Susan Ferro-Novick thought of becoming a grade school teacher or physical therapist because a job requiring an advanced degree seemed unattainable, given that her mother didn't finish high school and her father didn't finish grade school. "It seemed like a big jump in one generation," she explains. But as an undergraduate at New York University, she realized that making the jump would allow her to do research, which is what she really wanted. "I like analyzing a problem and dissecting it from the ground up to figure out all the details," she says.

Ferro-Novick entered the Genetics Program at the University of California, Berkeley in 1976, and worked with Randy Schekman (who became an HHMI Investigator in 1991). Schekman studies vesicle traffic, which became Ferro-Novick's interest as well. Vesicles are sacs of membrane that contain cargo. They bud from a donor membrane and subsequently fuse with an acceptor membrane to release their contents.

Cargo is constantly moving within vesicle sacs from one site to another, but there is no delivery service within the cell to read the address to which each sac should be taken. "So the question is," Ferro-Novick says, "How does this happen?"

By the 1990s, scientists thought they had this figured out when proteins called SNAREs were discovered. These proteins stick out from membranes like arms. According to the SNARE hypothesis, vesicles arrive at their correct destination because a specific SNARE on a vesicle pairs with a target SNARE on the appropriate acceptor compartment. "This seemed nice and simple in 1993," Ferro-Novick says. "But as people learned more, this hypothesis seemed less plausible because SNARE pairing turned out to be more promiscuous than originally thought."

Looking for other mechanisms, Ferro-Novick's group analyzed yeast mutants with trafficking defects. They identified a protein complex called transport protein particle 1 (TRAPPI) that is needed for trafficking vesicles between the endoplasmic reticulum (a series of interconnected tubules where proteins are made) and Golgi in yeast.

Recently, Ferro-Novick's group found out how vesicles derived from the endoplasmic reticulum are marked. "We discovered that specificity is determined very early on, when the vesicle buds," she explains.

Proteins made in the endoplasmic reticulum are packaged into vesicles that have a unique coat, called COPII (coat protein II). Ferro-Novick discovered that one of the coat complex's subunits, Sec23, binds to TRAPP I. Because Sec23 is incorporated into COPII as part of a complex that recruits cargo into vesicles before the coat is polymerized, it labels a vesicle according to its contents. It then docks vesicles to Golgi membranes by interacting with TRAPP I. "It looks like the coat on the vesicle interacts with the machinery that tells the vesicle where to go," Ferro-Novick says. "This was a surprise because everyone thought the coat was removed before a vesicle could recognize its target membrane."

Like most basic research, this work is relevant to human disease. A disorder of bone formation called spondyloepiphyseal dysplasia tardia involves a mutation in a subunit of the human version of the TRAPP complex.

While Ferro-Novick was in graduate school in Berkeley, she married Peter Novick, who also worked in Schekman's lab. A month after their twin sons were born in 1999, the couple began to collaborate on a new project. The collaborative work addresses how the endoplasmic reticulum is delivered from mother to daughter cells. Using a genetic approach, the researchers have found several novel genes that encode some of the required proteins.