* Dyson Institute for Vision Research, Joan and Sanford Weill Medical College of Cornell University, New York, New York 10021, USA‡ Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA† Present address: Department of Ophthalmic Research, Cleveland Clinic Foundation, Cleveland, Ohio 44195, USA § e-mail: boulan@ mail. med. cornell. edu n higher eukaryotes, secretory and plasma-membrane proteins are transported from the endoplasmic reticulum (ER) to a central Golgi complex and subsequently packaged into membrane-bound carriers for delivery to the cell surface. The long-distance transport of post-Golgi organelles to the tips of axons or to developing hyphal extensions in Neurospora crassa shows an absolute requirement for microtubules and microtubule-associated motors1, 2. In contrast, microtubule disruption only moderately attenuates Golgi-toplasma-membrane transport in fibroblasts and randomizes surface delivery of select proteins in epithelial cells3, 4. The observed preservation of biosynthetic transport after microtubule disruption is probably due to the extensive fragmentation and redistribution of Golgi mini-stacks to regions immediately adjacent to both the ER and the plasma membrane3, 5, 6. Here we have designed experiments to test the hypothesis that when the characteristic central localization of the Golgi is preserved, microtubules, kinesin and the GTPase dynamin are essential for post-Golgi trafficking. We ruled out a pharmacological approach to tackling this problem because we found that microtubule antagonists caused dispersal of the Golgi complex before complete microtubule disassembly occurred (see

Supplementary Information). Instead, we microinjected functionblocking anti-kinesin antibodies HD and SUK-4 or cDNAs encoding a dominant-negative form of dynamin into cells expressing a green fluorescent protein (GFP)-tagged apical-membrane protein, p75. We show that kinesin and dynamin are required for different stages of post-Golgi transport.