Previous immunolocalization studies using many primate cultured cell lines demonstrated that a
microtubule-associated protein of M(r) approximately 210,000 which is now called
MAP4, is present along the length of microtubules in interphase and mitotic cells [Bulinski and Borisy (1980) J. Cell Biol. 87:802-808; DeBrabander et al. (1981) J. Cell Biol. 91:438-455]. Since
MAP4 has been implicated as a microtubule stabilizer, we asked whether all classes of microtubules possess an equal
complement of
MAP4. We have reexamined the cellular distribution of
MAP4, using both conventional double-label immunofluorescence and an antibody blocking technique [Schulze and Kirschner (1987) J. Cell Biol. 104:277-288] to highlight microtubules lacking, or depleted in,
MAP4. These techniques have revealed that thin processes extending from monkey kidney cells (TC-7), and those made by human
neuroblastoma cells (IMR-32) in response to
retinoic acid, are often deficient in
MAP4 immunoreactivity. Since both types of cellular processes contain stable microtubules, which are enriched in detyrosinated (Glu)
tubulin, we tested the ability of
MAP4 to bind to microtubules made from pure Glu and pure tyrosinated (Tyr)
tubulin in vitro.
MAP4 bound to both types of microtubules, and the similar saturation level of
MAP4 binding to Glu and Tyr microtubules suggested that differential binding to these forms of
tubulin does not contribute directly to a mechanism for segregation of
MAP4 on microtubules in vivo. In TC-7 cells, we also observed MAP4-depletion on single microtubules, distal regions of broad cytoplasmic extensions, and midbodies of dividing cells.
MAP4 depletion may reflect recent, rapid growth of microtubules to which
MAP4 has not yet bound, or the presence of other MAPs that may compete with
MAP4 for binding sites on the MT. We suggest that different levels of
MAP4 on microtubules may directly modulate microtubule dynamics within single cells, as well as other microtubule functions such as those involving microtubule motor activity.