New Road Map Through the Land of IBD
immunofluorescence microscopy to investigate the mecha-
nisms underlying the inherent intestinal tissue injury seen in
both ulcerative colitis (UC) and Crohn’s disease (CD). In this
study the authors recruited patients with active inflammatory
bowel disease (IBD), as defined by a CD activity index ?150
and colitis activity index ?9 for CD and UC, respectively.
Endoscopic samples were taken from macroscopically in-
flamed areas, confirmed by histological examination, and
compared with the normal mucosa from control subjects.
Frozen tissue sections from these biopsies were then serially
treated with 32 different fluorescent-labeled antibodies and
imaged. The resulting data for all antibodies were integrated
to generate a matrix of combinatorial molecular phenotypes
(CMP), consisting of the presence or absence for each of the
32 epitopes distributed in a topographic manner over the
entire section. As such, a 2D representation of protein asso-
ciations was created on the histological map of the intestinal
mucosa. Wilcoxon rank-sum and Student’s t-test were used to
identify CMPs that differed in a statistically significant man-
ner between the biological classes. Hub analysis was used to
identify markers that colocalized with a particular marker of
The authors report finding 1337 CMP motifs that dif-
fered between CD and controls, 2930 that differed between
UC and controls, and 2599 that differed between CD and UC.
As it is difficult to determine which motifs are pathogenic and
which are bystanders, the authors then asked focused ques-
tions about CD3?T cells to determine if specific combina-
tions of protein colocalizations might differentiate between
the disease cohorts. Through this analysis the authors deter-
mined that naive T cells are increased and resistant to apop-
totic signals in the inflamed mucosa in CD. However, the
regulation of apoptosis was distinct in various T-cell sub-
populations. Furthermore, CD4?CD25?T regulatory cells
were found to be elevated only in UC, whereas CD4?CD7?
memory T cells were increased in both CD and UC. Using
erndt et al recently reported a provocative new approach
that uses Multi-Epitope-Ligand-Cartographie (MELC)
hub analysis, the authors were able to identify colocalization
patterns for NF-?b and PARP that distinguished UC from
CD. In summary, the authors have presented a usage for the
MELC technology to interrogate protein networks in a topo-
graphical context to better understand IBD pathogenesis.
With the recent explosion in high-throughput “-omic” technol-
ogies, there is an ever-increasing and confusing morass of bio-
logical data that is accompanied by not insignificant false-posi-
tive and -negative rates. However, through the integration of
orthogonal “-omic” approaches we can better understand the
relationships between individual genes/proteins to differentiate
between relevant disease pathophysiology and epiphenomenon.1
Recent genome-wide association studies for IBD susceptibility
loci have identified several genes, most notably NOD2 and
IL-23R.2,3Transcriptome studies of endoscopic pinch biopsies
have demonstrated that genes involved in Th1 processes and
antigen presentation are upregulated, whereas in UC genes that
regulate biosynthesis, metabolism, and electrolyte transport are
downregulated.4Tissue proteome studies have demonstrated
increased proteins associated with signal transduction, stress
responses, and energy metabolism.5However, it is only through
the integration of the growing number of high-throughput data-
sets that we might extract additional relevant biological infor-
mation. This approach is illustrated in our recent colorectal
cancer biomarker discovery studies, which integrated plasma
proteome data and tissue transcriptomic data to prioritize candi-
date biomarkers for orthogonal antibody validation (Hung et al,
Traditional “shotgun” proteomics approaches are per-
formed on tissue lysates to determine global protein differ-
ences between 2 biological states.8Unfortunately, proteins
require proper colocalization within a framework that is or-
ganized in time and space for their functionality. Whereas
this information is not captured by traditional proteomics
approaches, the MELC technology is able to provide such
valuable toponomic data. Nonetheless, MELC does suffer
from being an “unsupervised supervised approach.” This
method is unsupervised in that the CMPs can be clustered
without knowledge of the disease classes. However, it is
supervised in that it is dependent on carefully characterized
antibodies against previously known epitopes. As such, the
MELC technology can only uncover novel colocalizations
Copyright © 2008 Crohn’s & Colitis Foundation of America, Inc.
Published online 14 February 2008 in Wiley InterScience (www.
Inflamm Bowel Dis ● Volume 14, Number 6, June 2008
and interactions between known proteins. Nonetheless, these Download full-text
2 approaches may be coupled in a complementary fashion.
Shotgun proteomics may be used to identify novel proteins of
interest, whereas MELC can place these proteins in the con-
text of a protein interaction network. Taken together, the
MELC technology can now add an extra dimension of infor-
mation to traditional proteomics approaches.
As highlighted previously, analysis of enormous high-
throughput datasets has yielded limited biological significance.
However, the integration of orthogonal “-omic” datasets can
yield additional biologically significant information. Indeed,
such an analysis might identify further biological importance
among the thousands of CMP motifs that were identified with
only 32 antibodies. As such, the MELC technology now adds
the toponome to our growing arsenal against IBD.
Kenneth E. Hung, MD, PhD
Instructor of Medicine
Harvard Medical School
1. Ge H, Walhout AJM, Vidal M. Integrating omic information: a bridge
between genomics and systems biology. Trends Genet. 2003;19:
2. Cho JH, Weaver CT. The genetics of inflammatory bowel disease.
3. Xavier RJ, Podolsky DK. Unravelling the pathogenesis of inflammatory
bowel disease. Nature. 2007;448:427–434.
4. Wu F, Dassopoulos T, Cope L, et al. Genome-wide gene expression
differences in Crohn’s disease and ulcerative colitis from endoscopic
pinch biopsies: insights into distinctive pathogenesis. Inflamm Bowel
5. Shkoda A, Werner T, Daniel H, et al. Differential protein expression
profile in the intestinal epithelium from patients with inflammatory
bowel disease. J Proteome Res. 2007;6:1114–1125.
6. Hung KE, Kho AT, Sarracino D, et al. Mass spectrometry-based study
of the plasma proteome in a mouse intestinal tumor model. J Proteome
7. Forrester S, Hung KE, Kuick R, et al. Low-volume, high-throughput
sandwich immunoassays for profiling plasma proteins in mice: identifi-
cation of early-stage systemic inflammation in a mouse model of intes-
tinal cancer. Mol Oncol. 2007;1:216–225.
8. Fournier ML, Gilmore JM, Martin-Brown SA, et al. Multidimen-
sional separations-based shotgun proteomics. Chem Rev. 2007;107:
Inflamm Bowel Dis ● Volume 14, Number 6, June 2008Please Supply Wording