One of the things that drew me into biology was the recombinant DNA
revolution. I was studying chemistry at the University of Illinois in
1977 when I took a class that discussed the new methods to rearrange
DNA. I was amazed by this and decided that I would like to work on
recombinant DNA and so left chemistry to follow this new wave of
biological discoveries waiting to happen. I did my graduate work at
MIT working on DNA repair and found time for side projects on the
development of new methods for generating recombinant DNA,
essentially my hobby. During this period I developed the first phasmid
cloning vectors that contained both a plasmid and phage origin, and
this understanding of phage biology became the crux of many future
discoveries.
I was a post-doctoral fellow in the Biochemistry Department at
Stanford University and by complete accident I cloned a gene involved
in synthesizing dNTPs needed for DNA replication. Since it was an
accident, I wasn't very interested in it. However, since I was prone
to working on extremely high-risk projects in hopes of doing something
important, I was in need of a safe project in case I ever hoped to
actually get a faculty position after my post-doc. So I followed up on
the ribonucleotide reductase genes, eventually discovering that they
were induced by DNA damage. This was an important discovery because it
became the basis for a number of genetic screens that led to the
discovery of a signal transduction pathway that senses and responds to
DNA damage. However, that was not the most important aspect of this
gene with respect to this story; it was also cell-cycle regulated,
which got me thinking about the cell cycle. Not long after that, Paul
Nurse, a leader of the cell-cycle field and co-winner of the 2001
Nobel Prize in medicine, gave a talk at UCSF and described isolating
the human homolog of a key cell-cycle kinase gene, Cdc2, by
introducing a human cDNA library into an S.pombe mutant. The
methods were primitive but the message was clear, that the core of the
cell cycle was functionally conserved and many human genes could be
isolated by complementation cloning in yeast. This led to a marriage
of my two interests, cloning technology and the cell cycle, and I
further developed new cloning methods to make human cDNA libraries in
yeast expression vectors and streamlined them to make them easy to
use.
When I started my faculty position at the Baylor College of
Medicine in Houston, the first experiment I performed was to introduce
these new libraries into yeast cell-cycle mutants. The first gene I
discovered was the human Cdc2 gene isolated by Nurse two years
previously. However, I also isolated in that screen a relative of this
kinase, Cdk2, which work from my lab and others subsequently revealed
to be the key protein controlling the G1 to S phase transition, the
critical transition for cancer. This was a key discovery and began my
lab's involvement in human cell-cycle control. I then applied my newly
developed cloning technologies to an emerging method for detecting
protein-protein interactions called the two-hybrid system. I converted
that system from a method for detecting interactions into a method for
cloning cDNAs encoding proteins that associate with a target protein.
Using Cdk2 as bait, my colleague Wade Harper and I isolated the
p21CIP1 gene and showed it to be the first of a family of Cdk
inhibitors. Bert Vogelstein's lab also identified the same gene as a
gene regulated by p53 and we published our papers back-to-back in
November of 1993. This was a watershed event in cell-cycle research
and helped forge the connection between p53 and cell-cycle control.
Our studies showed that p21 was a member of a family of inhibitors
that the whole field began to work on. We also discovered that one
member of this family, p57KIP2, was found to be mutant in individuals
with the familial overgrowth and cancer predisposition disease
Beckwith-Weidemann Syndrome. Our group was the first to demonstrate
that these inhibitors were key regulators of developmental growth
control being involved in everything from lens to muscle development.
Another important discovery came from looking for human genes that
could complement other yeast cell-cycle regulators. In 1996, we found
that human cyclin F could allow yeast cdc4 mutants to live. Cdc4
mutants fail to enter into S phase because they need to degrade the
Cdk inhibitor Sic1. Through a series of experiments we found that both
the human cyclin F and yeast Cdc4 proteins could bind to a protein
called Skp1 which was in humans and yeast. We discovered that Skp1 was
a core component of an E3 ubiquitin ligase complex and that it could
bind to a large family of proteins through a conserved motif, the
F-box. We set forth the F-box hypothesis, which stated that F-box
proteins were the specificity subunits for this ubiquitin ligase
complex we later called the SCF. At that time no one knew how proteins
were chosen for ubiquitination and subsequent destruction by the
proteosome, and only one E3 had been identified previously. I turned
to my expert biochemical collaborator Wade Harper for help and
together we established that the Cdc4 F-box protein could allow the
SCF to recognize phosphorylated Sic1 and ubiquitinate it. Importantly,
if we substituted a different F-box protein in, it would no longer
recognize Sic1 but would now recognize a new substrate. Since there
were hundreds of F-box proteins in the database, it opened up the
field of ubiquitin ligase specificity to a wide audience. Since our
first paper on F-box proteins four years ago, there have been over 100
papers implicating the SCF in numerous pathways including cell-cycle
control, cell morphology, amino acid control, DNA replication, gene
expression, the immune response, cell fate determination, hand and
foot development, AIDS, b-catenin regulation, plant circadian rhythms,
plant flowering, and auxin and jasmonate responses in plants. Thus,
the discovery of F-box proteins and the SCF pathway resulted in an
explosion of discoveries about the control of regulated protein
stability in biology.