Discovery of motors (1983-1986). In my first independent position - with NIH at the Marine Biological Laboratory in Woods Hole, MA - Tom Reese, Mike Sheetz, Ron Vale, and I worked together to elucidate the mechanism of axonal transport. By making technical improvements in computer enhanced light microscopy, and developing a method to dissociate squid axoplasm, we obtained real-time images of organelle movements along isolated "filaments", which we identified in the EM as single microtubules. My principal contribution to this work was to develop the methods and technology to image single, unstained microtubules in real time. This work established the basic role of microtubules in intracellular transport. We then set out to reconstitute vesicle transport along microtubules, the latter assembled in vitro from purified cow brain tubulin. We devised in vitro motility assays that revealed the existence and polarity of cytoplasmic motors. These assays established that distinct motors promote transport toward opposite ends of the microtubule, and enabled us to purify the plus-end motor, kinesin, led by Ron Vale, and the minus-end motor, cytoplasmic dynein, by myself. Overall, these studies raised a number of basic questions that we and other labs have been pursuing over the years: 1) What is the mechanism by which molecular motors walk along microtubules? 2) How do motors organize traffic in the cell? 3) How are motors linked to cellular cargos, and what exactly are these cargoes?

Biophysics of kinesin motility (1986-1993). Again in the context of collaborations, I produced a body of biophysical work that used the in vitro motility assays not, as before, to purify proteins, but instead to bring to light fundamental aspects of the molecular mechanism of kinesin motility. My initial experiments, using the inhibitor AMP-PNP, provided initial evidence that the two heads of kinesin work in an alternate fashion, i.e. that kinesin walks "arm-over-arm" along the microtubule. Prominent in the next phase of my work was the development and use of advanced light optical methods to reveal the molecular mechanics of kinesin walking. These studies depended on a series of collaborations. My first attempts at detecting the molecular scale motion produced by kinesin were pursued in collaboration with Drs. Jeff Gelles and Mike Sheetz. Using a novel centroid tracking program invented by Dr. Gelles, we computed the displacement of beads carrying multiple kinesins, and thereby demonstrated that kinesin tracks along single protofilaments. To determine whether bead movement could be driven by a single kinesin, I collaborated with Dr. Steven Block who at that time was investigating bacterial motility with optical tweezers. With Dr. Block, I interfaced optical tweezers to the microtubule imaging equipment in my lab, and developed a reproducible single kinesin motility assay: This assay utilized optical tweezers to place onto microtubules individual beads, each carrying a single kinesin. These studies demonstrated that a single kinesin is processive, i.e. in line with the "arm over arm" model that we proposed earlier. A direct outgrowth of this work was the demonstration, using an" optical trapping interferometer", that a single kinesin walks along the microtubule in a step-wise manner, pausing at 8 nm intervals-the tubulin dimer repeat. Karel Svoboda, a biophysics Ph.D. student with Dr. Block, led this work under our guidance.

Cargo-motor interactions (1993-present). Since 1993, the research on motors in my lab has centered on how kinesins and dynein organize intracellular traffic. While scores of kinesins have been cloned, their cargoes, and the nature of the cargo-motor linkage eluded us and the rest of the field until very recently. We recently published two reports, one on cytoplasmic dynein, the second on kinesin , which evolved independently from very different projects, yet brought us to the same conclusion: the motor-cargo linkers are soluble scaffold proteins with multiple cellular functions.
First project. With the goal of reconstituting vesicle transport from isolated components, we produced a body of work that exploited squid axoplasm, for its supply of highly purified vesicles that move efficiently along microtubules in vitro. This project entailed technical advances in vesicle isolations and motility assays, e.g. we devised methods to biochemically separate retrograde from anterograde vesicles. Among other findings, we elucidated how these two vesicle populations differentially employ kinesin and dynein motors to move in opposite directions within the axon. In the most recent work, published in 2001, we reconstituted dynactin-dependent dynein-driven vesicle transport using liposomes of defined composition. Using motility assays as the experimental endpoint, we show that the dynactin complex, and thereby dynein, is recruited to acidic phospholipids by axonal spectrin - a scaffold protein that binds directly to acidic phospholipids in the membranes of vesicles.
Second project. With the goal of identifying the cargoes and motor-cargo linkers for neuronal kinesins, my lab screened for and cloned families of kinesins expressed in brain (1993-6). In addition to characterizing a subset of these independently, I initiated a collaboration (1995) with Tom Rapoport in my department (Cell Biology, Harvard Medical School), for the purpose of identifying the cargo and the cargo-motor linker for the conventional kinesin isoform isolated in our screen. To this end, we combined our backgrounds - in membrane trafficking (Rapoport) and motors (Schnapp) - and formalized our collaboration by co-supervising a small group of postdoctoral fellows. This collaboration produced a pair of papers, led by Kristen Verhey, that made progress with respect to the cargo question. First, we showed that cargo-less kinesin in vivo is inhibited from interacting with microtubules, and we defined the molecular underpinnings of this regulation, in particular, the role played by the kinesin light chains. In the newest work, we identified the partner proteins of the kinesin light chain TPR motifs. These partners are JIP scaffolding proteins for the JNK family of MAP kinases; they appear to play a role in neurogenesis by scaffolding a signaling cascade activated by the Reelin ligand. By defining the molecular basis of the JIP-kinesin interaction, and showing that this interaction is essential for JIP localization to nerve terminals, we established a relationship between a kinesin, a motor-cargo linker molecule, and the cargo itself. In essence, we have learned that 1) kinesins (like dynein) are linked to their membranous cargo via soluble scaffolding proteins that have additional functions, and 2) motor proteins localize pre-assembled signaling pathways.
To read about similar findings from other laboratories, check out the kinesin home page

RNA localization (1995-present). Like axonal transport before 1985, how specific RNAs are transported and confined to particular subregions of the cell is an extremely interesting and far-reaching problem yet to be solved at a molecular level. Using Xenopus oocytes, we have been isolating the proteins that bind specifically to the localization element of a number of transcripts transported to the vegetal cortex. We made progress by identifying repeated UUCACs as cis-acting targeting signals that direct transcript localization in the context of diverse transcripts. On the basis of its affinity to these signals, we purified one of the first reported transacting factors for localization, a protein we call "Vera" (also known as "zipcode binding protein") that has four KH domains. Vera/ZBP is now recognized as a general factor involved in the localization of diverse transcripts in many different cells. Vera has extremely interesting properties: it associates with the ER, shuttles in and out of the nucleus, and is highly conserved (80% overall identity from frog to man), from Drosophila to man. We have recently turned to Drosophila for addressing the key question of how Vera implements the localization of specific mRNAs. We are using genetic, biochemical, and live cell imaging approaches to address the function of this protein.