Dynactin drives dynein without microtubule binding

Stalling secretase assembly A lzheimer’s disease might be slowed by inhibiting γ-secretase, the membrane protease complex that cleaves amyloid precursor protein (APP). Spasic et al. (page 629) now identify an endogenous inhibitor that prevents γ-secretase complex assembly and activity and thus might be targeted for therapy. APP cleavage by γ-secretase leads to amyloid plaque deposition, one possible cause of Alzheimer’s symptoms. γ-secretase is composed of four proteins—presenilin, nicastrin, PEN-2, and APH-1—which must come together for cleavage activity. Although all four components are present in the ER, their assembly into functional γ-secretase is somehow restricted; most of the active enzyme is found close to the cell surface. Assembly of γ-secretase begins with the binding of nicastrin to APH-1. This binding, Spasic and colleagues now fi nd, is prevented early in the secretion pathway by Rer1p, a membrane receptor that retrieves proteins from the Golgi back to the ER. Rer1p binds to nicastrin, thus interfering with nicastrin’s ability to bind APH-1. Decreasing the amount of Rer1p led to an increase in γ-secretase activity. Exactly what triggers Rer1p to release nicastrin and allow it to bind to APH-1, and subsequently to the other γ-secretase components, remains to be determined. Preventing this release might provide a means to reduce γ-secretase activity and thus amyloid plaque formation. D ynactin might bind to and organize microtubules. But it doesn’t need to bind microtubules to jump-start dynein motoring, Kim et al. report on page 641. Dynactin retrieves microtubule motors such as dynein from the cytoplasm and docks them onto their cargo. Dynactin also anchors the minus ends of microtubules on the centrosome. One dynactin isoform that is found in human neurons lacks its microtubule-binding domain (MBD). Kim and colleagues thus supposed that dynactin might be functional even without locking onto microtubules. They found that indeed cargo transport does not rely on the MBD. With or without a functional MBD, dynactin helped move four different physiological cargos, including vesicles and proteins, just as far along microtubules and at the same velocity and frequency. Dynactin’s MBD was previously thought to increase dynein’s processivity by giving it a second hand to hang onto microtubules. But this idea was based largely on studies of the transport of dynein-coated beads. This artifi cial cargo, or the way dynein and dynactin were attached to it, might have altered the normal relations between dynein and dynactin observed in a cell. Not all dynactin activities were unaffected by the loss of the MBD, however. Cells expressing the MBD-deleted form commonly arrested in prometaphase with multipolar spindles, suggesting a failure of microtubules to correctly conjoin in centrosomes and thus a role of dynactin in latching together microtubules.