Journal of Chromatography B, 849 (2007) 1–31

Review

Methods for samples preparation in proteomic research夽

Anna Bodzon-Kulakowska a , Anna Bierczynska-Krzysik a , Tomasz Dylag a , Anna Drabik a,b ,
Piotr Suder a,c , Marek Noga a , Justyna Jarzebinska a , Jerzy Silberring a,d,∗
a Department of Neurobiochemistry, Faculty of Chemistry, Jagiellonian University, Ingardena St. 3, 30-060 Krakow, Poland
b Institute of Medical Biochemistry, Medical School, Jagiellonian University, Krakow, Poland
c Regional Laboratory, Jagiellonian University, Krakow, Poland
d Centre for Polymer Chemistry, Polish Academy of Sciences, Zabrze, Poland

Received 12 June 2006; accepted 23 October 2006

Available online 20 November 2006

Abstract
Sample preparation is one of the most crucial processes in proteomics research. The results of the experiment depend on the condition of the
starting material. Therefore, the proper experimental model and careful sample preparation is vital to obtain significant and trustworthy results,
particularly in comparative proteomics, where we are usually looking for minor differences between experimental-, and control samples. In this
review we discuss problems associated with general strategies of samples preparation, and experimental demands for these processes.
© 2006 Elsevier B.V. All rights reserved.

Keywords: Sample; Proteomics; Analysis; Proteins; Peptides; Identification; Sequence

Contents

1. The source of samples in proteomic research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.1. Animal models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2. Animal tissue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3. Cell cultures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3.1. Between tissue and cell culture—laser capture microdissection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.4. Body fluids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.4.1. Blood, serum and plasma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.4.2. Cerebrospinal fluid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4.3. Saliva . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4.4. Synovial fluid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.5. Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.6. Bacterial samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.7. Viral samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2. Samples preparation—a general strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1. Methods of cell disruption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1.1. Mechanical homogenization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1.2. Ultrasonic homogenization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1.3. Pressure homogenization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

夽 This paper is part of a special volume entitled “Analytical Tools for Proteomics”, guest edited by Erich Heftmann.
∗ Corresponding author at: Department of Neurobiochemistry, Faculty of Chemistry, Jagiellonian University, Ingardena St. 3, 30-060 Krakow, Poland.
Tel.: +48 12 2927949; fax: +48 12 6340515.
E-mail address: [email protected] (J. Silberring).

1570-0232/$ – see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jchromb.2006.10.040
2 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31

2.1.4. Freeze–thaw homogenization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

2.1.5. Osmotic and detergent lysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2. Protein solubilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.1. Chaotropes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2.2. Detergents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2.3. Reductants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2.4. Protection from proteolysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.3. Removal of contaminants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3.1. Salts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3.2. Detergents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.3.3. Abundant proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.3.4. Lipids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.3.5. Polysaccharides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.3.6. Nucleic acids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.3.7. Other substances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.4. Protein enrichment methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.4.1. Precipitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.4.2. Centrifugation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.4.3. Electrophoretic methods of protein enrichment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.4.4. Membrane proteins enrichment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.4.5. Prefractionation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.4.6. Chromatographic techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.4.7. Solid-phase protein enrichment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3. Samples preparation guidelines for various proteomic techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.1. Electrophoretic methods of protein separation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.2. Capillary electrophoresis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.3. Sample preparation for high-performance liquid chromatography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.3.1. Top–down proteomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.3.2. Bottom-up proteomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.4. Mass spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.4.1. MALDI-MS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.4.2. SELDI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.4.3. Electrospray ionization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.5. Quantitative proteomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.6. Imaging mass spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4. Conclusions and future prospects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

1. The source of samples in proteomic research of protein abundance. In a single cell there could be only 10
copies of transcription factor at the bottom of this range, but at
Proteomics examines all proteins expressed in a cell, tissue or the other end we may expect up to 1,000,000 copies of a more
organism. Proteins carry out different functions and are respon- abundant protein. To deal with this problem, the most abundant
sible for maintaining homeostasis in organisms. Changes in their proteins could be removed or the complexity of the entire sample
composition could lead to pathological processes; so there is an could be reduced. Several methods of samples fractionation and
extensive interest in applying proteomics to the identification of techniques of proteins enrichment could be used to achieve this
disease markers. Tissues, cell lines, primary cell cultures and goal [2].
body fluids such as plasma or cerebrospinal fluid are used as a It has to be remembered that protein content in contrast
source of proteins. Another branch of proteomics is devoted to to genome, which is stable and identical in all cells of one
plants, bacteria and viruses. organism (apart from germ cells) is not even similar in vari-
The results of any experiment are dependent on the condition ous cell types. In fact, this difference is responsible for such
of the starting material. Therefore, choosing the proper experi- great diversity of the cells. Apart from this, changes in protein
mental model and preparing the sample carefully is crucial for composition could also occur in response to different stimuli
obtaining significant and trustworthy results. Sample prepara- and in different timepoints and space (cellular compartments).
tion is a matter of great importance, especially in comparative Thus, the aim of the experiment and an appropriate model has
proteomics, where we are usually looking for minor differences to be carefully considered to obtain reliable results. Below,
between experimental and control samples [1]. strategies and methodologies for samples preparation used in
One of the major obstacles associated with analyzing such proteomics are reviewed and their advantages and drawbacks are
complex material as a biological sample is the dynamic range discussed.
 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31 3

1.1. Animal models material. As a result, a clear supernatant should be obtained,

followed by its division into aliquots and freezing [6].
Recent advances in medical sciences would not be possible There is also a possibility of performing some of the research
without animal research. Animal models for human diseases on human post-mortem tissue, but apart from the ethical consid-
are indispensable in understanding the background and biology erations, a question remains as to whether pathophysiological
of a disease and in finding out the methods of its treatment. mechanism could indeed be evaluated in such material. Here,
Mice and rats, for several reasons, are the models of choice. data such as age, gender, ethnicity, medical history, agonal state,
These animals have relatively short lifespan, which enables us post-mortem and post-autopsy intervals have to be taken into
to study the progression of a disease. Well-characterized strains consideration. In case of brain tissue, many proteins are remark-
of animals are available. Physiological processes that occur in ably stable post-mortem or undergo degradation only to a minor
humans are often (but not always!) similar to those in rodents. degree; thus valid and practical measures of some of the param-
Thus, studies on the pathophysiology of various diseases at the eters may be performed in human brain [7].
proteome level are possible due to a relatively high similarity Biopsy might be another source of the tissue for proteomics
between the rats/mice and human proteins. Transgenic animals analysis. It seems to reflect the state of living organism, and
seem to be another great promise for obtaining appropriate and sometimes, collected material could be cultured for further
useful models [3]. experiments. Such samples, usually obtained during surgery,
It has to be remembered that these models cannot be consid- have to be frozen in liquid nitrogen and must be stored at −80 ◦ C
ered as a complete equivalent of human disorders, as they present prior analysis [8].
only some of their aspects. We have to be aware that, apart from
similarities, some systems could be different (for example, the 1.3. Cell cultures
rat steroid system is different from the human one). Aspects of
gender, weight, feeding, etc., also need to be taken into consider- Simplification of the sample is one of the greatest advantages
ation before establishing a model for a particular experiment [4]. of cell culture serving as an experimental model, especially in
proteomics. Tissue samples are invariably heterogeneous and,
1.2. Animal tissue thus, more complex. It is assumed that about 1 × 104 proteins
are expressed in one cell. In a tissue composed of different types
When a particular animal model has been established, usually of the cells this number is much higher.
the tissue connected with the disease is chosen for detailed anal- In contrast, selective pressure of the culture conditions, after
ysis. Every tissue has its own characteristics. For example, lipids one or two passages, tends to produce a homogenous culture
are particularly abundant in the brain and have to be eliminated of the most vigorous cell type. It means that this model allows
together with nucleic acids, in order to obtain results of high for studying the behavior of a single type of cell in the absence
quality. The most common method used here is selective pre- of the complexity of the entire tissue. This may help to reveal
cipitation of the proteins with acetone and trichloroacetic acid changes in low abundance proteins, which could be impossible
(TCA) [1]. in the whole tissue study.
During tissue preparation for proteomic analysis, it is impor- Experiments involving cell cultures very often test the influ-
tant to diminish its heterogeneity, as much as it is possible. The ence of potential medicines or toxic substances. Owing to the
sample should be pure and relevant. For example, in case of can- homogeneity of the culture cells in each dish, they are virtually
cer proteome analysis, it should be free of stroma, blood, serum, identical; therefore, examination of the influences and compar-
etc., and whenever possible, should represent only tumor cells ison with the control could be highly relevant and trustworthy.
[5]. When a specific part of the whole organ is isolated (for Reagents, to which cells are exposed, could be administered
example, striatum from the brain), it is important to preserve its directly at a defined concentration, which is almost impossible
regional and cellular specificity and not extract too much of the during in vivo experiments. This technique also ensures for sig-
surrounding tissue [1]. nificant control of the environmental conditions [9].
Fresh tissue should be freed from connective tissue and fat Primary cell culture is derived either from enzymatic or
and preferably perfused with an ice cold saline prior to excision mechanical dispersal of the tissue, or by outgrowth of migrating
or at least rinsed right after it. It should be well minced with cells from a tissue fragment. Cells capable of proliferation under
surgical scissors in the freshly prepared lysis buffer containing particular conditions, after appropriate time, reach the conflu-
chaotropes, detergents, reductants and protease inhibitors. If a ence and form a monolayer. At this stage the culture shows the
tissue contains a large proportion of connective tissue, it should closest morphological resemblance to the parent tissue. A pri-
be frozen in liquid nitrogen, ground to fine powder in a mortar mary cell culture becomes a cell line after the first passage.
and placed in the lysis buffer. Continuous cell line may be obtained after transformation of
Such prepared tissue should be placed in ground glass tissue normal cell line either spontaneously or by chemical or viral
grinder and homogenized until a uniform homogenate, without induction. Those cells are usually aneuploid and could have
any visible tissue particles is formed. Then the sample should be unlimited culture lifespan. A number of the properties of contin-
left for some time at room temperature and in darkness to allow uous cell lines, such as reduced serum requirement and reduced
each constituent of the sample to solubilize. The homogenate density limitation of growth, are associated with malignant trans-
should then be centrifuged to remove nucleic acids and insoluble formations [10].
4 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31

Preparation of cell culture for 2D gel electrophoresis (2-DE), only the most abundant cellular proteins were detected with this
one of the major methods of proteins separation in proteomic, is method; so its sensitivity needs further improvement [1].
not a trivial task. Witzmann et al. [11], using primary hepatocyte
cell culture, showed that the recovery of the cells from mono- 1.4. Body fluids
layer cell culture by scraping, washing and centrifugal pelleting
of the cells, followed by solubilization, results in the introduction Body fluids are an important source providing vital infor-
of significant variability between the samples. Proteins localized mation on the function of living organisms. For example, cere-
in cytosolic, cytoskeletal or external compartments lost over half brospinal fluid (CSF), being in a direct contact with nervous
of their abundance during this procedure. tissue, has long been considered as reflecting dynamic changes
According to their research, direct solubilization of cells in in the Central Nervous System (CNS). Body fluids such as blood,
the cell culture dish in lysis buffer, after removal of medium, is a CSF and saliva are relatively easily available; thus they are com-
best way of preparing the sample for proteomics analysis. Dur- monly used for clinical diagnosis. In particular, the use of saliva
ing this procedure, lysis buffer should be added directly to the does not raise any serious ethical questions and can be taken even
cell culture dish and left in the incubator for 1 h with intermittent by a non-trained personnel for fast analysis of, e.g. narcotics.
manual agitation. After solubilization, the entire volume of liq- Moreover, sensitivity of the nowadays methods is so high that
uid should be placed in a tube and sonicated. Sonication should only a small amount of these samples is required for analysis.
be carried out every 15 min for 1 h and the obtained solution As they are a great potential source of diagnostic data, proper
should be stored at −80 ◦ C until analysis. preparation of such samples for analysis is important. The first
In samples prepared in this way, 2-DE gel analysis detected difficulty is associated with the broad dynamic range of compo-
and matched an average of 1388 proteins compared to an average nents present in body fluids [15]. One of the major challenges is
of 899 proteins in washed/scraped/pelleted cell sample. the reproducibility of the two-dimensional gel electrophoresis,
It has to be remembered that limited complexity, lack of which is still the main method of proteome analysis, as this
homeostatic regulation from nervous and endocrine system, procedure requires several steps including gel transfer, strict
loss of three-dimensional organization of the tissue and spe- temperature control, appropriate number of replicates, selec-
cific cell interaction characteristic of its histology may lead to tion of additives and an experienced operator. Such studies were
some differences in cell behavior between cultured cells and described by Terry and Desiderio [16], who demonstrated that
their counterparts in vivo. It means that discoveries done on this the analysis of human CSF by 2D gels can achieve a high level
kind of models demand further confirmation by referring back of within-sample and between-sample reproducibility.
to the original tissue.
1.4.1. Blood, serum and plasma
1.3.1. Between tissue and cell culture—laser capture A simple Medline search for “serum” and “proteomics”
microdissection reveals ca. 460 papers describing various approaches for the
Laser capture microdissection (LCM) is a technology that identification of blood proteins. Each strategy seems to be
permits the isolation of selected cells or groups of cells from individually developed or modified by a particular research
a thin tissue section mounted on the glass slide. We can say group.
that samples obtained by this method are somewhere between Preparation of the blood samples differs depending on a
tissue and cell culture. During this process a narrow infrared laser method chosen and a purpose of the analysis. Serum is a fraction
beam is shone through a heat-sensitive transparent polymer film of blood, obtained after clotting and centrifugation of the whole
(thermoplastic membrane), which contacts the tissue section. blood, without addition of anticoagulants. Plasma is a fraction
Laser causes local melting of the polymer and adhesion of the obtained after collection of blood using various coagulants. The
selected cells to it. Cells can then be removed from the section, major drawbacks of blood samples are its complexity and the
together with the polymer [12]. presence of a huge amount of major protein “contaminants” such
This method of reduction of the sample complexity is crucial as albumin, immunoglobulins and haptoglobin. These proteins
in the analysis of heterogeneous samples such as solid tumor comprise a major fraction of the blood and therefore, special
tissue. For the study of cancer, it is very important to isolate care should be taken to remove them before analysis [17].
malignant cells away from their surroundings, including nor- The complexity of blood and methodological problems, with
mal, inflammatory or reactive cells [13], and this method permits its analysis at a reproducible and reliable level, initiated a
obtaining homogenous populations of those cells. LCM allows HUPO (The Human Proteome Organisation) recommendation
for microscopic verification of the material and then, for selec- for preparation of such samples prior to proteome analysis.
tive transfer and recovery of the cells from histological tissue HUPO-initiated a pilot phase, which evaluated advantages and
sections with greater speed and precision than the manual dis- limitations of many depletion, fractionation and MS (mass spec-
section methods [14]. trometry) technology platforms. Reference specimens of human
LCM increases anatomical specificity of the sample, but with serum and EDTA (ethylene diamine tetraacetic acid), heparin
this method only a very small amount of the sample could be and citrate-anticoagulated plasma were compared in laborato-
obtained. One LCM experiment usually consists of 3000 laser ries. The panel recommends use of plasma instead of serum,
shots, which involves approximately 15,000 epithelial cells, with EDTA (or citrate) for anticoagulation. To improve reso-
while for 2-DE, 100,000 or more cells are essential. So far, lution, sensitivity and reproducibility of peptide identifications
 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31 5

and protein matches, a combination of depletion, fractionation removes low-abundance proteins, including cytokines. Detailed
and MS/MS (tandem mass spectrometry) technologies is rec- information about abundant protein depletion could be find in
ommended, with explicit criteria for evaluation of spectra, use Section 2.3.
of search algorithms and integration of homologous protein Application of derivatized cellulose for generating protein
matches [18]. profiles of human serum samples was demonstrated by Feuer-
As it was said before, abundant proteins present in a blood stein et al. [29]. The technique allows for high enrichment of
sample sometimes need to be removed prior further analy- sample without depletion of albumin and immunoglobulin, and
sis. The work described by Zolotarjova et al. [19] addresses sample elution prior to MS analysis.
some of the potential problems in depleting proteins in typical Wasinger et al. described preseparation of human plasma
biomarker studies. The authors conclude that significant differ- samples, performed with the aid of the membrane-based prepar-
ences were noted between the depletion techniques employed, ative electrophoresis technology platform [30], which served
and this should be considered based on the expectations set dur- for albumin depletion. Various anticoagulants were tested here,
ing experimental design. Some of those techniques are described including EDTA, citrate and heparin. Another strategy, for com-
below. prehensive profiling of human plasma and serum proteomes,
One of the strategies involves depletion of albumin on the termed as protein array pixelation, was described by Tang et al.
dye-based columns or removal of immunoglobulin G (IgG) on a The approach consists of three sequential high-resolution protein
protein A-immobilized column [20]. Another reasonable solu- prefractionation methods (major protein depletion, solution iso-
tion to the above mentioned problems might be retentate chro- electrofocusing and one dimensional electrophoresis (1-DE)),
matography and surface-enhanced laser desorption/ionisation followed by nanocapillary reversed phase (RP) tryptic peptide
(SELDI) concept. This method based on a MALDI-TOF separation prior to MS/MS analysis [31].
(Matrix-assisted laser desorption/ionization – time of flight) A multidimensional and non-denaturing proteome-
methodology applies chips with modified surfaces. Samples are separation procedure using microplate technology was also
bound to the chip surface and the unbound part of the biological presented [32]. In the first dimension, the sample under study
material is washed out. Using various surfaces, one can per- was separated into 96 fractions by size-exclusion chromatogra-
form a more selective analysis, thus substantially decreasing the phy (SEC). In the second dimension, the fractions of the first
complexity of samples [21,22]. dimension were transferred by the liquid-handling device to
A method that allows for the reduction of the protein con- 96 parallel anion exchange chromatography columns. In this
centration range within a complex mixture, such as neat serum, way, the proteins were conserved in their native states and
through the simultaneous dilution of high-abundance proteins were distributed in 2400 liquid fractions. The fractions were
and the concentration of low abundance ones in a single simple subjected to MALDI-MS, and their tryptic digests to both
step was given by Guerrier et al. [23]. This methodology utilizes MALDI- and LC–ESI-MS/MS. The method was applied to
solid-phase ligand libraries of large diversity. With a controlled separate normal human serum proteome. Within 255 fractions
sample-to-ligand ratio, it was possible to modulate the relative exhibiting the highest protein concentrations, 742 proteins were
concentration of proteins such that a large number of peptides or identified by LC–ESI-MS/MS peptide sequence tags.
proteins that are normally not detectable by classical analytical Peptidomics in blood is another subject of special inter-
methods were found. est, as in neuroscience, where particular peptides/neuropeptides
Recent developments suggest several other approaches in are measured predominantly by RIA (Radio Immuno Assay)
serum analysis, including antibody-based microarrays and other or ELISA (Enzyme-Linked ImmunoSorbent Assay). There are
affinity-based agents such as aptamers [24]. A similar approach also several strategies that allow for such studies. One of them
has been applied by Ahmed et al. for tracking of serum pro- is peptidomics platform, coupling magnetic-based, automated
teins isoforms as biomarkers of ovarian cancer [25]. Serum solid-phase extraction of small peptides with a high-resolution
samples were preseparated on the Affigel Blue, prior to the MALDI-TOF mass spectrometric readout [33,34]. Another way
IEF (isoelectric focusing) separation. The combination of step- of profiling of blood peptides was described utilizing presepa-
wise IgG and albumin depletion by affinity chromatography ration with various physical methods, followed by Differential
and ultrahigh-efficiency capillary liquid chromatography sep- Peptide Display strategy for a semi-quantitative peptides, profil-
arations coupled to ion trap-tandem mass spectrometry enabled ing [35]. A “reversed” strategy was applied by Lowenthal et al.
identification of 2392 proteins from a single plasma sample [36] to investigate blood peptides bound to albumin. First, albu-
with an estimated confidence level of >94% and an additional min was removed by a solid-phase affinity captured under native
2198 proteins with an estimated confidence level of 80% [26]. binding and washing conditions. Captured albumin-associated
The authors reported that more than 80% of the observed pro- proteins and peptides were separated by gel electrophoresis
teins demonstrated interactions with IgG and/or albumin. This and subjected to iterative MS sequencing by microcapillary
result is consistent with another report where the investigators reversed-phase tandem MS.
concluded that though serum depletion of highly abundant pro- Another aspect is analysis of cell components in the body flu-
teins significantly increased the number of proteins identified, ids. As this topic is beyond the scope of our review, we will only
both the degree of sample complexity and this depletion method mention an example given by Pasini et al. [37], who described
resulted in a non-selective loss of other proteins [27], and with the identification of the membrane and cytosolic proteome of red
that of Granger et al. [28], who found that albumin depletion blood cells. A total of 340 membrane proteins and 252 soluble
6 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31

proteins were identified, validated and categorized in terms of Bearing in mind the low abundance of many CSF compo-
subcellular localization, protein family and function. Splice iso- nents, preconcentration of the fluid is often necessary. Such work
forms of proteins were identified and polypeptides that migrated has already been reported in 1960s by Kaplan and Johnstone
with anomalously high or low apparent molecular weights could [45], where earlier papers on this topic are also cited.
be grouped into either ubiquitinylated, partially degraded and An interesting approach utilizing bottom–up proteomics and
ester-linked complexes. two-dimensional LC–MS/MS for the analysis of human ven-
tricular CSF was given by Wenner et al. [46]. The neat fluid
1.4.2. Cerebrospinal fluid samples withdrawn from neurologically normal elderly persons
The CSF shows a similar protein content as blood plasma. The were treated with trypsin, followed by C18 solid-phase extrac-
major difference is their lower concentration in CSF [38]. As the tion. Tryptic CSF peptides were separated by 2D-LC–MS/MS,
brain is in direct contact with CSF, the biochemical changes in and individual samples were compared to one another. Using
the nervous system might be reflected in the fluid, which implies this strategy, it was possible to identify 249 CSF proteins from
CSF analysis as a potential diagnostic tool. Sample preparation 10 subjects. Of these proteins, 38% were unique to individual
is a challenge, similar to the above described blood plasma and patients, whereas only 6% were common to all 10 subjects. The
begins already at the moment of collection, a procedure often results clearly suggest substantial subject-to-subject variability
affected by blood contamination [39]. in the CSF proteome. Another thorough analysis of individual
The difference between blood and CSF withdrawal is that human samples was performed by Finehout et al. [47]. The
the latter is done under aseptical conditions, most often without applied procedure included lumbar puncture, storage of CSF
participation of laboratory personnel, and under certain pressure at −70 ◦ C and ethanol precipitation of proteins from the 250 ␮l
(surgery, etc.). Therefore, the CSF sample might be out of a strict aliquots. The obtained 2D gels contained 600 identified spots
control (e.g. temperature and time of storage), thus gaining many representing 82 different proteins. Of these 82 proteins identi-
fluctuations in protein/peptide content. fied, 25 have not appeared in any previously published 2-DE
Numerous excellent reviews on CSF analysis were recently map of CSF and 11 have not been previously reported to exist
published. Yuan and Desiderio [40] compared several sample in CSF. This paper shows the potential of such approach, uti-
preparation methods and also discussed an improvement in lizing small CSF aliquots and simple preconcentration of the
confidence level for determining differential spots in compar- sample.
ative proteomics. Another work [41] describes in details the
procedure for CSF preparation and analysis, including sam- 1.4.3. Saliva
ple handling, separation, analysis and data interpretation. The Saliva is one of the most easily available human body fluid.
authors were able to identify more than 480 spots separated It may be easily, safely and non-invasively collected. As a pro-
on the 2-D gels using MALDI-TOF and ESI (electrospray teomic sample, saliva does not need any special preparation,
ionization) linked to nanochromatography. In general, the 2- like in the case of blood or serum. Usually ultrafiltration and
D gels are clearer when the fluid is preseparated and more initial purification on the RP column or microcolumn are suf-
spots are detected. These authors also utilized the advantage ficient to obtain proteins and peptides of interest. Depending
of the different hydrophobic properties of CSF proteins, and on a diagnosed population and a goal of analyses, operating
a reversed-phase solid-phase extraction (SPE) cartridge was personnel should remember that saliva, like almost every human-
used to prefractionate human lumbar CSF proteins into three derived fluid, is potentially biohazardous material and following
separate fractions prior to the two-dimensional gel electrophore- the safety rules during the entire procedure is a must.
sis [42]. Davidsson et al. [43] applied liquid-phase isoelectric Saliva contains well-known proteins such as lysozyme and
focussing for CSF preseparation, prior to 2-DE, thus leading alpha-amylase in comparably huge amounts. Besides, it seems
to the detection of low-abundant proteins. Several proteins, to contain other proteins and peptides that may be used for diag-
including cystatin C, IgM-kappa, hemopexin, acetyl-coenzyme nostics of the condition of the human organism while applying
A carboxylase-alpha and alpha-1-acid glycoprotein, were iden- fast, proteomic assays. Even based on the major proteins, apply-
tified in prefractionated CSF but not in unfractionated CSF. ing the proteomic method such as 2D electrophoresis supported
Low-abundant forms of post-translationally modified proteins, by high throughput MALDI-MS, on saliva sample shows a great
e.g. alpha-1-acid glycoprotein and alpha-2-HS glycoprotein, complication. Hirtz et al. detected that about 140 electrophoretic
can be enriched, and thus are better resolved and detected on spots correspond to alpha-amylase isoforms. About 90 of them
the 2D gel. Liquid-phase IEF, as a prefractionation step prior correspond to full-length post-translationally modified protein;
to 2-DE, reduces sample complexity, facilitates detection of the rest are probably products of truncation of amylase before
less abundant protein components, and increases the protein secretion [48]. To interesting proteins from the immunological
loads and the protein amount in each gel spot for MALDI-MS point of view, we should include defensins found in this fluid.
analysis. Defensins may play an important role in the protection against
Several prefractionation methods, involving ethanol, TCA, microorganism infections caused by food or drinks [49].
and TCA–acetone precipitation were compared to direct 2D- In the recent years, we observe a rapid increase of interest in
PAGE by Hansson et al. [44] in CSF analysis. The results suggest saliva as a potential diagnostic fluid. On the basis of the novel
that, with respect to protein recovery and purification potential, techniques, scientists from the University of Minnesota created
ethanol precipitation was found to be most efficient. a catalog of 437 saliva proteins, which is a good reference source
 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31 7

