Unbiased Thiol-Labeling and Top-Down Proteomic Analyses Implicate Multiple Proteins in the Late Steps of Regulated Secretion

Abstract

Regulated exocytosis enables temporal and spatial control over the secretion of biologically active compounds; however, the mechanism by which Ca²⁺ modulates different stages of exocytosis is still poorly understood. For an unbiased, top-down proteomic approach, select thiol- reactive reagents were used to investigate this process in release-ready native secretory vesicles. We previously characterized a biphasic effect of these reagents on Ca²⁺-triggered exocytosis: low doses potentiated Ca²⁺ sensitivity, whereas high doses inhibited Ca²⁺ sensitivity and extent of vesicle fusion. Capitalizing on this novel potentiating effect, we have now identified fluorescent thiol- reactive reagents producing the same effects: Lucifer yellow iodoacetamide, monobromobimane, and dibromobimane. Top-down proteomic analyses of fluorescently labeled proteins from total and cholesterol-enriched vesicle membrane fractions using two-dimensional gel electrophoresis coupled with mass spectrometry identified several candidate targets, some of which have been previously linked to the late steps of regulated exocytosis and some of which are novel. Initial validation studies indicate that Rab proteins are involved in the modulation of Ca²⁺ sensitivity, and thus the efficiency of membrane fusion, which may, in part, be linked to their previously identified upstream roles in vesicle docking.

Keywords: 2-dimensional gel electrophoresis; Rab GTPase; calcium; exocytosis; secretory vesicle; thiol-reactivity.

Figure 1

Promotion of the Ca²⁺-triggered steps of exocytosis with fluorescent thiol reagents. (A) Ca²⁺ activity curves (n = 7–9) for cortical vesicle (CV)–CV fusion after 20 min treatment with 750 µM LYIA, 2 mM mBB or 500 µM dBB. Fusion kinetics in response to 24.3 ± 2.2 µM [Ca²⁺]_free (inset; n = 6–8). (B) Sr²⁺ activity curves (n = 6 or 7) for CV–CV fusion after 20 min treatment with 750 µM LYIA, 2 mM mBB or 500 µM dBB. Fusion kinetics in response to 3110 µM [Sr²⁺]_free (inset; n = 5 or 6). (C–E) Summary of concentration-dependent effects of LYIA (C), mBB (D) and dBB (E) on the Ca²⁺-sensitivity of CV–CV fusion. Data presented as mean ± SEM; statistical analysis by one-way ANOVA with Bonferroni multiple comparison test versus control (* p< 0.05, ** p< 0.01, *** p < 0.001).

Figure 2

Total membrane proteome of CV treated with fluorescent thiol reagents. (A) Average two-dimensional gel electrophoresis (2DE) gel image of total membrane proteome from untreated CV resolved by mini 3–10 NL IPG and large 10%–14% SDS-PAGE format, stained for total protein with Sypro Ruby (n = 10). The average PVDF blot images (n = 4) scanned for LYIA (B), mBB (C) and dBB (D); addition of negatively charged LYIA shifts pIs of proteoforms relative to other labels. Open and closed arrows indicate background labeling of CV proteins that are normally released to form the fertilization envelope. Open arrow head indicates low abundance, poorly resolved peptides co-migrating at the gel front. Boxed regions of interest contain spots labeled by all three fluorescent reagents are shown in greater detail in Figure 3.

Figure 3

Fluorescent thiol-labeled spots excised from total membrane proteome.To improve resolution of protein spots, pooled CV membrane protein samples for each condition were resolved in duplicate by large format 2DE (i.e., 17 cm 3–10 NL IPG and large 10%–14% SDS-PAGE). A montage of boxed regions of interest indicated in Figure 2 is shown from representative gel image of untreated CV stained for total protein (Sypro ruby) and representative PVDF blot images for LYIA, mBB and dBB; addition of negatively charged LYIA shifts pIs of proteoforms relative to other labels. Protein spots reproducibly labeled with LYIA, mBB and dBB in both mini and large formats, as indicated by numbers in total protein images and arrowheads in thiol-labelled images, were excised for identification. The contrast level on each panel has been adjusted to visualize low-abundance-labeled spots; for full blot images refer to Figure 2.

Figure 4

Cholesterol-enriched membrane proteome of CV treated with fluorescent thiol reagents. (A) Average 2DE gel image of cholesterol-enriched membrane proteome from untreated CV resolved on mini 3-10 NL IPG and large 10%–14% SDS-PAGE, stained with Sypro Ruby (n = 6). Boxed regions correspond to the same regions of interest shown for total membrane proteome. (B) A montage of boxed regions is shown from average gel image of untreated CV stained for total protein (Sypro ruby; n = 6) and average PVDF blot images (n = 3) scanned for LYIA, mBB and dBB. Protein spots corresponding to labeled proteins from the total membrane proteome are numbered accordingly. Open and closed arrows indicate background labeling of CV content proteins that are normally released to form the fertilization envelope. An open arrow head indicates labeled peptides co-migrating at the dye front. Protein spots reproducibly labeled with LYIA, mBB and dBB, as indicated by bold numbers in the total protein image and arrowheads in thiol-labelled images, were excised for identification. The contrast level on each panel has been adjusted to visualize low-abundance-labeled spots; for full blot images refer to Figure S4.

Figure 5

GTP has minimal impact on the efficiency of Ca²⁺-triggered exocytosis. (A) Ca²⁺ activity curves for standard CV–CV fusion assays supplemented with GTP (n = 5–6) and fusion kinetics in response to 33.7 ± 2.7 µM [Ca²⁺]_free (inset; n = 3). (B) Ca²⁺ activity curves for standard CV–CV fusion assay supplemented with GTPγS (n = 4) and fusion kinetics in response to 33.7 ± 2.7 µM [Ca²⁺]_free (inset; n = 3).

Figure 6

Rab-GTPase modulates the efficiency of Ca²⁺-triggered exocytosis. (A) Ca²⁺ activity curves for the standard CV–CV fusion assay supplemented with 100 µM GTPγS (CTRL; n = 11), plus the addition of 5 µg/mL RAB (n = 5), 50 µg/mL RAB (n = 5), and 200 µg/mL RAB (n = 6) or 200 µg/mL scrambled (SC; n = 4) peptide for 30 min. Fusion kinetics (inset; n = 4–9) in response to 41.8 ± 3.5 µM [Ca²⁺]_free shown for CTRL, 200 µg/mL RAB and 200 µg/mL SC peptide. (B) Ca²⁺ activity curves for the modified settle assay supplemented with 100 µM GTPγS (CTRL; n = 3), plus the addition of 200 µg/mL RAB (n = 2) or 200 µg/mL SC (n = 2) peptide for 30 min. (C) Ca²⁺ activity curves for standard CV–CV fusion assay pre-treated with 20 mM IA for 20 min. IA treatment was followed by a second incubation in fresh BIM supplemented with 100 µM GTPγS (IA control; n = 5), plus the addition of 200 µg/mL RAB (n = 5) or 200 µg/mL SC (n = 3) peptide for 30 min. Fusion kinetics (inset; n = 3–4) in response to 45.4 ± 6.2 µM [Ca²⁺]_free are shown.