Visualization of localized store-operated calcium entry in mouse astrocytes. Close proximity to the endoplasmic reticulum

Abstract

Unloading of endoplasmic reticulum (ER) Ca(2+) stores activates influx of extracellular Ca(2+) through 'store-operated' Ca(2+) channels (SOCs) in the plasma membrane (PM) of most cells, including astrocytes. A key unresolved issue concerning SOC function is their spatial relationship to ER Ca(2+) stores. Here, using high resolution imaging with the membrane-associated Ca(2+) indicator, FFP-18, it is shown that store-operated Ca(2+) entry (SOCE) in primary cultured mouse cortical astrocytes occurs at plasma membrane-ER junctions. In the absence of extracellular Ca(2+), depletion of ER Ca(2+) stores using cyclopiazonic acid, an ER Ca(2+)-ATPase inhibitor, and caffeine transiently increases the sub-plasma-membrane Ca(2+) concentration ([Ca(2+)](SPM)) within a restricted space between the plasma membrane and adjacent ER. Restoration of extracellular Ca(2+) causes localized Ca(2+) influx that first increases [Ca(2+)](SPM) in the same restricted regions and then, with a delay, in ER-free regions. Antisense knockdown of the TRPC1 gene, proposed to encode endogenous SOCs, markedly reduces SOCE measured with Fura-2. High resolution immunocytochemistry with anti-TRPC1 antibody reveals that these TRPC-encoded SOCs are confined to the PM microdomains adjacent to the underlying 'junctional' ER. Thus, Ca(2+) entry through TRPC-encoded SOCs is closely linked, not only functionally, but also structurally, to the ER Ca(2+) stores.

Figure 1. Illustration of the method used to image sub-PM microdomains in cells loaded with FFP-18

Diagram shows a cross-section (X–Z plane) of a small region at the periphery of an FFP-18-loaded cell. There are two wide-field fluorescence light paths. Light in path ‘A’ is emitted from dye in the PM facing the cytosol at the bottom and top. This dye signals changes in the [Ca²⁺] near the PM. Light in path ‘B’ is emitted from dye in the PM and ER membranes facing the tiny PM–ER cytosolic spaces at the bottom and at the top. Localization of dye molecules in the ER membrane facing the ER lumen is negligible, as confirmed by the relatively uniform resting FFP-18 ratio image shown in Fig. 2Da.

Figure 2. SOCE-induced local [Ca²⁺]_SPM signal in an astrocyte

A, FFP-18 (F₃₆₀) image of a portion of cell. To better visualize the ER, the image contrast was increased so that the ER appears to contain much more dye than is actually the case. N, nucleus. Scale bar = 25 μm. B, time course of ratio (F₃₄₀/F₃₈₀) signal in regions outlined in A (red and blue); times of treatment with CPA (10 μm) + CAF (10 mm) and Ca²⁺-free solution are indicated. RY (1 μm) was applied 10 min before the traces shown and was maintained throughout the experiment. Inset shows the early rising phase of the F₃₄₀/F₃₈₀ ratio after Ca²⁺ restoration. C, DiOC image of the boxed portion of A. Arrowheads point to ER (including mitochondria). Most mitochondria, which are brightly stained by DiOC, appeared to lie on the ER. D, ratio (F₃₄₀/F₃₈₀) images (a–j) captured at the times indicated in B. E, linescan of CPA + CAF-induced changes in ratio (F₃₄₀/F₃₈₀), shown in B, along the white dotted line in C and F. The time scale is variable. G, linescan with high time scale resolution (from the boxed portion in E). Scale bar = 10 s. Comparable results were obtained in 12 other cells.

Figure 3. High K⁺-induced [Ca²⁺]_SPM signal in an astrocyte

A, FFP-18 (F₃₆₀) image of a small portion of cell; scale bar = 25 μm. B, linescan of high K⁺-induced changes in ratio along the white dotted line in A. C, time courses of ratio (F₃₄₀/F₃₈₀) signal in regions outlined in A (red and blue); time of treatment with 50 mm KCl is indicated. Inset shows the early rising phase of the F₃₄₀/F₃₈₀ ratio. D, ratio (F₃₄₀/F₃₈₀) images captured at the times indicated in C inset. Comparable results were obtained in eight other cells.