for further analyses [50]. Moreover, saliva seems to be the most frequently used acetone or 10% trichloroacetic acid in acetone
convenient source of markers for cancer and other diseases [51]. [64,70]. Islam et al. demonstrated that introduction of 10% TCA
Nowadays, a few biomarkers of tumors have been found. One of alone is not sufficient to remove contaminants and suggested
the good examples of the usefulness of this fluid may be the sali- TCA cleaning with sonication in the presence of glass beads,
vary c-erbB-2 protein, which proved to be a reliable marker of and brief grinding [64].
the breast malignant cancer [52]. Other diagnostic targets using As in the other biological samples, variability of proteins in
saliva as the source of markers may be the following: osteoporo- their pI range, abundance, solubility, hydrophobicity and other
sis [53], preterm labor [54,55], exposure to organophosphate features mentioned above makes them difficult to separate by
pesticides [56], drug testing [57,58] and various periodontal dis- classical two-dimensional gel electrophoresis. An alternative
eases [59,60]. separation method can be the liquid chromatography technique
connected on-line to mass spectrometry (LC–MS/MS).
1.4.4. Synovial fluid Typically, analysis of the protein complement proceeds
This type of fluid is rather not used in proteomics. Procedure through the phases of extraction, prefractionation, separation,
of synovial biopsy is not very difficult, but its application in mass spectrometry and identification [71]. General procedure for
detection of a disease based on the proteins and peptides con- sample preparation in proteomic research strongly depends on
tent is limited. At present, synovial fluid is used in diagnostics the plant type, its fragment being analyzed (leaf, fruit, sap, etc.)
of joint’s infectious diseases after detection of bacterial DNA or even, as mentioned above, on stage of the plant development.
[61]. Sometimes, analysis of synovial fluid can be useful in sar- To show one exemplary way of handling plant sample, a proto-
coidoses identification, but diagnostic success depends on the col for sample preparation of plant material from rice embryo
detection of CD 4(+) lymphocytes and other cells during histo- (O. sativa) and its further analysis by 2D electrophoresis are
logical investigations [62]. Synovial fluid is also useful in the briefly presented below. This procedure is described in details
diagnosis of, e.g. rheumatoid arthritis and other types of inflam- by Fukuda et al. [72]. The authors applied chemical homogeniza-
matory processes [63]. So far this human-derived fluid is not tion with solution consisting of urea, thiourea, Ampholine pH
involved in strictly proteomic investigations. 3–10, CHAPS (3-[(3-Cholamidopropyl)-dimethyl-ammonio]-
1-propane sulfonate), 2-mercaptoethanol and PVP (polyviny-
1.5. Plants lopolypyrrolidone), followed by boiling at 100 ◦ C, and centrifu-
gation. After discarding the supernatant, hexane was added to
Characteristic property of plant cell is its cell wall, mostly remove lipids and this step was repeated three times. Samples
made up of cellulose and its derivatives. Young plant cells are prepared in this way were analyzed by 2-DE.
surrounded with primary cell wall; in some plants and cell types,
a rigid secondary cell wall is present between the plant cell 1.6. Bacterial samples
and the primary wall after the phase of development. Gener-
ally, disruption of a cell wall and protein release is crucial for Pathogenic bacteria are an interesting object for proteomic
analytical success. Various chemical and physical techniques study in search of proteins having vaccine and diagnostic signifi-
are used to destroy cell wall, for example, lysing buffer, sonica- cance, determining novel targets for drug design and elucidating
tion, freeze–thawing and high-speed blending [64]. In particular, the cause of antibiotic or chemical resistances of these organ-
mature plants need special treatment. Islam et al. presented a isms.
procedure of extracting proteins from mature rice leaves for two- During sample preparation, problems can arise in disrupt-
dimensional gel electrophoresis with superior resolution [64]. ing bacterial cells, due to the presence of thick cell walls and
The plant cell wall is a dynamic structure and plays a key-role polysaccharide capsule in certain bacterial groups [73]. Some
in the plant life cycle. About 10% of the cell wall mass consists bacteria could simply be lysed by the constituents of the lysis
of cell wall proteins (CWP) [65] involved in signaling, interac- buffer, but others must be disrupted mechanically (by, e.g. son-
tions with plasma membrane proteins and modifications of the ication). Sometimes, removal of the cell wall by enzymatic
cell wall components [66]. Extraction of CWP is a challenge for digestion is necessary. It has to be remembered that 2-DE anal-
proteomics: to date available cell wall proteomes include only ysis or another technique, used to separate and analyze bacterial
labile and loosely bound proteins [67,68] and there is no effi- proteins, will reflect the proteome of the bacteria at the time when
cient procedure for the extraction of the strongly bound CWP proteins were solubilized. It means that all manipulations, such
[68]. as centrifugation, may stress the bacteria and thus, influence the
Most of the research is conducted on Thale cress (Arabidopsis protein pattern [74].
thaliana) and rice (Oryza sativa), which are the model plants of Another promise of bacterial proteomics, arises from the fact
a relatively small genome, which have been sequenced for A. that the genomes of some bacteria, e.g. Escherichia coli, are now
thaliana and the sequencing of the genome of rice is in progress. determined; so the complentary proteome analysis may bring
Another specific feature of plant proteome analysis is the some new interesting facts concerning cellular metabolism. The
presence of non-proteinaceous contaminants specific to the relatively small size of bacterial genomes makes it likely that we
plant, such as polyphenols, lipids, organic acids, terpenes or obtain a complete description of the free living organism from
pigments [69] that can interfere with separation methods. There- its genes to its complementary proteins and their functions, in
fore, cleaning procedures are desirable; for instance, the most the next few years [73].
8 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31

1.7. Viral samples [76] and the tissue is much too tough for ultrasonic process-
ing. Mechanical homogenizers are also chosen for processing
Recently, some attempts have been made to examine the pro- of hard or filamentous tissues, such as bones, teeth or cartilages
teome of viruses. Yoder et al. [75] analyzed the proteome of [77,78].
Vaccinia virus. Two fractions of viral proteins were prepared: In some cases mechanical homogenization may result in loss
membrane fraction containing soluble proteins and a fraction of the activity of the investigated material, particularly when it is
enriched with the cores and insoluble proteins. Those two frac- heat-sensitive and the cooling during processing is ineffective.
tions were prepared via treatment with detergent and centrifu- In the case of dispersing of human breast tumor tissue and calf
gation. Sixty-three different proteins were identified during this uterus, mechanical homogenization leads to the rapid denatura-
study. tion of the estrogen and progesterone receptors [79]. In this case,
Such research could be crucial for our understaining of virus detergent lysis at low temperature did not lead to any significant
biology and should help to discover new antiviral drugs and loss of the biological activity [79].
vaccines. Rotor–stator homogenizers can effectively break up animal-
derived, as well as plant tissues. In the case of plant tissues, where
2. Samples preparation—a general strategy cells are covered with strong cell walls, mechanical homogeniza-
tion seems to be one of the best methods of their disruption [80].
2.1. Methods of cell disruption For special applications, such as releasing chromosomes from
the plant cells, mechanical homogenization can be supported by
Homogenization is one of the steps allowing for preparation addition of a lysis buffer [81].
of any biological material as a sample for the proteomic analysis. Using rotor–stator homogenizers, we should remember that
The term “homogenization” covers many meanings such as mix- to gain optimal results, the tissue should be precut to slices, the
ing, stirring, dispersing, emulsifying, but in general, it means: size of which is slightly smaller than the diameter of the applied
receiving sample of the same composition and structure in the stator. Larger pieces of the sample, especially in the case of
whole volume. By applying homogenization in the procedure of rough or fibrous tissues, may clog generator’s inlet and make
sample preparation, we assume that the sample should change effective homogenization impossible.
its physical properties without any changes in the chemistry of To homogenize dry samples using mechanical processing,
components. open blade homogenizers, also called as blenders, are used.
Homogenization methods used for the proteomics purposes Rotating blades are closed in glass or metal chamber accord-
can be divided into five major categories: ing to the safety rules. Blenders can be used to dry or liquified
samples, and in case of non-satisfactory results, sample usu-
1. mechanical; ally needs to be processed by another apparatus, e.g. ultrasonic
2. ultrasonic; homogenizer, to receive optimal homogeneity [82]. This obser-
3. pressure; vation confirms the investigations of the homogenization method
4. freeze–thaw; of the skeletal muscles to receive high enzymatic activity in the
5. osmotic and detergent lysis. final solution. In this case, glass–glass homogenizer seems to be
much more effective than other methods (blender + detergent,
2.1.1. Mechanical homogenization blender + sonicator or blender + teflon pestle) [83].
Mechanical homogenization can be realized by at least two
types of devices: so-called rotor–stator homogenizers and open 2.1.2. Ultrasonic homogenization
blade mills. Ultrasonic homogenizers, also called as disintegrators or
Rotor–stator homogenizers are one of the best homogenizing sonificators, are based on the piezoelectric effect while gen-
tools applied in the laboratories. They can homogenize samples erating the high energy or ultrasonic wave, interacting with
in the volumes from 0.01 ml to about 20 l, depending on the the sample. Energy, resolved after explosion/implosion of gas
tip and power of a motor applied. Homogenizing tips can eas- microbubbles, effectively destroys solid particles such as cells.
ily be cleaned and sterilized. Use of disposable tips completely This method is widely used in laboratories and manufactur-
eliminates cross-contamination of the series of samples. Heat ing practices. Ultrasonic devices are used for homogenizing,
transfer to the processed mixture is low to moderate but usually emulsifying [84], dispersing [85], extracting or suspending of
needs external cooling. Sample loss is minimal in comparison the mixtures [86], and even for cleaning small metal parts in
to pressure processors (French presses), and very small amounts electronics [87]. For preparation of the sample, ultrasonic disin-
of samples can easily be homogenized. tegrators are successfully used to homogenize cells, after cell
This kind of homogenization is widely used for various culturing or after isolation from the organism. For example,
tissues and cells. Depending on the chemical resistance of after detailed studies on the usefulness of ultrasonic homoge-
a cutting tool, it is possible to homogenize samples under nization on the leukocytes, Fauth et al. confirmed that this type
strongly acidic or basic conditions to prevent degradation by of homogenization did not affect enzymatic activity of 13 inves-
endogenous enzymes. A good example could be the investi- tigated enzymes [88]. However, the procedure may lead to the
gation of high-energy phosphates in myocardial tissue where disruption of non-covalently bound molecular clusters (like mul-
mechanical homogenization occurs in 0.4 M perchloric acid tienzyme complexes) [89].
 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31 9

Ultrasonic devices are mainly used to homogenize small were efficient only towards granulocytes, which allowed the
pieces of soft tissues (brain, blood, liver). Plant cells and determination of bacteria viability [98]. Depending on the condi-
microorganisms can also be effectively homogenized by ultra- tions, osmotic lysis can be used also for microbial cell disruption.
sonic processors. One of the examples can be isolation of the In case of staphylococci, after addition of lysostaphin to hyper-
ribonucleotide reductase from Streptomyces aeurofaciens where tonic solution, lysis is as effective as ultrasonic homogenization
ultrasonic processor was very effective towards disruption of [99]. According to another report, addition of lysozyme to the
bacteria cells [90]. Tough and dense tissues are not recom- buffer supports osmotic lysis of Pseudomonas sp. [100] and
mended to homogenize using this equipment. other bacteria.
Detergent lysis is used for almost every type of cells,
2.1.3. Pressure homogenization viral envelopes and subcellular structures. The most commonly
Pressure homogenizer, also called as French press, is an effec- applied detergents are Triton X-100, Tween 80, Nonidet P-
tive system for homogenization of eukaryotic cells as well as 40 (NP 40) and saponin. Similarly, freeze–thaw and osmotic
microorganisms in suspension. lysis process, in proper buffer conditions, causes rapid perme-
French press is often applied for the preparation of the cell abilization of cell membranes and does not change the native
membranes for further experiments, e.g. in case of Leptospirosis conformation of intracellular antigens. This feature is often use-
pulmonary [91], or E. coli K+ /H+ transmembrane transport sys- ful during staining of the internal proteins of the cell. Among
tem [92]. Pressure homogenization is one of the most effective the typical applications of detergent lysis are, e.g. release of the
homogenizing system towards microbial and plant cells in sus- endogenous substances from bacterial organelles [101], protein
pension and is widely used in biological and biotechnological staining after cell permeabilization [102,103] and disruption of
laboratories to fast purification of the desirable proteins from the cells [104].
culture [93]. Some important molecules, such as mRNA, cannot
be obtained from the homogenate after using French press or 2.2. Protein solubilization
glass–glass milling [94].
Because of the construction, this type of homogenization is Protein solubilization process is widely quoted among the
ineffective towards tissues or organs without previous prepara- protocols of special importance applied in each proteomic sam-
tion in another type of homogenizer. ple preparation procedure. Regardless of the further separation
technique, this step strongly affects quality of the final results
2.1.4. Freeze–thaw homogenization and thus determines the success of the entire experiment.
This type of homogenization uses effect of ice crystals for- Once isolated, proteins in their native state are often insoluble.
mation in the tissue during freezing process. The method is rel- Breaking interactions involved in protein aggregation, e.g. disul-
atively fast, effective and, what is also important, does not intro- fide/hydrogen bonds, van der Waals forces, ionic and hydropho-
duce any external impurities into the sample because freezed bic interactions, enables disruption of proteins into a solution of
solution is isolated from the external environment. Freeze–thaw individual polypeptides and thus promotes their solubilization
homogenization is effective towards most of the bacterial, plant [105,106].
and animal cells in water solution and may be used as an Considering the great diversity of heterogeneity of proteins
additional or final step after mechanical or ultrasonic homog- and sample-source related interfering contaminants in biologi-
enization. cal extracts, simultaneous solubilization of all proteins remains
Some microbial cells, preconditioned in starvation mediums, a great challenge. Integration of proteins into membranes and
are resistant to homogenization by freeze–thaw method, as in the their association, and forming complexes with other proteins
case of Vibrio parahaemolyticus [95]. Another inconvenience or nucleic acids hamper the process significantly. Numer-
of this method is a possibility of causing the changes in activ- ous attempts undertaken during the past years flourished in
ity or properties of bioactive molecules (enzymes, membrane the development of strategies enabling the identification of
proteins) after a few freeze–thaw cycles. Such changes were the so far “unreachable” proteins, e.g. membrane [107,108],
confirmed in the case of G-protein coupled receptor kinases, acidic/basic [109–111], high/low molecular weight [112,113] or
␤-arrestins and other proteins [96]. low-abundant [114] and thus allowed for a more complete pro-
teome analysis. Despite the progress in the field, one needs to
2.1.5. Osmotic and detergent lysis remember that there is no single approach that may be multiplied
These methods of disruption of cells utilize osmotic pressure or copied. Each sample and conditions require a unique, experi-
or detergent interactions to destroy cells’ walls and membranes. mentally determined treatment. To avoid protein modifications,
They are also efficient for homogenization of nuclear and mito- aggregation or precipitation resulting in the occurrence of arti-
chondrial membranes in cell extracts. Osmotic lysis is often facts and subsequent protein loss, sample solubilization process
used to disrupt blood cells. It may be useful for RNA extrac- implicates the use of chaotropes (e.g. urea and/or thiourea),
tion, even from bacteria like Brucella abortus internalized in detergents (e.g. 3-[(3-Cholamidopropyl)-dimethyl-ammonio]-
macrophages [97], or to determine survival of Staphylococcus 1-propane sulfonate (CHAPS) or Triton X-100), reducing agents
aureus after phagosytosis by human granulocytes. After phago- (dithiothreitol/dithioerythritol (DTT/DTE) or tributylphosphine
cytosis, granulocytes were osmotically lyzed, which led to the (TBP)) and protease inhibitors in a sample buffer [115]. Their
release of staphyllococcal cells. Conditions for osmotic lysis proper use, together with the optimized cell disruption method,
10 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31

dissolution and concentration techniques determines effective- horizontal streaks may appear. To solve the problem associated
ness of solubilization [116]. with the presence of SDS, zwitterionic and non-ionic detergents
became widespread alternatives [114,125,126].
2.2.1. Chaotropes Uncharged detergents, mild and relatively non-denaturing
Chaotropes disrupt hydrogen bonds and hydrophilic inter- such as Triton X-100, NP-40 and dodecyl maltoside were among
actions enabling proteins to unfold with all ionizable groups the most widely applied in the present day proteomics, to
exposed to solution. The reagents applied within this group ensure protein solubilization and prevent aggregation [127,128].
are not as diversified as detergents. A neutral chaotropic agent, Further studies demonstrated the generally better solubilizing
urea, is used at high concentrations ranging from 5 to 9 M to power of zwitterionic detergents [129,130], although Luche
effectively disrupt secondary protein structure. As indicated by recently proved the non-ionic detergents, dodecyl maltoside and
Rabilloud et al. [117], addition of thiourea to the denaturing decaethylene glycol mono hexadecyl ether to be more efficient
solution containing urea, allows for substantial improvement of [126]. Also Taylor and Pfeiffer found the non-ionic n-dodecyl-
protein solubility, manifested in an increased number of pro- ␤-d-maltoside and the zwitterionic amidosulfobetaine ASB-14
tein traces visualized on 2D gels. These are, however, mostly to be more effective in solubilizing myelin proteins than the
water-soluble and several transmembrane proteins [118–120]. commonly used zwitterionic CHAPS [131]. In turn, Babu et
Inclusion of thiourea to the sample buffer decreases solubiliza- al. reported a distinct detergent 1,2-diheptanoyl-sn-glycero-3-
tion of urea. Therefore, when combined with 2 M thiourea, urea phosphatidyl choline (DHPC), to be even more potent in solubi-
concentration should not exceed 5–7 M [121]. While performing lizing integral membrane proteins than sulfobetaines or ASB-14
2-DE, thiourea should also be included in rehydration buffer to [132]. The non-charged reagent, stable in a wide pH range,
ensure solubility of all extracted proteins. Nevertheless, because may be useful in generating proteomic maps for most complex
the reagent may hinder binding SDS (sodium dodecyl sulfate) organelles including sarco(endo)plasmic reticulum.
to proteins, it has to be omitted in the equilibration buffers prior The advantage of zwitterionic detergents is that they combine
to the separation in the second dimension [122]. properties of detergents of other classes enabling efficient dis-
Urea and thiourea may hydrolyze to cyanate and thiocyanate, ruption of protein aggregates. Offering the low-denaturing and
respectively. This may result in modification of proteins and net-zero charge characteristics of non-ionic detergents, zwit-
hence, evokes artifactual charge heterogeneity. The process is terions also efficiently disrupt protein aggregation. Although
promoted by heat; therefore, samples containing the chaotropes Triton X-100 and NP-40 were less effective in solubilizing
should not exceed temperatures higher than 37 ◦ C. Thus, car- very hydrophobic proteins, zwitterions and sulfobetaines substi-
rier ampholytes, known to act as cyanate scavengers, are often tuted them successfully [133,134], allowing combination with
included in the urea buffer [115]. urea/thiourea and solubilization of membrane—but not integral
Charged chaotropic agent, guanidine hydrochloride proteins. The sulfobetaine CHAPS is most commonly applied
(GdnHCl), is another choice for the extraction medium for in proteomic studies nowadays due to its high solubility and
2-DE however, as it interferes with IEF, it needs to be removed a relative lack of detergent-induced artifacts. Its concentration
by dialysis against urea and thiourea. This may result in the ranges between 2 and 8% in 8 M urea. Conventionally for 2-DE,
loss of some classes of proteins. 4% CHAPS is used. Other alkylsulfobetaines such as N-decyl-
N-N -dimethyl-3-ammonio-1-propane sulfonate (SB 3-10) have
2.2.2. Detergents also been applied, however, these are relatively insoluble in high
Detergents and amphipathic molecules disrupt hydrophobic concentrations of urea [135]. On the contrary, 4-octylbenzol
interactions, thus enabling protein extraction and solubilization. amidosulfobetaine and ASB-14 are well compatible with 7 M
With respect to the ionic character of the hydrophilic group, urea and thiourea and are reported to have superior properties
they are classified into several groups: ionic (e.g. anionic [136].
sodium dodecyl sulfate (SDS)), non-ionic (uncharged, e.g. Combining various detergents and chaotropes may also be
octyl glucoside, dodecyl maltoside and Triton X-100) or beneficial. Chevallet et al. obtained best results with a denaturing
zwitterionic (having both positively and negatively charged solution containing urea, thiourea and synthesized zwitterionic
groups with a net charge of zero, e.g. CHAPS, 3-[(3-Chola- detergents and amidosulfobetaines with an alkyl tail containing
midopropyl)dimethylammonio]-2-hydroxy-1-propanesulfonate 14–16 carbons [129]. The amidosulfobetaine type ASB-14 and
(CHAPSO), tetradecanoylamidopropyl-dimethylammonio- mixed alkyl-akryl tail C8ø allowed solubilization of multiple
butanesulfonate (ASB-14)). Applicable concentrations of membrane, transmembrane and hydrophobic proteins from A.
detergents range from 1 to 4%, and the exact content of thaliana. Moreover, the reagents used for fractionation of mem-
solubilization solution needs to be verified in accordance to the brane proteins followed by 2-DE and combined with 7 M urea
method of choice for protein separation (some reagents may and 2.5 M thiourea, allowed solubilization of integral membrane
interfere with subsequent steps). proteins of E. coli and A. thaliana by Santoni et al. [137] and
Ionic SDS, highly efficient in solubilizing hydrophobic Moloy et al. [138].
and membrane proteins, interferes with non-denaturing elec- Solubilization of the proteins associated with membranes,
trophoresis and isoelectric focusing step and therefore, cannot existing in close contact with membrane lipids and forming
be used for 2-DE unless diluted and replaced with, e.g. CHAPS, membrane complexes, constitutes the greatest challenge for the
Triton X-100 or Nonidet P-40 (NP-40) [123,124]. Otherwise researchers nowadays. Although the number of identified mem-
 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31 11

