Altered Levels of Amyloid Precursor Protein Intracellular Domain-interacting Proteins in Alzheimer Disease Arunabha Chakrabarti, MSc,* Atri Chatterjee, MBBS,w Mohor B. Sengupta, MSc,* Partha Chattopadhyay, MD,w and Debashis Mukhopadhyay, PhD* Background: The amyloid precursor protein intracellular domain (AICD) is an intrinsically unstructured molecule with functional promiscuity that plays an important role in determining the fate of the neurons during its degeneration. Its association with Alzheimer disease (AD) recently played a key role in propelling scientists to choose AICD as a major molecule of interest. Although several studies have been conducted elucidating AICD’s participation in inducing neurodegenerative outcomes in AD condition, much remains to be deciphered regarding the linkage of AICD with cellular pathways in the AD scenario. Results: In the present study, we have pulled down interactors of nonphosphorylated AICD from the cerebrospinal fluid of AD subjects, identified them by matrix assisted laser desorption ion- ization mass spectrometry, and subsequently studied the differ- ential expression of these interactors in AD and control cases by 2-dimensional difference gel electrophoresis. The study has yielded some AICD-interactors that are differentially expressed in the AD cases and are involved in diverse cellular functions. Conclusions: This proteomic-based approach highlights the first step in finding the possible cellular pathways engaged in AD pathophysiology on the basis of interaction of one or more of their members with AICD. Key Words: AD, AICD, pull-down, MALDI, DIGE (Alzheimer Dis Assoc Disord 2013;00:000–000) A myloid precursor protein intracellular domain (AICD) is emerging as one of the key molecules of interest in Alzheimer disease (AD) research. 1 This intracellular frag- ment of amyloid precursor protein (APP) plays an impor- tant role in determining cell fate in neurodegeneration. 2 It can interact with other cellular adaptors and is shown to participate in transcriptional transactivation, Ab produc- tion, APP trafficking, and cellular apoptosis, among others. 2 AICD is also involved in the modulation of gene expression, cytoskeletal dynamics, 3 cellular calcium homeostasis, 4 and in exerting synaptic toxicity by the upregulation of glycogen synthase kinase-3b expression. 5 Most of the molecular functions of AICD are known to be dependent on its interaction with other proteins, charac- teristic of an intrinsically unstructured protein. In previous studies of AICD-interacting proteins, mainly the phos- phorylated forms of AICD were targeted. 6–9 In a real sce- nario, however, it is expected that there should be a balance of its phosphorylated and nonphosphorylated forms and both the forms are expected to interact with other proteins and subsequently deliver their intracellular effects with equal probability. 1 Keeping this possibility in mind, this study was designed to determine the interacting partners of nonphosphorylated AICD, and a bacterial expression sys- tem was used to express and purify AICD so as to minimize any untoward phosphorylation of the protein. Pull-down experiments were conducted using CSF proteins, as targets to determine the binding partners of nonphosphorylated AICD. Identified AICD-interacting partners from CSF were followed up using a 2-dimensional differential in-gel electrophoresis (2D-DIGE) approach to determine differentially expressed AICD interactors in CSF of AD compared with non-AD controls. There has always been a debate regarding CSF studies as to whether the samples should be pooled or each sample run individually. As our sample size was small, DIGE experiments were conducted with both the approaches. Individual DIGE experiments for the 3 AD samples, compared with non-AD samples, were carried out, and a single DIGE experiment with 8 pooled AD-CSF samples, compared with equal number of non-AD-CSF, was con- ducted in parallel. Combination of these 2 approaches ensured that there was a minimum possibility of false- positive results because of inherent biological variation of the clinical samples, although expressions of low abundant proteins were not obscured because of pooling. Here, the altered expression of those previously found AICD-inter- acting partners in CSF, which were found unambiguously both in the case of pooled samples and individual samples, are being reported. In this study, CSF was used not only because it has attracted attention an excellent AD biomarker, but also because the blood-brain barrier allows little exchange of proteins with serum and therefore findings in the CSF are likely to reflect brain related processes. 10–14 As a con- sequence change in the protein synthesis of the brain should also consistently and specifically be reflected in CSF. Although it is not known whether AICD is present or not in the CSF of AD patients, a very recent study showed the presence of this fragment in the brain tissue. 15 Received for publication May 4, 2012; accepted August 12, 2013. From the *Structural Genomics Division, Saha Institute of Nuclear Physics; and wDepartment of Medicine, Nilratan Sircar Medical College, Kolkata, West Bengal, India. A. Chakrabarti and A. Chatterjee contributed equally. Supported by the SPGHGD project, DAE, Government of India; Council of Scientific and Industrial Research (CSIR), Government of India. Partially supported by the SPGHGD project, DAE, Government of India. The authors declare no conflicts of interest. Reprints: Debashis Mukhopadhyay, PhD, Structural Genomics Division, Saha Institute of Nuclear Physics, 1/AF, Bidhannagar, Kolkata 700064, West Bengal, India (e-mail: debashis.mukhopadhyay@ saha.ac.in). Supplemental Digital Content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Website, www.alzheimerjournal.com). Copyright r 2013 by Lippincott Williams & Wilkins ORIGINAL ARTICLE Alzheimer Dis Assoc Disord  Volume 00, Number 00, ’’ 2013 www.alzheimerjournal.com | 1