Pathology – Research and Practice 200 (2004) 95–103
Array technology and proteomics in autoimmune diseases
Veit Krenna,*, Iver Petersena, Thomas H. auplb, Axel Koepenikc, Christiane Blinda,
Manfred Dietela, Zoltan Konthurd, Karl Skrinerb
aDepartment of Pathology, Universit. atsklinikum Charit! e, Schumannstrasse 20/21, Berlin 10117, Germany
bDepartment of Rheumatology, University Hospital Charit! e, Berlin, Germany
cOligene GmbH, Berlin, Germany
dMax-Planck-Institute of Molecular Genetics, Dahlem, Berlin, Germany
Received 8 December 2003; accepted 2 February 2004
Two new technologies (tissue microarrays (TMAs) and proteomics) have generated a great amount of data in life
science. High-density TMAs allow for the simultaneous analysis of proteins and RNA by various methods
(immunohistochemistry, in situ hybridization, FISH) on a large scale and under highly standardized conditions.
Proteomics includes a variety of techniques that are partly high throughput. These techniques aim at the innovation of
proteins, the description of the domain structure, the determination of protein sequences and epitope characterization,
and ultimately the definition of protein function and protein reactivities in immunologic processes. Proteins that have
been characterized accordingly require validation mostly at the morphologic level of defined tissue, linking proteomics
to TMAs. In autoimmune diseases, array-based antigenic fingerprinting of autoantibodies will drive the development
and the selection of antigen-specific diagnostic tools and therapies. The powerful combination of genomics and
proteomics formed in tissue arrays has the potential to change the way the biology of autoimmunity is studied. Novel
targets of drug discovery, based on antigen-specific therapies to induce anergy, or regulatory T-cells using the targeted
autoantigens of individual patients could be developed in the coming decades.
r 2004 Elsevier GmbH. All rights reserved.
Keywords: Tissue array; Proteomics; Rheumatoid arthritis; Osteoarthritis; Synovitis score
In this review, we give an overview regarding two new
technologies, tissue microarrays (TMAs) and proteo-
mics, which have generated an extremely large set of
data in life science. In this context, we would like to
answer the question of how these techniques may
contribute to the identification of new pathogenetic
antigens involved in inflammatory and autoimmune
diseases. Following the completion of the human
genome project, a variety of techniques have been
developed that are generally molecular, miniaturized,
and high throughput [8,15,19]. These techniques allow
for the simultaneous analysis of a huge number of genes
or gene products in a single experiment. Good examples
for miniaturized high-throughput techniques are high-
density TMAs, which permit the simultaneous analysis
of proteins by immunohistochemistry or in situ hybri-
dization on a large scale and under highly standardized
conditions. This technique was described as early as
1986 . Proteomics includes a variety of techniques
that are partly high throughput. These techniques aim at
the innovation of proteins, the description of the
domain structure, thedetermination ofprotein
ARTICLE IN PRESS
*Corresponding author. Tel.: +30-450-536-134; fax: +30-450-536-
E-mail address: firstname.lastname@example.org (V. Krenn).
0344-0338/$-see front matter r 2004 Elsevier GmbH. All rights reserved.
ARTICLE IN PRESS
sequences and epitope characterization, and ultimately
the definition of protein function and protein reactivities
in immunologic processes.
Proteins that have been characterized accordingly
require validation mostly at the morphologic level of
defined tissue, linking proteomics to TMAs, identifying
in this way new candidate molecules for therapy and
Proteomics and immunomics in autoimmune
Autoimmune disease affects 3% of the population in
Western countries [6,12]. Over the last few decades,
considerable progress has been made in understanding
immune function, i.e., the role of the major histocom-
patibility complex and the nature of T-cell activation
that confers specificity to immune responses. However,
the underlying dysregulation leading to autoimmunity
remains largely unknown. Studies of monocygotic twins
have shown low concordance rates for most autoim-
mune diseases. Thus, environmental factors and a
certain autoimmune prone genetic background induce
This topic is further complicated by the variability of
the autoantigens. Multiple autoantigens have been
identified in Sjogren’s syndrome and systemic lupus
erythematosus. Their specific roles in the initiation,
perpetuation, and pathophysiology are poorly under-
stood. For many other autoimmune diseases, including
rheumatoid arthritis (RA) and psoriasis, the targeted
autoantigens remain unidentified, despite extensive
experimental efforts. A dysregulation of B-cells seems
to be based on a high level of bcl [5,26], which therefore
do not undergo apoptosis and are able to produce
autoantibodies  against certain proteins that are
mostly abundant, highly conserved, and/or modified
and able to form big complexes often associated with
RNA and situated in apoptotic blebs.