brane proteins is still minor, in comparison to water-soluble performed by Navarre et al. [150], resulted in identification of
ones, the growing interest in the field and development of new intrinsic plasma membrane proteins from 1 to 12 transmembrane
detergents are promising for further research. Till date, there domains and positive GRAVY value. Another cationic deter-
are several membrane complexes purified and crystallized. The gent, benzyl hexadecyl ammonium chloride (16-BAC) in sub-
detailed list of those has been presented by Kashino in his review, sequent combination with SDS-PAGE, the system introduced
including the detergents used for protein solubilization [139]. by MacFarlane [151], was employed in numerous researches
The author points out that the choice of not only detergents but [107,108,152,153], improving separation, resolution and iden-
also their concentrations is significant. The paper describes the tification of integral membrane and basic proteins.
advantages, limitations and applications of SDS, CHAPS, Triton Regardless of recent advances and development of new deter-
X-100, Tween 20, n-octyl-␤-d-glucoside (OG), n-dodecyl-␤-d- gents, there is still no simple procedure allowing simultaneous
maltoside (DDM), and n-heptyl-␤-d-thioglucoside (HTG), to solubilization of the complex set of soluble, and membrane pro-
name a few. teins. The most frequent combinations of reagents for protein
Hydrophobic proteins are not recovered with the use of solubilization are presented by Govorun and Archakov [154].
CHAPS; therefore, other detergents need to be applied in 2D Seddon et al. [155] in turn, focus on the relevant molecular
gels. An excellent comparison of various reagents used for properties of detergents and lipids, and summarize different
separation of hydrophobic proteins from extrinsic ones (Tri- reconstitution and solubilization methods of membrane proteins,
ton X-100, Triton X-114, carbonate, chloroform/methanol), and implicating the strengths and weaknesses of the chosen reagents.
the efficiency of new zwitterionic detergents (øC5-øC8, C8ø, For more detailed information on basic aspects of detergent
ASB14) was reported by Santoni et al. [140]. In another report, physical chemistry, see the review by Garavito and Ferguson-
this group demonstrates the advantages of membrane washing Miller [156].
and the use of zwitterionic detergent C8ø against Triton X- Sonication may accelerate the protein solubilization process,
114 fractionation combined with CHAPS solubilization [137]. which usually requires several hours. In order not to overheat
Successful fractionation and improved recovery of hydrophobic the sample and prevent protein degradation and modifications
proteins on gels are also reported after the protein pretreatment in solutions containing urea, the burst should not last more than
with alkaline solution containing sodium carbonate or Triton few seconds.
X-100/KBR [138,141].
Blue Native (BN) polyacrylamide gel electrophoresis devel- 2.2.3. Reductants
oped by Schagger and von Jagov [142], based on the introduction Reductants disrupt disulfide bonds between cysteine
of Coomassie dyes to induce a charge shift on the proteins, residues, thus, promote unfolding of proteins and enable anal-
employed a combination of a mild detergent, lauryl maltoside, ysis of single subunits of proteins. Conventionally, sulfhydryl
and a zwitterionic salt aminocaproic acid to improve solubiliza- reducing agents: dithothreitol (DTT), dithioerythritol (DTE) are
tion and isolation of membrane complexes greater than 1 MDa applied in the sample preparation protocol. DTT and DTE are
under non-denaturing conditions. The approach was often impli- used at concentrations ranging from 20 to 100 mM. The free-
cated in proteomic research [143]. Henderson et al. successfully thiol-containing regents, week acids, are charged particularly
extended the technique to the analysis of pyruvate dehydroge- at alkaline pH, therefore may migrate out of the pH gradi-
nase complex [144]. As to the applied detergents, modifications ent while performing isoelectric focusing. This results in the
considered replacing laurylmaltoside with Triton X-100, allow- decrease in reductant’s concentration and may cause reoxidation
ing solubilization of the membrane-bound complex without its of sulfhydryl groups and loss of solubility for certain proteins
dissociation and the use of lower concentration of aminocaproic [121].
acid. Triton X-100 and sodium deoxycholate (DOC) were pre- More recently, phosphines, e.g. tributhylphosphine and tris-
viously successfully utilized in the native electrophoresis of carboxyethylphosphine (TCEP) in concentration of 2 mM were
hydrophobic proteins. The choice of detergents for BN elec- introduced as remedies for the problems associated with the
trophoresis is still limited as compared to IEF [145]. Detergents use of thiol reagents. Application of non-charged phosphines
used for solubilization of membrane proteins include mainly n- benefits when alkaline gradient is performed. The reagent signif-
dodecyl-maltoside, Triton X-100 and digitonin, used for, e.g. icantly increases solubilization of proteins during IEF, including
analysis of mitochondrial respiratory complexes. Eubel et al. keratins, and unlike DTT, does not interact with the alkylating
[146,147] cite also other detergents suitable for the native sol- substrates such as 4-vinylpyridine and acrylamide [157]. Thus,
ubilization of proteins, e.g. octyl glucoside, Brij 96, saponin, reduction and alkylation may be performed in a single step. The
Big CHAPS, C12E5/8, n-decanoylsucrose and NP-40. Digitonin reagents are discussed in details by Govorun and Archakov [154]
proved to be a very suitable detergent for the solubilization and and Herbert [121]. Previously, ␤-mercaptoethanol was used,
stabilization of supercomplexes of Arabidopsis mitochondria. In however it needs to be used at higher concentrations and can
combination with BN-PAGE, nine photosystem supercomplexes produce artifacts [158].
were resolved by Heinemeyer et al. [148] and incubation of
membranes with sublytic amounts of digitonin improved separa- 2.2.4. Protection from proteolysis
tion of plasma membranes from other membranes [149]. Finally, Proteases regulate many essential biological functions
protein separation using the cationic detergent cetyl trimethyl including influencing physiological processes, cell cycle and
ammonium bromide (CTAB) and SDS in second dimension, apoptosis, to name a few. According to Rawlings et al. [159]
12 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31

there are nearly 700 proteases and their homologs defined in the 2.3. Removal of contaminants
human genome that may be classified to one of the metallo-,
serine-, cysteine-, or aspartyl proteases groups. If not inhibited, The pH and ionic strength of sample solutions consider-
liberated/activated endogenous proteases during cell membrane ably influence protein solubility. Therefore, buffers, salts and
disruption, are responsible for uncontrolled enzymatic proteins detergents are included in sample solutions. They often tend to
degradation. Such proteolysis may produce artifacts and hence interfere with further protein separation steps, inhibit the diges-
complicate further analysis. tion process, collide with the mass spectrometry analysis, or
Olivieri et al. [160] showed major differences in 2D patterns complicate data analysis significantly, thus need to be removed
of red blood cell membranes, with and without application of at a proper time of analysis. An excellent review on sample
protease inhibitors. Substantial proteolytic action in untreated preparation for peptides and proteins in biological matrices was
cells resulted in poor recovery of high molecular weight proteins recently presented by Visser et al. [176].
and the peptide mixture barely extended a molecular mass of
50 kDa. As the problem arises in early stage of sample prepara- 2.3.1. Salts
tion, it concerns not only 2-DE but also other techniques involved Salts naturally occur in body fluids such as plasma, cere-
in proteome analysis. Denaturants employed while performing brospinal fluid and urine, or may be added into the sample
sample preparation, tend to inhibit majority but the most resistant buffer to prevent protein precipitation. Salts migrate away from
protease. Proteases, however, are more resistant to denaturation proteins during isoelectric focusing, thus contributing to their
than most other proteins. precipitation and aggregation. Moreover, a high electrical cur-
Protein degradation may be minimized by quick and small- rent carried by the salt load, interferes with electrophoretic
scale tissue extraction [161], boiling the sample in SDS buffer separation of proteins and reduces the efficiency of 2-DE [177].
with the high-pH Tris-base, or, on the contrary, lowering Hence, if present in concentrations >100 mM, salts should be
the pH and performing ice-cold precipitation in, e.g. 20% removed prior to IEF. Cup loading tolerates a slightly higher
trichloroacetic acid. Alternatively, denaturation in boiling in salt concentration [115]. It is also possible to dilute sample
water [162], focused microwave irradiation [163] and the use below the critical concentration and apply larger sample vol-
of organic solvents [164] may be applied to inhibit proteases ume on the immobilized pH gradient (IPG) gel. Sample dilution
activity as described by Ivanov and Yatskin [165]. While active is also advised prior to capillary electrophoresis (CE), provided
in high concentration of urea, proteases may effectively be that proteins of interest are present at detectable concentrations
inhibited by addition of thiourea to a lysis solution. More- [178].
over, concentration-dependent efficiency of thiourea in inhibi- Most often, salt removal is being accomplished via (spin,
tion of the proteolysis of sensitive substrates and solubilizing micro) dialysis [179,180], ultrafiltration [181,182], gel filtration,
proteins was underlined by Castellanos-Serra and Paz-Lago precipitation with TCA or organic solvents [161] and solid-phase
[166]. In another experiment, heat shock proteins (sHsps), were extraction. Other alternative is the use of commercially available
found to protect proteins in vitro from proteolytic degradation clean-up kits [183].
[167]. Dialysis is an effective method enabling extraction of pep-
In general, addition of specific protease inhibitors (e.g. tides/proteins from biological matrices. This procedure has how-
phenylmethylsulfonyl fluoride (PMSF), aminoethyl benzyl- ever, some drawbacks: is time consuming, difficult to automate,
sulfonyl fluoride (AEBSF), ethylene diamine tetraacetic acid requires large volumes of solutions, and may result in sample
(EDTA), pepstatin, benzamidine, leupeptin, aprotinin) or cock- degradation or loss. Dialysis is usually followed by protein pre-
tails with a broader activity spectrum, is recommended during cipitation or concentration in vacuum. Spin dialysis is faster,
cell disruption and subsequent preparation [168–171]. no extra sample volume is needed, but also protein loss may
As observed by Finnie and Svensson [172], protein degra- occur due to the adsorption on a dialysis membrane. The tech-
dation, minor during protein extraction, was considerably nique should be applied before urea and detergents are added to
increased when isoelectric focusing (IEF) separation was per- sample solution.
formed. In this case proteolysis may be almost completely pre- Microdialysis, based on size and shape differences, is suit-
vented by using cup loading to apply proteins to the IEF strip able for smaller sample amounts with lower dialysis flow-
and inclusion of protease inhibitors in the IEF reswelling buffer. rates. Unlike dialysis, microdialysis may also be performed in
Protease inhibitors should be applied with precaution, as it was vivo [184,185]. Samples obtained using the method are, how-
reported that they may modify proteins, introduce charge trains ever, diluted and sample recovery is about 100-fold lower than
and adducts, and hence interfere with further peptide studies achieved with solid-phase extraction [176]. As a rule, the higher
[173,174]. A variety of the most commonly applied protease mass of the component, the lower recovery through microdial-
inhibitors in 2-DE, including their advantages and limitations, ysis membrane. The advantage of the technique might be a
have been described in details in [116]. possibility of its on-line coupling to LC [186–188].
For liquid chromatographic separation, protease inhibitors Ultrafiltration, similarly to dialysis, implicates the use of
are advised to be added both in binding and elution buffers, membrane, but here it plays a role of a sieving device and not a
maintained at 0–4 ◦ C. Oh-Ishi and Maeda suggest GdnHCl as the barrier between two liquid phases with different characteristics.
efficient reagent for inhibiting protease activity and the endoge- The multi-step operation is efficient for sample concentration
nous proteases in cells [175]. and purification [189], but is difficult to automate. Ultrafiltration
 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31 13

is considered superior in protein recovery to common precipita- of proteins from human body fluids (plasma, urine, amniotic
tion techniques [190] or dialysis [191]. fluid and tears) implicating the use of centrifugal filter device
Gel filtration, usually carried out with the use of Sephadex and the sample buffer containing CHAPS for efficient lipid and
[192], is also acceptable as a salt removal method with efficiency salts removal. A wide range of techniques used for salt removal
comparable to ultrafiltration and better than precipitation [190]. including fast-protein liquid chromatography (FPLC), desalting
It was also reported that it may not achieve sufficient desalting columns, SPE, ultrafiltration or dialysis was also proposed by
for mass spectrometry purposes, and the procedure results in an Visser et al. [176].
excessive sample dilution [193]. Finally, Chambers reported the automated, high throughput
Protein precipitation followed by resuspension in sample use of nickel and glutathione discs for protein purification [200].
solution belongs to the most commonly applied procedures
enabling removal of contaminants such as salts, lipids, polysac- 2.3.2. Detergents
charides, detergents, nucleic acids, etc., that may interfere with Most common detergent removal methods include dialysis,
further analytical steps. There is currently no method that would gel filtration chromatography, hydrophobic adsorption chro-
allow precipitating all proteins and, consequently, only precipi- matography and protein precipitation. For detergents with high
tated proteins can be further resolubilized. Therefore, precipita- critical micelle concentration (CMC) and/or small aggregation
tion should be avoided when screening for a complete proteome numbers, dialysis is usually the preferred choice. Detergents can
is required. Most commonly, precipitation with TCA, acetone, be extracted against a detergent-free buffer in about 200 excess
chloroform/methanol, ammonium sulfate or combinations of the over a period of days. For a wider spectrum of detergents present
above are being performed. TCA/acetone precipitation method, in the sample, gel filtration can be applied. This results, how-
the most popular for 2-DE, is more effective than any of the ever in a considerable sample dilution. In turn, ion-exchange
reagents used alone. Precipitation with acetone (75% final con- chromatography effectively excludes non-ionic and zwitterionic
centration) is less powerful, but enables easier protein resuspen- detergents, although Zischka et al. reported its successful appli-
sion. For overviews on precipitation methods, including general cation also for SDS removal [201].
procedures and limitations the references [176,181,194] are rec- Furthermore, SDS can be removed with nanoscale
ommended. hydrophilic phase chromatography [202] or acetone precipita-
Solid-phase extraction is a fast, versatile, easy to use, and tion. When carried out at −20 ◦ C, the process is more effec-
easy to automate sample preparation technique [176]. It enables tive than at room temperature. Finally, Dong et al. developed
both concentration and purification of the sample. Most often, ceramic hydroxyapatite (HAP) chromatography for the com-
the technique bases on the reversed-phase separation mechanism plete removal of SDS bound to soluble or membrane proteins
employing C18 resin. Solid-phase filled tips may also serve as [203].
miniature chromatography columns for microscale solid-phase As to the zwittergents removal, Hannam et al. reported equal
extraction (␮SPE), and are widely applied in proteomic research efficacy of gel filtration chromatography and a detergent affinity
to desalt, concentrate, and fractionate peptides and proteins bead chromatography column, lightly dominating over capabili-
[110,195,196]. Prior to SPE, centrifugation, filtration or pre- ties of dialysis. Also SPE was found efficient in CHAPS removal
cipitation are advised to be performed, in order to remove the from dilute protein solutions, offering significant advantages
contaminants that may block the cartridge during extraction. over standard dialysis or gel filtration methods [204,205]. Sev-
Several interesting reports on comparison and evaluation eral of the above methods including nickel columns and His tags
of various desalting techniques have been published recently. were reviewed by Seddon et al. [155] in details.
Yuana and Desiderio [177] tested the mentioned desalting meth- Moreover, there are commercially available kits, e.g. deter-
ods. Smaller proteins loss was reported while performing ultra- gent precipitation reagents or gels effective for binding and
filtration, mainly due to the adsorption on a filter, as com- removal milligram quantities of various detergents from pro-
pared with dialysis. Column salt removal method enabled the tein solutions (e.g. Extracti-Gel® D Detergent Removing Gel
highest protein recovery. Alternatively, commercially available and the SDS-OutTM SDS Precipitation Reagent and Kit, Pierce)
microcentrifuge filtration devices (spin filters) can be applied [206]. Hydrophobic adsorption employing the use of insoluble
to wash away contaminating species and to resuspend proteins resin (e.g. CALBIOSORBTM , Calbiochem) can also be used to
in buffers compatible with digestion [197]. Lazar et al. evalu- remove excess detergent.
ated centrifugal filters, reversed-phase high-performance liquid
chromatography (HPLC), and size-exclusion HPLC. The latter, 2.3.3. Abundant proteins
using aqueous acetonitrile as the mobile phase directly coupled Protein concentration in biological samples may vary even
to ESI-MS, provided the best performance [198]. more than 10 orders of magnitude [207]. Proteomic analyses
The efficiency of four other desalting procedures (desalting are, hence, more complicated and detection of least abundant
column packed with Sephadex G-100, on-target washing, cen- proteins is hampered by those molecules present at higher con-
trifugal filter devices and microcolumns C18 ) was carried out centration. Inter alias, high-abundant proteins prevent optimal
by Salplachta et al. [193]. For intact proteins, the experiments focusing, limit loading capacity of low-abundant species and
showed that the best desalting procedure was the application tend to mask considerable areas on the 2-DE gels. Plasma, serum
of microcolumns C18 , pipette tips and centrifugal filter devices. and CSF, protein sources of great importance to biomedicine,
Moreover, Joo et al. [199] developed a method for extraction diagnostics and therapeutics (see Section 1.4), contain up to
14 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31

90% of highly abundant proteins such as albumin, immunoglob- casein- and bovine IgG-specific immobilized Sepharose enabled
ulins (IgG and IgA), antitrypsin, transferrin, transthyretin, ␣1- Yamada et al. to identify several low-abundant proteins of spe-
antitrypsin, hemopexin or haptoglobin. Removal of those pro- cial physiological relevance [223]. Alternatively, an affinity spin
teins may increase detection of other molecules present at low tube filter technique can be applied to enrich the low-abundant
concentrations, however, it may also result in a loss of other biomarkers [224].
proteins, hindering identification of holistic alterations in the A preparative electrophoresis system, Gradiflow, is being
analyzed proteomes [208]. employed for depletion of albumin under native and denatured
Various strategies have been presented for the removal of conditions (see also Section 1.4). The technique enables sep-
high-abundant proteins [209], most of which base on affin- aration of proteins on the basis of their molecular weight and
ity chromatography employing dye-ligands, their derivatives charge. Hence, separation of the majority of plasma proteins
[210,211], mimetic ligands [212,213], proteins A and G [214], characterized by the pI close to that of albumin may be carried
and antibodies (immunoaffinity depletion) [215]. Cibacron Blue out successfully [225]. Albumin can be also effectively removed
columns and their derivatives are commonly used to bind albu- by isoelectric trapping [226] and peptide affinity column chro-
min whereas immunoglobins are excluded based on their inter- matography [213].
actions between with proteins G and A [216,217]. Among the wide range of applicable precipitation methods,
Other dye ligands include Procion Red He3B, Reactive Blue ammonium sulfate fractionation was reported most efficient in
MRB, Reactive Green H4G, Reactive Green HE4BD, Reactive albumin removal [181].
Yellow M8G and Reactive Brown M4R all of which can be Finally, solid-phase extraction constitutes another promising
coupled to solid supports [5]. Application of the dye-employing tool to reduce differences in proteins concentration, and thus
methods is, however, limited due to the fact that the dyes and enhancing the possibility to detect and analyze low-abundant
high-abundant proteins themselves tend to bind low molecular species [177,227].
weight proteins, lipoproteins, and enzymes present in a sam-
ple [177,191]. Hence, removal of high-abundant proteins results 2.3.4. Lipids
also in non-specific loss of other species. This effect is called Similarly to salts, lipids are widely present in biological fluids
“albumin sponge effect” and sometimes could be prevented by such as plasma. Numerous proteins are complexed with lipids,
sample dilution with a buffer containing acetonitryle [218,219]. and this interaction reduces their solubility and might affect the
Approaches based on ultrafiltration proved to be less successful pI and MW. Moreover, by forming complexes with detergents,
in high-abundant proteins removal [220]. lipids reduce protein enrichment/separation efficacy. Most often,
Recently, Ahmed and Rice demonstrated the use of affinity if 2-DE separation is to be performed, the use of centrifugal
dyes in conjunction with a supporting matrix, ProtoClear and filter device and the sample buffer including CHAPS allows for
Affi-Gel Blue, along in combination with Protein A (Aurum efficient lipid and salt removal, thus ensuring high percentage
serum protein mini kit, Bio-Rad), which proved efficient in of proteins recovery and high-quality separation.
removing high-abundance proteins without a significant loss Joo et al. compared the use of CHAPS and subsequent cen-
of protein profile or number of protein spots [5]. The loss trifugation to sample boiling in SDS and dialysis [228]. Appli-
of associated proteins was reported to be dependent on the cation of CHAPS allowed visualization of more proteins than
treatment duration. Furthermore, the authors suggest Agilent achieved with the classical method, no streaking was observed
multiple affinity removal system (Agilent Technologies) as capa- on gels. Heating samples at presence of SDS cannot be per-
ble of binding and retaining six highly abundant proteins and formed if proteins are to be resolved by IEF [105,121]. Alterna-
enabling enhanced detection of low-abundant molecules in a tively, precipitation in acetone or combination of TCA/acetone
high throughput manner. ProtoClear technique was reported to removes lipids efficiently. Lawless also reported a precipita-
be far more specific at clearing albumin and immunoglobulin tion technique employing acetonitrile supplemented with 1%
G from human serum samples than Cibracon Blue Dye chro- TFA and 1% n-nonyl-␤-d-glucopyranoside, which was found
matography [219]. especially helpful in dissolving membrane proteins and lipids
Furthermore, Govorukhina et al. evaluated capacity of vari- [229]. Moreover, Watkins et al. introduced a new method for
ous columns to remove albumin and/or IgG from human serum delipidation of human serum lipoproteins involving the use of a
[210]. HiTrap Blue and protein G columns in combination were reversed-phase C18 solid-phase extraction cartridge. The method
found more effective than Aurum columns. Another kind of of delipidation produced a higher and more reproducible pro-
affinity chromatography technology enabling high-throughput tein yield than the conventional liquid–liquid methanol-diethyl
proteomic removal of abundant proteins from serum implicates ether delipidation technique, and implemented a fast, sequen-
the use of SwellGel protein A/G and Cibacron blue discs or resin tial desalting and delipidation of the lipoproteins for subsequent
combination [200]. For more information concerning prepara- mass spectrometric analysis [230].
tion of body fluids, the readers should refer to Section 1.4.
Immunoaffinity-based protein subtraction chromatography 2.3.5. Polysaccharides
(IASC) described by Pieper at al. was shown to effectively Polysaccharides and nucleic acids may interact with carrier
and reproducibly remove multiple, abundant proteins present ampholytes causing streaking visible on 2D gels. Moreover,
in plasma and serum, enabling visualization of a vast number their presence in a sample buffer can result in viscous solutions,
of lower abundance proteins [222]. Similarly, the use of ␤- clogging the pores of the polyacrylamide gels, thus causing
 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31 15

