When John Heuser finished medical school at the height of the Vietnam War he was immediately eligible for the draft. Luckily for him, and for cell biology, he fulfilled his military service at the National Institutes of Health (NIH) in the United States Public Health Service. It was there that he brought the concept of “membrane recycling” to light.
He brought to Thomas Reese's lab at the NIH a postdoctoral project he had started with Sir Bernard Katz at University College, London—an attempt to capture a picture of neurotransmitter “quanta” being released as Katz had proposed. The only approach available to him at the time, Heuser recalls, “was to stimulate the nerve like hell and throw it into fixative.” But he soon realized, “the chances were still almost zero of catching it.” The method, however, gave him images of nerve terminals at the frog neuromuscular junction that had “weird vesicles and membranous cisternae” inside. Heuser, now at Washington University (St. Louis, Missouri), recalls that other scientists who saw the images said the overstimulated nerves were “disgusting,” “just destroyed,” and “not relevant to anything.”
Heuser had a hunch, however, that the internal structures were not just signs of degradation, but instead were products of endocytosis. Together with Tom Reese, he decided to investigate his idea using horseradish peroxidase (HRP), the endocytic tracer that was just coming into its own as a powerful marker in electron microscopy. They stimulated frog nerve terminals while bathing them in extracellular HRP and found that HRP first appeared in clathrin-coated vesicles that formed from the nerve terminal plasma membrane. These vesicles then coalesced, and the HRP showed up in the internal cisternae (now known to be endosomes). Finally, the HRP ended up in a new population of synaptic vesicles as they reformed (Heuser and Reese, 1973).
This evidence, along with the pair's meticulous accounting of membrane fluxes between synaptic vesicles, plasma membrane, and cisternae, argued strongly for a rapid recycling of synaptic vesicle membrane via endocytosis. The model figure at the end of the paper headed straight into textbooks and the paper received more than 1,300 citations.
Heuser says he chose the term “recycling” deliberately, both because of the new environmental movement and because it made a critical distinction: “The synaptic vesicle is not like a cola bottle that never loses its integrity when returned to the factory to be filled again. Instead, it melts into the plasma membrane and is completely reformed, like an aluminum beer can.”
This laid the groundwork for what would be called the “kiss-and-run” hypothesis: that synaptic vesicles could deliver their cargo by fusing slightly with the membrane and then reform by pinching back off (Fesce et al., 1994). These different views led to a decade of competition between the two groups. Heuser says, “We were using the same preparations, and the results were identical, too. But we had opposite interpretations.”
Heuser and his colleagues substantiated their model with studies showing a correlation between synaptic vesicle exocytosis and quantal transmitter release (Heuser et al., 1979), thus confirming the one vesicle–one quantum theory. By then, Steinman et al. (1976) had shown that in nonneuronal cells so much membrane was coming into the cell (cells pinocytosed their entire cell surface area in ∼30 min) that there must be a general recycling flow back to the plasma membrane. This work therefore defined a recycling pathway with the plasma membrane as destination rather than source. Meanwhile, Heuser went on to capture beautiful images of the structural changes that clathrin goes through during receptor-mediated endocytic events (Heuser and Evans, 1980).
For synaptic vesicle exocytosis, the question of kiss-and-run versus full fusion plus recycling is still very much up in the air. Some researchers now believe that both forms of exocytosis occur, but that cells use kiss-and-run when vesicles are in short supply (Wightman and Haynes, 2004). Heuser isn't so sure. “Kissing and running isn't an option for membrane compartments as tiny as synaptic vesicles,” he says. By the time these vesicles exocytose they lack any protective coat that could maintain their shape. After fusion, “the surface tension on such small membrane spheres is probably so great that their exocytosis cannot be reversible. I started thinking that after our 1973 paper, and still believe it.”