Johns Hopkins
Kimmel
Cancer Center scientists have completed the first draft
of the genetic code for breast and colon cancers. Their
report, published online in the Sept. 7 issue of Science
Express, identifies close to 200 mutated genes, now
linked to these cancers, most of which were not previously
recognized as associated with tumor initiation, growth,
spread or control.
"Just as sequencing the human genome laid the
groundwork for subsequent research in genetics, these data
lay the foundation for decades of research on colon and
breast cancers," said Victor Velculescu, assistant
professor of oncology at the Johns Hopkins Kimmel Cancer
Center.
Although gene discoveries by independent scientists
scattered around the world have provided clues, Velculescu
says that relatively few genes have been shown to be
altered in cancers. The Johns Hopkins gene hunters say the
number of genes that were altered in breast and colorectal
cancer genomes surprised them. "We expected to find a
handful of genes, not 200," said Tobias Sjoblom, a lead
author and postdoctoral fellow at the Kimmel Cancer
Center.
Despite the potential rewards envisioned by cancer
biologists, efforts to map cancer genes have drawn
criticism from other scientists who say that funding
dollars should be spent on projects yielding more immediate
benefits for detection and treatment.
Kenneth Kinzler, professor of oncology and co-director
of the Ludwig Center at Johns Hopkins, said, "These are
good debates to have, [but] we are convinced that this kind
of study will provide one of the best road maps possible
for beating cancer. Who would pass up the opportunity to
read the enemy's game plan?"
Some gene alterations already have led to successful
detection and treatment strategies. These include the
breast cancer drug Herceptin, which targets a breast cancer
cell receptor made by the Her2/neu gene, and blood tests
for hereditary colon cancer, based on the APC gene and
others identified by the Johns Hopkins group.
Bert Vogelstein, an investigator at the Howard Hughes
Medical Institute and co-director of the Ludwig Center at
Johns Hopkins, said, "Cancer scientists recognize that
merely identifying pieces of DNA that have a role in the
disease is a beginning, not an end to our work. But by
using a more systematic method to identify genes that play
an essential role in cancer, we will be able to guide that
work."
The Johns Hopkins team began its project with 11
samples each from breast and colon cancers removed from
patients during surgery.
Within each tumor cell, billions of individual
chemicals called nucleotides pair together in a
preprogrammed fashion to build the rungs of a DNA ladder
that compose genetic instructions. Changes called mutations
in the nucleotides can create coding errors that transform
a normal cell into a cancerous one.
To locate the altered nucleotides, the scientists
compared the genetic code of their tumor samples with
normal ones. First, they used the Human Genome Project to
identify the sequences of best-known genes — more
than 13,000 in all — roughly two-thirds of the total
number of genes identified by the HGP. The actual number of
human genes is still in dispute but is estimated to be
about 20,000.
Then, in each tumor, the scientists examined the DNA
code of these 13,000 genes by dividing each gene into
overlapping sections, about 10 per gene, to get 130,000
sections for analysis. Each segment was amplified through a
process called polymerase chain reaction, purified and its
sequence determined using more than 3 million biochemical
reactions. The sequences were fed through computer software
that matches up normal sequences with those from tumor
samples. The software highlighted more than 800,000
suspicious regions that were visually inspected, one by
one, to verify that they were true mutations that altered
protein code rather than normal variations or minor changes
with no effect on the gene product.
In total, the Johns Hopkins team combed through 465
million nucleotides — several encyclopedias' worth of
letters — to find approximately 1,500 DNA nucleotides
that differed from the normal code in important ways.
Virtually all these mistakes were mere single nucleotide
"typos." Some 200 genes were significantly mutated; the
mutated genes in breast and colon cancers were almost
completely distinct, suggesting very different pathways for
the development of each of these cancer types.
Kinzler said, "This gives us some understanding of why
breast and colon cancers, and most likely other cancers as
well, are very different diseases and develop through
different processes. When we say this will drive cancer
research for the next couple of decades, this is one of the
reasons. Now researchers will study how these mutations
occur in breast and colon cancers, perhaps searching for
environmental agents or cellular processes that drive these
changes."
The Johns Hopkins team also found that the average
number of mutant genes in each cancer is about 100, and at
least 20 of them are likely to be crucial for tumor
formation. "Each cancer has a different blueprint. No two
patients are identical," Velculescu said.
Other cancers also can be evaluated using the Johns
Hopkins approach, which the researchers say has been
developed over the past two decades and made possible
through recent advances in DNA sequencing and
bioinformatics.
"These findings," Vogelstein said, "will guide and
provide support for future comprehensive genetic studies
including those envisioned by the Cancer Genome Atlas
Project." Future research will include performing similar
analyses on other tumor types, charting the pathways
through which each mutant gene acts and looking for common
mutations that can be targeted with cancer drugs or used to
detect the disease earlier.
This research was supported by the Virginia and D.K.
Ludwig Fund for Cancer Research, National Institutes of
Health, Department of Defense, Pew Charitable Trusts,
Palmetto Health Foundation, Maryland Cigarette Restitution
Fund, State of Ohio Biomedical Research and Technology
Transfer Commission, Clayton Fund, Blaustein Foundation,
National Colorectal Cancer Research Alliance, Strang Cancer
Prevention Center, Avon Foundation, Flight Attendant
Medical Research Institute and V Foundation for Cancer
Research.
The Johns Hopkins research team also includes Sian
Jones, Laura D. Wood, D. Williams Parsons, Jimmy Lin,
Thomas Barber, Diana Mandelker, Rebecca J. Leary, Janine
Ptak, Natalie Silliman, Steve Szabo, Giovanni Parmigiani,
Ben Ho Park and Nickolas Papadopoulos. Other authors
include Phillip Buckhaults, Christopher Farrell and Paul
Meeh, University of South Carolina; Sanford D. Markowitz,
Case Western Reserve University, University Hospitals of
Cleveland and Howard Hughes Medical Institute; Joseph
Willis and Dawn Dawson, Case Western Reserve University and
University Hospitals of Cleveland; James K.V. Willson and
Adi F. Gazdar, University of Texas Southwestern Medical
Center; James Hartigan, Agencourt Bioscience Corp.; Leo Wu
and Changsheng Liu, SoftGenetics; and Kurtis E. Bachman,
University of Maryland Greenebaum Cancer Cancer.