Background

Co-IP is extensively used to unravel the intricate relationship between protein complexes and various chromatin transactions during replication, transcription, and genome maintenance. However, it is challenging to keep labile protein-protein interactions intact during extraction, immunoprecipitation and washing steps of a Co-IP experiment. One way to stabilize labile protein interactions is to treat cells with cell-permeable reversible chemical cross-linker such as dithiobis-succinimidyl propionate prior to cell lysis (Smith et al., 2011). As this approach is associated with shortcomings such as inefficient extraction and nonspecific protein trapping, Co-IP without crosslinking (native-IP) is preferred.

A nuclear protein can be distributed to various sub-nuclear compartments or chromatin regions that require varying degree of stringency for its extractions and solubilization. For example, TOP1 is present in nucleoplasm, associated with chromatin, and also localized to nucleolus. Extraction of such proteins into multiple sub-nuclear fractions using buffers with narrow incremental stringencies would preserve more labile interactions than extracting them in either a single fraction with high stringency buffer or multiple fractions using buffers with large changes in stringencies (Figure 1A). For proteome-wide interactome analysis the key to successful Co-IP of labile interacting partners is to keep small differences in stringencies of the buffers used for sequential extraction to prevent unnecessary exposure of protein complexes that can be extracted with low stringency buffers to high stringency buffers. We applied such a sequential nuclear extraction strategy to efficiently extract, Co-IP, and identify proteins that interact with TOP1 (Figures 1B and 1C) (Husain et al., 2016).

TOP1 is an enzyme that relieves superhelical tension when two DNA strands are separated during transcription and replication, and prevent accumulation of negative supercoils, which if not relieved, may facilitate the formation of non-canonical DNA secondary structures. The non-canonical DNA secondary structures are implicated in transcription-associated mutagenesis, triplet repeat instability and activation induced cytidine deaminase-dependent immunoglobulin gene diversification (Kim et al., 2012; Kobayashi et al., 2009 and 2011). To understand the molecular mechanisms involved in TOP1-associated genomic instability, we performed Co-IP and identified proteins that interact with TOP1 (Figure 1). For these Co-IP experiments, we used transfectants of the TOP1-deficient mouse B-lymphocytic leukemia cell-line P388/CPT45 that stably expresses either GFP (green fluorescent protein)-tagged human TOP1 (GFPhTOP1) or GFP alone. Nuclear lysates from these cells were immunoprecipitated using the GFP-Trap^® system (Figure 1C). Although this protocol was developed for Co-IP of a GFP-fused proteins, it also worked successfully with FLAG^®-tagged proteins. The protocol described here is easy to perform and can be used as described or with minor modifications to Co-IP interacting partners of most chromatin-associated protein.


Figure 1. Native co-immunopreipitaion of TOP1-associated proteins. A. Schematic representation of sequential extraction of TOP1 protein complexes from P388/CPT45-GFPhTOP1 cells. P388/CPT45 cell expressing GFP alone was used as control. B. Analysis of TOP1 in soluble fractions (S1 to S4) and insoluble pellet (P) by western blotting with TOP1 and H3 antibodies. Asterisk (*) indicates a non-specific protein band. C. Co-immunoprecipitation of SMARCA4, subunits of histone chaperone FACT (SSRP1 and SUPT16H), and H3K4me3 with TOP1. The nuclei from P388/CPT45 transfectants expressing either GFP (P388/CPT45-GFP) or GFP-TOP1 (P388/CPT45-GFPhTOP1) were immunoprecipitated with GFP-Trap^® agarose, and 5-10% of the co-immunoprecipitated proteins were electrophoresed on 4-20% SDS–PAGE gradient gel, followed by western blot with indicated antibodies. The positions of nearest MW marker bands and their size are also shown on the left.