either precipitation or extended focusing times, and resulting 2.4. Protein enrichment methods
in horizontal streaking. Furthermore, some polysaccharides are
negatively charged and thus may form complexes with proteins Biological samples used as a source for proteomics analysis
by electrostatic interactions. are usually very complex. As it was mentioned before, proteins
In order to exclude polysaccharides from the sample, precipi- concentration range in a single sample is usually beyond the
tation in TCA, acetone, ammonium sulfate or phenol/ammonium dynamic range of any single analytical method, and any exam-
acetate, followed by centrifugation may be beneficial. High- ined proteome could have a large and unknown complexity.
speed ultracentrifugation is applied when the removal of larger Hence, prior to analysis, it is desirable to reduce the complexity
polysaccharides is required. These molecules can block the of the sample by its prefractionation, or to enrich it with proteins
matrix of chromatographic materials and the pores of mem- of our interest.
branes [231]. Furthermore, similar methods as for nucleic acids The fundamental idea of prefractionation is to isolate sample
removal are advised in case of lipids and polysaccharides. into distinguishable fractions containing restricted numbers of
molecules. The sample can be fractionated using a variety of
2.3.6. Nucleic acids approaches including precipitation, centrifugation, liquid chro-
Nucleic acids can interfere with carrier ampholytes and pro- matography and electrophoresis-based methods, filtration, and
teins, and may contribute to the poor recovery of DNA-, RNA- velocity or equilibrium sedimentation. The selection of the tech-
binding protein and evoke horizontal streaks on 2D gels [114]. nique strongly depends on the nature of sample to be analyzed,
Silver staining used for visualization of samples separated via 2- and the object of the study.
DE detects also nucleic acids, if present in the gel, which results The enrichment methods allow for increasing the concentra-
in background smearing. Moreover, nucleic acids increase sam- tion of proteins of interest. This statement is really important
ple viscosity, clog the pores of polyacrylamide gels and affect in the proteomics study, because usually low-abundant proteins
the accuracy of sample loading. carry valuable diagnostic information and are responsible for
In order to remove DNA and RNA, digestion with protease- processes ongoing in the cells. The conventional method of sam-
free DNase and RNase is often applied [115]. The treatment ple concentration by simple evaporation results, along with the
reduces nucleic acids to mono- and oligonucleotides. One should proteins, in buffer components concentration (e.g. salts and other
be aware that DNAses and RNases may appear on 2D pat- contaminants). Therefore, universal, more convoluted methods
terns. Alternatively, protein precipitation from the solution is of protein enrichment are necessary. It should be remembered
advised. Proteins associated with nucleic acids may be lost from that during any enrichment process, conditions must be stable to
the sample, unless the nucleic acid fraction is extracted with avoid protein interactions among the rest of mixture components
the detergent cocktail as presented by Giavalisco et al. [232]. (e.g. non-specific interactions with other proteins).
According to Rabilloud [233], ultracentrifugation and addition
of basic polyamine, e.g. spermine, is also effective in removal 2.4.1. Precipitation
of large nucleic acids, as well as high MW proteins. High-ionic The most common methods of protein enrichment and
strength extraction and high-pH extraction appear to be potent purification rely on selective precipitation using acetone,
in minimizing interactions between negatively charged nucleic trichloroacetic acid, ethanol, isopropanol, diethylether, chloro-
acids and positively charged proteins. A convenient alternative form/methanol, ammonium sulfate, polyethylene glycol (PEG),
utilizing QIAShredder (QIAgen) and subsequent centrifugation and a number of commercially available affinity precipita-
was reported by Leimgruber et al. [114]. tion kits [190,237,238]. Ammonium sulfate is expected to be
the most widespread precipitant utilized, which causes pro-
2.3.7. Other substances tein destabilization. This is known as the “salting-out” effect.
Endogenous, small ionic molecules; nucleotides metabo- Addition of various organic solvents promotes an increased elec-
lites, phospholipids present in cell lysates are often negatively trostatic attraction between particles of opposite charge in the
charged, resulting in poor focusing towards anode. Other dis- sample solution. This effect leads to protein precipitation. Pre-
turbances during protein separation may also be evoked by cipitation can be promoted by addition of the organic polymers
insoluble material, e.g. organelles clogging gel pores. Non- such as PEG. The precipitate recovery rely on redissolving in a
proteinaceous impurities may form complexes with proteins smaller volume, followed by centrifugation or filtration. Other
hampering their solubilization. Therefore, in order to remove type of protein enrichment routine is immunoprecipitation. The
the contaminants, TCA/acetone precipitation or other salt- principle is based on the utilization of antibodies that are selec-
excluding techniques are effectively performed. Alternatively, tive for one or a group of proteins with a similar epitope (e.g.
high-speed centrifugation [234,235] can be applied. phospho or glycoproteins) [239] (see also Section 2.4).
The presence of phenols observed in plant tissues, may mod-
ify proteins through an enzyme-catalyzed oxidative reaction. 2.4.2. Centrifugation
Oxidation may be prevented with the use of reductants while One of the simplest methods of protein enrichment is ultra-
performing tissue extraction. Furthermore, protein precipitation centrifugation. Separation of cell substructures can be attained
with TCA, followed by extraction of phenols with ice-cold ace- by the series of runs at different centrifugal forces, or in
tone or phenol adsorption to polyvinylopolypyrrolidone (PVP) sucrose/mannitol gradient, which allows separation of different
[236], are advantageous. cellular or tissue material, according to the density character-
16 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31

istics of the structure [240]. This technique is useful for con- One of the latest technology platforms that have been devel-
centration of mitochondrial, membrane, nuclear or other locally oped for proteomics includes, e.g. Rotofor, the multicompart-
abundant proteins. mental instrument, capable of fractionating proteins according
to their pI [227]. The Gradiflow is another multifunctional elec-
2.4.3. Electrophoretic methods of protein enrichment trokinetic membrane device that can be used for proteins sep-
One-dimensional gel electrophoresis (1-DE) separation of aration based on the differences in mobility, pI and the size of
proteins by their size is the traditional manner for protein enrich- proteins. One of the major drawbacks of this strategy is that the
ment and analysis (see also Section 3.1). Two-dimensional gel fractions collected from above equipments contain high amount
electrophoresis (2-DE) has the benefit that it enables simulta- of ampholytes, which, fortunately, can be removing by micro-
neous visualization of hundreds of protein spots, their post- columns filled with C18 material.
transational modifications, and quantification of protein lev- It was shown by Petsev et al. that proteins can be separated
els. Reproducibility of protein patterns between laboratories on a small-scale without use of more expensive chemicals or
is more difficult, because of protocols variations, artifacts and molecules, such as antibodies or synthetic ampholytes [249].
technology. Separation based on 2-DE technique dedicated to Field gradient device separation mechanism is based on the
hydrophobic and membrane proteins, as well as alkaline and opposition of two or more forces, from which one is constant as
low-molecular weight polypeptides, possess some limitations. a function of distance along fluid channel, while the other force
Even though, membrane proteins containing up to 12 transmem- is changed gradually or stepwise.
brane helices, have successfully been resolved [125]. Free-flow electrophoresis (FFE) is another method for puri-
fying cells and subcellular organelles, but it is rarely used for
2.4.4. Membrane proteins enrichment protein enrichment because of high diffusion effect. FFE sepa-
The membrane proteins are of great importance for pro- rates charged particles ranging in size from molecular to cellular
teomics, because they represent receptors, transporters, chan- dimensions, according to their electrophoretic mobility or pI
nels, and they participate in a variety of significant cellular [227,250,251].
mechanisms. Because of their function, they are considered as
a major pharmaceutical target, and ability of their detection is 2.4.6. Chromatographic techniques
of particular interest. Unfortunately, a vast under-representation Another most applied proteomic system for proteins enrich-
of membrane proteins has been observed during whole cell pro- ment is chromatographic separation, which is proficient to
teome analysis. reduce the complexity of the sample by separating proteins
Membrane proteins are usually enriched by ultracentrifuga- according to their charge, hydrophobicity, size, or specificity.
tion in sucrose gradient, lectin affinity chromatography in com- Solid-phase chromatographic techniques are capable of protein
bination with centrifugation, silica beads or biotinylation and fractionation with low, as well as very high selectivity depending
interaction with immobilized streptavidin [241]. Solubilization on the adsorbent and conditions of adsorption–elution selec-
of this fraction has to be improved by using special detergents, tion. Affinity chromatography utilizes highly specific biological
and the choice of them depends on the nature of experiment. interactions such as that between antigen and antibody, recep-
Ferro et al. used combination of chloroform and methanol to tor and ligand, or enzyme and its substrate, or inhibitor and
extract hydrophobic chloroplast membrane proteins [242]. The it allows for very efficient protein enrichment. This technique
aqueous two-phase system, which employed detergents DDM, offers many advantages over conventional chromatography, for
Triton X-114 or PEG, was used for membrane proteins enrich- example, specificity and selectivity to name the few.
ment by Sivars and Tjerneld [243] and Everberg et al. [244]. Several technologies have been designed as important tools
This technique requires the selective binding of one or more pro- for biological research, including proteome- and inhibitor-based
teins of interest to the one of the incompatible aqueous phases. affinity chromatography, along with activity-based profiling.
Detailed description of different detergents used during this kind These techniques reduce sample complexity to access the less
of analysis, could be found in Section 2.3. abundant proteins, and to help understand pathological pro-
Identification of membrane proteins is not an easy task due cesses [252]. During inhibitor-based procedure, the target pro-
to the lack of tryptic cleavage sites across transmembrane chain tein is attached to the solid support, covered with inhibitor
fragments. Enzymatic digestion often results in large, hydropho- molecule. Proteome-based affinity chromatography relies on
bic pieces, which hinder identification. To enlarge sequence searching the protein of interest within the captured compounds.
coverage, a mixture of proteases and cyanogen bromide with Finally, the activity-based affinity profiling is achieved by com-
addition of detergents could be performed [245,246]. petition of target protein with standard probe (e.g. biotin or
fluorophore) [253].
2.4.5. Prefractionation For investigation of post-translational modifications
Preparative liquid electrophoretic methodologies are other (PTM’s), a variety of combinations of the affinity-based enrich-
fractionation systems for proteome profiling. The fact that pro- ment and extraction methods, multidimensional separation
tein fractions are collected here in a liquid form, promotes techniques and mass spectrometry are applied. More then 300
application of such methodologies, because sample handling types of PTM’s are currently known [254].
is diminished, thus the risk of it loss or degradation is reduced Glycosylation, one of the most common PTM’s, plays funda-
[247,248]. mental role in a diverse set of biological processes, such as proper
 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31 17

folding, signaling pathways associated with the transformation Solid-phase extraction also provides sample enrichment
of a normal cell to a cancer cell, the immune response and cellu- and purification. There are plenty of solid-phase microex-
lar regulation. In one study the multi-lectin column was used to traction systems (SMECs), which enable simple and rapid
enrich glycoproteins from human serum [255]. All lectins have extraction technique, including ZipTip, ZipPlate, Gelloader and
ability to bind certain monosacharides such as mannose, glucose MassPREP PROtarget [196,269]. Another easy sample prefrac-
or fucose. It was found that the multi-lectin column was highly tionation method was reported [270] which is based on neutral
specific for O- and N-linked glycans present in serum. Lectin beads of Sephadex to isolate proteins according to their isoelec-
affinity chromatography may be utilized to purify a variety of tric points.
proteins. Fluorescent staining methods were developed as well, Antibody arrays are very useful for profiling biomarker can-
in order to visualize and analyze glycoproteins and phosphopro- didates in large sets of biological samples. A variety of substrates
teins [256]. and methods of antibody attachment have been used [271,272].
The reversible phosphorylation, which occurs on serine, thre- Moreover, the entire libraries of antibody microarrays are cre-
onine and tyrosine residues, is a basis for regulation of funda- ated [254,273,274].
mental cellular functions, such as DNA replication, cell division, Surface-enhanced laser desorption/ionization (SELDI) is
cell cycle control, transcription, translation, protein localization, also a useful tool for protein enrichment, since the sample prepa-
energy metabolism and signal transduction. The knowledge of ration procedure involves surface prefractionation of protein
the phosphorylation status of all proteins at a given time would mixtures on the derivatized target plates, in order to maximize
be one of the major issues for understanding of the physiological the number of detectable peaks [227,271,275].
and pathological role of phosphoproteins. Detection of phospho- An additional method of sample fractionation is laser cap-
peptides is possible by scanning for the neutral loss from peptide ture microdissection (see Section 1) which allows separation
during MS/MS analysis under positive and negative ion modes from neighboring cells in biopsy material, cells in cultures, etc.
[257]. LCM is still an effort-demanding procedure that yields limited
Traditional metabolic labeling methods use the radioactive amounts of material and so it is not fully suitable for testing a
isotopes 32 P and 33 P [258]. The most recent approach is based large number of samples [271].
on chemical modifications of phosphate groups [259], and appli- The above applications and technologies demonstrate the
cation of antibodies against specific phosphoaminoacids [260]. value and potential of protein enrichment in proteomics. Further
Phosphopeptides, likewise glycoproteins, calcium-binding improvements to the methods should broaden their exploitation
and histidine-exposed proteins can be selectively enriched by and create large impact on proteomic research.
immobilized metal affinity chromatography (IMAC), which
reduces the complexity of the protein mixture [227,261]. IMAC 3. Samples preparation guidelines for various
is based on formation of coordinate bonds between basic groups proteomic techniques
on protein surface, and metal ions (e.g. Fe, Co, Ni, Cu, Zn, Al)
immobilized on chromatographic beads. Elution of bound pro- During sample preparation for proteomics study it has to be
teins is undertaken by lowering the pH or using chelating agent, remembered that realistically, no single method could be applied
such as EDTA. Analysis of protein phosphorylation was intro- to all possible samples, and there is always necessity to optimize
duced based on covalent modifications with biotin at the site of the procedure for particular samples. All applied procedures
phosphorylation [262]. The DIGE (difference gel electrophore- should be as simple as possible and, even more important, repro-
sis) technology allowed separation of phosphoproteins with high ducible. It is necessary to avoid proteins loss, degradation and
resolving power [263]. Phosphoproteins could be also analyzed modifications.
through stable isotope labeling by amino acids in cell culture Before starting the experiment, it is advisable to know the
[257]. fundamental problems that might appear at each step of the pro-
Ubiquitination is a process of biological labeling proteins cedure. This is crucial, due to the fact that each next step is based
which are destined for destruction. His6-tagged ubiquitin facili- on the quality of the previous one, and if the mistake is done at
tate affinity isolation of ubiquitin-conjugated compounds [264]. the beginning, there is no way to improve the results at the end.
Wide spectrum of affinity ligands are utilized for a variety of
applications, but there is a requirement for a thorough knowledge 3.1. Electrophoretic methods of protein separation
of the sample content before applying affinity-based approach
[265,266]. Polyacrylamide gel electrophoresis in SDS (SDS-PAGE) was
described for the first time in 1949 [276]. Separation is achieved
2.4.7. Solid-phase protein enrichment once the electric field is applied to a solution containing a protein
A novel chromatographic carrier has been developed to per- that has a net positive or negative charge. The protein migrates
form adsorption and purification of proteins [267]. Zeolite sur- at a rate that depends on its net charge, size and shape [277].
face interacts with proteins through chelation with Co2+ on Presently, one-dimensional separation is often used as a pre-
zeolite nanocrystals [268]. The most important advantage of fractionating technique in proteomic approach, because of its
this procedure is lack of the co-concentration of salts during the insufficient resolution [278,279].
enrichment process, and the peptides adsorbed on the surface of The traditional two-dimensional gel electrophoresis method
the zeolite can be directly analyzed by mass spectrometry. was introduced in 1974 [280] and is now one of the most
18 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31

commonly applied techniques in proteomics. This method is separate components present in the sample. Because of that,
based on orthogonal separation of proteins according to different this technique is best suitable for separation of charged com-
physicochemical principles. 2-DE enables separation of com- pounds. In contradiction to gel electrophoresis, in capillary
plex proteins mixtures in respect to their pI, molecular weight, electrophoresis the separation process is conducted inside a
solubility and relative abundance [115]. Depending on the gel fused-silica capillary of the internal diameter between 50 and
size and pH gradient, it can simultaneously resolve more than 100 ␮m. Interactions between ions in the electrolyte and charged
5000 proteins [281]. Moreover, 2-DE provides information on groups situated on the capillary walls are responsible for the
protein changes in their expression level, isoforms and post- phenomenon of electroosmothic flow (EOF), which drives elec-
translational modifications [282]. trolyte through capillary. The motion of compounds during the
Detection limit depends on the dye applied for visualiza- separation results from both electroosmothic flow and their indi-
tion of the proteins. The present detection limit achieves less vidual electrophoretic mobilities.
than 1 ng of protein [115]. Another advantage of 2-DE is cre- It should be noted here that term capillary electrophoresis
ation of protein-patterns (profiles), that can be examined using is often used instead of a more precise “Capillary Zone Elec-
image analysis software [283]. Protein patterns may be simpli- trophoresis” (CZE) which refers to a family of closely related
fied as a result of the most abundant proteins depletion methods separation techniques employing narrow capillaries. Within this
[221,284–286]. text the term CE will be used as a shorthand of CZE.
The success of any of the protein separation and purifica- Major features of capillary electrophoresis include extremely
tion techniques is largely dependent on protein solubilization high separation efficiency, sensitivity, short analysis time and
method [121,287,288]. A more “traditional” sample lysis pro- low sample consumption, as compared to capillary high-
cedure for 2-DE, involves cell or tissue disruption in the presence performance liquid chromatography (cHPLC). CE can be cou-
of high concentrations of urea, reducing agents and detergents. pled to the mass spectrometer either with ESI (on-line) or
The immobilized pH gradient strip (IPG strip) is then rehydrated MALDI (off-line or on-line) ion source, which makes it a fea-
with sample, and proteins are separated. Unfortunately, the opti- sible method to use in proteomics. Drawbacks include complex
mal lysis conditions for 2-DE, are not compatible with MS. procedures for preparation of capillaries, their coating and prepa-
Detection of the separated proteins is usually accomplished ration of necessary interfaces between CE and, e.g. mass spec-
by the use of a visible stain, whereas newer approaches apply trometers. Detailed description of these connections is beyond
the fluorescent dyes [289–295]. Some of the fluorescent dyes the scope of this review and a more detailed description is rec-
have been designed for detection of the post-translational mod- ommended, e.g. [302].
ifications [296,297]. Here we will focus on sample preparation prior to CE analy-
The major problem associated with 2-DE is the reproducibil- sis, as reported in literature. In a recent article Fliser at al. [303]
ity of samples separation. It is particularly crucial in comparative published results on CE–MS analysis of crude urine samples.
proteomics, where control sample and experimental one have Multiple peptide signals visible on 2D electrophoregram plots
to be compared. To address this problem, a new method, two- prove the capillary electrophoresis is capable of removing
dimensional difference gel electrophoresis (2-DIGE) has been light-, and highly mobile contaminants (like salts, present in
developed. This technology allows for separation of two sam- samples at high concentration) during the run. However, high
ples on the same gel simultaneously, due to labeling the proteins salt content causes a decrease in efficiency of separation, and
from different samples with different cyanine dyes prior to the desalting utilising reversed-phase [304] and/or anion-exchange
first dimension [1]. This approach removes gel-to-gel variabil- solid-phase extraction [305] is beneficial for the final outcome
ity, and is valuable in distinguishing differences in migration of the analysis.
due to pI or molecular mass [263,298,299]. For statistical pur- Whole-cell or tissue digests obtained in the bottom–up pro-
poses, 2-DIGE gels also utilize pooled standards to normalize teomics approach reflect problems similar to the mentioned
measurements of protein abundance across multiple gels in the above—complex mixtures of peptides in buffers with high salt
experiment [300]. concentration, and denaturing compounds, such as urea. Not sur-
Two-dimensional Blue Native/SDS gel electrophoresis (2D prisingly, the reported CE separations of cell-line [306] and body
BN/SDS-PAGE) merges IEF proteins in their native state with a fluid digests [307] employ simple sample pretreatment proto-
second denaturing dimension. This method of protein separation cols, limited to desalting with C18 solid-phase extraction and/or
was designed to study dynamics and interactions of membrane removal of insoluble fractions after centrifugation.
proteins [143,301]. Whereas CE separation of peptides and oligopeptides in
Although there is a diversity of emerging proteomic tech- crude samples is possible, analysis of intact protein mixtures
niques, there is still no appropriate method that can replace is more troublesome. In their recent study, Moini and Huang
electrophoretic approach. [308] applied CE for separation of E. coli cell lysate proteins. In
the initial studies, direct injection of lysate resulted in capillary
3.2. Capillary electrophoresis clogging caused by protein precipitation at low pH of acidic elec-
trolyte used. To overcome this problem, the lysate was divided
Capillary electrophoresis is one of the liquid-phase sepa- into basic and acidic fractions by precipitation with 0.2% acetic
ration techniques. Like other electrophoretic approaches (i.e. acid. Both fractions were analyzed separately—basic proteins
gel electrophoresis), it utilizes electrostatic forces to drive and at low pH conditions and acidic at high pH. Different capillary
 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31 19

coating was applied in both cases (APS (aminopropyltrime- are identified directly by means of tandem mass spectrometry. In
toxysilane) and cellulose, respectively). Even though, the protein this approach, liquid chromatography may be used for separation
mixture was found to be too complex for 1D separation, thus the of proteins prior to mass spectrometry analysis. In the bottom–up
procedure was repeated with lysate of E. coli ribosomes with approach, protein, or protein mixture is digested. Single- or
good results. multidimensional liquid chromatography coupled to mass spec-
Capillary isoelectric focusing (CIEF) comprises another trometry is then used for separation of peptide mixtures and
important technique in the capillary electrophoresis “family” identification of their compounds. In this chapter we will sum-
of methodologies. Similarly to isoelectric focusing, it utilizes marize exemplary sample preparation procedures that may be
mixture of ampholytes to create linear pH gradient in the useful in both top-down, and bottom-up proteomics strategies.
electric field, and thus separates compounds according to their
isoelectric point. Its main advantage of the technique is its abil- 3.3.1. Top–down proteomics
ity to preconcentrate compounds at their pI. However, because In an examplary top–down approach, Wang et al. [309]
application of ampholytes suppresses MS ionization, direct analyzed yeast cytosol proteins by reversed-phase liquid chro-
application of CIEF in proteomics is limited due to difficulties matography. Sample preparation included cell disruption by
in coupling to a mass spectrometer. Therefore, this technique sonification, followed by centrifugation and desalting using
is best suited as a sample preconcentration and preparation 5 kDa cut-off cellulose membrane. Protein solution was sub-
technique for further CE–MS or LC–MS analysis in the 2-DE jected to denaturation and reduction in 20 mM Tris, 8 M urea and
workflow. Detailed description of complex instrumentation for 0.1 M DTT followed by alkylation by iodoacetamide. Prior to
linking CIEF to MS, CE–MS or LC–MS is beyond the scope RP–LC–MS/MS analysis, the buffer was exchanged to 10 mM
of this article—more information may be obtained in the work Tris and 2 M urea using size-exclusion column. Desalting of
by Simpson and Smith [302]. proteins and buffer exchange by centrifugal ultrafiltration and
anion-exchange chromatography prior to RP separation was
3.3. Sample preparation for high-performance liquid employed by Li et al. during identification of human plasma
chromatography proteins [310].
Moritz et al. [311] shown the free-flow electrophoresis-RP-
Nowadays, high-performance liquid chromatography is an HPLC (FFE IEF-RP-HPLC) approach for the analysis of human
important separation technique in proteomics. It can easily be plasma proteins. In this approach, fractions obtained by FFE
coupled to mass spectrometry, which makes it a perfect tool IEF (Free-Flow Electrophoresis IsoElectricpoit Focusing) were
for separation of proteins and peptides directly prior to mass directly injected onto RP column. Ampholytes and buffers/salts
analysis. Compatibility of solvents used in the reversed-phase abundant in the IEF fraction were not retained by the column,
chromatographic separations makes this hyphenated technique and thus easily removed from the system.
most commonly used in the final stage of proteomics analy-
sis workflow. Other liquid chromatography subtypes, includ- 3.3.2. Bottom-up proteomics
ing size-exclusion, ion exchange and affinity separations are Although LC separation of proteins is increasingly common
commonly used during consecutive steps of sample prepara- as protein separation technique in the top–down proteomics,
tion, clean-up, enrichment and prefractionation. Most chromato- the 2D gel approach is still considered as basic proteomic strat-
graphic approaches are tolerant to moderate concentration of egy. Reversed-phase liquid chromatography coupled to a tandem
contaminants, such as weak buffers. In this part, we will sum- mass spectrometer, is well established and commonly used pro-
marize several examplary sample pretreatment approaches used cedure for identification of the in-gel digested proteins. Apart
prior to injection onto LC column. from optional drying in a vacuum centrifuge/lyophilization and
Firstly, it should be noted that samples injected onto solubilization in the mobile phase prior to injection onto LC
chromatographic column cannot contain insoluble particles column, this approach usually does not require any additional
or dispersed molecules that may cause column clogging and sample preparation steps. Recent results include identification
malfunction. Such contaminants are usually removed by of cancer cell-line proteins [312,313], or mitochondrial protein
centrifugation and/or sample filtration using spin-filters (45 ␮m complexes [314]. However, additional step of cleaning of protein
pores). In addition, samples should not contain buffers affecting digest by, e.g. RP SPE (Reversed Phase Solid Phase Extrac-
LC separation, e.g. samples injected onto column should not be tion) is recommended in some cases. In the work, published by
dissolved in buffer with higher eluting strength than of mobile Rappsilber et al. [315] the authors introduced disposable C18
phase. High concentration of detergents should be avoided RP extraction tips for use in conjuction with MALDI, nanoESI
in case of RP separation whereas samples injected on the and RP-LC–MS approaches. The rationale for application of
ion-exchange column should not contain high contraction of RP SPE prior to LC separation is removal of insoluble parti-
background salts and other ionic contaminants that might disturb cles, purification from salts and sample preconcentration by the
ionic equilibrium. Volatile buffers such as ammonium acetate elution in small volume of organic solvents, followed by vac-
or ammonium bicarbonate, are recommended in this case. uum centrifugation. For similar reasons, commercially available
Liquid chromatography may be used both in top–down and capillary chromatography systems include trapping precolumns,
bottom–up proteomics approaches. In the first case, protein sam- where the sample is purified, desalted and preconcentrated prior
ple is separated and then individual proteins (or simple mixtures) to injection onto capillary column.
20 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31

