Most studies of cytoskeletal organelles have concentrated on molecular analyses of abundant and biochemically accessible structures. In many of the classical cases, however, the nature of the system chosen has precluded a concurrent genetic analysis. The mitotic spindle of the yeast Saccharomyces cerevisiae is one example of an organelle that can be studied by both classical and molecular genetics. We show here that this microtubule structure also can be examined biochemically. The spindle can be isolated by selective extractions of yeast cells by using adaptations of methods successfully applied to animal cells. In this way, microtubule-associated proteins of the yeast spindle are identified.Microtubule-associated proteins are the focus of much attention, for it seems plausible that these minor components of microtubule structures may be the factors that help to define specific parameters of the repertoire of microtubule organelles: state of assembly, localization, organization, interaction, and regulation offunction. This suggestion arises from the properties of associated proteins in vitro and from unique localizations of some of them in vivo. It has not yet been possible to integrate these sorts of results into a unified picture of precise microtubule functions in vivo. For this reason, increasing interest is focused on those organisms with genetic systems that may serve to bridge the gulf between molecular characterizations and cellular functions.The budding yeast Saccharomyces provides an especially attractive system for such studies. Its mitotic cell cycle has been well characterized by electron microscopy (1-3), light microscopy (4), and immunofluorescence microscopy (5).Mutations and suppressors of mutations in the f-tubulin gene have been defined, as have a variety of conditional mutations which lead to cell-cycle arrests (6-8). Tubulin can be purified from yeast extracts (both Saccharomyces cerevisiae and Saccharomyces uvarum), relying largely upon the in vitro assembly reaction that has been developed for isolating tubulin from several types ofanimal cells and tissues (9). That the yeast tubulin assembles under the same conditions as those applied to other tubulins demonstrates a common property, but there are important differences as well. In particular, Kilmartin (9) noted a different number ofprotofilaments in the assembled microtubules and different sensitivities to cold and to microtubule-depolymerizing drugs. In the filamentous fungus Aspergillus nidulans (10), focus has been primarily on the tubulin components of its microtubule systems; no associated proteins have been identified.In this paper, we have used an alternative approach to identifying cytoplasmic microtubule components. We describe here the successful application to yeast of a technique that has been used to define microtubule-associated proteins in a variety of vertebrate cells (11). This technique has been used to identify species, cell-type, and cell cycle-specific associated proteins (12,13). Antisera to some of these p...