropomodulin fragment, promoting increases in both a-helix and b-sheet structure. We speculate that this conformational change in tropomodulin could convert tropomodulin from a dynamic to a permanent actin cap.

PUBLICATIONS

Greenfield, N.J., Fowler, V.M. Tropomyosin requires an intact N-terminal coiled coil to interact with tropomodulin. Biophys. J. 82:2580, 2002.

Littlefield, R., Fowler, V.M. Measurement of thin filament lengths by distributed deconvolution analysis of fluorescence images. Biophys. J. 82:2548, 2002.

Mardahl-Dumesnil, M., Fowler, V.M. Thin filaments elongate from their pointed ends during myofibril assembly in Drosophila indirect flight muscle. J. Cell Biol. 155:1043, 2001.