Strategies employing direct proteolysis of biological sam- might release remainings from the manufacturing process, thus
ples, apart from much higher complexity of peptide mixture, increasing background in the mass spectrometer. Another impor-
are very similar in contents to the gel-spot digests. In the recent tant problem might be adsorption of a minute amount of mate-
work, Sun et al. [316] compared 1D-SDS-PAGE followed by rial on the tube/pipette tip walls. It seems that the commonly
proteolytic digestion and 1D-LC–MS/MS approach with 1D- accepted solution to this problem is application of the siliconized
and 2D-LC–MS/MS, for direct analysis of human urinary pro- tubes of smaller volume (e.g. 250–500 ␮l), instead of 1.5 ml.
tein lysates. In all cases, sample preparation protocols were very Silicone produces several characteristic peaks in the mass spec-
similar. Protein content was extracted by acetone precipitation, trometer but the sample loss is lower. Moreover, sample should
followed by in-gel or in-solution reduction, alkylation and diges- be processed as soon as possible after freezing the purified mate-
tion. For both approaches, no further sample pretreatment was rial. In many cases, sample loss is so significant after 4–6 days
performed, except for lyophilization and dissolution in LC buffer of storage that components are not detectable (R. Ekman, R.
directly prior to use. Similar approach in the bottom–up analy- Persson, G. Karlsson, J. Silberring, personal communication).
sis of human serum proteins was employed by Li et al. [310].
In proteomics analysis of transcription factors bound proteins, 3.4.1. MALDI-MS
additional desalting of the digest by RP solid-phase extraction, MALDI mass spectrometry is one of the basic proteomic
prior to lyophilization was performed [317], because the high techniques and is extensively used for peptide mass fingerprint-
salt content of the digestion buffer may affect peptide binding to ing, due to its speed, sensitivity, accuracy, satisfactory tolerance
the SCX (strong cation exchange) column in first dimension of to impurities and ease of automation [323]. In order to obtain
2D-LC run. The same approach was employed by Lominadze a peptide map, sample proteins are separated by 2D-PAGE and
et al. [318] in analysis of human neurophil granules’ proteins. visualized by staining. Coomassie Brilliant Blue stain is robust
Ramstrom et al. [319] analyzed human cerebrospinal fluid tryp- and MALDI-compatible, although cannot be used for detection
tic digest using RP-LC coupled to FT-ICR (Fourier-transform of the low-abundant proteins (ca. 100 ng detection limit). Sil-
Ion Cyclotron Resonance) mass spectrometry. In this work, the ver staining is much more sensitive (low nanograms) but, until
digest was desalted using commercially available RP solid-phase recently, was not suitable for MS due to the presence of glu-
extraction. Organic solvents after SPE were removed by vacuum taraldehyde [324]. This was overcome by development of MS-
centrifugation. compatible silver staining methods [325]. Fluorescent labels,
such as SYPRO Ruby can also be applied, that exhibit sensitiv-
3.4. Mass spectrometry ity similar to silver staining, have broad linear dynamic range
and are fully compatible with MALDI-MS. Further information,
For most proteomic applications, mass spectrometry is the including detailed protocols on the use of various staining meth-
ultimate phase of the analytical process, and is supposed to pro- ods can be found elsewhere [326–328]. After electrophoretic
vide the reliable end-data. Their quality, however, is directly separation, proteins are subjected to in-gel proteolytic digestion
dependent on the quality of the input from all earlier sample in order to obtain peptide maps that are further analyzed by
preparation/processing steps. In general, samples entering the MALDI-MS. Digestion of proteins directly on MALDI target
MS should be of the highest possible purity, not too complex, plates was also reported [329,330].
and deprived of compounds that compete with the analyte for For successful MALDI analysis, the key point is the proper
ionization or cause signal suppression, such as inorganic salts, choice of matrix and sample deposition method, to achieve high-
chaotropic agents, detergents, polymers and non-volatile com- est possible sensitivity and accuracy. Matrix purity should be
ponents. Matrix-assisted laser desorption/ionization (MALDI) of at least 99%, otherwise re-crystallization is recommended as
and electrospray ionization techniques are most useful for pro- impurities may negatively affect the formation of analyte/matrix
teomic studies (for a description of MS instrumentation and crystals. To minimize the risk of cross-contamination, applica-
application in proteomics see, e.g. [320,321]. MALDI is gen- tion of disposable (and prespotted with matrix) targets may be
erally more salt-tolerant than electrospray and a useful list of considered, instead of the multiple-use stainless-steel devices.
accepted contaminants may be found on the Internet [322]. The most popular matrices for proteomic applications include
Regardless of ionization technique, all operations must be per- ␣-cyano-4-hydroxycinnamic acid (CHCA) [331], sinapinic acid
formed with the highest cleanliness to avoid introduction of (SA) [332] and 2,5-dihydrobenzoic acid (DHB) [333]. Typically,
keratins or other contaminants that impair protein identifica- CHCA is preferred for analysis of peptide maps, SA works best
tion, and the reagents should be of highest possible purity to for larger proteins and DHB is usually used for hydrophobic,
avoid high background, obscuring spectra quality. Care should glyco- and phosphopeptides, but these are only general guide-
be taken to avoid material loss during preparatory steps. How- lines. For some applications, combinations of different matrices
ever, depending on the type of experiment and its ultimate goal, were found useful [334,335].
samples may be prefractionated in order to limit the dynamic A vast number of sample preparation protocols were devised
range of concentrations to unmask low-level components (see that differ in the order of application of matrix and sample solu-
Sections 2.3 and 2.4). tion, their concentrations and solvents used. The most popular
Application of the proper quality of assay tubes (and plas- include the “dried droplet” method [336], in which the matrix
tics in general) cannot be neglected. There is a broad dis- solution is mixed with the sample, and the resulting mixture is
cussion among laboratories concerning this issue as the tubes deposited on the target plate. In the “thin layer” method [337],
 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31 21

a drop of analyte solution is deposited on a matrix-covered tar- to ensure maximum concentration. Therefore, in order to max-
get surface, thereby improving accuracy and sensitivity. Useful imize their final concentration on the target plate, analytes may
tips on the choice of matrix and sample preparation techniques be eluted with a matrix-containing solvent [340,349]. Although
were given by Kussmann et al. [330], and a detailed study on each transfer of low-concentrated samples between tubes and
the conditions affecting the efficiency of CHCA can be found columns may lead to significant material loss [348], the bene-
in [338]. Depending on the protocol, the matrix may be acidi- fits of thorough purification might well outweigh the risks. The
fied with trifluoroacetic acid to facilitate protonation. Generally, use of chromatographic beads immersed in the analyzed solu-
there is no universal sample preparation protocol that could tion was reported as well. The beads were then transferred to the
be used for each and every type of sample, and in more dif- MALDI plate, where the bound material was eluted and analyzed
ficult cases (low analyte and high salt concentration) careful [350]. A sophisticated and efficient means of sample purifica-
optimization might be necessary if standard procedures fail tion and separation is offered by the LC–MALDI technology
[330]. that combines the advantages of capillary liquid chromatogra-
Although it may be possible to record spectra of native bio- phy (LC) with the sensitivity and accuracy of MALDI-MS.
logical samples, prior to analysis they might be desalted, in In high-throughput screening (HTS) studies, automated
order to avoid signal suppression by high salt contamination. machines are used for sample preparation and deposition on tar-
For sinapinic acid or CHCA it is possible to decrease salt con- get plates that offer unparalleled precision in low-volume oper-
centration by on-target washing, i.e. after the sample and matrix ations, speed and elimination of human errors [351–357]. These
were deposited on the target plate and co-crystallized. It is a devices make possible the efficient application of LC–MALDI
cheap and simple method that takes advantage of the fact that, for confident protein identification and, most importantly, are
unlike inorganic ions, the crystallized matrix with incorporated able to operate on picoliter sample volumes. Some of the robotic
peptides or proteins is water-insoluble. Here a ca. 5 ␮l droplet of platforms enable automatic completion of other stages of the
deionized water, 0.1% TFA or 5% formic acid, is deposited on experiment as well, such as protein digestion [358–360], or chro-
the surface of a sample dried on the target plate. The droplet is matographic separation [357,361].
removed after several seconds and the procedure can be repeated Apart from typical stainless-steel MALDI target plates, pre-
[339]. In particular, the thin-layer method offers the possibility of structured targets are available that include hydrophilic anchors
effective on-target washing since the analyte and impurities are distributed over a hydrophobic surface in order to concen-
localized on top of the matrix layer. The use of DHB excludes trate the analyte on the spot centre, thus increasing sen-
washing as this matrix is water-soluble. The efficiency of on- sitivity. For prestructured targets, dedicated protocols were
target washing is dependent on the type of contaminants, their developed [362–364]. Recently, disposable matrix-precoated
solubility and accessibility for wash solution [340]. It should be AnchorChipTM (Bruker Daltonik, Germany) targets with cal-
noted that some analyte molecules can also be washed away. A ibration spots were introduced [365]. Their major advantages
comparison of different protocols can be found in [341]. The include the ease of automated sample spotting, elimination of
addition of nitrocellulose to the matrix was found to strengthen cross-contamination risk, and possibility of several months’ long
binding to target surface and to make it more resistant against storage. Various in-house methods of modifications of the target
contaminants, thus allowing more effective washing [342]. Mod- surface were also reported [343,366–369] and reviewed recently
ifications of the target plate surface are also possible to facilitate [370].
on-target purification [343]. Furthermore, the atmospheric pressure MALDI (AP-
If on-target washing gives unsatisfactory results, efficient salt MALDI) was introduced in 2000 [371]. Although its applica-
removal may be achieved through dialysis or the use of commer- tions in complex proteomic studies are still limited, it should
cially available or home-made chromatographic microcolumns be noted here that for AP-MALDI typical matrices may be
(see also Section 2.3). Desalting is often performed together with used, such as CHCA or SA. However, the use of liquid matri-
concentration, as is the case with reverse-phase microcolumns. ces, including water as matrix was also reported in combination
In popular Millipore ZipTips, the stationary phase is placed in with an infrared laser [372,373]. This gives the possibility of the
the outlet of a pipette tip, and elution occurs in the microliter on-line combination of liquid chromatography and AP-MALDI
volumes [344]. Their convenience stems from the ease of oper- [374,375].
ation, as washing and elution are performed by simply plunging
and releasing the pipette plunger. Therefore, they are compatible 3.4.2. SELDI
with many robotic platforms for proteomics. Chromatographic As a conceptual modification of MALDI-TOF measure-
microcolumns can also be prepared in-house, using gel loading ments, surface-enhanced laser desorption/ionization technique
pipette tips, fused silica capillaries or small columns equipped is marketed by Ciphergen that combines chromatographic sepa-
with frits and filled with beads of different stationary phases ration and mass spectral measurement for proteomic profiling
[330,345–347]. Independent on the shape, microcolumns offer and biomarker discovery (see also Section 1.4). The SELDI
the possibility of stepwise elution of bound components, which chip contains chromatographic coating of selected type (i.e.
lowers sample complexity. A useful comparison of experimen- hydrophobic, ion-exchange, metal-binding, etc.), on which sam-
tal conditions for the use of RP microcolumns (bed volume, ple components of a given type are captured. Unbound com-
solvents, etc.) can be found in [348]. In general, best results are pounds are washed off, thus contaminants are removed and
obtained if the sample is eluted in the lowest possible volume sample complexity is markedly reduced. After application of
22 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31

a proper energy-absorbing matrix, such as CHCA or SA, pro- ical organic solvents used in RP separations include methanol
teins bound to stationary phase are analyzed for MS profil- or acetonitrile with an ion-pairing additive to strengthen the
ing. The company also released chips with covalently bound analyte-stationary phase interaction. A list of various ion-pairing
matrices. Thus, background in the low m/z range is elimi- agents and their extensive characteristics, including pros and
nated and small molecules can be successfully analyzed by this cons of the use of most popular TFA, can be found in [384].
technique. Based on a sound literature overview, the author concluded that,
The great advantage of SELDI lies in its ability to remove salts despite its drawbacks, TFA is probably the best ion-pairing agent
and other impurities prior to MS analysis, thanks to which crude available. On the other hand, TFA may cause significant signal
sample can be analyzed, such as urine [376,377], cerebrospinal suppression in the mass spectrometer, thus decreasing sensitivity
fluid [378,379], serum [380,381], etc. However, sample prepara- of measurements.
tion steps such as denaturation or depletion of high-abundance Direct coupling of ESI-MS with other types of chromatog-
proteins may be considered [382] (compare Section 2.3). For raphy (i.e. ion exchange, size-exclusion, affinity, etc.) is less
reviews on the application of SELDI in proteomic profiling, see frequently used [384], since they require non-volatile inorganic
[383]. buffers. Nevertheless, these types of chromatography are often
used in biochemical purifications (using volatile buffers that
3.4.3. Electrospray ionization overcome such problems) as they preserve the function of pro-
Electrospray (ESI) is a “soft” method of ionization, which teins, therefore, in order to facilitate ESI-MS analysis of eluents,
rarely promotes spontaneous fragmentation of analytes. How- on-line desalting methods were developed based on microdial-
ever, electrospray is often coupled with ion trap or quadrupole ysis [385], or other principles [186,386].
analysers as this combination enables efficient peptide sequenc- For the efficient analysis of highly complex proteomic sam-
ing by induced fragmentation (MS/MS). This makes ESI well ples, non-ESI-compatible chromatographic techniques may be
suited for peptide identification in complex mixtures, such as combined with RP separation and electrospray detection. In
protein digests. In an ESI ion source the ionization process particular, strong cation exchange (SCX) chromatography is
occurs under atmospheric pressure and the sample is introduced usually used as a part of two-dimensional systems, where sam-
as liquid, therefore electrospray may be coupled directly to liq- ple components are first separated on an SCX column and then
uid chromatography systems. These features make ESI a useful transferred to the RP column. In this setup, although increasing
tool in proteomics. salt concentrations are used to elute fractions from the SCX
As already mentioned, the ESI ion source is sensitive to column, the analytes are then trapped on the RP stationary
inorganic salts that cause signal suppression and adduct forma- phase and eluted in an MS-friendly water/organic eluent system
tion, lowering sensitivity and signal-to-noise ratio. Additionally, directly to the ion source. If complex protein digests applied
some inorganic salts, such as phosphates, could precipitate in to the first dimension already contain concentrated buffers that
the heated capillary of the ESI source leading to its permanent could interfere with binding to the SCX column, the use of an
damage. Certainly, any solid particles (also from the bleeding additional first-in-line RP precolumn might be considered for
chromatographic columns!) in the analyzed samples must be effective desalting. On-line hyphenation of 2D-LC system with
strictly avoided. In proteomic applications, the positive ion mode ESI-MS, although still technically challenging, is the basis of
is mostly used, in which ionization is based on protonation of the multidimensional protein identification technology (Mud-
the analyte molecules, therefore the sample is often acidified, PIT) [387] and is more and more routinely applied to complex
e.g. with formic or trifluoroacetic (TFA) acids, to facilitate ion proteomic projects (see [388,389] for methodological reviews
formation. and [390–393] for recent applications). More detailed guide-
Desalting is the main stage of sample preparation for ESI lines for sample preparation prior to LC separation can be found
measurement (a list of accepted salt concentrations can be in Section 3.2.
found in [322]). For direct sample infusion, in which sample Very recently, a new ionization technique was devised termed
is pumped into the ion source through a syringe pump, typi- desorption electrospray ionization (DESI). Here the charged
cal desalting strategies may be used, as described in Section droplets of solvent are sprayed onto the analyzed object, so
2.3. Nevertheless, such high-volume approach is rarely used in that molecules present on its surface are ionized [394]. DESI
proteomics where only minute amounts of sample are typically can be applied to solids, liquids (including complex biologi-
available. cal samples) and adsorbed gases. In proteomics, proteins and
In many, if not most, ESI-MS-based proteomic studies, the protein complexes can be detected, and peptide maps can be
spectrometer is coupled on-line to a chromatographic system, analyzed including MS/MS fragmentation. Importantly, DESI
where contaminants are efficiently removed. Chromatographic apparently does not require sophisticated sample pretreatment
separation also lowers sample complexity, which makes the anal- and tissue sections could be analyzed directly. For liquids, such
ysis more sensitive for low-level components. In such case, as urine or plasma, several microliters are deposited on an appro-
however, final eluents must be chosen carefully so that they priate surface and left to dry before analysis. The applications
do not interfere with ionization. Usually sample components of DESI cover, among others, clinical diagnostics, especially in
are introduced into the ion source after they are eluted from dermatology, microorganism characterization and MS imaging.
the reversed-phase (RP) column, in which water/organic solvent Up-to-date reviews of this promising technique can be found in
systems are used that are easily accepted by the ESI source. Typ- [395,396].
 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31 23

3.5. Quantitative proteomics are pooled and proteolyzed. Derivatized tryptic peptides are then
isolated from by affinity chromatography thanks to the presence
Based on ESI ionization, quantitative proteomic analyses are of biotin moiety in the ICAT reagent molecule, and analyzed by
possible to measure relative abundances of peptides or proteins LC–MS. However, biotin might obscure the analysis by shifting
in samples taken from biological fluids, cells or organisms in peptide masses to a higher mass range and by generation of addi-
different states, e.g. controls versus diseased. Typically, compo- tional fragments in MS/MS spectra. In an updated version of the
nents of one sample are labeled with the light reagent of standard ICAT experiment, the labeling reagent was immobilized on solid
isotopic composition, while components of the other sample are support by a photocleavable linker so that modified peptides
modified with the heavy reagent, enriched in stable heavy iso- could be more easily separated by filtration and bond cleavage
topes of a selected element, e.g. containing 2 H instead of 1 H [409]. A drawback of all proton/deuterium containing labels
atoms. Labeled samples are pooled, digested, subjected to addi- stems from differences in their behavior in RP chromatography,
tional prefractionation if needed (the order of operations may as deuterated species elute a bit earlier than corresponding ones,
differ between methods), and analyzed by LC–MS. In the spectra containing only 1 H atoms [410]. These problems were addressed
pairs of peaks are observed whose m/z values differ proportion- by the introduction of 12 C/13 C labeling system in which, addi-
ally to the molecular weights of the labels, and whose intensities tionally, the biotin tag may be easily removed from the labeled
correspond to relative abundances of the peptides/proteins. For peptides in acidic environment after they have been isolated by
in-depth description of quantitative proteomics and its methods, affinity chromatography (cleavable ICAT), in order to lower the
see [397–400]. overall size of the tag prior to MS analysis and improve fragmen-
From the sample preparation point of view, the labels may be tation efficiency. This approach proved effective in the analysis
introduced to the sample in vivo during cell growth (metabolic of complex proteomes [411]. What is more, the synthesis of
labeling, SILAC—see below) or after proteins have been iso- further types of ICAT reagents was reported [412,413].
lated from the biological material (enzymatic labeling, ICAT, Recently, the isobaric tags for relative an absolute quantita-
iTRAQ). Potential errors in quantification may result from dif- tion (iTRAQ) were developed by which four samples can be
ferences in preparation of samples to be compared, e.g. during analyzed simultaneously [414]. This allows for analyzing sam-
protein isolation from crude biological material, their concen- ple composition, e.g. in different time points or in duplicates.
tration or fractionation. These errors can be minimized if the Proteolytic peptides are labeled with amine-specific isobaric
number of operations is limited, or if as many operations as tags which, upon induced fragmentation in the mass spectrome-
possible are conducted on pooled samples. ter, yield different reporter ions in 114–117 m/z range. Reporter
In one approach of in vivo labeling, compared cell lines are ions have different isotopic compositions and hence molecular
grown on media containing only either 14 N or 15 N isotopes weights but thanks to balance groups the tags are isobaric and no
[401]. The function of the isotope labels may also be served by shift in mass spectra is observed for different ions, which simpli-
modified amino acids (stable isotope labeling by amino acids in fies the analysis, including the MS/MS measurement. Although
cell culture, SILAC). In the latter and more popular approach, this technique is relatively new, it was applied in several inter-
one population of cells is grown on a standard medium, con- esting projects [415–418]. An informative recent overview can
taining all necessary amino acids in their typical isotopic com- be found in [419].
positions, while the other population is grown on a medium An absolute quantification (AQUA) strategy was also devel-
containing a selected amino acid labeled with heavy atoms, e.g. oped, in which a synthetic standard peptide is introduced into the
Leu-d3 instead of Leu-d0 [402] but other amino acids can also cell lysates at known concentration [420]. The peptide contains
be used [403,404]. Simultaneous analysis of three distinct states stable heavy isotopes and mimics a tryptic peptide present in the
by differentially labeled arginine derivatives was also performed protein of interest. The selected reaction monitoring (SRM) mea-
[405], and the heavy metal SILAC approach was used for iden- surement during MS analysis allows detection and quantitative
tification and quantitation of methylation sites [406]. In both assessment of the native peptide, compared to the isotopically
these in vivo methods protein isolation, proteolytic cleavage and enriched standard.
chromatographic separations are performed on pooled samples,
which helps minimize errors. 3.6. Imaging mass spectrometry
Among in vitro labeling methods, probably the most straight-
forward and universal is enzymatic labeling, where one of Imaging mass spectrometry is a new tool for revealing the
the samples to be compared is proteolyzed in the isotopically spatial distribution and relative concentration of compounds in
enriched H2 18 O instead of usual water [407]. This approach does biological samples such as tissue sections. It seems to be a valu-
not require any special sample preparation compared to routine able method in comparative studies, where profiles and images of
proteomic strategies but has some limitations as the mass of pep- tissue sections in different stages could be matched. It enables to
tide increases. Lower resolution at higher m/z does not allow for find the differences in protein/peptide expression between those
a satisfactory separation of the peak envelope. stages, without the necessity of knowing which proteins have
On the other hand, popular isotope-coded affinity tag (ICAT) changed. Because of that, this technique seems to be very useful
technique is a multistage process, in which the heavy reagent for biomarker discovery [421].
contains eight deuterium atoms instead of eight 1 H atoms and Development of MS imaging was possible only due to advan-
labeling occurs via a thiol-reactive group [408]. Labeled samples tages of mass spectrometers with MALDI ion source [422,423]
24 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31

or TOF SIMS (Secondary Ion Mass Spectrometry) [424] for matrix deposition, the blotted area should be rinsed with water, to
peptides and smaller molecules. It allows for soft ionization of remove tissue fragments, cell debris, blood, and salts. Caution
the components of the sample, which means that the molecules must be taken here, because more hydrophobic proteins and
which possessed relatively high masses, like proteins (but also peptides that are weakly bound to the membrane could be lost
peptides), could be observed. Moreover, the ionization is usually [430].
not multicharge, as ESI source, so relatively simple spectra of Apart from blotting, MS imaging could be performed directly
complex mixtures can be obtained. Another advantage of this on thin tissue section, which eliminates the problems associated
technique is its great sensitivity, which means that even proteins with blotting procedures. Thin tissue sections are obtained from
present in the femtomole range could be detected. the snap-frozen tissue samples using a cryostat. The thickness
In MS imaging the thin section of tissue, or the membrane of the slices is not critical and can be adjusted to assure easy
with direct tissue blotting, is placed on a MALDI target plate, handling. Typically, for mammalian tissue, the thickness should
then the matrix is applied on the surface. Inside the spectrometer, be about 10–20 ␮m, i.e. in the order of mammalian cell diameter,
laser beam causes the ionization of proteins, which are present so the majority of the cells are cut open, exposing intracellular
exactly in this particular section, and in the place where the contents. The temperature of the sample during cutting should
laser beam is focused. For every such point a spectrum could be be kept between −5 and −25 ◦ C. The exact value depends on the
obtained. The mass to charge ratio can range from 1000 to more tissue type (for example, fatty tissues require lower temperature
than 100,000. Using this method, a profile of a given spatial to obtain high-quality sections). Usually, slicing is performed at
point on the surface of the tissue section could be generated −15 ◦ C and slices are 12 ␮m thick.
and then, from the intensity of a given m/z value monitored in In traditional tissue sectioning before cutting, the tissue is
each spectrum, a density map or image could be constructed. embedded in the optimum cutting temperature polymer (OCT)
Virtually, all signals from the section could serve to generate or agar, which stabilizes the tissue and provides its smooth sur-
a specific density map for this section. Proteins of interest can face. It is important to avoid cutting the OTC with the cryostat
be then identified based on its peptide map, after isolation and blade, as this could leave a thin film of the polymer on the top
trypsin digestion. of the section, which may cause poor MALDI-MS analysis.
All those steps lead to create specific molecular image of Tissue slices are then thaw-mounted on the target plate. The
the tissue, which provides useful information about local pro- preferred way to do it is to put the tissue section on the cold
teomic/peptidomic composition, relative abundance and spatial target plate and then quickly warm them together. With this
distribution of the components in the tissue. A more detailed method there is no loss of water-soluble proteins. It is important
description of the method can be found elsewhere [425–427]. to transfer the sample very carefully to the target plate to preserve
Sample preparation is a crucial step in MS imaging. Espe- its native shape. Slices are then allowed to dry in a vacuum
cially, it is important to maintain the integrity of the spatial desiccator for 1 h [421,430,431].
arrangements of all compounds. Any mistake done here may Cells obtained by LCM may also be the source of sample for
cause delocalization and degradation of the analytes. At the imaging MS analysis. Following microdissection, polymer with
beginning, it is essential to surgically remove the tissue samples adhered cells may be placed directly on a target MALDI plate
very carefully to retain its native shape. Fresh tissue samples [12].
must be frozen immediately after dissection. Usually liquid As already mentioned, to obtain the spectra, matrix must co-
nitrogen is used here. The sample, loosely wrapped in aluminum crystallize with the sample, so it has to be deposited on the
foil should be gently lowered into the liquid gas over a period surface of the tissue section, or blotted area. Three main condi-
of 30–60 s. The foil helps to stabilize more fragile fragments tions have to be fulfilled to obtain the high-quality images. First,
of tissue and to protect it against the adhesion to the container the process of covering the surface of the sample with the matrix
walls. Sample prepared in this way may be stored at −80 ◦ C cannot change the native localization of proteins. Second, matrix
until analysis. solution has to wet the tissue surface in order to form crystals
As it was mentioned before, in the MS imaging experiment, with the proteins. Last, the size of crystals must be smaller than
tissue sections may be examined directly or a protein blot may be the image resolution [421].
analyzed. Material obtained from the laser capture microdissec- There are two ways in which matrix can be placed on the sam-
tion experiment could also be used [12]. The blotting procedure ple surface: “drop deposition” method, and covering the surface
is a quick and easy way to generate global protein profiles from with using a glass spray nebulizer. The simplest way is to deposit
a given tissue and the obtained protein profiles seem to be repro- a drop of matrix using an automatic pipette, at a given coordinate
ducible between animals of the same strain [428]. Proteins can of the section. Typically, about 100–200 nl of matrix is used.
be transferred from the fresh tissue by contact blotting on an In the glass spray nebulizer, about 500 ␮l of matrix solu-
active surface, such as C18 [429], or a carbon-filled polyethy- tion should be sprayed on the section at the distance of about
lene membrane [430]. 15–30 cm. Then the surface is allowed to dry at room temper-
During this procedure, the membrane is wetted in methanol ature. This spray cycle should be repeated up to 10 times with
for 30 s and then attached to the target plate by a conductive air-drying in between, to obtain homogenous crystal field. It
tape. Fresh cut tissue section is placed on the membrane and is important to remember that excessive wetting of the sample
contact blotting is performed for 5 min. During this time, the should be avoided in order to protect proteins from delocaliza-
tissue should be covered by a glass slide to avoid drying. Before tion.
 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31 25