Proteomics is the global high-throughput analysis of
protein expression, function, and interactions of pro-
teins expressed in cells, tissues, or organism .
Proteomics reaches far beyond transcriptomics, as it
also includes post-translational and structural modifica-
tions. With our entrance into the proteomic era, it is
essential to develop novel tools that link proteomics
with immunomics, a holistic approach to studying all
immunologic proteins or peptides targeted by the
immune system. Protein arrays and other multiplex
screening technologies represent powerful tools for
studying the pathophysiology and the specificity of
autoimmune responses. It might become possible to
identify an antigenic fingerprint of each patient. This
might be a crucial step to target the major pathogenetic
antigen, unraveling that multiple proteins or peptides
modified by intrinsic or environmental factors are able
to induce certain autoimmune diseases in conjunction
with the individual genetic background.
The advent of array technology, including TMAs, and
proteomics provides an excellent opportunity to study
the pathology and pathogenesis of autoimmune dis-
Arrays and TMAs
TMAs are paraffin blocks that contain a high number
of cylindrical tissue biopsies. These cylindrical tissue
biopsies are obtained from individual donor paraffin-
embedded tissue blocks and placed into a recipient block
(Figs. 1 and 2). In the recipient block (tissue array), the
cylindrical cores are embedded according to defined
array coordinates (Fig. 2). In general, an array does not
necessarily consist of tissue fragments. Arrays also
include all molecules immobilized on a matrix at specific
coordinates. Figs. 3 and 4 show schematically peptide-,
protein-, antibody-, cell-, and tissue arrays.
Despite the small size of the individual cylindrical
specimens, studies have shown that results obtained
from TMA analysis are highly representative of their
donor tissues . They facilitate a rapid assessment of
the clinical relevance of molecular markers by enabling
the simultaneous analysis of hundreds of tissue speci-
mens. By defining the expression pattern of candidate
genes, TMAs are ideally suited for proteomics- and
genomics-based research and drug target discovery
[14,15,27,32]. Most of the applications of the TMA
technology have their origin in the field of cancer
research. Examples include the analysis of the frequency
of molecular alterations in malignant tumors, the
investigation of tumor progression, the identification
of prognostic factors, and finally the validation of new
genes defined as diagnostic targets [16,20].
The main part of work related to the production of
TMAs is even for a pathologist a rather time-consuming
process and consists of several steps. First, tissues and
tissue paraffin blocks that are potentially useful for
TMA production are selected. Hematoxylin and eosin
(H&E) -stained tissue sections are obtained from a large
series of donor blocks, which are retrieved from the
archive. The H&E-stained sections have to be viewed,
and relevant areas have to be marked. Subsequently, the
selected blocks and the marked slides are handed over to
the arraying device [1,4,16].
V. Krenn et al. / Pathology – Research and Practice 200 (2004) 95–10396
ARTICLE IN PRESS
The process of TMA production itself is simple and
can easily be carried out in a routine histopathology
laboratory. Principally, the hand-operated devices con-
sist of a block holder that can be positioned with high
precision, and of two hollow needles. Holes are punched
into empty recipient paraffin blocks using one of the
needles. The second needle is used to take tissue
cylinders from the ‘‘donor’’ paraffin block. These tissue
Fig. 1. Macroscopic aspect of two HE-stained sections from TMAs with different probe diameters and numbers of specimens.
Fig. 2. Stepwise fabrication of TMAs.
V. Krenn et al. / Pathology – Research and Practice 200 (2004) 95–10397
ARTICLE IN PRESS
cylinders are subsequently transferred into the prepared
holes of the recipient block at specific coordinates (Fig. 2).
Diameters ranging from 0.6 to 2mm have become the
standard sizes for tissue punches. Between 20 (Fig. 1)
and 1000 cylindrical tissue biopsies can be transferred
into a single recipient block [20,29]. Since the cylindrical
cores have small diameters, this technology causes only
minimal damage to the original tissue block so that it
can still be used for further diagnostic purposes.
