Researchers at Johns Hopkins took advantage of a new
technique that reads the makeup of proteins to identify
nearly all chemical changes nature makes by adding
phosphate to proteins manufactured in human cells.
The Johns Hopkins team then added its list of these
so-called phosphorylation events to lists compiled by
others and created a database publicly available on the Web
— PhosphoMotif Finder at
www.hprd.org/PhosphoMotif_finder — to help speed
the work of researchers around the world.
"Finding so many at one time is a huge advance," said
Akhilesh Pandey, an associate professor at the
McKusick-Nathans Institute of Genetic Medicine at Johns
Hopkins. "Phosphorylation is essential for controlling
chemical reactions in our cells' protein factories, and
phosphorylation gone awry has been implicated in several
diseases. The ability to study more than one
phosphorylation at a time will help us understand some of
these diseases — including cancers — sooner.
"What we have here is about 20 years' worth of lots of
work in one searchable list," Pandey said. A report on all
the newly identified protein alterations is published in
the Feb. 13 issue of the Proceedings of the National
Academy of Sciences, and a report on the database appears
in the March issue of Nature Biotechnology.
Pandey's team used electron transfer dissociation
tandem mass spectrometry, a technology that breaks apart
proteins into small fragments, separates them by size and
identifies the fragments based on their mass (their size
and weight). The process improves on previous techniques by
breaking up proteins more gently and keeping chemical
modifications like phosphorylation intact. Previous
spectrometry methods were "just too rough" on the delicate
protein alterations and sheared them right off, Pandey
said. "We had to guess where they might be, and nobody
wants to chase false leads based on wrong guesses."
Pandey says the original goal of the research was to
identify accurately as many protein changes as possible
using the new technology.
"But to see how well we measured up, we had to compare
our findings to what already was published, and there was
just no clean, easy way of doing that because there were
reports all over the place," he said. "That's when we
decided to go through and consolidate just about everything
on phosphorylation that was out there."
Working with human kidney cells, the researchers
fished out the thousands of different proteins and analyzed
them by electron transfer dissociation, resulting in a net
total of 1,435 phosphorylations. Comparing these 1,435 to
the 20 years' of published data, they discovered that about
80 percent of what they found had never been reported.
The team then constructed an online search tool,
PhosphoMotif Finder, which was incorporated into their
previously established Human Protein Reference Database.
Human Protein Reference Database now contains about 16,000
phosphorylation sites described in the literature, and the
PhosphoMotif Finder tool allows any researcher to find
potential phosphorylation sites in any protein of
interest.
"The power of this technique is not just in the
numbers," Pandey said. "Rather, we've found what you might
call new information about old proteins, and we hope the
new data will help researchers study their favorite
proteins in greater depth. After all, there's no sense in
reinventing the wheel."
Pandey and his team now are curious about other
chemical modifications of proteins, which are the "business
end" products of our genes.
"There is evidence of other, more fragile
modifications that until now no one has been able to get a
handle on because they're way too hard to work with. Now we
have the tools to probe further," he said.
The research was funded by the U.S. Public Health
Service; Canadian Institute of Health Research; National
Institutes of Health; National Heart, Lung and Blood
Institute; and National Multiple Sclerosis Society.
Authors on the Proceedings of the National Academy of
Sciences paper are Henrik Molina and Pandey, of Johns
Hopkins; Suresh Mathivanan, of the Institute of
Bioinformatics in Bangalore, India; and David Horn and Ning
Tang of Agilent Technologies in Santa Clara, Calif. Authors
on the Nature Biotechnology paper are Ramars Amanchy,
Balamurugan Periaswamy, Raghunath Reddy and Pandey, all of
Johns Hopkins; and Mathivanan and Sudhir Gopal Tattikota,
of the Institute of Bioinformatics in Bangalore, India.