Contact

Duke Health News

919-660-1306

DURHAM, N.C. – In findings that may enhance efforts to
starve tumors, Duke University Medical Center researchers say
they have generated antibodies in rabbits that inhibit the same
cellular target as angiostatin and actually surpass the natural
protein's ability to prevent cell growth in the laboratory.

The researchers said they will have to develop "humanized"
antibodies – ones that won't be attacked by patients' immune
systems – before any clinical application of their findings is
possible.

Angiostatin appears to be one of the most important
molecules in controlling angiogenesis, the complex process of
creating new blood vessels, the scientists said. While
angiostatin and drugs impacting other angiogenesis pathways are
already being tested in patients, the researchers said the new
understanding of the process may lead to improved attempts to
cut off blood flow to tumors or to replenish normal tissues
such as the heart.

The researchers had already shown that angiostatin binds to
an enzyme called ATP-synthase that they found on the cell
surface, but until now they did not know whether this binding
actually caused angiostatin's effects. The new study, funded by
Duke University, is the first proof that ATP-synthase does, in
fact, act as a receptor for angiostatin, the researchers
reported in Tuesday's issue of the Proceedings of the National
Academy of Sciences.

"In this study we've shown for the first time that the
cell-surface ATP-synthase enzyme retains its normal activity –
creating ATP – and that it is inhibited by angiostatin and the
antibodies we've generated," said first author Tammy Moser,
research associate in the department of pathology at Duke
University Medical Center. "This is a major new
observation."

The Duke scientists, led by Moser, reported in 1999 that
angiostatin binds to ATP-synthase on the surface of endothelial
cells, the cells that line blood vessels. Previously,
ATP-synthase was known only to exist inside cells, and its only
known role was as an energy-producing molecule.

"The bottom line is that the proliferation of new blood
vessels is energy dependent," said Dr. Salvatore Pizzo, the
study's principal investigator and chair of pathology at Duke.
"If you didn't need ATP-synthase on the cell surface, then
inhibiting it wouldn't change proliferation."

The researchers have now shown that the cell-surface
ATP-synthase is identical to that inside the cell's powerhouses
– tiny sacs called mitochondria. They also reported that
angiostatin controls blood vessel cell growth by interfering
with the enzyme's normal function, turning adenosine
diphosphate (ADP) into adenosine triphosphate (ATP), which
stores energy the cell can use to power other processes.

"The ATP-synthase enzyme is used as a receptor in its active
form by angiostatin," said Moser. "This represents a new
biological paradigm – that the receptor function is targeted,
not just its structure."

The goal of angiogenesis-based therapies is to control the
growth of new blood vessels, which ultimately depends on the
proliferation of endothelial cells. In cancer,
anti-angiogenesis treatment would stop vessel growth to starve
a tumor, while in heart disease and other circulatory
conditions, blood vessel growth caused by
angiogenesis-enhancing treatments could help restore blood flow
to oxygen-starved tissues.

"The fundamental difference between targeting the
angiostatin receptor compared to other angiogenesis mechanisms,
such as the vascular endothelial growth factor (VEG-F)
receptor, is that the body compensates for those other
pathways," explained Pizzo, also a member of the Duke
Comprehensive Cancer Center. "The beauty of targeting the
angiostatin receptor – and we now know that's what ATP-synthase
is – is that it doesn't matter what drive creates angiogenesis,
you can shut it down."

The new study is the first to show that antibodies that
recognize ATP-synthase can mimic angiostatin's effects, the
researchers said.

"Our identification of surface ATP-synthase as angiostatin's
binding site was greeted with much skepticism in 1999, and at
the time we speculated that the enzyme served as angiostatin's
receptor," Pizzo said. "With this paper, no one will be able to
say that ATP-synthase is not angiostatin's receptor or that
this enzyme is not a target for therapy development."

Antibodies are an attractive alternative for therapy because
they are easier to make in a functional form than a complex
protein like angiostatin, the researchers said. There may be
non-protein small molecules that mimic angiostatin, but unlike
angiostatin or the antibodies, they could get into a cell and
shut down the vital mitochondrial ATP-synthase as well as the
growth-controlling surface enzyme.

The researchers created antibodies for their laboratory
experiments by exposing rabbits to the subunits of recombinant
human ATP-synthase (alpha and beta) that angiostatin binds.
Because the rabbit's immune system perceives the enzyme
subunits as foreign, the animal develops antibodies that
recognize and bind to those specific proteins. "Recombinant"
means that the human enzyme was produced by using E. coli
bacteria as an ATP-synthase factory.

To prove that angiostatin targets ATP-synthase's normal
function, Moser developed a laboratory test that measures the
enzyme's rate of ATP production, in collaboration with
co-author Dennis Cheek of the Duke University School of
Nursing. Moser used this test to measure the effects on
ATP-synthase of angiostatin, the alpha and beta-subunit
antibodies and control antibodies.

Even without isolating the most potent rabbit antibodies,
the researchers found that purified serum from the exposed
rabbits reduced the activity of endothelial cells' surface
ATP-synthase by roughly 60 percent, while angiostatin reduced
ATP production by about 81 percent. The "polyclonal" rabbit
antibodies also slowed growth of cultured endothelial cells up
to twice as much as angiostatin. Serum from rabbits not exposed
to the ATP-synthase did not affect cell growth rates or ATP
production.

The scientists added that studies in animal models of blood
vessel growth support their laboratory findings, but those
results have not yet been published.

Other authors on the study are co-first author Daniel Kenan,
Timothy Ashley, Julie Roy, Michael Goodman and Uma Misra of the
department of pathology at Duke.

Duke University will hold the rights to these discoveries
through a pending patent application.