Motor proteins of the eukaryotic cytoskeleton

Motility is a hallmark of life. Even cells that are incapable of active movement within their environment perform essential intracellular motility processes. A number of mechanisms have evolved to generate the mechanical forces required to drive biological motility. A particularly successful and ubiquitous mechanism of biological force production utilizes mechanochemical enzymes, or “motor proteins.” These enzymes convert chemical energy, typically in the form of adenosine triphosphate (ATP), into mechanical force.

Many of the movements, both cellular and subcellular, of eukaryotic cells are generated by the activities of motor proteins that act upon rigid cytoskeletal fibers. Three superfamilies of cytoskeletal motor proteins have been recognized. Motors of the myosin superfamily act upon actin filaments to generate cell surface contractions and other morphological changes, vesicle motility, cytoplasmic streaming, and muscle cell contraction. Members of the dynein and kinesin microtubule-based motor superfamilies move vesicles and organelles within cells, cause the beating of flagella and cilia, and act within the mitotic and meiotic spindles to segregate replicated chromosomes to progeny cells. Members of each motor superfamily can be recognized by the presence of a unique and conserved amino acid sequence that forms the ATP-binding and force-producing domain of the molecule. These conserved “motor domains” are linked to a diverse array of polypeptide sequences that presumably impart functional specificity to the various motors. Each of these superfamilies is populated with numerous distinct members that perform different cellular roles. In recent years, molecular genetic techniques have revealed novel motor proteins more quickly than our ability to determine their functions. The talks in this session of the German–American Frontiers of Science meeting addressed the cellular roles of some of these motor proteins.

Mitotic Motors of the Yeast Saccharomyces cerevisiae.

The segregation of replicated chromosomes in eukaryotic cells is accomplished by the mitotic spindle, a microtubule-based, force-producing organelle. For mitotically dividing S. cerevisiae (budding yeast) cells, the only essential role of microtubules appears to be for spindle function. Not surprisingly, all but one of the seven microtubule-based motors (one dynein and six kinesin-related) encoded by the S. cerevisiae genome have been linked to spindle function (Table 1). The exception, Smy1p, performs a nonessential role in polarized cell growth (1, 2). Due to functional overlap or redundancy, none of these seven motors are individually essential for cell viability. In this simple organism, every essential spindle movement is accomplished by more than one motor.

Table 1.

Experimentally determined roles for S. cerevisiae microtubule-based motors

Motor	Role(s)	Ref(s).
Dynein	Spindle elongation	21
	Spindle positioning	22, 23
Kinesin-related motors
Cin8p and Kip1p	Spindle assembly	24, 25
(BimC motors that  overlap in function)	Spindle elongation	21
Kip2p	Spindle positioning (antagonizes Kip3p and Kar3p)	10
Kip3p	Spindle positioning	10, 11
Kar3p	Spindle assembly	6, 26
	Antagonizes Cin8p and Kip1p to restrain spindle elongation until anaphase	5, 6, 9
	Kinetochore movement	27
	Spindle positioning	10
	Karyogamy (nuclear fusion during mating)	26
Smy1p	Polarized cell growth	1, 2

The mitotic spindle is a bipolar structure composed of two half-spindles that face each other with their microtubules interdigitating. Bipolar spindle assembly requires the actions of kinesin-related proteins of the BimC family (3, 4). The two BimC motors in S. cerevisiae, Cin8p and Kip1p, also produce an outwardly directed force acting upon the spindle poles that causes the elongation of the spindle during anaphase. During the bipolar spindle assembly stages prior to anaphase, this BimC-derived outward force is balanced by an inwardly directed force acting upon the poles. The inward force is produced in part by kinesin-related Kar3p in S. cerevisiae and by related motors in other eukaryotes (5–8). Therefore, the assembly of the bipolar spindle is achieved by motors that antagonize each other by exerting force in different directions (9).

During mitosis, spindles are specifically positioned within eukaryotic cells by motile processes. Proper positioning of the spindle is often essential for cell division and differentiation processes. The asymmetric cell division characteristic of budding yeast requires that the spindle be positioned and properly oriented at the neck separating the mother and bud cell bodies. Four of the seven S. cerevisiae microtubule-based motors have been implicated in this process (Table 1; refs. 10 and 11). Interestingly, an antagonistic relationship between motor activities appears to be important for positioning of the spindle as well as for spindle assembly. In the absence of the dynein motor, the spindle positioning activities of Kip3p and Kar3p are antagonized by the actions of Kip2p, a motor normally required for proper spindle positioning (10).

The Role of Dynein in Assembly of Mitotic Spindles.

Spindle bipolarity is essential to direct the segregation of a duplicated set of chromosomes into two progeny cells. One of the fundamental questions in assembly of mitotic spindles is how the two spindle poles are formed. In some spindles, these poles consist of two centrosomes (animal cells), or spindle pole bodies (fungal cells), that define the two ends of the spindle and nucleate and regulate spindle microtubules. However, in other spindles, poles seem to form de novo without the apparent direction of a centrosome or related structure (12). The assembly of poles de novo can be studied in concentrated extracts of Xenopus eggs where, following the addition of chromatin, poles assemble by a microtubule-based motor-driven process. It appears that poles form by the sorting of randomly nucleated microtubules into a bipolar structure (13). This process requires the activity of the cytoplasmic dynein motor. Microtubules possess a structural polarity with minus ends associated with the poles of spindles. Cytoplasmic dynein motors move toward the minus or poleward ends of microtubules. It appears that a number of dyneins clustered together, perhaps on a membrane vesicle, can collect microtubules together into two poles of uniform microtubule polarity. The ability of motors to sort microtubules into polar structures is an interesting example of self-organization in biology.

Myosins on the Move.

In recent years, the number of identified myosin superfamily members has increased exponentially. It has become clear that virtually all eukaryotic cells, not only muscle cells, express a multitude of different myosin molecules (14). The myosins share a conserved motor domain that consists of separate head and neck regions. The head region contains the ATP- and actin-binding sites and exhibits actin-activated ATPase activity. The neck region consists of an extended α-helix of variable length that binds between one and six light polypeptide chains of calmodulin or calmodulin-related proteins. This neck region is proposed to serve as a lever arm for force production (15). Physiological modification of the light chains (i.e., by phosphorylation) contributes to the regulation of motor function. During its mechanochemical cycle, the myosin responsible for muscle contraction does not move processively along the actin filament. Instead, it holds on for only a short period and then spends considerable time detached from the filament (16). Therefore, to achieve continuous movement along actin filaments, a high density of myosin motors is required. Accordingly, muscle myosin self-assembles into filaments. The multitude of newly discovered myosin molecules are believed not to form filaments, and the mechanochemical properties for most remain to be investigated. In addition to the motor domain, the different myosin molecules contain diverse tail domains that are postulated to specify function, perhaps by determining the target of force generation. These tail domains frequently contain amino acid sequence motifs that are also found in other proteins (17). The elucidation of the targets to which the different myosins bind represents an important challenge for the future. The tail domain of a myosin identified from rat tissue, myr5, serves to negatively regulate signal transduction by the small Ras-related G protein Rho (18). Rho regulates the organization of the actin cytoskeleton and various other cellular processes (19, 20). This finding suggests the fascinating possibility that initiation and flow of information might be coupled to directed force production along actin filaments.