Biology Reference
In-Depth Information
these few proteins has been hard to come by. This has
meant that systems biological models suffered from
a paucity of hard parameters, and instead usually had to
make do with very rough estimates of the identities,
abundances, localization and modification states of the
involved proteins. Modern MS-based proteomics is now
ready to change this situation completely.
Its success in protein analysis comes as the last chapter
in the very long history of mass spectrometry, which began
with the observation of Kanalstrahlen (anode rays) by
Eugen Goldstein in 1886 and the construction of the first
mass spectrometer by Francis William Aston in 1919. The
first application to amino acids by Carl-Ove Andersson
dates back to 1958. Later, both the quadrupole and the
three-dimensional ion traps were developed by Wolfgang
Paul, for which in 1989 he received the Nobel Prize in
Physics, together with Hans Georg Dehmelt. However, the
breakthrough for MS in biology came with the develop-
ment of soft ionization technologies that enabled gentle
transfer of peptides or proteins into the mass spectrometer,
for which the Nobel Prize in Chemistry in 2002 was
awarded. The emergence of MS as a powerful 'omics'
discipline was also enabled by continuous developments in
sample preparation, separation technologies and break-
throughs in the capabilities of the mass spectrometers
themselves, some of which are detailed below. In parallel
with these improvements on the 'wet side', data analysis
and computational strategies on the 'in silico side' over the
last 20 years have been just as important, as they allow the
identification of peptides in sequence databases from
a minimum of mass and fragmentation information. Orig-
inally applied to one peptide at a time in a manual fashion,
these algorithms now deal with hundreds of thousands of
peptides in multifaceted projects and require large-scale
data management issues to be addressed which are just as
demanding as they are in the other 'omics' technologies.
The development of relative and absolute quantification
methods over the last decade has been particularly crucial
to proteomics. Using the latest proteomics technologies, it
is now possible to quantify essentially complete proteomes
of model organisms such as yeast
[2]
. More complex
organisms are also coming within reach
[3
individual proteins in single cells, for instance by means of
immunostaining
[10]
or protein tagging
[11]
.
One of the most important areas for MS-based proteo-
mics is the analysis of post-translational modifications
(PTMs)
[12,13]
. During recent years, MS-based proteomics
has revealed an unexpected diversity and extent of protein
modifications. For example, phosphorylation turns out to
occur not only on a few key proteins but on thousands of
them, which possibly also applies to less studied PTMs.
How to model their regulatory roles will long be a key
challenge for systems biology.
MS-based proteomics now for the first time opens up
the entire universe of cellular proteins to detailed study.
Protein amount, localization, modification state, turnover
and interactions can all be measured with increasing
precision and increasingly sophisticated approaches, as
detailed below. There is a unique opportunity to employ
these data as a crucial underpinning for building accurate
and comprehensive models of the cell
[14]
.
MS-BASED PROTEOMICS WORKFLOW
The analysis of complex protein mixtures is very difficult.
Accordingly, the field of MS-based proteomics has been
made possible by seminal advances in technology that have
helped to overcome a number of critical challenges.
Together, they have resulted in a generic and general
'shotgun' workflow that can be applied to any source of
proteins and almost any problem that can be addressed by
MS-based proteomics (
Figure 1.1
). Here we explain the
principles of this workflow, but also point out variations to
the general theme.
Until the 1980s proteins or peptides were largely
incompatible with MS, as they could not be transferred into
the vacuum of the mass spectrometer without being
destroyed. Two alternative approaches solved this funda-
mental problem: electrospray ionization (ESI), for which
a share of the 2002 Nobel Prize in Chemistry was awarded
to John B. Fenn, and matrix-assisted laser desorption/
ionization (MALDI). MALDI involves embedding the
analyte in a solid matrix of an organic compound, followed
by transfer into the vacuum system. A laser pulse then
excites the matrix molecules, leading to their desorption
along with the ionized analyte molecules, whose mass is
measured in a time-of-flight (TOF) analyzer
[15]
.In
contrast, in electrospray a stream of liquid is dispersed into
a charged aerosol when high voltage is applied to the
emitter. Solvent molecules in aerosol droplets rapidly
evaporate, and charged analyte molecules are then trans-
ferred into the vacuum of the mass spectrometer, where
they finally arrive as 'naked' ions
[16]
.
Even with appropriate ionization techniques at hand,
large intact proteins are usually difficult to handle, there-
fore the standard MS-based proteomic workflow follows
5]
. However,
quantitative proteomics not only permits precise proteome
quantification in one state compared to another (termed
'expression proteomics' and providing data conceptually
similar to transcriptomics) but also enables 'functional
proteomics', when combined with appropriate biochemical
workflows. This can, for example, identify specific protein
interactions or
e
reveal
the composition of subcellular
structures
[6
8]
. Together, these methods allow the pro-
teome to be studied in space and time, something that
cannot easily be done on a large scale and in an unbiased
manner by other technologies
[9]
. The resulting proteomic
data perfectly complement large-scale studies following
e
