Calcium regulates the association between mitochondria and a smooth subdomain of the endoplasmic reticulum

Abstract

Association between the ER and mitochondria has long been observed, and the formation of close contacts between ER and mitochondria is necessary for the ER-mediated sequestration of cytosolic calcium by mitochondria. Autocrine motility factor receptor (AMF-R) is a marker for a smooth subdomain of the ER, shown here by confocal microscopy to be distinct from, yet closely associated with the calnexin- or calreticulin-labeled ER. By EM, smooth ER AMF-R tubules exhibit direct interactions with mitochondria, identifying them as a mitochondria-associated smooth ER subdomain. In digitonin-permeabilized MDCK cells, the addition of rat liver cytosol stimulates the dissociation of smooth ER and mitochondria under conditions of low calcium. Using BAPTA chelators of various affinities and CaEGTA buffers of defined free Ca(2+) concentrations and quantitative confocal microscopy, we show that free calcium concentrations <100 nM favor dissociation, whereas those >1 microM favor close association between these two organelles. Therefore, we describe a cellular mechanism that facilitates the close association of this smooth ER subdomain and mitochondria when cytosolic free calcium rises above physiological levels.

Figure 1

Smooth ER AMF-R tubules exhibit a close relationship to the calnexin- and calreticulin-labeled ER. MDCK cells were fixed and double-immunofluorescently labeled for either calnexin (A) or calreticulin (D) and AMF-R (B and E), and were imaged by confocal microscopy. Merged images (C and F) show that the distribution of AMF-R–labeled tubules is distinct from, but always in proximity to, the calnexin- or calreticulin-labeled ER (arrowheads). Bar, 10 μm.

Figure 2

The AMF-R–labeled smooth ER subdomain is selectively associated with mitochondria. MDCK cells were fixed and double-immunofluorescently labeled for Mt-HSP70 (A,D, and G), together with AMF-R (B), calnexin (E), or calreticulin (N), and imaged by confocal microscopy. Merged images (C, F, and I) show that the association of smooth ER AMF-R tubules with mitochondria is significantly greater than either the calnexin- or calreticulin-labeled ER. Bar, 20 μm.

Figure 3

AMF-R tubules and mitochondria exhibit a close association by EM. MDCK cells were postembedding immunogold-labeled for AMF-R. AMF-R–labeling (arrows) is associated with smooth tubules closely associated with mitochondria. Bars: (A) 0.2 μM; (B and C) 0.1 μM.

Figure 4

Quantitative analysis of the distance between ER tubules and mitochondria. The minimal distance of smooth and rough ER tubules labeled or not for AMF-R, as indicated, was determined from EM micrographs of AMF-R–labeled MDCK cells. The number of ER tubules located from 0–200, 200–400, 400–600, 600–800, 800–1,000, and >1,000 nm of a mitochondrion within the same cell is presented. The total number of tubules identified per condition (n) is indicated.

Figure 5

Rat liver cytosol dissociates AMF-R tubules and mitochondria in semipermeabilized MDCK cells. MDCK cells were permeabilized by digitonin and then incubated with buffer A (A, B, and C) or buffer A supplemented with 5 mg/ml rat liver cytosol (D, E, and F) at 37°C for 60 min. Fixed cells were immunofluorescently labeled for AMF-R (A and D; red) and Mt-HSP70 (B and E; green) and imaged by confocal microscopy. Dual color merged images are presented in C and F. Extensive overlap between smooth ER AMF-R tubules and mitochondria is observed in the presence of buffer alone (C), but following addition of cytosol (F) extensive AMF-R labeling not associated with mitochondria (in red) can be visualized. Bar, 20 μM.

Figure 6

Ca²+ regulates cytosol-mediated dissociation of smooth ER AMF-R tubules and mitochondria. A, The percent dissociation of smooth ER tubules (± SEM) from mitochondria in semi-permeabilized MDCK cells was quantified from confocal images of cells double-labeled for AMF-R and Mt-HSP70 (for details see Materials and Methods). MDCK cells were permeabilized with digitonin and incubated for 60 min in the presence of buffer A alone or buffer A supplemented with 5 mg/ml rat liver cytosol or 5 mg/ml heat-inactivated cytosol. B, Semipermeabilized MDCK cells were incubated for 60 min with buffer A supplemented with 0.5 mM Ca²⁺ (+0.5 mM Ca), with cytosol (+cytosol), or with cytosol (C) plus 0.1, 0.25, or 0.5 mM Ca²⁺, as indicated. To assess the role of free Ca²⁺ in cytosol-mediated dissociation of smooth ER AMF-R tubules from mitochondria, semipermeabilized MDCK cells were incubated with buffer A alone, or with buffer A supplemented with cytosol (+cytosol), with cytosol containing 0.5 mM Ca²⁺ (+C+0.5 Ca), or with cytosol, 0.5 mM Ca²⁺, and 100 μM of various BAPTAs (B+C+0.5 Ca+BAPTA) with Ca²⁺ dissociation constants (K _d) of 40 nM, 160 nM, 635 nM, and 1.6 μM, as indicated. Similar results were obtained from at least three independent experiments and data from representative experiments is presented. C, Using data presented in Table , free [Ca²⁺]_SOL measured using Fura-2 in CaEGTA buffers 2-7 was plotted against the extent of dissociation of smooth ER AMF-R tubules from mitochondria. Free [Ca²⁺]_SOL above 100 nM is associated with increasing association between the two organelles.