Both methods give spectra of comparable quality but translational modifications, as they play vital role in signal
when high-resolution image is demanded, homogenous coating transduction. The goal is no longer proteins identification but a
obtained by glass spray nebulizer is preferred. better recognition and understanding of global processes, occur-
Sinapinic acid (3,5-dimethoxy-4-hydroxycinnamic acid, SA) ring within cells. Preparation of the samples dedicated for such
(saturated solution in a solution of 50/50/0.1 acetonitrile, water purpose should also be mild and suitable for extraction of non-
and trifluoroacetic acid, v/v/v) seems to be the best matrix for covalent complexes between proteins, peptides, nucleic acids
higher molecular weight proteins. Replacement the acetonitrile and metabolic products. Application of activated surfaces, e.g.
with ethanol also gives spectra of high quality [432]. SELDI concept, might also be a future direction for proteins
Samples are usually analyzed in the linear mode and mass prepurification directly on a MALDI plate. Two-dimensional
spectra are usually at an average of 250–1000 laser shots aimed gel electrophoresis might soon be replaced or at least limited,
randomly at different positions across the spot. Usually, internal by a combination of other methods such as isoelectrofocusing in
calibration is performed by the addition of proteins of known solution, followed by, e.g. one-dimensional PAGE and capillary
molecular weight. LC–MS/MS. This approach might gain better reproducibility
Signals correspond to proteins and peptides found in given and much better dynamic range of proteins to be identified. In
tissue. Mass to charge ratio is in a range from 2 to 100 kDa general, a number of optimized procedures are necessary to com-
and signals represent a wide range of intensities within three pare data between laboratories, and to gain good reproducibility.
orders of magnitude. To generate a map of signals, the section Another question is availability of the highly advanced instru-
surface is covered by grid and each spectrum is taken at each grid mentation and expertise, often not available in the university
coordinate with image resolution limited to the laser spot size, laboratories. Therefore, it might be advisable to establish well-
which, for commercially available instruments, is about 50 ␮m equipped core facilities while samples preparation and presep-
in diameter. aration procedures will be performed by the end-users. As the
As examples of application of this new proteomics technique, direction of proteomics switches from simple identification of
we could mention imaging of the rat brain [421] and search- proteins to differential proteomics and functional studies of pos-
ing for tumor-specific biomarkers in colon and prostate cancers sible disease markers, there is no need for the end-user to keep
[433]. This kind of study might have a great importance for clin- equipment and staff to identify protein profiles just for one par-
icians because it could permit molecular assessment of tumor ticular project.
biopsies, with the potential to identify subpopulations of cells
that are not based on the cellular phenotype determined micro- Acknowledgement
scopically. In the future, assessment of surgical margins could be
possible due to molecular assessment using MS imaging [434]. This work was sponsored by a grant from the International
Centre for Genetic Engineering and Biotechnology, Trieste, Italy
4. Conclusions and future prospects (No. CRP/POL05-02).

The results published during recent years suggest that pro- References
teomics is still in its infancy. There is no standardized strategy
for samples preparation, proteins purification and separation. [1] W.M. Freeman, S.E. Hemby, Neurochem. Res. 29 (6) (2004) 1065.
Moreover, various mass spectrometers are used for proteins [2] S.D. Patterson, Biotechniques 35 (3) (2003) 440.
identification, and each instrument and laboratory produces its [3] C. Janus, D. Westaway, Physiol. Behav. 73 (5) (2001) 873.
own set of more or less specific data. Despite the methodolog- [4] L. Paulson, R. Persson, G. Karlsson, J. Silberring, A. Bierczynska-
Krzysik, R. Ekman, A. Westman-Brinkmalm, J. Mass Spectrom. 40 (2)
ical problems, bioinformatics is still the Achilles tendon of the (2005) 202.
global analysis of proteins. In particular, a mixture of peptide [5] N. Ahmed, G.E. Rice, J. Chromatogr. B Anal. Technol. Biomed. Life Sci.
maps that belongs to several proteins cannot be identified using 815 (1–2) (2005) 39.
presently existing tools. [6] T.H. Steinberg, C.R. Hart, W.F. Patton, in: J.M. Walker (Ed.), The Pro-
Future prospects will keep in focus systems biology where teomics Protocols Handbook, Humana Press, Totowa, 2005, p. 31.
[7] M.R. Hynd, J.M. Lewohl, H.L. Scott, P.R. Dodd, Neurochemistry 85 (3)
the entire networks are identified for faster and more effi- (2003) 543.
cient diagnostics dedicated for individual patients (personalized [8] C. Melle, D. Osterloh, G. Ernst, B. Schimmel, A. Bleul, F. von Eggeling,
medicine). Int. J. Mol. Med. 16 (1) (2005) 11.
The best strategy for the future would be NO SAMPLE [9] R. Kikkawa, M. Fujikawa, T. Yamamoto, Y. Hamada, H. Yamada, I. Horii,
PREPARATION at all, but today instrumentation (mass spec- J. Toxicol. Sci. 31 (1) (2006) 23.
[10] R.I. Freshney, Culture of Animal Cells. A Manual of Basic Technique,
trometers) do not allow for direct identification/quantitation of Wiley-Liss, New York, 1990.
hundreds of proteins present in complex biological mixtures. [11] F.A. Witzmann, J.W. Clack, K. Geiss, S. Hussain, M.J. Juhl, C.M. Rice,
Until then, a high-throughput strategy must be involved, with an C. Wang, Electrophoresis 23 (14) (2002) 2223.
increasing number of microchip technologies (lab-on-the-chip), [12] B.J. Xu, R.M. Caprioli, M.E. Sanders, R.A. Jensen, J. Am. Soc. Mass.
simultaneously and automatically preseparating a large number Spectrom. 13 (11) (2002) 1292.
[13] M.A. Tangrea, B.S. Wallis, J.W. Gillespie, G. Gannot, M.R. Emmert-
of components. Such process will certainly be more focused Buck, R.F. Chuagui, Expert Rev. Proteomics 1 (2) (2004) 185.
on isolation of organelles, dedicated proteomics/peptidomics [14] T.H. Steinberg, C.R. Hart, W.F. Patton, in: J.M. Walker (Ed.), The Pro-
such as phosphopeptides, glycated proteins and other post- teomics Protocols Handbook, Humana Press, Totowa, 2005, p. 59.
26 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31

[15] G.L. Hortin, S.A. Jortani, J.C. Ritchie Jr., R. Valdes Jr., D.W. Chan, Clin. [50] H. Xie, N.L. Rhodus, R.J. Griffin, J.V. Carlis, T.J. Griffin, Mol. Cell.
Chem. 52 (7) (2006) 1218. Proteomics 4 (11) (2005) 1826.
[16] D.E. Terry, D. Desiderio, Proteomics 3 (2003) 1962. [51] R.R. Drake, L.H. Cazare, O.J. Semmes, J.T. Wadsworth, Expert Rev. Mol.
[17] M. West-Nielsen, E.V. Hogdall, E. Marchiori, C.K. Hogdall, C. Schou, Diagn. 5 (1) (2005) 93.
N.H. Heegaard, Anal. Chem. 77 (2005) 5114. [52] C. Streckfus, L. Bigler, Adv. Dent. Res. 18 (1) (2005) 17.
[18] G.S. Omenn, et al., Proteomics 5 (2005) 3226. [53] S. Koka, M.D. Forde, S. Khosla, Int. J. Prosthodont. 19 (1) (2006) 53.
[19] N. Zolotarjova, J. Martosella, G. Nicol, J. Bailey, B.E. Boyes, W.C. Bar- [54] R.L. Goldenberg, A.R. Goepfert, P.S. Ramsey, Am. J. Obstet. Gynecol.
rett, Proteomics 5 (2005) 3304. 192 (5 Suppl.) (2005) 36.
[20] N.I. Govorukhina, T.H. Reijmers, S.O. Nyangoma, A.G. van der Zee, [55] P.S. Ramsey, W.W. Andrews, Clin. Perinatol. 30 (4) (2003) 701.
R.C. Jansen, R. Bischoff, J. Chromatogr. A 1120 (1–2) (2006) 142. [56] D. Wessels, D.B. Barr, P. Mendola, Environ. Health Perspect. 111 (16)
[21] K. Junker, F. von Eggeling, J. Muller, T. Steiner, J. Schubert, Urologe A. (2003) 1939.
45 (3) (2006) 305. [57] D.J. Crouch, Forensic Sci. Int. 150 (2–3) (2005) 165.
[22] F. Simpkins, J.A. Czechowicz, L. Liotta, E.C. Kohn, Pharmacogenomics [58] A.G. Verstraete, Ther. Drug Monit. 26 (2) (2004) 200.
6 (2005) 647. [59] M. Taba Jr., J. Kinney, A.S. Kim, W.V. Giannobile, Dent. Clin. N. Am.
[23] L. Guerrier, V. Thulasiraman, A. Castagna, F. Fortis, S. Lin, L. Lomas, 49 (3) (2005) 551.
P.G. Righetti, E. Boschetti, J. Chromatogr. B. Anal. Technol. Biomed. [60] N. Ozmeric, Clin. Chim. Acta 343 (1–2) (2004) 1.
Life Sci. 833 (2006) 33. [61] H.C. Gerard, Z. Wang, G.F. Wang, H. El-Gabalawy, R. Goldbach-Mansky,
[24] M.M. Elrick, J.L. Walgren, M.D. Mitchell, D.C. Thompson, Basic Clin. Y. Li, W. Majeed, H. Zhang, N. Ngai, A.P. Hudson, H.R. Schumacher,
Pharmacol. Toxicol. 98 (2006) 432. Arthritis Rheum. 44 (2001) 1689.
[25] N. Ahmed, K.T. Oliva, G. Barker, P. Hoffmann, S. Reeve, I.A. Smith, [62] L. Newman, C. Rose, L. Maier, N. Engl. J. Med. 336 (1997) 1224.
M.A. Quinn, G.E. Rice, Proteomics 5 (2005) 4625. [63] B. Bresnihan, Arthritis Res. Ther. 5 (6) (2003) 271.
[26] Y. Shen, J. Kim, E.F. Strittmatter, J.M. Jacobs, D.G. Camp 2nd, R. Fang, [64] N. Islam, M. Lonsdale, N.M. Upadhyaya, T.J. Higgins, H. Hirano, R.
N. Tolie, R.J. Moore, R.D. Smith, Proteomics 5 (2005) 4034. Akhurst, Proteomics 4 (2004) 1903.
[27] A.K. Yocum, K. Yu, T. Oe, I.A. Blair, J. Proteome Res. 4 (2005) 1722. [65] G. Borderies, E. Jamet, C. Lafitte, M. Rossignol, A. Jauneau, G. Boudart,
[28] J. Granger, J. Siddiqui, S. Copeland, D. Remick, Proteomics 5 (2005) B. Monsarrat, M.T. Esquerré-Tugayé, A. Boudet, R. Pont-Lezica, Elec-
4713. trophoresis 24 (2003) 3421.
[29] I. Feuerstein, M. Rainer, K. Bernardo, G. Stecher, C.W. Huck, K. Kofler, [66] E. Jamet, H. Canut, G. Boudart, R.F. Pont-Lezica, Trends Plant Sci. 11
A. Pelzer, W. Horninger, H. Klocker, G. Bartsch, G.K. Bonn, J. Proteome (1) (2006) 33.
Res. 4 (2005) 2320. [67] G. Borderies, E. Jamet, C. Lafitte, M. Rossignol, A. Jauneau, G. Boudart,
[30] V.C. Wasinger, V.L. Locke, M.J. Raftery, M. Larance, D. Rothemund, A. B. Monsarrat, M.T. Esquerré-Tugayé, A. Boudet, R. Pont-Lezica, Elec-
Liew, I. Bate, M. Guilhaus, Proteomics 5 (2005) 3397. trophoresis 24 (2003) 3421.
[31] H.Y. Tang, N. Ali-Khan, L.A. Echan, N. Levenkova, J.J. Rux, D.W. Spe- [68] E. Jamet, H. Canut, G. Boudart, R.F. Pont-Lezica, Trends Plant Sci. 11
icher, Proteomics 5 (2005) 3329. (1) (2006) 33.
[32] A. Horn, S. Kreusch, R. Bublitz, H. Hoppe, G.A. Cumme, M. Schulze, [69] W. Wang, M. Scali, R. Vignani, A. Spadafora, E. Sensi, S. Mazzuca, M.
T. Moore, G. Ditze, H. Rhode, Proteomics 6 (2006) 559. Cresti, Electrophoresis 24 (2003) 2369.
[33] J. Villanueva, J. Philip, D. Entenberg, C. Chaparro, A.M.K. Tanwar, E.C. [70] J.K. Rose1, S. Bashir, J.J. Giovannoni, M.M. Jahn, R.S. Saravanan, Plant
Holland, P. Tempst, Anal. Chem. 76 (2004) 1560. J. 39 (2004) 715.
[34] J. Villanueva, J. Philip, C.A. Chaparro, Y. Li, R. Toledo-Crow, L. [71] R.P. Newton, A.G. Brenton, C.J. Smith, E. Dudley, Phytochemistry 65
DeNoyer, M. Fleisher, R.J. Robbins, P. Tempst, J. Proteome Res. 4 (2005) (2004) 1449.
1060. [72] M. Fukuda, N. Islam, S.H. Woo, A. Yamagishi, M. Takaoka, H. Hirano,
[35] H. Tammen, I. Schulte, R. Hess, C. Menzel, M. Kellmann, P. Schulz- Electrophoresis 24 (2003) 1319.
Knappe, Comb. Chem. High Throughput Screen. 8 (2005) 725. [73] P. Cash, Anal Chim. Acta 372 (1998) 121.
[36] M.S. Lowenthal, A.I. Mehta, K. Frogale, R.W. Bandle, R.P. Araujo, B.L. [74] T.H. Steinberg, C.R. Hart, W.F. Patton, in: J.M. Walker (Ed.), The Pro-
Hood, T.D. Veenstra, T.P. Conrads, P. Goldsmith, D. Fishman, E.F. Pet- teomics Protocols Handbook, Humana Press, Totowa, 2005, p. 19.
ricoin 3rd, L.A. Liotta, Clin. Chem. 51 (2005) 1933. [75] J.D. Yoder, T.S. Chen, C.R. Gagnier, S. Vemulapalli, C.S. Maier, D.E.
[37] E.M. Pasini, M. Kirkegaard, P. Mortensen, H.U. Lutz, A. Thomas, M. Hruby, Virol. J. 3 (1) (2006) 10.
Mann, Blood 108 (3) (2006) 791. [76] H.E. Botker, H.H. Kimose, P. Helligso, T.T. Nielsen, J. Mol. Cell Cardiol.
[38] A.R. Aldred, C.M. Brack, G. Schreiber, Comp. Biochem. Physiol. 111B 26 (1) (1994) 41.
(1995) 1. [77] F. Guilak, L.G. Alexopoulos, M.A. Haider, H.P. Ting-Beall, L.A. Setton,
[39] A.R. Consiglio, A.B. Lucion, Brain Res. Brain Res. Protoc. 5 (2000) Ann. Biomed. Eng. 33 (10) (2005) 1312.
109. [78] R. D’Souza, L.R. Brown, J.R. Newland, B.M. Levy, L.B. Lachman, Arch.
[40] X. Yuan, D.M. Desiderio, J. Chromatogr. B Anal. Technol. Biomed. Life Oral. Biol. 34 (5) (1989) 307.
Sci. 815 (2005) 179. [79] E.D. Ryan, W.H. Walker, J. Immunoassay 1 (4) (1980) 463.
[41] A. Sickmann, W. Dormeyer, S. Wortelkamp, D. Woitalla, W. Kuhn, H.E. [80] K.H. van het Hof, B.C. de Boer, L.B. Tijburg, B.R. Lucius, I. Zijp, C.E.
Meyer, J. Chromatogr. B 771 (2002) 167. West, J.G. Hautvast, J.A. Weststrate, J. Nutr. 130 (5) (2000) 1189.
[42] X. Yuan, D.M. Desiderio, Proteomics 5 (2005) 541. [81] M.A. Lysak, J. Cihalikova, M. Kubalakova, H. Simkova, G. Kunzel, J.
[43] P. Davidsson, S. Folkesson, M. Christiansson, M. Lindbjer, B. Dellheden, Dolezel, Chromosome Res. 7 (6) (1999) 431.
K. Blennow, A. Westman-Brinkmalm, Rapid Commun. Mass Spectrom. [82] G.R. Hashemitabar, G.R. Razmi, A. Naghibi, J. Vet. Med. B. Infect. Dis.
16 (2002) 2083. Vet. Public Health 52 (5) (2005) 243.
[44] S.F. Hansson, M. Puchades, K. Blennow, M. Sjogren, P. Davidsson, Pro- [83] M. Grace, L. Fletcher, S.K. Powers, M. Hughes, J. Coombes, J. Sports
teome Sci. 2 (2004) 7. Med. Phys. Fitness 36 (4) (1996) 281.
[45] A. Kaplan, M. Johnstone, Clin. Chem. 12 (1966) 717. [84] A. Martinac, J. Filipovic-Grcic, B. Perissutti, D. Voinovich, Z. Pavelic, J.
[46] B.R. Wenner, M.A. Lovell, B.C. Lynn, J. Proteome Res. 3 (2004) 97. Microencapsul. 22 (5) (2005) 549.
[47] E.J. Finehout, Z. Franck, K.H. Lee, Electrophoresis 25 (2004) 2564. [85] L.F. Pacheco, A.M. Carmona-Ribeiro, J. Colloid Interface Sci. 258 (1)
[48] C. Hirtz, F. Chevalier, D. Centeno, V. Rofidal, J.C. Egea, M. Rossignol, (2003) 146.
N. Sommerer, D. Deville de Periere, Proteomics 5 (17) (2005) 4597. [86] R.J. Townsend, M. Hill, N.R. Harris, N.M. White, Ultrasonics 42 (1–9)
[49] Y. Abiko, M. Nishimura, T. Kaku, Med. Electron Microsc. 36 (4) (2003) (2004) 319.
247. [87] B. Niemczewski, Ultrason Sonochem. 14 (1) (2007) 13.
 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31 27