Figure 4. Effect of antisense oligos for the TRPC1 gene on TRPC1 protein expression, SOCE and astrocyte proliferation

A, Western blot of TRPC1 expression in control cells (no oligos), and in astrocytes treated with NS- or AS-oligos. Proteins (10 μg/lane) were separated on 7.5% polyacrylamine gel, blotted, and probed with specific anti-TRPC1 antibody. Blots were later incubated with anti-β-actin antibodies to verify uniform protein loading. B, data are normalized to the amount of β-actin and are expressed as means ± s.e.m. from 15 fetuses (3 litters). *P < 0.001 versus NS. C, representative records showing time course of [Ca²⁺]_cyt changes in cells with disrupted TRPC1 expression (red) and in non-transfected cells (blue). Red and blue records correspond to spatially averaged changes in [Ca²⁺]_cyt within the small red and blue ovals, respectively, in the fura-2 fluorescence images in Ea. CPA (10 μm) was applied to the cells in the absence and presence of extracellular Ca²⁺, as indicated. D, summarized data showing the amplitude of SOCE induced by CPA. Data are means ± s.e.m. (n = 58 cells; 10 coverslips). *P < 0.001 versus NS. Each bar shows data from 9 to 12 fetuses from 3 litters (3–4 fetuses/litter). E, Fura-2 fluorescence (F₃₆₀) image (a) showing the cells in which [Ca²⁺]_cyt was measured (scale bar = 25 μm). Following the Ca²⁺ imaging experiment, the same cells were immunocytochemically stained for TRPC1 (b) and then labelled with DAPI (c). Two DAPI-labelled cells were TRPC1 positive, indicating they were not transfected with AS-oligos; six cells were non-fluorescent indicating inhibition of TRPC1 expression (b). F, cell numbers were determined before (Basal) and after incubating 60 h in control growth medium (10% FBS) (Contr) or media containing NS- or AS-oligos. Data are means ± s.e.m. of 4 experiments/litters (53 coverslips). *P < 0.001 versus NS.

Figure 5. Immunofluorescent localization of TRPC proteins in astrocytes

A and B, low magnification images of cells crossreacted with anti-TRPC1 antibody (A); ER-Tracker was later used to stain the ER (B). C, fluorescence detected when the primary antibody was preincubated with TRPC1 peptide. D, high magnification images of a portion of an astrocyte triple labelled with anti-TRPC1 antibody (a), anti-SERCA-2 antibody (b), and ER-Tracker (c). All three labels show similar reticular distributions. E, images of a portion of another astrocyte double labelled with anti-TRPC4 antibodies (a), and ER-Tracker (b). F, high power images of a portion of astrocyte double labelled with anti-PM Ca²⁺-ATPase (a) and anti-SERCA-2 antibodies (b). G, images of a portion of astrocyte double labelled with anti-L-type Ca²⁺ channel antibody (a) and ER-Tracker (b). H, images of a portion of a non-permeabilized astrocyte triple labelled with anti-TRPC1 antibody (a), anti-SERCA-2 antibody (b), and ER-Tracker (c). Scale bars = 25 μm (A and C), 5 μm (Da, Ea and Fa), and 10 μm (Ga and Ha). Similar results were obtained in 18 cells (A and B), 53 cells (D), 37 cells (E), 14 cells (F), 11 cells (G), and 15 cells (H) from 16 fetuses (4 litters).

Figure 6. Coimmunoprecipitation of endogenous TRPC1 with SERCA2b and IP₃R-1 in mouse cortical astrocytes

A, TRPC1 immunoprecipitate (IP) was generated and probed with anti-SERCA2b and anti-IP₃R type 1 antibodies. Control beads were prepared with rabbit γ-globulin (γ-Gl). Gels were loaded with homogenate (Homog, 1st lane; 2 μg protein for SERCA2b and 10 μg protein for IP₃R-1), IP pellet (IP P, 2nd lane; 10 μl of IP protein solution for all co-IPs); post-IP supernatant (IP S, 3rd lane, same volume as 1st lane) and pellet eluted from γ-globulin-IP beads (Con P, 4th lane). B, SERCA2b and IP₃R-1 IPs probed with TRPC1 antibody. Reactivity to anti-TRPC1 was blocked by preincubating the antibody with the peptide (not shown). Molecular masses are indicated in kDa.