TMAs and detection of molecular targets
Any immunohistochemical staining (Fig. 5), in situ
hybridization, or other molecular detection techniques
developed for whole tissue sections can also be adapted
to TMA sections . The limiting factor for antigen
detection is usually the nature, the quality, and the
variability of tissue fixation. TMA sections are most
commonly used for the detection of proteins by
Fig. 3. Principle of an array in the broadest sense: peptides, proteins antibodies, individual cells, and tissue fragments may be
‘‘arrayed’’ on a matrix that serves as a device for defined protein interactions/reactivities.
Fig. 4. Genome-wide protein expression and immunoscreening analysis using high-density filters and protein microarrays. The
cDNA expression library and proteins are arrayed in high-density protein filters using automated robot technology and are screened
for autoantibody binding. Positive clones are purified and re-arrayed on fast slides and detected on the protein chip with RGS-HIS-
V. Krenn et al. / Pathology – Research and Practice 200 (2004) 95–10398
ARTICLE IN PRESS
immunohistochemistry (Fig. 5). The TMA provides
several advantages. First, a large number of TMA
sections containing different types of tissues, including
normal tissues, different tumor types, or cell lines, can
be produced for testing and optimizing antigen retrieval,
antibody titers, and detection systems . Second, the
reproducibility of the staining reaction, as well as the
speed and reliability of the interpretation, is improved,
because all the tissues are on the same slide. Third,
consecutive slides can be stained with H&E for
morphology or with other antibodies against the same
or other molecular targets. Alternatively, immunohisto-
chemical double staining might be carried out by
detecting different antigens simultaneously in one
defined tissue area. Therefore, immunohistochemically
analyzed TMAs permit the comparison of multiple
antibody stainings in virtually the same histologically
controlled regions of the tissues.
TMAs can be used for a variety of applications in
clinical practice and research. Previously, TMAs were
mostly obtained from normal tissue, neoplastic tissue,
and tissue from degenerative diseases. However, the
application of TMAs is not limited to specific diseases.
Tissue samples of different autoimmune and inflamma-
tory diseases can be arrayed to obtain an ‘‘autoimmune/
inflammatory TMA’’. For example, an RA/synovitis
TMA was created, containing normal synovial tissue,
synovialitis from degenerative (osteoarthritis) and in-
flammatory joint diseases, including RA, reactive and
psoriatic arthritis (Fig. 5). In this ‘‘Synovitis TMA’’,
synovial tissue graded according to the synovitis score
 was used to include all degrees of inflammatory
infiltration of joint diseases. The major problem of
synovitis lies in its heterogeneity, which is a problem
particularly in small tissue samples. Focal inflammatory
alterations may be lacking in small biopsies. To over-
come this problem, tissue biopsy cores were taken from
the three relevant compartments of synovitis (synovial
cell lining, resident cells, and inflammatory infiltrate) of
graded synovial tissue samples. Samples from different
colitis types have been included in the ‘‘Colitis TMA’’.
Here again, heterogeneity of the inflammatory infiltrate
is an intrinsic problem. In that case, tissue biopsy cores
were taken from the epithelium, from the lamina
propria, and from areas of dense inflammatory infiltra-
tion. Therefore, ‘‘inflammatory’’ and ‘‘autoimmune’’
TMAs have to include all relevant compartments of the
Fig. 5. Microscopic aspects of a ‘‘Synovitis TMA’’ with CD52 immunohistochemistry shown to be up-regulated in RA using a
DNA-microarray approach. (A, B) Synovial tissue specimen from a patient with osteoarthritis (synovitis score 1/9 points). The
specimen contains the synovial lining cell layer, as well as the synovial stroma. (C, D) Synovial tissue specimen from a patient with
RA (synovitis score 8/9 points). This probe contains synovial lining cell layer, synovial stroma, and inflammatory infiltrate (original
magnification A, C 50?, B, D 200?).
V. Krenn et al. / Pathology – Research and Practice 200 (2004) 95–103 99
ARTICLE IN PRESS
inflamed tissue and allow for the analysis of antigen
presentation and localization, as well as for the
characterization of the phenotype of inflammation.