[88] U. Fauth, T. Schlechtriemen, W. Heinrichs, I. Puente-Gonzalez, M. Hal- [130] G.H. Perdew, S.W. Schaup, D.P. Selivonchick, Anal. Biochem. 135 (1983)
magyi, Eur. J. Clin. Chem. Clin. Biochem. 31 (1) (1993) 5. 453.
[89] C.G. Mitchell, Biochem. J. 313 (1996) 769. [131] C.M. Taylor, S.E. Pfeiffer, Proteomics 3 (7) (2003) 1303.
[90] P. Racay, M. Kollarova, Biochem. Mol. Biol. Int. 38 (3) (1996) 493. [132] G.J. Babu, D. Wheeler, O. Alzate, M. Periasamy, Anal. Biochem. 325
[91] J.E. Nally, J.P. Whitelegge, R. Aguilera, M.M. Pereira, D.R. Blanco, M.A. (2004) 121.
Lovett, Proteomics (1) (2005) 144. [133] V.M. Molloy, T. Rabilloud, Electrophoresis 21 (2000) 1054.
[92] M.V. Radchenko, K. Tanaka, R. Waditee, S. Oshimi, Y. Matsuzaki, M. [134] M.P. Molloy, Anal. Biochem. 280 (1) (2000) 1.
Fukuhara, H. Kobayashi, T. Takabe, T. Nakamura, J. Biol. Chem. 281 [135] B.A. Stanley, I. Neverova, H.A. Brown, J.E. Van Eyk, Proteomics 3 (2003)
(29) (2006) 19822. 815.
[93] R.R. Ishmukhametov, M.A. Galkin, S.B. Vik, Biochim. Biophys. Acta [136] R. Henningsen, B.L. Gale, K.M. Straub, D.C. DeNagel, Proteomics 2
1706 (1–2) (2005) 110. (11) (2002) 1479.
[94] M. Lopez de Heredia, R.P. Jansen, BMC Biochem. 1 (2004) 5. [137] V. Santoni, T. Rabilloud, P. Doumas, D. Rouquie, M. Mansion, S. Kieffer,
[95] H.C. Wong, C.N. Chang, M.Y. Chen, J. Food Prot. 67 (6) (2004) 1243. J. Garin, M. Rossignol, Electrophoresis 20 (4–5) (1999) 705.
[96] M. Grange-Midroit, J.A. Garcia-Sevilla, M. Ferrer-Alcon, R. La Harpe, [138] M.P. Molloy, B.R. Herbert, M.B. Slade, T. Rabilloud, A.S. Nouwens, K.L.
C. Walzer, J. Guimon, Brain Res. Mol. Brain Res. 101 (1–2) (2002) 39. Williams, A.A. Gooley, Eur. J. Biochem. 267 (2000) 2871.
[97] J. Covert, L. Eskra, G. Splitter, J. Microbiol. Methods 60 (3) (2005) 383. [139] Y. Kashino, J. Chromatogr. B 797 (1–2) (2003) 191.
[98] M. Trautmann, C. Meincke, K. Vogt, M. Ruhnke, A.M. Lajous-Petter, [140] V. Santoni, S. Kieffer, D. Desclaux, F. Masson, T. Rabilloud, Elec-
Infection 20 (6) (1992) 350. trophoresis 21 (16) (2000) 3329.
[99] R.D. Miller, D.Y. Fung, Can. J. Microbiol. 22 (9) (1976) 1215. [141] V. Santoni, P. Doumas, D. Rouquie, M. Mansion, T. Rabilloud, M. Rossig-
[100] C.G. Mitchell, Biochem. J. 313 (1996) 769. nol, Biochimie 81 (1999) 655.
[101] N. Bakalara, A.M. Simpson, L. Simpson, J. Biol. Chem. 264 (31) (1989) [142] H. Schagger, G. von Jagov, Anal. Biochem. 199 (1991) 223.
18679. [143] D. Neff, N.A. Dencher, Biochem. Biophys. Res. Commun. 259 (3) (1999)
[102] I. Surgucheva, J. Bryan, Cell Motil. Cytoskeleton 32 (3) (1995) 233. 569.
[103] D.D. Vandre, F.M. Davis, P.N. Rao, G.G. Borisy, Proc. Natl. Acad. Sci. [144] N.S. Henderson, L.G. Nijtmans, J.G. Lindsay, E. Lamantea, M. Zeviani,
U.S.A. 81 (14) (1984) 4439. I.J. Holt, Electrophoresis 21 (2000) 2925.
[104] W. Rodgers, Biotechniques 32 (5) (2002) 1044. [145] H. Eubel, H.P. Braun, A.H. Millar, Plant Methods 1 (1) (2005) 11.
[105] T. Rabilloud, Electrophoresis 17 (5) (1996) 813. [146] H. Eubel, L. Jaensch, H.P. Braun, Plant Physiol. 1331 (2003) 274.
[106] M.R. Wilkins, K.L. Williams, R.D. Appel, D.F. Hochstrasser (Eds.), [147] H. Eubel, J. Heinemeyer, H.P. Braun, Plant Physiol. 134 (2004) 1450.
Proteome Research: New Frontiers in Functional Genomics, Springer, [148] J. Heinemeyer, H. Eubel, D. Wehmhoner, L. Jansch, H.P. Braun, Plants
Verlag, Berlin/Heidelberg/New York, 1997, p. 13. Photochem. 65 (12) (2004) 1683.
[107] I. Guillemin, M. Becker, K. Ociepka, E. Friauf, H.G. Nothwang, Pro- [149] P.A. Nielsen, J.V. Olsen, A.V. Podtelejnikov, J.R. Andersen, M. Mann,
teomics 5 (2005) 35. J.R. Wisniewski, Mol. Cell. Proteomics 4 (4) (2005) 402.
[108] A. Bierczynska-Krzysik, S.U. Kang, J. Silberring, G. Lubec, Neurochem. [150] C. Navarre, H. Degand, K.L. Bennett, J.S. Crawford, E. Mortz, M. Boutry,
Int. 49 (3) (2006) 245. Proteomics 2 (12) (2002) 1706.
[109] D. Boyd-Kimball, A. Castegna, R. Sultana, H.F. Poon, R. Petroze, B.C. [151] D.E. MacFarlane, Mol. Pharmacol. 22 (1982) 580.
Lynn, J.B. Klein, D.A. Butterfield, Brain Res. 1044 (2) (2005) 206. [152] H.D. Coughenour, R.S. Spaulding, C.M. Thompson, Proteomics 4 (2004)
[110] S.H. Bae, A.G. Harris, P.G. Hains, H. Chen, D.E. Garfin, S.L. Hazell, 3141.
Y.K. Paik, B.J. Walsh, S.J. Cordwell, Proteomics 3 (5) (2003) 569. [153] J. Hartinger, K. Stenius, D. Hogemann, R. Jahn, Anal. Biochem. 240 (1)
[111] S. Hoving, B. Gerrits, H. Voshol, D. Muller, R.C. Roberts, J. van Oostrum, (1996) 126.
Proteomics 2 (2) (2002) 127. [154] V.M. Govorun, A.I. Archakov, Biochemistry 67 (10) (2002) 1109.
[112] J. Culvenor, N.T. Ilaya, M.T. Ryan, L. Canterford, D.E. Hoke, N.A. [155] A.M. Seddon, P. Curnow, P.J. Booth, Biochim. Biophys. Acta 1666 (1–2)
Williamson, C.A. McLean, C.L. Masters, G. Evin, Eur. J. Biochem. 271 (2004) 105.
(2) (2004) 375. [156] R.M. Garavito, S. Ferguson-Miller, J. Biol. Chem. 276 (35) (2001) 32403.
[113] M. Fountoulakis, J.F. Juranville, D. Roder, S. Evers, P. Berndt, H. Langen, [157] B.R. Herbert, M.P. Molloy, A.A. Gooley, B.J. Walsh, W.G. Bryson, K.L.
Electrophoresis 19 (10) (1998) 1819. Williams, Electrophoresis (5) (1998) 845.
[114] R. Leimgruber, J.P. Malone, M.R. Radabaugh, M.L. LaPort, B.N. Violand, [158] T. Marshall, K.M. Williams, Anal. Biochem. 139 (2) (1984) 502.
J.B. Monahan, Proteomics 2 (2) (2002) 135. [159] N.D. Rawlings, D.P. Tolle, A.J. Barrett, Nucleic Acids Res. 32 (2004),
[115] A. Gorg, W. Weiss, M.J. Dunn, Proteomics 4 (2004) 3665. Database issue: D160.
[116] http://www.mnhn.fr/mnhn/bpy/2DMS/2DE.pdf. [160] E. Olivieri, B. Herbert, P.G. Righetti, Electrophoresis 22 (3) (2001)
[117] T. Rabilloud, C. Adessi, A. Giraudel, J. Lunardi, Electrophoresis 18 (3–4) 560.
(1997) 307. [161] R.J. Simpson, Purifying Proteins for Proteomics: A Laboratory Manual,
[118] B. Lanne, F. Potthast, A. Hoglund, H. Brockenhuus von Lowenhielm, CSHL Press, New York, 2004, p. 393.
A.C. Nystrom, F. Nilsson, B. Dahllof, Proteomics 1 (7) (2001) 819. [162] N. Minamino, J. Tanaka, H. Kuwahara, T. Kihara, Y. Satomi, M. Mat-
[119] C. Pasquali, I. Fialka, L.A. Huber, Electrophoresis 18 (1997) 2573. subae, T. Takao, J. Chromatogr. B 792 (2003) 33.
[120] I. Fialka, Electrophoresis 18 (14) (1997) 2582. [163] M. Svensson, K. Skold, P. Svenningsson, P.E. Andren, J. Proteome Res.
[121] B. Herbert, Electrophoresis 20 (4–5) (1999) 660. 2 (2) (2003) 213.
[122] T. Rabilloud, C. Adessi, A. Giraudel, J. Lunardi, Electrophoresis 18 (3–4) [164] E. Clynen, G. Baggerman, D. Veelaert, A. Cerstiaens, D. Van der Horst,
(1997) 307. L. Harthoorn, R. Derua, E. Waelkens, A. De Loof, L. Schoofs, Eur. J.
[123] G.F. Ames, K. Nikaido, Biochemistry 15 (3) (1976) 616. Biochem. 7 (2001) 1929.
[124] D.L. Wilson, M.E. Hall, G.C. Stone, R.W. Rubin, Anal. Biochem. 83 (1) [165] V.T. Ivanov, O.N. Yatskin, Expert Rev. Proteomics 2 (4) (2005) 463.
(1977) 33. [166] L. Castellanos-Serra, D. Paz-Lago, Electrophoresis 11 (2002) 1745.
[125] T. Rabilloud, T. Blisnick, M. Heller, S. Luche, R. Aebersold, J. Lunardi, [167] M.J. Han, J.W. Lee, S.Y. Lee, J. Proteome Res. 4 (6) (2005) 2429.
C. Braun-Breton, Electrophoresis 20 (18) (1999) 3603. [168] A.J. Rai, C.A. Gelfand, B.C. Haywood, D.J. Warunek, J. Yi, M.D.
[126] S. Luche, V. Santoni, T. Rabilloud, Proteomics 3 (3) (2003) 249. Schuchard, R.J. Mehigh, S.L. Cockrill, G.B. Scott, H. Tammen, P. Schulz-
[127] P.H. O’Farrel, J. Biol. Chem. 250 (1975) 4007. Knappe, D.W. Speicher, F. Vitzthum, B.B. Haab, G. Siest, D.W. Chan,
[128] J. Klose, Humangenetik 26 (3) (1975) 231. Proteomics 13 (2005) 3262.
[129] M. Chevallet, V. Santoni, A. Poinas, D. Rouquie, A. Fuchs, S. Kieffer, M. [169] J.W. Yang, P. Suder,. Silberring, G. Lubec, Neurochem. Int. 47 (3) (2005)
Rossignol, J. Lunardi, J. Garin, Electrophoresis 19 (11) (1998) 1901. 159.
28 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31

[170] B.C. Yoo, S.H. Kim, N. Cairns, M. Fountoulakis, G. Lubec, Biochem. [209] D. Sitnikov, D. Chan, E. Thibaudeau, M. Pinard, J.M. Hunter, J. Chro-
Biophys. Res. Commun. 280 (1) (2001) 249. matogr. B 832 (1) (2006) 41.
[171] D.E. Terry, E. Umstot, D.M. Desiderio, J. Am. Soc. Mass Spectrom. 6 [210] N.I. Govorukhina, A. Keizer-Gunnink, A.G. van der Zee, S. de Jong, H.W.
(2004) 784. de Bruijn, R. Bischoff, J. Chromatogr. A 1009 (2003) 171.
[172] Ch. Finnie, B. Svensson, Anal. Biochem. 311 (2002) 182. [211] E. Gianazza, P. Arnaud, J. Biochem. 201 (1982) 189.
[173] W.M. Willard, S.E. Hemby, Neurochem. Res. 29 (2004) 1065. [212] S. Kabir, Immunol. Invest. 31 (2002) 263.
[174] M.J. Dunn, Gel Electrophoresis of Proteins, BIOS Scientific Publishers [213] A. Sato, D.J. Sexton, L.A. Morganelli, E.H. Cohen, Q.L. Wu, G.P. Conley,
Ltd./Alden Press, Oxford, 1993. Z. Streltsova, S.W. Lee, M. Devlin, D.B. DeOliveira, J. Enright, R.B. Kent,
[175] M. Oh-Ishi, T. Maeda, J. Chromatogr. B 771 (1–2) (2002) 49. C.R. Wescott, T.C. Ransohoff, A.C. Ley, R.C. Ladner, Biotechnol. Prog.
[176] N.F. Visser, H. Lingeman, H. Irth, Anal. Bioanal. Chem. 382 (2005) 535. 18 (2002) 182.
[177] X. Yuana, D.M. Desiderio, J. Chromatogr. B 815 (2005) 179. [214] M. Fountoulakis, J.F. Juranville, L. Jiang, D. Avila, D. Roder, P. Jakob,
[178] M.A. Jenkins, S. Ratnaike, J. Biochem. Biophys. Methods 41 (1) (1999) P. Berndt, S. Evers, H. Langen, Amino Acids 27 (2004) 249.
31. [215] B.A. Chromy, A.D. Gonzales, J. Perkins, M.W. Choi, M.H. Corzett,
[179] T. Manabe, H. Miyamoto, K. Inoue, M. Nakatsu, M. Arai, Electrophoresis B.C. Chang, C.H. Corzett, S.L. McCutchen-Maloney, J. Proteome Res. 3
20 (1999) 3677. (2004) 1120.
[180] C.C. Lai, G.R. Her, Rapid Commun. Mass Spectrom. 14 (21) (2000) [216] J. Raymackers, A. Daniels, V. De Brabandere, C. Missiaen, M. Dauwe, P.
2012. Verhaert, E. Vanmechelen, L. mecheus, Electrophoresis 21 (2000) 2266.
[181] L. Jiang, L. Hea, M. Fountoulakis, J. Chromatogr. A 1023 (2004) 317. [217] H. Liao, J. Wu, E. Kuhn, W. Chin, B. Chang, M.D. Jones, S. O’Neil, K.R.
[182] A. Sickmann, W. Dormeyer, S. Wortelkamp, D. Woitalla, W. Kuhn, H.E. Clauser, J. Karl, F. Hasler, R. Roubenoff, W. Zolg, B.C. Guild, Arthritis
Meyer, Electrophoresis 13 (2000) 2721. Rheum. 50 (12) (2004) 3792.
[183] K. Zuobi-Hasona, P.J Crowley, A. Hasona, A.S. Bleiweis, L.J. Brady, [218] O. Chertov, J.T. Simpson, A. Biragyn, T.P. Conrads, T.D. Veenstra, R.J.
Electrophoresis 6 (2005) 1200. Fisher, Experts Rev. Proteomics 2 (2005) 139.
[184] J.M. Zeitzer, A. Morales-Villagran, N.T. Maidment, E.J. Behnke, L.C. [219] A.I. Metha, S. Ross, M.S. Lowenthal, Dis. Markers 19 (2003) 1.
Ackerson, F. Lopez-Rodriguez, I. Fried, J. Engel Jr., C.L. Wilson, Sleep [220] H.M. Georgiou, G.E. Rice, M.S. Baker, Proteomics 1 (12) (2001) 1503.
29 (4) (2006) 455. [221] B.A. Lollo, S. Harvey, J. Liao, A.C. Stevens, R. Wagenknecht, R. Sayen,
[185] A.L. Freed, J.D. Cooper, M.I. Davies, S.M. Lunte, J. Neurosci. Methods J. Whaley, F.G. Sajjadi, Electrophoresis 20 (4–5) (1999) 854.
109 (2001) 23. [222] R. Pieper, Q. Su, C.L. Gatlin, S.T. Huang, N.L. Anderson, S. Steiner,
[186] D.J. Wilson, L. Konermann, Anal. Chem. 77 (21) (2005) 6887. Proteomics 3 (4) (2003) 422.
[187] J. Bengtsson, B. Jansson, M. Hammarlund-Udenaes, Rapid. Commun. [223] M. Yamada, K. Murakami, J.C. Wallingford, Y. Yuki, Electrophoresis 23
Mass Spectrom. 19 (15) (2005) 2116. (2002) 1153.
[188] J.A. Jakubowski, N.G. Hatcher, J.V. Sweedler, J. Mass Spectrom. 40 (7) [224] Y.Y. Wang, P. Cheng, D.W. Chan, Proteomics 3 (2003) 243.
(2005) 924. [225] D.L. Rothemund, V.L. Locke, A. Liew, T.M. Thomas, V. Wasinger, D.B.
[189] B. Lignot, V. Lahogue, P. Bourseau, J. Biotechnol. 103 (3) (2003) 281. Rylatt, Proteomics 3 (3) (2003) 279.
[190] P. Tantipaiboonwong, S. Sinchaikul, S. Sriyam, S. Phutrakul, S.T. Chen, [226] B. Herbert, P.G. Righetti, Electrophoresis 21 (2000) 3639.
Proteomics 5 (4) (2005) 1140. [227] P.G. Righetti, A. Castagna, P. Antonioli, E. Boschetti, Electrophoresis 26
[191] B.N. Hammack, G.P. Owens, M.P. Burgoon, D.H. Gilden, Mult. Scler. 9 (2005) 297.
(5) (2003) 472. [228] W.A. Joo, D.Y. Lee, C.W. Kim, Biosci. Biotechnol. Biochem. 67 (7)
[192] J. Ouyang, J. Wang, D. Deng, Q. Long, X. Wang, Protein Expr. Purif. 32 (2003) 1574.
(1) (2003) 28. [229] M.K. Lawless, S. Hopkins, M.K. Anwer, J. Chromatogr. B 707 (1998)
[193] J. Salplachta, P. Rehulka, J. Chmelik, J. Mass Spectrom. 39 (12) (2004) 213.
1395. [230] L.K. Watkins, P.V. Bondarenko, D.C. Barbacci, S. Song, S.L. Cockrill,
[194] M. Zeller, W. Winkler, H. Hayden, M. Diestinger, M. Eliasen, B. D.H. Russell, R.D. Macfarlane, J. Chromatogr. A 840 (2) (1999) 183.
Gesslbauer, I. Miller, M. Chang, A. Kungl, E. Roth, R. Oehler, Elec- [231] M. Fountoulakis, Mass Spectrom. Rev. 23 (4) (2004) 231.
trophoresis 26 (2005) 2481. [232] P. Giavalisco, E. Nordhoff, L.H. Lehrach, J. Gobom, J. Klose, Elec-
[195] S.M. Stevens Jr., A.D. Zharikova, L. Prokai, Brain Res. Mol. Brain Res. trophoresis 24 (1–2) (2003) 207.
117 (2) (2003) 116. [233] T. Rabilloud, Methods Mol. Biol. 112 (1999) 9.
[196] T. Baczek, J. Pharm. Biomed. Anal. 34 (5) (2004) 851. [234] N.A. Guzman, J. Chromatogr. B 749 (2000) 197.
[197] L.L. Manza, S.L. Stamer, A.M. Ham, S.G. Codreanu, D.C. Liebler, Pro- [235] Z. Zhang, E. Carpenter, X. Puyan, N.J. Dovichi, Electrophoresis 22 (6)
teomics 5 (7) (2005) 1742. (2001) 1127.
[198] A.C. Lazar, L. Wang, W.A. Blattler, G. Amphlett, J.M. Lambert, W. [236] P.K. Koonjul, W.F. Brandt, J.M. Farrant, G.G. Lindsey, Nucleic Acids
Zhang, Rapid Commun. Mass Spectrom. 19 (13) (2005) 1806. Res. 27 (3) (1999) 915.
[199] W.A. Joo, D.Y. Lee, C.W. Kim, Biosci. Biotechnol. Biochem. 267 (7) [237] Y.Y. Chen, S.Y. Lin, Y.Y. Yeh, H.H. Hsiao, C.Y. Wu, S.T. Chen, A.H.J.
(2003) 1574. Wang, Electrophoresis 26 (2005) 2117.
[200] S.P. Chambers, Drug Discov. Today 7 (14) (2002) 759. [238] M. Zellner, W. Winkler, H. Hayden, M. Diestinger, M. Eliasen, B.
[201] H. Zischka, C.J. Gloeckner, C. Klein, S. Willmann, M. Swiatek-de Lange, Gesslbauer, I. Miller, M. Chang, A. Kungl, E. Roth, R. Oehler, Elec-
M. Ueffing, Proteomics 4 (12) (2004) 3776. trophoresis 26 (2005) 2481.
[202] S. Claverol, O. Burlet-Schiltz, J.E. Gairin, B. Monsarrat, Mol. Cell. Pro- [239] M. Cross, T. Nguyen, V. Bogdanoska, E. Reynolds, J.A. Hamilton, Pro-
teomics 2 (8) (2003) 483. teomics 5 (2005) 4754.
[203] M. Dong, L.G. Baggetto, P. Falson, M. Le Maire, F. Penin, Anal. Biochem. [240] A.J. Link, T.C. Fleischer, C.M. Weaver, V.R. Gerbasi, J.L. Jennings, Meth-
247 (2) (1997) 333. ods 35 (2005) 274.
[204] C. Hannam, G.L. Lange, A. Mellors, Anal. Biochem. 258 (2) (1998) 246. [241] W. Zhang, G. Zhou, Y. Zhao, M.A. White, Y. Zhao, Electrophoresis 24
[205] S.W. Hall, S.R. vandenBerg, Prep. Biochem. 19 (1) (1989) 1. (2003) 2855.
[206] https://fscimage.fishersci.com/webimages FSC/downloads/pierce [242] M. Ferro, D. Seigneurin-Berny, N. Rolland, A. Chapel, D. Salvi, J. Garin,
detergent.pdf. J. Joyard, Electrophoresis 21 (2000) 3517.
[207] N.L. Anderson, N.G. Anderson, Mol. Cell. Proteomics 1 (11) (2002) 845, [243] U. Sivars, F. Tjerneld, Biochem. Biophys. Acta 1474 (2000) 133.
Erratum in: Mol. Cell. Proteomics 2 (1) (2003) 50. [244] H. Everberg, U. Sivars, C. Emanuelsson, C. Persson, A. Englund, L.
[208] A.I. Mehta, S. Ross, M.S. Lowenthal, V. Fusaro, D.A. Fishman, E.F. Haneskog, P. Lipniunas, M. Jornten-Karlsson, F. Tjerneld, J. Chromatogr.
Petricoin 3rd, L.A. Liotta, Dis. Markers 19 (1) (2003–2004) 1. A 1029 (2004) 113.
 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31 29

