Polarized TIRFM reveals changes in plasma membrane topology before and during granule fusion

Abstract

We have recently developed a combination of polarization and total internal reflection fluorescence microscopy (pTIRFM) to monitor changes in plasma membrane topology occurring after fusion of chromaffin granules. In this report, pTIRFM is further exploited to reveal two major findings in regards to the secretory pathway in bovine chromaffin cells. First, we show that changes in membrane topology are sometimes detected even prior to fusion. This occurs with high probability in a small subset of granules that appear in the evanescent field during the experiment. On these occasions, the plasma membrane invaginates with the movement just preceding the appearance of a granule in the evanescent field. Such events may represent a direct interaction of the granule with the plasma membrane. Second, we show that the topological fate of the post-fusion, granule/plasma membrane intermediate is regulated by divalent cation. When Sr2+ is used instead of Ca2+ to trigger exocytosis, membrane topology in the exocytotic region is stabilized with significant curvature and indentation.

Fig. 1

Granules at sites of plasma membrane invaginations viewed by electron microscopy. A–C Images taken by EM are shown of granules closely apposed to plasma membrane invaginations (A) or appearing to contact the plasma membrane at sites of invagination (B, C)

Fig. 2

Plasma membrane topological changes detected prior to movement of granules into the evanescent field. A A series of images are shown of the NPY-Cer and apposed P/S and P+2S intensities for an event where a granule moves into the field and subsequently fuses. Time after application of the high K⁺ stimulus is indicated. Interval between frames is 450 ms. B A sudden increase in P/S intensity (and decrease in P+2S intensity) is evident (at 11.70 s, circled area) immediately before the gradual increase in NPY-Cer intensity that signals a granule moving into the field (line a). Circle diameter is 876 nm. C NPY-Cer, P/S, and P+2S intensities are indicated for a different event. Line “a” indicates the time before the P/S increase and P+2S decline begin (at 18.90 s). The P/S increase is most significant between 21 and 22 s. The P+2S decreases gradually between 19 and 21 s. Both measures reach a steady-state after 22 s. The NPY-Cer increase begins around 22 s. In this example, the granule does not eventually fuse

Fig. 3

Plasma membrane topological changes upon fusion become long-lived in the presence of strontium. Extracellular Ca²⁺ was replaced by 10 mM Sr²⁺ in the high K⁺ depolarizing solution. A With fusion of a granule and release of NPY-Cer, a long-lived punctate increase in P/S is observed, along with a transient drop in P+2S. Circle radius is 800 nm. B Data from A is presented in a graph. C The P/S and P+2S of diI-membranes of individual fusion events from control (n = 50) and Sr²⁺ stimulated cells (n = 25) were normalized, aligned to the pre-fusion frame, averaged. The data were also normalized to the average of 10 pre-fusion frames. The average P/S is different between the two groups at every time point after 0.9 s (P < 0.05, Student’s t test). The P+2S for the two groups are similar. Error bars are not shown for clarity. D A cumulative frequency histogram was generated to compare the remaining P/S increase from individual fusion events in control (Ca²⁺) and Sr²⁺-treated cells at 22.5 s after fusion. With Sr²⁺ replacing external Ca²⁺, a greater fraction of fusion events were resistant to curvature decay. The distributions are significantly different (P < 0.05, Mann–Whitney test). The percent increase in P/S was calculated by taking the difference between the P/S at a particular frame (P/S _f) and the average P/S of 10 pre-fusion frames (Ave_pre), divided by the average; (P/S _f − Ave_pre)/Ave_pre. Control data were adapted from (Anantharam et al. 2010). E The results from C are interpreted in the cartoons