However, it does not help to find new, still unknown
autoantigens. For this purpose, various different ‘‘high’’
and ‘‘low throughput’’ proteomics technologies have
‘‘High’’ and ‘‘low throughput’’ technologies in
The most important feature of an assay is its capacity
to screen all possible immunoreactive compounds, i.e.,
thousands of proteins and peptides, for their ability to
interact with autoantibodies in a reasonable time. Many
sensitive assays are routinely used, including immuno-
blotting, two-dimensional gel electrophoresis (2DE),
and enzyme-linked immunosorbent assay (ELISA).
However, these assays cannot be applied to high-
throughput proteomics. Other techniques, including
microfluidics and mass spectroscopy, are under active
development, but are unlikely to be distributed to
persons other than academic experts who are able to
handle new methods such as the isotype-coded affinity
tag reagent method and mass spectroscopy [18,28].
In proteomics, complex mixtures of proteins obtained
from cells and tissues are separated by 2DE. A pH
gradient is used in the first dimension, and a conven-
tional SDS-PAGE in the second dimension. Finally,
protein and peptide spots are visualized by silver
staining or other staining methods. Western blotting of
these gels and immunoblot analyses using antibodies
obtained from autoimmune serum samples may help to
identify new autoantigens . This approach is easy to
perform by experienced researchers with a biochemical
background and can lead to remarkable results if the
correct disease-specific compartment is analyzed. This
technique is highly standardized, and immunoreactive
protein spots are analyzed by amino acid sequencing,
mass spectroscopy, or tandem mass spectroscopy. A
major disadvantage of 2DE is that the amount of
proteins seen on a gel is about 10–50 times lower than
the amount of proteins present in the starting material.
Furthermore, the nature of the immunoreactive proteins
cannot be readily identified on the basis of their charge
or molecular mass.
Different from 2DE, automated phage display allows
for the screening of large numbers of proteins and forms
a link to the encoded genes. George P. Smith [30,31]
pioneered the phage display of combinatorial peptide
libraries and affinity selection technologies. This has
turned out to be a powerful tool for isolating ligands,
affinity chromatography, studying protein–protein in-
teractions, epitope mapping of antibodies, and isolating
antibody fragments [7,9]. High-throughput technologies
for automated library handling and phage display have
been developed. These use picking–spotting robots and
a module for pin-based magnetic particle handling. This
system enables interaction screenings in an individual
way, so that each serum sample can be screened for
autoantigens (peptide or proteins) displayed on different
phages and selected in multiple screening rounds. The
whole screening procedure for autoantigens can be
automatized. Different phages are in use and are already
available on the market (T7, M13), displaying multiple
cDNA or peptide libraries . One problem of the
selected phages is that the identity of the gene is
unknown and requires further characterization. Single
phage clones have to be picked and analyzed in ELISA
or filter screens; positive phages are further sequenced,
and the insert is subcloned into suitable expression
vectors. This may be overcome by combining this with
new gene array hybridization, where the selected
positive phages are hybridized to identify the gene.
However, after selection of a phage and identification of
the cDNA, the sequence has to be part of an expression
clone bank. Furthermore, the epitope of the gene,
displayed on the phage, cannot be identified in this
The production of thousands of unique proteins using
arrayed cDNA expression libraries cloned in bacterial or
yeast expression vectors, displayed in a miniaturized
format, may eliminate the need to express individual
autoantigen-containing clones. High-throughput sub-
cloning of open reading frame libraries and character-
ization of clones by sequencing and hybridization
represents a major technical challenge and has been
described . A major goal is to achieve the standar-
dized form of thousands of verified recombinant
proteins spotted on filters or in other microformats.
Recently, protein filters have been designed on the basis
of a cDNA expression library hEX1 from human fetal
brain [2,3] and screened with antibody fragments from a
phage display antibody library, and antigens have been
identified. The human genome encodes about 120,000
proteins, not counting multiple isoforms . An
interesting solution to spot these proteins on a chip
was given by Walter et al. . His-tagged proteins are
expressed in liquid bacterial cultures, purified by nickel
chromatography, and applied to a polyvinylidene-
difluoride surface using ink-jet technology prior to
probing the array with an antibody solution. However,
these proteins are not modified post-translationally. The
production, purification, and spotting of every protein
on individual arrays are supported by picking–spotting
robots (Fig. 4).
Large numbers of recombinant proteins of known
identity will simplify the finding of autoantigens,
because they will overcome the problems of protein levels
in natural tissue and will lead to good reproducibility.