[245] B.A. van Montfort, B. Canas, R. Duurkens, J. Godovac-Zimmermann, [289] T. Rabilloud, Anal. Chem. 72 (2000) 48A.
G.T. Robilard, J. Mass Spectrom. 37 (2002) 322. [290] J.C. Nishihara, K.M. Champion, Electrophoresis 23 (2002) 2203.
[246] B.A. van Montfort, M.K. Doeven, B. Canas, R. Duurkens, L.M. Veenhoff, [291] J.X. Yan, R. Wait, T. Berkelman, R.A. Harry, J.A. Westbrook, C.H.
B. Poolman, G.T. Robilard, Biochem. Biophys. Acta 1555 (2002) 111. Wheeler, M.J. Dunn, Electrophoresis 21 (2000) 3666.
[247] I.D. Cruzado-Park, J. Biomol. Technol. 16 (4) (2005) 311. [292] W. Schlags, M. Walther, M. Masree, M. Kratzel, C.R. Noe, B. Lachman,
[248] B.J. Cargile, J.R. Sevinsky, A.S. Essader, J.L. Stephenson Jr., J.I. Bundy, Electrophoresis 26 (2005) 246.
J. Biomol. Technol. 16 (2005) 181. [293] R. Wastermeier, R. Marouga, Biosci. Rep. 25 (2005) 19.
[249] D.N. Petsev, G.P. Lopez, C.F. Ivory, S.S. Sibbett, Lab Chip 5 (2005) 587. [294] V. Neuhoff, N. Arold, D. Taube, W. Ehrhardt, Electrophoresis 9 (1988)
[250] H.J. Lee, E.Y. Lee, M.S. Kwon, Y.K. Paik, Curr. Opin. Chem. Biol. 10 2550262.
(2006) 42. [295] T. Rabilloud, J.M. Strub, S. Luche, A. van Dorsselear, J. Lunardi, Pro-
[251] C. Obermaier, V. Jankowski, C. Schmutzler, J. Bauer, R. Wildgruber, M. teomics 1 (2001) 699.
Infanger, J. Kohrle, E. Krause, G. Weber, D. Grimm, Electrophoresis 26 [296] T.H. Steinberg, K.P.O. Top, K.N. Berggren, C. Kemper, L. Jones, Z.J.
(2005) 2109. Diwu, R.P. Haugland, W.F. Patton, Proteomics 1 (2001) 841.
[252] K. Szardenings, B. Li, L. Ma, M. Wu, Drug Discov. Today: Technol. 1 [297] T.H. Steinberg, B.J. Agnew, K.R. Gee, W.Y. Leung, T. Goodman, B.
(2004) 9. Hendrickson, J.M. Beechem, R.P. Haugland, W.F. Patton, Proteomics 3
[253] P. Lescuyer, D.F. Hochstrasser, J.C. Sanchez, Electrophoresis 25 (2004) (2003) 1128.
1125. [298] B.K. Ndimba, S. Chivasa, W.J. Simon, A.R. Slabas, Proteomics 5 (2005)
[254] O.N. Jensen, Curr. Opin. Chem. Biol. 8 (2004) 33. 4185.
[255] Z. Yang, W.S. Hancock, J. Chromatogr. A 1053 (2004) 79. [299] T. Stasyk, S. Marandell, R. Bakry, I. Feuerstein, C.W. Huck, G. Stecher,
[256] W.F. Patton, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 771 G.K. Bonn, L.A. Huber, Electrophoresis 26 (2005) 2850.
(2002) 3. [300] A. Alban, S.O. David, L. Bjorkesten, C. Anderson, E. Sloge, S. Lewis, I.
[257] M. Mukherij, Expert Rev. Proteomics 2 (2005) 117. Currie, Proteomics 3 (2003) 36.
[258] L.D. Luo, P.J. Wirth, Electrophoresis 14 (1993) 127. [301] D. Claeys, K. Geering, B.J. Meyer, Electrophoresis 26 (2005) 1189.
[259] H. Zhou, J.D. Watts, R. Aebersold, Nat. Biotechnol. 19 (2001) 375. [302] D.C. Simpson, R.D. Smith, Electrophoresis 26 (2005) 1291.
[260] H. Kaufmann, J.E. Bailey, M. Fussenegger, Proteomics 1 (2001) 194. [303] D. Fliser, S. Wittke, H. Mischak, Electrophoresis 26 (2005) 2708.
[261] A. Schlosser, J. Bodem, D. Bossemeyer, I. Grummt, W.D. Lehmann, [304] S. Wittke, D. Fliser, M. Haubitz, S. Bartel, R. Krebs, F. Hausadel, M.
Proteomics 2 (2002) 911. Hillmann, I. Golovko, P. Koester, H. Haller, T. Kaiser, H. Mischak, E.M.
[262] Z. Huilin, J.D. Watts, R. Aebersold, Nat. Biotechnol. 19 (2001) 375. Weissinger, J. Chromatogr. A 1013 (2003) 173.
[263] R. Raggiaschi, C. Lorenzetto, E. Diodato, A. Cariocasole, S. Gotta, G.C. [305] T. Kaiser, A. Hermann, J.T. Kielstein, S. Wittke, S. Bartel, R. Krebs,
Terstappen, Proteomics 6 (2006) 748. F. Hausadel, M. Hillmann, I. Golovko, P. Koester, H. Haller, E.M.
[264] J. Peng, D. Schwartz, J.E. Elias, C.C. Thoreen, D. Cheng, G. Marisischky, Weissinger, D. Fliser, H. Mischak, J. Chromatogr. A 1013 (2003)
J. Roelofs, D. Finley, S.P. Gygi, Nat. Biotechnol. 21 (2003) 921. 157.
[265] W.C. Lee, K.H. Lee, Anal. Biochem. 324 (2004) 1. [306] J. Zhang, H. Hu, M. Gao, P. Yang, X. Zhang, Electrophoresis 25 (2004)
[266] X. Sun, J.F. Chiu, Q.Y. He, Expert Rev. Proteomic 5 (2005) 649. 2374.
[267] Y. Zhang, X. Wang, W. Shan, B. Wu, H. Fan, X. Yu, Y. Tang, P. Yang, [307] M. Wetterhall, M. Palmblad, P. Hakansson, K.E. Markides, J. Bergquist,
Angew. Chem. 44 (2005) 615. Proteome Res. 1 (2002) 361.
[268] H. Chiku, M. Matsui, S. Murakami, Y. Kiyozumi, F. Mizukami, K. Sak- [308] M. Moini, H. Huang, Electrophoresis 25 (2004) 1981.
aguchia, Anal. Biochem. 15 (2003) 80. [309] Y. Wang, B.M. Balgley, P.A. Rudnick, C.S. Lee, J. Chromatogr. A 1073
[269] L. Wallman, S. Ekstrom, G. Marko-Varga, T. Laurel, J. Nielsson, Elec- (2005) 35.
trophoresis 25 (2004) 3778. [310] X. Li, Y. Gong, Y. Wang, S. Wu1, Y. Cai1, P. He, Z. Lu, W. Ying, Y.
[270] A. Gorg, G. Boguht, A. Kopf, G. Reil, H. Paral, W. Weiss, Proteomics 2 Zhang, L. Jiao, H. He, Z. Zhang, F. He, X. Zhao, X. Qian, Proteomics 5
(2002) 1652. (2005) 3423.
[271] L.F. Steel, B.B. Haab, S.M. Hanash, J. Chromatogr. B 815 (2005) 275. [311] R.L. Moritz, A.B. Clippingdale, E.A. Kapp, J.S. Eddes, H. Ji, S. Gilbert,
[272] J. Ingvarsson, M. Lindstedt, C.A.K. Borrenbaeck, C. Wingren, J. Pro- L.M. Connolly, R.J. Simpson, Proteomics 5 (2005) 3402.
teome Res. 5 (2006) 171. [312] H. Wu, C.L. Pan, Y.C. Yao, S.S. Chang, S.L. Li, T.F. Wu, Proteomics 6
[273] P. Angenendt, Drug Discov. Today 7 (2005) 503. (2006) 826.
[274] H.K. Binz, A. Pluckthun, Curr. Opin. Biotechnol. 16 (2005) 459. [313] G. Zhou, H. Li, Y. Gong, Y. Zhao, J. Cheng, P. Lee, Y. Zhao, Proteomics
[275] O. Chertov, J.T. Simpson, A. Biragyn, T.P. Conrads, T.D. Veenstra, R.J. 5 (2005) 3814.
Fisher, Experts Rev. Proteomics 2 (2005) 139. [314] A.S. Fandino, I. Rais, M. Vollmer, H. Elgass, H. Schagger, M. Karas, J.
[276] H. Hoch, J. Biochem. 45 (1949) 285. Mass Spectrom. 40 (2005) 1223.
[277] G.T. Hoehn, A.F. Sufferdini, Crit. Care Med. 33 (2005) S444. [315] J. Rappsilber, Y. Ishihama, M. Mann, Anal. Chem. 75 (2003) 663.
[278] C.R. Evans, J.W. Jorgenson, Anal. Bionanal. Chem. 378 (2004) 1952. [316] W. Sun, F. Li, S. Wu, X. Wang, D. Zheng, J. Wang, Y. Gao, Proteomics
[279] A. Guttman, M. Varoglu, J. Khandurina, Drug Discov. Today 9 (2004) 5 (2005) 4994.
136. [317] S.L. Sanders, J. Jennings, A. Canutescu, A.J. Link, P.A. Wei, Mol. Cell
[280] P.H. O’Farrell, J. Biol. Chem. 250 (1974) 4007. Biol. 22 (2002) 4723.
[281] A. Gorg, C. Obermaier, G. Bought, A. Harder, B. Scheibe, R. Wildgruber, [318] G. Lominadze, D.W. Powell, G.C. Luerman, A.J. Link, R.A. Ward, K.R.
W. Wiess, Electrophoresis 21 (2000) 1037. McLeish, Mol. Cell. Proteomics 4 (2005) 1503.
[282] M. Mukherij, Expert Rev. Proteomics 2 (2005) 17. [319] M. Ramstrom, M. Palmblad, K.E. Markides, P. Hakansson, J. Bergquist,
[283] P.S. Arora, H. Yamagiwa, A. Srivastava, M.E. Bolander, G. Sarkar, J. Proteomics 3 (2003) 184.
Orthop. Sci. 10 (2005) 160. [320] V.H. Wysocki, K.A. Resing, Q. Zhang, G. Cheng, Methods 35 (2005)
[284] L. Thandikkaran, M.A. Siegenthaler, D. Crettaz, P.A. Queloz, P. Schnei- 211.
der, J.D. Tissot, Proteomics 5 (2005) 3019. [321] B. Domon, R. Aebersold, Science 312 (2006) 212.
[285] Q. Fiu, C.P. Garnham, S.T. Elliott, D.E. Bovenkamp, J.E. Van Eyk, Pro- [322] Scripps Center for Mass Spectrometry, http://masspec.scripps.edu/
teomics 5 (2005) 2656. Services/Proteomics/salttol.php.
[286] P.G. Righatti, A. Castagna, B. Herbert, G. Candiano, Biosci. Rep. 25 [323] B. Thiede, W. Hohenwarter, A. Krah, J. Mattow, M. Schmid, F. Schmidt,
(2005) 3. P.R. Jungblut, Methods 35 (2005) 237.
[287] D.E. Macfarlane, Anal. Biochem. 176 (1989) 457. [324] J. Poland, T. Rabilloud, S. Pranav, in: J.M. Walker (Ed.), The Proteomics
[288] I. Rais, M. Karas, H. Schagger, Proteomics 4 (2004) 2567. Protocols Handbook, Humana Press, Totowa, 2005, p. 177.
30 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31

[325] A. Shevchenko, M. Wilm, O. Vorm, M. Mann, Anal. Chem. 68 (1996) [364] E. Nordhoff, M. Schurenberg, G. Thiele, C. Lubbert, K.D. Kloeppel, D.
850; Theiss, H. Lehrach, J. Gobom, Int. J. Mass Spectrom. 226 (2003) 163.
J.X. Yan, R. Wait, T. Berkelman, R.A. Harry, J.A. Westbrook, C.H. [365] http://www.bdal.de/data/files/temp/pac brochure.pdf.
Wheeler, M.J. Dunn, Electrophoresis 21 (2000) 3666. [366] K.C. Hung, H. Ding, B. Guo, Anal. Chem. 71 (1999) 518.
[326] B. Schulenberg, N. Ahnert, W.F. Patton, in: J.M. Walker (Ed.), The Pro- [367] X. Li, M. Wilm, T. Franz, Proteomics 5 (2005) 1460.
teomics Protocols Handbook, Humana Press, Totowa, 2005, p. 209. [368] H. Wei, S.L. Dean, M.C. Parkin, K. Nolkrantz, J.P. O’Callaghan, R.T.
[327] T.H. Steinberg, C.R. Hart, W.F. Patton, in: J.M. Walker (Ed.), The Pro- Kennedy, J. Mass Spectrom. 40 (2005) 1338.
teomics Protocols Handbook, Humana Press, Totowa, 2005, p. 215. [369] T. Redeby, J. Roeraade, A. Emmer, Rapid Commun. Mass Spectrom. 18
[328] R. Westermeier, R. Marouga, Biosci. Rep. 25 (2005) 19. (2004) 1161.
[329] T.G. Kleno, C.M. Andreasen, H.O. Kjeldal, L.R. Leonardsen, T.N. Krogh, [370] C. Afonso, N. Budimir, F. Fournier, J.C. Tabet, Curr. Pharma. Des. 11
P.F. Nielsen, M.V. Sorensen, O.N. Jensen, Anal. Chem. 76 (2004) 3576. (2005) 2559.
[330] M. Kussmann, E. Nordhoff, H. Rahbek-Nielsen, S. Haebel, M. Rossel- [371] V.V. Laiko, M.A. Baldwin, A.L. Burlingame, Anal. Chem. 72 (2000) 652.
Larsen, L. Jakobsen, J. Gobom, E. Mirgorodskaya, A. Kroll-Kristensen, [372] V.V. Laiko, N.I. Taranenko, V.D. Berkout, M.A. Yakshin, C.R. Prasad,
L. Palm, P. Roepstorff, J. Mass Spectrom. 32 (1997) 593. H.S. Lee, V.M. Doroshenko, J. Am. Soc. Mass Spectrom. 13 (2002) 354.
[331] R.C. Beavis, T. Chaudhary, B.T. Chait, Org. Mass Spectrom. 27 (1992) [373] S.J. Lawson, K.K. Murray, Rapid Commun. Mass Spectrom. 16 (2002)
156. 1248.
[332] R.C. Beavis, B.T. Chait, Rapid Commun. Mass Spectrom. 3 (1989) 432. [374] J.M. Daniel, V.V. Laiko, V.M. Doroshenko, R. Zenobi, Anal. Biochem.
[333] K. Strupat, M. Karas, F. Hillenkamp, Int. J. Mass Spectrom. Ion Processes 383 (2005) 895.
111 (1991) 89. [375] J.M. Daniel, S. Ehala, S.D. Friess, R. Zenobi, Analyst 129 (2004) 574.
[334] F. Gonnet, G. Lemaitre, G. Waksman, J. Tortajada, Proteome Sci. 1 (2003) [376] M.T. Nguyen, G.F. Ross, C.L. Dent, P. Devarajan, Am. J. Nephrol. 25
2. (2005) 318.
[335] S. Laugesen, P. Roepstorff, J. Am. Soc. Mass Spectrom. 14 (2003) 992. [377] W. Liu, M. Guan, D. Wu, Y. Zhang, Z. Wu, M. Xu, Y. Lu, Eur. Urol. 47
[336] M. Karas, F. Hillenkamp, Anal. Chem. 60 (1988) 2299. (2005) 456.
[337] O. Vorm, P. Roepstorff, M. Mann, Anal. Chem. 66 (1994) 3281. [378] S. Ranganathan, E. Williams, P. Ganchev, V. Gopalakrishnan, D. Lacomis,
[338] S.L. Cohen, B.T. Chait, Anal. Chem. 38 (1996) 31. L. Urbinelli, K. Newhall, M.E. Cudkowicz, R.H.J. Brown, R. Bowser, J.
[339] O.N. Jensen, M. Wilm, A. Shevchenko, M. Mann, in: A.J. Link (Ed.), Neurochem. 95 (2005) 1461.
Methods in Molecular Biology, Humana Press, Totowa, NJ, 1999, p. 571. [379] U. Ruetschi, H. Zetterberg, V.N. Podust, J. Gottfries, S. Li, A. Hviid
[340] J. Gobom, E. Nordhoff, E. Mirgorodskaya, R. Ekman, P. Roepstorff, J. Simonsen, J. McGuire, M. Karlsson, L. Rymo, H. Davies, L. Minthon,
Mass Spectrom. 34 (1999) 105. K. Blennow, Exp. Neurol. 196 (2005) 273.
[341] P. Rehulka, J. Salplachta, J. Chmelik, J. Mass Spectrom. 38 (2003) 1267. [380] D.W. Ho, Z.F. Yang, B.Y. Wong, D.L. Kwong, J.S. Sham, W.I. Wei, A.P.
[342] A. Shevchenko, M. Wilm, O. Vorm, M. Mann, Anal. Chem. 68 (1996) Yuen, Cancer. 107 (1) (2006) 99.
850. [381] J.Y. Engwegen, H.H. Helgason, A. Cats, N. Harris, J.M. Bonfrer, J.H.
[343] A.H. Brockman, B.S. Dodd, R. Orlando, Anal. Chem. 69 (1997) 4716. Schellens, J.H. Beijnen, World J. Gastroenterol. 12 (2006) 1536.
[344] Millipore, ZipTip Pipette Tips User Guide http://www.millipore. [382] N. Guerreiro, B. Gomez-Mancilla, S. Charmont, Proteome Sci. 4 (2006)
com/userguides.nsf/docs/p36110. 7.
[345] F. Rusconi, J.M. Schmitter, J. Rossier, M. le Maire, Anal. Chem. 70 (1998) [383] V. Seibert, M.P. Ebert, T. Buschmann, Brief. Funct. Genomic Proteomic
3046. 4 (2005) 16.
[346] Y. Ishihama, J. Rappsilber, M. Mann, J. Proteome Res. 5 (2006) 988. [384] M.C. Garcia, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 825
[347] J. Rappsilber, Y. Ishihama, M. Mann, Anal. Chem. 75 (2003) 663. (2005) 111.
[348] H. Erdjument-Bromage, M. Lui, L. Lacomis, A. Grewal, R.S. Annan, [385] S. Canarelli, I. Fisch, R. Freitag, J. Chromatogr. A 948 (2002) 139.
D.E. McNulty, S.A. Carr, P. Tempst, J. Chromatogr. A 826 (1998) [386] P. Lindberg, A.P. Dahlin, S.K. Bergstrom, S. Thorslund, P.E. Andren, F.
167. Nikolajeff, J. Bergquist, Electrophoresis 27 (11) (2006) 2075.
[349] H. Zhang, P.E. Andren, R.M. Caprioli, J. Mass Spectrom. 30 (1995) 1768. [387] M.P. Washburn, D. Wolters, J.R. Yates, Nat. Biotechnol. 19 (2001) 242.
[350] K. Gevaert, H. Demol, M. Puype, D. Broekaert, S. De Boeck, T. [388] C. Delahunty, J.R. Yates, Methods 35 (2005) 248.
Houthaeve, J. Vandekerckhove, Electrophoresis 18 (1997) 2950. [389] T. Kislinger, A.O. Gramolini, D.H. MacLennan, A. Emili, J. Am. Soc.
[351] T. Miliotis, S. Kjellstrom, J. Nilsson, T. Laurell, L.E. Edholm, G. Marko- Mass Spectrom. 16 (2005) 1207.
Varga, J. Mass Spectrom. 35 (2000) 369. [390] E.M. Bayer, A.R. Bottrill, J. Walshaw, M. Vigouroux, M.J. Naldrett, C.L.
[352] T. Miliotis, S. Kjellstrom, P. Onnerfjord, J. Nilsson, T. Laurell, L.E. Thomas, A.J. Maule, Proteomics 6 (2006) 301.
Edholm, G. Marko-Varga, J. Chromatogr. A 886 (2000) 99. [391] G. Cagney, S. Park, C. Chung, B. Tong, C. O’Dushlaine, D.C. Shields,
[353] H.S. Chen, T. Rejtar, V. Andreev, E. Moskovets, B.L. Karger, Anal. Chem. A. Emili, J. Proteome Res. 4 (2005) 1757.
77 (2005) 2323. [392] M.E. Regonesi, M. Del Favero, F. Basilico, F. Briani, L. Benazzi, P. Tor-
[354] Y.X. Wang, Y. Zhou, B.M. Balgley, J.W. Cooper, C.S. Lee, D.L. DeVoe, tora, P. Mauri, G. Deho, Biochimie 88 (2006) 151.
Electrophoresis 26 (2005) 3631. [393] X.W. Zhou, B.F. Kafsack, R.N. Cole, P. Beckett, R.F. Shen, V.B. Car-
[355] B. Zhang, C. McDonald, L. Li, Anal. Chem. 76 (2004) 992. ruthers, J. Biol. Chem. 280 (2005) 34233.
[356] C. Ericson, Q.T. Phung, D.M. Horn, E.C. Peters, J.R. Fitchett, S.B. [394] Z. Takats, J.M. Wiseman, B. Gologan, R.G. Cooks, Science 306 (2004)
Ficarro, A.R. Salomon, L.M. Brill, A. Brock, Anal. Chem. 75 (2003) 471.
2309. [395] Z. Takats, J.M. Wiseman, R.G. Cooks, J. Mass Spectrom. 40 (2005) 1261.
[357] K.W. Ro, J. Liu, D.R. Knapp, J. Chromatogr. A 1111 (2006) 40. [396] R.G. Cooks, Z. Ouyang, Z. Takats, J.M. Wiseman, Science 311 (2006)
[358] D. Ericsson, S. Ekstrom, J. Nilsson, J. Bergquist, G. Marko-Varga, T. 1566.
Laurell, Proteomics 1 (2001) 1072. [397] W. Yan, S.S. Chen, Brief. Funct. Genomic Proteomic 4 (2005) 27.
[359] T. Laurell, G. Marko-Varga, S. Ekstrom, M. Bengtsson, J. Nilsson, J. [398] S. Julka, F.E. Regnier, Brief. Funct. Genomic Proteomic 4 (2005) 158.
Biotechnol. 82 (2001) 161. [399] S.E. Ong, M. Mann, Nat. Chem. Biol. 1 (2005) 252.
[360] H. Wu, J. Zhai, Y. Tian, H. Lu, X. Wang, W. Jia, B. Liu, P. Yang, Y. Xu, [400] H. Zhou, R. Boyle, R. Aebersold, Methods Mol. Biol. 261 (2004) 511.
H. Wang, Lab Chip 4 (2004) 588. [401] T.P. Conrads, K. Alving, T.D. Veenstra, M.E. Belov, G.A. Anderson, D.J.
[361] M.H. Fortier, E. Bonneil, P. Goodley, P. Thibault, Anal. Chem. 77 (2005) Anderson, M.S. Lipton, L. Pasa-Tolic, H.R. Udseth, W.B. Chrisler, B.D.
1631. Thrall, R.D. Smith, Anal. Chem. 73 (2001) 2132.
[362] T. Redeby, A. Emmer, Anal. Bioanal. Chem. 381 (2005) 225. [402] S.E. Ong, B. Blagoev, I. Kratchmarova, D.B. Kristensen, H. Steen, A.
[363] L.P. Oehlers, Rapid Commun. Mass Spectrom. 19 (2005) 752. Pandey, M. Mann, Mol. Cell. Proteomics 1 (2002) 376.
 A. Bodzon-Kulakowska et al. / J. Chromatogr. B 849 (2007) 1–31 31

[403] S.E. Ong, I. Kratchmarova, M. Mann, J. Proteome Res. 2 (2003) 173. [417] L. DeSouza, G. Diehl, M.J. Rodrigues, J. Guo, A.D. Romaschin, T.J.
[404] N. Ibarrola, H. Molina, A. Iwahori, A. Pandey, J. Biol. Chem. 279 (2004) Colgan, K.W. Siu, J. Proteome Res. 4 (2005) 377.
15805. [418] Y.S. Cong, E. Fan, E. Wang, Mech. Ageing Dev. 127 (2006) 332.
[405] B. Blagoev, S.E. Ong, I. Kratchmarova, M. Mann, Nat. Biotechnol. 22 [419] L.R. Zieske, J. Exp. Bot. 57 (2006) 1501.
(2004) 1139. [420] S.A. Gerber, J. Rush, O. Stemmen, M.W. Kirschner, S.P. Gygi, Proc. Natl.
[406] S.E. Ong, G. Mittler, M. Mann, Nat. Methods 1 (2004) 119. Acad. Sci. U.S.A. 100 (2003) 6940.
[407] M. Schnolzer, P. Jedrzejewski, W.D. Lehmann, Electrophoresis 17 (1996) [421] P. Chaurand, S.A. Schwartz, R.M. Caprioli, Curr. Opin. Chem. Biol. 6 (5)
945. (2002) 676.
[408] S.P. Gygi, B. Rist, S.A. Gerber, F. Turecek, M.H. Gelb, R. Aebersold, [422] K. Tanaka, H. Waki, Y. Ido, S. Akita, Y. Joshida, T. Joshida, Rapid Com-
Nat. Biotechnol. 17 (1999) 994. mun. Mass Spectrom. 2 (1988) 151.
[409] H. Zhou, J.A. Ranish, J.D. Watts, R. Aebersold, Nat. Biotechnol. 20 [423] F. Hillenkamp, M. Haras, R.C. Beavis, B.T. Hait, Anal. Chem. 63 (24)
(2002) 512. (1991) 1193A.
[410] R. Zhang, C.S. Sioma, S. Wang, F.E. Regnier, Anal. Chem. 73 (2001) [424] A.F. Altelaar, J. van Minnen, C.R. Jimenez, R.M. Heeren, S.R. Piersma,
5142. Anal. Chem. 77 (3) (2005) 735.
[411] T.M. Pawlik, D.H. Hawke, Y. Liu, S. Krishnamurthy, H. Fritsche, K.K. [425] M. Stoeckli, P. Chaurand, D.E. Hallahan, R.M. Caprioli, Nat. Med. 7 (4)
Hunt, H.M. Kuerer, BMC Cancer 6 (2006) 68. (2001) 493.
[412] A.H. Fauq, R. Kache, M.A. Khan, I.E. Vega, Bioconjug. Chem. 17 (2006) [426] R.L. Caldwell, R.M. Caprioli, Mol. Cell. Proteomics 4 (4) (2005) 394.
248. [427] P. Chaurand, R.M. Caprioli, Electrophoresis 23 (18) (2002) 3125.
[413] P. Bottari, R. Aebersold, F. Turecek, M.H. Gleb, Bioconjug. Chem. 15 [428] P. Chaurand, B.B. DaGue, S.G. Ma, S. Kasper, R.M. Caprioli, Biochem-
(2004) 380. istry 40 (2001) 9725.
[414] P.L. Ross, Y.N. Huang, J.N. Marchese, B. Williamson, K. Parker, S. Hat- [429] R.M. Caprioli, T.B. Farmer, J. Gile, J. Anal. Chem. 69 (1997) 4751.
tan, N. Khainovski, S. Pillai, S. Dey, S. Daniels, S. Purkayastha, P. Juhasz, [430] P. Chaurand, M. Stoeckli, R.M. Caprioli, Anal. Chem. 71 (1999) 5263.
S. Martin, M. Bartlet-Jones, F. He, A. Jacobson, D.J. Pappin, Mol. Cell. [431] S.A. Schwartz, M.L. Reyzer, R.M. Caprioli, Anal. Chem. 71 (1999)
Proteomics 3 (2004) 1154. 5263.
[415] F. Petti, A. Thelemann, J. Kahler, S. McCormack, L. Castaldo, T. Hunt, [432] S.A. Schwartz, M.L. Reyzer, R.M. Caprioli, J. Mass Spectrom. 38 (7)
L. Nuwaysir, L. Zeiske, H. Haack, L. Sullivan, A. Garton, J.D. Haley, (2003) 699.
Mol. Cancer Ther. 4 (2005) 1186. [433] P. Chaurand, B.B. DaGue, R.S. Pearsall, D.W. Threadgill, R.M. Caprioli,
[416] V.G. Keshamouni, G. Michailidis, C.S. Grasso, S. Anthwal, J.R. Strahler, Proteomics 1 (2001) 1320.
A. Walker, D.A. Arenberg, R.C. Reddy, S. Akulapalli, V.J. Thannickal, [434] N. Masumori, T.Z. Thomas, P. Chaurand, T. Case, M. Paul, S. Kasper,
T.J. Standiford, P.C. Andrews, G.S. Omenn, J. Proteome Res. 5 (2006) R.M. Caprioli, T. Tsukamoto, S.B. Shappell, R.J. Matusik, Cancer Res.
1143. 61 (2001) 2239.