Toponomics Analysis of Drug-Induced Changes in Arachidonic Acid-Dependent Signaling Pathways during Spinal Nociceptive Processing Bona Linke, † Sandra Pierre, † Ovidiu Coste, † Carlo Angioni, † Wiebke Becker, † Thorsten Ju ¨ rgen Maier, ‡ Dieter Steinhilber, ‡ Claus Wittpoth, § Gerd Geisslinger, † and Klaus Scholich* ,† Pharmazentrum Frankfurt, ZAFES, Institute of Clinical Pharmacology, Klinikum der Goethe-Universita ¨t Frankfurt, Germany, Institute of Pharmaceutical Chemistry, Goethe-Universita ¨t Frankfurt, Germany, and MelTec GmbH&Co KG, Magdeburg, Germany Received February 9, 2009 Abstract: Multi-Epitope-Ligand-Carthography (MELC) al- lows consecutive immunohistochemical visualization of up to 100 proteins on the same tissue sample. Subsequent biomathematical analysis of these images allows a quan- titative description of changes in protein networks. We used the MELC technology to study the effect of the nonopioid analgesic drug dipyrone on protein network proﬁles associated with arachidonic acid-dependent sig- naling pathways. MELC analysis with 31 different ﬂuo- rescence-labeled tags was used to compare the effect of dipyrone on protein networks in spinal cords of mice with zymosan-induced hyperalgesia, a common model for inﬂammatory pain. We found that the number of motifs which describe the colocalization of 5-lipoxygenase (5- LO) or 12-LO with other proteins increased disproportion- ally after dipyrone treatment. Activation of 5-LO and 12- LO induces their translocation to membrane compartments which was also reﬂected by MELC results. Although no changes in 5-LO or 12-LO expression were seen by Western blot analysis or by immunohistochemistry in spinal cords of dipyrone-treated mice, the activation of both enzymes was veriﬁed by determining LO-products. Spinal amounts of 5(S)-hydroxyeicosatetraenoic acid (HETE) and 12(S)-HETE, which are generated by 5-LO and 12-LO, respectively, were signiﬁcantly increased in spinal cords of dipyrone-treated animals. In primary spinal cord neurons, dipyrone selectively and dose-dependently in- creased 5(S)-(HETE) and 12(S)-HETE synthesis. Thus, we show for the ﬁrst time that monitoring protein network proﬁles by topological proteomic analysis is a useful tool to identify mechanisms of drug actions. Keywords: Toponome • protein networks • immunohis- tochemistry • dipyrone • lipoxygenase • arachidonic acid • pain • spinal cord Introduction Multi-Epitope-Ligand-Carthography (MELC) is a novel imag- ing technology which utilizes biomathematical tools to identify and quantify protein complexes after consecutive immunohis- tochemical visualization of up to 100 proteins on the same sample. 1,2 The detection of targets is achieved by using ﬂuorescence-labeled tag libraries that may comprise antibodies, lectins, proteins or toxins. After application and visualization of each tag, the dye is bleached and relabeling of the same sample with the next tag occurs. The different colocalization combinations of the tags are described as “combinatorial molecular phenotypes (CMP)” and are used to describe statisti- cal changes between different treatment groups. 3-6 On the basis of this statistical analysis, it is possible to quantitatively describe changes in the composition and distribution of protein complexes in an intact tissue sample. Thereby new insights in the protein interactions and protein network organization, referred to as the toponome, 1,2 are generated. The advantage of the MELC system over other common proteomic tools is based on the fact that analysis of protein networks occurs in samples with an intact cell or tissue structure. The usefulness of this technology to describe qualitative changes regarding immune cells during inﬂammatory skin and bowel diseases as well as colorectal cancer has already been demonstrated. 3-5 More recently, we showed that the quantitative analysis of CMPs in the MELC system allows the identiﬁcation of changes on the subcellular level in tissue samples. Using an animal model for acute and inﬂammatory pain, we detected and veriﬁed rare and intermittent activity-dependent changes at spinal cords synapses. 6 Here, we aimed to investigate whether the MELC technology can be used as screening tool to study the inﬂuence of drug treatment on protein networks in an in vivo setting using dipyrone as example. Dipyrone is a potent nonopioid analgesic and antipyretic drug that has been used clinically for more than 100 years. Over the last 20 years, several groups have demon- strated that dipyrone inhibits in vitro 7,8 and in vivo 9,10 the activity of cyclooxygenases (COX). COX converts arachidonic * To whom correspondence should be addressed. Klaus Scholich Phar- mazentrum Frankfurt, ZAFES, Institute of Clinical Pharmacology, Klinikum der Goethe-Universita ¨t Frankfurt, Theodor Stern Kai 7, 60590 Frankfurt, Germany. Tel.: 49-69-6301-83103. Fax: 49-69-6301-83378. E-mail: scholich@ em.uni-frankfurt.de. † Institute of Clinical Pharmacology, Klinikum der Goethe-Universita ¨t Frankfurt. ‡ Institute of Pharmaceutical Chemistry, Goethe-Universita ¨t Frankfurt. § MelTec GmbH&Co KG. 10.1021/pr900106v CCC: $40.75  2009 American Chemical Society Journal of Proteome Research 2009, 8, 4851–4859 4851 Published on Web 08/24/2009