V. Krenn et al. / Pathology – Research and Practice 200 (2004) 95–103100
ARTICLE IN PRESS
However, these recombinant proteins lack naturally
occurring post-translational and putatively disease-
relevant modifications, which are often targets of
autoantibodies. Future techniques may overcome this
problem by using modified enzymes on the filter, or so-
called living arrays, where cell arrays have been reported
in which the proteins of interest are produced by
induction in bacteria or yeast or cells as the host .
New liquid approaches enable real-time millisecond
quantification of binding kinetics between bead-bound
antigens and the antibodies. In addition, they have the
potential to permit sensitive detection of low-affinity
biomolecular interactions, which is impossible with
other proteomics technologies, including phage plaque-
and colony-based strategies. The Luminex system
involves the use of distinct sets of spectrally resolvable
fluorescent beads . Each fluorescently identifiable set
of beads is conjugated to a different antigen, antibody
serum, incubated with the test sample, and then
identified using a bench top flow cytometer. Bead
approaches have the potential to use arrays of sig-
nificant complexity for large-scale peptide and protein
Several companies have developed microfluid tech-
nologies for multiplex proteomic analysis, which are
now available to laboratories. The protein bound to the
bead cannot show immunologic epitopes as in immuno-
blots when the antigen is denaturized. When it comes to
studying epitopes on autoantigens, it is advantageous to
use surface-bound proteins (as part of a well, matrix, or
planar surface). In the last chapter of this reveiw, we will
focus on future directions and applications of proteo-
Application of proteomics technology to
investigate the autoimmune response in patients
with rheumatic diseases
In the future, protein array technology could be
applied to clinical diagnostics. Protein arrays allow for
the analysis of the specificity of the autoantigen response
against the same panel of putative autoantibodies in
many patients (Figs. 3 and 4).
It may take years to change traditional ELISA and
immunoblot techniques into antigen microarrays, either
on planar surfaces or in other formats such as liquid
chips. However, multiplex parallel testing is the key to
obtaining relevant information for diagnosis. Multiple
tests are needed to optimize antigen production and
purification, different surfaces and bead-based techni-
ques in solution might overcome structural problems for
detection and, most important, quantification. More-
over, complex data sets have to be analyzed in order to
handle the extensive validation of large amounts of
serum samples. New chip-based diagnostic tests have
been published for allergy and autoimmune diseases, but
it will take time until these technologies can be applied
to regular clinical testing [13,24,25]. The greatest
impact of immunomics in the field of rheumatology will
be the search for autoantibodies. This technology
permits the rapid, simultaneous detection of thousands
of autoantibody immunoreactivities of serum samples
. One major impact for clinics will be to characterize
specific autoantigenic patterns, i.e., ‘‘autoantigenic
Guiding therapies with autoantigenic profiling may be
possible to predict the outcome of conventional and new
biologic therapies such as TNF-a or anti-CD20 anti-
bodies in RA, which are often not effective in a number
of patients. Using sera from animal models of auto-
immune disease and from patients in their early stage of
disease, antigens inducing autoimmunity might be
found. After characterizing the subclass of antibody, it
is possible to evaluate the pathogenic mechanism, the
formation of complement fixing immune complexes, or
the induction of an allergic IgE response. An important
goal in the near future will be to expand the array
technologies. In this way, it is possible to define all
autoantigens from autoimmune diseases, including RA,
diabetes mellitus, and autoimmune diseases of the skin
and allergies and make them useable for high-through-
put diagnostic screening.
The present work was supported by the SFB 421
(Project Z3: Protektive and pathologische Folgen der
Antigenverarbeitung) and by Berlinflame (Projekt B2),
Tissue arrays (e.g., synovitis, colitis, carcinomas) are
commercially available through oligene GmbH, Berlin
 H. Battifora, The multitumor (sausage) tissue block:
novel method for immunohistochemical antibody testing,
Lab. Invest. 55 (1986) 244–248.
 K. Bussow, D. Cahill, W. Nietfeld, D. Bancroft, E.
Scherzinger, H. Lehrach, G. Walter, A method for global
protein expression and antibody screening on high-
density filters of an arrayed cDNA library, Nucleic Acids
Res. 26 (1998) 5007–5008.
