Background

A number of proteins are secreted from the cell into the extracellular environment via various secretory pathways. These pathways include the conventional secretory pathway following ER-Golgi-plasma membrane route (Lee et al., 2004) and multiple unconventional pathways, such as lysosomal exocytosis, translocation across the plasma membrane and exosome and ectosome release (Zhang and Schekman, 2013). To study these pathways, it is often necessary to analyze proteins secreted from cultured cells into the media. Proteins may be secreted either in free form or in association with vesicular membrane structures such as ectosomes and exosomes (Zhang and Schekman, 2013). Whether these proteins are membrane-associated or not determines the type of approach used for their analysis. To isolate membrane-associated proteins, media is usually subjected to differential centrifugation procedure upon which the protein-containing membranes are sedimented, concentrated and purified from the constituents of the media, which allows for unhampered protein analysis (Momen-Heravi et al., 2013). However, the analysis of free-form proteins is a more challenging task (Chevallet et al., 2007). Two major obstacles are 1) the presence of substantial amounts of bovine serum albumin (BSA) in the serum or other supplements added to most cell culture media, which may mask the protein of interest; and 2) low abundance in the media of at least some of the secreted proteins, including mHtt. This protocol has been developed in order to study secreted mHtt and may be useful for analyzing other low-abundance, free-form proteins in the media.

Protocol optimization
To tackle the issue of excessive BSA, we first monitored cell viability in various media with partial or total reduction of BSA. In the case of Neuro2A cells stably expressing N-terminal 571 amino acids of mHtt with 72 glutamines (Neuro2A-mHtt cells), the optimal medium was Opti-MEM, which not only preserved viability of the cells, but also enhanced the secretion of mutant huntingtin (mHtt) as compared to other media, such as DMEM without supplemented fetal bovine serum (FBS) and HBSS (Figure 1). Moreover, we observed that applying conditioned Opti-MEM (Opti-MEM pre-incubated with naïve Neuro2A cells which do not express mHtt) dramatically enhanced the secretion of mHtt from Neuro2A-mHtt cells, possibly due to enrichment of the media with factors influencing secretion (Figure 1). However, primary cortical neurons were not viable in the presence of Opti-MEM, whereas they survived well in Neurobasal medium, regardless of the addition of BSA-containing B27 supplement (Table 1). Since high amounts of B27 supplement lead to the appearance of BSA accumulates in the region of the membrane where mHtt normally migrates (even when BSA was partially removed by immunoprecipitation) (Figure 2), we opted for Neurobasal media with reduced amount of B27 (0.2%).

To address the issue of low abundance of mHtt in the media, we compared several approaches to concentrate the media and thus enrich mHtt. While TCA precipitation of proteins and immunoprecipitation of mHtt from the media did not give satisfactory results, ultrafiltration of the media proved to be the method of choice for both Neuro2A cells and neurons. We were then able to detect mHtt by immunoblotting (Figure 1).


Figure 1. Optimizing conditions for detection of mHtt secreted from Neuro2A-mHtt cells. 800,000 mHtt-Neuro2A cells were incubated for 4 h in the following media: DMEM, Opti-MEM, conditioned Opti-MEM and HBSS. Cells were lysed, and media were collected, concentrated and analyzed by SDS-PAGE/immunoblotting using anti-Htt antibody. Tubulin was used as a loading control for the cell lysates. The fifth lane represents concentrated conditioned media before being applied on mHtt-expressing Neuro2A.

Table 1. Cells from Figure 2 were visually monitored for viability



Figure 2. Optimizing conditions for detection of mHtt in primary cortical neurons. 700,000 cells were incubated overnight in Neurobasal media with 2% B27 supplement (NB 2% B27), plain Neurobasal media (NB), Opti-MEM, Neurobasal media with 0.4 or 0.2% B27, a mix of Opti-MEM and Neurobasal media at 1:1, or Neurobasal media with 2% B27 where BSA was partially depleted by immunoprecipitation using anti-BSA agarose beads (NB 2% B27 -BSA). The media were collected, concentrated and analyzed by SDS-PAGE/Western blotting. Membrane was stained by Ponceau S to reveal proteins. Large accumulations of proteins (BSA) in the region where mHtt normally migrates (arrow) revealed that the conditions for lanes 1 and 7 were not optimal for mHtt-immunodetection.