|
A
glance at most of the scientific, financial and general
media will tell us that the human genome has been
sequenced; we are now in the 'post-genomic' era. What
does this mean for the pharmaceutical industry and
how can this landmark event be exploited in terms
of drug discovery?
The
proteome: a whole host of proteins
The so-called central dogma of biology tells
us that DNA codes for RNA, which in turn codes for
proteins. In an oversimplified version of this process,
one gene encodes for one protein and therefore the
human genome defines the complement of all the expressed
proteins of an individual - or the proteome.
Not surprisingly, the generation of a protein from
a gene sequence, as well as its eventual function,
are subject to complex processes. Primarily, variations
in the way a gene is spliced after transcription can
generate different proteins. Subsequently, post-translational
modifications (changes that occur after a protein
has been produced from its mRNA), such as glycosylation
and phosphorylation, are essential for the correct
function of the protein. Furthermore, proteins interact
with each other (and other macromolecules) in complex
ways.
In a disease state, each of these elements is subject
to change. These changes can influence the expression
of other genes and the characteristics of other proteins.
Most importantly, this cannot be predicted from the
human genome.
It is the study of these processes to which proteomics
addresses itself - the exploitation of the central
dogma.
Proteomics: a range of technologies (and challenges)
In order to have relevance to drug discovery, proteomics
must be capable of analysing differences in several
protein characteristics induced by different disease
states:
- Presence
-
Relative abundance
-
Modifications
-
Protein-protein interactions
Proteomics comprises a group of evolving technologies
designed to identify these differences as they occur
between normal and disease states:
-
2D gel electrophoresis
-
Mass spectrometry
-
Bioinformatics
-
Protein "chips"
The
challenge presented by the application of these technologies
cannot be understated. The task is considered by some
to be more difficult than the sequencing of
the human genome.
Approximately 30,000 genes have been identified in
the human genome. Salomon Z Langer, VP of Molecular
Biology and Drug Discovery at Compugen told IMS HEALTH's
R&Dfocus
in an interview that a gene will produce an average
of four splice variants.
Therefore, there could be 120,000 or more unique
proteins in the human proteome. This, coupled with
differences in post-translational modifications and
protein interactions, has led Trevor Hawkins, Director
of the US Department of Energy's Joint Genome Institute,
to speculate that the human proteome may prove to
be ten times more complex than the genome.
One of the first challenges for proteomics is to establish
routine, reliable and efficient technologies for the
acquisition and analysis of data. To fulfil these
criteria, the technologies need to facilitate consistent
sample preparation, automation, and assimilation of
the information generated. Essentially, reproducible
high-throughput technologies are required.
The protagonists: collaborate or die
In the light of the complexity of the task, it is
no surprise that nearly all proteomics projects involve
collaborations. Table 1 details the main combined
projects currently being undertaken.
|
Table
1: Proteomics consortia
|
Partners
|
Value
|
Nature
of Collaboration
|
| Myriad,
Hitachi, Oracle |
$185
million |
To
map entire proteome |
| Applera
Corp*, Washington Univ., Geneva Univ. |
|
To
generate and evolve methodologies |
| Proteome
Consortium (Univ. of Michigan) |
$12
million |
Protein
identification |
| BioMerieux-Pierre
Fabre, CNRS, various French universities
|
|
Using
proteomics to discover cancer drug targets
|
| Centre
for Proteome Analysis, Denmark |
|
Method
development, therapeutic and diagnostic
design |
| Celera
Genomics*, Compaq, Sandia Natl Lab. |
$10
million |
To
build proteomics computer facility |
|
(*part of PE Biosystems)
As can be seen from Table 1, the collaborations are
being set up to address different aspects of the technological
challenges. The Myriad/Hitachi/Oracle agreement combines
the proteomics capabilities of Myriad, the electronics
expertise of Hitachi and the software know-how of
Oracle. It was announced that the alliance intends
to map the human proteome by 2004 and assemble a database
of the information.
The Applera-Academic axis aims to establish the Proteomics
Research Centre, an institute to develop methodologies
in proteomics and make them commercially available.
The centre is to use ICAT (Isotope Coded Affinity
Tags) technology, a platform for the quantification
of proteins in complex mixtures developed by the University
of Washington, with additional expertise from the
University of Geneva.
Large pharmaceutical companies and others are also
teaming up with biotechnology companies providing
proteomics tools (see Table 2).
|
Table
2: Collaborations implementing proprietary proteomics
technologies
|
Partners
|
Indication
|
Date
Initiated
|
| Pfizer,
Oxford GlycoSciences |
Atherosclerosis,
Alzheimer's disease |
1999
|
| Medarex,
OGS |
Cancer
and others |
2000
|
| Bayer,
OGS |
Respiratory
disease |
2000
|
| Merck
& Co., OGS |
Diabetes
|
1999
|
| OGS,
Oxford Univ |
Rheumatoid
arthritis |
1997
|
| Novartis,
Cubist |
Infection
|
1999
|
| Novartis,
Myriad |
Cardiovascular
disease |
1995
|
| Proteome
Sciences, Geneva Univ., Buckingham Univ.
|
Type
II diabetes |
1999
|
| Curagen,
Biogen, Genentech |
|
1997
|
| Curagen,
Gemini Genomics |
|
2000
|
| Curagen,
Cor Therapeutics |
Cardiovascular
disease |
1999
|
| M-phasys,
Graffinity |
G
protein coupled receptors |
2001
|
|
Source:
R&Dfocus
Another approach is to limit research to diseases
caused by specific organisms. A proprietary
proteomics database has been launched by Hybrigenics,
comprising information on all known protein-protein
interactions between HIV-1 and human lymphocytes.
Using the same protein interaction map technology,
the company is also investigating such interactions
in other organisms, including hepatitis C virus and
Helicobacter pylori.
HUPO: the next big hope?
In parallel with commercial projects, an organization
has been established hoping to initiate for the proteome
what the Human Genome Mapping Project did for the
genome. The Human Proteome Organization (HUPO)
is an international consortium of academic and industrial
partners, established in February 2001. South Korea
was the only government to commit itself to the project
as of April 2001.
HUPO held its first annual meeting in April 2001,
and announced that it had access to approximately
$1 billion of collective funding. It aims to identify
every expressed protein and construct recombinant
versions of these within the next 5-10 years. HUPO
is intent on establishing a publicly funded Human
Proteome Project.
On a promise
At present, no therapies on the market have been discovered
using proteomics technologies. The promise of the
new technology, however, lies in its ability to identify
the targets of drugs, as opposed to genes - which
are not the entities drugs traditionally act upon.
It is hoped that the new technology will identify
differentially expressed or modified proteins as new
drug targets.
Proteomics is in many respects the next logical step
in a chain of events driven by the sequencing of the
human genome. It remains unclear whether efforts such
as the Human Proteome Project, which rely on the sequencing
of all expressed proteins rather than addressing the
dynamic nature of protein function, can yield tangible
pharmaceuticals.
|