V. Krenn et al. / Pathology – Research and Practice 200 (2004) 95–103 101
ARTICLE IN PRESS
 K. Bussow, Z. Konthur, A. Lueking, H. Lehrach,
G. Walter, Protein array technology. Potential use in
medical diagnostics, Am. J. Pharmacogenomics 1 (2001)
 J.K. Chan, C.S. Wong, W.T. Ku, M.Y. Kwan, Reflec-
tions on the use of controls in immunohistochemistry
and proposal for application of a multitissue spring-
 G.S. Cheema, V. Roschke, D.M. Hilbert, W. Stohl,
Elevated serum B lymphocyte stimulator levels in patients
with systemic immune-based rheumatic diseases, Arthritis
Rheum. 44 (2001) 1313–1319.
 G.S. Cooper, B.C. Stroehla, The epidemiology of auto-
immune diseases, Autoimmun. Rev. 2 (2003) 119–125.
 R. Crameri, R. Kodzius, Z. Konthur, H. Lehrach, K.
Blaser, G. Walter, Tapping allergen repertoires by
advanced cloning technologies, Int. Arch. Allergy Im-
munol. 124 (2001) 43–47.
 B.L. Daugherty, S.J. Siciliano, M.S. Springer, Radiola-
beled chemokine binding assays, Methods Mol. Biol. 138
 P.A. De Ciechi, C.S. Devine, S.C. Lee, S.C. Howard, P.O.
Olins, M.H. Caparon, Utilization of multiple phage
display libraries for the identification of dissimilar peptide
motifs that bind to a B7-1 monoclonal antibody, Mol.
Diversity 1 (1996) 79–86.
 R.J. Fulton, R.L. McDade, P.L. Smith, L.J. Kienker,
J.R. KettmanJr, Advanced
with the FlowMetrix system, Clin. Chem. 43 (1997)
 S.P. Gygi, G.L. Corthals, Y. Zhang, Y. Rochon, R.
Aebersold, Evaluation of two-dimensional gel electro-
phoresis-based proteome analysis technology, PNAS 15
 D.L. Jacobson, S.J. Gange, N.R. Rose, N.M. Graham,
Epidemiology and estimated population burden of
selected autoimmune disease in the United States, Clin.
Immunol. Immunopathol. 84 (1997) 223–243.
 B. Jahn-Schmid, C. Harwanegg, R. Hiller, B. Bohle, C.
Ebner, O. Scheiner, M.W. Mueller, Allergen microarray:
comparison of microarray using recombinant allergens
with conventional diagnostic methods to detect allergen-
specific serum immunoglobulin, Clin. Exp. Allergy 33
 O.P. Kallioniemi, Biochip technologies in cancer research,
Ann. Med. 33 (2001) 142–147.
 G.L. Kenyon, D.M. DeMarini, E. Fuchs, D.J. Galas, J.F.
Kirsch, T.S. Leyh, W.H. Moos, G.A. Petsko, D. Ringe,
G.M. Rubin, L.C. Sheahan, National Research Council
Steering Committee, , Defining the mandate of proteo-
mics in the post-genomics era: workshop report, Mol.
Cell. Proteomics 1 (2002) 763–780.
 J. Kononen, L. Bubendorf, A. Kallioniemi, M. Barlund,
P. Schraml, S. Leighton, J. Torhorst, M.J. Mihatsch, G.
Sauter, O.P. Kallioniemi, Tissue microarrays for high-
throughput molecular profiling of tumor specimens, Nat.
Med. 4 (1998) 844–847.
 V. Krenn, L. Morawietz, T. Haupl, J. Neidel, I. Petersen,
A. Konig, Grading of chronic synovitis—a histopatholo-
Diag. Pathol.4 (2000)
gical grading system for molecular and diagnostic
pathology, Pathol. Res. Pract. 198 (2002) 317–325.
 R.M. LoPachin, R.C. Jones, T.A. Patterson, W. Slikker,
D.S. Barber, Application of proteomics to the study of
molecular mechanisms in neurotoxicology, Neurotoxicol-
ogy 24 (2003) 761–775.
 P. Lorenz, P. Ruschpler, D. Koczan, P. Stiehl, H.J.
Thiesen, From transcriptome to proteome: differentially
expressed proteins identified in synovial tissue of patients
suffering from rheumatoid arthritis and osteoarthritis by
an initial screen with a panel of 791 antibodies,
Proteomics 3 (2003) 991–1002.
