By Robert Sanders, Media Relations | 17 February 2009

BERKELEY — Researchers from the University of California, Berkeley,
and the University of Iowa have turned a relatively benign virus into
a highly infectious form that is ideal as a carrier for gene therapy.

In its first gene therapy test, it completely cured human cystic
fibrosis lung tissue in culture.

UC Berkeley researchers forced the adeno-associated virus to evolve
so as to bind to sialic acid on the surface of lung cells, making it
easier for the virus to infect them. The forced mutaion (red)
surrounds the mouth (green) of the receptor that binds sialic acid.
(Schaffer lab/UC Berkeley)This success with the benign adeno-
associated virus (AAV), published this week in the online early
edition of the journal Proceedings of the National Academy of
Sciences, overcomes a major problem of earlier virus-based gene
therapy for cystic fibrosis, and sets the stage for tests in advanced
animal models of the disease.

"I think it is worthwhile thinking about clinical therapy at the
levels of infection we are achieving," said coauthor David Schaffer,
professor of chemical engineering at UC Berkeley.

A new pig model of cystic fibrosis developed last year by Schaffer's
colleague, pulmonologist Joseph Zabner of the University of Iowa
Hospitals and Clinics in Iowa City, will provide a key test of the
virus as a carrier of a gene to replace the mutated gene responsible
for the disease.

"If we are able to show that efficient gene transfer can result in
gene therapy, if we can cure the lung disease of pigs that have been
genetically engineered to have cystic fibrosis lung disease, we
should have a real chance of curing cystic fibrosis in humans,"
Zabner said in an e-mail.

Schaffer's lab is collaborating with groups elsewhere to adapt the
virus to gene therapy for other diseases, including Alzheimer's
disease and amyotrophic lateral sclerosis (Lou Gehrig's disease).

"Both of those are situations where improvements in the properties of
the vehicle can have a significant impact on the success of the
therapy,"
Schaffer said.

These lung cells from a cystic fibrosis patient have been infected
with the evolved virus carrying a correct copy of the CFTR gene. The
cells with a green interior are expressing high levels of normal CFTR
protein, which is a chloride ion transporter that is defective in CF
patients because of a mutation in the CFTR gene. The chloride ion
transport of these cells was completely repaired with the virus.
(Schaffer lab/UC Berkeley)Cystic fibrosis (CF) is a common hereditary
disease that affects the body's mucus membranes, in particular the
lungs, resulting in difficulty breathing and typically in death
before the age of 40 from lung or organ failure. One in 4,000
children in the United States is born with CF.

Schaffer and his UC Berkeley colleagues collaborated with Zabner's
laboratory to test a technique developed by Schaffer to force the
evolution of a virus in ways that make it more effective in gene
therapy.

Two years ago, Schaffer and colleagues used the technique to create a
variant of AAV that more easily avoids the immune system, allowing
the virus to remain in the body long enough to deliver a gene to its
intended target.

Using the same technique, the team produced a variant of AAV that is
several hundred times more effective at entering lung cells than the
natural virus.

Schaffer's technique involves making many mutations in AAV, culturing
these variants with cells, and then taking those with specific
improved properties
- in this case, the ability to infect lung cells - and repeating the
process.

"We probably conducted about six rounds of evolution in which we
infected the lung epithelial cultures in Iowa, they sent it back to
us, we recovered the viral sequences, made new viruses and sent them
back again," Schaffer said. "It was iterative rounds of infection and
selection for improved infection that finally led to this enhancement
of function."

The main problem in CF is a mutation in the CFTR (cystic fibrosis
transmembrane conductance regulator) gene that results in a defective
chloride ion channel in the body's cells. This, in turn, creates a
chloride ion imbalance in the cell, which interferes with water
transport in and out of the cell. In the lungs, this causes the mucus
that lines the lung surface to become thick and sticky. Breathing
becomes difficult if the mucus is not loosened, often by vigorous
pounding on the chest, and coughed out.
Respiratory infections are common, and lung failure often results.
The ion channel defect also affects digestion, leading to nutritional
deficiency.

According to Schaffer, previous attempts to deliver a normal CFTR
gene to lung cells by means of a virus failed either because the
immune system mopped up the virus before it had a chance to deliver
its cargo, as was the case with adenovirus; or because the virus was
inefficient at delivering the gene to cells, the case with adeno-
associated virus. Most respiratory viruses tend to have low infection
rates, apparently because they would otherwise quickly wipe out their
host, Schaffer said.

Schaffer's technique forced the normally benign AAV, which has
already infected over 90 percent of people without any harmful side
effects, through rounds of directed evolution to increase its
infectivity several hundred-fold.

"We devised a way to evolve viruses that are released from the
natural constraints of evolution and have the freedom or ability to
evolve toward properties that are more useful for medical
application," he said. "In human lung tissue, it completely rescued
the chloride ion transport properties of the cells after delivering
the correct copy of the CFTR gene to replace the mutated copy of the
gene that is present in cystic fibrosis patients."

In this case, the infectious AAV strain developed two major changes:
Thanks to a mutation on the viral surface, it was able to bind to
different receptors or bind to a more plentiful receptor on the cell
surface; and it also acquired a mutation that gave the virus an
enhanced ability to make it past the cell surface membrane, slip past
the lipid bilayer and reach the inside the cell.

"Neither change alone was enough; it had to be the combination of the
two that resulted in the improved properties," said Schaffer. "If we
decided to use rational design, we wouldn't have known this. So, we
left it up to evolution to discover the answer."

Coauthors with Schaffer and Zabner were James T. Koerber, a graduate
student with UC Berkeley's chemical engineering and bioengineering
departments and with the Helen Wills Neuroscience Institute; research
scientist Katherine J. D. A. Excoffon and David D. Dickey of the
University of Iowa; Matthew Murtha and Brian K. Kaspar of The
Research Institute at Nationwide Children's Hospital in Columbus,
Ohio; and Saf Keshavjee of Toronto General Hospital and the
University of Toronto.

The research was funded by the National Institutes of Health and the
Cystic Fibrosis Foundation.