 H. Moch, T. Kononen, O.P. Kallioniemi, G. Sauter,
Tissue microarrays: what will they bring to molecular
and anatomic pathology? Adv. Anat. Pathol. 8 (2001)
 C.A. Von Muhlen, E.M. Tan, Autoantibodies in the
diagnosis of systemic rheumatic diseases, Semin. Arthritis
Rheum. 24 (1995) 323–358.
 J. Packeisen, E. Korsching, H. Herbst, W. Boecker, H.
Mol. Pathol. 56 (2003) 198–204.
 A. Pandey, M. Mann, Proteomics to study genes and
genomes, Nature 405 (2000) 837–846.
 W.H. Robinson, C. DiGennaro, W. Hueber, B.B. Haab,
M. Kamachi, E.J. Dean, S. Fournel, D. Fong, M.C.
Genovese, H.E. de Vegvar, K. Skriner, D.L. Hirschberg,
R.I. Morris, S. Muller, G.J. Pruijn, W.J. van Venrooij,
J.S. Smolen, P.O. Brown, L. Steinman, P.J. Utz,
Autoantigen microarrays for multiplex characterization
of autoantibody responses, Nat. Med. 8 (2002) 295–301.
 W.H. Robinson, L. Steinman, P.J. Utz, Protein arrays for
autoantibody profiling and fine-specificity mapping,
Proteomics 3 (2003) 2077–2084.
 V. Roschke, S. Sosnovtseva, C.D. Ward, J.S. Hong, R.
Smith, V. Albert, W. Stohl, K.P. Baker, S. Ullrich, B.
Nardelli, D.M. Hilbert, T.S. Migone, BLyS and APRIL
form biologically active heterotrimers that are expressed
in patients with systemic
diseases, J. Immunol. 169 (2002) 4314–4321.
 G. Sauter, M. Mirlacher, Tissue microarrays for pre-
dictive molecular pathology, J. Clin. Pathol. 55 (2002)
 F. Schmidt, S. Donahoe, K. Hagens, J. Mattow, U.E.
Schaible, S.H. Kaufmann, R. Aebersold, P.R. Jungblut,
Complementary analysis of the Mycobacterium tubercu-
losis proteome by two-dimensional electrophoresis and
isotope coded affinity tag technology, Mol. Cell. Proteo-
mics, 3 (2004) 24–42.
 R. Simon, G. Sauter, Tissue microarrays for miniaturized
high-throughput molecular profiling of tumors, Exp.
Hematol. 30 (2002) 1365–1372.
 G.P. Smith, Filamentous fusion phage: novel expression
vectors that display cloned antigens on the virion surface,
Science 228 (1985) 1315–1317.
 G.P. Smith, V.A. Petrenko, Phage display, Chem. Rev. 97
 J. Torhorst, C. Bucher, J. Kononen, P. Haas, M. Zuber,
O.R. Kochli, F. Mross, H. Dieterich, H. Moch,
M. Mihatsch, O.P. Kallioniemi, G. Sauter, Tissue
V. Krenn et al. / Pathology – Research and Practice 200 (2004) 95–103102
ARTICLE IN PRESS Download full-text
microarrays for rapid linking of molecular changes to
clinical endpoints, Am. J. Pathol. 159 (2001) 2249–2256.
 P. Uetz, L. Giot, G. Cagney, T.A. Mansfield, R.S.
Judson, J.R. Knight, D. Lockshon, V. Narayan, M.
Srinivasan, P. Pochart, A. Qureshi-Emili, Y. Li, B.
Godwin, D. Conover, T. Kalbfleisch, G. Vijayadamodar,
M. Yang, M. Johnston, S. Fields, J.M. Rothberg,
Acomprehensive analysis of protein-protein interactions
in Saccharomyces cerevisiae, Nature 403 (2000) 623–627.
 J.C. Venter, M.D. Adams, E.W. Myers, P.W. Li, R.J.
Mural, G.G. Sutton, et al., The sequence of the human
genome, Science 291 (2001) 1304–1351.
 G. Walter, K. Bussow, A. Lueking, J. Glokler, High-
throughput protein arrays: prospects for molecular
diagnostics, Trends Mol. Med. 8 (2002) 250–253.
 G. Walter, Z. Konthur, H. Lehrach, High-throughput
screening of surface displayed gene products, Comb.
Chem. High Throughput Screen 4 (2001) 193–205.
V. Krenn et al. / Pathology – Research and Practice 200 (2004) 95–103 103